Te ara tika o te hauora hapori

Journal of the Medical Association Vol 134 | No 1538 | 9 July 2021 Bumper Issue

COVID-19 pandemic studies of relevance to Aotearoa New Zealand Te ara tika o te hauora hapori Publication information published by the New Zealand Medical Association

NZMJ Editor NZMA Chair Professor Frank Frizelle Dr Alistair Humphrey

NZMJ Production Editor NZMA Communications Manager Richard Beer Simon Bradford

Other enquiries to: To contribute to the NZMJ, first read: NZMA www.nzma.org.nz/journal/contribute PO Box 156 © NZMA 2021 The Terrace Wellington 6140 Cover photo and editorial thumbnail Phone: (04) 472 4741 courtesy of the Ministry of Health

To subscribe to the NZMJ, email [email protected]

Subscription to the New Zealand Medical Journal is free and automatic to NZMA members. Private subscription is available to institutions, to people who are not medical practitioners, and to medical practitioners who live outside New Zealand. Subscription rates are below. All access to the NZMJ is by login and password, but IP access is available to some subscribers. Read our Conditions of access for subscribers for further information www.nzma.org.nz/journal/subscribe/conditions-of-access If you are a member or a subscriber and have not yet received your login and password, or wish to receive email alerts, please email: [email protected] The NZMA also publishes the NZMJ Digest. This online magazine is sent out to members and subscribers six times a year and contains selected material from the NZMJ, along with all obituaries, summaries of all articles, and other NZMA and health sector news and information. Subscription rates for 2021 New Zealand subscription rates Overseas subscription rates Individuals* $349 Individual $486 Institutions $604 Institutions $650 Individual article $33 Individual article $33

*NZ individual subscribers must not be doctors (access is via NZMA Membership) New Zealand rates include GST. No GST is included in international rates. Note, subscription for part of a year is available at pro rata rates. Please email [email protected] for more information. Individual articles are available for purchase by emailing [email protected]

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA www.nzma.org.nz/journal contents

EDITORIALS 68 9 Making sure the New Zealand Bumper issue of COVID-19 border is not our Achilles heel: pandemic studies of relevance to repeated cross-sectional COVID-19 Aotearoa New Zealand surveys in primary care Nick Wilson, Jennifer A Summers, Kyle Eggleton, Nam Bui, Leah Grout, Michael G Baker Felicity Goodyear-Smith ARTICLES 77 An NP-led pilot telehealth 18 programme to facilitate guideline- Deficient hand washing amenities directed medical therapy for heart in public in the time of failure with reduced ejection the COVID-19 pandemic: fraction during the COVID-19 a multi-regional survey pandemic Nick Wilson, George Thomson Andy McLachlan, Chris Aldridge, 28 Mary Morgan, Mayanna Lund, Ruvin Gabriel, Valerio Malez Māori and Pacific people in New Zealand have higher risk of 89 hospitalisation for COVID-19 Empty waiting rooms: Nicholas Steyn, Rachelle N Binny, the New Zealand general practice Kate Hannah, Shaun C Hendy, Alex James, experience with telehealth during Audrey Lustig, Kannan Ridings, the COVID-19 pandemic Michael J Plank, Andrew Sporle Geraldine Wilson, Olivia Currie, 44 Susan Bidwell, Baraah Saeed, Higher perceived stress and Anthony Dowell, Andrew Adiguna Halim, Les Toop, Ann Richardson, exacerbated motor symptoms in Ruth Savage, Ben Hudson Parkinson’s disease during the COVID-19 lockdown in New Zealand 102 Rebekah L Blakemore, Maddie J Pascoe, Emergency COVID-19 funding to Kyla-Louise Horne, Leslie Livingston, general practices in early 2020: Bob N Young, Beth Elias, Marie Goulden, lessons for future allocation Sophie Grenfell, Daniel J Myall, to support equity Toni L Pitcher, John C Dalrymple-Alford, Vanessa Selak, Sue Crengle, Campbell J Le Heron, Tim J Anderson, Matire Harwood, Samantha Murton, Michael R MacAskill Peter Crampton 52 111 Life during lockdown: COVID-19 and the impact on a qualitative study of low-income urology service provision at Capital New Zealanders' experience during & Coast District Health Board the COVID-19 pandemic Simon Lambracos, Lance Yuan, Kimberley Choi, Namratha Giridharan, Andrew Kennedy-Smith Abigail Cartmell, Dominique Lum, Louise Signal, Viliami Puloka, Rose Crossin, Lesley Gray, Cheryl Davies, Michael Baker, Amanda Kvalsvig

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 3 www.nzma.org.nz/journal contents

120 139 Impact of the COVID-19 Teleophthalmology in the pandemic lockdown on public post-coronavirus era sector ophthalmic work by Francesc March de Ribot, New Zealand’s ophthalmologists Anna March de Ribot, Daniel A R Scott, Peter W Hadden, Kelechi Ogbuehi, Ruth Large Graham A Wilson 144 VIEWPOINTS COVID-19 serology: use and interpretation in New Zealand 128 Gary N McAuliffe, Timothy K Blackmore Addressing structural discrimination: prioritising people CLINICAL CORRESPONDENCE with mental health and addiction 148 issues during the COVID-19 Dangers of a single pellet pandemic Jeong Ha, Olga Korduke, Helen Lockett, Ashley Koning, Megan Rodney, Peter Stiven Cameron Lacey, Susanna Every-Palmer, Kate M Scott, Ruth Cunningham, 100 YEARS AGO Tony Dowell, Linda Smith, Alison Masters, Arran Culver, Stephen Chambers 150 Notes on a Case of 135 “Brodie's Abscess” or How were medical students “Circumscribed Abscess” of the from , Lower End of the Left Femur New Zealand, involved in their 1921 COVID-19 response? Matthew J Cowie, Cicely V M Barron, Anna G Bergin, Noella N Farrell, India G Hansen

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 4 www.nzma.org.nz/journal summaries

Deficient hand washing amenities in public toilets in the time of the COVID-19 pandemic: a multi-regional survey Nick Wilson, George Thomson This study found major gaps in the way New Zealand public toilets are designed and serviced that could contribute to the risk of infectious disease transmission. This issue is relevant to helping with control of any COVID-19 outbreaks but also other diseases such as the more common norovirus infection. Nevertheless, hand hygiene may only be of relatively minor importance for COVID-19 control when compared to reducing aerosol transmission via improved ventilation, mask use and physical distancing. Māori and Pacific people in New Zealand have higher risk of hospitalisation for COVID-19 Nicholas Steyn, Rachelle N Binny, Kate Hannah, Shaun C Hendy, Alex James, Audrey Lustig, Kannan Ridings, Michael J Plank, Andrew Sporle We use data on cases of COVID-19 in Aotearoa New Zealand up to September 2020 to estimate the risk of hospitalisation for different age and ethnicity groups. After controlling for age and pre-existing conditions, we find that Māori and Pacific people face significantly higher risk of hospitalisation with COVID-19. Our work focuses on the likelihood of hospitalisation given infection, but other risk factors may mean that Māori and Pacific communities also face greater risk of infection. Higher perceived stress and exacerbated motor symptoms in Parkinson’s disease during the COVID-19 lockdown in New Zealand Rebekah L Blakemore, Maddie J Pascoe, Kyla-Louise Horne, Leslie Livingston, Bob N Young, Beth Elias, Marie Goulden, Sophie Grenfell, Daniel J Myall, Toni L Pitcher, John C Dalrymple-Alford, Campbell J Le Heron, Tim J Anderson, Michael R MacAskill This manuscript describes findings from a study conducted during New Zealand’s strict and relatively successful lockdown period to contain the SARS-CoV-2 virus. We surveyed our Parkinson’s disease (PD) and healthy control cohorts who are part of our established PD longitudinal study at the New Zealand Brain Research Institute. Here we report strong evidence of an association between higher levels of perceived stress and worsening of PD motor symptoms and higher perceived stress in PD than controls since the lockdown began. This report highlights the close interaction between stress and altered movement function in PD, indicating that monitoring and minimising stress levels during the pandemic may be an important adjunct strategy to improve motor function in PD.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 5 www.nzma.org.nz/journal summaries

Life during lockdown: a qualitative study of low-income New Zealanders' experience during the COVID-19 pandemic Kimberley Choi, Namratha Giridharan, Abigail Cartmell, Dominique Lum, Louise Signal, Viliami Puloka, Rose Crossin, Lesley Gray, Cheryl Davies, Michael Baker, Amanda Kvalsvig This research explores the experience of 27 low-income New Zealanders during the COVID-19 pandemic lockdown. Life during lockdown was challenging for study participants who were fearful of the virus and experienced mental distress and isolation. Participants were also resourceful and resilient including using technology, self-help techniques and support from others. New Zealand’s welfare state ensured participants had access to health services and welfare payments, but there were challenges. Despite welcome increases in welfare payments these did not fully meet participants’ needs, and support from charitable organisations was critical. Participants were overwhelmingly positive about the Government’s response and advised the Government to take the same approach in the future. An early and hard lockdown, the welfare state, compassion and clearly communicated leadership were keys to a successful pandemic response for the low-income people in this study. Making sure the New Zealand border is not our Achilles heel: repeated cross-sectional COVID-19 surveys in primary care Kyle Eggleton, Nam Bui, Felicity Goodyear-Smith This is paper presents a primary care perspective on border openings in the context of the COVID-19 pandemic. The study itself is part of an international collaboration designed to analyse and disseminate concerns of primary care in relation to the pandemic. Results from this study show that primary care has expressed repeated concerns about opening the border and see the border as being an ‘Achilles heel’ in our defence against COVID. And the fragile nature of the border means that primary care believes that effective strategies to reduce local transmission must be maintained. An NP-led pilot telehealth programme to facilitate guideline- directed medical therapy for heart failure with reduced ejection fraction during the COVID pandemic Andy McLachlan, Chris Aldridge, Mary Morgan, Mayanna Lund, Ruvin Gabriel, Valerio Malez For most patients, the home monitoring/telephone process resulted in rapid titration and less need for clinic review. Patients found the process acceptable and 60% of clinic visits were able to be held remotely, saving patients both time and money. Titration rates and markers of improved outcomes improved across cardiac imaging, biochemical and clinical findings and were comparable to most real-world clinical reports. Although this is not a novel or inno- vative process, it is not usual care in New Zealand. However, this simple and straightforward process could be replicated across DHBs.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 6 www.nzma.org.nz/journal summaries

Empty waiting rooms: the New Zealand general practice experience with telehealth during the COVID-19 pandemic Geraldine Wilson, Olivia Currie, Susan Bidwell, Baraah Saeed, Anthony Dowell, Andrew Adiguna Halim, Les Toop, Ann Richardson, Ruth Savage, Ben Hudson In response to the COVID-19 pandemic, in early 2020 New Zealand general practices rapidly changed in the way they delivered healthcare, including using telehealth (eg, by phone or video) instead of seeing patients in person. For most patients and primary care practices this was convenient and allowed for patients to be safely given the care they required. There were problems where practices and patients did not have access to suitable technology to use telehealth, and some patient groups were found to be more disadvantaged. To make sure future telehealth consultations can be used by all in an equitable way, we propose that practices are given adequate funding and training; patients need to be given clear commu- nication about how telehealth works and skills and access to technology to use it. Telehealth worked best where there were existing relationships between the practice and patients, and we recommend that healthcare provided by telehealth is integrated with existing primary healthcare services. Emergency COVID-19 funding to general practices in early 2020: lessons for future allocation to support equity Vanessa Selak, Sue Crengle, Matire Harwood, Samantha Murton, Peter Crampton There were serious concerns about the financial sustainability of general practices during the COVID-19 national lockdown in early 2020, which led to the Ministry of Health to provide emergency funding to support general practices in March and April 2020. We sought to describe the distribution of this emergency funding according to the proportion of high needs patients (Māori, Pacific and those living in areas with the highest level of socioeconomic deprivation) in each practice. We found that although the March payment was higher for practices with higher proportions of patients with high needs relative to those with lower proportions of such patients, the April payment was similar irrespective of patient need. In order to meet its aspirations to support equitable health outcomes, the Ministry of Health should apply pro-equity resource allocation in all emergency circumstances. COVID-19 and the impact on urology service provision at Capital & Coast District Health Board Simon Lambracos, Lance Yuan, Andrew Kennedy-Smith Our study demonstrated how restrictions as a result of the COVID-19 pandemic restrictions affected urology service provision in the Capital & Coast District Health Board region. The adjustments that were implemented to our service showed how we could function more efficiently, economically and environmentally. It also set a potential precedent for optimising healthcare service provision in other district health boards across the country on a routine basis as well in response to a crisis. Impact of the COVID-19 pandemic lockdown on public sector ophthalmic work by New Zealand’s ophthalmologists Daniel A R Scott, Peter W Hadden, Graham A Wilson Publicly employed ophthalmologists experienced dramatic reductions to elective clinic and operating volumes during the COVID-19 lockdown. National-level information confirmed clinic and elective operating volumes reduced to 38.2% and 11.5% of usual service volumes during April 2020. The return of service delivery volumes back to normal within two months supports the value of a COVID-19 elimination strategy in New Zealand. A 17.9-fold increase in the volume of non-contact/virtual visits allowed ongoing management without risking virus transmission for selected patients. At a personal level, the lockdown resulted in reported physical health benefits for ophthalmologists.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 7 www.nzma.org.nz/journal summaries

Addressing structural discrimination: prioritising people with mental health and addiction issues during the COVID-19 pandemic Helen Lockett, Ashley Koning, Cameron Lacey, Susanna Every-Palmer, Kate M Scott, Ruth Cunningham, Tony Dowell, Linda Smith, Alison Masters, Arran Culver, Stephen Chambers This paper resulted in people with mental health and addiction issues being recognised as a priority group for COVID-19 vaccinations. People with mental health and addiction issues have a significantly reduced life expectancy, much of which can be attributed to physical illnesses that, if recognised earlier and treated fairly, could have been mitigated. Structural discrimination worsens these physical health inequities. Structural discrimination is where the policies and practices of health care organisations impact unfairly on access to and quality of care. An expert advisory group, convened as part of the Aotearoa Equally Well collabo- rative, considered findings of a literature review on the vulnerability of people with mental health and addiction issues of contracting and dying from COVID-19. The group concluded mental health and addiction issues should be recognised as underlying health conditions that increase COVID-19 vulnerability, and that people with these issues should be prioritised for vaccination. The COVID-19 vaccination sequencing framework adopted the recommenda- tions, placing people with mental health and addiction issues in vaccination group 2 or 3. How were medical students from Christchurch, New Zealand, involved in their COVID-19 response? Matthew J Cowie, Cicely V M Barron, Anna G Bergin, Noella N Farrell, India G Hansen Medical students from the , Christchurch School of Medicine were involved in their local COVID-19 response. A group of ten students helped with the assessment of indi- viduals at community-based assessment centres or mobile testing units. They primarily helped assess and test individuals alongside experienced healthcare workers. The students gained valuable clinical and public health experience. Key learning points were the risks with pandemic involvement, identifying local barriers to healthcare and developing an appre- ciation for an evolving health response. Overall, students felt that preparation for future involvement could benefit further pandemic responses. COVID-19 serology: use and interpretation in New Zealand Gary N McAuliffe, Timothy K Blackmore We describe how serology is used to test for COVID -19 infections in New Zealand and how the different available tests fit in with the body’s response to infection. We provide some guidance to interpretation of the different tests for New Zealand doctors. Dangers of a single pellet Jeong Ha, Olga Korduke, Megan Rodney, Peter Stiven In this case study a patient presented to a rural New Zealand with a shotgun wound to the thigh. Although selective non-operative management is described throughout the liter- ature, this presents issues within the context of a rural hospital. It is uncommon for clinicians to encounter gunshot wounds in New Zealand. The decision to commence explorative surgery was made and perforation of the bowel from a stray pellet was found.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 8 www.nzma.org.nz/journal editorial

Bumper issue of COVID-19 pandemic studies of relevance to Aotearoa New Zealand Nick Wilson, Jennifer A Summers, Leah Grout, Michael G Baker

ABSTRACT In response to the COVID-19 pandemic, Aotearoa New Zealand adopted a clear ‘elimination strategy’, which has (up to June 2021) been very successful in both health and economic terms compared to other OECD countries. Nevertheless, the pandemic response has still been a very major shock to the New Zealand health system. This issue of the New Zealand Medical Journal has 14 new pandemic-related articles. Some of this work can help inform vaccination prioritisation decisions and inform preparations of primary and secondary care services and social services for any future raising of levels in the pandemic Alert Level system. Particularly strong themes are around the value (and challenges) of telehealth services, and also the need for responses throughout the health system to ensure health equity and support for the most vulnerable citizens.

s with a number of other jurisdic- Zealand topped a ‘normalcy index’ that tions in the Asia–Pacific region, New assessed return to ‘pre-pandemic life’.13 AZealand adopted tight border controls However, a full and proper analysis of New and other stringent public health and social Zealand’s elimination strategy response to the 1 measures (PHSMs) to control the COVID-19 COVID-19 pandemic will need to take account pandemic. The country’s clearly articulated of a multi-year perspective. In particular, it 2 COVID-19 ‘elimination strategy’ has been will need to be done after COVID-19 vacci- 3–5 remarkably successful. Despite occasional nation coverage has stabilised in New border system failures that have caused out- Zealand and comparable OECD countries. 6 breaks, the country has regained its elimina- Indeed, the country is still at risk of large tion status after each instance (at least up to outbreaks until it achieves high vaccination late June 2021). coverage (it was near the bottom of the OECD Indeed, New Zealand has the lowest on 24 June 2021 for people fully vaccinated14 COVID-19 cumulative death rate in the OECD and equity goals were not being met15). (data from the Worldometers website7 on 26 The Government also needs to upgrade the June 2021). It has also had the lowest level of outdated Alert Level system,16 integrate ‘excess deaths’ among OECD countries8 and mass masking in a systematic manner17 and within a grouping of 29 high-income coun- enhance border protections, along with other tries.9 Similarly, New Zealand was one of potential upgrades.18,6 There are of course only a few high-income countries where life numerous lessons for the future in terms of expectancy actually increased between 2018 enhancing New Zealand’s pandemic response and 2020, with pandemic-related reductions capabilities.19 in the others.10 New Zealand has also done better than the What are some of the health OECD average in terms of average changes impacts of the pandemic in quarterly GDP (from Q1 2020 to Q1 2021 response in New Zealand? and with higher growth in the first quarter of Much research relating to COVID-19 and 2021).11 It also had relatively lower increases New Zealand has already been reported, in unemployment than the OECD average.12 including the psychological distress asso- Also, relative to other OECD countries, New ciated with raised Alert Levels.20 There

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 9 www.nzma.org.nz/journal editorial

was also an increase in alcohol-related response Alert Level system in response to emergencies involving ambulance staff outbreaks or the threat of outbreaks (Table attendances in 2020,21 and increased smoking 1). A particularly strong theme is around the levels in some groups.22 Potential adverse expanded use of telehealth services, with impacts of increased COVID-19-related this being the major theme for three articles unemployment onto cardiovascular disease and also considered in three others. The have been modelled.23 Publications have overall impression given is that telehealth also considered equity issues around health services were very useful when Alert Levels service impacts,24 and the perspectives of were raised, albeit with various limitations Māori16,25,26 and Pasifika.27 Although it appears and issues of concern raised (eg, risk of that cancer care services were disrupted increasing inequities). Some authors consider by the pandemic response, this was rela- that increased routine use of telehealth in tively minor overall (eg, “an 8% year-to-date some areas of healthcare delivery may have decrease in radiation therapy attendances”).24 long-term efficiency benefits. One study also There was also “little evidence of differential includes qualitative data on the use of tele- impact of COVID-19 on access to cancer diag- health for contacting Māori patients through nosis and care between ethnic groups,” but a marae clinic.36 for lung cancer there was a decrease in new A notable feature of this body of new 24 diagnoses among Māori. articles is that many consider aspects of Identified benefits of the response equity in terms of ethnic or income ineq- included that the experience of the raised uities,32,35,37 those with chronic/underlying Alert Levels had positive psycho-social conditions,32,33 how government funding aspects for some people.28 There was also a support for general practices was not consis- reduction in 2020 in infectious respiratory tently pro-equity during the response38 and diseases.29–31 These reductions in infectious whether telehealth in primary care exacer- diseases may have long-term implications bates inequities.36 With regard to the latter for disease control (eg, the value of staying it might be that, for well-designed telehealth at home when unwell and mask wearing on services (as argued for in one article39), there public transport in winter months). could be long-term equity benefits if these new services can be used to reduce waiting What do the new studies in this times and improve service delivery to Journal show? underserved communities. But to facilitate The 14 new articles relating to COVID-19 in this, further improvements could be made this issue of the New Zealand Medical Journal to internet broadband and mobile phone span epidemiology and public health (n=4), access across the country (as per some of the secondary care services (n=3), telehealth difficulties identified in one study36). services (n=3) and various other COVID-19 issues (n=4) (see Table 1 for brief summary Conclusions details). Particularly notable is the evidence In response to the COVID-19 pandemic, for increased risk of hospitalisation from Aotearoa New Zealand adopted a clear COVID-19 for Māori and Pasifika by Steyn elimination strategy, which has (up to June et al.32 This work has immediate rele- 2021) been very successful in both health vance to prioritisation with the current and economic terms compared to other COVID-19 vaccine rollout, as does the article OECD countries. Nevertheless, the pandemic presenting the case for prioritising those response has still been a very major shock with mental health and addiction issues by to the New Zealand health system. This issue Lockett et al.33 Also of substantial current of the Journal includes work that can help relevance is thinking around the importance inform vaccination prioritisation decisions of border controls by Eggleton et al34 and and preparations of primary and secondary the health and social support needed for care and social services for any future raising low-income people if raised Alert Levels are of levels in the Alert Level system. Partic- required again (the work by Choi et al35). ularly strong themes are around the value (and challenges) of telehealth services, and The articles relating to secondary care also the need for responses throughout the provision can all provide lessons if New health system to ensure health equity and Zealand needs to go up levels in the COVID-19 support for the most vulnerable citizens.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 10 www.nzma.org.nz/journal editorial

Table 1: New studies in this issue of this Journal on COVID-19 pandemic-related issues in Aotearoa New Zealand.

Topic and authors Key findings

Epidemiology and public health Risk of hospitalisation This analysis of 1,829 COVID-19 cases in New Zealand reported that Māori with COVID-19 in New had 2.5 times greater odds of hospitalisation and Pacific people 3 times Zealand (Steyn et al32) greater odds than non-Māori, non-Pacific people (after controlling for age and pre-existing conditions). The authors concluded that “structural inequi- ties and systemic racism in the healthcare system mean that Māori and Pa- cific communities face a much greater health burden from COVID-19. Older people and those with pre-existing health conditions are also at greater risk.” The authors state that these findings should inform future decisions around prioritisation for vaccination.

Prioritising people This article reports on the work of an expert advisory group convened as with mental health part of the Aotearoa Equally Well collaborative. It found that “evidence and addiction issues indicates an association between mental health and addiction issues and in- for vaccination (Lock- fection risk and worse outcomes.” “The group concluded mental health and ett et al33) addiction issues should be recognised as underlying health conditions that increase COVID-19 vulnerability, and that people with these issues should be prioritised for vaccination.” The authors argue that “addressing these inequities must be integral in modern health policy—including our COVID-19 pandemic response.”

Views on border con- From three surveys of primary care practices, this study reported increasing trol and other control support for “opening a trans-Tasman border but not internationally.” Two strategies (Eggleton broad themes were for making sure that the border is not an Achilles heel et al34) and effective strategies to reduce local transmission. Sub-themes included community control, tracing and testing individuals and vaccinating pop- ulation. An issue raised concerned the need to prevent pandemic spread from New Zealand: “Would be scared of NZ taking it into Pacific Islands after measles problems.”

Handwashing ameni- This study concluded that “although handwashing is probably a much less ties in public toilets critical COVID-19 control intervention than reducing aerosol transmission, it (Wilson and Thom- should still be strongly supported. Yet this survey found multiple deficiencies son40) with handwashing amenities at public toilets and only modest improve- ments since a previous survey.”

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 11 www.nzma.org.nz/journal editorial

Table 1: New studies in this issue of this Journal on COVID-19 pandemic-related issues in Aotearoa New Zealand (continued).

Topic and authors Key findings

Secondary care Impact of the raised This study surveyed 134 people with PD and 49 healthy controls, and re- Alert Levels on pa- ported that perceived stress was higher in PD patients than controls and “in tients with Parkinson’s those reporting a worsening of tremor, balance/gait, dyskinesia and bradyki- disease (PD) (Blake- nesia compared to those indicating no change during the COVID-19 lock- more et al41) down.” The authors conclude that “Reducing stressors may be an important adjunct treatment strategy to improve motor function in PD.”

Delivery of ophthal- This study surveyed ophthalmologists nationwide and found that a large mology services and majority of respondents (82% and 98% respectively) reduced elective clinic the raised Alert Levels and surgical volumes by at least 75%. National-level information confirmed (Scott et al42) major reductions in clinics (down to 38.2% of normal) and elective operating volumes (down to 11.5%), with virtual visits increasing 18-fold. However, recovery was rapid with: “Elective clinic and elective operating volumes promptly recovered to usual volumes on the second month post lockdown.” In terms of telehealth, the authors note that “this form of service delivery may have a greater role in our overburdened public health system for the future.”

Impact on a urology This study detailed how the raised Alert Levels resulted in “an overall reduc- service from the raised tion in service delivery and a reorientation to non-contact outpatient consul- Alert Levels (Lambrac- tations.” But this was “mitigated by proactive outsourcing of elective surgery os et al43) to a private hospital and a dramatic shift to virtual consultations.” The authors report that this experience can inform crisis response management for the future but also the potential benefits of telehealth going forward: “Furthermore, with regard to the virtual consultation platform, the data also suggest ways in which our practice can be adapted on a routine basis in the future, in order to increase efficiency and to provide a service that is both economic to the patient and environmentally prudent.” A notable feature of this study was how the telehealth aspect was estimated to have saved “an average 22.7km of travel per patient,” with benefits for the environment, and out-of-pocket travel costs for patients.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 12 www.nzma.org.nz/journal editorial

Table 1: New studies in this issue of this Journal on COVID-19 pandemic-related issues in Aotearoa New Zealand (continued).

Topic and authors Key findings

Experiences with telehealth (see also the articles on urology services and ophthalmology services in the two rows directly above, and the study by Choi et al in the next subsection below) Experiences with tele- This study reported on nationwide surveys of New Zealand general prac- health in primary care tice teams. It reported that telehealth consultations were “most successful (Wilson et al36) where there was a pre-existing relationship between healthcare provid- er and patient.” But various barriers identified included “technological challenges, communication difficulties for those with hearing impairments, concern regarding the cost and difficulty in making online payments.” The authors noted that the experiences described were “consistent with other international work showing that telehealth risks increasing inequity” (eg, “it can create extra barriers for those who are already disadvantaged, such as those in rural areas, those with hearing impairment or cognitive decline and refugee and migrant populations who may have language barriers”). Of note was that after the Alert Level restrictions there was a “rapid move back to in-person care and ‘business as usual’ was felt by the GP teams to be driven by patient choice. So while telehealth may play an increasing role in the future, it is unlikely to fully replace in-person care.” The authors provide a number of recommendations for improving the use of telehealth in primary care settings.

A nurse practi- This study found that “for most patients, the home monitoring/telephone tioner-led telehealth process resulted in rapid titration and less need for clinic review. Patients programme for heart found the process acceptable and 60% of clinic visits were able to be held failure management remotely, saving patients both time and money.” Titration rates and markers (McLachlan et al37) of improved outcomes improved across cardiac imaging, biochemical and clinical findings and were comparable to most real-world clinical reports.” The authors suggest that this simple and straightforward process could be replicated across District Health Boards.

Telehealth and oph- This article considers some of the international literature around ‘teleoph- thalmology (March de thalmology’ and the New Zealand situation with respect to service demand Ribot et al39) and the impact of the raised Alert Levels. The authors argue the case that te- leophthalmology could improve the referral process, and if teleophthalmol- ogy is properly implemented, they anticipate “a 40% decrease in the number of referrals to public ophthalmology services in New Zealand, which would improve the workflow in ophthalmology departments of public by about 20%.” Limitations such as cost are discussed but overall the authors argue that “now is the moment to implement innovations so as not to leave anyone behind.”

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 13 www.nzma.org.nz/journal editorial

Table 1: New studies in this issue of this Journal on COVID-19 pandemic-related issues in Aotearoa New Zealand (continued).

Topic and authors Key findings

Other A qualitative study of This qualitative study used 27 interviews with low-income people in June–July low-income New Zea- 2020 (immediately after ‘lockdown’ was lifted). It reported that life during landers’ experiences lockdown was challenging for study participants. “They were fearful of the with raised Alert Levels virus and experienced mental distress and isolation. Most participants felt (Choi et al35) safe at home and reported coping financially while still experiencing financial stress. Participants were resourceful and resilient. They coped with lockdown by using technology, self-help techniques and support from others.” The study found that, although participants had access to health services and welfare payments, “welfare payments did not fully meet participants’ needs, and support from charitable organisations was critical.” Nevertheless, participants were “overwhelmingly positive about the Government’s response and advised the Government to take the same approach in the future.” The study authors concluded that “An early and hard lockdown, the welfare state, compassion and clearly communicated leadership were keys to a successful lockdown for the low-income people in this study.” They also note that capturing the experi- ence of low-income people during pandemics “is critical to ensuring inequities in pandemic impact are mitigated.”

Ministry of Health This study reported that initial emergency financial support in March 2020 (MOH) funding of gen- for general practices was higher for those with more high-needs patients. eral practices (Selak But this was not the case for the funding in April 2020. The authors argue et al38) that “in the future, the MOH should apply pro-equity resource allocation in all emergencies, as with other circumstances.” The article provides valuable context in terms of the inequitable burden of COVID-19 according to ethnicity and also evidence for the wider problem of inadequate New Zealand Govern- ment funding of health services according to need.

Review of COVID-19 se- This article provides a review of the use of COVID-19 serology in the New rology in the New Zea- Zealand context. “Testing may provide useful information in public health in- land context (McAuliffe vestigations or select cases of post-infectious complications and is necessary and Blackmore44) for overseas travel to some destinations.” But the authors note that “test reliability varies substantially according to the testing scenario.” Importantly they note that “the role of post-vaccination serology testing as a correlate of immunity has not yet been determined,” and make an argument for clinical microbiologist advice for interpretation in “high-consequence cases.”

Medical student contri- This article describes how a group of New Zealand medical students were bution to the COVID-19 involved in a local COVID-19 response. It identified both the helpful contri- response (Cowie et bution the students made to the response, alongside the “valuable clinical al45) and public health experience” gained. They reported that “we found our involvement rewarding, whether it was on the frontline or not, and the level of risk balanced well with learning opportunities.” Home visits for COVID-19 testing were also considered valuable from a learning perspective: “These visits let us view living situations from the centre of a patient’s home. This left a lasting impact on many of us and cemented a strong reminder of how risk factors and living conditions can impact upon health.”

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 14 www.nzma.org.nz/journal editorial

Competing interests: One of these 14 COVID-19 related articles described here involved the first author of this Editorial (ie, the survey by Wilson and Thomson). He is also the sibling of the third author in the survey by Scott et al. The last author of this editorial was also a contributor to the article by Choi et al. Acknowledgements: Professor Baker acknowledges funding support from the Health Research Council of New Zealand (20/1066). Author information: Nick Wilson: Professor, Department of Public Health, University of Otago Wellington. Jennifer Summers: Senior Research Fellow, Department of Public Health, University of Otago Wellington. Leah Grout: Research Fellow, Department of Public Health, University of Otago Wellington. Michael Baker: Professor, Department of Public Health, University of Otago Wellington. Corresponding author: Professor Nick Wilson, Director of the BODE3 Programme, Department of Public Health, University of Otago Wellington [email protected] URL: www.nzma.org.nz/journal-articles/bumper-issue-of-covid-19-pandemic-studies-of-relevance- to-aotearoa-new-zealand

REFERENCES 1. Hale T, Angrist N, response: a descriptive (26 June 2021 data). https:// Goldszmidt R, Kira B, Peth- epidemiological study. www.worldometers. erick A, Phillips T, Webster Lancet Public Health 2020. info/coronavirus/. S, Cameron-Blake E, Hallas 4. Baker MG, Wilson N, 8. The Economist. Tracking L, Majumdar S, Tatlow H. Anglemyer A. Success- covid-19 excess deaths A global panel database of ful elimination of across countries (11 pandemic policies (Oxford Covid-19 transmission May 2021 update). The COVID-19 Government in New Zealand. N Engl Economist. https:// Response Tracker). (Data J Med 2020;(7 August) www.economist.com/ for 19 June 2021 at: doi:101056/NEJMc2025203. graphic-detail/coronavi- https://ourworldindata. 5. Baker M, Wilson N, Blakely rus-excess-deaths-tracker org/grapher/covid-strin- T. Elimination may be the 9. Islam N, Shkolnikov VM, gency-index). Nat Hum optimal response strategy Acosta RJ, Klimkin I, Kawa- Behav 2021;5:529–38. for covid-19 and other chi I, Irizarry RA, Alicandro 2. Baker M, Kvalsvig A, emerging pandemic diseas- G, Khunti K, Yates T, Jdanov Verrall A, Telfar-Barnard es. BMJ 2020;371:m4907. DA, White M, Lewington L, Wilson N. New Zealand’s 6. Grout L, Katar A, Ait S, Lacey B. Excess deaths elimination strategy for Ouakrim D, Summers associated with covid-19 the COVID-19 pandemic J, Kvalsvig A, Baker M, pandemic in 2020: age and what is required to Blakely T, Wilson N. and sex disaggregated make it work. N Z Med J Estimating the failure risk time series analysis in 29 2020;133(1512):10-14. of quarantine systems high income countries. 3. Jefferies S, French N, for preventing COVID-19 BMJ 2021;373:n1137. Gilkison C, Graham G, Hope outbreaks in Australia and 10. Woolf S, Masters R, Aron V, Marshall J, McElnay New Zealand. medRxiv L. Effect of the covid-19 C, McNeill A, Muellner P, 2021;(3 July). https://www. pandemic in 2020 on Paine S, Prasad N, Scott medrxiv.org/content/10.11 life expectancy across J, Sherwood J, Yang L, 01/2021.02.17.21251946v4 populations in the USA Priest P. COVID-19 in 7. Worldometers [Internet]. and other high income New Zealand and the COVID-19 Coronavirus countries: simulations of impact of the national Pandemic. Worldometer provisional mortality data.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 15 www.nzma.org.nz/journal editorial

BMJ 2021;373:n1343. peoples. Lancet Reg Health One 2021;16:e0246053. 11. OECD [Internet]. Quarterly West Pac In press. 24. Gurney JK, Millar E, Dunn GDP. OECD, 2021 (Accessed 17. Kvalsvig A, Wilson N, Chan A, Pirie R, Mako M, Mander- 12 June 2021). https://data. L, Febery S, Roberts S, Betty son J, Hardie C, Jackson C, oecd.org/gdp/quarterly-gdp. B, Baker M. Mass masking: North R, Ruka M, Scott N, htm#indicator-chart. an alternative to a second Sarfati D. The impact of 12. OECD [Internet]. lockdown in Aotearoa. N Z the COVID-19 pandemic Unemployment rate. Med J 2020;133(1517):8-13. on cancer diagnosis and OECD, 2021 (Accessed 18. Kvalsvig A, Wilson service access in New 25 June 2021). https:// N, Baker M. Urgently Zealand-a country pursuing data.oecd.org/unemp/ upgrading NZ’s Covid-19 COVID-19 elimination. unemployment-rate.htm. response. Public Health Lancet Reg Health West Pac 2021;10:100127. 13. The Economist [Internet]. Expert (Blog) 2021;(24 Covid-19 data: The global June). https://blogs.otago. 25. McLeod M, Gurney J, normalcy index. The ac.nz/pubhealthexpert/ Harris R, Cormack D, Economist 2021;(1 July). urgently-upgrading-nzs- King P. COVID-19: we https://www.economist. covid-19-response/. must not forget about com/graphic-detail/ 19. Kvalsvig A, Baker M. How Indigenous health and tracking-the-return-to- Aotearoa New Zealand equity. Aust N Z J Public normalcy-after-covid-19. rapidly revised its Covid-19 Health 2020;44:253-56. 14. Our World in Data [Inter- response strategy: lessons 26. Manuirirangi K, Jarman net]. Share of people for the next pandemic J. The Taranaki COVID-19 fully vaccinated against plan. J Roy Soc N Z response from a Maori COVID-19. Our World in 2021;31;51(Suppl1):S143-66. perspective: lessons Data 2021;(24 June data). 20. Every-Palmer S, Jenkins M, for mainstream health https://ourworldindata.org/ Gendall P, Hoek J, Beagle- providers in Aotearoa explorers/coronavirus-da- hole B, Bell C, Williman New Zealand. N Z Med J ta-explorer?tab=table&zoom- J, Rapsey C, Stanley J. 2021;134;(1533):122-24. ToSelection=true&picker- Psychological distress, 27. Ioane J, Percival T, Laban Sort=desc&pickerMetric=- anxiety, family violence, W, Lambie I. All-of-commu- total_cases&Metric=Peo- suicidality, and wellbeing nity by all-of-government: ple+fully+vaccinated&In- in New Zealand during reaching Pacific people terval=7-day+rolling+av- the COVID-19 lockdown: A in Aotearoa New Zealand erage&Relative+to+Pop- cross-sectional study. PLoS during the COVID-19 ulation=true&Align+out- One 2020;15:e0241658. pandemic. N Z Med J breaks=false&coun- 21. Truebridge N. Alco- 2021;134;(1533):96-103. try=~OWID_WRL. hol-related emergency 28. Jenkins M, Hoek J, Jenkin 15. Cheng D. Covid 19 coro- department visits increased G, Gendall P, Stanley J, navirus: Derek Cheng: in 2020 - data. Radio New Beaglehole B, Bell C, Rapsey How vaccination delivery Zealand 2021;(18 June). C, Every-Palmer S. Silver failed to meet expectations. https://www.rnz.co.nz/ linings of the COVID-19 New Zealand Herald news/national/445055/ lockdown in New Zealand. 2021;(2 July). https:// alcohol-related-emergen- PLoS One 2021;16:e0249678. www.nzherald.co.nz/nz/ cy-department-visits-in- 29. Huang QS, Wood T, Jelley politics/covid-19-coronavi- creased-in-2020-data. L, Jennings T, Jefferies rus-derek-cheng-how-vac- 22. Gendall P, Hoek J, Stanley S, Daniells K, Nesdale cination-deliv- J, Jenkins M, Every-Palmer A, Dowell T, Turner N, ery-failed-to-meet-ex- S. Changes in Tobacco Use Campbell-Stokes P, Balm pectations/ During the 2020 COVID- M, Dobinson HC, Grant GMJ4KU3SLYPC7L6HIL- 19 Lockdown in New CC, James S, Aminisani N, EITT2WCM/. Zealand. Nicotine Tob Ralston J, Gunn W, Bocacao 16. Kvalsvig A, Wilson N, Res 2021;23:866-71. J, Danielewicz J, Moncrieff Davies C, Timu-Parata 23. Nghiem N, Wilson N. Poten- T, McNeill A, Lopez L, Waite C, Signal V, Baker M. tial impact of COVID-19 B, Kiedrzynski T, Schrader Expansion of a national related unemployment on H, Gray R, Cook K, Currin Covid-19 alert level system increased cardiovascular D, Engelbrecht C, Tapurau to improve population disease in a high-income W, Emmerton L, Martin health and uphold the country: Modeling health M, Baker MG, Taylor S, values of Indigenous loss, cost and equity. PLoS Trenholme A, Wong C,

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 16 www.nzma.org.nz/journal editorial

Lawrence S, McArthur C, 2021;134;(1538):128-134. Teleophthalmology in the Stanley A, Roberts S, Rahna- 34. Eggleton K, Bui N, Good- post-coronavirus era. N Z ma F, Bennett J, Mansell year-Smith F. Making sure Med J 2021;134;(1538):139- C, Dilcher M, Werno A, the New Zealand border 143. Grant J, van der Linden is not our Achilles heel: 40. Wilson N, Thomson G. A, Youngblood B, Thomas repeated cross-sectional Deficient handwashing PG, Consortium NP, Webby COVID-19 surveys in amenities in public toilets RJ. Impact of the COVID- primary care. N Z Med J in the time of the COVID-19 19 nonpharmaceutical 2021;134;(1538):68-76. pandemic: a multi-region- interventions on influenza 35. Choi K, Giridharan N, al survey. N Z Med J and other respiratory viral Cartmell A, Lum D, Signal 2021;134;(1538):18-27. infections in New Zealand. L, Puloka V, Crossin R, 41. Blakemore R, Pascoe M, Nat Commun 2021;12:1001. Gray L, Davies C, Baker Horne K-L, Livingston L, 30. Trenholme A, Webb R, M, Kvalsvig A. Life during Young B, Elias B, Goulden Lawrence S, Arrol S, lockdown: a qualitative M, Grenfell S, Myall D, Taylor S, Ameratunga S, study of low-income New Pitcher T, Dalrymple-Alford Byrnes CA. COVID-19 and Zealanders’ experience J, Le Heron C, Anderson Infant Hospitalizations during the COVID-19 T, MacAskill M. Higher for Seasonal Respiratory pandemic. N Z Med J perceived stress and exac- Virus Infections, New 2021;134;(1538):52-67. erbated motor symptoms in Zealand, 2020. Emerg 36. Wilson G, Currie O, Bidwell Parkinson’s disease during Infect Dis 2021;27:641-43. S, Saeed B, Dowell A, Halim the COVID-19 lockdown 31. Duffy E, Thomas M, Hills A, Toop L, Richardson A, in New Zealand. N Z Med T, Ritchie S. The impacts of Savage R, Hudson B. Empty J 2021;134;(1538):44-51. New Zealand’s COVID-19 waiting rooms: The New 42. Scott D, Hadden P, Wilson epidemic response on Zealand general practice G. Impact of the COVID-19 community antibiotic use experience with telehealth pandemic lockdown on and hospitalisation for during the COVID-19 public sector ophthalmic pneumonia, peritonsillar pandemic. N Z Med J work by New Zealand’s abscess and rheumatic 2021;134;( 1538):89-101. ophthalmologists. N Z Med fever. Lancet Reg Health 37. McLachlan A, Aldridge C, J 2021;134;(1538):120-127. West Pac 2021;12:100162. Morgan M, Lund M, Gabriel 43. Lambracos S, Yuan L, 32. Steyn N, Binny R, Hannah R, Malez V. An NP-led pilot Kennedy-Smith A. COVID- K, Hendy S, James A, telehealth programme to 19 and the impact on Lustig A, Ridings K, Plank facilitate guideline-directed Urology service provision M, Sporle A. Māori and medical therapy for heart at Capital & Coast District Pacific people in New failure with reduced Health Board. N Z Med J Zealand have higher ejection fraction during the 2021;134;(1538):111-119. risk of hospitalisation COVID-19 pandemic. N Z 44. McAuliffe G, Blackmore for COVID-19. N Z Med J Med J 2021;134;(1538):77-88. T. COVID-19 serology: 2021;134;(1538):28-43. 38. Selak V, Crengle S, Harwood use and interpretation in 33. Lockett H, Koning A, M, Murton S, Crampton New Zealand. N Z Med J Lacey C, Every-Palmer S, P. Emergency COVID-19 2021;134;(1538):144-147. Scott K, Cunningham R, funding to general practic- 45. Cowie M, Barron C, Dowell T, Smith L, Masters es in early 2020. Lessons Bergin A, Farrell N, A, Culver A, Chambers for future allocation to Hansen I. How were S. Addressing structural support equity. N Z Med J medical students from discrimination: Prioritis- 2021;134;(1538):102-110. Christchurch, New Zealand, ing people with mental 39. March de Ribot F, involved in their COVID- health and addiction March de Ribot A, 19 response? N Z Med J issues during the COVID- Ogbuehi K, Large R. 2021;134;(1538):135-138. 19 pandemic. N Z Med J

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 17 www.nzma.org.nz/journal article

Deficient handwashing amenities in public toilets in the time of the COVID-19 pandemic: a multi-regional survey Nick Wilson, George Thomson

ABSTRACT AIMS: To identify the extent of the provision of handwashing amenities in public toilets at the time of the COVID-19 pandemic, and also to make comparisons with a related pre-pandemic survey. METHODS: We collected data from 400 facilities that were open to the public: all those in three contiguous city council territories (228) and a further convenience sample of 172 around other parts of New Zealand. Comparisons were made with the data on the same facilities included in a 2012/2013 survey. RESULTS: Of the toilets in this survey, 2.5% had no water for handwashing and 14.8% had no soap. There was COVID-19-related health messaging signage in 19.5% of toilets, with posters of the COVID-19 QR code used for contact tracing in 12.3%, and generic handwashing signage in 1.8%. The handwashing water had ‘no-touch’ activation at 28.0% of toilets, and 18.5% of toilets had no-touch bowl flushing. Toilet bowl lids were not present at 32.8%, and 2.3% of toilets had damage that would impair their functionality (eg, broken toilet seats). This new survey found significantly increased provision of soap (risk ratio = 1.47; 95%CI: 1.25 to 1.72), but no increased provision of water, at the 128 sites that had also been examined in the previous survey. CONCLUSIONS: Although handwashing is probably a much less critical COVID-19 control intervention than reducing aerosol transmission, it should still be strongly supported. Yet this survey found multiple deficiencies with handwashing amenities at public toilets and only modest improvements since a previous survey.

he COVID-19 pandemic has focused washing for COVID-19 control, even though international attention on non-phar- the current evidence indicates that the role maceutical interventions to reduce of contaminated surfaces in transmission T 4 pandemic spread prior to vaccination roll- is likely to be small when compared to the out. These interventions include hygiene inhalation route of transmission. In public practices such as appropriate handwashing, toilets, which are often unheated and where which is an evidence-based measure for the surfaces are generally impermeable, preventing respiratory virus transmission.1 cooler temperatures and surfaces more More specifically, the World Health Orga- retentive to coronavirus may increase the nization (WHO) has issued guidance on role of contaminated surfaces in trans- COVID-19 and hygiene/,2 which mission compared to other living and covers the need for handwashing amenities working situations.5,6 with water and soap. Improved hygiene is also beneficial for The COVID-19 pandemic virus (SARS- reducing the spread of other respiratory CoV-2) appears to survive much longer on viruses such as seasonal influenza and human skin than influenza A virus.3 Expert norovirus infection. For example, one commentary has recommended hand- review identified six studies that implicated

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 18 www.nzma.org.nz/journal article

surfaces as primary sources of messaging) in New Zealand public toilets human norovirus infection.7 at the time of the COVID-19 pandemic. We WHO also recommends having separate also aimed to report on other toilet features toilets “for people with suspected or relevant to hygiene and COVID-19 trans- confirmed SARS-CoV-2 infection” and that mission, and to make comparisons with 16 “the toilet should be flushed with the lid a pre-pandemic survey that included a down to prevent droplet splatter and aerosol sample of the same facilities. clouds.”8 The latter recommendation is By way of context, New Zealand has previ- supported by data from a COVID-19 quar- ously reported shortages of public toilets, antine room study, which found that “the especially in areas with high numbers of inner walls of toilet bowl and sewer inlet international tourists.17,18 Previous research were the most contaminated sites with the has also shown deficiencies with public highest viral loads.”8 Another such study toilets in the country, in terms of lacking reported that “there was extensive envi- handwashing water (4%) and soap (39%).16,19 ronmental contamination by 1 SARS-CoV-2 Another study of toilets in one New Zealand patient,” with toilet bowl and sink samples city (mainly at cafés and public facilities) being positive for viral RNA.9 A study found that some had no handwashing using genomic sequencing also implicated facilities (2%) and no soap (13%).20 Within exposure to sewage as a risk factor for SARS- this sample, the lack of soap was highest CoV-2 infection.10 There has also been one in the public toilets, at 38%. Another study study indicating circumstantial evidence of of primary school toilets also reported that faecal aerosol transmission of COVID-19 via only 28% had facilities meeting the relevant an apartment drainage system,11 similar to code of practice (eg, there was a lack of hot an outbreak from faecal aerosols of SARS- water, lack of drying facilities and lack of CoV-1 in 2003.12 soap).21 Inadequate hand hygiene has also In New Zealand, there was community been recorded, with users in spread of the pandemic virus (SARS-CoV-2) on New Zealand sometimes not washing hands 22 a number of occasions during the 2020 year, (13%) or using soap (28%). but elimination was successfully achieved In New Zealand there are a range of public and re-achieved after a series of border toilet types. These include both unisex and/ control failures.13,14 These successes arose or single-sex facilities (male and/or female). largely from a combination of tight border Unisex facilities contain a toilet bowl and controls (quarantine and isolation), a strict usually handwashing amenities, and less lockdown and widespread testing/contact frequently a . Single-sex toilet facilities tracing. Actions related to hygiene included: can contain separate cubicles (with toilet • mass media messages relating to bowls, and sometimes with their own hand- handwashing, cough etiquette, staying washing amenities) and, typically, common home when sick and mask use. handwashing amenities. Male toilet facilities usually include . • actions by organisations to install hand sanitisers and posters with hygiene messaging in workplaces and Methods various public places Definitions • public toilets were all closed at the A ‘toilet facility’ was the sampling unit in highest lockdown level this study. We defined this as where there • actions by some local government was an opening door from the outside agencies to increase soap availability world to a discrete toilet area (ie, a ‘male in the public toilets they maintain (eg, toilet’, a ‘female toilet’, a ‘unisex toilet’ or a Napier City Council15) and to include ‘urinal-only toilet’). In some situations there posters with hygiene messaging in could be multiple toilet facilities that make these facilities. up a ‘toilet complex’. A ‘toilet complex’ was defined as where there was at least one Given the latter point on public toilets, toilet facility or multiple facilities, either in the aim of this study was to identify the the same connected structure or within five provision of water and soap for hand- metres of each other (eg, a male toilet and a washing (along with related health female toilet, or two or more unisex toilets).

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 19 www.nzma.org.nz/journal article

A ‘toilet cubicle’ was defined as the enclosed Requirements for touching (or not touching) area with a toilet bowl and a lockable door. taps/buttons/levers that activated tap water Survey sampling and flushed the toilet were noted. The presence of lids for toilet bowls was docu- We attempted to include all the public toilet mented, given concerns around potential facilities in three contiguous cities (the ‘three virus dispersion when flushing when the lid council area survey’) with a combined popu- is not down (see the introduction). lation of 389,000 (Wellington City, Hutt City and Porirua City). We conducted additional Analysis convenience sampling in cities and rural We compared the results of the three areas, based on author travel plans for other council area survey with the supplementary reasons around New Zealand from 18 July convenience sample. Comparisons were 2020 to 2 January 2021. This sampling was also made with the exact same facilities for only toilets that could be accessed within involved in the previous survey conducted those plans and was not of all toilets in any in 2012/2013 (albeit excluding the 14.7% convenience sample local authority area. We (22/150) that were not readily accessible or expanded the sample to a total of 400 toilets, that had subsequently been closed down). which we estimated would be at least 10% of Statistical analysis used OpenEpi (v 3.01) and the facilities nationwide. But we also aimed Mid-P exact values were calculated (2-tailed). to replicate a previous survey of amenities in public toilets in the lower North Island in National denominator estimation To provide context for our survey, we 2012/2013.16 The public toilets were located by: estimated the total number of public • using online city council maps of toilet complexes in New Zealand using the public toilets, for the three council following steps: area survey23–25 • For city-based local authorities, • using Google Maps to locate a city/ we used as a basis the facilities we town/district, and then searching for surveyed in the three council area ‘public toilets’ in the map survey (n=131 or 3.4 per 10,000 • using the smartphone app version of population). 26 CamperMate • For the other local authorities (LAs), • watching for roadside public toilet we used data from Google Maps in signs when travelling by car. three LAs in the Wairarapa region The sample excluded temporary toilets, (n=13 facilities) and scaled from our portable toilets and public toilets that estimate of the sensitivity of such were not directly open to the outside (ie, data on Google Maps from the three which were inside of other buildings, such council area survey (at 61.8% or as shopping complexes, council-owned 81/131, giving a scaling factor of 1.62). buildings (eg, libraries) and railway or ferry This gave 21 complexes, or 4.3 per buildings, some of which were signed as for 10,000 population. ‘patrons only’). Where toilet facilities were • We then extrapolated the three council closed or were being cleaned, we attempted area survey results to the 12 other city repeat visits where this was convenient. LAs in New Zealand, and we extrapo- Data collection lated the Wairarapa results to the other At each toilet complex, we surveyed all the 51 LAs that were council districts. male and unisex toilet facilities. Data were collected on the availability of water and Results soap for handwashing either in the toilet Survey results for 2020/21 cubicles or toilet facility. ‘Soap’ included Data could not be collected at seven toilet liquid, foam and cake forms, and we aimed complexes in the three council area survey to separately identify alternatives (eg, alcohol that we intended to be complete, due to gel). Toilets with only empty soap containers closure for repairs and/or being padlocked were counted as without soap. We also shut—the latter finding suggestive that these photographed all health-related signage (eg, locked toilet complexes were only opened relating to COVID-19 and handwashing). for specific sporting events.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 20 www.nzma.org.nz/journal article

Our total survey comprised 400 toilet facil- toilet seats, Figure S3-1; broken toilet ities at 242 toilet complexes. The distribution rolls, Figure S3-2; destroyed liquid soap of these 400 facilities spanned major urban dispensers, Figure S3-3). areas (49% of the sample), other urban The majority of toilets had no health-re- areas (40%) and small town and rural areas lated signage (72.8%). Out of those with (11%) (Table 1). Most of the city councils in signage, some form of COVID-19-related the country were included in the sampling health messaging was the most common (62%), but only 26% of the district councils type (19.5%), followed by the NZ COVID were included. We estimated that the total Tracer App QR code used to facilitate contact number of such toilet complexes nationwide tracing (12.3%), generic handwashing was around 1,104 in city councils and 760 signage (1.8%) and then non-smoking in district councils (ie, around 1,864 in the signage (1.3%) (see Supplementary File 2 whole country, or 3.7 per 10,000 population). for examples of these posters). Signs with So, our sample was estimated at 13.0% of COVID-19 health messaging were more the estimated total (242/1,864), which was common in the convenience sample than greater than our target of a 10% sample. in the three council area survey sample Our estimate of the total number is slightly (RR=1.81; 95%CI=1.21 to 2.71; p=0.0040). higher than an OpenStreetMap estimate of There was no signage that promoted toilet 27 1,740 in 2016. lid lowering prior to flushing. Most of the sample of 400 toilet facilities From a qualitative perspective, we noted were from the survey of three contiguous that several COVID-19 signs in tourist areas city councils (n=228), relative to the addi- were in Chinese language (Supplementary tional convenience sample (n=172). The File 2, Figure S2-1), and a few handwashing former group consisted of toilet facilities signs included te reo Māori (Supplementary that were all in the lower North Island, more File 2, Figure S2-3). We also noted that some likely to be in major urban areas and more automatic water and soap dispensers took likely to be unisex (vs male toilets) (Table 1). some time to activate and then dispensed The proportion of all the 400 toilet facil- too little soap or too little water for a satis- ities without water for handwashing was factory handwash (ie, repeat activation was 2.5%, with this being higher in the conve- required). nience sample than the three council area Comparing survey results for survey sample (5.2% vs 0.4%; risk ratio (RR)=11.9; 95%CI=1.53 to 93.3; p=0.0030). 2020/21 with 2012/13 The comparison of the exact same toilet None of the toilets without running water facilities involved in the previous survey in had any other methods of hand sanitation 2012/13 is shown in Table 2. There was no (eg, alcohol gel) that could replace use of improvement in the availability of water soap with water. Absence of soap was 14.8% for handwashing, but soap availability overall, and this absence was also higher in improved significantly from 59% to 86% the convenience sample vs the three council (RR=1.47; 95%CI: 1.25 to 1.72; p<0.0000001). area survey sample (21.5% vs 9.6%; RR=2.23; 95%CI=1.37 to 3.64; p=0.0011). Most soap was dispensed as a liquid or foam (Supple- Discussion mentary File 1, Figure S1-3), but at 6.0% of Main findings toilets it was available in a cake/bar form. A This survey found a deficient provision number of toilets had containers for liquid of both soap (14.8% of toilet facilities soap that were empty (Table 1). with none) and, to a lesser extent, water ‘No-touch’ activation was available for (2.5% with none). These findings suggest handwashing water at 28.0% of facilities, that although there has been a statisti- for toilet bowl flushing at 18.5% and for cally significant improvement in soap urinal flushing at 80.5%. Toilet bowl lids provision (but not water provision) in the were not present for 32.8% (many were eight-year period since the previous survey, designed or built this way) (Supplementary the attempts by some local government File 3, Figure S3-1), and 2.3% of toilets had agencies to increase soap provision at the damage that would impair their function- time of the COVID-19 pandemic15 need to be ality (eg, Supplementary File 3: broken further augmented.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 21 www.nzma.org.nz/journal article

Table 1: Full results for the 2020/2021 survey of public toilet facilities in New Zealand (n=400 surveyed), showing column percentages.

Attempted complete sample of three city Additional convenience LAs (n=228 unless indi- sample (n=172 unless cated otherwise) indicated otherwise) Total (n=400)

Characteristic N % N % N % Comments

Type, location Toilet complexes (ie, some with multiple That is the average toilet complex had 1.7 separate toilet fa- toilet facilities) cilities (400/242) (median=1 facility; range: 1 to 6 facilities). All 131 – 111 – 242 – but one complex had no user charges; and one complex had an office with a supervisor.

Male toilet facility 44 19.3% 65 37.8% 109 27.3% Four of these were urinals only.

Unisex toilet facility 184 80.7% 107 62.2% 291 72.8%

Toilet facilities in the North Island 228 100.0% 126 73.3% 354 88.5% The range was from Auckland to Wellington.

In the 0 0.0% 46 26.7% 46 11.5% The range was from Picton to Christchurch.

In a city council territory 228 100.0% 40 23.3% 268 67.0% 8 out of the 13 city councils in NZ were included (62%).

In a district council territory 14 out of the 53 district councils in New Zealand were includ- 0 0.0% 132 76.7% 132 33.0% ed (26%).

In a major urban areas (100,000+ popula- These were: Auckland City (n=5), Hutt City (n=50), Wellington 185 81.1% 9 5.2% 194 48.5% tion)* City (n=135), Christchurch City (n=4).

In a small, medium or large urban areas 38 16.7% 123 71.5% 161 40.3% (1,000 to 99,999 population)*

In a small town or rural area (<1,000 pop- 5 2.2% 40 23.3% 45 11.3% ulation)*

Water for handwashing Water not available 1 0.4% 9 5.2% 10 2.5%

Automatic, no-touch water delivery 104 45.6% 8 4.7% 112 28.0%

Lever mechanism for tap 5 2.2% 11 6.4% 16 4.0% This is a subset of the above row.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 22 www.nzma.org.nz/journal article

Table 1: Full results for the 2020/2021 survey of public toilet facilities in New Zealand (n=400 surveyed), showing column percentages (continued).

Attempted complete sample of three city Additional convenience LAs (n=228 unless indi- sample (n=172 unless cated otherwise) indicated otherwise) Total (n=400)

Characteristic N % N % N % Comments

Soap Not available 22 9.6% 37 21.5% 59 14.8%

Dispenser not working / empty 12 5.3% 3 1.7% 15 3.8% This is a subset of the above row.

Bar/cake soap only 0 0.0% 24 14.0% 24 6.0%

Toilet bowls Automatic flushing (no need to use a but- Denominator excluded urinal only toilet facilities 58/226 25.7% 14/163 8.6% 72/389 18.5% ton or lever)

Lid missing 78/240 32.5% 68/205 33.2% 146/445 32.8% Denominator includes all separate toilet bowls in toilet cubicles

Urinal flushing Automatic flushing 52/55 94.5% 43/63 68.3% 95/118 80.5% Other urinals required a button/level/cord to be used

Notable facility damage Damage 6 2.6% 3 1.7% 9 2.3% See footnote for details.**

Health-related signage Any COVID-19-related behavioural messaging Excluding QR codes—see below. There was an example of a hand- 33 14.5% 45 26.2% 78 19.5% washing sign involving soap in a toilet with no soap available.

Any COVID-19 QR code signage Inside or on outside wall/door. We included one sign that had 27 11.8% 22 12.8% 49 12.3% fallen onto the floor.

Any handwashing signage 7 3.1% 0 0.0% 7 1.8% That is generic signage, not COVID-19 specific

Any non-smoking signage 2 0.9% 3 1.7% 5 1.3%

No health-related signage 175 76.8% 116 67.4% 291 72.8% None of the four categories above.

* Using the New Zealand Statistical Standard for Geographic Areas 2018. ** Damage included broken seats, a toilet bowl lid with a hole, soap dispenser container remnants, the light not working and a toilet roll holder on the ground.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 23 www.nzma.org.nz/journal article

The higher absence of water (5.2% vs water and soap dispensing; and no-touch 0.4%) and soap provision (21.5% vs 9.6%) toilet flushing and hand-drier activation). in the convenience sample versus the Automation could extend to the toilet bowl three council area survey sample is likely lid being closed before flushing. to reflect the more modern amenities in Ensuring that all toilet bowls have lids the latter and/or a higher-quality main- (missing for 32.8% in this survey, often by tenance schedule. As New Zealand is design) is also desirable, along with messaging highly urbanised, we suspect that the true to close the lids prior to flushing (see the nationwide results would be closer to those introduction for the rationale for lid closing). of the three council area survey than the convenience sample ones. However, as Study strengths and limitations provincial and rural areas have a propor- This is the largest such survey to date in tionately greater number of domestic and New Zealand that we know of, and it was international tourists in comparison to the able to compare a sub-sample of the same resident population, which has a greater toilet facilities after an eight-year period. impact on public toilets, there appears to It was also conducted at a time when there be a further and continued long-term need was heightened need for hygiene, due to the for central government investment in, and COVID-19 pandemic. regulation of, rural and small-town toilets. Nevertheless, the study was limited by The consistent need for a high or very high only attempting to be complete in three prioritisation of toilet infrastructure for contiguous council areas, with the rest being tourists across rural regions and roads indi- convenience sampling. This was owing cates a large backlog of investment.27 to this being an unfunded study with no The relatively low level of health-related budget for travel. The convenience sampling signage, especially COVID-19-related health is likely to have involved surveying facilities messaging (only 19.5%), NZ COVID Tracer that were more accessible to the researchers App QR code signage (only 12.3%) and generic by being on or near main roads. These handwashing signage (only 1.8%), was a may be relatively newer and have a better problematic finding for health promotion maintenance schedule than facilities in and disease control. These deficits are likely small rural towns. This may have resulted to represent both a long-term lack of public in some under-estimation from the conve- signage relating to handwashing and an inad- nience sample of the extent of the problems equate response to the COVID-19 threat. outside urban areas (eg, with water and soap availability). Also of note from a hygiene perspective was the limited extent of no-touch activation The study was also limited to male and of devices (ie, for handwashing water at unisex toilets. Internationally, there are 28.0% and for toilet bowl flushing at 18.5%). indications that the quality of public toilet 28 These should ideally be expanded with facilities for females may be poorer. The the potential long-term goal being to have limitation may have had implications for maximally no-touch amenities (including the surveyed quality of facilities, especially automated door opening and locking via for small town and rural areas, which had a hand waving in front of a sensor; no-touch smaller proportion of unisex toilets.

Table 2: Comparison of the same public toilet facilities in the two surveys (not including those demol- ished or closed at the time of the second survey; all facilities were in the lower North Island).

Previous survey in This survey in 2012/1316 2020/21

Characteristic N % N % P-value Water available for handwashing 123/128 96.1% 123/128 96.1% 1.0

Soap available** 75/128 58.6% 110/128 85.9% <0.0000001

* If a toilet complex had changed in the number of toilet facilities (eg, expanded from two to three unisex facilities), we only compared the exact same number of facilities as in the original survey. ** In liquid, foam or bar/cake forms.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 24 www.nzma.org.nz/journal article

Potential research and policy could use their powers under the Health implications Act 1956 to ensure that unhealthy facilities are improved. A possible way to ensure A fundamental research issue is to better mandated national standards are estab- quantify the risks of infectious disease lished, monitored and maintained is through transmission associated with use of public the central government’s current Three toilets (eg, from any aerosolisation of faeces Waters Reform Programme.31 The under- and from touching contaminated surfaces). lying principles for standards of public toilet This is not only relevant to SARS-CoV-2 but quality and provision have been discussed also other infectious diseases (eg, norovirus in a United Nations report,32 and the World infection and seasonal influenza). Bank has reported on design, operation, Nevertheless, surveys such as this could management and regulatory guidelines.33 be improved by being made fully random at Part of the required role for local author- the national level and collecting additional ities in communicable disease control could data on the facilities to compare them with be regulations on the presence of standard local standards (eg, as per New Zealand health messages and signage in relevant ones,29 albeit from 1999) or more state-of- public facilities, to help address the major the-art Japanese designs.30 We suggest that deficit that we found. New Zealand’s nation- regular (at least five yearly) nationwide al-level legislation requiring smokefree signs surveys funded by the central government (eg, on school grounds) is an example that are needed to better establish the health could be followed.34 and other risks, and as a way to audit compliance with current and future regu- Central government could also boost lations. Research on ways to minimise the funding support for the provision and vandalism of public toilets is also needed, quality of toilets, at least partly from border as in the past this has been reported as a charges collected from international tourists barrier to using soap dispensers in New (a funding system already in place in New Zealand.15 Artwork inside toilets and on Zealand). The extent of the funding needed exterior walls is used in some New Zealand should be seen in the context of the huge toilets (Supplementary File 1, Figure S1-1) costs of COVID-19 and future pandemics, and might be worth expanding if it is found and from other respiratory viruses such to be effective against vandalism. as seasonal influenza and from norovirus infection. Some toilets are still without basic elements of hygiene. Policy goals for local government could be to ensure all facilities Conclusions have water and soap, and to move towards To conclude, despite the serious threat and designs that are maximally no-touch. great costs of the COVID-19 pandemic, and Built-in redundancy (eg, two separate soap although there has been some improvement dispensers) may be desirable to minimise in soap provision in the eight-year period the risk of running out of soap, and cakes of since the previous survey, attempts by some soap should probably be avoided as these local government agencies to increase such are more vulnerable to theft. If fully auto- provisions need to be further augmented. mated taps are not installed, then tap levers There are also other design and mainte- or floor pedals for activating water flow nance deficiencies that would improve could be an alternative. Levers allow users hygiene in public toilets. There is a major to use the back of their hands and can also scope for improving health messaging be more suited than conventional taps for at these sites, and this might be a quick, people with disabilities such as arthritis. low-cost intervention to assist pandemic Central government could set and better control in any future outbreaks. enforce minimum standards for coun- cil-owned public toilet facilities and the Supplementary files extent of their provision. Currently, local • View Supplementary File 1. government authorities largely regulate • View Supplementary File 2. the quality of their own toilet facilities, although theoretically District Health Boards • View Supplementary File 3.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 25 www.nzma.org.nz/journal article

Competing interests: Nil. Author information: Nick Wilson: Department of Public Health, University of Otago, Wellington, New Zealand. George Thomson: Department of Public Health, University of Otago, Wellington, New Zealand. Corresponding author: Nick Wilson, Department of Public Health, University of Otago, Wellington, New Zealand [email protected] URL: www.nzma.org.nz/journal-articles/deficient-handwashing-amenities-in-public-toilets-in-the- time-of-the-covid-19-pandemic-a-multi-regional-survey-open-access

REFERENCES 1. Jefferson T, Del Mar CB, Why coronavirus survives populated Community: Dooley L, Ferroni E, Al-An- longer on impermeable Guangzhou, China, April sary LA, Bawazeer GA, van than porous surfaces. 2020. Clin Infect Dis 2020. Driel ML, Jones MA, Thorn- Phys Fluids 2021;33. 11. Kang M, Wei J, Yuan J, Guo ing S, Beller EM, Clark J, 6. Mecenas P, Bastos R, J, Zhang Y, Hang J, Qu Y, Hoffmann TC, Glasziou Vallinoto ACR, Normando Qian H, Zhuang Y, Chen X, PP, Conly JM. Physical D. Effects of temperature Peng X, Shi T, Wang J, Wu interventions to interrupt and humidity on the J, Song T, He J, Li Y, Zhong or reduce the spread spread of COVID-19: A N. Probable Evidence of of respiratory viruses. systematic review. PLoS Fecal Aerosol Transmission Cochrane Database System One 2020;15:e0238339. of SARS-CoV-2 in a High- Rev 2020;11:CD006207. 7. Leone CM, Tang C, Sharp J, Rise Building. Ann Intern 2. World Health Organization Jiang X, Fraser A. Presence Med 2020;173:974-80. & United Nations Children’s of human noroviruses 12. Yu IT, Li Y, Wong TW, Tam Fund (UNICEF) [Internet]. on bathroom surfaces: a W, Chan AT, Lee JH, Leung Water, sanitation, hygiene, review of the literature. DY, Ho T. Evidence of and waste management Int J Environ Health airborne transmission of for SARS-CoV-2, the virus Res 2016;26:420-32. the severe acute respirato- that causes COVID-19: 8. Hu X, Xing Y, Ni W, Zhang F, ry syndrome virus. N Engl interim guidance, 29 Lu S, Wang Z, Gao R, Jiang J Med 2004;350:1731-9. July 2020. World Health F. Environmental contam- 13. Baker M, Wilson N, Blakely Organization. Available ination by SARS-CoV-2 of T. Elimination may be the from: https://apps.who.int/ an imported case during optimal response strategy iris/handle/10665/333560 incubation period. Sci Total for covid-19 and other 3. Hirose R, Ikegaya H, Naito Environ 2020;742:140620. emerging pandemic diseas- Y, Watanabe N, Yoshida T, 9. Ong SWX, Tan YK, Chia es. BMJ 2020;371:m4907. Bandou R, Daidoji T, Itoh PY, Lee TH, Ng OT, Wong doi: 10.1136/bmj.m4907. Y, Nakaya T. Survival of MSY, Marimuthu K. Air, 14. Wilson N, Grout L, Kvalsvig SARS-CoV-2 and influenza Surface Environmental, A, Baker M. Time to Stop virus on the human skin: and Personal Protective Dodging Bullets? NZ’s Eight Importance of hand Equipment Contamination Recent Border Control Fail- hygiene in COVID-19. Clin by Severe Acute Respirato- ures. Public Health Expert Infect Disease 2020:Online ry Syndrome Coronavirus (Blog) 2020;(16 November). Oct 3, 10.1093/cid/ciaa517. 2 (SARS-CoV-2) From a Available from: https:// 4. Lewis D. COVID-19 rarely Symptomatic Patient. JAMA blogs.otago.ac.nz/pubheal- spreads through surfac- 2020;323:1610-12. thexpert/2020/11/16/ es. So why are we still 10. Yuan J, Chen Z, Gong C, time-to-stop-dodging- deep cleaning? Nature Liu H, Li B, Li K, Chen X, bullets-nzs-eight-recent- 2021;(E-publication Xu C, Jing Q, Liu G, Qin P, border-control-failures/ 29 January). Available Liu Y, Zhong Y, Huang L, 15. Radio New Zealand from: https://www. Zhu BP, Yang Z. Sewage [Internet]. Coronavirus: nature.com/articles/ as a Possible Transmis- Napier to reinstall public d41586-021-00251-4. sion Vehicle During a toilet soap. Radio New 5. Chatterjee S, Murallidharan Coronavirus Disease 2019 Zealand 2020;(6 March). J, Agrawal A, Bhardwaja R. Outbreak in a Densely https://www.rnz.co.nz/

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 26 www.nzma.org.nz/journal article

news/national/411152/ hygiene facilities in a toilets for sanitation coronavirus-napier-to-re- developed country. J Public access in urban public install-public-toilet-soap Health (Oxf) 2012;34:483-8. spaces: A systematic 16. Wilson N, Thomson G. 22. Garbutt C, Simmons G, review Utilities Policy Neglecting the basics? Patrick D, Miller T. The 2021;70:June 2021, 101186. Survey of water and soap public hand hygiene 29. Standards New Zealand availability in council-op- practices of New Zealand- [Internet]. NZS 4241:1999, erated public toilets in ers: a national survey. N Public toilets. Published: New Zealand. N Z Med J Z Med J 2007;120:U2810. 30/03/1999. Available from: 2013;126(1376):110-4. 23. Hutt City [Internet]. Public https://shop.standards.govt. 17. Kuprienko D. More toilets in Lower Hutt (map). nz/catalog/4241%3A199 toilets and changing (Accessed 1 November 9%28NZS%29/view rooms proposed for 2020). Available from: 30. BBC [Internet]. The Queenstown Lakes https://huttcity.maps.arcgis. Japanese project rethink- district. Southland Times com/apps/opsdashboard/ ing society through its 2018;(15 March). Available index.html#/ab818e582df toilets. BBC News 2020;(7 from: https://www.stuff. f48e3be6ddd392486f1d0 December). https:// co.nz/southland-times/ 24. Porirua City [Internet]. www.bbc.com/reel/ news/102260583/ Find a public toilet video/p090jqwf/the-japa- more-toilets-and-chang- (map). (Accessed 1 nese-project-rethinking-so- ing-rooms-planned-for- November 2020). ciety-through-its-toilets queenstown-lakes-district Available from: https:// 31. Department of Internal 18. Cropp A. Councils estimate poriruacity.govt.nz/contact/ Affairs. Central/Local $1.4b bill for urgent tour- neighbourhood-issues/ Government Three Waters ism infrastructure. Stuff public-toilets/ Reform Programme. 2017;(14 March). Available 25. Wellington City Council Wellington: Department from: https://www.stuff. [Internet]. Public toilets in of Internal Affairs, 2021. co.nz/business/90404870/ Wellington: Public toilet 32. Heller L, UN Human Rights councils-estimate--14- locations. (Accessed 1 Council. Human rights b-bill-for-urgent-tour- November 2020). Available to water and sanitation ism-infrastructure from: https://wellington. in spheres of life beyond 19. Wilson N, Thomson G. govt.nz/community-sup- the household with an Additional evidence port-and-resources/ emphasis on public spaces for concern about the resources-and-facilities/ A/HRC/42/47. New York: quality of public toilets public-toilets United Nations, 2019. in New Zealand. N Z 26. CamperMate [Inter- 33. Cardone R, Schrecongost Med J 2013;126:110. net]. CamperMate. A, Gilsdorf R. Shared and 20. Norris P, Choi T, Lee R, Lu Available from: https:// Public Toilets: Champi- A, Png C, Yang S, Priest P. www.campermate.co.nz/ oning Delivery Models Adequacy of handwash- 27. Deloitte. National Tour- That Work. Washington: ing facilities in public ism Infrastructure and World Bank, 2018. toilets in Dunedin, New Investment Assessment 34. Thomson G, Wilson N. Zealand. Aust N Z J Public – Appendices document. Smokefree signage at Health 2012;36:194-5. Wellington: Tourism children’s playgrounds: 21. Reeves LM, Priest PC, Industry Aotearoa, 2017. Field observations Poore MR. School toilets: 28. Deister Moreira F, Rezende and comparison with facilitating hand hygiene? S, Passos F. On-street Google Street View. Tob A review of primary school Induc Dis 2017;15:37.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 27 www.nzma.org.nz/journal article

Māori and Pacific people in New Zealand have a higher risk of hospitalisation for COVID-19 Nicholas Steyn, Rachelle N Binny, Kate Hannah, Shaun C Hendy, Alex James, Audrey Lustig, Kannan Ridings, Michael J Plank, Andrew Sporle

ABSTRACT AIMS: We aim to quantify differences in clinical outcomes from COVID-19 infection in Aotearoa New Zealand by ethnicity and with a focus on risk of hospitalisation. METHODS: We used data on age, ethnicity, deprivation index, pre-existing health conditions and clinical outcomes on 1,829 COVID-19 cases reported in New Zealand. We used a logistic regression model to calculate odds ratios for the risk of hospitalisation by ethnicity. We also considered length of hospital stay and risk of fatality. RESULTS: After controlling for age and pre-existing conditions, we found that Māori have 2.50 times greater odds of hospitalisation (95% CI 1.39–4.51) than non-Māori non-Pacific people. Pacific people have three times greater odds (95% CI 1.75–5.33). CONCLUSIONS: Structural inequities and systemic racism in the healthcare system mean that Māori and Pacific communities face a much greater health burden from COVID-19. Older people and those with pre-existing health conditions are also at greater risk. This should inform future policy decisions including prioritising groups for vaccination.

p to 25 September 2020, New Zealand resulted from workplace, community, public had reported 1,829 confirmed and transport and household transmission, probable cases of COVID-19, a disease rather than being associated with inter- U 1 caused by a novel coronavirus originating in national travel. The August cluster had Wuhan, China. The majority of these cases a higher proportion of cases in under-20- were associated with one of two outbreaks year-olds and a lower proportion of cases in of sustained community transmission: the over-60-year-olds than the earlier outbreak first in March/April 2020 and the second in (Figure 1). It also contained a much higher August/September 2020. Up to 22 May 2020, proportion of cases among the Pacific and there were 1,504 confirmed and probable Māori populations than the first outbreak.1, cases, of which 573 had a recent history of 2 Multigenerational living is proportionately international travel. Between 22 May and greater in Pacific peoples as a population, 11 August, there were 65 reported cases, all but the lack of high-quality suitable housing of which were in detected in international means that their homes are often over- arrivals and contained in government-man- crowded.3 Pacific people also experience aged isolation facilities. Between 11 August poorer access to healthcare4 and are at a and 25 September, 260 cases were reported, greater risk of clinically severe outcomes with the majority linked to a large cluster in from COVID-19 infection.5 Auckland. Historically, Māori and Pacific commu- The August/September cluster differed nities both in New Zealand and in the Pacific substantially from the initial outbreak in have had worse experiences of pandemics. March/April 2020. The vast majority of cases During the 2009 H1N1 influenza pandemic,

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 28 www.nzma.org.nz/journal article

the rate of infection for Māori was twice data are imperfect, and the number of that for Pākehā and more severe.6 Our cases is relatively small, but they are none- recent research estimated similar inequities theless the best data currently available would occur in the infection fatality rate for to understand differences in risk from COVID-19.5 infection of COVID-19 between ethnicities New Zealand’s effective public health in New Zealand. The results are important response to the pandemic limited the for future policy decisions and pandemic number of COVID-19 fatalities during 2020 planning: for example, identification of to 25, 7 which corresponds to a fatality rate priority groups for vaccination against of five deaths per million people. This means COVID-19. that there are insufficient empirical data at present to reliably estimate differences Methods in the infection fatality rate by ethnicity. We developed three separate risk models Here, we aim to determine whether there to quantify the risk of hospitalisation, length are significant differences by ethnicity in of hospital stay and fatality risk. Each model the risk of clinically severe outcomes from used the same methodology and set of COVID-19, measured by the hospitalisation predictor variables. rate and length of hospital stay. We take a data-driven approach by using information Data Case data was obtained from the EpiSurv that is routinely collected for all cases of database on all 1,829 confirmed and probable COVID-19 in New Zealand. The available

Figure 1: Age–ethnicity structure of New Zealand’s two major outbreaks of COVID-19 using prioritised ethnicity. The plots on the right give the number of cases per 1,000 people in that age–ethnicity group- ing. Population data from Census 2018.2

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 29 www.nzma.org.nz/journal article

cases of COVID-19 reported in New Zealand groups: Māori, Pacific, Asian, NZ European/ up to 25 September 2020. EpiSurv is New Other. Due to the limitations of analysing Zealand’s national notifiable disease surveil- small numbers of cases and to avoid over- lance database, operated by Environmental fitting, individuals whose ethnicity was Science and Research (ESR) on behalf of the recorded as Middle Eastern/Latin American/ Ministry of Health.8 EpiSurv collates noti- African (n=49 cases, 1 hospitalisation) or fiable disease information, including case Other (n=5 cases, 1 hospitalisation) were demographics, clinical features and risk assigned to the NZ European/Other ethnicity factors, in real time. The data for COVID-19 group. Individuals for whom total ethnicity cases include hospitalisation status and dates, data was missing (n=29 cases, 1 hospi- clinical outcome (eg, recovered, death), age, talisation) were assigned to the ethnicity sex, presence/absence of several underlying recorded in the ‘prioritised ethnicity’ field health conditions (see next paragraph), Stats in EpiSurv. Of the 29 cases with missing NZ meshblock of current home address and total ethnicity data, prioritised ethnicity self-reported ethnicity (Table 1). Ethnicity was recorded as Māori for one case, Pacific information in EpiSurv is collected on the for two cases, Asian for 11 cases and NZ standard COVID-19 case report form,8 where European/Other for 15 cases. Of all 1,829 it is described as ‘core surveillance data’. cases, 1,719 (94%) had a single ethnicity 9 The responses are then prioritised to a recorded, 102 (5.6%) had two ethnicities single response using the Ministry of Health’s recorded and eight (0.4%) had three ethnic- Ethnicity Data Protocols.10 The ethnicity infor- ities recorded. A breakdown of the number mation in the Ministry of Health sourced data of cases in the data set by ethnicities is includes multiple ethnicity fields sourced by shown in Appendix Table 1. linking EpiSurv data to the National Health Of the 120 hospitalised cases, only 102 had Index (NHI) data collection. listed discharge dates, which were required The data on underlying health conditions for analysis on the length of hospital stay. were simplified into a binary variable indi- Five of the 18 cases without discharge dates cating whether the individual had at least resulted in death, so their discharge dates one of the following conditions: chronic lung were set to the date of death. The remaining disease, cardiovascular disease, diabetes, 13 cases (two who had not recovered by 25 immunodeficiency, asthma or malignancy. September 2020 and 11 with no discharge These conditions were chosen because they date recorded) were excluded from the are all recorded in the EpiSurv dataset8 and length of stay analysis. One additional case are known to be associated with increased was excluded because the discharge date risk of COVID-19 hospitalisation.11 We recorded was prior to the hospitalisation did not consider the effects of multiple date. This resulted in a sample of 106 cases underlying health conditions, due to the with a recorded length of stay in hospital limitations of analysing such small numbers (Figure 2). Of the 14 excluded cases, seven (see section Discussion for the associated (50%) were Pacific people, despite Pacific limitations). Of the 1,829 cases, 269 cases people only making up 18% of hospital- (14.7%) had one of the above conditions isations. This reduced the sample size for recorded; 55 cases (3.0%) had two conditions Pacific people and likely biased the results. recorded; four cases (0.2%) had three condi- Model selection tions recorded; and two cases (0.1%) had For each of the three models, we carried four conditions recorded. out a simple analysis to determine which The meshblock numbers of residential predictor variables to include in the model. addresses were used to allocate a measure We used a logistic regression to determine of geographic and socioeconomic depri- which of ethnicity, underlying health vation, based on the New Zealand Index conditions, sex, age and deprivation index of Deprivation (NZDep18).12 This was not should be included. We used Akaike infor- available for 34 cases, so any models that mation criterion (AIC) and the area under include deprivation index had a sample size the receiver operating characteristic curve of 1,795 cases and 114 hospitalisations. (AUC) for model selection. Using AIC is a Total ethnicity data were used to assign standard, likelihood-based procedure for individuals into one or more of the following model selection that quantifies how parsi-

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 30 www.nzma.org.nz/journal article

Table 1: Summary of case data. Deprivation index was used in its raw index form in the model but has been presented as quintiles for ease of interpretation, with the 1st quintile representing those that re- side in a meshblock with the lowest socioeconomic deprivation and the 5th quintile the highest depriva- tion. Age is also presented in discretised brackets. The use of total ethnicity data means sums over these rows will be greater than the totals where some cases are recorded as having multiple ethnicities.

Hospitalised Mean length Died Total of stay (days) Overall 120 (6.6%) 8.2 25 (1.4%) 1,829

Ethnicity Māori 18 (10.1%) 9.4 4 (2.2%) 178

Pacific 21 (10.0%) 11.6 1 (0.5%) 210

Asian 15 (5.0%) 4.1 0 (0.0%) 300

NZ European/Other 69 (5.5%) 7.8 21 (1.7%) 1,259

Health Status Underlying condition 47 (14.2%) 9.0 13 (3.9%) 330

No underlying condition 73 (4.9%) 7.6 12 (0.8%) 1,499

Sex Male 57 (6.9%) 9.2 14 (1.7%) 823

Female 63 (6.3%) 7.3 11 (1.1%) 1,006

Age 0–19 4 (1.7%) 2.8 0 (0.0%) 239

20–39 23 (3.1%) 2.6 0 (0.0%) 732

40–59 43 (8.0%) 8.1 2 (0.4%) 538

60–79 38 (13.6%) 8.6 10 (3.6%) 279

80+ 12 (29.3%) 19.2 13 (31.7%) 41

Deprivation 1st quintile (least) 17 (4.0%) 10 3 (0.7%) 422

2nd quintile 35 (8.3%) 6.6 5 (1.2%) 421

3rd quintile 11 (3.7%) 9.2 2 (0.7%) 300

4th quintile 33 (8.8%) 6.5 1 (0.3%) 374

5th quintile (most) 18 (6.5%) 10.4 14 (5.0%) 278

Missing 6 (17.6%) 13.5 0 (0.0%) 34

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 31 www.nzma.org.nz/journal article

moniously the model describes the data and consistently identified as a significant penalises models with too many variables.13 predictor variable in all three models. Using AUC measures how accurately the model priority ethnicity neglects important infor- predicts the outcome of interest (in this case, mation on individuals who were in multiple hospitalisation) for cases in the dataset.14 ethnicity groups.15 For example, there The complete model was: were 19 individuals who were recorded logit( P(hospitalised) ) ~ age + ethnicity + as Māori and Pacific, none of whom were sex + has underlying conditions + dep index hospitalised. In the standard prioritisation routine, these individuals were classified Ethnicity was treated as a categorical as Māori and did not, therefore, contribute variable with individuals belonging to one to model estimates for Pacific people. This of Māori, Pacific, Asian or NZ European/ undercounted Pacific cases potentially Other. In the case of multiple recorded created age-related biases in the results for ethnicities the standard Ministry of Health Pacific people, as younger Pacific people prioritisation was used10 for the model are more likely to report multiple ethnic- selection phase, as there was insufficient ities.16 To account for this, we reran each data to consider all ethnicity combinations model using different ethnicity prioriti- (see section Estimating the effect of ethnicity sation orderings (Table 2). Odds ratios and for estimation of effect sizes using multiple confidence intervals on the odds ratios were ethnicity data). The NZ European/Other obtained by exponentiating the coefficient group is used as the baseline group so that estimates and confidence intervals on the resulting odds ratios are interpreted as coefficient estimates for each risk factor. ‘difference in risk relative to NZ European/ Other’. Because AIC requires all models to Length of stay and risk of fatality have the same sample size, the 34 records In addition to the risk of hospitalisation, missing a deprivation index were removed we used a linear model to consider the effect during model selection. of these variables on length of hospital stay: Estimating the effect of ethnicity length of hospital stay ~ age + ethnicity During the model selection phase, + sex + has underlying conditions + dep ethnicity (using prioritised ethnicity) was index

Figure 2: Sankey diagram of case data that was included/excluded from the length of hospital stay analysis. Those with valid discharge dates or death dates (n=106) were included in the analysis. The remaining cases were excluded (n=14)—of these, one was Māori, seven were Pacific, one was Asian, five were NZ European/Other and zero had multiple ethnicities recorded. Data are for cases reported up to 25 September 2020.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 32 www.nzma.org.nz/journal article

Finally, despite very limited data, we also 8. Cases with a recent overseas travel considered fatality risk under the same history (n=707 cases) were excluded framework: from the dataset. logit{P(death)} ~ age + ethnicity + sex + 9. Cases with missing length of hospital has underlying conditions + dep index stay data were assumed to have Because there were no fatalities in Asian a length of stay of zero days (the people, in this final model we combined shortest stay in the dataset). the Asian and NZ European/Other ethnicity 10. Cases with missing length of hospital groups. For both these models, we used stay data were assumed to have length the same methodology as for the risk of of stay of 52 days (the longest stay in hospitalisation model (ie, we used standard the dataset). ethnicity prioritisation to identify signif- icant predictor variables then re-analysed Results the contribution of these predictor vari- ables under different ethnicity prioritisation Risk of hospitalisation For risk of hospitalisation, the model orderings). containing age, ethnicity and the presence Sensitivity analysis of underlying health conditions as predictor To check how robust our conclusions variables gave the most parsimonious fit were with respect to our assumptions about (lowest AIC). This model also has the same ethnicity data and other potential sources predictive power (similar AUC) as more of bias, we performed a sensitivity analysis complex models (Table 3). Including inter- by re-running the preferred models for risk action terms did not improve the model fit of hospitalisation and length of hospital stay as measured by AIC. Age was always the under each of the following assumptions: strongest predictor of hospitalisation and 1. Cases with primary ethnicity recorded was included in all models. After age had as Middle Eastern/Latin American/ been accounted for, the best two-variable African (n=49 cases, 1 hospitalisation) model also included ethnicity. or Other (n=5 cases, 1 hospitalisation) Coefficient estimates associated with sex were excluded from the dataset. were always close to zero and had consis- 2. Cases with missing total ethnicity data tently large p-values, indicating that sex (n=29 cases, 1 hospitalisation) were was not a strong predictor of hospital- excluded from the dataset. isation in New Zealand’s COVID-19 cases. 3. Cases satisfying either one or two This is contrary to some international above were excluded from the evidence that suggests men suffer worse 11 dataset. clinical outcomes on average. Deprivation index was only statistically significant 4. Cases with missing total ethnicity data when considered alongside age, not were assumed to be Māori. ethnicity. Deprivation index and ethnicity 5. Cases with missing total ethnicity data were slightly correlated, so this suggests the were assumed to be Pacific. effect of deprivation index was partially 6. Cases with missing total ethnicity data captured by ethnicity. Different age groups were assumed to be Asian. were represented differently across 7. Cases with missing total ethnicity data different indexes of deprivation, suggesting were assumed to be NZ European/ that a model containing a deprivation Other. index–age interaction term may be suitable.

Table 2: Ethnicity prioritisation ordering depending on the ethnicity effect being estimated.

Ethnicity effect being estimated Prioritisation ordering Māori Māori, Pacific, Asian, NZ European/Other

Pacific Pacific, Māori, Asian, NZ European/Other

Asian Asian, Māori, Pacific, NZ European/Other

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 33 www.nzma.org.nz/journal article

This was tested and the resulting coeffi- on average, there is a 20.7-year age gap cients were not statistically significant. between Māori and NZ European/Other, and Māori and Pacific people are known to a 25.2-year age gap between Pacific and NZ have higher rates of multi-morbidity and European/Other, at the same level of risk. underdiagnosis of comorbid conditions.17,18,19 These estimates should be used with caution This suggests that including a term in the because they assume that age has the model for the interaction between ethnicity same proportional effect in each ethnicity and presence of underlying health condi- (see section Discussion for limitations and tions could be important. However, this sources of bias). term was found to be not statistically The results of the sensitivity analysis significant. (see Appendix Table 2) showed that the Table 4 and Figure 3 show the results main conclusions were robust to different for the preferred model for risk of hospi- assumptions. The magnitude of the odds talisation. Age was associated with a 4.5% ratios for Māori and Pacific people could increase in odds of hospitalisation per be slightly smaller than those in Figure 3 additional year. The presence of at least one under different assumptions about missing underlying health condition increased the ethnicity data or ethnicity groupings. For odds of hospitalisation by 1.74 times (95% scenarios 1–7 described in Methods, the CI 1.14–2.65, p=0.01). After controlling for odds ratio for Māori was always statisti- age and underlying conditions, Māori and cally significant and varied between 2.15 Pacific people had substantially higher odds (95% CI 1.20–3.86), if cases with missing of being hospitalised for COVID-19 than total ethnicity data were assumed to be other ethnicities: Māori 2.5 times higher Māori, and 2.50 (95% CI 1.39–4.51), under odds (95% CI 1.39–4.51, p=0.002) and Pacific the default model. The odds ratio for Pacific people 3.06 times higher odds (95% CI people was always statistically significant 1.75–5.33, p=8×10-5). Asian people were also and varied between 2.78 (95 CI 1.61–4.80) at higher risk, with 1.35 times higher odds, and 3.06 (95% CI 1.75–5.33) under scenarios although this result was not statistically 1–7. If the EpiSurv ethnicity field (which is significant (95% CI 0.74–2.48, p=0.33). more up to date but only allows the priority ordering with Māori as priority ethnicity) The odds ratios for different ethnicities was used instead of Ministry of Health total shown in Figure 3 represent the increase in ethnicity data, the odds ratio for Māori was risk after controlling for underlying health 2.68 (95% CI 1.48–4.83), which is larger than conditions, which are present in higher in Figure 3. Excluding cases with a recent rates in Māori and Pacific people.17 In the international travel history (scenario 8) 1,829 cases in the data, there was only a very increased the odds ratio for Māori and for small correlation between having under- Pacific people to 2.51 (95% CI 1.28–4.93) and lying conditions recorded and either Māori 3.20 (95% CI 1.73–5.94) respectively. The ethnicity (Pearson’s r-squared r2=0.07)or odds ratio for Asian people was not statisti- Pacific ethnicity (r2=0.02), so the results were cally significant under any of the scenarios not affected by multi-collinearity in these tested. variables. The model can be used to estimate the Length of hospital stay probability of hospitalisation following For length of hospital stay, the model infection with COVID-19 for an individual of containing only age and ethnicity as a given age and ethnicity and the presence/ predictor variables gave the most parsi- absence of underlying health conditions monious fit (lowest AIC). Age was a more (see Figure 4 and see Appendix Figure 1 for important factor than in the probability of confidence intervals). It can also be used to hospitalisation model, with an additional estimate the age at which Māori or Pacific year of age predicting an additional 0.22 -6 cases had the same risk of hospitalisation as days (95% CI 0.14–0.31 days, p=2×10 ) in those at a specific reference age in the NZ hospital on average. When used as the European/Other group, after we controlled priority ethnicity, Māori are expected for the presence or absence of underlying to spend 4.9 days (95% CI 0.02–9.7 days, health conditions (Table 5). This shows that, p=0.052) longer in hospital than NZ European/Other, and Pacific people are

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 34 www.nzma.org.nz/journal article

Table 3: AIC and AUC values for the eight models for risk of hospitalisation with lowest AIC, as well as the age-only and ethnicity-only models. Smaller values of AIC indicate a more parsimonious model fit; larger values of AUC indicate better predictive power.

Model AIC AUC Model AIC AUC Age + Eth + HasCond 755 0.762 Age + Eth + Sex 761 0.758

Age + Eth + HasCond 757 0.763 Age + Eth + Dep 761 0.757 + Sex

Age + Eth + HasCond 757 0.762 Age + Eth + Sex + Dep 763 0.758 + Dep

Age + Eth + HasCond + 758 0.763 Age 779 0.728 Sex + Dep

Age + Eth 759 0.757 Eth 844 0.577

Table 4: Results of the preferred model (age, ethnicity, underlying conditions) for risk of hospitalisation under each ethnicity prioritisation ordering. Coefficient estimates for ethnicity that is not the priority (grey text) should be treated with caution. These models use the data from all cases (1,829 individuals).

Coefficient Estimate (p-value)

Priority AIC Intercept Age HasCond Māori Pacific Asian ethnicity Māori 789 -5.205 0.044 0.553 0.918 1.185 0.327 (5 x 10-53) (4 x 10-15) (0.01) (0.002) (3 x 10-5) (0.292)

Pacific 789 -5.204 0.044 0.548 0.985 1.118 0.327 (6 x 10-53) (4 x 10-15) (0.01) (0.01) (8 x 10-5) (0.291)

Asian 788 -5.198 0.044 0.553 0.918 1.227 0.306 (5 x 10-53) (4 x 10-15) (0.01) (0.02) (2 x 10-5) (0.332)

Figure 3: Odds ratios and 95% confidence intervals for the considered risk factors. The odds ratio for underlying conditions was taken from the model with Māori as the priority ethnicity, but these results change very little under different prioritisations. In the same model, the odds ratio for an additional year of age was 1.045 (1.034, 1.057). Analysis based on cases reported up to 25 September 2020.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 35 www.nzma.org.nz/journal article

Figure 4: Estimated probability of hospitalisation by age and ethnicity, with and without underlying health conditions. Analysis based on cases reported up to 25 September 2020.

Table 5: Age differences between ethnicities at the same level of risk of hospitalisation. Each row shows a reference age for NZ European/Other and the corresponding age [95% CI] at which Māori and Pacific people have the same predicted risk of hospitalisation as NZ European/Other. Note that, after we control for underlying health conditions, the average age difference between NZ European/Other and Māori at the same level of risk is always 20.7 years and the average age difference between NZ European/Other and Pacific people at the same level of risk is always 25.2 years, but the size of the confidence intervals varies slightly with age.

No underlying health conditions At least one underlying health condition

NZ Euro/ Māori age with Pacific age NZ Euro/ Māori age with Pacific age Other age same risk with same risk Other age same risk with same risk 60 39.3 [26.2, 51.8] 34.8 [22.7, 45.9] 60 39.3 [24.5, 52.0] 34.8 [20.3, 47.1]

65 44.3 [31.6, 57.0] 39.8 [28.2, 51.2] 65 44.3 [30.1, 56.9] 39.8 [26.0, 52.0]

70 49.3 [36.8, 62.4] 44.8 [33.5, 56.5] 70 49.3 [35.7, 62.0] 44.8 [31.6, 57.1]

75 54.3 [42.0, 68.0] 49.8 [38.6, 62.0] 75 54.3 [41.1, 67.3] 49.8 [37.1, 62.2]

80 59.3 [46.9, 73.7] 54.8 [43.6, 67.7] 80 59.3 [46.4, 72.6] 54.8 [42.4, 67.6]

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 36 www.nzma.org.nz/journal article

expected to spend 5.2 days (95% CI 0.08–10.2 rates and age, with one paper estimating days, p=0.049) longer in hospital than NZ an increase in probability of death of 12.9% European/Other. Length of hospital stay for per year of age.20 This is comparable to our Asian people was not significantly different results (although changes in the infection from NZ European/Other. fatality rate are not identical to changes in The sensitivity analyses (Appendix Table odds, they are close at small probabilities). 3) showed that the difference in length The number of fatalities was too small of hospital stay for Māori was sometimes to draw any concrete conclusions on the marginally statistically significant at the relationship between risk of fatality and p=0.05 level and sometimes not statisti- ethnicity. There were no models where cally significant, with average length of ethnicity was a consistently statisti- stay varying between 4.4 days and 6.1 days cally significant predictor of fatality risk. longer than NZ European. The difference for However, this is most likely due to inade- Pacific people was statistically significant quate statistical power of analysing such under most scenarios. Under scenarios 1–8 small numbers. Furthermore, the majority described in Methods, the average length of of fatalities are linked to aged care facilities, stay for Pacific people varied between 5.0 and so are not representative of the type days and 5.7 days longer than NZ European. of fatalities that would occur if COVID-19 The average length of stay for Pacific people were to become more widespread in the was sensitive to assumptions about cases community. with missing or invalid length of stay data (scenarios 9–10) because Pacific people Discussion were disproportionately represented in Structural bias and systemic racism are this cohort. If cases with missing data widespread in healthcare systems and were assumed to have length of stay zero are basic determinants of ethnic health days (the smallest value in the data), the inequities in New Zealand and interna- difference in length of stay for Pacific people tionally.4,21 New Zealand’s experience with was not statistically significant. If cases the COVID-19 epidemic indicates that Māori with missing data were assumed to have and Pacific people are at much greater risk length of stay 52 days (the largest value in of hospitalisation following infection with the data), the difference in length of stay for COVID-19. It is widely understood from Pacific people was highly significant, with overseas experience that the risk of hospi- an average stay 14.9 days longer than NZ talisation for COVID-19 increases rapidly European. These two scenarios are opposite with age. However, the effects of ethnicity extremes and reality is likely to lie some- in New Zealand are not as well understood. where between them. Our results show that an 80-year-old patient Risk of fatality with COVID-19 in the NZ European/Other For risk of fatality, the model containing group without reported comorbidities has only age and deprivation index as predictor the same predicted risk of hospitalisation as variables gave the most parsimonious fit a 59.3-year-old (95% CI 46.9–73.7 years old) (lowest AIC). In this model, an additional patient in the Māori group without reported year of age increased the odds of fatality by comorbidities. Similarly, an 80-year-old 15.9% (95% CI 11.5%–20.4%, p=3×10-14). A patient in the NZ European/Other group unit increase in deprivation index was asso- without reported comorbidities has the ciated with a 0.80% (95% CI 0.33%–1.27%, same predicted risk of hospitalisation as a p=0.001) increase in the odds of fatality. The 54.7-year-old (95% CI 43.6–67.7 years old) difference in deprivation score between the patient in the Pacific group without reported 1st and 4th quintiles in the dataset was 146. comorbidities. Similar differences are seen This means that the model predicts that an across all ages and for cases with at least individual at the 80th percentile of depri- one reported comorbidity (Table 6). These vation has 3.19 (95% CI 1.62–6.31) times differences in age-specific risk are broadly the odds of fatality as someone at the 20th consistent with earlier estimates of ineq- percentile in this dataset. uities in the COVID-19 infection fatality rate.5 International evidence suggests a linear Our analysis suggested that average length relationship between log infection fatality of hospital stay could be longer for Māori

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 37 www.nzma.org.nz/journal article

and Pacific people than for NZ European/ earlier age.17,18,19 These factors have not been Other, but the data was insufficient to draw accounted for in the model and are likely strong conclusions. to exacerbate the risk of clinically severe We have only considered the risk of outcomes from COVID-19. It is possible that being hospitalised given an individual was some of the observed risk of hospitalisation infected with COVID-19. The likelihood of for Māori could be explained by unre- hospitalisation will depend on prevailing ported, undiagnosed or multiple comorbid admission policies in each hospital. These conditions, in which case the odds ratios policies may vary across the country and for ethnicity that we reported could be over time, but we ignored this variation overestimates. in this analysis. The overall risk of being Testing rates and contact tracing were hospitalised also depends on the like- much higher in the second outbreak in lihood of infection, which is specifically August/September 2020 than in the first not included in our calculations. COVID-19 outbreak in March/April 2020, meaning can spread quickly in communities with that more mild cases of COVID-19 would higher levels of workplace, community have been identified in the second outbreak or whānau interaction, crowded housing, compared with the first. As this second insecure employment and decreased access outbreak disproportionately affected Pacific to healthcare or COVID-19 testing. These are and Māori people, the model may underes- frequently the same individuals, groups or timate their relative risk of hospitalisation. communities that are at higher risk of hospi- Our model is fitted to data from a period talisation and fatality if infected, meaning in which the prevalence of COVID-19 was there is additional potential burden of the low and healthcare services had adequate epidemic on these people. capacity. Systemic racism within the When fitting each model, we assigned healthcare system could further exacerbate each individual to only one ethnicity, as the inequities in outcomes if COVID-19 prev- small number of cases precluded investi- alence were to increase and if healthcare 22,17,23 gation of all combinations of ethnic identity. capacity were overstretched. Depri- This means that our results cannot be used vation index was assigned according to multiplicatively to estimate the risk of the meshblock of each individual’s home hospitalisation for an individual belonging address. This may be a good proxy for to multiple ethnicities. Other effects are general current socioeconomic deprivation multiplicative in the odds. For example, an on average, but the small number of cases in individual with reported comorbid condi- the dataset may not be sufficient for this to tions has odds of hospitalisation that are apply. Geographic measures of deprivation 74% greater than another individual of the are widely used and useful because they same age and ethnicity without reported simply require an address to provide the comorbid conditions. information. However, such information may not represent the socioeconomic expe- We have presented the results of a simple riences of an individual over their lifetime.24 analysis that ignores several potential sources of bias and additional inequities The level of ethnic group classification (Table 6). For example, the recording and used here involved broad categories that the analysis of the effect of comorbid condi- define populations with diverse experi- tions are crude. Different health conditions ences, cultures, nationalities, exposure to have significantly different effects, and racism and immigration histories. The level the presence of multiple health conditions of ethnicity data available and the absence may increase risk further. We did not have of migration information (other than recent a sufficient number of cases to estimate overseas travel) precluded a more nuanced the effect of individual health conditions understanding of the hospitalisation risks or combinations of conditions. Māori within these broad ethnic categories. and Pacific people have lower life expec- Understanding the potential impact of the tancy, higher rates of multi-morbidity and epidemic and informing the delivery of the respiratory illness, higher rates of under-re- vaccination programme requires complete porting of comorbid conditions and typically and detailed ethnicity information to be experience adverse health outcomes at an included in the routinely available data.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 38 www.nzma.org.nz/journal article

Table 6: Sources of potential bias and their likely direction of effect on model predictions for the hospi- talisation odds ratio for Māori and Pacific people.↑ and ↓ indicate that the source of bias is likely to mean that the model underestimates or overestimates respectively the odds ratios for Māori and Pacific people.

Source of bias and likely direction of effect on Contextual remarks hospitalisation odds ratio for Māori and Pacific people Outdated or inaccurate total ethnicity data ↑ Using the EpiSurv ethnicity field (which is only prioritised ethnicity using the first priority or- dering in Table 2) results a larger estimated odds ratio for Māori. Multimorbidity and underreporting of ↓ Māori and Pacific people have higher rates of comorbid conditions multi-morbidity and under-reporting of comor- bid conditions. If Māori and Pacific cases in the dataset have multiple comorbid conditions or comorbid conditions that are not reported, the re- ported odds ratio could overestimate the relative risk of hospitalisation. Change in COVID-19 testing over time ↑ In the first wave, which was dominated by NZ European cases, testing and contact tracing rates were lower, and testing was less accessible than in the second wave, which had more Māori and Pacific cases. If more mild cases were missed in the first wave than in the second, this could make the hospitalisation risk appear lower in the sec- ond wave. This could mean the model underesti- mates the risk for Māori and Pacific people. Change in threshold for hospitalisation ̶ There is no clear evidence that the threshold for over time hospitalisation with COVID-19 has changed over time. It is possible that the introduction of hotel quarantine facilities for community cases in the second wave meant mild cases were less likely to be hospitalised. If this were the case, the model could underestimate the risk for Māori and Pacific people. Overrepresentation of international travel- ↑ Excluding overseas cases from the analysis lers in dataset increased the odds ratio for Māori and Pacific people.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 39 www.nzma.org.nz/journal article

This is currently not the case, yet these use only two or three predictor variables. groups have high risks of poor outcomes This highlights the variables with the largest from COVID-19 infection. Ideally, ethnicity impacts on the results, but necessarily information should be either collected at the ignores factors that could have important time of testing or sourced from the existing effects on risk. If in future New Zealand has NHI information. The collection of high- significantly more hospitalisations from quality ethnicity information can be done COVID-19, the analysis should be rerun to quickly and simply, even in busy clinical take account of the additional data. With a settings. This study has also highlighted larger number of cases, the model selection the differential impact of missing data on phase of our approach could include more understanding the course and impact of the variables in the model. Our approach uses epidemic, which is important for informing information that is routinely collected for interventions, including the vaccine delivery all cases of COVID-19 in New Zealand, so it programme. Data completeness checks and would be straightforward to run with an follow-ups of missing data are simple quality updated dataset. control mechanisms for improving the reli- After controlling for age, presence of ability of routinely collected but essential underlying health conditions and socio- information. economic deprivation, we conclude that The results we have presented are from a Māori and Pacific people have substantially relatively small number of cases that may higher risk of hospitalisation for COVID-19. not be representative of the New Zealand We have previously estimated that Māori population, due to the limited spread of and Pacific people would experience higher these outbreaks. Consequently, although infection fatality rates from COVID-19.5 Our our results are based on all cases for which new results add to the imperative for New data are available, caution should be used Zealand’s COVID-19 response to include a when generalising the results to other focus on measures to protect high-risk groups groups or the wider community. The small and to prevent the large-scale inequities number of cases and hospitalisations also in health outcomes that would result from makes it difficult to separate the effects of widespread community transmission.25 Our different variables: for example, the effect of results also have clear implications for identi- belonging to multiple ethnicities or having fying priority groups for vaccination against multiple comorbid conditions recorded. COVID-19, for which planning is currently We have used a likelihood-based approach underway. They demonstrate that it will (AIC) that penalises the use of models with be essential to account for ethnicity when too many variables. The results we have targeting vaccination to age groups based on presented are from very simple models that their risk of clinically severe infection.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 40 www.nzma.org.nz/journal article

Appendix

Appendix Table 1: Breakdown of the number of cases in the data set by ethnicity showing cases with a single ethnicity (bold numbers) and cases with two ethnicities (non-bold numbers). In addition to the 1,821 single and dual ethnicity cases represented in the table, there were 8 cases with three ethnicities, of which 6 were Māori, Pacific and NZ European/other, and 2 were Pacific, Asian and NZ European/other.

Māori Pacific Asian NZ Euro/other

Māori 99 13 1 59

Pacific 13 166 8 15

Asian 1 8 283 6

NZ Euro/other 59 15 6 1,171 Total 172 202 298 1,251

Appendix Table 2: Results of sensitivity analysis for the model for probability of hospitalisation. View Appendix Table 2.

Appendix Table 3: Results of sensitivity analysis for the model for length of hospital stay. View Appendix Table 3.

Appendix Figure 1: Model results including 95% confidence intervals for probability of hospitalisation by age and ethnicity for individuals without underlying health conditions (top row of plots) and with underlying health conditions (bottom row of plots).

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 41 www.nzma.org.nz/journal article

Competing interests: Mr Steyn, Dr Plank, Dr Lustig, Dr James, Dr Ridings, Dr Hendy, Dr Hannah and Dr Binny report grants from Te Pūnaha Matatini and the Ministry of Business, Innovation and Employment during the conduct of the study. Acknowledgements The authors acknowledge the support of Stats NZ, ESR and the Ministry of Health in supply- ing data in support of this work. In particular, we would like to acknowledge Laura Cleary for her work in providing data on total ethnicity and meshblock of home address. The authors are grateful to Melissa McLeod, Ricci Harris, Anja Mizdrak, Patricia Priest and three anonymous reviewers for comments on an earlier version of this manuscript. This work was funded by the New Zealand Ministry of Business, Innovation and Employment and Te Pūnaha Matatini, Centre of Research Excellence in Complex Systems. Andrew Sporle is also funded by Health Research Council Project Grant 20/1018. Author information: Nicholas Steyn: School of Mathematics and Statistics, University of Canterbury, Christchurch, New Zealand; Department of Physics, University of Auckland, Auckland, New Zealand; Te Pūnaha Maatini: Centre of Research Excellence in Complex Systems, New Zealand. Rachelle N Binny: Manaaki Whenua, Lincoln, New Zealand; Te Pūnaha Matatini: Centre of Research Excellence in Complex Systems, New Zealand. Kate Hannah: Department of Physics, University of Auckland, Auckland, New Zealand; Te Pūnaha Matatini: Centre of Research Excellence in Complex Systems, New Zealand. Shaun C Hendy: Department of Physics, University of Auckland, Auckland, New Zealand; Te Pūnaha Matatini: Centre of Research Excellence in Complex Systems, New Zealand. Alex James: School of Mathematics and Statistics, University of Canterbury, Christchurch, New Zealand; Te Pūnaha Matatini: Centre of Research Excellence in Complex Systems, New Zealand. Audrey Lustig: Manaaki Whenua, Lincoln, New Zealand; Te Pūnaha Matatini: Centre of Research Excellence in Complex Systems, New Zealand. Kannan Ridings: Department of Physics, University of Auckland, Auckland, New Zealand; Te Pūnaha Matatini: Centre of Research Excellence in Complex Systems, New Zealand. Michael J Plank: School of Mathematics and Statistics, University of Canterbury, Christchurch, New Zealand; Te Pūnaha Matatini: Centre of Research Excellence in Complex Systems, New Zealand. Andrew Sporle: Department of Statistics, University of Auckland, Auckland, New Zealand; iNZight Analytics Ltd., Auckland, New Zealand. Corresponding author: Michael J Plank, School of Mathematics and Statistics, University of Canterbury, Christchurch 8140, New Zealand, +64 3 3692462 [email protected] URL: www.nzma.org.nz/journal-articles/maori-and-pacific-people-in-new-zealand-have-a-higher- risk-of-hospitalisation-for-covid-19-open-access

REFERENCES 1. ESR [Internet]. NZ COVID- New Zealand: Bridget Kukutai T, et al. Estimated 19 Dashboard 2020. Williams Books; 2017. inequities in COVID-19 Available from: nzcovid- 4. Talamaivao N, Harris R, infection fatality rates dashboard.esr.cri.nz. Cormack D, Paine S-J, King by ethnicity for Aotearoa 2. Stats NZ [Internet]. P. Racism and health in New Zealand. New Census 2018 Usually Aotearoa New Zealand: Zealand Medical Journal. Resident Population 2018. a systematic review of 2020;133(1520):28-39. Available from: nzdot- quantitative studies. New 6. Wilson N, Barnard LT, stat.stats.govt.nz. Zealand Medical Journal. Summers JA, Shanks 3. Salesa D. Island Time 2020;133(1520):55-68. GD, Baker MG. Differ- - New Zealand’s Pacific 5. Steyn N, Binny RN, Hannah ential mortality rates Futures. Wellington, K, Hendy SC, James A, by ethnicity in 3 influ-

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 42 www.nzma.org.nz/journal article

enza pandemics over a December 2019. Univer- itz-Katz G. Assessing the century, New Zealand. sity of Otago; 2019. age specificity of infection Emerging Infectious 13. Anderson D, Burnham fatality rates for COVID- Diseases. 2012;18(1):71-7. K. Model selection and 19: systematic review, 7. Jefferies S, French N, multi-model inference. meta-analysis, and public Gilkison C, Graham G, Hope New York: Springer; 2004. policy implications. European Journal of V, Marshall J, et al. COVID- 14. Fawcett T. An introduction Epidemiology. 2020. 19 in New Zealand and to ROC analysis. Pattern the impact of the national Recognition Letters. 21. Paradies Y, Ben J, Denson response: a descriptive 2006;27(8):861-74. N, Elias A, Priest N, epidemiological study. Pieterse A, et al. Racism as 15. Cormack D. The politics The Lancet Public Health. a determinant of health: and practice of counting: 2020;5(11):e612-e23. a systematic review and ethnicity in official statistics meta-analysis. PLOS One. 8. ESR [Internet]. EpiSurv: in Aotearoa/New Zealand: 2015;10(9):e0138511. Coronavirus (COVID-19) Te Rōpū Rangahau Hauora case report form 2020. a Eru Pōmare; 2010. 22. Robson B, Harris R. Hauora: Available from: https:// Māori Standards of Health 16. Stats NZ [Internet]. surv.esr.cri.nz/episurv/ IV. A study of the years Census 2018 ethnic group CaseReportForms/Coro- 2000–2005. Wellington: Te summaries 2018. Available navirus_Sep2020.pdf. Ropu Rangahau Hauora from: https://www.stats. a Eru Pomare. 2007. 9. Stats NZ [Internet]. 2018 govt.nz/information-re- Census: Design of forms leases/2018-census-eth- 23. Cormack D, Harris R, Wellington, New Zealand nic-groups-dataset. Stanley J, Lacey C, Jones 2018. Available from: R, Curtis E. Ethnic bias 17. Ministry of Health. Wai https://www.stats.govt. amongst medical students 2575 Māori Health Trends nz/methods/2018-cen- in Aotearoa/New Zealand: Report. Wellington, sus-design-of-forms. Findings from the Bias and New Zealand; 2019. 10. Ministry of Health [Inter- Decision Making in Medi- 18. Yon Y, Crimmins EM. net]. HISO 10001:2017 cine (BDMM) study. PLOS Cohort morbidity hypoth- Ethnicity Data Protocols One. 2018;13(8):e0201168. esis: health inequalities of 2017. Available from: 24. Galobardes B, Shaw M, older Māori and non-Māori https://www.health. Lawlor DA, Lynch JW, in New Zealand. New govt.nz/publication/ Smith GD. Indicators of Zealand Population hiso-100012017-ethnic- socioeconomic position Review. 2014;40:63. ity-data-protocols. (part 1). Journal of Epide- 19. Reid J, Taylor-Moore K, 11. Williamson EJ, Walker miology and Community Varona G. Towards a AJ, Bhaskaran K, Bacon S, Health. 2006;60(1):7-12. social-structural model Bates C, Morton CE, et al. 25. Ministry of Health for understanding current Factors associated with [Internet]. Annual Data disparities in Maori health COVID-19-related death Explorer 2018/19: New and well-being. Journal using OpenSAFELY. Nature. Zealand Health Survey of Loss and Trauma. 2020;584(7821):430-6. 2019. Available from: 2014;19(6):514-36. 12. Atkinson J, Salmond C, https://minhealthnz. 20. Levin AT, Hanage WP, Crampton P. NZDep2018 shinyapps.io/nz-health-sur- Owusu-Boaitey N, Cochran Index of Deprivation, vey-2018-19-annu- KB, Walsh SP, Meyerow- Interim Research Report, al-data-explorer/.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 43 www.nzma.org.nz/journal article

Higher perceived stress and exacerbated motor symptoms in Parkinson’s disease during the COVID-19 lockdown in New Zealand Rebekah L Blakemore, Maddie J Pascoe, Kyla-Louise Horne, Leslie Livingston, Bob N Young, Beth Elias, Marie Goulden, Sophie Grenfell, Daniel J Myall, Toni L Pitcher, John C Dalrymple-Alford, Campbell J Le Heron, Tim J Anderson, Michael R MacAskill

ABSTRACT AIMS: Stress plays a key role in Parkinson’s disease (PD) by acting on the dopaminergic system and worsening patients’ motor function. The impact of New Zealand’s strict lockdown measures to contain COVID-19 on perceived stress and PD motor symptoms remains unknown. Here we examined the relationship between perceived levels of stress, changes in physical activity levels and PD motor symptoms during lockdown. METHODS: During lockdown, 134 participants with PD and 49 controls completed a survey assessing perceived stress, self-reported changes in PD motor symptoms and physical activity duration and intensity prior to and during lockdown. RESULTS: Perceived stress was higher in PD than controls, and in those reporting a worsening of tremor, balance/gait, dyskinesia and bradykinesia compared to those indicating no change during the COVID-19 lockdown. These effects were not modulated by physical activity. CONCLUSIONS: Reducing stressors may be an important adjunct treatment strategy to improve motor function in PD.

s the COVID-19 pandemic continues the dopaminergic system and worsening to grip the globe, there is concern for motor symptoms.4–7 Heightened stress can Athe increased vulnerability of those reduce the efficacy of levodopa treatment, with chronic disease, including Parkinson’s resulting in further worsening of motor disease (PD).1 New Zealand was placed symptoms.8 However, physical activity can under strict lockdown for 33 days. Residents counteract the detrimental effects of stress were required to stay home except to exer- and might even exert neuroprotective cise or access essential services. Although effects.9,10 The COVID-19 lockdown period crucial to control the spread of the virus, may have therefore placed this cohort, these drastic measures had the potential to relative to those without PD, at greater significantly impact on mental and physical risk of experiencing higher stress, which, wellbeing. together with potential changes in physical People with PD may have an impaired activity, could exacerbate interactions ability to cope with sudden changes to between stress and their already compro- everyday life due to their cognitive and mised motor system. Our objectives were motor inflexibility,2,3 which stems from to examine whether during New Zealand’s nigrostriatal dopamine depletion, a patho- stringent lockdown period (1) perceived logical hallmark of the disease. Stress also stress was higher in PD than controls, (2) plays an important role in PD by acting on there was a direct relationship between

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 44 www.nzma.org.nz/journal article

higher perceived stress and worsening of PD All participants gave informed consent. motor symptoms and (3) these effects were The study was conducted within an existing modulated by physical activity. longitudinal PD study that was approved by the Health and Disability Ethics Committee Material and methods of the New Zealand Ministry of Health. Participants Measures Participants were recruited from the Participants completed a suite of scales established PD longitudinal study at the New and questionnaires as part of the survey. Zealand Brain Research Institute (NZBRI). Study data were collected and managed 15 One hundred and forty-nine participants using REDCap electronic data capture tools with PD and 51 healthy controls were hosted at NZBRI. We report findings from invited to take part in a survey during the three scales: First, we created the PD motor nationwide lockdown, either online or by symptom scale (PDMSS) to examine self-re- phone. Of the 200 people invited, 134 PD ported perceived changes in four motor (M=72 years, SD=7) and 49 controls (M=78 symptoms since lockdown began: tremor, years, SD=7) completed the survey (a dyskinesia, balance/gait and bradykinesia. response rate of 92%). All PD participants For each motor symptom, PD participants had previously completed comprehensive indicated whether they had noticed any neuropsychological11 and MDS-UPDRS12 changes in their symptom presentation assessments as part of their participation in during lockdown compared to before the NZBRI longitudinal PD study (Table 1). lockdown. If so, two further questions probed whether that symptom was ‘better’ Data collection occurred between 17 April or ‘worse’, and the magnitude of that change and 14 May 2020, during New Zealand’s was measured on a four-point Likert scale Alert Level 4 and Level 3 periods.14 At Alert (slight, mild, moderate, substantial). Level 4 (commencing 25 March), New Zealanders were in strict lockdown unless Second, the perceived stress scale (PSS- 16,17 providing an essential service. At Alert Level 10) assessed participants’ (PD and 3 (beginning 27 April and lasting 17 days), controls) perceived stress levels during restrictions were slightly loosened, but lockdown. Participants self-reported how vulnerable individuals (those over 70 years frequently they had felt a certain way on or immunocompromised) were advised to a five-point Likert scale (from 0 (‘Never’) continue self-isolating where possible. At to 4 (‘Very often’). For the present study, the commencement of data collection, Alert the wording of each question was inten- Level 4 had been in place for 22 days. As far tionally altered from ‘In the last month’ as we are aware, none of our participants to ‘Since lockdown began’. One further tested positive for COVID-19. question required participants to report

Table 1: Percentage distribution of self-reported changes (worsening) in PD motor symptoms during the COVID-19 lockdown. Clinical characteristicsa of our patient cohort (n=134): mean disease duration 11 years (SD=6); Hoehn–Yahr score13 2.4 (SD=0.6); last pre-lockdown Part III MDS-UPDRS12 ‘ON’ PD medica- tion 34.0 (SD=13.2).

Motor Symptom (% reporting symptom) No change Slight Mild Moderate Severe Tremor (89%) 70% 12% 7% 7% 4%

Dyskinesia (88%) 74% 11% 8% 7% 0%

Bradykinesia (98%) 64% 14% 10% 8% 4%

Balance/gait (98%) 53% 15% 15% 13% 4%

aThese scores were obtained during each participant’s most recent assessment visit as part of the NZBRI longitudinal PD study, on average 12 months (SD=10) prior to the present data collection.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 45 www.nzma.org.nz/journal article

whether their stress levels were higher, upon reasonable request from any qualified lower or about the same compared to before investigator. lockdown began. Third, we created the physical activity Results levels questionnaire (PALQ) to assess the Mean perceived stress levels in both frequency, duration and intensity of each groups were relatively low (PD, M=12.1, participant’s physical activity during the SD=6.4; controls, M=8.6, SD=6.0); however, preceding seven days (during lockdown) perceived stress was higher in those with and in the seven days prior to lockdown. Parkinson’s than controls (by 3.6 points, This questionnaire was based on items 95% CI [1.5, 5.7], probability of perceived included in the International Physical stress being higher in Parkinson’s P>99%; 18 Activity Questionnaire. PD participants Figure 1A). The majority of PD participants also indicated whether they had partic- (69%) reported their stress levels were about ipated in an exercise group, class or the same during lockdown compared to programme specifically for people with PD the period immediately prior to lockdown. before the lockdown, and if so, whether Twenty-two percent felt their stress levels they had been able to continue this in some were higher and 9% felt they were lower. form during lockdown. Similar results were found for controls Statistical analyses (stress levels about the same 67%; higher, Perceived stress, physical activity 20%; lower, 12%). levels and self-reported changes in motor Despite being in lockdown, both PD symptoms during lockdown were examined and controls continued being physically using Bayesian multilevel modelling. The active, with no evidence of a difference probabilistic language Stan was used along in the overall amount of time spent being with the R packages rstan (v2.19.3) and active in the current week compared to the brms (v2.13.0) to fit Bayesian models and week prior to lockdown, and no evidence generate estimates within the R statistical of a difference between groups (Figure environment (v4.0.119). First, group differ- 1B). Almost half the PD participants (49%) ences in perceived stress and physical considered their physical activity levels activity levels (total hours active over the during lockdown to be light in intensity, 42% week and intensity) were analysed. In these moderate in intensity and 9% indicated their models, group was included as a predictor physical activity was vigorous, involving and the intercept varied by participant. hard physical effort. Similar physical activity The total hours active and physical activity intensities were reported for controls intensity models additionally included (light, 51%; moderate, 42%; vigorous, 7%), a group-by-timepoint interaction term as well as prior to lockdown. Twenty-three (before versus during lockdown). Next, percent of PD participants indicated they we examined whether there was a rela- were attending a PD-specific exercise group tionship between perceived stress, PD motor prior to lockdown, of which 37% said they symptom changes and physical activity. For were able to continue during lockdown each symptom (tremor, dyskinesia, balance/ via, for example, online classes. There was gait, bradykinesia), scores of symptom wors- no evidence of a relationship between ening from the PDMSS were collapsed across perceived stress and physical activity ‘slight’ to ‘substantial’ levels to generate a duration or intensity during lockdown for summary score of ‘Worsened’ versus ‘No either group. change’ in symptoms during lockdown. The In those with PD, we examined any resulting motor symptom change scores self-reported changes since lockdown in for each symptom were modelled with a the motor symptoms of tremor, dyski- primary predictor of PSS score. Results are nesia, bradykinesia and balance/gait. Of reported as the means of the posterior distri- the 89% of PD participants who self-re- bution, together with 95% credible intervals ported tremor, 70% indicated no changes (CI) and the probability of the parameter in their tremor since lockdown began. being greater than zero. Anonymised data The remaining 30% reported a worsening that support the findings of this study will that ranged from slight to severe changes be shared by the corresponding author (Table 1). Dyskinesia and bradykinesia

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 46 www.nzma.org.nz/journal article

also showed similar patterns of change. Of during the pandemic.26–28 The low levels the 98% of participants who self-reported of stress in this study are consistent with balance/gait problems, almost half indicated our findings of no change in depression or these symptoms were exacerbated during anxiety levels in the same PD and control lockdown. cohort during lockdown compared to Next we examined whether subjective pre-lockdown (unpublished observations), motor symptom changes were associated possibly due to the effectiveness of the New with perceived levels of stress. As shown Zealand government’s policy and strategy 29 in Figure 1C–F, scores on the perceived to contain the virus. The relative success stress scale were higher in those reporting of this swift and stringent response may a worsening of tremor (by 2.4 points, 95% have contributed to the mental and physical CI [0.7, 4.2], probability of perceived stress wellbeing of residents and thus influenced being higher in those reporting a worsening the results we report here. However, the of tremor P>99%), dyskinesia (2.1 points, low perceived stress levels could also reflect 95% CI [0.1, 4], P=98%), balance/gait (1.8 the older age of our participants and a points, 95% CI [0.2, 3.4], P=99%) and brady- larger percentage of male responders (58%), kinesia (1.6 points, 95% CI [-0.01, 3.17], as perceived stress during the pandemic P=97%) compared to those indicating these is reported to be higher in females and 26,28 symptoms did not change during lockdown. younger people. Including physical activity duration and New Zealand’s strict lockdown did not intensity as predictors in these analyses did appear to impact on participants’ reported not improve the model fits. physical activity levels, perhaps in part because explicit provision for outdoor Discussion exercise was a feature of government guidelines. Moreover, the effects of stress Consistent with previous literature on motor symptom presentation and the showing stress plays a key role in PD and between-group differences in perceived alters motor system functioning,4–8 perceived stress during lockdown were not modulated stress was higher in PD than healthy by physical activity duration or intensity. controls and, importantly, higher in PD These findings contrast with other reports participants reporting a worsening of motor that demonstrate reduced physical activity symptoms during lockdown compared to since the pandemic in PD,24,25 which was those reporting no impact of lockdown on associated with worsening of symptoms motor symptoms. Over a quarter of PD (combining both motor and non-motor participants reported changes in one or symptoms together).25 The lack of support more motor symptoms during lockdown. for previous research showing beneficial A worsening of PD motor symptoms since effects of physical activity on stress9,10,30 may the onset of the COVID-19 pandemic has be explained by the already low average been reported by others.20–24 Consistent with stress levels of our groups. van der Heide et al,25 our results extend these findings to show a direct relationship Given New Zealand’s unique pandemic between higher perceived stress and dete- response to COVID-19, our findings may rioration of motor function in PD; however, not generalise to other PD and older-aged they cannot demonstrate causality (ie, healthy control cohorts around the world. whether increased symptom presentation Nonetheless, monitoring and addressing increases stress or vice versa). stress levels, particularly as this pandemic continues, may be an important adjunct Interestingly, the levels of perceived stress strategy to mitigate exacerbation of reported here appear lower than that found symptoms and improve motor function in the general population across the globe in PD.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 47 www.nzma.org.nz/journal article

Figure 1: (A) Total perceived stress scores (PSS) since lockdown began for PD patients and controls. (B) Duration of time spent being physically active over the week immediately prior to lockdown and one week during lock- down for patients and controls. Includes physical activity level intensities ranging from light to vigorous. Data for time (hours) were log transformed before analysis. (C–F) PSS scores for those participants reporting a worsening of their motor symptoms (Worsened) compared to those participants reporting no change in their symptom presentation (No Change), for (C) tremor, (D) dyskinesia, (E) bradykinesia and (F) balance/gait. Box plots illustrate the median and quartiles; individual patient and control data are also shown (grey circles).

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 48 www.nzma.org.nz/journal article

Competing interests: Nil. Acknowledgements: We acknowledge people who have collected earlier data for the existing New Zealand Brain Research Institute’s longitudinal Parkinson’s disease study, including Saskia van Stockum, Charlotte Graham, Krysta Callander, Megan Livingstone and Meisha Nicolson. Author information: Rebekah L Blakemore: PhD; New Zealand Brain Research Institute, Christchurch, New Zea- land; School of Physical Education, Sport and Exercise Sciences, University of Otago, Duned- in, New Zealand; Brain Health Research Centre, University of Otago, Dunedin, New Zealand. Maddie J Pascoe: BSc; New Zealand Brain Research Institute, Christchurch, New Zealand. Kyla-Louise Horne: PhD; New Zealand Brain Research Institute, Christchurch, New Zealand; Department of Medicine, University of Otago, Christchurch, New Zealand; Brain Research New Zealand – Rangahau Roro Aotearoa Centre of Research Excellence. Leslie Livingston: BA (Hons); New Zealand Brain Research Institute, Christchurch, New Zea- land; Department of Medicine, University of Otago, Christchurch, New Zealand Bob N Young: MSc; New Zealand Brain Research Institute, Christchurch, New Zealand. Beth Elias: MA; New Zealand Brain Research Institute, Christchurch, New Zealand. Marie Goulden: BN PGCert; New Zealand Brain Research Institute, Christchurch, New Zea- land; Brain Research New Zealand – Rangahau Roro Aotearoa Centre of Research Excellence Sophie Grenfell: MSc; New Zealand Brain Research Institute, Christchurch, New Zealand. Daniel J Myall: PhD; New Zealand Brain Research Institute, Christchurch, New Zealand. Toni L Pitcher: New Zealand Brain Research Institute, Christchurch, New Zealand; Depart- ment of Medicine, University of Otago, Christchurch, New Zealand; Brain Research New Zealand – Rangahau Roro Aotearoa Centre of Research Excellence. John C Dalrymple-Alford: PhD; New Zealand Brain Research Institute, Christchurch, New Zealand; Brain Research New Zealand – Rangahau Roro Aotearoa Centre of Research Excel- lence; School of Psychology, University of Canterbury, New Zealand. Campbell J Le Heron: FRACP, DPhil; Department of Medicine, University of Otago, Christchurch, New Zealand; Department of Neurology, Christchurch Hospital, Christchurch, New Zealand. Tim J Anderson: FRACP, MD; New Zealand Brain Research Institute, Christchurch, New Zealand. Michael R MacAskill: PhD; Department of Medicine, University of Otago, Christchurch, New Zealand; Brain Research New Zealand – Rangahau Roro Aotearoa Centre of Research Excellence; Department of Neurology, Christchurch Hospital, Christchurch, New Zealand. Corresponding author: Dr Rebekah Blakemore, School of Physical Education, Sport and Exercise Sciences, University of Otago, 55 Union Street, Dunedin, 9016, New Zealand, +643 479 8387 [email protected] URL: www.nzma.org.nz/journal-articles/higher-perceived-stress-and-exacerbated-motor-symp- toms-in-parkinsons-disease-during-the-covid-19-lockdown-in-new-zealand-open-access

REFERENCES 1. Helmich, R C & Bloem, B R. H J. Cognitive and motor but increases impulsivity in The impact of the COVID-19 shifting aptitude disorder patients with Parkinson’s pandemic on Parkinson’s in Parkinson’s disease. disease. Neuropsychologia disease: Hidden sorrows Journal of Neurology, 41, 1431-1441 (2003). and emerging opportuni- Neurosurgery & Psychiatry 4. Blakemore, R L, MacAskill, ties. Journal of Parkinson’s 47, 443-453 (1984). M R, Shoorangiz, R & disease 10, 351-354 (2020). 3. Cools, R, Barker, R A, Saha- Anderson, T J. Stress-evok- 2. Cools, A R, van den Berck- kian, B J & Robbins, T W. ing emotional stimuli en, J H, Horstink, M W, van L-Dopa medication remedi- exaggerate deficits in motor Spaendonck, K P & Berger, ates cognitive inflexibility, function in Parkinson’s

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 49 www.nzma.org.nz/journal article

disease. Neuropsycholo- format, and clinimetric A community-based gia 112, 66-76 (2018). testing plan. Movement case-control study. 5. Hemmerle, A M, Dickerson, Disorders 22, 41-47 (2007). Movement Disorders J W, Herman, J P & Seroogy, 13. Hoehn, M M & Yahr, 35, 1287-1292 (2020). K B. Stress exacerbates M D. Parkinsonism: 21. Prasad, S et al. Parkin- experimental Parkinson’s Onset, progression and son’s disease and disease. Molecular Psychi- mortality. Neurology COVID-19: Perceptions atry 19, 638-640 (2014). 17, 424-442 (1967). and implications in 6. Metz, G A. Stress as a 14. New Zealand Government patients and caregivers. modulator of motor system [Internet]. Alert system Movement Disorders function and pathology. overview: Unite against 35, 912-914 (2020). Reviews in the Neurosci- COVID-19. Available from: 22. Schirinzi, T et al. ences 18, 209-222 (2007). https://covid19.govt.nz/ Self-reported needs of 7. Smith, L K, Jadavji, N M, alert-system/alert-sys- patients with Parkinson’s Colwell, K L, Perehudoff, tem-overview/ (2020). disease during COVID- S K & Metz, G A. Stress 15. Harris, P A et al. Research 19 emergency in Italy. accelerates neural degen- electronic data capture Neurological Sciences eration and exaggerates (REDCap): A metada- 41, 1373-1375 (2020). motor symptoms in a ta-driven methodology 23. Brown, E G et al. The rat model of Parkinson’s and workflow process for effect of the COVID-19 disease. European providing translational pandemic on people with Journal of Neuroscience research informatics Parkinson’s disease. medRx- 27, 2133-2146 (2008). support. Journal of Biomed- iv, 2020.2007.2014.20153023 8. Zach, H, Dirkx, M F, ical Informatics 42, (2020). Pasman, J W, Bloem, B R 377-381 (2009). 24. Shalash, A et al. Mental & Helmich, R C. Cognitive 16. Cohen, S, Kamarck, T & health, physical activity, stress reduces the effect of Mermelstein, R. A global and quality of life in levodopa on Parkinson’s measure of perceived Parkinson’s disease during resting tremor. CNS stress. Journal of Health COVID-19 pandemic. Neuroscience & Therapeu- and Social Behavior Movement Disorders tics 23, 209-215 (2017). 24, 385-396 (1983). 35, 1097-1099 (2020). 9. Smith, A D & Zigmond, M 17. Cohen, S & William- 25. van der Heide, A, Meinders, J. Can the brain be protect- son, G. in The Social M J, Bloem, B R & Helmich, ed through exercise? Psychology of Health: R C. The impact of the Lessons from an animal Claremont Symposium on COVID-19 pandemic on model of parkinsonism. Applied Social Psycholo- psychological distress, Experimental Neurology gy (eds S Spacapan & S physical activity, and 184, 31-39 (2003). Oskamp) (Sage, 1988). symptom severity in 10. Adlard, P A & Cotman, C W. 18. Lee, P H, Macfarlane, D Parkinson’s disease. Jour- Voluntary exercise protects J, Lam, T H & Stewart, S nal of Parkinson’s disease against stress-induced M. Validity of the inter- Pre-press, 1-10 (2020). decreases in brain-derived national physical activity 26. Limcaoco, R S G, Mateos, neurotrophic factor protein questionnaire short form E M, Fernandez, J M & expression. Neuroscience (IPAQ-SF): A systematic Roncero, C. Anxiety, worry 124, 985-992 (2004). review. International Jour- and perceived stress in the 11. Wood, K-L et al. Different nal of Behavioral Nutrition world due to the COVID-19 PD-MCI criteria and risk and Physical Activity pandemic, March 2020. of dementia in Parkinson’s 8, 115 (2011). Preliminary results. medRx- disease: 4-year longitudinal 19. R Core Team. R: A iv, 2020.2004.2003.20043992 study. npj Parkinson’s Language and Environment (2020). disease 2, 15027 (2016). for Statistical Comput- 27. Grover, S et al. Psycho- 12. Goetz, C G et al. Movement ing. Vienna, Austria: R logical impact of Disorder Society-spon- Foundation for Statistical COVID-19 lockdown: An sored revision of the Computing (2015). online survey from India. Unified Parkinson’s 20. Cilia, R et al. Effects of Indian Journal of Psychi- Disease Rating Scale COVID-19 on Parkinson’s atry 62, 354-362 (2020). (MDS-UPDRS): Process, disease clinical features: 28. Rossi, R et al. COVID-19

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 50 www.nzma.org.nz/journal article

pandemic and lockdown a team of 5 million: Dimeo, F, Wüstenberg, T & measures impact on New Zealand beats back Ströhle, A. The stress-buff- mental health among COVID-19, March-June ering effect of acute the general population 2020 (Innovations for exercise: Evidence for HPA in Italy. Frontiers in Successful Societies, axis negative feedback. Psychiatry 11, 790 (2020). Princeton University, 2020). Psychoneuroendocrinol- 29. Cameron, B. Captaining 30. Zschucke, E, Renneberg, B, ogy 51, 414-425 (2015).

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 51 www.nzma.org.nz/journal article

Life during lockdown: a qualitative study of low-income New Zealanders’ experience during the COVID-19 pandemic Kimberley Choi, Namratha Giridharan, Abigail Cartmell, Dominique Lum, Louise Signal, Viliami Puloka, Rose Crossin, Lesley Gray, Cheryl Davies, Michael Baker, Amanda Kvalsvig

ABSTRACT AIM: This research explores the experience of low-income New Zealanders during the COVID-19 pandemic lockdown and their advice to the Government about addressing future pandemics. New Zealand had a rapid and effective lockdown that meant the virtual elimination of community transmission. METHOD: Twenty-seven semi-structured interviews were undertaken with low-income people in June– July 2020, immediately after lockdown was lifted. RESULTS: Life during lockdown was challenging for study participants. They were fearful of the virus and experienced mental distress and isolation. Most participants felt safe at home and reported coping financially while still experiencing financial stress. Participants were resourceful and resilient. They coped with lockdown by using technology, self-help techniques and support from others. New Zealand’s welfare state ensured participants had access to health services and welfare payments, but there were challenges. Welfare payments did not fully meet participants’ needs, and support from charitable organisations was critical. Participants were overwhelmingly positive about the Government’s response and advised the Government to take the same approach in the future. This is a particularly reassuring finding from some of the most vulnerable New Zealanders. CONCLUSIONS: An early and hard lockdown, the welfare state, compassion and clearly communicated leadership were keys to a successful lockdown for the low-income people in this study. Research of the experience of low-income people during pandemics is critical to ensuring inequities in pandemic impact are mitigated.

he 2020 COVID-19 pandemic necessi- stigma.”1 Since the study reported here, tated a global public health response there has been a proliferation of research Tresulting in unprecedented nation- on the impact of the COVID-19 lockdown wide lockdowns. Research on the public’s that identifies harms to individuals’ mental experience of pandemics is relatively wellbeing and their ability to access medical limited, and until 2020 it did not address and dental care, food and social support.2–7 the scale of the current lockdowns. A rapid Qualitative studies on people’s experience review of evidence of the psychological im- have largely focused on people with specific pact of quarantine prior to COVID-19 found health conditions or on specific aspects of that impacts were “wide-ranging, substan- their lives. Three qualitative studies from tial and can be long lasting.”1 Key stressors Aotearoa New Zealand focused on aspects of were: “quarantine duration, infection fears, healthcare provision.8-10 frustration, boredom, inadequate supplies, Public health is inherently political as it inadequate information, financial loss, and requires the organised efforts of society.11

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 52 www.nzma.org.nz/journal article

Globally, the pandemic has seen a range of Grants and the newly introduced COVID-19 political leadership and policy responses Wage Subsidy. Community Services Cards and subsequent public health outcomes. A enable people receiving low-income to key risk in any public health crisis (including get subsidies across a range of services, COVID-19) is exacerbating existing ineq- including general practitioner (GP) appoint- uities.12,13 Understanding the COVID-19 ments and low-cost medical scripts (NZ$5). experience of vulnerable people will assist Many New Zealanders also rely on char- in ensuring equitable pandemic responses. itable organisations for their basic needs Given that people living in poverty suffer (eg, food banks and the City Mission).24 worse outcomes during pandemics14–16 and These services continued to operate during are more isolated in society,17 research lockdown, adapting and augmenting their examining their experience is needed to services as needed.25 protect them during pandemics. New Zealand had its first reported case In 2020 New Zealand eliminated of COVID-19 on 28 February 2020, a month community transmission of COVID-19 after the first cases were reported in using a rapid and stringent lockdown.18,19 Europe.26 New Zealand’s initial approach to This research explores low-income New the pandemic followed the 2017 influenza Zealanders’ experiences of the Government’s pandemic response plan.27 It proved inef- COVID-19 policy response and lockdown fective and an elimination strategy was during March–May 2020. Further, it developed.28 Prime Minister Jacinda Ardern explores participants’ views on the Govern- announced a new Alert Level System on ment’s response and their advice to the 21 March urging New Zealanders to “be Government about how to deal with future strong, be kind, and unite against Covid- pandemics. 19.”29,30 Alert Levels 4 and 3 both involved lockdowns at home, colloquially termed Context the ‘bubble’.31 Housing was found for the homeless within days of the Alert Level New Zealand is a unitary welfare state System announcement. with largely publicly funded healthcare and relatively simple institutional arrange- These lockdown restrictions lasted seven ments for health. The lead agency is the weeks, from 25 March until 13 May 2020, national Ministry of Health led by the at which time the country moved to Alert Director-General of Health.20 Regional public Level 2. Alert Level 2 was much less socially health services are undertaken by district restrictive. On 8 June the country moved health boards. New Zealand’s economic to Alert Level 1, which allowed unre- policy has been strongly neo-liberal in the stricted movement within New Zealand past three decades, which has resulted in but tight border restrictions were retained. welfare payments well below the living Throughout Alert Levels 4 and 3, daily 1pm wage.21,22 The Labour-led government at updates were televised, usually featuring the time of the crisis was clearly moving the Prime Minister and the Director-General in a more social democratic direction, with of Health, who were the key government budgets focused on wellbeing, growth in leaders in the COVID-19 response. New government and increased welfare benefits. Zealand’s response has been characterised 32 Yet, prior to COVID-19, nearly 10% of work- as relying on “science and empathy.” ing-age New Zealanders were receiving a Ardern’s key messages included, “we must ‘main benefit’.23 Table 1 outlines the context go hard, and go early,” and, “our team of 5 at the time of this study. See https://covid19. million,” referring to the efforts of the entire 30,33 govt.nz/ for current information. population. Work and Income New Zealand (WINZ) The initiation of Alert Level 4 in New is the key government agency responsible Zealand on 25 March brought widespread for benefit allocation. WINZ played a key economic and social consequences. The role in the COVID-19 response through the Government provided the COVID-19 Wage provision of existing benefits: Working Subsidy to employers so they could continue Age Benefits (increased during lockdown), paying their staff during the lockdown. Winter Energy Payment (increased during Despite this, the number of working-age lockdown), Hardship Assistance, Food people signing up to benefits during the

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 53 www.nzma.org.nz/journal article

Table 1: The New Zealand Context.

Agencies

Work and Income New Zealand (WINZ) Government agency under the Ministry of Social Develop- ment that offers income support and provides training for job seekers and employers.

https://www.workandincome.govt.nz/

City Mission One of many charitable organisations that operates in major cities throughout New Zealand. Offers support to a wide range of people in hardship. Further information found at:

https://www.aucklandcitymission.org.nz/

Food Banks Range of services and charities in New Zealand provide free food parcels to whānau (extended family). Many services remained open to support their communities as essential services. This range of services can be seen at:

https://www.foodbank.co.nz/

Benefits

Working Age Benefits Government benefits for people aged 18–64. The three major categories are Jobseeker Support, Sole Parent Support and the Supported Living Payment (for people with long-term health conditions/disabilities or their carers). In response to COVID-19, all Working Age Benefits were increased by NZ$25 per week.

https://www.msd.govt.nz/documents/about-msd-and-our- work/publications-resources/statistics/benefit/2020/bene- fit-fact-sheets/benefit-fact-sheets-snapshot-june-2020.pdf

Winter Energy Payment From 1 May to 1 October WINZ provides additional weekly payments to beneficiaries to cover extra heating costs over the winter months, as many people on the benefit cannot afford to heat their homes. This was doubled in response to COVID-19 to NZ$40.91 singles NZ$63.64 for couples or people with dependents to acknowledge people would need to spend more on heating during lockdown.

https://www.workandincome.govt.nz/products/a-z-benefits/ winter-energy-payment.html

Hardship Assistance Emergency and one-off payments from WINZ to help people in acute financial hardship. Covers payments such as Special Needs Grants and Recoverable Assistance Payments.

https://www.workandincome.govt.nz/products/a-z-benefits/ special-needs-grant.html https://www.workandincome.govt.nz/products/a-z-benefits/ recoverable-assistance-payment-grant.html

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 54 www.nzma.org.nz/journal article

Table 1: The New Zealand Context (continued).

Food Grants Grant for people on low-income or a benefit who need help paying for food. Weekly rate is dependent on circumstance.

https://www.workandincome.govt.nz/eligibility/urgent-costs/ food.html

Community Services Card Allows people receiving low-income to have subsidised access to services such as GP appointments, medical scripts, pools and gyms.

https://www.workandincome.govt.nz/products/a-z-benefits/ community-services-card.html

Wage Subsidy Available to New Zealand businesses that experienced a greater than 40% decline in revenue during a 30-day period due to COVID-19. In return the businesses were expected to keep their employees at 80% of their usual wage. This scheme was initially for a 12-week period but businesses were then able to apply for another 8-week extension. Indi- viduals who had lost employment due to COVID-19 were also able to apply for income relief payments of up to NZ$490 for up to 12 weeks.

https://www.workandincome.govt.nz/covid-19/wage-subsidy/ index.html

Timeline*

28 February • First case reported

14 March • 14-days self-isolation at border (except Pacific)

19 March • Border closed to all but returning New Zealanders • 100-people gathering limit

21 March • Four-level Alert Level System announced • New Zealand at Alert Level 2

23 March • New Zealand at Alert Level 3

25 March • New Zealand at Alert Level 4 • National State of Emergency declared

27 April • New Zealand at Alert Level 3

13 May • New Zealand at Alert Level 2

8 June • New Zealand at Alert Level 1

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 55 www.nzma.org.nz/journal article

Table 1: The New Zealand Context (continued).

Alert Level System at Time of First Lockdown*

Alert Level 4: Eliminate • People instructed to stay at home • Educational facilities closed Sustained and intensive transmission • Businesses closed except for essential services (eg, supermarkets, pharmacies, clinics) and lifeline utilities Widespread outbreaks • Rationing of supplies and requisitioning of facilities • Travel severely limited • Major reprioritisation of healthcare services

Alert Level 3: Restrict • Travel in areas with clusters or community transmission limited Community transmission occurring, or • Affected educational facilities closed multiple clusters break out • Mass gatherings cancelled • Public venues closed (eg, libraries, museums, cinemas, food courts, gyms, pools, amusement parks) • Alternative ways of working required, and some non-es- sential businesses should close • Non-face-to-face primary care consultations • Non-acute (elective) services and procedures in hospi- tals deferred and healthcare staff reprioritised

Alert Level 2: Reduce • Border entry measures maximised • Further restrictions on mass gatherings High risk of importing COVID-19, or • Physical distancing on public transport (eg, leave the uptick in imported cases, or uptick in seat next to you empty if you can) household transmission, or single or • Limit non-essential travel around New Zealand isolated cluster outbreak • Employers start alternative ways of working if possible (eg, remote working, shift-based working, physical dis- tancing within the workplace, staggering meal breaks, flexible leave arrangements) • Business continuity plans activated • High-risk people advised to remain at home (eg, those over 70 or those with other existing medical conditions)

Alert Level 1: Prepare • Border entry measures to minimise risk of importing COVID-19 cases applied Heightened risk of importing COVID-19, • Contact tracing or sporadic imported cases, or isolated • Stringent self-isolation and quarantine household transmission associated • Intensive testing for COVID-19 with imported cases • Physical distancing encouraged • Mass gatherings over 500 cancelled • Stay home if you’re sick and report flu-like symptoms • Wash and dry hands, cough into elbow and don’t touch your face

*https://covid19.govt.nz/alert-system/history-of-the-covid-19-alert-system/

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 56 www.nzma.org.nz/journal article

one-month period from March to April participants had experienced abuse, home- changed at a rate almost double that of lessness or drug addiction, or they had been the previous 24 years.23 Hardship Assis- through the justice system. All participants tance payments rose sharply.34 Food banks spent their lockdown at private dwellings, reported a 100% increase in demand during except for four, of whom one lived in a the lockdown period.35 Charitable organi- refugee trust home, one a City Mission sations expressed concern over the rising home, one an institutional care home and number of those seeking support due to one lived in their car. 36 unemployment. Data were collected through semi-struc- tured interviews (the interview guide is Methods available from the authors on request). This qualitative study included a Questions focused on life during lockdown, purposeful sample of 27 low-income New how participants coped, what would have Zealanders aged 18 years or older. The made lockdown easier, participants’ views research was planned during lockdown by a of the Government’s COVID-19 response new research team working by Zoom. Inter- and its impact on their whānau (extended views were undertaken immediately after family) and their advice to the Prime lockdown, between 30 June and 31 July 2020. Minister and Director-General of Health People were selected through community about how to deal with pandemics in the organisations, including the City Mission, future. Interviews took around 30 minutes. in two cities in New Zealand (Auckland They were audio-recorded with permission. and Christchurch); Auckland population Ethical approval for this study was obtained ~1.6million and Christchurch population from the University of Otago Human Ethics ~340,000 (www.stats.govt.nz/2018-census). Committee (D20/182). Of the participants, 15 (56%) were female This study is framed by political theory, and 12 (44%) were male. Ethnicity was particularly theory of political economy.38,39 determined by self-identification, with Applying a political economy perspective some participants identifying with multiple requires an analysis of political discourse ethnicities. In total, seven (26%) participants and action, including an analysis of the identified as Māori, 15 (56%) as New Zealand role of the state.21 Key to this is the extent European, two as Dutch and one each as to which the Government response was Cook Island Māori, Iranian, British, Indian, neo-liberal (centred on individual rights, the Fiji Indian and Filipino. The sampling rights of the market and reduced state inter- strategy included over-sampling of Māori in vention) versus social democratic (centred order to ensure a strong Indigenous voice on collective rights, addressing inequity, in the data. Participants’ average age was redistribution of resources and increased 52 years, with the youngest being 30 and state intervention). the oldest 64. All participants had a New This paper explores the impact of Zealand Index of Socioeconomic Deprivation the COVID-19 policy response from the for Individuals (NZiDep) score of four or perspective of some of the most vulnerable five (five is maximum deprivation) during New Zealanders. It examines whether the the nationwide COVID-19 Alert Levels 4–3 Government’s response protected the rights 37 lockdown. Twenty-three participants of these low-income New Zealanders and the (85%) were on a government benefit prior participants’ views of the response and their to lockdown. Four participants lost their job advice to the Government about how to deal during lockdown, two of whom commenced with future pandemics. receiving a government benefit. Data were transcribed then coded and Participants lived in a wide range of analysed using content analysis. Each circumstances typical of low-income transcript was coded independently by two communities. The sample includes sole researchers. Analysis involved discussion parents, people seeking asylum or refuge in across the research team identifying key New Zealand and people with disabilities themes. The transcripts of Māori partic- or long-term health conditions (eg, cerebral ipants were independently coded and palsy, epilepsy, depression, arthritis). Some analysed by a Māori researcher (DL).

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 57 www.nzma.org.nz/journal article

“I just don’t think I could live like Results that, but we had, we just had to... Findings are presented below. Quotes are ‘cause it’s the way it was.” (NM, C11) used to illustrate key points. Participant “I would never want to be put in that coding is as follows: M=Māori, NM=non- situation ever again because that was Māori, A=Auckland, C=Christchurch. hard, it was very stressful… when we Impact of the outbreak on people’s need to see our families, they won’t lives open the door for us.” (M, A14) The majority of participants sponta- One Māori participant commented on neously discussed feelings of anxiety related how difficult it was not being able to phys- to COVID-19 infection. In particular, people ically attend tangihanga (funerals) for expressed anxiety around contracting whānau members who passed away during COVID-19 and infecting vulnerable relatives lockdown. Instead, they had to watch tangi- in their bubble. While lockdown restricted hanga on live stream. almost all daily activities besides essential As the lockdown continued, feelings of services, anxiety was felt during these isolation resulted in significant mental essential interactions. It caused some people distress for some individuals. This was to restrict their activities even further: for particularly so for participants with pre-ex- example, not venturing past their letterbox isting mental health conditions such as for the entire lockdown. depression and anxiety, which were exac- “The fear of my partner, with him erbated by the lockdown. All but one of being sick… that was quite scary, you those participants in a solo bubble reported know, with him and how sick he is. feelings of isolation, considerably more Just having that fear of anyone being than those in joint bubbles. These feelings near him or coming near our house worsened as the lockdown and the length of or even just walking up our shared isolation was extended. driveway was enough to scare me.” “I think that with loneliness and (M, C4) depression, that’s what it is. It slowly Most participants expressed sadness creeps in… I think that’s how it is.” and hopelessness surrounding the impacts (NM, C14) of COVID-19 around the world and the “So everyone, they got depressed, we response of overseas leaders to the probably got depressed, everyone pandemic. A few stated they felt extremely that I knew was depressed. So it was distressed. a really really bad thing, ya know… “[It was] like a war really, with an but essential.” (NM, C12) invisible enemy.” (NM, C12) Although they still reported feeling While a minority disengaged from the isolated from loved ones, many long-term news, others found it made them appreciate beneficiaries said they felt more prepared New Zealand more when they compared to cope with a life in lockdown. They noted themselves to people in other countries that that life on the benefit was similar to were faring worse in the pandemic. lockdown, as they did not work and interact Experience of lockdown with many people on a daily basis. Effects on wellbeing “It was a lot easier I think for bene- ficiaries during the lockdown, ‘cause The majority of participants found we’re used to staying at home, ya lockdown had a negative effect on their know?” (NM, A2) mental wellbeing. This included feelings of immense boredom, confinement and Similarly, those who suffered or who were emotional volatility. However, the key effect suffering from loss and trauma noted that identified across most participants was the impact of COVID-19 was less of a concern that of isolation. People felt isolated from in contrast to their other problems. They their whānau, their communities and the expressed being prepared for trauma. day-to-day social interactions they had from “I’m used to it. I’m used to a long and their usual activities that they could no painful journey.” (NM, A7) longer pursue.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 58 www.nzma.org.nz/journal article

Participants who became unemployed as “I was just stressing out, hard a result of the lockdown appeared to suffer out, stressing all day every day, considerable emotional distress due to loss just over food really, especially of financial freedom and the social support after having nothing, no help from they received from their participation in anyone.” (M, A5) paid employment. One participant, who spent their “I had physical support but no lockdown in an institutional care home, emotional or mental health support had an extremely negative experience at all. No psychological support.” of lockdown. They expressed feeling (NM, C6) ill-treated by the staff and distrustful However, a number of participants noted and fearful of the institution. “You don’t positive effects on their wellbeing. These complain unless you’ve got somewhere included having less traffic on the road, else to go” (NM, C12). Although they felt more time to connect with their whānau the restrictions were necessary, they also and an opportunity to reflect on life and be felt negatively towards the Government’s appreciative of things they normally took for communication of the pandemic. They felt granted. unsupported by their institution during lockdown. “It was all up to individuals” “[Lockdown] gave me an opportunity (NM, C12). to think about a lot of things, espe- cially about a lot of others who were Financial effects a lot less fortunate than myself.” The majority of participants stated that (NM, C14) they coped financially during lockdown “I like it actually, spending more and were able to pay their bills and provide time with the family… in a way for themselves and their whānau. This it’s quite pleasing. Yeah, just more was aided by reduced expenditure due interaction, communication with the to the restrictions, alongside the raised kids.” (M, A16) payment across all government benefits and the increased Winter Energy Payment. Most participants felt safe at home. For However, many stated they endured the few who did not, this came from inse- financial stress and had to juggle spending curity in those they shared their bubble to pay the bills. This was the case for five of with, concerns over others’ adherence to the seven Māori participants, considerably COVID-19 restrictions, personality conflicts, more than the proportion of non-Māori. homelessness or failure of the institution Despite the increase to the Winter Energy in which they lived to keep them safe. Payment, this stress usually related to the Although some expressed being previous power bill, which, due to the larger number victims of domestic violence, none disclosed of people staying at home, was more such experience during lockdown—but expensive. The financial stress also influ- participants were not asked directly about enced some participants’ access to medical whether they experienced any domestic services. One participant found they had violence during lockdown. to choose between paying for their medical One participant found themself homeless scripts or paying the bills. for the entire lockdown. They slept in “So we have to [be] careful. 15–20 their car because they did not feel safe in dollars goes like this and we have to their previous place of residence, a shared plan. On the benefit, we can only pay boarding house for previous offenders. They the rent, pay two [cell]phone bills, also lost their job due to lockdown. pay for wifi.” (NM, A3) “My boss never rang me back… I There was a difference between those don’t know what happened. I lost who had been receiving a benefit for a long everything that day.” (M, A5) period and those who had lost their job due They experienced many hardships, to COVID-19. Those who were receiving a including hunger, being cold, loneliness, benefit prior to lockdown displayed financial isolation and stress over not knowing when resilience with comments such as, “we help would come. pretty much just lived as normal” (NM, C15),

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 59 www.nzma.org.nz/journal article

and, “I’m always watching my money” (NM, participants stayed home when they ran out C1). Those who became unemployed due to of inhalers and others relied on self-coping COVID-19 and fell into financial hardship mechanisms for mental distress rather than appeared to struggle a lot more and spoke of seeking external support. a sense of loss over financial freedom they “And you started wondering where had previously taken for granted. do you go? Or who do you see? I “That was my job gone and I had to suppose there’s help-lines and things survive on the [Wage] Subsidy and like that, but I’m very wary of those I had to sign up with the Jobseekers sorts of things.” (NM, C14) [benefit] as well because what money The participants who accessed health I was getting from subsidy wasn’t services reported mixed feedback on their paying my power and also my food experiences. Although some found it easier and my rent. It wasn’t enough there.” to access teleconsults and prescriptions (M, A14) sent electronically, others felt distressed Access to food when seeking support over the phone. One Despite many people reporting that they parti­cipant found accessing healthcare coped financially, the majority of partici- for their partner with a chronic health pants accessed food banks during lockdown. condition difficult because they had no Two participants reported not being able to personal vehicle, because of changes in the access food. Over half of participants made bus timetable and also because of perceived comments expressing stress when accessing restrictions placed on tertiary healthcare. food. Participants noted the difficulty of “There was a few scared moments accessing supermarkets, such as waiting because with the lockdown, the in long lines that deterred them from access to the hospital wasn’t that shopping. easy… [my partner] needs oxygen “Ya have to shop differently, ya know? now and then, so it was pretty hard Which makes it hard, especially when to try and get an ambulance.” (M, C4) you’re not too well yourself and you The majority of participants used social need proper food.” (NM, A2) services such as the City Mission, WINZ, Some participants spoke of their financial refugee trusts and food banks. Participants restrictions and increased food prices. This who used these services commented they resulted in some participants being unable had been provided with accommodation, to access healthy food or using up their financial aid, food and employment. Social existing food. services also provided many with a sense of community, which they felt significantly “[I] always had access to food but didn’t disconnected from during lockdown. always have access to good food.” (NM, C6) “I was a real lost cause when I Some participants expressed distress over arrived [at the City Mission]. You others hoarding supplies, whereas others know, homeless and everything, had found they were reassured by supermarkets lost everything really… if it wasn’t for being well stocked. the Mission, I don’t know where my Access to services life would be right now. I mean it’s Most participants accessed some form of like a big family here.” (NM, C2) service during lockdown. This included GP Three participants felt they were treated teleconsults (a largely new development disrespectfully when accessing WINZ. One of in response to COVID-19), pharmacists, these participants was homeless throughout psychiatrists and other social services. A key the lockdown therefore their main concern concern for the participants was the lack of was having access to food. Desperation was information around availability of health a barrier to being able to receive help and and social services. Some also expressed prevented them from wanting to reach out unwillingness to access services due to fear to WINZ again. of contracting COVID-19, fear of burdening “WINZ just hung up on me and the health system and stigma around the declined me… it might have been me, use of such services. For example, some my voice. I was hungry. I was getting

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 60 www.nzma.org.nz/journal article

annoyed because they kept asking all “[I’d] give myself lectures saying, these questions about work and work ‘come on, come on, this is going to be and work and I was hungry.” (M, A5) okay’.” (NM, C1) Another participant, in a better financial Participants mentioned using exercise, position, persisted, after which they were household chores and indoor hobbies to able to get the help they needed. occupy themselves. Many participants “She made me feel that I shouldn’t mentioned their ability to get out of the deserve to have a food grant… house to go for a walk helped them to cope when I hung up, I said, ‘I’m going to with feelings of isolation and confinement. have to ring back and get someone “I had to go for my walk… if I got to listen to the conversation I had stuck... not [able to] get some vitamin with the lady’… the big boss lady (of C [sic], some fresh air into me, I don’t WINZ) phoned me back and apolo- know where I’d be today.” (NM, C8) gised.” (M, A14) The majority of participants found staying Two participants experienced stress connected to their whānau and commu- surrounding the delay in visa applications or nities was important to support themselves. effects of unemployment on their eligibility Due to the social restriction of lockdown, to live in New Zealand. technology played a key role in connecting Some participants had difficulty accessing people to their sources of support. Partici- food banks because they did not realise food pants mentioned calling and texting their banks were an essential service or because loved ones frequently. This was particularly of the rules imposed by the food banks. Due important for those participants in solo to having no physical address or ability bubbles or who had loved ones overseas. to drive his car, one participant who was “Ringing my mum yeah every day. I homeless during lockdown could not access think that was the biggest, it was just these services. They had to survive mainly having that support on the phone.” off bread and water for the entire lockdown. (M, C4) “I had no money. I had a vehicle but Technology was key in enabling no gas. I tried to walk there [to a community groups to adapt to the lockdown. church supplying food parcels] and Religious groups were able to stream their get food but they wouldn’t let me. services, and social support groups, like the Needed a vehicle, stay in the vehicle. City Mission, provided chat forums and daily They put [the food] in themselves… activities for their communities. They said you can’t come in, you “Our church started having online have to be at home… I didn’t have an church… they were really good for address for them to drop it off to.” getting people through.” (NM, C10) (M, A5) Some participants commented that Coping strategies connecting through technology was not the Although most participants experi- same, stating they missed physical human enced mental distress, very few mentioned interaction. accessing mental health services. The “I don’t know what that is, but it’s majority of participants coped using like an emotion aye, a feeling, a vibe... self-help techniques. These included A human thing. It’s like when you methods such as self-talk and relaxation meet someone and you shake hands... techniques to calm themselves when feeling and I don’t know what it is, but you overwhelmed. The self-talk techniques were get this warm glow that comes over described by participants with comments you.” (NM, C14) such as: Although not specifically asked, two “‘calm down, now calm down’. So participants noted they did not have access I’d sort of calm myself down and I‘d to the internet during the lockdown. For think, ‘no, it won’t be for long, ya these participants it was a key concern and know, just do as they say’. So I did.” a barrier to connecting with whānau and (NM, C7) accessing education.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 61 www.nzma.org.nz/journal article

Opinions of the Government’s However, a few participants did report response breaking restrictions for mental health reasons: for example, visiting friends or Participants were overwhelmingly whānau to socialise. Most expressed guilt positive about the Government’s response and an understanding that this was wrong. to COVID-19. Most expressed trust in the But they felt they could no longer cope. Government and appreciation for its rapid and effective handling of the pandemic. “I “You try not to break the rules, but, think they did a top job” (NM, C14). you know, I live on my own. So yeah. So sometimes I went and visited Participants were quick to contrast New friends with a mask on ‘cause I’d just Zealand’s position in the global pandemic had enough, you know?” (NM, A7) with other countries. This helped them reflect on the lockdown in a positive Only one participant reported not manner. Participants reflected that New adhering to restrictions, due to their distrust Zealand’s response ‘led the world stage’. in the Government. Many spoke as a member of the ‘team of five The vast majority of participants were million’. extremely happy with the communica- “We got to help the health of each tions from the Government through the other. Keep yourself healthy and then lockdown. Most participants engaged with look after each other.” (M, A15) and appreciated having the daily 1pm televised media briefings and found the Participants appeared to be reluctant to communication clear and up to date. criticise the Government even when there were breaches at the border. Some partici- “I did follow everyday… everyday at pants expressed anger towards politicians one o’clock. That helped… that helped who undermined the Government’s deci- me understand it.” (NM, C11) sions. “Shut up, Simon [the leader of the Participants for whom English was not opposition]” (NM, C2). their first language also reported that the “I think we’re very blessed in New television subtitles were sufficient to gather Zealand. You know, compared to the a clear picture of the situation. A couple of other countries, I think we’re very participants made negative comments. “Too very lucky.” (NM, C1) much… not COVID [again], I’m sick of the news” (NM, A1). Many participants specifically praised Prime Minister Jacinda Ardern and Direc- Advice for future lockdowns or tor-General of Health Dr Ashley Bloomfield. pandemics Positive comments centred around valuing The vast majority of participants were individuals’ lives, feeling truly cared about positive about the Government’s response to and the calm, collected and personable the pandemic. Hence, when asked whether manner displayed by Ardern and they had any advice for future pandemics, Bloomfield. the majority of participants called for the “I really [want to stress] the thank- Government to continue with their current fulness, you know, and gratefulness approach. for what Jacinda Ardern has done. “Exactly how they did this. They did For me, my family, my friends. excellent.” (NM, C15) Magnificent. So I’m very, very grateful Over a quarter of participants expressed for that.” (M, A14) concern over a resurgence of COVID-19 in Participants overwhelmingly viewed the the community because of border insecurity. Government’s restrictions as necessary to Most of these participants wanted stricter prevent the spread of the virus, and they border procedures and testing prior to reported adhering to them to keep others in travelling to New Zealand. One participant their bubbles and the community safe. called for complete border closure. Others “If we keep to the social distance... requested more financial and practical then maybe we could come out support. stronger… the whole of New Zealand “I’ll first say the Government should could come out stronger.” (M, A14) provide the basics first. Food, gloves,

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 62 www.nzma.org.nz/journal article

masks. Because how can anyone be pants stating they coped financially. Food safe without these?” (NM, A3) insecurity was common. Clearly the pre-ex- Some participants expressed gratitude isting and increased government welfare at the increase in benefits over the winter payments were critical: yet this research months and advised for them to stay high suggests they were still not sufficient. Parti­ throughout the year. A few participants cipants who lost their jobs during lockdown suggested the need for more social services reported struggling more than beneficiaries, and mental health support during lockdown. both with the financial hardship and their sense of loss. This finding emphasises the “For people like me, if somebody [did] gap in the standard of living between bene- ring up say, ‘how ya feeling what ya ficiaries and those in employment. It also need? You need help?’, that would underscores the importance of government make me happy.” (NM, C5) and employer commitment to maintaining One participant, who stood out in their people in work during a public health crisis, negative opinion towards the response, including the importance of the COVID-19 expressed deep distrust in the Government. Wage Subsidy. “I’d sack half of what’s in Parliament”. They Participants were largely resourceful and called the communication surrounding resilient. They coped with lockdown by using COVID-19 ‘propagram’, claiming that “[you] technology, self-help techniques, health and can’t trust them as far as you can kick social services and support from whānau and [them]” (NM, A1). their communities. New Zealand’s lockdown allowed people to go outside for exercise, Discussion which helped manage their isolation and Life during the COVID-19 lockdown confinement. This research suggests that, if was challenging for the low-income New possible, people should be able to go outside Zealanders in this study. Participants were during lockdown. While technology may fearful of the virus. Lockdown impacted not totally bridge lockdown’s social divide, negatively on their mental health, as it is a key resource that the vast majority of reported elsewhere.1,3,6 Feelings of isolation participants were able to use well. However, were common. Beneficiaries reported solutions are needed for those without feeling more prepared to cope with internet access. lockdown than others because they are ‘used Most participants accessed some to staying at home’, a finding that high- form of social or health service during lights the social isolation that beneficiaries lockdown—unsurprisingly, given partic- 17 routinely endure. A number of people ipant recruitment was through community noted ‘silver linings’ to the lockdown, as services. Some participants expressed 3 identified by Every-Palmer et al. concern about the lack of information Most people felt safe at home, except about the available services and a reluc- for one homeless person and one in insti- tance to use services for a range of reasons, tutional care, two key areas for critical including infection fear and stigma (previ- attention. Although some participants had ously reported stressors during quarantine)1 previously experienced domestic violence, and not wanting to burden the system. they did not report further incidents during Government social services were insuffi- lockdown. However, Every-Palmer et al cient for participants’ basic needs, and it noted elevated levels of domestic violence was necessary for these critical functions to during lockdown.3 Safety at home, or having be augmented by charitable social services, a home at all, are key concerns. The Govern- including in the provision of food. There ment’s ability to house many of the homeless were a number of barriers to social service during this crisis proves that solutions can provision, including access, delays, rules be found to seemingly intractable problems. that were difficult to negotiate and disre- It is hoped that effective solutions can be spect from staff. It is concerning that some maintained and more easily found in future participants were already experiencing without being prompted by a crisis. such severe hardship that the pandemic Financial stress was common, particularly lockdown was perceived as having only a for Māori, despite the majority of partici- moderate impact on their lives.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 63 www.nzma.org.nz/journal article

In relation to health services, some was effective and timely. However, not participants found the innovation of tele- all low-income people access services. consults and electronic prescriptions Therefore, it is likely that these results easier, but others found them distressing. underestimate the impacts of lockdown on Previous research suggests that telehealth those unable to access services. Over-sam- is not suitable for all people or all issues.8 pling of Māori participants ensured a One participant found it hard to access stronger Indigenous voice in the data. Given tertiary healthcare for her partner with a the over-representation of Māori among chronic health condition. At a time of crisis low-income New Zealanders, it is likely in the health system, these low-income that Māori bore a heavier burden than the New Zealanders were able to access New non-Māori population. Although this paper Zealand’s publicly funded, largely free presents a study from New Zealand, this health services and utilise innovations, research has valuable insights for other albeit not without some challenges. This jurisdictions with marginalised populations. may not be the case in more neo-liberal This research demonstrates the impor- countries with privatised healthcare, such as tance of eliciting the views of society’s most the US. vulnerable citizens during a public health These low-income New Zealanders emergency. Undertaking such research can were overwhelmingly positive about the yield valuable information for strength- Government’s response to COVID-19. They ening responses and better meeting the expressed high levels of trust, adhered needs of vulnerable people during and to restrictions and willingly participated between periods of crisis. Future research as part of the team of five million New including Indigenous people, children and Zealanders working together to beat the young people and people not connected to virus. This is a particularly reassuring social services is needed. Also, this research finding from some of the most vulnerable suggests the needs of those in institutional New Zealanders. So too is the participants’ care and those who become unemployed advice for the Government to take the same as a result of pandemics need to be better approach in any future pandemic. The study understood. Follow-up studies are required findings, and the specific advice participants to explore the long-term impacts. had for the Government, suggest the need If the greatness of a nation can be for stricter border controls, further financial judged by how it treats its most vulnerable and practical support for low-income New members, Aotearoa New Zealand appears Zealanders and specific focus on the needs to have shown considerable strength during of particularly vulnerable people, such as the COVID-19 pandemic, at least according the homeless and those in institutional care. to the participants in this research. The keys This research was undertaken rela- to this success were the policy of going hard tively early in the global pandemic and and going early, the more social democratic therefore presents the immediate impacts policy direction and the compassionate of lockdown. It was conceived, funded and clearly communicated leadership. and developed during the lockdown. Due Research on life during lockdown for people to New Zealand’s success in controlling receiving low-incomes is essential to guide the virus, it was possible to undertake the future responses to pandemics or other research face-to-face soon after the end of emergencies. Researching the experience of the initial lockdown, which reduced the vulnerable members of society during and risk of recall bias but precluded assessment immediately after crises, such as pandemics, of the longer-term impacts of COVID-19. is critical to ensuring that their rights as Recruiting participants through community citizens are protected and that inequities in organisations, particularly the City Mission, the impact of such crises are mitigated.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 64 www.nzma.org.nz/journal article

Competing interests: Nil. Acknowledgements: The COVID-19 pandemic and the Government’s subsequent response has had ongoing impacts on the lives of all New Zealanders. The researchers gratefully thank all the parti­ cipants who shared their time and personal stories in the middle of ongoing difficult times. We also thank all community group leaders and coordinators who helped us recruit the participants. This research was funded by a grant from the Health Research Council of New Zealand, Grant No 20/1066. Author information: Kimberley Choi: Health, Environment & Infection Research Unit (HEIRU), Department of Public Health, University of Otago, Wellington, New Zealand. Namratha Giridharan: Health, Environment & Infection Research Unit (HEIRU), Department of Public Health, University of Otago, Wellington, New Zealand. Abigail Cartmell: Health, Environment & Infection Research Unit (HEIRU), Department of Public Health, University of Otago, Wellington, New Zealand. Dominique Lum: Health, Environment & Infection Research Unit (HEIRU), Department of Public Health, University of Otago, Wellington, New Zealand. Louise Signal: Health, Environment & Infection Research Unit (HEIRU), Department of Public Health, University of Otago, Wellington, New Zealand. Viliami Puloka: Health, Environment & Infection Research Unit (HEIRU), Department of Public Health, University of Otago, Wellington, New Zealand. Rose Crossin: Department of Population Health, University of Otago, Christchurch, New Zealand. Lesley Gray: Department of Primary Health Care and General Practice, University of Otago, New Zealand. Cheryl Davies: Health, Environment & Infection Research Unit (HEIRU), Department of Public Health, University of Otago, Wellington, New Zealand; Tu Kotahi Māori Asthma Trust, 7-9 Barnes Street, Seaview, Lower Hutt, Wellington, New Zealand. Michael Baker: Health, Environment & Infection Research Unit (HEIRU), Department of Public Health, University of Otago, Wellington, New Zealand. Amanda Kvalsvig: Health, Environment & Infection Research Unit (HEIRU), Department of Public Health, University of Otago, Wellington, New Zealand. Corresponding author: Louise Signal: Health, Environment & Infection Research Unit (HEIRU), Department of Public Health, University of Otago, Wellington, PO Box 7343, Wellington South 6242, New Zealand [email protected] URL: www.nzma.org.nz/journal-articles/life-during-lockdown-a-qualitative-study-of-low-income- new-zealanders-experience-during-the-covid-19-pandemic-open-access

REFERENCES 1. Brooks SK, Webster and Pakistan: results of 4. Giebel C, Lord K, Cooper RK, Smith LE, et al. The pre-COVID and COVID-19 C, et al. A UK survey of psychological impact of lockdown stakeholder COVID-19 related social quarantine and how to engagements. BMJ Glob support closures and their reduce it: rapid review of 2020;5(8):e003042. effects on older people, the evidence. The Lancet 3. Every-Palmer S, Jenkins people with dementia, 2020;395(10227):912-20. M, Gendall P, et al. and carers. International 2. Ahmed SAS, Ajisola M, Psychological distress, Journal of Geriatric Psychi- Azeem K, et al. Impact anxiety, family violence, atry 2021;36(3):393-402. of the societal response suicidality, and wellbeing 5. Hamadani JD, Hasan MI, to COVID-19 on access in New Zealand during Baldi AJ, et al. Immediate to healthcare for the COVID-19 lockdown: A impact of stay-at-home non-COVID-19 health issues cross-sectional study. PLoS orders to control COVID-19 in slum communities of ONE 2020;15(11):e0241658. transmission on socioeco- Bangladesh, Kenya, Nigeria nomic conditions, food

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 65 www.nzma.org.nz/journal article

insecurity, mental health, Social isolation and mental 21. Signal L, Jenkin G, Foug- and intimate partner health. Journal of Clinical ere G, et al. The politics violence in Bangladeshi Nursing 2020(https://doi. of health promotion. women and their families: org/10.1111/jocn.15290). In: Signal L, Ratima an interrupted time series. 13. Smith JA, Judd J. COVID- M, eds. Promoting The Lancet Global Health 19: vulnerability and Health in Aotearoa New 2020;8(11):e1380-e89. the power of privilege Zealand. Dunedin: Otago 6. Marroquín B, Vine V, in a pandemic. Health University Press 2015. Morgan R. Mental health Promotion Journal of 22. Boston J. Redesigning during the COVID-19 Australia 2020;31(2):158. the welfare state. Policy pandemic: Effects of 14. Grantz KH, Rane MS, Quarterly 2019;15(1) stay-at-home policies, social Salje H, et al. Disparities 23. Ministry of Social Devel- distancing behavior, and in influenza mortality opment. Evidence Brief: social resources. Psychiatry and transmission related The impact of COVID-19 Research 2020;293:113419. to sociodemographic on benefit receipt rates 7. Peloso RM, Pini NIP, factors within Chicago in historic perspective. Sundfeld Neto D, et al. in the pandemic of 1918. Wellington: Ministry How does the quar- Proceedings of the National of Social Development antine resulting from Academy of Sciences Research Archives, 2020. COVID-19 impact dental 2016;113(48):13839-44. 24. McLeod J. The New Zealand appointments and patient 15. Rutter PD, Mytton OT, Mak cause report: Shape anxiety levels? Brazilian M, et al. Socio-economic of the Charity Sector. Oral Research 2020;34 disparities in mortality Wellington: JBWere, 2017. 8. Imlach F, McKinlay E, due to pandemic influenza 25. Franks J. Coronavirus: Middleton L, et al. Tele- in England. International Where to get help, food health consultations in Journal of Public Health parcels during Covid-19 general practice during 2012;57(4):745-50. alert level 3: Stuff http:// a pandemic lockdown: 16. Simonsen L, Spreeu- www.stuff.co.nz; 2020 survey and interviews on wenberg P, Lustig R, [accessed 2 June 2021]. patient experiences and et al. Global mortality 26. Jernigan DB, CDC COVID-19 preferences. BMC Family estimates for the 2009 Response Team. Update: Practice 2020;21(1):1-14. Influenza Pandemic from public health response to 9. Imlach F, McKinlay E, the GLaMOR project: a the coronavirus disease Kennedy J, et al. Seek- modeling study. PLoS Med 2019 outbreak—United ing healthcare during 2013;10(11):e1001558. States, February 24, lockdown: challenges, 17. Stewart MJ, Makwarimba 2020. Morbidity and opportunities and lessons E, Reutter LI, et al. Poverty, Mortality Weekly Report for the future. International sense of belonging and 2020;69(8):216. Journal of Health Policy and experiences of social 27. Ministry of Health. New Management April 13, 2021. isolation. Journal of Poverty Zealand Influenza Pandem- 10. Elers C, Jayan P, Elers P, 2009;13(2):173-95. ic Plan: A framework for et al. Negotiating Health 18. Martin H, Tokalau T. NZ action (2nd ed). Wellington: Amidst COVID-19 Lock- marks 100 days since last Ministry of Health, 2017. down in Low-income community transmission 28. Baker MG, Kvalsvig A, Communities in Aotearoa Covid-19 case: Stuff; 2020 Verrall AJ, et al. New New Zealand. Health [accessed 2 June 2021]. Zealand’s COVID-19 Communication 19. Summers J, Cheng H-Y, elimination strategy. Med J 2021;36(1):109-15. Lin H-H, et al. Potential Aust 2020;213(198):10.5694. 11. Acheson D. Public health in lessons from the Taiwan 29. Government NZ. History of England: The report of the and New Zealand health the COVID-19 Alert System Committee of Inquiry into responses to the COVID-19 https://covid19.govt.nz/ the Future Development of pandemic. The Lancet alert-levels-and-updates/ the Public Health Function. Regional Health-Western history-of-the-covid-19- London: Committee of Pacific 2020:100044. alert-system/ Wellington: Inquiry into the Future 20. Carpinter P. History of New Zealand Govern- Development of the Public the Welfare State in New ment; 2021 [accessed Health Function 1988. Zealand. Wellington: New 2 June 2021]. 12. Usher K, Bhullar N, Jackson Zealand Treasury, 2012. 30. Ardern J. PM Address - D. Life in the pandemic:

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 66 www.nzma.org.nz/journal article

Covid-19 Update: Beehive beehive.govt.nz/release/ [accessed 2 June 2021]. Speeches, 21 March. major-steps-taken-protect- 36. Forbes S. Coronavirus: http://www.beehive. new-zealanders-covid-19 Salvation Army report govt.nz Wellington; 2020 2020 [accessed 2 highlights social impact of [accessed 2 June 2021]. June 2021]. Covid-19 lockdown https:// 31. Franks J. Dr Tristram 34. Ministry of Social Devel- www.stuff.co.nz: Stuff; 2020 Ingham, the brains opment. Benefit Fact [accessed 2 June 2021]. behind the Covid-19 Sheets: Snapshot-June 37. Salmond C, Crampton ‘bubble’ https://www.stuff. 2020 Quarter https:// P, King P, et al. NZiDep: co.nz/national/health/ www.msd.govt.nz/ a New Zealand index of coronavirus/122163720/ about-msd-and-our-work/ socioeconomic depri- dr-tristram-ingham- publications-resources/ vation for individuals. the-brains-behind-the- statistics/benefit/index.html Social Science & Medicine covid19-bubble: Stuff; 2020 Wellington: Ministry of 2006;62(6):1474-85. [accessed 2 June 2021]. Social Development; 2020 38. Bambra C, Fox D, [accessed 2 June 2021]. 32. Khalil S. Coronavirus: Scott-Samuel A. Towards How New Zealand 35. Franks J. Coronavirus: a politics of health. Health relied on science and Demand for food banks Promotion International empathy http://www.bbc. double pre-Covid levels, 2005;20(2):187-93. doi: com/news/: BBC; 2020 and still growing https:// 10.1093/heapro/dah608. [accessed 2 June 2021]. www.stuff.co.nz/ 39. Signal L. The politics of national/122093506/ 33. Ardren J. Major steps health promotion: insights coronavirus-demand- taken to protect New from political theory. for-food-banks-double- Zealanders from COVID- Health Promotion Interna- precovid-levels-and- 19. Press statement 14 tional 1998;13(3):257-63. March 2020. https://www. still-growing: Stuff; 2020

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 67 www.nzma.org.nz/journal article

Making sure the New Zealand border is not our Achilles heel: repeated cross-sectional COVID-19 surveys in primary care Kyle Eggleton, Nam Bui, Felicity Goodyear-Smith

ABSTRACT AIM: Quick COVID-19 Surveys are an international collaboration designed to rapidly analyse and disseminate a primary care perspective on the pandemic and associated health response. In this paper we present results from surveys relating to opening the New Zealand border. METHOD: Three surveys were distributed to primary care practices between May and December 2020. A range of primary care member organisations distributed the survey augmented by snowballing. Quantitative data were analysed using descriptive statistics and qualitative data through an inductive process and grouped into themes. RESULTS: Respondents became increasingly supportive of opening a trans-Tasman border but not internationally. Two broad themes were evident: (1) making sure that the border is not an Achilles heel and (2) effective strategies to reduce local transmission. These themes highlight primary care’s concerns around management of the border and the management of local spread respectively. CONCLUSION: The results highlight concerns around border control from a primary care perspective. The border control issues raised by primary care have proven to be prophetic at times and reflect the role that primary care has as observers of society. The survey mechanism provides a template for rapidly eliciting a primary care voice for future health issues.

s is well-known, in December 2019 Our surveys were supported by a number the Chinese authorities advised the of primary care organisations, including AWorld Health Organization of cases the Royal New Zealand College of General of pneumonia of unknown cause, originat- Practice (RNZCGP), the Royal New Zealand ing in Wuhan, Hubei province. This was sub- College of Urgent Care (RNZCUC), General sequently identified as caused by a severe Practice New Zealand (GPNZ), the Rural acute respiratory syndrome coronavirus General Practice Network (RGPN) and the 2 (named SARS-CoV-2), and a pandemic of Practice Managers and Administrators coronavirus 2019 (COVID-19) was declared Association of New Zealand (PMAANZ). on 11 March 2020. Summary findings were rapidly available In parallel with researchers from the after each survey closed and disseminated United States,1 Canada2 and Australia,3 from to key policymakers, including the Chief May 2020 we conducted regular (fortnightly Science Adviser for the Ministry of Health or monthly) surveys of New Zealand general and the Director-General of Health, and the practices on the impact of the pandemic on New Zealand media. primary care.4 Each fortnightly or monthly The participants were not intended to be survey was termed a ‘Series’ and sequen- representative of all New Zealand primary tially labelled. care practices. As indicated above, there are

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 68 www.nzma.org.nz/journal article

many organisations that represent aspects Participant recruitment of New Zealand primary care, as well as the Links to the survey were disseminated via 30 primary health organisations (PHOs) to the RNZCGP, the RNZCUC, GPNZ, RGPN, the which practices may belong, but no single PMAANZ, the New Zealand Medical Associ- organisation speaks for all. The overall aim ation, several PHOs not aligned with GPNZ of the project was to provide an opportunity and Facebook groups, including GPs Down for the voices of primary care practice staff Under, New Zealand Women in Medicine, to be heard by policymakers. GPs for GPs and Health Forum NZ. Respon- The first line of defence against COVID-19 dents could also sign up for alerts to be sent is border control, and once elimination has each new survey link. A snowballing method been achieved, ongoing primary healthcare was used—participants are invited to pass approaches (public health and primary the link on to their primary care colleagues. care), particularly COVID-19 testing and Should they receive invites from multiple contact-tracing, are required to prevent or sources, they were asked to complete the address border breaches from incoming survey only once. travellers.5 Survey design Our borders were restricted on 16 March The survey in each series was based on the 2020, and then closed to all but New Zealand United States core questions plus one to three citizens and residents on 20 March. By 23 additional (‘flash’) questions determined by May 2020 (Series 1) New Zealand had moved local contexts. Both quantitative responses, down to Alert Level 2 and the curve had flat- such as Likert scales and free-text options, tened. There had been 1,473 community and were included. Demographics included 31 managed isolation and quaratine (MIQ) their professional role (doctor, nurse, nurse cases, with 21 deceased. By 5 June (Series 2) paractitioner, practice maanger) and the size there had been no more community cases and nature of the practice (general practice, for two weeks, and the country moved urgent care, rural or urban). down to Alert Level 1 on 9 June. In August a community cluster from a border breach Survey delivery The participant information sheet led to the Alert Level being raised (Auckland was accessed at the start of the survey. to 3, the rest of the country to 2), until it Completing the survey was implicit consent. returned to Alert Level 1 on 8 October. By 9 Surveys were launched by the United States December (Series 10) community spread had team at the Larry A Green Center using again ceased, and all cases were in MIQ.6 SurveyMonkey. No stored information is The aim of this paper is to present the identifiable. Secure results were shared in findings of responses from New Zealand a password-protected file using Filelocker. primary care doctors, nurses and practice All analyses of New Zealand data were managers regarding border control issues conducted by the New Zealand research on 23 May, 5 June and 9 December 2020. team. Questions analysed in this paper Method Series 1, Series 2 and Series 10 surveys The overall project uses a repeated included the flash question, ‘Do you think cross-sectional survey design with a mixture it is safe to open up the country?’ In Series of quantitative and qualitative data. 1 the options were ‘No’, ‘Yes’, ‘It depends’, Ethical approval and ‘Unsure’. In Series 2 and Series 10, the Approval was granted by the University choice of answers was refined to ‘No’, ‘Yes of Auckland Human Participants Ethics for domestic travel’, ‘Yes for trans-Tasman Committee on 11/05/20 for three years. travel’, ‘Yes for Pacific Islands travel’ and Reference 024659. ‘Yes for all international travel’. In all cases Participants a free-text box was available for comments regarding the reasons for their response. Participants were doctors, nurses, nurse practitioners and practice managers The surveys started after the New Zealand working in New Zealand general practice, State of Emergency had been lifted on 14 May urgent care or other primary care settings. and community spread had been virtually Secondary care clinicians were excluded. eliminated. Series 1 was launched when New

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 69 www.nzma.org.nz/journal article

Zealand was under Alert Level 2 (23 May); allowing COVID-19 to re-enter the country. Series 2 when stepping down from Alert There were three broad approaches, Level 2 to 1 (5 June); and Series 10 after New or sub-themes, to managing the border Zealand had been under Alert Level 1 for suggested by participants: some weeks, following a second community • stopping it before the border outbreak in August (9 December). At Alert • strengthening border controls Level 1, the disease is contained in New Zealand but uncontrolled overseas, with • opening with restrictions. sporadic imported cases and possible occa- ‘Stopping it before the border’ related to sional isolated community transmission. Key participants’ concerns that there is too much events are shown in Figure 1. COVID-19 overseas or that global control Analyses needs to occur first before New Zealand can allow people to enter: for example, “Inter- The quantitative responses were analysed national situation still unstable, NZ locally using descriptive statistics. The free-text stable” [GP participant]. Often participants responses to the question about opening expressed fear, uncertainty or anxiety that up the border in the three surveys were our health system would not cope with collated and coded independently by two another wave of COVID-19: “We are not ready researchers (KE and FG). An inductive for the second wave” [GP participant]. The content analysis approach was taken experiences from the first wave of COVID-19 to coding with text categorised and in New Zealand meant that participants then organised into themes through an would prefer to avoid further lockdowns, abstraction process. Differences in coding maintain a strict border and wait until vacci- were discussed between KE and FG before nation enabled more open travel. finalising the themes. “Too risky to lose the gains we have made. Not sure the country could Results cope with a level 4 lockdown again.” There were 170, 153 and 64 participants [GP participant] in Series 1, Series 2 and Series 10 respec- This latter quote, pertaining to losing the tively (Table 1), with free-text responses to gains, related to other concepts of sacrifice the question made in 100%, 65% and 30% of or suffering that led to greater outcomes, their responses. and that opening the border would undo the Quantitative responses to the question hard work done by so many people. about whether it was safe to open up the “The risk of bringing a covid case to country are shown in Table 2. Free-text NZ is too high. It will jeopardise all responses to the option ‘It depends’ in Series we have sacrificed and achieved so 1 led to qualifying ‘Yes’ with ‘domestic’, far.” [Practice manager participant] ’trans-Tasman’ or ‘international travel’ in Series 2, and in Series 10 a Pacific Islands Concerns around the permeability of the option was also added. It can be seen that border led to calls for greater strengthening by December 2020 just over a quarter were of border controls for returning residents comfortable to open up to Australia and 42% and border workers. Air crew and people to the Pacific, but none wished to open the entering on special work visas were iden- border further. tified as risks. The importance of strict quarantine procedures for everyone, as well During analysis of the qualitative data, as ongoing monitoring of returnees after two main themes were identified: (1) making leaving quarantine, was highlighted. sure that the border is not an Achilles heel and (2) effective strategies to reduce local “I do not believe quarantine or self transmission. isolation is monitored appropri- ately in Auckland. There are many Making sure that the border is anecdotal instances… of the laxity of not an Achilles heel Auckland quarantine. Is close ongoing Participants were uniformly opposed monitoring of these people occurring to opening up the country to interna- after they leave quarantine/isolation?” tional travel (Table 2). The border was [Practice manager participant] generally seen as a significant risk in

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 70 www.nzma.org.nz/journal article

Figure 1: Key events relating to New Zealand border controls and social restrictions.

Data extracted from NZ Doctor’s timeline (https://www.nzdoctor.co.nz/timeline-coronavirus) and the history of COVID-19 Alert System on covid19.govt (https://covid19.govt.nz/alert-system/history-of-the-covid-19-alert-system/).

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 71 www.nzma.org.nz/journal article

Table 1: Participants in Series 1, 2 and 10.

Series 1 Series 2 Series 10

Total Qualitative Total Qualitative Total Qualitative responses responses responses GPs and 123 (72%) 121 (71%) 85 (55%) 52 (34%) 50 (78%) 14 (22%) urgent care doctors

Nurse 2 (1%) 2 (1%) 0 0 0 0 practitioners

Practice 21 (12%) 21 (12%) 16 (11%) 9 (6%) 7 (11%) 3 (5%) nurses

Practice 27 (16%) 27 (16%) 58 (40%) 38 (25%) 7 (11%) 2 (3%) managers

Total* 170 (100%) 170 (100%) 153 (100%) 99 (65%) 64 (100%) 19 (30%)

*There are several respondents who indicate multiple roles.

Table 2: Responses to, ‘Do you think it is safe to open up the country?’

Series 1 Series 2 Series 3 No 71 (42%) 18 (12%) 10 (16%)

Yes 23 (14%) - -

It depends… 68 (40%) - -

Unsure 7 (4%) - -

Yes, domestic - 116 (76%) 47 (73%)

Yes, trans-Tasman - 15 (10%) 17 (27%)

Yes, Pacific Islands - - 27 (42%)

Yes, international - 4 (2%) 0

Total 170 (100%) 153 (100%) 64

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 72 www.nzma.org.nz/journal article

The risks of the border were identified by “We need to be clear about the effects one participant as being the Achilles heel of going to Level 2, particularly in of New Zealand’s elimination strategy: “the opening bars to ascertain if that will border is the Achilles heel of the plan for flush out further cases and possibly elimination failing” [GP participant]. lead to clusters as has happened in Despite proposals to open travel bubbles overseas countries with similar low with Australia and Pacific nations, the to zero numbers at the time. We need majority of participants were hesitant and 2 x 2 weeks of zero cases to be certain wanted to see restrictions in place before because of the incubation/ duration of opening. Key to this was ensuring that there infection period.” [Practice manager was adequate control in Australia before a participant] travel bubble could be considered. Ongoing Workplaces and general practice waiting sporadic cases in Australia meant a general rooms were potential sites where ongoing reluctance to allow travel, and, when transmission could occur and participants combined with the fatigue caused by lock- were reluctant to see a delayed spike in downs that many of the participants were cases occurring: “We don’t want a delayed seeing in the public, this meant that the spike” [GP participant]. Of concern to a risks of further COVID-19 cases arising from couple of participants was the observation Australia were perceived as being too high. that New Zealanders had become quite “I’d like to see a trans-Tasman/Pacific complacent towards COVID-19 and this bubble, but with ongoing new infec- could lead to further outbreaks. tions in Australia, and a significant “Those of us who have worked relaxation/fatigue with lockdown, throughout can see that so many not sure this will be achieved.” [GP people have relaxed their stan- participant] dards already, so we could have an A further concern for at least one partic- outbreak quite easily.” [Practice ipant was the prospect of New Zealanders manager participant] spreading COVID-19 into the Pacific and the Effective tracing and testing individuals impact that might occur on a fragile health was deemed a critical component of system. reducing local transmission prior to border “Would be scared of NZ taking it opening. One important element that into Pacific Islands after measles related to this was gold standard contact problems.” [GP participant] tracing: for example, “Contact tracing needs to be perfect” [Practice manager Although overseas travel within bubbles participant]. was seen as potentially fraught, a few partic- ipants in Series 2 felt that it was necessary Other elements included ongoing surveil- to open up travel in order to protect the lance testing and strict guidelines on when economy. to self-isolate and get tested. One partic- ipant identified the need for adequate Effective strategies to reduce local provision of sick leave to enable people to transmission self-isolate and not feel pressured to come This theme related to three intercon- into work. nected areas, or sub-themes, of eliminating “Government leadership to ensure COVID-19 in the community before border workers/population supported to opening could be considered: self-isolate with even minor respi- • community control ratory symptoms (eg, extended • tracing and testing individuals access to sick-leave).” [GP • vaccinating population. participant] Participants’ views of community control The final sub-theme of effective strategies were around ensuring that COVID-19 was vaccination. Participants did not see was effectively eliminated, primarily by any hope on the horizon with opening up being certain that adequate time occurred the border until an effective vaccination between decisions around Alert Level programme had been rolled-out. Opening changes. the border in the absence of vaccinations

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 73 www.nzma.org.nz/journal article

would mean that the sacrifices would be for practices to reflect on the sacrifices that nothing. have occurred to eliminate COVID-19. “I would expect a vaccine and Participants viewed the sacrifice in fragile a better funded and structured terms, reflecting their position as observers healthcare system first. We have no of society. This framing is consistent with immunity and no vaccine; the risks the proposition that solidarity becomes would be the same for our popu- more tenuous when sub-groups of the lation as at the beginning of the population feel invulnerable and fail 14 pandemic. If they open up before any to adhere to public health messaging of the above, then the lockdown and (eg, failure to social distance or border it all entailed was for nothing.” [GP breaches) or are given certain privileges participant] (eg, air crew). New Zealand academics have not been Discussion unanimously supporting border controls and social restrictions, and since February The importance of border control, along 2020 a small group of scientists and clini- with quarantine, to curb the spread of cians have advocated ‘COVID-19 Plan B’, pandemics has been recognised throughout protesting that New Zealand should not history.7 Rapid border control is seen as the ‘hunker in a bunker’ and shut ourselves off front-line strategy.5 Border control measures from the rest of the world.15 They continue enforced in China have been shown to to maintain a Facebook page promoting full dramatically limit spread,8 and not insti- opening of borders. Results from our study gating travel restrictions is likely to have indicate that general practice staff are not led to accelerated spread in Spain, Italy and convinced by the data promulgated by this Central Europe.9 group. Given that COVID-19 Plan B now also A study of different implementation strat- advocates against government roll-out of egies in seven Western-Pacific countries the COVID-19 vaccination programme,16 it (Hong Kong Special Administrative Region, is important to note that this vocal group Japan, Malaysia, Shanghai, Singapore, South appears to have had little influence on GP Korea, Taiwan) found that implementation opinion. of border control measures, along with case- finding by rapid tests and social distancing Strengths and limitations measures, was associated with bringing Our repeated Quick COVID-19 Surveys outbreaks under control,10 and a Taiwanese give primary care practices a voice. Rapid study records eliminating an early outbreak analyses and dissemination to key ministry using border control along with enhanced officials, primary care organisations and the surveillance, case detection with contact media has given them opportunity to impact tracing, quarantine and population-based on policy. Findings have been disseminated interventions, such as the use of face through TV, radio and written media and masks.11 have included Ministry of Health responses on how some of the expressed concerns will New Zealand researchers recognise be addressed, such as access to tests and that the border is our Achilles heel, and personal protective clothing and funding for incoming cases may lead to re-emergence additional workload: for example, testing of community transmission. Kvalsvig et al12 or vaccination delivery (see https://covid- recommend increased risk management 19-pc.auckland.ac.nz/media/). However, it with strategies that minimise incoming should also be noted that the sample size is infections, risk of missed cases or contacts relatively small and not representative of and consequences of infected or suscep- the whole practice staff population, and the tible individuals mixing with and infecting rapid analyses are inevitably ‘rough and others. Other researchers identify that a ready’. surveillance system with a very high level of routine testing is required to detect ongoing Implications breaches at our borders.13 Our participants have proved to be Concerns around the Achilles-heel nature prophetic on occasion. For example, their of the border has also led primary care concern that New Zealanders are too

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 74 www.nzma.org.nz/journal article

relaxed about the possibility of community spread has recently come to pass, with the Conclusion February/March 2021 lockdown due to those These Quick COVID-19 Survey results told to self-isolate breaking the rules. Other have highlighted concerns around border examples are their caution towards opening control from a primary care perspective. up to Australia (which also continues to The border control issues raised by primary have community outbreaks) and needing care have proven to be prophetic at times. to keep borders closed with Pacific coun- Often a primary care perspective is lacking tries to protect them. Primary care health in policy decisions and the rapid analysis professionals are at the interface between and dissemination of the surveys has the the public and government. Utilising potential to address this to some degree. rapid surveys, such as the Quick COVID-19 Greater attention could be applied to the Surveys, provides an opportunity for poli- use of rapid, repeated primary care-based cymakers to understand a primary care surveys in the future, in order to under- perspective that is often grounded in prag- stand a primary care position on an evolving matic reality. public health matter.

Competing interests: Nil. Acknowledgements: We would like to acknowledge our international colleagues, Prof Kirsty Douglas, Drs Kath- leen O’Brien, Katelyn Barnes and Sally Hall from Australia; Dr Rebecca Etz, Ms Sarah Reves and Mr Jonathan O’Neal from the United States; and Dr Sabrina Wong from Canada. Our appreciation to all individuals and organisations who have delivered our survey links and disseminated our results, and lastly a big thanks to everyone who have completed our sur- veys, either once or on multiple occasions. Author information: Kyle Eggleton: Senior Lecturer, Department of General Practice & Primary Health Care, University of Auckland, Auckland, New Zealand. Nam Bui: Research Fellow, Department of General Practice & Primary Health Care, University of Auckland, Auckland, New Zealand. Felicity Goodyear-Smith: Department of General Practice & Primary Health Care, University of Auckland, Auckland, New Zealand. Corresponding author: Dr Kyle Eggleton, Department of General Practice & Primary Health Care, University of Auckland, PB 92019, Auckland, New Zealand, +64 21 686 487 [email protected] URL: www.nzma.org.nz/journal-articles/making-sure-the-new-zealand-border-is-not-our-achilles- heel-repeated-cross-sectional-covid-19-surveys-in-primary-care-open-access

REFERENCES 1. Johns Hopkins University VA, US2021; [cited 2021 Practice [Internet]. of Medicine [Internet]. Feb 24]. Available from: COVID-19 General Practice COVID-19 Data Repository https://www.green-center. Clinicians: 5 Minute Survey by the Center for Systems org/covid-survey. Canberra, Australia: Austra- Science and Engineering 3. Primary and Integrated lian National University; (CSSE) at Johns Hopkins Health Care Innovations 2021; [cited 2021 Feb 25] University Baltimore, Network [Internet]. Canadi- Available from: https:// US2020. Available from: an Quick COVID-19 Primary medicalschool.anu.edu. https://github.com/ Care Survey Canada: SPOR au/research/projects/ CSSEGISandData/COVID-19 PIHCI Canada 2020; [cited covid-19-general-prac- accessed 29 May 2020]. 2021 Feb 25]. Available tice-clinicians-5-min- 2. Larry A Green Center from: http://spor-pihci. ute-survey accessed [Internet]. QUICK COVID- com/resources/covid-19/ 5. Department of General 19 SURVEY Richmond, 4. Academic Unit of General Practice and Primary

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 75 www.nzma.org.nz/journal article

Health Care [Internet]. travel restrictions. Comput ing the re-emergent Quick COVID-19 New Methods Biomech Biomed COVID-19 pandemic after Zealand Primary Care Engin 2020;23(11):710-17. elimination: modelling Survey Auckland, New doi: https://dx.doi.org/10.10 study of combined Zealand: Univeristy of 80/10255842.2020.1759560 primary care and hospital Auckland; 2021; [cited 10. Yeoh EK, Chong KC, Chiew surveillance. NZ Med J 2021 Feb 25]. Available CJ, et al. Assessing the 2020;133(1524):28-39. from: https://covid-19-pc. impact of non-pharma- 14. Elcheroth G, Drury J. auckland.ac.nz/results/ ceutical interventions on Collective resilience in 6. Goodyear-Smith F, Kinder the transmissibility and times of crisis: Lessons K, Eden AR, et al. Primary severity of COVID-19 during from the literature for care perspectives on the first five months in the socially effective responses pandemic politics. Glob Western Pacific Region. One to the pandemic. Br J Soc Public Health 2021:1-16. Health 2021;12:100213. doi: Psychol 2020;59(3):703- doi: https://dx.doi.org/10.10 https://dx.doi.org/10.1016/j. 13. doi: https://doi. 80/17441692.2021.1876751 onehlt.2021.100213 org/10.1111/bjso.12403 7. Piret J, Boivin G. 11. Cheng HY, Chueh YN, 15. The Economist [Internet]. Pandemics Throughout Chen CM, et al. Taiwan’s Economic Intelligence History. Front Microbiol COVID-19 response: Unit’s Democracy Index; 2020;11:631736. doi: https:// Timely case detection and [cited 2017 Oct]. Available dx.doi.org/10.3389/ quarantine, January to from: https://infographics. fmicb.2020.631736 June 2020. J Formos Med economist.com/2017/ 8. Wells CR, Sah P, Moghadas Assoc 2020;02:02. doi: DemocracyIndex/ SM, et al. Impact of inter- https://dx.doi.org/10.1016/j. 16. Outdoots Party [Internet]. national travel and border jfma.2020.10.023 Grey S. Open Letter to control measures on the 12. Kvalsvig A, Summers J, PM, Health Ministers and global spread of the novel Gray L, et al. COVID-19 the Attorney General 2019 coronavirus outbreak. outbreaks in Aotearoa of NZ re Pfizer vaccine Proc Natl Acad Sci USA New Zealand: urgent “comirnaty” re breaches of 2020;117(13):7504-09. doi: action is required to Medicines Act, misleading https://dx.doi.org/10.1073/ address systematic causes and deceptive claims pnas.2002616117 and consequences of and other matters. 2021 9. Linka K, Peirlinck M, Sahli border failures. NZ Med 31 Mar. Available from: Costabal F, et al. Outbreak J 2020;133(1527):8-14. https://www.outdoorsparty. dynamics of COVID-19 in 13. Wilson N, Schwehm M, co.nz/sue-grey-open-let- Europe and the effect of Verrall AJ, et al. Detect- ter-to-prime-minister/

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 76 www.nzma.org.nz/journal article

An NP-led pilot telehealth programme to facilitate guideline-directed medical therapy for heart failure with reduced ejection fraction during the COVID-19 pandemic Andy McLachlan, Chris Aldridge, Mary Morgan, Mayanna Lund, Ruvin Gabriel, Valerio Malez

ABSTRACT AIMS: Heart failure with reduced ejection fraction (HFrEF) is associated with poor outcomes. While several medications are beneficial, achieving optimal guideline-directed medical therapy (GDMT) is challenging. COVID-19 created a need to explore new ways to deliver care. METHODS: Fifty consecutive patients were taught to identify fluid congestion and monitor their vital signs using BP monitors and electronic scales with NP-led telephone support. Quantitative data were collected and a patient experience interview was performed. RESULTS: The majority (76%) of the cohort (male, 76%; Māori/Pacific, 58%) had a new diagnosis of HFrEF, with 90% having severe left ventricular (LV) dysfunction. There were 216 contacts (129 (60%) by telephone), which eliminated travelling, (time saved, 2.12 hours per patient), petrol costs ($58.17 per patient), traffic pollution (607 Kg of CO2) and time off work. Most (75%) received contact within two weeks and 75% were optimally titrated within two months. Improvements in systolic BP (SBP) (124mmHg to 116mmHg), pulse (78 bpm to 70 bpm) and N-terminal pro-brain natriuretic peptide (NT-proBNP) (292 to 65) were identified. Of the 43 patients who had a follow-up transthoracic echocardiogram (TTE), 33 (77%) showed important improvement in left ventricular ejection fraction (LVEF). CONCLUSIONS: Patients found the process acceptable and experienced rapid titration with less need for clinic review with titration rates comparable with most real-world reports.

eart failure with reduced ejection pharmacotherapies must be initiated in a fraction (HFrEF) is increasing in timely manner and titrated to maximally prevalence1 and is associated with tolerated doses, and they often require a H 2 3 inequality, significant costs, ill health number of clinic appointments. In addition, and preventable mortality.4 Several evi- careful monitoring for side effects, alongside dence-based interventions, including phar- a focus on patient empowerment, cannot macological, device and care strategies,5 im- be overlooked.10 Perhaps because of this prove quality of life and survival and reduce additional complexity of care, and despite hospitalisation.6 However, the delivery of a strong evidence base, guideline-directed these treatments can be challenging, as it is medical therapy (GDMT) is inconsistently de- impacted by system,7 clinician8 and patient livered by healthcare teams11 and, if offered, factors.9 To achieve the maximal benefits, is not always tolerable to patients.12

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 77 www.nzma.org.nz/journal article

The heart failure service at Counties electronic scales alongside nurse prac- Manukau District Health Board (CMDHB) titioner (NP)-led telephone support for includes a multidisciplinary team and has patients with HFrEF. a strong focus on titrating evidence-based The goal was to titrate medications medications to guideline-directed doses. safely with a target of two weeks between Timely titration may be limited by a variety enrolment and first contact and two weeks of factors, such as clinic volumes or barriers between medication changes. This had to be to patients attending clinics or filling achieved within the existing workflow and prescriptions. be acceptable for patients. Pragmatic criteria Different models of care have been were developed to identify patients who had attempted to improve GDMT, with multi- a clinical indication for titration and would disciplinary, HFrEF-specific clinics being be able to engage with this approach. 13,14 effective. Telehealth has also shown Patients were identified during an acute promise, particularly in patients with HF admission by the cardiology team and complex comorbidities and socioeconomic asked to participate if they met the inclusion 15 barriers to access. criteria. A commitment to engage with the On the 23 March 2020, as the COVID-19 home monitoring system was agreed by the pandemic swept across the globe, Aotearoa patient and the HF team. New Zealand went into level 3 lockdown, Inclusion criteria: and then a full level 4 lockdown two • Patients willing to participate with days later. The future of healthcare was symptomatic HFrEF (left ventricular uncertain. With a cohort of patients ejection fraction (LVEF) <40%) and currently in hospital being treated for acute requiring titration heart failure, normal outpatient titration of GDMT could not continue as usual. Both • Ability to collect prescriptions from a patients and healthcare providers, anxious community pharmacy and have blood about the risk of spreading the virus, limited tests collected every two weeks face-to-face contact. Telephone support • Patients who can understand instruc- for titration of heart failure (HF) therapy tions in English was suggested. Although neither new nor • Arm diameter between 22cm and novel,16 it had never been attempted in 42cm our department as a method of titrating • Weight <200kg HF medications. Under lockdown, a rapid and pragmatic response was required, with Exclusion criteria: early reports suggesting that the pandemic • Chronic kidney disease (CKD; eGFR was having a significant negative impact <30 ml/min/1.73m2) on patients attending for care.17 Remote • Hyperkalaemia (K+ >5.4mmol/l) management raised concerns regarding an • Hypotension (SBP <90mmhg) inability to safely optimise GDMT without • More than first degree heart block vital sign measurements and laboratory with no pacemaker results.18 A clinical care plan was developed to manage patients safely during this period. • Severe aortic stenosis This paper describes the feasibility of tele- • End stage heart failure and not for phone support by the HF team with the use active titration of scales and blood pressure (BP) monitoring • Atrial or Ventricular arrhythmia that devices to augment decision-making. could interfere with the accuracy of the BP and HR monitor Methods The team introduced the process and Using the latest decision pathway for discussed the need for frequent dose optimisation of heart failure treatment,6 adjustments and blood tests following dose our aims were to facilitate titration while adjustments. Self-help material included limiting in-person clinic visits by using ‘How has your breathing been in the last patient self-monitoring with a package that 2–3 days’ (a visual scale) and the book included funded home BP monitors and Living Well with Heart Failure (available

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 78 www.nzma.org.nz/journal article

from www.heartfoundation.org.nz), which tient department (Google Maps was used facilitates monitoring symptoms and vital to calculate distance). Standard car petrol signs. A HF action plan and guidance for usage was used to calculate petrol costs. The daily checks looking for signs and symptoms travel time was based on off peak traffic of decompensated HF was introduced. volumes to calculate a conservative estimate The blood pressure monitors (Omron of time saved. HEM-RML31) and electronic scales (capacity Data collected include: 200kg) were given to each patient following • Baseline a practical demonstration. • Demographics A booked fortnightly telephone call from the NP or clinical nurse specialist (CNS) • Vital signs and weight was agreed. Clinical support and guidance • NT-proBNP, renal function and were available from a consultant cardiol- electrolytes ogist. The first New Zealand lockdown was • Date of HF recent hospitalisation coming to an end by the time we started; but • Transthoracic echocardiogram contact was still uncertain because patients (TTE) assessment of LVEF preferred not to come to outpatient reviews, and health teams were still being advised to • Baseline medication use/doses maintain virtual reviews, where possible. • Follow-up However, we relaxed the non-contact rule, • Number of contacts and a face-to-face option was made available • Time to contact if required. Each patient, where possible, • Time to maximal tolerated GDMT met the HF team member who would support them at the beginning of the trial. • Reason for variation in titration Some patients preferred email contact and • Hospitalisations/deaths text, although telephone support was the • Change in clinical parameters most common way of communicating. Up-titration was facilitated by a new elec- Results tronic ePrescription and eLabform process Between 7 March and 5 August 2020, that had been fast tracked into clinical use. 52 patients were enrolled with HFrEF Data was statistically analysed using and agreed to take part in the trial. Two excel and a QI Macro with support from patients accepted but died before any the statistics team from Ko Awatea, outpatient contact was initiated and were CMDHB’s centre for health innovation and excluded, leaving 50 patients in the cohort improvement. For each hypothesis test, we for analysis. set an alpha value of 0.05 and used a paired Support for medication concordance, t-test to compare the variables before and alcohol harm reduction and smoking after the trial participation, after having cessation support was offered to all. verified that the difference between pairs were normally distributed. Table 1: Demographics. Quantitative data was collected in a secure database, and a patient experience Total =50 (%) telephone interview was conducted by Ko Awatea, independently of the cardiology Male 38 (76%) team. Mean age (all) 58.9 years Patient symptoms, clinical findings and Female 65.1 years any change in the HF plan were notified Male 56.9 years in real time using a secure, electronic template that communicated directly to the Māori 12 (24%) designated primary care provider. Pacific Islanders 17 (34%) The accessibility benefits to the patient Others 21 (42%) from the virtual consultation was calcu- lated based on distance travelled from Compared to the population of CMDHB: Māori 16%, the patient’s home address to the outpa- Pacific 34% and others 63%.19

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 79 www.nzma.org.nz/journal article

Each participant had an assessment of of 10.8 days): 50% were contacted within 9 their left ventricular ejection fraction (LVEF) days, 75% contacted within 14 days and all either at the time of hospital admission or were contacted by 28 days. prior to clinic referral. Almost all (90%) were The majority (60%) of enrolled patients classified as having severe left ventricular had been in hospital within the last six 4 dysfunction (LVEF <30%), with 48 (96%) months with a primary diagnosis of decom- reporting being symptomatic, New York pensated HF: 50% of patients were enrolled Heart Association (NYHA) class 2 or above. within two months from their most recent This is a high-risk cohort of patients with hospitalisation and 75% within four months. significant HFrEF with a prognosis worse than most cancers.20 The virtual consultation effect During the period of the trial, 216 contacts Time to first contact were made: 129 (60%) by telephone and 87 The time to contact was consistent, with (40%) face to face. low standard variation (5 days over a mean By eliminating the need to travel to the outpatient department, we estimated each Table 2: Baseline characteristics. patient saved on average 2.12 hours and travelled 73.6 fewer kilometres. This equates Total=50, (%) to savings in travel costs of $2,908 during the

pilot, or $58.17 per patient. Total CO2 emis- New diagnosis of 38 (76%) sions were reduced by 607 Kg, which would HFrEF have required 27.9 medium-sized trees to absorb this amount within one year. History of CVD 21 (42%) GDMT summary Hypertension 34 (68%) Within two months of entering the trial Atrial fibrillation/ 12 (24%) (56 days), 75% of patients were deemed to be optimally titrated, with 88% achieving ≥50% flutter of target dose of renin angiotensin blocker, Obstructive sleep 6 (12%) 74% achieving ≥50% of target dose of beta apnoea blocker and 62% being on spironolactone (MRA). Type 2 diabetes 22 (44%) The use of Entresto (ARNI), a novel nepri- HbA1c (mmol/l) Mean 64 (range 43–100) sylin inhibitor/angiotensin receptor blocker agent, increased from 12% to 40% and Body mass index (kg/m2) Mean 32 (range 18–59) diuretic use fell from 46% to 26%. CKD (eGFR <50 ml/ 11 (22%) Reasons for variation in GDMT min/1.73m2) Over one third (17 (38%)) of patients Implantable defibrilla- 4 (8%) were up titrated in a step-like fashion with tor (ICD) in situ regular monitoring and tolerated the process uneventfully. A resting heart rate consis- Reported non-concor- 8 (16%) tently less than 60bpm limited titration in dance 11 patients (22%). A resting systolic blood pressure less than 90mmhg or symptomatic Current smoking 10 (20%) hypotension limited titration in nine (18%), Harmful alcohol use 8 (16%) and a further four (8%) developed significant hyperkalaemia or a deterioration in eGFR, requiring stopping or reducing the dose of Table 3: Baseline left ventricular ejection func- tion. GDMT. Two patients (4%) had significant comorbidity related to cancer therapy, which LVEF N =50 (%) delayed contact due to frequent hospitalisa- tions. Despite multiple attempts, five (10%) Less than 20​% 23 (46%)​ patients were either unable to be consis- 21–30%​ 22 (44%)​ tently contacted or declined to optimise any therapy. Cardiology outpatient clinic non-at- 31–35%​ 5 (10%)​ tendance rates are approximately 20%.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 80 www.nzma.org.nz/journal article

HF outcomes ty-three patients (66%) showed improved With a mean of 203 days (range 140–264 LV function with 10 (20%) moving from days) follow-up, there were no deaths severe LVEF <30% to moderate LVEF >35% during the course of the pilot. and 23 (46%) improving to mild dysfunction, defined as LVEF >40% (p 0.0001). Twenty-five admissions in 19 patients (38%) were recorded with four admissions Patient experience (8%) related to HF. Fourteen patients consented to a tele- Clinical outcomes phone interview: half female, half male, four Māori, four Pacific, five Pākehā and one During the evaluation there were Asian (Appendix Figure 1). The majority of significant reductions in systolic blood patients expressed confidence in using and pressure (p. 0.004), heart rate (p. 0.002) and reading the BP monitor from home. Patients NT-proBNP (p. 0.001) (Figure 1). Other indi- reported feeling empowered and having cators, such as eGFR, serum potassium and increased motivation to manage their health NYHA class, showed no significant changes conditions since participating in the trial. (p. 0.2). Patient experience with clinicians was Assessment of LVEF positive, with the majority describing appre- Forty-three patients (86%) received a ciation for the interaction and rapport. They follow-up TTE after titration to maximum expressed an understanding of the changes tolerated GDMT. to their medication and felt that the alter- Ten patients (20%) had continued severe ations were beneficial to their health. The left ventricular (LV) dysfunction and were booklet provided by the service facilitated referred back into the HF clinic for device patients’ understanding and acceptance therapy or further GDMT optimisation. Thir- of changes to medication. There were a few patients who specifically needed more support, including a better understanding of Table 4: Cause of heart failure. the detail of what they needed to do, when Cardiomyopathy N=50 (%) they needed to do it and what to expect. Ischaemic 14 (28%) Discussion Non-ischaemic 36 (72%) In New Zealand, approximately 5,500 Dilated (not further patients generate about 12,000 hospital 17 (34%) admissions for HF each year. The average defined) length of stay is five days, and the overall Accelerated heart rate 9 (18%) costs associated with HF account for 1.5–2% of the total health budget, most of which Alcohol 4 (8%) is for inpatient care.3 It is estimated that Valvular 3 (6%) approximately 20% of the population will develop HF in their lifetime, which places a Anthracycline 2 (4%) significant burden on individuals, commu- Sarcoid 1 (2%) nities and health services.21 Having skilled teams provide gold-standard and individ- ualised HF assessment and management Table 5: Optimal tolerated titration dose of GDMT. is vital work, but many patients remain 22 ACE/ARB/ Beta MRA underserved. During the COVID-19 ARNI blocker pandemic, this became even more chal- lenging, and the HF team introduced an Target 28 (56%) 16 (32%) 24 (48%) inexpensive, home-based approach to heart 50–99% 16 (34%) 21 (42%) 7 (14%) failure care. We achieved rates of timely of target GDMT optimisation—at least as good as dose many contemporary clinical studies—and we managed to do this within existing Low/ 6 (12%) 13 (26%) 19 (38%) workloads and good engagement from most none patients.23

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 81 www.nzma.org.nz/journal article

It is projected that the number of people reduced heart failure hospitalisations.6 with heart failure will increase as people Technological advances have allowed live longer and can access more effective increasingly sophisticated methods to treatments for coronary heart disease remotely monitor and manage heart failure. 24 associated with a reduction in mortality. Simple telephone-based remote assess- Inequality persists, with the mortality rate ments, stand-alone home-based systems, from heart failure for male and female implanted devices with advanced haemo- Māori aged over 65 years being signifi- dynamic monitoring features and now cantly higher than for non-Māori. Rates wearable technologies have opened up a of hospitalisation for heart failure among range of opportunities to facilitate patient Māori in this age group are also signifi- management.27 It is relatively easy to collect cantly higher than for non-Māori. Māori data remotely, but it has been a challenge are significantly younger on admission to to find a way to integrate continuous data hospital for heart failure than New Zealand streams into already overloaded systems of Europeans (62 years compared to 78 years). care, and to convert remote data into better Morbidity and mortality from heart failure decision-making that improves the outcome for Pacific peoples is approximately twice or experience of care.28,29 as high compared to the total population.25 Both structured telephone support and Over half (58%) of our pilot group were telemonitoring have been reported as being Māori or Pacific and we were able to show effective in reducing all-cause mortality active engagement and equitable outcomes and heart failure related hospitalisations, as compared to the non-Māori or non-Pacific well as improving quality of life, reducing participants. healthcare costs and enhancing GDMT.30 A recent New Zealand-wide cohort study A positive effect was noted using tele- of patients with acute coronary syndrome health to monitor heart failure patients in a (ACS) showed that rates of GDMT in those home environment, with patients showing with reduced LVEF was low at one year an increased self-awareness around post discharge—only 34% and 35% received managing their condition31 and increased ≥50% target doses of ACEi/ARB and beta- empowerment and confidence.32 Essential 26 blockers respectively. Suboptimal use of characteristics of effective telehealth include GDMT therefore persists despite its associ- clinical feedback in the form of ‘teachable ation with improved patient mortality and

Figure 1: Comparison between BP, HR and BNP at baseline compared to follow-up.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 82 www.nzma.org.nz/journal article

moments’, a system which is easy and quick immediately after lockdown. This addi- to use and patients perceiving tangible tional workload will likely be unsustainable benefits from the system.33 when face-to-face clinics are again running Interviews suggest that telehealth is to full capacity. Additional staff and moni- generally acceptable to most patients, toring device resources will be required including Māori and their whānau.34 for a sustainable programme. The patients Despite this, there is a high rate of tele- accepted the process and appreciated the health refusal among patients, which is efforts of the HF team to reassure and not well understood.35 The perception for support them through these challenging many, particularly older and less tech- times. aware people, is that telehealth is remote, Limitations cold and distant, and many people prefer a This was a small feasibility pilot project, more personal touch from their healthcare delivered rapidly and in very uncertain provider.36 However, we did not find this times by an enthusiastic team of HF focused to be a significant issue, perhaps as the clinicians. The patients were a selected COVID-19 environment increased accept- cohort and may not represent the entire HF ability of alternative models of care. cohort. The benefits of telemonitoring are The strengths of this report are the dependent on a number of factors. inclusion of participants representative of Researchers have not consistently shown our HF cohort and the wider community. We positive outcomes, which has led commen- performed telephone interviews to under- tators to criticise speed of uptake and stand the patient experience and a further implementation.37 Increased costs and follow-up telephone interview is planned. clinical workloads have also been reported.33 The exclusion of non-English speaking It is possible the benefits seen in telemoni- participants impacted on our ability to toring trials could be due to enhancement of provide an equitable service, and we have the underpinning clinical service rather than since explored the use of interpreters the telemonitoring communication itself.38 using three-way telephone conversations. Because of increased communication The majority of patients were able to be with patients, which often occurs outside contacted, although access to mobile phones of formally booked clinic times, the CMDHB was inconsistent for some and did limit pilot resulted in additional workload for engagement. the clinicians. This was offset by fewer Further research is needed to under- face-to-face appointments during and stand the components of this observational

Figure 2: LVEF at baseline compared to follow-up.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 83 www.nzma.org.nz/journal article

study that can be used as an adjunct to good saving patients both time and money. quality heart failure care. Titration rates and markers of improved outcomes improved across cardiac imaging, Conclusion biochemical and clinical findings, and were comparable to most real-world clinical For most patients, the home monitoring/ reports. Although this is not a novel or inno- telephone process resulted in rapid titration vative process, it is not usual care in New and less need for clinic review. Patients Zealand. This simple and straightforward found the process acceptable and 60% process could be replicated across DHBs. of clinic visits were able to be held remotely,

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 84 www.nzma.org.nz/journal article

Appendix Appendix

Appendix Figure 1: Patient experience questionnaire page one. View the complete Appendix Figure 1. Appendix Figure 1: Patient experience questionnaire

Blood pressure monitoring questions for telephone interviews Proposed telephone interviews with patients at approximately day three (to identify immediate concerns) and follow-up interviews at the date that drop outs tend to occur

Patient details Anonymised Name: Age: Suburb you live in: Ethnic group/s: Zoom or Telephone Interview: Contact Number: Zoom Link: Availability for interview (Fill in details below) Date: Time:

Elevator pitch Hello/Kia ora/Talofa/ Malo e lelei/ Namaste/ Ni hao/ Fakaa alofa lahi atu my name is… from Counties Manukau Health. I am calling to see how you are finding doing your own blood pressure and weight checks from home? Are you happy to speak with me in an interview about how it is going?

• No… is there another time that would be better for me to call you? • No… ok thank you for your time, take care and stay safe, goodbye. • Yes… ok thank you… Are you available for 20 minutes now or would you like me to call you back?

Thank you, we will continue.

For the first time we are helping patients to check their own blood pressure and weight from home, while also making sure that you are safe and your health is not getting worse. Our patients’ feelings are important to us, so we need to make sure we support you as best we can, this interview will help us to do this. We would like to know your thoughts about checking your own blood pressure and weight changes from home and have some questions we would like to ask you. Your thoughts will also help us to understand what is going well, and not so well and how it may work better.

We are asking patients to take part in telephone interviews lasting around 20 minutes of your time.

We will not be audio recording our conversation, but will write notes as we are speaking. The information you share with us will be confidential and no one involved in your care will know that you have spoken with us. All information you provide is confidential and will not include your name or other personal details that identify you in any of our reports.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 85 www.nzma.org.nz/journal article

Competing interests: Dr Lund reports other from Amgen inc and personal fees from Novartis outside the submitted work. Acknowledgements: Te Hao Apaapa-Timu-Programme Manager Co-design, Ko Awatea. Lucy Wong Improvement Advisor Co-Design, Pharmacist, NZRegPharm, Ko Awatea. Author information: Andy McLachlan: Nurse Practitioner, CMDHB. Chris Aldridge: Nurse Practitioner, CMDHB. Mary Morgan: Clinical Nurse Specialist, CMDHB. Mayanna Lund: FRACP, Cardiologist, CMDHB. Ruvin Gabriel: FRACP, Cardiologist, CMDHB. Valerio Malez: Senior Improvement Advisor, Ko Awatea . Corresponding author: Andy McLachlan Nurse Practitioner, CMDHB [email protected] URL: www.nzma.org.nz/journal-articles/an-np-led-pilot-telehealth-programme-to-facilitate-guide- line-directed-medical-therapy-for-heart-failure-with-reduced-ejection-fraction-during-the- -covid-19-pandemic-open-access

REFERENCES 1. Savarese G, Lund LH. The Heart Failure Clinic: 9. Fiuzat M, Ezekowitz Global Public Health A Consensus Statement J, Alemayehu W, et al. Burden of Heart Failure. of the Heart Failure Assessment of Limita- Card Fail Rev. 2017. Society of America. J Card tions to Optimization doi:10.15420/cfr.2016:25:2 Fail. 2008. doi:10.1016/j. of Guideline-Directed 2. Inkrot S. Comparison of cardfail.2008.10.005 Medical Therapy in Heart New Zealand European 6. Maddox TM, Januzzi Failure from the GUIDE- vs New Zealand maori JL, Allen LA, et al. 2021 IT Trial: A Secondary patients with acute heart Update to the 2017 ACC Analysis of a Randomized failure: The New Zealand Expert Consensus Decision Clinical Trial. JAMA heart failure registry. Eur Pathway for Optimization Cardiol. 2020. doi:10.1001/ J Cardiovasc Nurs. 2014. of Heart Failure Treat- jamacardio.2020.0640 3. Braunschweig F, Cowie ment: Answers to 10 10. Yancy CW, Januzzi JL, MR, Auricchio A. What Pivotal Issues About Heart Allen LA, et al. 2017 ACC are the costs of heart Failure With Reduced Expert Consensus Decision failure? Europace. Ejection Fraction. J Am Coll Pathway for Optimization 2011. doi:10.1093/ Cardiol. 2021. doi:10.1016/j. of Heart Failure Treatment: europace/eur081 jacc.2020.11.022 Answers to 10 Pivotal Issues About Heart 4. CSANZ Heart Failure 7. Driscoll A, Meagher S, Failure With Reduced Guidelines Working Kennedy R, et al. What is Ejection Fraction. J Am Coll Group N, Atherton MBBS the impact of systems of Cardiol. 2018. doi:10.1016/j. JJ, Sindone BMed A, et al. care for heart failure on jacc.2017.11.025 National Heart Foundation patients diagnosed with of Australia and Cardiac heart failure: A systematic 11. Shanbhag D, Graham ID, Society of Australia and review. BMC Cardiovasc Harlos K, et al. Effective- New Zealand: Guidelines Disord. 2016. doi:10.1186/ ness of implementation for the Prevention, Detec- s12872-016-0371-7 interventions in improving tion, and Management of 8. Jarjour M, Henri C, de physician adherence to Heart Failure in Australia Denus S, et al. Care Gaps guideline recommenda- 2018 Contents. Hear in Adherence to Heart tions in heart failure: A Lung Circ. 2018;27:1123- Failure Guidelines: Clinical systematic review. BMJ 1208. doi:10.1016/j. Inertia or Physiological Open. 2018. doi:10.1136/ hlc.2018.06.1042 Limitations? JACC Hear bmjopen-2017-017765 5. Hauptman PJ, Rich MW, Fail. 2020. doi:10.1016/j. 12. Cornelio C, Palo KE Di. Heidenreich PA, et al. jchf.2020.04.019 Guideline-Directed Medical

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 86 www.nzma.org.nz/journal article

Therapy in Hospitalized cardfail.2020.09.021 25. Identifying patients with Heart Failure Patients: Still 19. Winnard D, Lee M MG. heart failure - BPJ Issue Underprescribed Despite Demographic Profile: 2013 50. https://bpac.org.nz/ Updated Guidelines and Census, Population of bpj/2013/february/identi- Over 20 Years of Evidence. Counties Manukau.; 2015. fying-heart-failure.aspx. J Card Fail. 2018;24(8):S100. Accessed February 1, 2021. 20. Mamas MA, Sperrin M, doi:10.1016/j. Watson MC, et al. Do 26. Chan D, Doughty RN, Lund cardfail.2018.07.380 patients have worse M, Lee M, Kerr AJ. Target 13. Rice H, Say R, Betihavas outcomes in heart failure Doses of Secondary Preven- V. The effect of nurse-led than in cancer? A primary tion Medications Are Not education on hospitalisa- care-based cohort study Being Achieved in Patients tion, readmission, quality with 10-year follow-up in With Reduced Left Ventric- of life and cost in adults Scotland. Eur J Heart Fail. ular Ejection Fraction After with heart failure. A 2017. doi:10.1002/ejhf.822 Acute Coronary Syndrome systematic review. (ANZACS-QI 34). Hear Lung 21. Wall R, Bell A, Devlin G, Patient Educ Couns. Circ. 2020. doi:10.1016/j. Lawrenson R. Diagnosis 2018;101(3):363-374. hlc.2020.03.013 and treatment of heart doi:https://doi.org/10.1016/j. failure in Māori and New 27. Brahmbhatt DH, Cowie MR. pec.2017.10.002 Zealand Europeans at Remote Management of 14. Gandhi S, Mosleh W, the . Heart Failure: An Overview Sharma UC, Demers C, N Z Med J. 2013. of Telemonitoring Technol- Farkouh ME, Schwalm ogies. Card Fail Rev. 2019. 22. Chan D, Doughty RN, JD. Multidisciplinary doi:10.15420/cfr.2019.5.3 Mazengarb J, McLachlan Heart Failure Clinics Are A, Kerr AJ. Heart failure 28. Andrès E, Talha S, Zulfiqar Associated With Lower clinics improve use of A-A, et al. Current Research Heart Failure Hospital- evidence-based heart and New Perspectives of ization and Mortality: failure therapies in Telemedicine in Chronic Systematic Review and patients with reduced Heart Failure: Narrative Meta-analysis. Can J ejection fraction following Review and Points of Cardiol. 2017. doi:10.1016/j. acute coronary syndrome Interest for the Clini- cjca.2017.05.011 (ANZACS-QI 48). N Z Med cian. J Clin Med. 2018. 15. Gorodeski EZ, Goyal P, J. 2020;133(1516):58-71. doi:10.3390/jcm7120544 Cox ZL, et al. Virtual Visits https://www.nzma.org. 29. Fairbrother P, Ure J, Hanley for Care of Patients with nz/journal-articles/ J, et al. Telemonitoring Heart Failure in the Era heart-failure-clinics-im- for chronic heart failure: of COVID-19: A Statement prove-use-of-evidence- The views of patients from the Heart Failure based-heart-failure-ther- and healthcare profes- Society of America. J Card apies-in-patients-with- sionals - a qualitative Fail. 2020. doi:10.1016/j. reduced-ejection-fraction- study. J Clin Nurs. 2014. cardfail.2020.04.008 following-acute-coronary-s- doi:10.1111/jocn.12137 16. Hannah J, Humphrey G, yndrome-anzacs-qi-48. 30. Sousa C, Leite S, Lagido R, Doughty R, McGrinder Accessed February 7, 2021. Ferreira L, Silva-Cardoso J, H, Bos N, Bowman C. 23. Greene SJ, Butler J, Maciel MJ. Telemonitoring Telehealth in Heart Failure Albert NM, et al. Medical in heart failure: A state- Management: A Proof of Therapy for Heart Failure of-the-art review. Rev Port Principle Study. Hear Lung With Reduced Ejection Cardiol. 2014. doi:10.1016/j. Circ. 2010. doi:10.1016/j. Fraction: The CHAMP- repc.2013.10.013 hlc.2010.04.028 HF Registry. J Am Coll 31. Varon C, Alao M, Minter J, 17. Wong B, El-Jack S, Cardiol. 2018;72(4):351- et al. Telehealth on heart Armstrong G. Pandemic 366. doi:10.1016/j. failure: results of the Recap control: getting to the heart jacc.2018.04.070 project. J Telemed Telecare. of unintended conse- 24. Sidney S, Quesenberry 2015. doi:10.1177/13 quences. N Z Med J. 2020. CP, Jaffe MG, et al. Recent 57633X15577310 18. Ugolini S, Mondesir FL, trends in cardiovascular 32. Inglis SC, Clark RA, McAl- Maires C, et al. Virtual mortality in the United ister FA, Stewart S, Cleland Outpatient Heart Failure States and public health JGF. Which components of Care - Lessons From the goals. JAMA Cardiol. heart failure programmes Covid-19 Era. J Card 2016. doi:10.1001/ are effective? A systematic Fail. 2020. doi:10.1016/j. jamacardio.2016.1326 review and meta-analysis

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 87 www.nzma.org.nz/journal article

of the outcomes of struc- of a telehealth-enabled 2017. doi:10.1186/ tured telephone support chronic care manage- s12872-017-0594-2 or telemonitoring as the ment service to support 37. Chaudhry SI, Mattera JA, primary component of people with long-term Curtis JP, et al. Telemon- chronic heart failure conditions at home. J itoring in Patients with management in 8323 Telemed Telecare. 2012. Heart Failure. N Engl J patients: Abridged Coc. doi:10.1258/jtt.2012.SFT112 Med. 2010. doi:10.1056/ Eur J Heart Fail. 2011. 35. Woo K, Dowding D. Factors nejmoa1010029 doi:10.1093/eurjhf/hfr039 affecting the acceptance 38. Pinnock H, Hanley J, 33. Seto E, Leonard KJ, Cafazzo of telehealth services by McCloughan L, et al. JA, Barnsley J, Masino C, heart failure patients: Effectiveness of telemon- Ross HJ. Perceptions and An integrative review. itoring integrated into experiences of heart failure Telemed e-Health. 2018. existing clinical services patients and clinicians on doi:10.1089/tmj.2017.0080 on hospital admission for the use of mobile phone- 36. Greenhalgh T, A’Court C, exacerbation of chronic based telemonitoring. J Shaw S. Understanding obstructive pulmonary Med Internet Res. 2012. heart failure; explaining disease: Researcher blind, doi:10.2196/jmir.1912 telehealth - A hermeneu- multicentre, randomised 34. Venter A, Burns R, Hefford tic systematic review. controlled trial. BMJ. 2013. M, Ehrenberg N. Results BMC Cardiovasc Disord. doi:10.1136/bmj.f6070

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 88 www.nzma.org.nz/journal article

Empty waiting rooms: the New Zealand general practice experience with telehealth during the COVID-19 pandemic Geraldine Wilson, Olivia Currie, Susan Bidwell, Baraah Saeed, Anthony Dowell, Andrew Adiguna Halim, Les Toop, Ann Richardson, Ruth Savage, Ben Hudson

ABSTRACT AIM: The primary care response to the coronavirus disease 2019 (COVID-19) pandemic in early 2020 required significant changes to the delivery of healthcare by general practices. This study explores the experiences of New Zealand general practice teams in their use of telehealth during the early stages of the COVID-19 pandemic in New Zealand. METHOD: We qualitatively analysed a subtheme on telehealth of the General Practice Pandemic Experience New Zealand (GPPENZ) study, where general practice team members across the country were invited to participate in five surveys between 8 May 2020 to 27 August 2020. RESULTS: 164 participants enrolled in the study during survey one, with 78 (48%) completing all surveys. Five telehealth themes were identified: benefits, limitations, paying for consults, changes over time and plans for future use. Benefits included rapid triage, convenience and efficiency, and limitations included financial and technical barriers for practices and patients and concerns about clinical risk. Respondents rapidly returned to in-person consultations and wanted clarification of conditions suited to telehealth, better infrastructure and funding. CONCLUSION: To equitably sustain telehealth use, the following are required: adequate funding, training, processes communicated to patients, improved patient access to technology and technological literacy, virtual physical examination methods and integration with existing primary health care services.

he primary care response to the of remote healthcare through use of tele- coronavirus disease 2019 (COVID-19) communication devices, such as phones pandemic in early 2020 required and smart-phones, sometimes containing T 3 significant changes by general practices video. Over 48 hours, many practices rapidly throughout New Zealand. A major part of adopted telehealth consultations, managing this change was the swift move to telehealth people virtually at home where possible, to consultations, replacing a large proportion reduce the transmission of COVID-19 within of in-person consultations. healthcare centres.1,4 A single-practice retro- A switch to virtual consulting was recom- spective audit from Dunedin, New Zealand, mended by the Royal New Zealand College during the first two weeks of Alert Level of General Practitioners (RNZCGP)1 just days 4 lockdown found an increase in virtual before the government announcement of the consultations, to 79% from 30% in the same country moving to Alert Level 4 (lockdown) two-week period the year prior, with a 5 on 25 March 2020 due to increasing five-fold increase in phone consultations. community transmission of COVID-19.2 Tele- Current literature on telehealth use in health has been described as the provision COVID-19 consists largely of opinion pieces

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 89 www.nzma.org.nz/journal article

or single-institution findings.6,7 However, sations to encourage representation from qualitative interview findings from a large these groups. group of Finnish general practitioners The surveys were designed using Qual- explored the telehealth experience under tricsXM software.10 The surveys consisted of 8 person-centred care. Importantly, a New open-ended questions allowing for free-text Zealand study exploring patient experi- responses about participants’ experiences ences during COVID-19, including their during the COVID-19 pandemic in New experience of telehealth through an online Zealand. Questions requiring numerical survey and focus groups, provides a comple- answers were also included to explore the mentary insight into telehealth during this extent to which telehealth was used. The 9 time. No published New Zealand data have surveys explored participants’ personal and as yet qualitatively explored the impact professional experiences of the pandemic of this swift change to telemedicine on and included prompts about changes to primary care practice teams. This is the the delivery of patient care, use of tele- first qualitative analysis of the experience health, challenges, innovations, personal of telehealth from a large group of primary and professional supports, health and healthcare professionals throughout a coun- safety preparedness for staff and patients, try’s health system. staff wellbeing and business and financial aspects. Methods Participant textual responses and General practitioners (GP), nurse prac- comments were extracted from Qualtrics10 titioners (NP), practice nurses (N) and and analysed using NVivo as a coding practice managers (P) were invited framework.11 A team of researchers (GW, SB, to participate in the General Practice AD, BS and AH) conducted the analysis of the Pandemic Experience New Zealand data using simple descriptive statistics and a (GPPENZ) study, which followed the same thematic analysis within a constant compar- group of participants through a series of ative approach.12 First the responses were five online surveys from 8 May 2020 to 27 read in their entirety to develop a structured August 2020 (Figure 1). The invitation was framework and identify emerging themes. distributed widely through the RNZCGP, This was conducted in multiple rounds New Zealand Nurses Organisation (NZNO), with independent coding of themes by the primary health organisations (PHOs), peer researchers and final blinded input by AD groups and medical and social media. until all coders were satisfied with each code Recruitment was also targeted through definition. A codebook was developed. For Māori, Pasifika and rural medical organi- this paper, telehealth content was specif-

Figure 1: Timeline of General Practice Pandemic Experience New Zealand (GPPENZ) study 2020.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 90 www.nzma.org.nz/journal article

ically coded for by GW and BS (with each managed virtually and identifying those peer-reviewing the other’s coding). The who needed in-person consultations. Tele- primary analysis for this paper consisted of health permitted staff in some practices to questions directly related to telehealth (see take turns working from home to minimise Appendix), and further secondary analysis infection risk. Use of patient portals was performed where telehealth was coded increased, with one practice reporting for in more general survey questions. A around 1,000 new patients joining up. framework for analysis was developed People who were well connected electron- by GW, SB and TD, who also conducted a ically and younger populations “embraced thematic content analysis drawn from the the technology” (N19) and were able to relevant codes. This analysis was reviewed benefit most: by the authors. “Most people have no issues. The Ethical approval was obtained from majority of our patient population The University of Otago Human Ethics fortunately have access to suitable Committee (reference number D10/114). technology.” (GP35) Convenience, saving time and the lack of Results transport costs were also advantages for Participant characteristics and demo- rural people and those with work or family graphics from survey one is shown in Table commitments: 1. There was consistent participation from “There is definitely a cohort of all occupations over the five surveys. Partic- patients who love this model due to ipants from practices in Canterbury were being rural and the vast distances over-represented in the sample, at 37.8% of needed to travel… It most certainly survey one. Responses were received from has a place in delivering patient throughout New Zealand from participants care.” (P6) 13 affiliated with 80% of all PHOs. Respondents also mentioned specific We defined five major telehealth themes: presenting issues that could be discussed benefits, limitations, paying for telehealth remotely. Sending through photos of skin consultations, changes over time and lesions or injuries was a notable success for perceived future use. Excerpts from survey some. Another benefit was the increased responses are identified by the discipline of efficiency for following-up patients who the respondent and an index number (eg, were well known, including being “very ‘GP80’ is the eightieth general practitioner). useful for mental health issues especially Key benefits of telehealth during when one can see the patients” (GP65). COVID-19 Changes to previous health and social The aim and key benefit of the swift welfare processes supported telehealth, change to virtual consultations forced enabling a wider reach than would by the lockdown was to ensure that otherwise have been possible. Accident practices were able to keep providing Compensation Corporation (ACC) and Work healthcare while reducing the spread of and Income (WINZ) reviews and nurse illness through avoiding in-person contact, practitioner prescribing were all opened to especially in waiting rooms. Some were telehealth rather than, as previously, being already well prepared to offer telehealth limited to in-person appointments: options; GP80 commented that “we have “[Previously] to Covid, restrictions been using phone, email and text for many around my prescribing practice years.” Others were taken by surprise and included that I could only prescribe scrambled to get appropriate systems and in face-to-face situations. This has equipment in place. been changed over Covid to support Although the speed and urgency of the e-health options for consultations.” transition was stressful for some, there (NP10) was early recognition of the benefits for “Telehealth was used proactively both practices and patients. Triaging all by practices to contact patients patients became standard practice, further they were concerned about. Special increasing the number that could be efforts were made for ‘patients

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 91 www.nzma.org.nz/journal article

Table 1: Participants.

Total (%) Survey 1 164

Survey 2 136 (82.9%)

Survey 3 118 (72%)

Survey 4 112 (68.3%)

Survey 5 91 (55.5%)

Completed all surveys (1–5) 78 (48%)

Demographics Mean age (SD, range) 49.9 (SD 10.65, 25–71)

Female 125 (76.2%)

Ethnicity (total count*) European 144 (87.8%)

Māori 9 (5.5%)

Pacific Peoples 5 (3.0%)

Asian 12 (7.3%)

MELAA 2 (1.2%)

Occupation General practitioner 93 (56.7%)

Practice nurse 38 (23.2%)

Nurse practitioner 11 (6.7%)

Practice manager 18 (11%)

Practice manager and nurse (dual role) 4 (2.4%)

Type of practice# Urban 115 (70.1%)

Rural 34 (20.7%)

Other (eg, mixed) 14 (8.5%)

Practice size Full time equivalent GPs mean (range) 5.1 (0–20)

Employment status Employees 72 (43.9%)

Owner or partner 45 (27.4%)

Contractor 28 (17.1%)

Other 9 (5.5%)

* Total count of ethnicities will be greater than the number of respondents because one person can identify as belonging to multiple ethnicities. # Consistent with RNZCGP national GP workforce data.14 MELAA = Middle Eastern/Latin American/African.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 92 www.nzma.org.nz/journal article

with complex, chronic conditions both internet and cell phone connections to develop tailored plans with them and so presented equal difficulties for both and their whānau and ensure their practices and patients: medical needs were catered for and “Due to being rural and some of our that there was a plan in place’.” patients not having great internet or (GP32) cellphone coverage (or none), getting Telehealth was used in specific healthcare in contact with patients at the time initiatives for Māori, with one practice of their scheduled call has proved a proactively contacting Māori through their challenge. We have sometimes had to marae clinic: make multiple calls to get in contact.” “We texted all our marae clinic (P6) patients to advise how they could Additionally, certain patient groups were access healthcare. We set up a closed reported as being excluded from using tele- [Facebook] group to communicate health successfully. Many older people did with marae clinic patients.” (P3) not have smartphones or internet connec- Another advantage of telehealth for Māori tions, and it was unsuitable for patients with was that family groups could participate cognitive impairment, hearing difficulties from home: and those with limited English: “Māori have used telehealth “…very difficult to understand frequently as enabled whānau partici- accents etc over the phone… [a] lack pation in their own whare.” (N10) of body language in these patients who often have poor health literacy The limitations of telehealth makes this very difficult…” (GP5) The financial and technical barriers Respondents reported that some patients to telehealth were perceived as a major across all age ranges asked for in-person limitation by many respondents. Many consultations for greater confidentiality, a practices reported patients who could better feeling of rapport or being able to not ring for an appointment or access express themselves more coherently. messages. There were reports of patients sharing phones with others, or only being There were numerous concerns about the contactable through social media: clinical risk of not seeing patients in person. Video calling was described as being “a poor “Many have no data and cannot use substitute for face-to-face clinical exam- virtual care. Many others change ination” (GP8). There was a “fear of missing phone numbers or have no credit so something that might have been picked up cannot access texts or messages.” at in person visit” (GP20) and it was “difficult (GP1) to assess severity without pulse and temp There were difficulties with slow internet, checks” (GP83). Virtual consultations were poor sound or picture quality, incorrect also said to have a “lack of rapport… partic- phone numbers and calls that went unan- ularly difficult for patients with language swered at the time the practice had arranged barriers” (GP76) and patients who were new to call: or not well known. Others noted that, even “Around 50% have poor quality though telehealth was promoted for its time internet or devices, and/or difficulties saving and convenience, it was “inefficient using (eg, no sound or picture). if patient subsequently needs to be seen” Consultations slow and very limited.” (GP20). Moreover, some issues would always (GP51) need to be addressed in person: Technical issues were not confined to “A lot of screening cannot be done patients; practices everywhere had technical over the phone (eg, smears, vaginal issues, with cameras and other equipment examinations, blood pressure, being in short supply, systems that were not diabetes foot examination, before enabled for video calling and systems that school check, proper cardiovascular would not link-up to work together. Rural checks, plus various acute assess- areas had generally poor infrastructure for ments, cancer assessments).” (GP89)

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 93 www.nzma.org.nz/journal article

Paying for telehealth standard fee. A simple question can Patients who benefitted the most from not lead down an email trail and it is having to take time off work or travel long hard to know when to mention a fee.” distances were reported to pay readily and (GP7) promptly for the greater convenience and Another issue arose when an in-person efficient use of their time: visit was needed after a telehealth “We had little objection to paying for appointment. There was a perceived level of telehealth and we are actively trying resistance from patients to pay for both: to promote this method of delivering “A problem though if in fact they healthcare as it allows both patient need to come in after the telephone and healthcare worker more flexi- consult. Do we charge twice?” (GP62) bility. Last week I had two consults Changes in use of telehealth over with patients on their long commute to work!” (GP47) time during the pandemic Respondents reported a decline in the However, there were also many comments proportion of consultations conducted by about patient resistance to paying for tele- telehealth between survey two and survey health consultations. Respondents reported three, corresponding with the change from that telehealth appeared to be viewed as an COVID-19 Alert Level 2 to 1.2 There was no inferior form of appointment and that some subsequent change in proportion during patients didn’t understand “the time and survey four. There was a strong feeling from level of care that a health professional can respondents that they needed to catch-up take to deal with a patient’s health” (N7). on concerns that patients had put on hold Telehealth appeared to be viewed as an because they wanted to discuss them in inferior form of appointment: person: “Patients are complaining and “…rebound effect—things that could refusing to pay for telephone consults have waited during Level 4 but as they feel they should only have to now have become more urgent for pay if they see the doctor face to face. in-person consult.” (GP72) There seems to be an expectation that Moreover, as respondents pointed out, telephone consults are the same as Dr the ability to undertake telehealth work had triage and should be free even if they initially been aided by having good data on are a full consult.” (GP4) patients: Practices learned quickly that it was “We were able to manage most things necessary to give a clear explanation about remotely precisely because we have payment before the appointment and make high-quality data on our patients— payment easy: most had up-to-date BP, height, “We noticed early on patients weight and bloods, so it was not complaining about paying for hard to make do for three months. telehealth services, and so now Now we are having to update those the reception team explains every for patients, so that when there booking that there is the same costs is a recurrence of COVID in the as a face-to-face consult, so there is community we will be able to do it no confusion or surprise.” (GP80) again.” (GP63) Nonetheless, there were complications Apart from these purely clinical consid- in certain situations. The same patients erations, some respondents reported that who had no internet access were unable to they tended to prefer seeing patients in pay by online banking. Charging for email person and believed many patients felt the exchanges with patients was also difficult to same way, as “face-to-face appointments standardise: are in demand” (GP54). One GP succinctly “I find it really hard to charge for remarked: email questions and that can take “People crave human contact. Don’t up a lot of time—the work required you? Never underestimate its value.” varies so much it is hard to have a (GP81)

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 94 www.nzma.org.nz/journal article

Nevertheless, there was a generally “We are also preparing for the next positive expectation that telehealth would wave of COVID so needed to keep the have a place in the future as a “useful tool in expectation that some consults would the toolbox” (GP87). be via phone.” (N20) Perceived future use of telehealth Most respondents agreed that telehealth Discussion had proved its usefulness, and that over Based on a nationwide sample, this paper time it would become a normal component describes the experiences of general practice in the primary care model. Its greater flex- teams in New Zealand who used telehealth ibility and convenience were considered a as an emergency response tool during the major driving factor: COVID-19 pandemic to provide healthcare “It is going to be a permanent part of while minimising the potential spread of our practice moving forward, as the infection. general feedback from patients has Telehealth consultations were reported been that they like the accessibility of as being most successful where there was a it, and the lesser time it takes, instead pre-existing relationship between healthcare of waiting in a GP waiting room.” provider and patient, which was also found (N18) in a New Zealand patient experience study However, comments showed that, to performed at a similar time during early consolidate and enhance the benefits that stages of the pandemic.9 Another finding telehealth demonstrated, further changes that parallels that study is the need for would be necessary. Firstly, practices needed healthcare providers to clearly commu- to clarify exactly what they offered via nicate to their patients the process and telehealth, so that “patients have a better cost of telehealth.9 Many similar barriers understanding of what can be done” (GP50). were highlighted by patients, including These included: technological challenges, communication difficulties for those with hearing impair- “…non-acute consultations that do not ments, concern regarding the cost and require physical examination, point- difficulty in making online payments.9 of-care testing, etc… telehealth will eventually become the norm.” (NP8) The experience described through this study is consistent with other interna- Secondly, better infrastructure, more tional work showing that telehealth risks reliable technology for practices and increasing inequity.15,16 Potential healthcare improved access for patients was needed benefits of telehealth can be seen in those before telehealth could be implemented who are already well connected to tech- equitably: nology; however, it can create extra barriers “There is a place for this, but better for those who are already disadvantaged, technology is needed and better such as those in rural areas, those with support for when problems happen— hearing impairment or cognitive decline and for example, [the telecommunications refugee and migrant populations who may provider] lost our connection and we have language barriers. had no phones for three days!” (N27) Ultimately, despite initially high optimism Thirdly, the additional costs to practices of from general practice teams about the implementing telehealth options needed to ongoing use of telehealth, when COVID-19 be adequately recognised by funders: Alert Level restrictions eased, we found “We would need built in funded its use fell rapidly. The rapid move back to cameras and video software funded in-person care and ‘business as usual’ was by the DHB/PHO.” (GP16) felt by the GP teams to be driven by patient “Funding grant for improved IT, choice. So although telehealth may play an video consulting and patient portals.” increasing role in the future, it is unlikely to (GP62) fully replace in-person care. The cost of tele- health placed further financial stress on the Finally, having telehealth capability business model of many of these practices available was an important part of being during the COVID-19 pandemic. Additional pandemic ready:

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 95 www.nzma.org.nz/journal article

technology costs were largely borne by • Provision and enablers for tele- practices. health need to be considered when The strengths of this study include primary healthcare funding is revised reporting from not only GP respondents, following the New Zealand Health 18 but also nurses, nurse practitioners and and Disability System Review 2020. practice managers. We followed the same This could include suitable funding large group of participants through various to reduce barriers and innovative stages of the initial COVID-19 pandemic in ways of improving patients’ access to New Zealand and Alert Level changes from technology and their technological the end of lockdown. Limitations include literacy, especially for older patients 19 data collection being largely by textual and those with disabilities. survey only. However, from a logistical • For practices, we suggest a focus on viewpoint commencing during a pandemic being technology ready and devel- lockdown with busy healthcare teams, this oping protocols and training in was deemed the most feasible. Although telehealth for the GP team to ensure participant sampling was not stratified, we quality. New methods of doing virtual sought to have a variety of representation physical examinations could be and geographical spread from throughout developed with a new vocabulary for New Zealand. this. COVID-19 has thrown into sharp focus • For telehealth implementation to be the question of how telehealth can be most successful, we envisage inte- further integrated into general practice gration of telehealth with existing models of healthcare. Recent literature has primary care health services (aiding discussed key requirements for long-term continuity of care and whanaun- sustainability of telehealth post COVID-19, gatanga (relationships)), beneficial through “(a) developing a skilled work- sharing between networks and a force; (b) empowering consumers; (c) strong emphasis on reducing inequity. reforming funding; (d) improving the digital New Zealand general practices showed ecosystems; and (e) integrating telehealth agility and adaptability in embracing 17 into routine care.” Based on the findings of telehealth during the initial stages of the our study, we recommend the following: COVID-19 pandemic. It is important to • The effective use of telehealth reflect on their experience and then ensure depends on both individual practice adequate funding and practice processes and patient capabilities, and we are in place so telehealth can not only be recommend good communication swiftly used in future pandemics, but also with patients regarding expectations, become an integral part of the model of care processes and costs. in everyday general practice.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 96 www.nzma.org.nz/journal article

Appendix Primary qualitative analysis of telehealth survey questions • What limitations have your patients experienced in the use of technology for telehealth? • What place do you think telehealth will have in your practice for delivering patient care in the future? • What is your impression of patients willingness to pay for telehealth services? • Our recent survey responses have revealed low levels of ongoing use of telehealth since dropping down to Alert Level 1. Why do you think this has occurred? General GPPENZ survey questions for secondary analysis Note: Those with * not included in secondary analysis for this telehealth paper. Survey 1 Date: 8 May to 4 June 2020 These questions relate to the last two months of the Covid-19 Pandemic in New Zealand. • Can you describe how the Covid-19 pandemic has affected your feelings until now?* • What changes have you and your team made to how you practice in response to the Covid-19 pandemic? • What opportunities have you experienced over this time? Including the use of virtual consultations and new technology. • What challenges have you experienced over this time? • What do you think has been successful so far? • If you knew what you know now, what might you have done differently?* • Do you have any comments regarding the information you have received about Covid- 19 and which sources you have found most reliable and useful?* • What has the practice you work in done to protect the health and wellbeing of staff? e.g. personal protective equipment (PPE), stress management, anxiety reduction.* • What effect has the Covid-19 pandemic had on staffing levels at your main practice?* • Do you have any comments on the overall health system response to the Covid-19 pandemic over the last few months?* • Do you have any further thoughts or comments?* Survey 2 Date: 28 May to 18 June 2020 These questions relate to the Covid-19 Pandemic in New Zealand. • What are your impressions about non-Covid 19 related health conditions during this pandemic for your patients and your practice? • What changes have you and your team made to support Māori, Pasifika and high health needs groups since the beginning of the Covid-19 pandemic? • Have you had any significant events, incidents or near misses that have (or could have) caused harm to a patient, as a result of the Covid-19 pandemic? • If yes, you can describe the event in the question below. Please describe any events: • Since you completed the last GPPEC or GPPENZ survey, what changes have you and your team made to how you practice? Including changes associated with the move to Alert Level 2. • What do you see are the issues for your practice as you move through the winter months? • Do you have any further thoughts or comments?

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 97 www.nzma.org.nz/journal article

Survey 3 Date: 18 June to 9 July 2020 These questions relate to Alert Level 1 during the Covid-19 Pandemic in New Zealand. • Since the move to Alert Level 1, how have you been feeling about the Covid-19 pandemic? • Since the move to Covid-19 Alert Level 1, what changes have you and your team made to how you practise? • Please describe how respiratory and non-respiratory patients are currently streamed in your practice? • Please include challenges this has posed. These questions relate to the Covid-19 Pandemic in New Zealand. • What changes have been made at your practice to the way prescriptions are managed, since the beginning of the Covid-19 pandemic until now?* • What experience have you and your patients had with access to and management from secondary care during the Covid-19 pandemic? • What was the immediate and longer term financial impact of the Covid-19 pandemic on yourself and your practice? • Do you have any further thoughts or comments? Survey 4 Date: 9th July to 30th July 2020 These questions relate to the Covid-19 Pandemic in New Zealand. • Do you have any comments about the personal and/or professional support you have received since the beginning of the Covid-19 pandemic? Please include whether it was helpful or not. • What additional personal or professional support would you have liked to receive during the Covid-19 pandemic so far? * • In your patient interactions and consultations, what common issues have you observed about the financial effects of the Covid-19 pandemic on their healthcare? These questions relate to the time since the move to Alert level 1 in the Covid-19 pandemic (12am 9th June 2020). • Since Alert Level 1, what has the main practice you work in done to promote stress reduction or maintain psychological wellbeing for staff? * • Since the move to Alert level 1, do you have any comments regarding the clarity and consistency of information you have received, and which sources you have found the most useful? * • Now that we are in Alert Level 1, have staffing levels at your main acticepr returned to pre-Covid levels? If not, please comment on how it is different. * These questions relate to the change in Ministry of Health case definition for Covid-19 from Wednesday 24th June 2020. • Since the most recent case definition change, what changes have you made to how you practice and access patient Covid-19 testing. • Please include any challenges this has posed. • Since the most recent case definition change, how are you managing patients with low risk respiratory illness, including any PPE you use? • Do you have any further thoughts or comments?

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 98 www.nzma.org.nz/journal article

Survey 5 Date: 6th August to 27th August 2020 These questions relate to the Covid-19 Pandemic in New Zealand. • From your experience during the Covid-19 pandemic, what learning, changes or inno- vations would you like to see embedded in future general practice, in your practice and or generally? • Do you have any thoughts on if and how Māori, Pasifika and high needs health groups have been particularly affected over the last few months of the Covid-19 pandemic? • In what ways have you and your team adapted to support Māori, Pasifika and high health needs groups since the beginning of the Covid-19 pandemic? • Have you had any other significant events, incidents or near misses that have (or could have) caused harm to a patient, as a result of the Covid-19 pandemic? • If yes, you can describe the event in the question below. Please describe any events: • In the last month, have you been aware of any of your patients delaying seeking care, or presenting late due to the ongoing effects of the Covid-19 pandemic? If so, please describe further. • How do you think the healthcare system should be organising surveillance testing for Covid-19 at this time (with no evidence of community transmission)? * These questions relate to a potential second wave of Covid-19 infections. • In your main practice, do you feel adequately prepared for a potential second wave of Covid-19 community tramismitted infections? • If yes, how have you and your practice prepared? • If ‘maybe’ or ‘no’, what do you feel you and your practice would need to do to be prepared? • Have you considered and discussed as a practice how much PPE you would require to store for a potential second wave of Covid-19 infections? If so, please estimate this. * • What changes to General Practice funding would you like to see to help deal with a potential second wave of the Covid-19 pandemic? • What do you believe the wider New Zealand healthcare system should be doing now to prepare for a second wave of Covid-19 in New Zealand? • How concerned are you about a second lockdown as a consequence of uncontrolled community transmission of Covid-19, and the effect that a second lockdown would have on you, your practice and your patients? * • Do you have any further thoughts or comments? *

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 99 www.nzma.org.nz/journal article

Competing interest: Nil. Acknowledgments: Thank you to medical students Umaya Ranaweera and Harrison Beadel for their help with survey design and data collection. We thank Pegasus Health (Charitable) Ltd for providing statistical and quantitative analysis and consultation with the Director of Hauora Māori and Equity, Irihāpeti Mahuika. Lastly, we thank the large number of primary healthcare professionals who gave their time and insights by participating in this study during such a busy time in their working and personal lives. Author information: Geraldine Wilson: General Practitioner, Senior Research Fellow, Department of General Practice, University of Otago, Christchurch. Olivia Currie: General Practitioner, Senior Research Fellow, Department of General Practice, University of Otago, Christchurch. Susan Bidwell: Senior Research Fellow, Department of General Practice, University of Otago, Christchurch. Baraah Saeed: Medical Student, University of Otago, Christchurch. Anthony Dowell: Professor of Primary Health Care and General Practice, University of Otago, Wellington. Andrew Adiguna Halim: Analyst, Pegasus Health (Charitable) Limited. Les Toop: Professor of General Practice, Department of General Practice, University of Otago, Christchurch. Ann Richardson: Retired, Professor of Public Health. Ruth Savage: Senior Lecturer, Department of General Practice, University of Otago, Christchurch. Ben Hudson: Senior Lecturer Department of General Practice, University of Otago, Christchurch. Corresponding author: Dr Geraldine Wilson, Department of General Practice, University of Otago, Christchurch, PO Box 4345, Christchurch 8140, 03 364 3613 [email protected]. URL: www.nzma.org.nz/journal-articles/empty-waiting-rooms-the-new-zealand-general-practice- experience-with-telehealth-during-the-covid-19-pandemic-open-access

REFERENCES 1. Royal New Zealand College 3. Dorsey ER, Topol EJ. State tive Assessment of Rapid of General Practitioners of Telehealth. N Engl J System Transformation [Internet]. GPs open for Med. 2016;375:154-61. to Primary Care Video business – but changing 4. Baddock K. COVID-19-the Visits at an Academic the way they see patients frontline (a GP perspective). Medical Center. Ann Intern Wellington: Royal New N Z Med J. 2020;133:8-10. Med. 2020;173:527-535. Zealand College of General 5. Atmore C, Stokes T. Turning 8. Verhoeven V, Tsakitzidis Practitioners; 2020 [24 on a dime-pre- and post- G, Philips H, Van Royen P. Nov 2020]. Available COVID-19 consultation Impact of the COVID-19 from: https://www. patterns in an urban pandemic on the core rnzcgp.org.nz/RNZCGP/ general practice. N Z functions of primary News/College_news/2020/ Med J. 2020;133:65-75. care: will the cure be GPs-open-for-business.aspx worse than the disease? 6. Marshall M, Howe A, 2. New Zealand Government A qualitative interview Howsam G, et al. COVID-19: [Internet]. Unite against study in Flemish GPs. BMJ a danger and an oppor- COVID-19: History of the Open. 2020;10:e039674. tunity for the future of COVID-19 Alert System general practice. Br J Gen 9. Imlach F, McKinlay E, [24 Nov 2020]. Available Pract. 2020;70:270-271. Middleton L, et al. Tele- from: https://covid19.govt. health consultations in 7. Srinivasan M, Asch S, nz/alert-system/history-of- general practice during the-covid-19-alert-system/ Vilendrer S, et al. Qualita-

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 100 www.nzma.org.nz/journal article

a pandemic lockdown: Health Organisations. Last Health. JMIR Mhealth survey and interviews updated 7 Jan 2020. Avail- Uhealth. 2020;8:e14512. on patient experiences able from: https://www. 17. Thomas EE, Haydon and preferences. BMC health.govt.nz/our-work/ HM, Mehrotra A, et Fam Pract. 2020;21:269. primary-health-care/ al. Building on the 10. Qualtrics, Qualtrics version about-primary-health-or- momentum: Sustaining May-August 2020, First ganisations telehealth beyond COVID- release: 2005, Copyright 14. The Royal New Zealand 19. J Telemed Telecare. Year 2020, Provo, Utah, College of General Prac- 2020:1357633X20960638. USA. Available from: titioners [Internet]. 2018 18. Health and Disability https://www.qualtrics.com general practice workforce System Review. 2020. 11. QSR International (1999) survey, New Zealand, 2019. Health and Disabil- NVivo Qualitative Data Available from: https:// ity System Review Analysis Software [Version rnzcgp.org.nz/gpdocs/ –Final Report –Pūrongo 12]. Available from https:// New-website/Publications/ Whakamutunga. Welling- qsrinternational.com/ GP-Workforce/Workforce- ton: HDSR. Available from: nvivo/nvivo-products/ Survey2018Report1-re- www.systemreview.health. vised-July-20194web.pdf 12. Kolb SM. Grounded govt.nz/final-report. theory and the constant 15. Mishori R, Antono B. 19. Tan LF, Ho Wen Teng V, comparative method: Telehealth, Rural America, Seetharaman SK, Yip AW. valid research strategies and the Digital Divide. Facilitating telehealth for for educators. Journal J Ambul Care Manage. older adults during the of Emerging Trends in 2020;43:319-322. COVID-19 pandemic and Educational Research and 16. Brewer LC, Fortuna KL, beyond: Strategies from Policy Studies. 2012;3:83-6. Jones C, et al. Back to the a Singapore geriatric 13. Ministry of Health Future: Achieving Health center. Geriatr Gerontol [Internet]. About Primary Equity Through Health Int. 2020;20:993-995. Informatics and Digital

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 101 www.nzma.org.nz/journal article

Emergency COVID-19 funding to general practices in early 2020: lessons for future allocation to support equity Vanessa Selak, Sue Crengle, Matire Harwood, Samantha Murton, Peter Crampton

ABSTRACT AIM: To (1) describe the distribution of Ministry of Health (MOH) COVID-19 emergency funding to general practices in March and April 2020 and (2) consider whether further funding to general practices should be allocated differently to support equity for patients. METHODS: Emergency funding allocation criteria and funding amounts by general practice were obtained from the MOH. Practices were stratified according to their proportion of high-needs enrolled patients (Māori, Pacific or living in an area with the highest quintile of socioeconomic deprivation). Funding per practice was calculated for separate and total payments according to practice stratum of high-needs enrolled patients. RESULTS: The median combined March and April funding for general practices with 80% high-needs patients was 28% higher per practice ($36,674 vs $28,686) and 48% higher per patient ($10.50 vs $7.11) compared with the funding received by general practices with fewer than 20% high-needs patients. Although the March allocation did increase funding for high-needs patients, the April allocation did not. CONCLUSIONS: Emergency support funding for general practices was organised by the MOH at short notice and in exceptional circumstances. In the future, the MOH should apply pro-equity resource allocation in all emergencies, as with other circumstances.

eneral practitioners (GPs) operate out Funding to general practices from the of general practices, most of which Ministry of Health (MOH) for the delivery Gare privately owned and funded by a of primary care services has mainly been in combination of patient fees and government the form of capitation payments since the subsidies. The vision of the 2001 Primary early 2000s.3 These payments are deter- Health Care Strategy included that “prima- mined by the enrolled practice population ry health care services will focus on better and the capitation funding formula, which health for a population, and actively work to has evolved over time.3 Currently, practices reduce health inequalities between different are entitled to receive a standard capi- groups.”1 A key mechanism for supporting tation payment that varies according to the health equity is reducing access barriers to number and characteristics of their enrolled primary healthcare services.1 The Primary patients (age group, gender, whether or Health Care Strategy recognised that patient not the patient has a High Use Health Card fees were a significant barrier to general [HUHC]) and whether or not the practice practice services for low-income families. is an Access practice.4 Practices can opt to The strategy sought to reduce these by receive the following additional capitation increasing government subsidies to general payments according to the following funding practice.2 schemes:

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 102 www.nzma.org.nz/journal article

• Very low-cost access (VLCA): If at mented on Monday 23 March, New Zealand least 50% of patients enrolled in the went into Alert Level 3 (Restrict) because practice are high-needs (ie, Māori, community transmission of COVID-19 was Pacific or living in an area with the suspected, and after a further 48 hours, on highest quintile of socioeconomic Wednesday 25 March, New Zealand was in deprivation) and the practice agrees Alert Level 4 (Lockdown).5 Virtually over- to charge zero fees for their patients night there was a 50–80% reduction in GP aged 0–13 years, and no more than consultations and, consequently, patient a specified upper limit for older fees and cash flow too. This was likely due patients. to fear and anxiety about catching COVID-19 • Community services card (CSC) (for and the (false) perception that general non-VLCA practices): If the practice practices were overwhelmed by patients 5 agrees to charge no more than a spec- with COVID-19. The MOH agreed to provide ified upper limit for patients aged 14 emergency funding to support general prac- years and older. tices in March and April 2020 due to serious concerns about their financial sustainability. • Zero fees for under 14s or 6s (for non-VLCA practices): If the practice The aims of this study were to describe the agrees to charge zero fees for their distribution of MOH COVID-19 emergency patients aged 0–13 or 0–5 years, funding to general practices in March and respectively.4 April 2020, and to consider whether future emergency funding to general practices Despite increasing government subsidies should be allocated differently to better to general practice, cost continues to be support equitable access for the patients a significant barrier to general practice experiencing the greatest financial access services.2 In 2018/2019, 13.4% of New barriers to GP services. Zealand adults reported that in the previous year they had a medical problem for which they hadn’t visited a GP due to cost.2 The Methods proportion reporting GP cost access barriers The following information was obtained is higher among Māori (21.9%, 95% CI 19.7– from the MOH regarding the emergency 24.2%) and Pacific people (19.4%, 95% CI funding that had been distributed by the 16.4–22.7%) than among Europeans (12.7%, MOH during March and April 2020 to 95% CI 11.8–13.6%).2 GP cost access barriers support general practices in their COVID-19 are also higher among people living in areas response: with the most socioeconomic deprivation • Total amount distributed (quintile 5: 19.2%, 95% CI 17.4–21.1%) than • Amount by general practice those living in areas with the least depri- vation (quintile 1: 9.5%, 95% CI 7.9–11.3%).2 • Criteria used to allocate the funding, in particular any patient and/or The first case of COVID-19 was confirmed practice characteristics that were in New Zealand on 28 February 2020.5 taken into consideration There was considerable activity and anxiety among GPs and general practices as they • Practice primary health organisation struggled to keep up with frequent changes (PHO) affiliation and funding scheme of advice from the MOH, to obtain adequate (VLCA, CSC, zero fees for under 14s or supplies of personal protective equipment 6s) (PPE) and influenza vaccines amid supply • Number of patients enrolled within and distribution issues, and to contain the each practice by each of the following potential risks of COVID-19 to their staff and variables: ethnicity (prioritised, patients.5 General practice faced further level 1), socioeconomic deprivation considerable challenges when, on Saturday quintile, age group, CSC holder status, 21 March, the Royal New Zealand College HUHC status, high needs of General Practitioners (RNZCGP) recom- Practices were categorised according mended that GPs switch from face-to-face to the proportion of their patients who to virtual consultations.5 Just as the switch are high-needs enrolled patients: 0–19%, to virtual consultations was being imple- 20–39%, 40–59%, 60–79%, 80%+. Funding per

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 103 www.nzma.org.nz/journal article

enrolled patient was calculated for separate geted patient), with a minimum payment and total payments according to practice of $2,500 per practice and an additional stratum of high-needs enrolled patients. All payment of $5,000 for practices with 50% funding figures were exclusive of goods and or more targeted patients. The payment services tax (GST). Thirty-six practices were in April was to support general practices’ excluded from the analysis (see results). business sustainability and was based on The information was requested under the the rurality of the practice ($4.69 per patient Official Information Act (Ref H202003461) in non-rural practices and $4.92 per patient on 18 May 2020 and was received on 16 June in rural practices). The third payment, 2020. $7.8 million in March, was distributed by district health boards (DHBs) to practices on a fee-for-service basis to fully fund general Results practice-based COVID-19 assessments. General practices received three sources Details of the third payment were not able to of emergency COVID-19 funding, with a be provided by the MOH and are therefore total value of $45.22 million, from the MOH not included in this analysis. during March and April in 2020 (Table 1). Data were provided for a total of 951 Two payments were distributed by the practices, of which 915 (96%) were included MOH: $15 million in March and $22.42 in this analysis. Of the 36 excluded prac- million in April. The payment in March tices, eight (0.8%) were excluded because was to support capacity and capability in the practice received funding in March primary care. That payment was based on but did not have enrolment data in April, the characteristics of the enrolled practice three (0.3%) because they did not receive population ($4.50 per targeted patient both March and April MOH payments and [ie, Māori, Pacific, NZDep quintile 5 and/ 25 (2.6%) because they only received the or >65 years of age], $1.50 per non-tar-

Table 1: General practice emergency funding, by package and allocation criteria.

Funding package Amount Allocation criteria

Allocated by the Ministry of Health

March 2020: Primary Care $15 million Patient Support – Capacity and • $4.50 per targeted EP (Māori, Pacific, NZDep quintile Capability Funding 5 and/or >65 years) • $1.50 for other EPs Practice • $2,500 minimum payment per practice • $5,000 additional payment for practices with 50% or more targeted EPs

April 2020: Additional $22.42 • $4.69 per EP (non-rural practice) immediate funding support million • $4.92 per EP (rural practice) for general practice business sustainability

Allocated by DHBs March 2020: General Practice $7.8 million Provided to DHBs on a PBFF basis to fully fund general based COVID-19 Assessments practice assessments and testing. DHBs set nationally consistent funding criteria and provided funding to PHOs and to general practices on a fee-for-service basis.

DHB = district health board, EP = enrolled patient, NZDep = New Zealand socioeconomic deprivation index, PHO = Primary Health Organisation, PBFF = population-based funding formula.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 104 www.nzma.org.nz/journal article

minimum MOH payment ($2,500) in March. funding. Practices with 80% or more high- Of the included practices, VLCA was the needs patients (with 7% of all enrolled funding scheme for 271 (30%), CSC and free patients) received 12% of the March, 7% of under 14 year olds for 587 (64%), free under the April and 9% of the combined funding. 14 or 6 years olds for 38 (4%), CSC only for The median March funding was $2.38 per 12 (1%) and seven (1%) had elected not to patient ($9,264 per practice) for practices receive any additional funding. Most prac- with fewer than 20% high-needs patients. tices (377, 41%) had under 20% high-needs For practices with 80% or more high-needs patients, with 287 (31%) having 20–39% patients, the median March funding was high-needs patients, 109 (12%) having $5.77 per patient ($20,092 per practice). The 40–59%, 73 (8%) having 60–79% and 69 (8%) median April funding was $4.69 per patient having 80% or more high-needs patients. irrespective of the proportion of high-needs The proportion of VLCA practices increased patients. The median April funding per from 4% (n=16) among practices with under practice ranged from $16,354 for practices 20% high-needs patients, up to 97% (n=67) with 80% or more high-needs patients and among practices with 80% or more high- $20,022 for practices with fewer than 20% needs patients. high-needs patients. In April 2020, a total of 4,721,710 patients The median combined March and April were enrolled in practices included in funding for practices with fewer than 20% this study (Table 2). Most patients (40%) high-needs patients was $28,686 per practice were enrolled in practices with under 20% and $7.11 per patient; corresponding figures high-needs patients, with 33% enrolled in for practices with 80% or more high-needs practices with 20–39% high-needs patients, patients were $36,674 per practice and 12% in practices with 40–59%, 8% in prac- $10.50 per patient, respectively. The median tices with 60–79% and 7% of patients combined March and April funding for enrolled in practices with 80% or more high- practices with 80% high needs patients was needs patients. The proportion of younger 28% higher per practice and 48% higher per patients (aged 0–19 years) increased from patient than the practices with fewer than 24% among patients enrolled with prac- 20% high-needs patients. tices with under 20% high-needs patients, up to 34% among patients enrolled in practices with 80% or more high-needs Discussion patients. The proportion of Māori and The median combined March and April Pacific patients enrolled with practices MOH COVID-19 emergency funding for with under 20% high-needs patients was general practices with 80% high-needs 6% and 2%, respectively, increasing to 40% patients was 28% higher per practice and 36%, respectively, in practices with 80% ($36,674 vs $28,686) and 48% higher per or more high-needs patients. Similarly, the patient ($10.50 vs $7.11) compared with proportion of those living in an area in the the funding received by general practices highest socioeconomic deprivation quintile with fewer than 20% high-needs patients. and those with a CSC increased from 5% and Although the March allocation did increase 15%, respectively, in practices with under funding for high-needs patients, the April 20% high-needs patients, up to 64% and allocation did not, thereby attenuating 38%, respectively, in practices with 80% or the pro-equity distribution of emergency more high-needs patients. The proportion of funding to general practices in terms of the patients with a HUHC card was low across needs of Māori, Pacific patients and those all strata of practices. living in areas of the highest socioeconomic deprivation. The total March amount (capacity and capability funding) paid to practices was While the burden of COVID-19 in New $14,849,697, and $22,311,373 was paid in Zealand has remained modest to date, April (business sustainability funding) (Table likely largely due to the national elimi- 6 3). Practices with fewer than 20% high-needs nation strategy, commentators have noted patients (with 40% of all enrolled patients) that “the health impact of COVID-19 will be received 30% of the March funding, 40% of greater for Māori” and “the strong focus the April funding and 36% of the combined on numerical age as a risk factor is to the

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 105 www.nzma.org.nz/journal article

Table 2: Patient characteristics by practice stratum of high-needs enrolled patients.

Characteristics % high-needs patients enrolled in the practice in March 2020 Total of patients enrolled in the practice in April 0-19% 20-39% 40–59% 60–79% 80–100% 2020 Number of 1,894,575 (40%) 1,561,350 (33%) 555,321 (12%) 366,046 (8%) 344,418 (7%) 4,721,710 people*

Age group (years) 0-19 463,519 (38%) 392,945 (32%) 150,193 (12%) 110,268 (9%) 118,092 (10%) 1,235,017 (26%)

20-39 467,718 (38%) 409,412 (33%) 154,727 (13%) 100,047 (8%) 100,119 (8%) 1,232,023 (26%)

40-59 515,327 (43%) 396,429 (33%) 135,362 (11%) 88,432 (7%) 76,943 (6%) 1,212,493 (26%)

60-79 366,847 (43%) 298,197 (35%) 95,709 (11%) 58,407 (7%) 43,285 (5%) 862,445 (18%)

80+ 81,164 (45%) 64,367 (36%) 19,330 (11%) 8,892 (5%) 5,979 (3%) 179,732 (4%)

Ethnicity (prioritised) Māori 118,172 (6%) 201,147 (13%) 128,180 (23%) 136,945 (37%) 136,194 (40%) 720,638 (15%)

Pacific 39,367 (2%) 65,695 (4%) 50,737 (9%) 63,447 (17%) 125,169 (36%) 344,415 (7%)

Asian 278,840 (15%) 224,232 (14%) 82,796 (15%) 43,744 (12%) 32,049 (9%) 661,661 (14%)

European 1,414,910 (75%) 1,036,471 (66%) 276,384 (50%) 115,801 (32%) 46,872 (14%) 2,890,438 (61%)

Other / not stated 43,286 (2%) 33,805 (2%) 17,224 (3%) 6,109 (2%) 4,134 (2%) 104,558 (2%)

Socioeconomic deprivation quintile 5 (most deprived) 90,863 (5%) 243,172 (16%) 174,131 (31%) 171,890 (47%) 221,088 (64%) 901,144 (19%)

4 231,393 (12%) 363,270 (23%) 134,516 (24%) 87,018 (24%) 58,081 (17%) 874,278 (19%)

3 366,343 (19%) 347,434 (22%) 97,435 (18%) 45,493 (12%) 27,878 (8%) 884,583 (19%)

2 501,693 (26%) 299,919 (19%) 74,360 (13%) 30,107 (8%) 17,061 (5%) 923,140 (20%)

1 (least de- 649,353 (34%) 263,202 (17%) 59,634 (11%) 20,049 (5%) 10,309 (3%) 1,002,547 (21%) prived)

Missing 54,930 (3%) 44,353 (3%) 15,245 (3%) 11,489 (3%) 10,001 (3%) 136,018 (3%)

High needs 231,098 (12%) 434,833 (28%) 273,230 (49%) 253,894 (69%) 298,615 (87%) 1,491,670 (32%)

CSC holders 281,640 (15%) 336088 (22%) 155082 (28%) 122442 (33%) 129854 (38%) 1025106 (22%)

HUHC holders 2,452 (0%) 4,722 (0%) 1,243 (0%) 693 (0%) 359 (0%) 9,469 (0%)

Practice funding scheme VLCA 87,923 (5%) 320,118 (21%) 341,997 (62%) 361,069 (99%) 334,360 (97%) 1,445,467 (31%)

CSC and free <14y 1,662,325 (88%) 1,181,961 (76%) 210,923 (38%) 4,675 (1%) 10,058 (3%) 3,069,942 (65%)

Free <6y or <14y 115,652 (6%) 56,411 (4%) 2,401 (0%) - - 174,464 (4%)

CSC only 6,303 (0%) 1,421 (0%) - 195 (0%) - 7,919 (0%)

None 22,372 (1%) 1,439 (0%) - 107 (0%) - 23,918 (1%)

Number of patients (%), % within columns (ie, within practice stratum of high-needs enrolled patients, or across total patients) unless otherwise specified. * % across rows. CSC=Community Services Card, High needs = Māori, Pacific or living in an area with the highest quintile of socioeconomic deprivation, HUHC=High Use Health Card, VLCA=Very low-cost access.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 106 www.nzma.org.nz/journal article

Table 3: Funding by practice stratum of high-needs enrolled patients.

% high-needs patients enrolled in the practice in March 2020 Total

0–19% 20–39% 40–59% 60–79% 80–100%

Total funding, $ (% across rows) March $4,516,647 (30%) $4,513,993 (30%) $2,305,574 (16%) $1,742,475 (12%) $1,770,938 (12%) $14,849,627

April $8,942,201 (40%) $7,380,727 (33%) $2,625,340 (12%) $1,737,385 (8%) $1,625,720 (7%) $22,311,373

Total $13,458,848 (36%) $11,894,720 (32%) $4,930,914 (13%) $3,479,859 (9%) $3,396,658 (9%) $37,160,999

Funding by practice (not adjusted by practice size) Median (Q1,Q3) $9,264 ($5,726, $15,626) $11,738 ($7,216, $19,306) $18,078 ( $11,983, $25.033) $17,674 $12,178, $20,092 ($13,781, $12,616 ($7,500, $27,980) $31,954) $20,612)

March Mean (SD) $11,980 ($8,314) $15,728 ($13,041) $24,265 ($17,374) $23,870 ($17,804) $25,666 ($18,634) $16,229 ($13,352)

Median (Q1,Q3) $19,271 ($11,022, $31,493,) $18,268 ($10,963, $32,891) $20,022 ($10,778, $28,651) $16,471 ($9,186, $28,464) $16,354 ($10,004, $18,676 ($10,780, $31,831) $31,371)

April Mean (SD) $23,719 ($16,961) $25,717 ($21,830) $25,277 ($22,322) $23,800 ($22,353) $23,561 ($21,547) $24,386 ($19,874)

Median (Q1,Q3) $28,686 ($16,434, $47,307) $29,662 ($18,369, $51,720) $38,947 ($23,015, $53,363) $33,971 ($20,940, $56,223) $36,674 ($24,022, $31,062 ($18,725, $64,983) $51,598)

Total Mean (SD) $35,700 ($25,120) $41,445 ($34,661) $53,425 ($43,451) $47,669 ($40,143) $49,227 ($40,097) $40,613 ($32,424)

Funding by patient enrolled in April Median (Q1,Q3) $2.38 ($2.21, $2.54) $2.75 ($2.61, $2.90) $4.40 ($4.00, $5.33) $5.18 ($4.54, $6.28) $5.77 ($4.97, $6.69) $2.70 ($2.41, $4.31)

March Mean (SD) $3.56 ($15.61) $3.45 ($5.30) $5.51 ($4.23) $6.85 ($8.53) $10.07 ($30.63) $4.51 ($13.80)

Median (Q1,Q3) $4.69 ($4.69, $4.69) $4.69 ($4.69, $4.69) $4.69 ($4.69, $4.69) $4.69 ($4.69, $4.69) $4.69 ($4.69, $4.69) $4.69 ($4.69, $4.69)

April Mean (SD) $4.73 ($0.09) $4.73 ($0.09) $4.73 ($0.08) $4.75 ($0.10) $4.73 ($0.09) $4.73 ($0.09)

Median (Q1,Q3) $7.11 ($6.91, $7.30) $7.47 ($7.31, $7.66) $9.14 ($8.71, $10.02) $9.98 ($9.34, $11.11) $10.50 ($9.71, $11.39) $7.42 ($7.13, $9.05)

Total Mean (SD) $8.29 ($15.61) $8.18 ($5.30) $10.24 ($4.23) $11.59 ($8.52) $14.80 ($30.62) $9.24 ($13.79)

CSC=Community Services Card, High needs = Māori, Pacific or living in an area with the highest quintile of socioeconomic deprivation, HUHC=High Use Health Card, Q1=first quartile, Q3=third quartile, SD=standard deviation, VLCA=Very Low Cost Access

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 107 www.nzma.org.nz/journal article

detriment of Indigenous populations”.7 A emergency circumstances. Understanding modelling study that sought to disentangle the reasons for the differences in allocation the effects of age structure and comorbidity between the March and April tranche would has estimated that, if there were widespread be useful and may help to support better community transmission in New Zealand, implementation of equity policy imperatives. the infection fatality rate of COVID-19 for Ideally reasons for these differences in allo- Māori is likely to be at least 50% higher than cation would be pro-actively shared by the for non-Māori.8 The authors of that study MOH in the first instance. Alternatively, or in consider that inequities in the burden of order to obtain more in-depth understanding, COVID-19 for Māori and Pacific compared the reasons for these differences could be with Europeans could be even greater ascertained through case study research. because 2009 influenza H1N1 pandemic The limitations of our study are that we hospitalisation and fatality rates were were not able to include the DHB funding higher for Māori and Pacific people than for for practice-based COVID-19 assessments. Europeans, and Māori and Pacific people We were also unable to take into account are more likely to experience multi-mor- baseline practice financial vulnerability. bidity, avoidable hospitalisations and racism Further, as noted by Hauora, the Waitangi than Europeans. Further, inequities in the Tribunal Health Services and Outcomes infection fatality rate of COVID-19 could Inquiry report on primary healthcare be even greater if differences by ethnicity claims, “In relation to the capitated funding in age-specific health outcomes or unmet formulas, we have found that the formulas healthcare needs are underestimated in disadvantage primary health organisations 8 available data. and providers that predominantly service These predictions of the likely inequitable high-needs populations and particularly burden of COVID-19 according to ethnicity impact on Māori-led primary health organ- have already become manifest in other isations and providers that predominantly countries. Black, Latinx and Indigenous serve these populations.”12 Māori providers, populations in the United States, as well as some of which have general practices, black, Asian and minority ethnic (BAME) “have been underfunded from the outset,”13 groups in England are experiencing an and the Crown (represented by the MOH) increased burden of COVID-19 compared has now committed to “engagement on a with whites.9, 10 An international study methodology for assessing the extent of across 13 countries also found inequities underfunding of Māori primary health in the burden of COVID-19 according to organisations and providers”.14 Finally, ethnicity and income.11 while we assume that additional funding We acknowledge that the emergency is needed with higher proportions of support funding for general practice was patients with high health needs, we have not organised by the MOH at short notice and considered how much additional funding in exceptional circumstances. However, is required according to additional need. although the March tranche was allocated in Further research exploring how much a way that was pro-equity, the April tranche additional funding is required according was not. In the future, the MOH should to additional need is required to ensure apply pro-equity resource allocation in all funding is truly pro-equity.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 108 www.nzma.org.nz/journal article

Competing interests: There was no external funding source for preparing this article. The views, opinions, find- ings and conclusions or recommendations expressed in this paper are strictly those of the authors. They do not necessarily reflect the views of the institution where the authors cur- rently work. The paper is presented not as policy, but with a view to inform and stimulate wider debate. Dr Murton is President of the RNZCGP and she also has an academic appoint- ment at the University of Otago; the views expressed here are her own and not those of the College. Prof Crengle reports other from New Zealand Ministry of Health, personal fees from RNZCGP and other from WellSouth PHN, outside the submitted work. Acknowledgements: Data provided by the Ministry of Health. Author information: Vanessa Selak: Senior Lecturer , Epidemiology & Biostatistics, University of Auckland , Auckland. Sue Crengle: Associate Professor, Preventive and Social Medicine, Dunedin School of Medicine, University of Otago , Dunedin. Matire Harwood: Associate Professor, General Practice and Primary Healthcare, University of Auckland, Auckland. Samantha Murton: Senior Lecturer, Primary Health Care and General Practice, University of Otago, Wellington. Peter Crampton: Professor, Kōhatu, Centre for Hauora Māori, University of Otago, Dunedin. Corresponding author: Vanessa Selak, Epidemiology & Biostatistics, University of Auckland , Private Bag 92019, Auckland 1142, +64 9 923 6509 [email protected] URL: www.nzma.org.nz/journal-articles/emergency-covid-19-funding-to-general-practices-in-ear- ly-2020-lessons-for-future-allocation-to-support-equity-open-access

REFERENCES 1. King A. The Primary Health 2018-58445514-PDF.pdf Chatterjee A. Covid-19 and Care Strategy. Wellington: {Accessed 28 July 2020) health equity - Time to Ministry of Health, 2001. Wellington: TAS, 2018. think big. NEJM 2020;DOI: 2. Ministry of Health. Annual 5. Baddock K. COVID-19—the 10.1056/NEJMp2021209 Data Explorer 2018/19: frontline (a GP perspec- 10. Patel P, Hiam L, Sowemimo New Zealand Health tive). NZMJ 2020;133:8-10. A, et al. Ethnicity and Survey [Data File]. URL: 6. Baker MG, Kvalsvig A, covid-19. Public Health https://minhealthnz. Verrall AJ, et al. New England’s review of shinyapps.io/nz-health-sur- Zealand’s elimination disparities in covid-19 is a vey-2018-19-annual-da- strategy for the COVID-19 serious missed opportunity. ta-explorer/. Wellington: pandemic and what is BMJ 2020;369:m2882. Ministry of Health, 2019. required to make it work. 11. Shadmi E, Chen Y, Dourado 3. Crampton P. The ongoing NZMJ 2020;133:10-4. I, et al. Health equity evolution of capitation 7. McLeod M, Gurney J, Harris and COVID-19: global funding for primary care: R, et al. COVID-19: we must perspectives. Int J Equity the December 2018 not forget about Indigenous Health 2020;19:104. PHO capitation funding health and equity. ANZ 12. Waitangi Tribunal. Hauora. changes for Community J Public Health 2020;doi: Report on Stage One of Services Card holders. 10.1111/1753-6405.13015 the Health Services and NZMJ 2019;132:69-78. 8. Steyn N, Binny R, Hannah Outcomes Kaupapa Inqui- 4. TAS. PHO Services K, et al. Estimated inequi- ry. WAI 2575. Waitangi Agreement Version 6.0 (1 ties in COVID-19 infection Tribunal Report. Available December 2018). https:// fatality rates by ethnicity from https://forms.justice. tas.health.nz/assets/ for Aotearoa New Zealand. govt.nz/search/Documents/ Primary-psaap-u14/ NZMJ 2020;133:28-39. WT/wt_DOC_152801817/ PHO-Services-Agreement- Hauora%20W.pdf [Accessed 9. Berkowitz SA, Cene CW, Version-6.0-1-December- 11 Sep 2020]. Wellington:

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 109 www.nzma.org.nz/journal article

Waitangi Tribunal 2019. 14. The Crown, Stage One https://forms.justice.govt. 13. Baker G, Baxter J, claimants to the Waitangi nz/search/Documents/ Crampton P. The primary Tribunal’s Health Services WT/wt_DOC_158716675/ healthcare claims to and Outcomes Kaupapa Wai%202575%2C%20 the Waitangi Tribunal. Inquiry (Wai 2575). Heads 3.2.0209(a).pdf [Accessed NZMJ 2019;132:7-13. of Agreement. Wai 2575, 11 Sep 2020]. Wellington: #3.2.209(a) Available from Ministry of Justice, 2020.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 110 www.nzma.org.nz/journal article

COVID-19 and the impact on urology service provision at Capital & Coast District Health Board Simon Lambracos, Lance Yuan, Andrew Kennedy-Smith

ABSTRACT AIM: To determine the impact on the Capital & Coast District Health Board (CCDHB) urology service of the implementation of nationwide healthcare restrictions in response to the COVID-19 pandemic. METHODS: This is an observational retrospective study over a 21 working day period during the implementation of National Hospital Response Framework Alert (NHRFA) level 2. We obtained patient data during this period and a corresponding control period prior to the pandemic. The data was focussed on the volume of operating theatre cases, outpatient consultations, procedural clinic appointments and the estimated avoided outpatient travel. RESULTS: Total urology admissions decreased by 27% during the 21-day NHRFA level 2 period. However, acute surgical procedures increased by 30% whereas elective surgical procedures decreased by 32%. Outpatient consultations overall decreased by 32% during NHRFA level 2 despite virtual phone consultations increasing by 274%. Procedural clinic appointments decreased by 85%. The virtual platform also saved each patient an estimated 22.7km of average travel. CONCLUSION: The data demonstrate the effects of restrictions in response to a crisis and set a precedent for future management in such scenarios. The data also show how service efficiency can be optimised while providing an environmentally friendly alternative for routine clinical practice.

he coronavirus disease 2019 all district health boards (DHBs) have been (COVID-19) pandemic has impacted on escalated to National Hospital Response Tsocieties and health systems across the Framework Alert (NHRFA) level 2. As far as world. Governmental responses to the crisis the surgical specialties in Capital & Coast and the direct effects of the disease have DHB (CCDHB) are concerned, this has had been expressed differently across countries. four major ramifications:1 Health services in societies that experienced 1. migration of outpatient clinic appoint- high disease incidence have been challenged ments to a virtual setting or an off-site or overwhelmed by the clinical problems of setting as necessary patients with COVID-19. 2. deferral of non-urgent pre-assess- On 25 March 2020, the New Zealand ments and non-urgent clinic patients government initiated a national level 4 according to urgency lockdown, when case numbers were still 3. activation of any outsourcing arrange- small, in response to the international ments reached and engagement with COVID-19 pandemic and after confirmation options for supporting ‘cold trauma’ of community transmission within New cases and less-complex urgent cancer Zealand. Consequently, the New Zealand surgery Ministry of Health introduced a parallel COVID-19 National Hospital Response 4. prioritisation of planned care surgery Framework, consisting of four alert levels and other interventions based on (Figure 1) that are structured according to the urgency while continuing health severity of impact on healthcare services.1 service delivery for patients not expected to require ICU/HDU. Since the start of the national lockdown,

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 111 www.nzma.org.nz/journal article

This report has been set out to demon- During the COVID-19 lockdown and strate the impact NHRFA level 2 and national NHRFA level 2, the urology service was level 4 has had on the service provided reorganised in anticipation of significant by the urology department at CCDHB and numbers of COVID-19 patients in Wellington report on innovative service delivery within regional Hospital, but not without recog- the constraints of national lockdown and nising that the outcome of the national hospital reorganisation in preparation for level 4 lockdown may effectively abort the the anticipated COVID-19 crisis. While the epidemic locally too. The service was recon- impending epidemic threatened to collapse figured thus: existing service delivery, potentially with • The urology department was divided negative outcomes from disease unrelated into two teams maintaining strict to COVID-19, the local response partially separation from each other. maintained continuity of urology services. • Outpatient clinics were converted Moreover, the changes in practice triggered wholesale from face-to-face consul- some unforeseen improved outcomes. tations to telemedicine consultations We have focussed on four aspects of where possible. the service that we believe have been • Elective office procedures were significantly affected by the changes made suspended during the lockdown secondary to NHRFA level 2 implementation. period. These four aspects include: • General anaesthetic urology surgery 1. the volume of operating theatre was substantially relocated to a cases (acute and elective), including nearby private hospital facility, desig- outsourced public elective cases nated a ‘non-COVID-19 hospital’ and during NHRFA level 2 performing cases with urgent clinical 2. the volume of outpatient consultations priority. (virtual and face-to-face) • Some general anaesthetic urology 3. the volume of procedural clinic surgery was continued at Wellington appointments (flexible cystoscopy, Regional Hospital: this included TRUS prostate biopsy and urody- acutely presenting patients and some namics clinics) elective cases with urgent clinical 4. the avoided journeys (km) by outpa- priority. tients travelling to clinic by using a • Teams alternated weekly between the virtual interface. Wellington Regional Hospital campus and the remote private hospital Method campus. The urology service at CCDHB, Wellington, • In the event of a member of one team is a public health service that directly serves becoming ill, that team would be stood a population of 500,000 people and provides down. secondary care services and a tertiary level The intention of this reorganisation was urology service for the wider lower North to achieve continued delivery of urology Island of New Zealand. The delivery of care service through the COVID-19 pandemic, is provided by four to five specialist urol- whether or not there were significant ogists/senior medical officers (SMOs) and numbers of COVID-19 cases. five junior staff/resident medical officers at This is an observational retrospective varying levels of experience and supported study over a period of 21 working days (ie, by nursing, allied health and administrative excluding weekends and public holidays) staff. from the start of the national level 4 Outpatient and inpatient urology lockdown (25 March to 27 April 2020). The services are delivered through Wellington service volumes during this time were and Kenepuru hospitals. Community compared with a similar 21 working day urology care is provided by the network period in March 2019 (15 March to 12 April of primary care general practitioners and 2019). The control period was adjusted to the community nursing service that are exclude the period of junior doctor strike distributed across the region.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 112 www.nzma.org.nz/journal article

Figure 1: Detailing the first two alert levels of the National Hospital Response Framework implemented by all DHBs in response to the COVID-19 pandemic.1

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 113 www.nzma.org.nz/journal article

activity in late April 2019 and the control period was considered a typical workload Results for the urology service at CCDHB. There were no episodes of staff illness during the study period resulting in both Coding data was retrieved from the teams being able to continue work commit- CCDHB information system for details ments as planned. However, through regarding urology patients admitted acutely the planned retirement of one SMO, the and electively under the urology service department workforce was reduced to within the two time periods. four SMOs in 2020 from five SMOs in 2019. Data on scheduled outpatient urology This left one SMO post vacant during the consultations and office procedures were lockdown period. Data are presented in collected manually during the two time absolute numbers and, where appropriate, periods. Telephonic advice and informal calculated on a per-SMO ratio to better consultations were not assessed. Scheduled determine the impact of the reduced SMO outpatient consultations included: workforce. • face-to-face consultations During the study period (25 March to 25 • telemedicine/telephone consultations Aril 2020) there were 77 urology admissions, (virtual clinic) of which 26 were acute admissions and 51 • office procedures. elective admissions. All elective admissions The carbon impact of virtual clinic proceeded with surgery, of which 31 were consultations by negating travel to and operated on at the outsource private hospital. from the hospital clinic is recognised. To This equates to 12.75 elective cases per SMO. estimate the average travel saved (km) by Of 26 acute presentations, 12 patients outpatients consulted using the virtual required acute surgery, which were clinic interface, compared to attending performed at Wellington Regional Hospital. face-to-face outpatient clinic consultations, During the control period in 2019 there patients’ domiciles were obtained using were 106 admissions, of which 75 admis- data from CCDHB coding. We used Google sions were for elective surgery, equating to Maps to estimate the distance from each 15 elective cases per SMO. There were 31 patient’s domicile to Wellington Regional acute admissions during the control period, Hospital. nine of which required acute surgery.

Figure 2: The number of admissions and operative cases for the control 2019 and study 2020 cohorts.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 114 www.nzma.org.nz/journal article

The data indicate a 27% overall decrease tations. This compared to a total of 578 in elective and acute admissions during urology outpatient consultations for the the national level 4 lockdown period in 2019 cohort (115.6 consultations per SMO; comparison to the corresponding 2019 136 virtual and 442 contact consults). This period. However, acute surgical procedures represented a 32% decrease in the total increased in both absolute numbers and number of outpatient consultations during in percentage, an increase of 30% from the level 4 lockdown when compared with the control period. Elective surgery decreased 2019 data, although only an 8% decrease in 32% during the level 4 lockdown period outpatients per SMO. Face-to-face consul- compared to the control period. These tations decreased by 96%, but there was a outcomes are presented in Figure 2. notable 274% increase in virtual consulta- Data on the specific surgeries was tions during the national level 4 lockdown examined and is presented in Figure 3. period compared to the 2019 time period. The specific case data indicate a deferment Office procedure consultations were of non-urgent elective stone surgery and significantly affected, as expected from some other non-urgent surgery, but a the reorganisation. The urology service relative increase (25%) in cancer surgery at CCDHB had already integrated urinary during the NHRFA level 2, compared to the biomarker assessment (CxBladder) to 2019 cohort (Figure 4). manage demand for flexible cystoscopy. CxBladder testing continued during the Standard outpatient consultations period of COVID-19 lockdown but did not decreased during the COVID-19 lockdown influence the data on flexible cystoscopy period. As expected from the reorganisation, procedures in either time period. There there was a marked shift from face-to-face were 192 office procedures performed in the consultations to virtual consultations during 2019 cohort compared to 28 office proce- the lockdown. There were 391 urology dures in the 2020 cohort (Table 1). outpatient consultations for the 2020 cohort (97.75 consultations per SMO), of which 373 The total travel distance saving for outpa- were virtual and 18 were contact consul- tients consulted in the 2020 cohort using a virtual platform was 6,828km. This equates

Figure 3: The raw data of acute and elective operative cases performed for the control 2019 and study 2020 cohorts.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 115 www.nzma.org.nz/journal article

Figure 4: The proportion of cancer and non-cancer procedures performed for both 2019 and 2020 cohorts.

Figure 5: The number of patients attending contact, virtual and procedural clinics for the control 2019 and study 2020 cohorts.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 116 www.nzma.org.nz/journal article

to an average 22.7km of travel saved per from this experience set a precedent for patient during the national level 4 lockdown crisis response management within the as a result of the virtual clinic interface department for the future. (Table 2). Admissions The total number of admissions decreased Discussion by 27% during the NHRFA level 2 period The urology service review of clinical in comparison to the corresponding 2019 activities during the COVID-19 lockdown period. It appears that the reduction in demonstrates an overall reduction in elective operating cases was largely respon- service delivery and a reorientation to sible for this decrease and that, despite non-contact outpatient consultations. The outsourcing elective cases to the private impact on throughput of the hospital-wide hospital, the service was unable to match preparations for the epidemic, closures and the usual workload. Furthermore, there reorganisation were mitigated by proactive was only a 16% decrease in acute admis- outsourcing of elective surgery to a private sions during NHRFA level 2. The volume of hospital and a dramatic shift to virtual acute admissions is likely to vary regardless consultations. This reorganisation was of the circumstances, but it could also conceived and implemented simultaneously be explained by a reduction in patients with the national lockdown. The outcomes presenting to hospital due to the perceived

Table 1: A breakdown of the number of patients attending procedural clinic consultations for the 2019 and 2020 cohorts.

Flexible cystos- Cohort TRUS biopsy copy Urodynamics Total 2019 65 107 20 192

2020 2 26 0 28

Table 2: Number of patients from the various regions serviced by urology at CCDHB with clinic ap- pointments during NHRFA level 2 and the distances these patients would have to travel to attend clinic appointments at the Wellington Regional Hospital. Distances averaged for satellite towns and calculated individually for patients within Wellington City.

Distance to hospital Suburb Number of patients (km) Total distance (km) Featherston 2 65 130

Lower Hutt 86 19 1,634

Martinborough 1 83 83

Masterton 2 101 202

Otaki 2 76 152

Paekākāriki 3 43 129

Paraparaumu 20 53 1,060

Porirua 29 23 667

Upper Hutt 32 37 1,184

Waikanae 12 63 756

Wainuiomata 3 26 78

Wellington Central 109 Calculated individually 753 Total 301 6,828

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 117 www.nzma.org.nz/journal article

risk of contracting COVID-19 in hospital, and patients could still receive essential elective therefore this may represent an unmet need. care in a safe environment. Not only does Operating theatre cases this mean that patients can receive elective surgical care in a timely manner, but also There was a 33% increase in acute oper- that a backlog of elective cases does not ating theatre cases during NHRFA level 2 develop once NHRFA level 2 is de-escalated. compared to the control 2019 period. This is most likely explained by the fact that NHRFA Outpatient clinics level 2 did not impose any restrictions on As per the NHRFA level 2 restrictions, all acute theatre cases, and this would therefore outpatient clinics should be deferred to a represent the acute surgical requirement at virtual or offsite setting and all non-urgent the given time. appointments should be deferred. As a Despite a 25% increase in urgent cancer result, in the urology department at CCDHB, cases during NHRFA level 2, there was, as the vast majority of outpatient consultations expected, a decrease in general elective were rescheduled as virtual phone consul- operating. This was calculated as 32% but tations and, where possible, non-urgent there are confounding factors that poten- face-to-face consultations were deferred tially exaggerate this difference: until normal service resumes. A select few urgent face-to-face appointments were Firstly, 7% of the elective cases for retained where physical examination or the 2019 cohort occurred in Kenepuru in-clinic investigations were required. Community Hospital (KCH), a regional hospital in Porirua catering for low-risk These alterations to the outpatient service elective cases only. The availability at KCH were demonstrated by a 274% increase in normally allows for simultaneous elective virtual consultations under NHRFA level 2 lists both there and at Wellington Regional compared with the corresponding period Hospital. However, during NHRFA level 2, in 2019. Consequently, this correlated with KCH was closed. a 96% decrease in contact consultations for the 2020 cohort. Secondly, in 2019 there were five consul- tants working in the urology department Procedural clinics as opposed to four consultants in 2020. The major difficulty in implementing This provided the opportunity for an NHRFA level 2 within the department was increased elective surgical capacity in 2019. attempting to accommodate the patients It therefore seemed more representative to that were due for procedural clinic appoint- demonstrate the difference in elective case ments. As our data suggest, procedural workload by calculating the number of cases clinic consultations were down by 85% per consultant. For the 2019 cohort this was for the 2020 cohort compared to the 2019 15 cases per consultant compared to 12.75 cohort. Only two TRUS prostate biopsy for the 2020 cohort. Each consultant was procedures and 26 flexible cystoscopies therefore averaging 2.25 elective cases less were performed during NHRFA level 2, all of during NHRFA level 2, which, considering which were subacute cases. Standard proce- the severe restrictions placed on elective dural clinics had to be deferred due to their case workload, suggests a less significant non-urgent nature. This in turn resulted in difference between the cohorts than was a significant backlog of procedural clinic initially indicated. appointments for after NHRFA level 2 and a Finally, it is also worth noting that, despite significant number of patients with delayed the decrease in elective case workload, over investigations. 60% of elective cases during NHRFA level Avoided travel 2 were outsourced to the private hospital. We have previously assessed the Of these, 45% were urgent cases treating acceptability to patients of virtual clinic potential or confirmed cancer diagnoses. consultations (unpublished) and the actual This represented a 25% increase in urgent travel savings. Although virtual consul- cases during NHRFA level 2 compared to tations limit the possibility of physical the 2019 period. This demonstrates that, examination and in-clinic procedures, it even with severe restrictions on performing does provide a significant economic and elective cases in the public sector, over 30 environmental benefit to the community.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 118 www.nzma.org.nz/journal article

As stated previously, by consulting with NHRFA level 2. They provide us with an patients in a virtual setting and avoiding insight into which aspects of the service face-to-face consultations at the hospital, are most heavily affected by the imposed travel reduced by an estimated 6,828km. restrictions and how best to manage This equates to an average 22.7km of travel these to moderate the overall impact of per patient saved, which would have had healthcare provision for the community. significant repercussions on the envi- The data also set a precedent for major ronment as well as on travel expense and crisis response management going forward journey time for outpatients. and highlights which areas might require particular attention in these scenarios. Conclusion Furthermore, with regard to the virtual consultation platform, the data also suggest Our data provide a snapshot interpre- ways in which our practice can be adapted tation of the altered workload relating to on a routine basis in the future, in order to admissions, operating theatre cases and increase efficiency and to provide a service outpatient clinics in the CCDHB urology that is both economic to the patient and department during the implementation of environmentally prudent.

Competing interests: Nil. Acknowledgements Capital & Coast DHB information system coding team for providing DHB patient data; Shelley De Boer, the urology clinic charge nurse, for providing outpatient clinic data. Author information: Simon Lambracos: General Surgery Department, Poole Hospital NHS Trust, Longfleet Road, Poole BH15 2JB, UK. Lance Yuan: Urology Department, Capital & Coast DHB, Riddiford Street, Newtown, Wellington 6011. Andrew Kennedy-Smith: Urology Department, Capital & Coast DHB, Riddiford Street, Newtown, Wellington 6011. Corresponding author: Mr Simon Lambracos, Surgical Trainee, NHS [email protected] URL www.nzma.org.nz/journal-articles/covid-19-and-the-impact-on-urology-service-provision-at- capital-coast-district-health-board-open-access

REFERENCES 1. Ministry of Health Nation- https://meras.midwife. sponse-Framework-4pm. al Hospital Response org.nz/wp-content/ pdf Framework version 1.0 uploads/sites/4/2020/03/ [Internet]. Available from: National-Hospital-Re-

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 119 www.nzma.org.nz/journal article

Impact of the COVID-19 pandemic lockdown on public sector ophthalmic work by New Zealand’s ophthalmologists Daniel A R Scott, Peter W Hadden, Graham A Wilson

ABSTRACT AIM: In response to the COVID-19 pandemic, the New Zealand government enforced a nationwide ‘alert level 4’ lockdown from 26 March to 27 April 2020. We assessed the impact of this lockdown on New Zealand’s public ophthalmology service. METHOD: An anonymous online survey was sent to all New Zealand-based fellows of the Royal Australian and New Zealand College of Ophthalmologists (RANZCO) after lockdown. Respondents provided retrospective assessment of practice patterns and their personal health during the COVID-19 lockdown. This was supported by national-level administrative data, allowing survey findings to be contextualised. RESULTS: Fifty-seven respondents (response rate 49%) working in the public health system participated. A large majority of respondents reduced elective clinic and surgical volumes by at least 75% (82% and 98%, respectively). National-level information confirmed clinic reduced to 38.2% of normal and elective operating volumes to 11.5%, with virtual visits increasing 17.9-fold. Elective clinic and elective operating volumes promptly recovered to usual volumes on the second month post lockdown. Most respondents (58%) followed the RANZCO triaging guideline, and 28% triaged emergencies only. At a personal level, respondents reported a significant physical health benefit (p<0.001) associated with the lockdown experience, but no change in mental health or social wellbeing. CONCLUSIONS: Publicly employed ophthalmologists experienced dramatic reductions to elective clinic and operating volumes during the COVID-19 lockdown. The prompt recovery of service delivery volumes back to pre-lockdown levels supports the value of a COVID-19 elimination strategy in New Zealand. Virtual visits for selected patients allowed ongoing management without risking virus transmission.

he virus SARS-CoV-2 (the cause of on 26 March, which eventually relaxed into COVID-19) began circulating in Wu- a less intense alert level 3 lockdown on 28 Than, China, in November 2019. It sub- April.3 sequently spread across the world, becom- Compared to other countries, New Zealand ing a global pandemic. Physical distancing was fortunate to have a number of distinct was the main strategy of limiting the spread advantages that favoured an elimination of COVID-19, leading to a number of nation- strategy. First, COVID-19 had a relatively wide lockdowns. In April, approximately late arrival to New Zealand, allowing us one-third of the world’s population was time to plan and learn from the experiences under COVID-19 lockdown orders or similar of countries like China, Korea, the United 1,2 movement restrictions. New Zealand ad- Kingdom, Iran and Italy; as a remote island opted an evidenced-based strategy of virus nation with defined and monitored borders, elimination. After a small number of cases our location enabled an effective border were identified in New Zealand, a nation- closure and quarantine of all returning wide ‘alert level 4’ lockdown period began New Zealand nationals; and, informed by

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 120 www.nzma.org.nz/journal article

evidence-based information and supported using the ESCRS survey as a scaffold for by public trust and adherence to health its design.8 The survey was shortened to and safety messages from our ‘team of five eight questions to maximise the response million’, the central government co-or- rate, and a final question allowing free-text dinated a ‘go early, go hard’ approach, answers let respondents leave further which included a quick and clear national comments. The survey was reviewed by lockdown instructing the entire population a clinical psychologist prior to its distri- to remain in their ‘bubbles’ (ie, usual family/ bution. The survey was sent to all New household). Testing increased, cases were Zealand-based RANZCO fellows by their traced and isolated, and fast and efficient professional body (the RANZCO) on 21 May, contact tracing was possible.4 with a second reminder six weeks later, As a result, New Zealand experienced a on 2 July. At the time of the survey, there limited COVID-19 disease burden compared were 138 New Zealand RANZCO fellows, to other high-income countries, such as with 117 (87%) confirmed as working either Australia, the Uinited Kingdom and Italy. part- or full-time in the New Zealand public The reduced strain on secondary health sector at local district health boards (DHBs). services in New Zealand is highlighted Survey access and collection utilised a by low rates of intensive care admissions self-administered anonymous Google survey (0.7%) and mortality (1.5%) attributable to form. Survey results were converted into a COVID-19.5 However, the reprioritisation of Microsoft Excel spreadsheet for analysis. healthcare services and deferral of routine The following national-level admin- care (including procedures/surgery)5 to istrative data were requested from the reduce virus transmission has undoubtedly Ministry of Health for the month of April led to secondary negative health impacts.3 2019 and April 2020: clinic volumes, elective Delays with routine ophthalmic care theatre volumes and clinic virtual visits during COVID-19 lockdown will increase (VVs). This information was received on 19 the attributable burden of preventable August 2020 following a request under the blindness, and has been highlighted for Official Information Act. Data for months conditions like age-related macular degen- May and June (2019 and 2020) were received eration (AMD).6 Ophthalmology practice on 12 April 2021. patterns in Europe and India have high- lighted a significant reduction in the elective Results work as a result of the COVID-19 pandemic, A total of 59 ophthalmologists responded with a large proportion of ophthalmologists to the survey, with two excluded due to 7,8 seeing only emergencies. International working only in private (ie, no public/DHB 9 ophthalmology societies, including the workload). The response rate was therefore Royal Australian and New Zealand College 49% (57/117), with respondents representing 10 of Ophthalmologists (RANZCO), created 12 DHBs from around New Zealand (Figure triaging guidelines to help local ophthal- 1). The median DHB job size of respondents mologists decide who needed care during was 0.62 full-time equivalent (Figure 1). lockdown. Elective clinic consultations reduced In this paper, we present the results of a substantially (p<0.001) during the lockdown, survey assessing the impact of a COVID-19 with 82% of respondents seeing 75% less lockdown on New Zealand’s public ophthal- than their usual volume of patients (Figure mology service. We aimed to assess the 2). National-level administrative data impact by measuring the reduction in records for the month of April indicate 2020 elective clinic and elective operating clinic consultation volumes were 38.2% of volumes. We also aimed to assess the April 2019. Service volumes in May and personal health impact of the lockdown for June 2020 recovered to 68.9% and 107.0% New Zealand ophthalmologists. respectively compared to the previous year. Clinic volumes for first specialist appoint- Methods ments (FSAs) experienced a more significant The study survey was created in consul- reduction compared to clinic follow-up tation with two consultant ophthalmologists appointments (Table 1).

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 121 www.nzma.org.nz/journal article

Elective surgery essentially stopped confirmed that the volume of VVs increased (p<0.001) during the lockdown, with 79% of by 17.9 times compared to April 2019, respondents performing zero elective oper- being 19.8% of all April 2020 clinic visits. A ations, and 98% performing less than 25% breakdown of VVs for 2020 showed 91.3% of their usual volume (Figure 2). Nation- were classified as follow-ups, with the al-level administrative data (which included remainder being FSAs. one week of the less intense lockdown at The New Zealand RANZCO triaging alert level 3) confirmed elective operating guidelines were reported to have been reduced to 11.5% of April 2019 service followed by 58% of respondents (p<0.001). volumes. The elective operating volumes Over a quarter (28%) triaged only vision/ in May and June 2020 recovered to 80.9% life-threatening emergencies, and 14% using and 115.3% respectively compared to the clinical judgement for triaging (ie, either previous year. Of interest, acute operating they were unaware of the clinical guideline volumes reduced to 43.1% of usual service or preferred to triage differently to the volumes in April. A similar pattern of guideline) (Figure 3). recovery was demonstrated in May and June For ophthalmologists, the impact of 2020, with acute operating recovering to the alert level 4 lockdown appeared 61.5% and 95% respectively of usual service to have had mixed impacts across the volumes (Table 1). three domains (mental, social wellbeing, VVs performed by telephone, e-mail, physical) of health. There were no reported application and video were performed significant mental health (p=0.81) or by 70% of respondents (p<0.001) (Figure social wellbeing (p=0.69) impacts from the 3). National-level administrative data COVID-19 lockdown. However, there was

Figure 1: Respondents by district health board (left); box and whisker graph of respondent workload, measured by full-time equivalent (right).

Figure 2: Percentage of elective clinic patient consultations (right) and elective surgeries performed (left) during alert level 4 lockdown compared to usual workload.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 122 www.nzma.org.nz/journal article

Table 1: Public sector ophthalmic service workload reduction by district health board (DHB) represent- ed as a percentage (%) compared to normal (national-level administrative data for April 2020 compared to April 2019).

Total clinic Clinic Clinic FU Elective operat- Acute operating % FSA % % ing % % National average 38.2 43.2 36.7 11.5 43.1

DHB median 26.8 37.2 26.1 7.6 37.5

DHB 1 45.5 48.2 44.6 15.1 35.1

DHB 2 87.1 80.9 89.0 33.3 100.0

DHB 3 80.4 56.2 90.1 19.5 40.0

DHB 4 25.2 25.6 25.0 3.2 86.4

DHB 5 19.9 40.0 16.4 8.1 n/a

DHB 6 26.9 31.7 24.3 6.6 14.3

DHB 7 96.7 60.7 84.6 0.0 n/a

DHB 8 26.8 45.8 29.1 0.0 150.0

DHB 9 26.5 35.9 24.8 8.8 0.0

DHB 10 43.6 51.1 41.5 0.0 50.0

DHB 11 26.2 54.2 22.7 8.2 60.0

DHB 12 22.5 26.1 21.6 8.7 0.0

DHB 13 22.4 37.2 19.7 20.9 137.5

DHB 14 33.3 33.3 33.3 45.5 n/a

DHB 15 15.9 11.0 17.9 0.0 100.0

DHB 16 31.0 37.3 28.8 7.6 45.8

DHB 17 42.2 19.2 48.9 0.0 n/a

DHB 18 22.7 9.1 26.1 3.2 n/a

DHB 19 11.4 29.0 7.4 0.0 0.0

n/a Not applicable, FSA first specialist assessment, FU follow-up.

Figure 3: Percentage of virtual visits during alert level 4 lockdown (left), and triaging method for the alert level 4 lockdown restrictions (right).

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 123 www.nzma.org.nz/journal article

a clear physical health (p<0.001) benefit oritised ophthalmic service has probably reported from the lockdown (Figure 4). increased the risk of avoidable blindness A number of key themes emerged on from reduced access in the international review of the thirteen free-text comments setting. left by respondents. Two respondents high- Research to date documenting reduced lighted the benefits of consultant (ie, senior clinic and elective surgical volumes during medical officer) triage for managing acute COVID-19 lockdowns have been published referrals increasing efficiency and remote in both India and Europe (ESCRS survey).7,8 management. A few respondents appreciated Nair et al found 70% of ophthalmolo- the “less frenetic pace of work”; however, gists in India stopped all clinical work almost half of respondents appreciated (clinic and operating) for their nationwide the delays generated from lockdown have lockdown. For ophthalmologists in India increased the “backlog of elective work”, who continued to see patients, over 80% putting a “strain on services long term”. classified their patient encounters as emergencies (eg, endophthalmitis, retinal Discussion detachment, trauma).7 In Europe, the ESCRS survey reported that one-third of European The COVID-19 alert level 4 lockdown ophthalmologists provided emergency dramatically reduced the clinic and surgical care only, with 11% and 58% of European workload for New Zealand’s publicly ophthalmologists stopping all clinic consul- employed ophthalmologists. Our retro- tations and operating, respectively.8 spective survey of practice patterns found that during the lockdown almost 80% of In the New Zealand public ophthalmic these ophthalmologists did not perform sector, elective surgical volume reduced by elective surgery, with a similar proportion almost 90%, which is less than European 7,8 reducing clinic volumes by 75%. Nation- and Indian survey data by comparison. al-level administrative data indicate clinic The reduction of elective clinic consultations consultation visits and elective operating was more modest compared to the elective volumes reduced to 38.2% and 11.5% of surgical volumes in New Zealand. There normal, respectively. was substantial variation between different DHBs (Table 1), which highlights a different The timely access to ophthalmic care has case-mix of patients and/or different triaging major impacts on the progression of eye strategy. conditions and their treatment outcomes.11 It is therefore not surprising that COVID-19 has New Zealand adopted the elimination led indirectly to increased rates of blindness strategy towards COVID-19, as opposed from delayed diagnosis/referral of acute and to the mitigation and suppression strat- 14 chronic ocular conditions internationally.6,12 egies adopted by European countries. The timing of COVID-19 during VISION 2020 The benefit of the New Zealand approach is ironic, given this was the year to celebrate afforded the country a relatively quick vision.13 Instead a scaled back and repri- return to normal domestic activity and

Figure 4: Impact of alert level 4 lockdown on mental health (left), physical health (middle) and social wellbeing (right).

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 124 www.nzma.org.nz/journal article

routine ophthalmic clinical care.15 In prin- social connectedness, but our survey results ciple, elimination focuses on preventing did not support these findings.21 pandemic virus introduction and ending The use of teleophthalmology (or VVs), local transmission, as opposed to ongoing which provides physical distancing protec- suppression/mitigation strategies that tions while continuing to offer care, involve ongoing physical distancing, increased 17.9-fold during the lockdown testing and contact tracing (and potentially month of April. Over 90% of patient VVs 14 repeated lockdowns). Public ophthal- were classified as follow-up consultations, mology service volumes recovered back and the literature has shown these patients to normal on the second month after (as opposed to first specialist assessments) lockdown. Although the impact of lockdown are more likely to engage with this type of to ophthalmic service delivery was signif- service delivery.22 Furthermore, the inte- icant, it was relatively short-lived, and this gration of acute teleophthalmology services is a credit to the success of our public health in France and the United Kingdom was able elimination policy. Furthermore, although to reduce in-person ophthalmology assess- New Zealand and most European coun- ments by 73%23 and 78%24 respectively. tries were in lockdown, the health systems If New Zealand re-entered another of many European countries were over- lockdown as a result of a large outbreak whelmed, and they experienced potentially from a border control failure, the appli- avoidable deaths and increased all-cause cation and translation of our research could mortality.16 help inform future ophthalmology service Most international ophthalmology soci- responses. The survey highlights the success eties around the world produced triaging of upscaling VVs and consultant-led triage to guidelines during the COVID-19 pandemic. reduce the risk of virus transmission. This There was a focus on delaying non-urgent form of service delivery may have a greater care and following recommended safety role in our overburdened public health 17 practices to reduce infection transmission. system for the future. Anecdotal unpub- The New Zealand RANZCO branch produced lished reports from across New Zealand a triaging guideline with recommenda- indicate a number of patients have gone tions for various ocular presentations blind in their only eye due to missed or 10 and ophthalmic diseases. These were delayed elective clinic appointments. Future followed by the majority of respondents in research could calculate the visual burden our survey, with those in a tertiary referral attributed to delayed clinic appointments/ region being 1.9 times more likely to follow surgery from the alert level 4 lockdown. It these guidelines. The variance in the would be interesting to then assess whether triaging of care, reflected by the workload the different ophthalmic workload reduc- reductions across DHBs (Table 1), may tions by DHB had any impact on the visual have led to inequities in healthcare access burden. and differences in ocular health outcomes The combination of survey findings and between regions. Triaging care during this national-level administrative data is a period was time consuming, challenging and strength of our study, and we compared involved many competing interests. New findings to the published literature in novel scoring algorithms18 and modelling Europe and India. The survey response rate tools19 may have a role in providing a more of 49% is in line with the online web-based objective assessment in the future. survey response rate of surgical doctors.25 Most ophthalmologists in our survey The survey could have been held closer reported physical health benefits during the to the end of lockdown to increase the lockdown, which is not surprising given the accuracy of the respondents’ answers. known health benefits of exercise and its The national-level administrative data for promotion as an essential activity during the April also captures one week of alert level 20 New Zealand lockdown. Previous research 3 restrictions, and so triaging of care may has shown that lockdowns generally have been loosened, given the country had produce negative benefits to mental health already stepped down from alert level 4 and an improved sense of community and lockdown restrictions.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 125 www.nzma.org.nz/journal article

Competing interests: Dr Hadden reports that he is the current chair of the New Zealand branch of the Royal Australian and New Zealand College of Ophthalmologists, which formulated prioritisation guidelines for ophthalmology during the pandemic. Acknowledgements: Thank you to the RANZCO NZ branch for the distribution of this survey and to the fellows who responded. Thank you to Brandan Letham (Clinical Psychologist, Hauora Tairāwhiti) for reviewing our survey. Thank you to Professor Nick Wilson (University of Otago, Public Health) for an expert review of this manuscript. Author information: Daniel A R Scott: Ophthalmology Non-Vocational Registrar, Department of Ophthalmology, Gisborne Hospital, Hauora Tairāwhiti, Gisborne. Peter W Hadden: Ophthalmologist, Department of Ophthalmology, New Zealand National Eye Centre, Faculty of Medicine and Health Sciences, University of Auckland, Auckland. Graham A Wilson: Ophthalmologist, Department of Ophthalmology, Gisborne Hospital, Hauora Tairāwhiti, Gisborne; Mātai Lab, Gisborne. Corresponding author: Dr Daniel Scott, 421 Ormond Road, Gisborne Hospital, Hauora Tairāwhiti, Gisborne, +64 273211946 [email protected] URL: www.nzma.org.nz/journal-articles/impact-of-the-covid-19-pandemic-lockdown-on-public- sector-ophthalmic-work-by-new-zealands-ophthalmologists-open-access

REFERENCES 1. Koh D. COVID-19 epidemiological study. 2020;5:e000525. lockdowns throughout Lancet Public Health. 2020. doi: 10.1136/ the world. Occup Med doi: https://doi.org/10.1016/ bmjophth-2020-000525. (Lond). 2020;70:322. S2468-2667(20)30225-5. 10. Royal Australian and New 2. Greyling T, Rossouw 6. Thomas DS, Warwick A, Zealand College of Ophthal- S, Adhikari T. A tale of Olvera-Barrios A, et al. mologists (RANZCO) three countries: How Estimating excess visual [Internet]. RANZCO New did Covid-19 lockdown loss in people with neovas- Zealand Branch Triage impact happiness? GLO cular age-related macular Guidelines – COVID-19 Discussion Paper, No. degeneration during the Level 4; [cited 2020 Apr]. 584. Essen: Global Labor COVID-19 pandemic. Available from: https:// Organisation (GLO); 2020. MedRxiv. 2020. doi: ranzco.edu/wp-con- 3. Baker M, Wilson N, https://doi.org/10.1101/ tent/uploads/2020/04/ Anglemyer A. Success- 2020.06.02.20120642. RANZCO-triage-NZ-mod- ful Elimination of 7. Nair AG, Gandhi RA, ification-level-4.pdf Covid-19 Transmission Natarajan S. Effect of 11. World Health Organisa- in New Zealand. N Engl COVID-19 related lockdown tion [Internet]. World J Med. 2020;383:e56. doi: on ophthalmic practice report on vision; [cited 10.1056/NEJMc2025203. and patient care in India: 2020 Oct 8]. Available 4. Henrickson M. Kiwis and Results of a survey. Indian J from: https://www.who. COVID: The Aotearoa Ophthalmol 2020;68:725-30. int/publications-detail/ New Zealand response 8. Nuijts R, Cochener-Lamard world-report-on-vision to the global pandemic. B, Kohnen T, Findl O. 12. Monti S, Delvino P, Bellis The International Jour- Covid-19 practice patterns E, et al. Impact of delayed nal of Community and of ophthalmologist. diagnoses at the time of Social Development. Eurotimes. 2020;25:I-XI. COVID-19: increased rate 2020;2:121-33. 9. Nguyen AX, Gervasio KA, of preventable bilateral 5. Jefferies S, French N, Wu AY. Differences in SARS- blindness in giant cell Gilkison C, et al. COVID-19 CoV-2 recommendations arteritis. Ann Rheum in New Zealand and the from major ophthalmol- Dis. 2020;79:1658-9. impact of the national ogy societies worldwide. 13. Bourne RRA. Vision 2020: response: a descriptive BMJ Open Ophthalmol. where are we? Curr Opin

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 126 www.nzma.org.nz/journal article

Ophthalmol. 2020;31:81-4. Ophthalmology Data Pinnock H, et al. Telephone 14. Baker M, Kvalsvig A, Verral Repository for Triaging consulting in primary care: A, et al. New Zealand’s Patients With Glaucoma a triangulated qualita- elimination strategy for and Clinic Appointments tive study of patients the COVID-19 pandemic During Pandemics Such as and providers. Br J Gen and what is required COVID-19. JAMA Ophthal- Pract. 2009;59:e209-18. to make it work. N Z mol. 2020;138:974-80. 23. Bourdon H, Jaillant R, Balli- Med J. 2020;133:10-4. 19. Boyd MJ, Scott DAR, no A, et al. Teleconsultation 15. Baker M, Kvalsvig A, Verrall Squirrell DM, Wilson GA. in primary ophthalmic AJ. New Zealand’s COVID-19 Proof-of-concept calcu- emergencies during the elimination strategy. Med lations to determine the COVID-19 lockdown in J Aust. 2020;213:198-200. health-adjusted life-year Paris: Experience with trade-off between intrav- 500 patients in March and 16. Piccininni M, Rohmann itreal anti-VEGF injections April 2020. J Fr Ophtal- JL, Foresti L, et al. Use and transmission of COVID- mol. 2020;43:577-85. of all cause mortality to 19. Clin Exp Ophthalmol. quantify the consequences 24. Kilduff CL, Thomas AA, 2020. doi: https://doi. of covid-19 in Nembro, Dugdill JD, et al. Creating org/10.1111/ceo.13855. Lombardy: descriptive the Moorfields’ virtual eye study. BMJ. 2020. doi: 20. Füzéki E, Groneberg casualty: video consulta- 10.1136/bmj.m1835. DA, Banzer W. Physical tions to provide emergency activity during COVID- teleophthalmology care 17. Nguyen AX, Gervasio KA, 19 induced lockdown: during and beyond the Wu AY. Differences in SARS- recommendations. J Occup COVID-19 pandemic. CoV-2 recommendations Med Toxicol. 2020;15:25. BMJ Health Care Inform. from major ophthalmol- 2020;27:e100179. ogy societies worldwide. 21. Sibley C, Greaves L, BMJ Open Ophthalmol. Satherley N, et al. Effects 25. Meyer VM, Benjamens S, 2020;5:e000525. of the COVID-19 pandemic Moumni ME, et al. Global doi: 10.1136/ and nationwide lockdown Overview of Response bmjophth-2020-000525. on trust, attitudes toward Rates in Patient and government, and Health Care Professional 18. Bommakanti N, Zhou Y, well-being. Am Psychol. Surveys in Surgery: A Ehrlich J, et al. Application 2020;75:618-30. Systematic Review. Ann of the Sight Outcomes Surg. 2020. doi: 10.1097/ Research Collaborative 22. McKinstry B, Watson P, SLA.0000000000004078.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 127 www.nzma.org.nz/journal viewpoint

Addressing structural discrimination: prioritising people with mental health and addiction issues during the COVID-19 pandemic Helen Lockett, Ashley Koning, Cameron Lacey, Susanna Every-Palmer, Kate M Scott, Ruth Cunningham, Tony Dowell, Linda Smith, Alison Masters, Arran Culver, Stephen Chambers

ABSTRACT Structural discrimination worsens physical health inequities and significantly reduces life expectancy for people with mental health and addiction issues. Aotearoa has recently made some notable changes in health policy by formally recognising the physical health needs of people with mental health and addiction issues. The COVID-19 vaccination sequencing framework provides an important opportunity to protect and promote the health of people with addiction and mental health issues. An expert advisory group, convened as part of the Aotearoa Equally Well collaborative, considered findings of a literature review on the vulnerability of people with mental health and addiction issues of contracting and dying from COVID- 19. Evidence indicates an association between mental health and addiction issues and infection risk and worse outcomes. The group concluded mental health and addiction issues should be recognised as underlying health conditions that increase COVID-19 vulnerability, and that people with these issues should be prioritised for vaccination. For too long the health system has failed to address the life expectancy gap of people with addiction and mental health issues. Now is an opportunity to change the kōrero. People with mental health and addiction issues experience significant physical health inequities. Addressing these inequities must be integral in modern health policy—including our COVID-19 pandemic response.

he recognition of mental health and Over the past five years there have been addiction issues as pre-existing health notable changes in Aotearoa New Zealand’s Tconditions that place people at higher health policy towards formally recognising risk of serious outcomes or illness in the the physical health inequities experienced Ministry of Health’s vaccine rollout plan (ie, by people with mental health and addiction Group 3) is critical to prevent further physi- issues. These include Living Well with cal health inequities for people with mental Diabetes, the five-year plan for people at health and addiction issues. high risk of or living with diabetes, which People with mental health and addiction identifies people with mental health and issues have two to three times the risk of addiction issues as a high-risk group and dying before the age of 65 compared to priority population for routine diabetes 2 the general population.1 Two-thirds of this screening. An evidence review around risk is caused by preventable and treatable cardiovascular disease (CVD) risk and physical illnesses. Policy-level changes, management in people with mental alongside practice changes, are crucial if this health and addiction issues informed significantly reduced life expectancy is to be new guidance for primary care CVD 3 addressed. risk assessment and management. This

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 128 www.nzma.org.nz/journal viewpoint

guidance prioritises people with mental Understanding COVID-19 health and addiction issues for CVD risk vulnerability assessment and ongoing management from At the end of 2020, the backbone team the age of 25.4 The New Zealand Cancer of the Aotearoa New Zealand Equally Well Action Plan 2019–20295 has a focus on collaborative convened a group of expert the need to improve the quality of cancer advisors to appraise the available evidence screening and treatment of people with on COVID-19 vulnerability and people with addiction and mental health issues. mental health and addiction issues and, The whole of the healthcare system is from this appraisal, to develop a position consistently identified as a major factor to statement. The review of the evidence was the physical health inequities experienced conducted by Te Pou, a national workforce by people with mental health and addiction centre for mental health, addiction and issues. This is particularly due to systemic disability in Aotearoa New Zealand. issues that impact on access to, and quality In January 2021, the position statement of, physical healthcare.6 At worst, this results and a summary of the evidence gathered in the denial of preventative services and were provided to the Ministry of Health’s routine care. Stigma and discrimination, policy team developing the COVID-19 particularly diagnostic overshadowing (the vaccination sequencing framework. This assumption that physical health symptoms information was also made available on the are related to a mental health or addiction Te Pou website.13,14 diagnosis) and clinicians seeing the person as their mental health or addiction diag- This literature review was crucial for nosis, means people are less likely to understanding both the extent of risk of receive the expected standard of healthcare. infection for COVID-19 and poorer health This leads to delayed or missed physical outcomes for people with addiction and health screening, care and treatment. mental health issues. It was also important Discrimination is experienced across the to look at whether any risks were only for health system, including but not limited people with existing physical health comor- to diabetes care, routine cancer screening, bidities (such as high body mass index blood pressure and cholesterol monitoring, [BMI], hypertension, cardiovascular disease surgical interventions for cardiovascular or diabetes), or whether people with mental diseases and vaccinations.7 health and addiction issues have elevated risks from COVID-19 independent of coex- There is also the significant and often isting issues. If the latter is true, it is crucial unrecognised intersection of ethnicity, that vaccination prioritisation explicitly lists mental health, addiction and physical mental health and addiction issues as under- health, with the largest inequities for Māori lying health conditions that place people at and Pasifika peoples living with mental greater risk. Other countries, including the health and addiction issues.8,9 There is an UK, Ireland, Germany and Denmark, have important opportunity for the Government identified and recognised a significantly to meet its Te Tiriti o Waitangi commit- greater risk in their respective vaccination ments to active protection, as described prioritisation frameworks.15 in Whakamaua Māori Health Action Plan 2020–2025.10 Te Pou conducted the rapid literature scan in December 2020 and drew on systematic Kia Kaha, Kia Māia, Kia Ora Aotearoa: reviews, where available, and individual Psychosocial and Mental Wellbeing Plan11 studies. The review sought to answer the identifies people with mental health following questions: For people with mental and addiction issues as a priority group health and addiction issues: in relation to the psychosocial impacts of COVID-19, and that they are more • is the risk of being infected with vulnerable to infection and negative COVID-19 higher compared to other outcomes from the virus. It is critical that groups? there is formal recognition and that this • what is the independent risk of hospi- recognition translates into vaccination talisation and death from COVID-19? policy and implementation, including work- The review found a significant volume of force education and training.12 high-quality research and evaluation that

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 129 www.nzma.org.nz/journal viewpoint

explores the relationship between COVID-19 people with mental health and addiction infection and outcomes for people experi- issues within the same priority tier for encing mental health and addiction issues. vaccination as people in the general popu- The research spans a wide range of coun- lation aged 70–74. In March 2021, Toubasi tries and settings and consists of systematic and colleagues published a meta-analysis literature reviews, quantitative survey data, exploring the relationship between pre-diag- qualitative narratives and retrospective nosis of mental health issues and COVID-19 cohort and case-control studies.14 This outcomes. They found people with mental included large studies in the US, the UK and health diagnoses were more likely to Europe. become seriously ill or to die from COVID-19, The review found that the risks of an association that remained significant infection were particularly high for people after adjusting for confounding variables. meeting diagnostic criteria for a mental People with diagnoses of schizophrenia, health or addiction issue in the past year schizotypal and delusional disorders had (including first diagnosis); people experi- higher COVID-19 mortality compared to encing multiple mental health or addiction people with mood disorders. The authors issues; and people accessing inpatient recommend prioritising people with mental 20 services. In terms of hospitalisation with health and addiction issues for vaccination. COVID-19, the risk for people with expe- Three key findings emerged from the Te rience of mental health and addiction issues Pou literature review: is significantly higher compared to people 1. People with mental health and not experiencing these issues, with the risk addiction issues are at higher risk even higher for people who meet criteria for of contracting COVID-19 than those 16 multiple diagnoses. The risk of dying from without. COVID-19 may be up to twice as high for 2. Once infected, people with mental people with experience of mental health and health and addiction issues have addiction issues, and even higher for people a higher risk of severe outcomes with certain diagnoses and more complex including hospitalisation and death. needs. For example, people experiencing psychosis or schizophrenia may have a risk 3. These risks, especially for people with from 2.7 to 4.4 times higher than people ‘severe mental illness and addiction without these experiences.16–18 diagnoses’, are evident even once esti- mates are adjusted for other known Subsequently, even more compelling risk factors for COVID-19 infection evidence has emerged. In February 2021, the and severe outcomes. These factors Robert Koch Institute published an umbrella include physical comorbidities, review of the evidence on the relative risks obesity and socioeconomic status.14 of hospitalisation and death for different underlying health conditions.19 People There is also emerging evidence from a with ‘severe mental illnesses’ (defined UK study that people with mental health as meeting diagnostic criteria for schizo- and addiction issues are much less likely 21 phrenia, depression and bi-polar disorder) to present for COVID-19 vaccination. The had the third highest risk of hospitalisation authors hypothesise that this could reflect once infected with COVID-19 compared to challenges in access. These experiences all other underlying health conditions [OR highlight the importance of specific infor- 2.10 (95% CI 1.2–3.7)]. Only heart failure mation and targeted communication for and organ transplantations had higher risk people and whānau, as well as to the work- ratios. Similarly, the risk of death for people force, to support vaccination uptake and with severe mental illnesses was higher implementation. than all other underlying health conditions As a result of the evidence review and [OR 2.9 (95% CI 1.3–6.6)], except for lung their knowledge and experience of the disease, heart failure and cancer (not in physical health inequities experienced by remission).19 The quality of the evidence people with mental health and addiction was considered strong. In response to the issues, the expert advisory group made the findings of the Robert Koch Institute review, following recommendations in the position the German government have prioritised statement.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 130 www.nzma.org.nz/journal viewpoint

• That people with experience of and addiction issues and accelerated ageing mental health and addiction issues remained after the study took account for be included in the COVID-19 vaccine childhood physical health, adversities, sequencing framework within the socioeconomic status, smoking and weight. priority group ‘people aged under 65 Accelerated aging is an important measure with underlying health conditions and as it provides a risk marker prior to the disabilities.’ onset or diagnosis of physical illnesses, such • The initial priority within this group as CVD or diabetes. should be adults (18 years and older) The evidence shows an independent currently accessing secondary mental association between infection and poor health and addiction services and outcomes from COVID-19 for people across people with long-term mental health the range of mental health and addiction and addiction issues in continuing diagnoses. When this is combined with the primary care. evidence on accelerated ageing and well-es- • That a specific information and tablished body of knowledge on coexisting communication programme be physical health issues and premature developed for this population, designed mortality, the case is sadly compelling. with people with lived experience and The evidence is clear. Mental health and cultural leaders, to support vaccination addiction issues result in earlier onset of uptake and implementation.13 chronic physical conditions, and if vacci- nation prioritisation is based on physical Prioritising the physical health of health conditions alone, then it will exac- people with mental health and erbate existing health inequities for people addiction issues with experience of mental health and The findings of the literature review addiction issues. We need to change the on COVID-19 vulnerability align with the kōrero. Health policy should identify people established New Zealand and international who experience mental health and addiction evidence of the higher prevalence of a wide issues ipso facto as a priority group. An range of coexisting physical health issues inclusive policy approach would enable for people with addiction and mental health people to take action at systems and practice issues.22–26 The reasons for these associa- levels to help achieve physical health equity tions have not yet been fully explained, but for people experiencing mental health and are thought to be based on causal mecha- addiction issues. nisms and shared determinants.27 Causal mechanisms include a variety of biological, Supplementary note psychological, pharmacological and service Since this manuscript was first submitted, delivery pathways, combined with racism the Ministry of Health has taken up the and discrimination.9,28 Shared determinants recommendations of the advisory group. include socioeconomic and environmental On 1 June 2021 the list of relevant under- factors, particularly childhood adversities lying health conditions that increase the and generic risk factors.27,29,30 risk of worse outcomes from COVID-19 was updated. This list now includes people who Another recent New Zealand study makes “have been diagnosed with severe mental an important contribution to knowledge in illness (which includes schizophrenia, this area.25 Four measures of ageing were major depressive disorder, bipolar or schi- examined across the prospective cohort of zoaffective disorder, and adults currently people at age 45. People with mental health accessing secondary and tertiary mental and addiction issues were found to be health and addiction services).” (Source: ageing up to five years ahead of their actual Ministry of Health, 2021 COVID-19 vaccine age. This association between mental health rollout groups, covid19.govt.nz.)

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 131 www.nzma.org.nz/journal viewpoint

Competing interests: Nil. Acknowledgements: The authors would like to acknowledge the Te Pou research team who conducted the litera- ture review. In particular: Talya Postelnik, Dr Angela Jury, Maria Basabas. Along with Kirstie Saumure (Ministry of Health – Manatū Hauora) and Matthew Jenkins (University of Otago, Wellington – Te Whare Wānanga o Otāgo ki Poneke), who assisted with literature searches. Author information: Helen Lockett: Strategic Policy Advisor, the Wise Group, Hamilton; Honorary Senior Research Fellow, Department of Public Health, University of Otago, Wellington – Te Whare Wānanga o Otāgo ki Poneke, Wellington; Honorary Research Fellow, Department of Psychological Medicine, University of Auckland – Te Whare Wānanga o Tāmaki Makaurau, Auckland. Ashley Koning: Principal Advisor Addiction, Te Pou, Auckland. Cameron Lacey (Te Atiawa): Senior Lecturer, Māori Indigenous Health Institute (MIHI), University of Otago, Christchurch – Te Whare Wānanga o Otāgo ki Ōtautahi, Christchurch Susanna Every-Palmer: of Department, Department of Psychological Medicine, University of Otago, Wellington – Te Whare Wānanga o Otāgo ki Poneke, Wellington. Kate M Scott: Head of Department, Department of Psychological Medicine, University of Otago – Te Whare Wānanga o Otāgo, Dunedin. Ruth Cunningham: Senior Research Fellow, Director EleMent Research Group, Department of Public Health, University of Otago, Wellington – Te Whare Wānanga o Otāgo ki Poneke, Wellington. Tony Dowell: Professor of Primary Health Care and General Practice, University of Otago, Wellington – Te Whare Wānanga o Otāgo ki Poneke, Wellington. Linda Smith: Consumer Advisor, Canterbury District Health Board – Te Poari Hauora ō Waitaha, Christchurch. Alison Masters: Medical Director, Mental Health, Addictions & Intellectual Disability Service – Te-Upoko-me-te-Karu-o-Te-Ika, Wellington. Arran Culver: Chief Clinical Advisor, Ministry of Health – Manatū Hauora, Wellington. Stephen Chambers: Professor (Pathology), Department of Pathology and Biomedical Science, University of Otago, Christchurch – Te Whare Wānanga o Otāgo ki Ōtautahi, Christchurch. Corresponding author: Dr Helen Lockett, PO Box 307, Waikato Mail Centre, Hamilton 3240, 0064 (0) 27 558 4658 [email protected] URL: www.nzma.org.nz/journal-articles/addressing-structural-discrimination-prioritising-peo- ple-with-mental-health-and-addiction-issues-during-the-covid-19-pandemic-open-access

REFERENCES 1. Cunningham R, Sarfati D, with-diabetes-oct15.pdf disease risk assessment Peterson D, et al. Prema- 3. Lockett H, Bagnall C, and management for ture mortality in adults Cunningham R, Arcus K. primary care . Wellington: using New Zealand psychi- Cardiovascular disease Ministry of Health; 2018 atric services. N Z Med J. risk and management in [cited 2021 Mar 4]. 36 p. 2014;127(1394):31-41. people who experience Available from: https:// 2. Ministry of Health [Inter- serious mental illness: An www.health.govt.nz/ net]. Living Well with evidence review [Internet]. system/files/documents/ Diabetes: A plan for people Hamilton: Te Pou; 2017 publications/cardiovas- at high risk of or living [cited 2021 Mar 11]. 7 p. cular-disease-risk-assess- with diabetes . Wellington: Available from: https:// ment-management-prima- Ministry of Health; 2015 www.tepou.co.nz/uploads/ ry-care-feb18-v4_0.pdf [cited 2021 Mar 4]. 39 p. files/resources/CVD-risk- 5. Ministry of Health [Inter- Available from: https:// in-people-with-SMI.pdf net]. New Zealand Cancer www.health.govt.nz/ 4. Ministry of Health Action Plan 2019-2029 system/files/documents/ [Internet]. Cardiovascular - Te Mahere mō te Mate publications/living-well- Pukupuku o Aotearoa

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 132 www.nzma.org.nz/journal viewpoint

2019–2029. Wellington: Mar 4]. 36 p. Available evidence on COVID-19 risk Ministry of Health; 2019 from: https://www. for people experiencing [cited 2021 Mar 4]. 76 p. tepou.co.nz/resources/ mental health and addic- Available from: https:// achieving-physical-health- tion issues. Auckland: Te www.health.govt.nz/ equity-for-people-with-ex- Pou; 2021 [cited 2021 Mar system/files/documents/ perience-of-mental-health- 12]. 10 p. Available from: publications/new-zea- and-addiction-issues--- https://www.tepou.co.nz/ land-cancer-action-plan-re- evidence-update-july-2020 resources/the-evidence- vised-january-2020.pdf 10. Ministry of Health on-covid-19-risk-for-peo- 6. Cunningham R, Sarfati [Internet]. Whakamaua: ple-experiencing-mental- D, Stanley J, et al. Cancer Māori Health Action Plan health-and-addiction-issues survival in the context of 2020–2025. Wellington: 15. De Picker LJ, Dias MC, mental illness: a national Ministry of Health; 2020. Benros ME, et al. Severe cohort study. Gen Hosp 78 p. Available from: mental illness and Psychiatry. 2015;37(6):501- https://www.health.govt. European COVID-19 vacci- 6. doi: 10.1016/j. nz/system/files/documents/ nation strategies. Lancet genhosppsych.2015.06.003. publications/whakam- Psychiatry. 2021:S2215- 7. Livingston JD. Structural aua-maori-health-ac- 0366(21)00046-8. stigma in health-care tion-plan-2020-2025-2.pdf doi: 10.1016/ contexts for people 11. Ministry of Health S2215-0366(21)00046-8. with mental health and [Internet]. Kia kaha, kia 16. Yang H, Chen W, Hu substance use issues: A māia, kia ora Aotearoa: Y, et al. Pre-pandemic literature review. Ottowa: COVID-19 psychosocial and psychiatric disorders Mental Health Commis- mental wellbeing plan. and risk of COVID-19: sion of Canada; 2020 Wellington: Ministry of a UK Biobank cohort [cited 2021 Mar 4]. 36 p. Health; 2020 [cited 2021 analysis. Lancet Healthy Available from: https:// Mar 4]. 53 p. Available Longev. 2020;1(2):e69- www.mentalhealthcom- from: https://www.health. e79. doi: 10.1016/ mission.ca/sites/default/ govt.nz/system/files/ S2666-7568(20)30013-1. files/2020-07/structural_ documents/publications/ 17. Fond G, Pauly V, Orleans V, stigma_in_healthcare_eng. kia-kaha-kia-maia-kia-ora- et al. Increased in-hospital pdf aotearoa-covid-19-psycho- mortality from COVID-19 in 8. Cunningham R, Rijnberg social-mental-wellbeing- patients with schizophre- V, Gibbs S. Premature plan-21dec2020-v2.pdf nia. Encephale. 2020 Jul mortality among people 12. Mazereel V, Van Assche 30:S0013-7006(20)30188- using secondary mental K, Detraux J, De Hert M. 3. doi: 10.1016/j. health and addiction COVID-19 vaccination for encep.2020.07.003. services in Aotearoa New people with severe mental 18. Nemani K, Li C, Olfson Zealand: Scoping a key illness: why, what, and M, Blessing EM, et al. performance indicator how? Lancet Psychiatry. Association of psychiatric [Internet]. Wellington: 2021:S2215-0366(20)30564- disorders with mortality Mental Health and 2. doi: 10.1016/ among patients with Addiction Key Performance S2215-0366(20)30564-2. COVID-19. JAMA Psychiatry. Indicator Programme; 13. COVID-19 vaccine expert 2021:e204442. doi: 10.1001/ 2020 [cited 2021 Mar 4]. advisory group [Internet]. jamapsychiatry.2020.4442. Available from: https://new. Position statement from 19. Robert Koch Institut mhakpi.health.nz/wp-con- the mental health and [Internet]. Epidemiolo- tent/uploads/2021/03/ addiction COVID-19 vaccine gisches Bulletin: Beschluss Final-Premature-mor- expert advisory group: der STIKO zur 2. tality-among-second- January 2021. Auckland: Aktualisierung der ary-MHA-service-us- Te Pou; 2021 [cited 2021 COVID-19-Impfempfehlung ers-in-NZ.pdf Mar 4]. 4 p. Available und die dazugehörige 9. Te Pou [Internet]. Achiev- from: https://www.tepou. wissenschaftliche Begründ- ing physical health equity co.nz/resources/position- ung. Berlin: Robert Koch for people with experience statement-from-the-mental- Institut; 2021 [cited 2021 of mental health and health-and-addiction-covid- Mar 5]. 84 p. Available addiction issues: Evidence 19-vaccine-expert-adviso- from: https://www.rki. update. Auckland: Te ry-group-january-2021 de/DE/Content/Infekt/ Pou; 2020 [cited 2021 14. Te Pou [Internet]. The EpidBull/Archiv/2021/

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 133 www.nzma.org.nz/journal viewpoint

Ausgaben/05_21. 24. Momen NC, Plana-Ripoll 28. Cunningham R, Stanley pdf?__blob=publicationFile O, Agerbo E, et al. Asso- J, Haitana T, et al. The 20. Toubasi AA, AbuAnzeh ciation between mental physical health of Māori RB, Tawileh HB, et al. disorders and subsequent with bipolar disorder. A meta-analysis: the medical conditions. N Engl Aust N Z J Psychiatry. mortality and severity J Med. 2020;382(18):1721- 2020;54(11):1107-1114. doi: of COVID-19 among 1731. doi: 10.1056/ 10.1177/0004867420954290. patients with mental NEJMoa1915784. 29. Te Pou o te Whakaaro Nui disorders. Psychiatry 25. Wertz J, Caspi A, Ambler [Internet]. The physical Research. 2021 Mar A, et al. Association of health of people with 3:113856. doi: 10.1016/j. history of psychopathology mental health conditions psychres.2021.113856. with accelerated aging at and/or addiction. Auckland: 21. The OpenSAFELY Collabo- midlife. JAMA Psychiatry. Te Pou o te Whakaaro rative, MacKenna B, Curtis 2021:e204626. doi: 10.1001/ Nui; 2017 [cited 2021 HJ, et al. Trends, regional jamapsychiatry.2020.4626. Mar 5]. 144 p. Available variation, and clinical 26. Richmond-Rakerd LS, from: https://www. characteristics of COVID-19 D’Souza S, Milne BJ, et al. tepou.co.nz/resources/ vaccine recipients: a retro- Longitudinal associations the-physical-health-of-peo- spective cohort study in of mental disorders with ple-with-mental-health-con- 23.4 million patients using physical diseases and ditions-and-or-ad- OpenSAFELY. medRxiv. mortality among 2.3 diction-evidence-up- 2021. doi: 10.1101/ million New Zealand date-december-2017 2021.01.25.21250356. citizens. JAMA Netw 30. Te Pou o te Whakaaro Nui 22. Lockett H, Jury A, Tuason Open. 2021;4(1):e2033448. [Internet]. The physical C, et al. Comorbidities doi: 10.1001/jamanet- health of people with a between mental and workopen.2020.33448. serious mental illness and/ physical health problems: 27. Scott KM, Lim C, Al-Hamza- or addiction: Evidence An analysis of the New wi A, et al. Association summary. Auckland: Zealand Health Survey of mental disorders with Te Pou o te Whakaaro data. New Zeal J Psychol. subsequent chronic Nui; 2014 [cited 2021 2018;47(3):5-11. physical conditions: World Mar 5]. 14 p. Available from: https://www. 23. Jury A, Koning A, Lai J, et mental health surveys tepou.co.nz/resources/ al. Hazardous drinking and from 17 Countries. JAMA the-physical-health-of-peo- general practitioner visits Psychiatry. 2016;73(2):150- ple-with-a-serious-mental- in the past year. N Z Med J. 8. doi: 10.1001/ illness-and-or-addiction- 2020 Jul 17;133(1518):43-53. jamapsychiatry.2015.2688. an-evidence-review

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 134 www.nzma.org.nz/journal viewpoint

How were medical students from Christchurch, New Zealand, involved in their COVID-19 response? Matthew J Cowie, Cicely V M Barron, Anna G Bergin, Noella N Farrell, India G Hansen

ABSTRACT Medical students from the University of Otago, Christchurch Department of Medicine were involved in their local COVID-19 response. A group of ten students helped with the assessment of individuals at community-based assessment centres or mobile testing units. They primarily helped assess and test individuals alongside experienced healthcare workers. The students gained valuable clinical and public health experience. Key learning points were the risks of pandemic involvement, identifying local barriers to healthcare and developing an appreciation for an evolving health response. Overall, students felt that preparation for future involvement could benefit further pandemic responses.

arly in the academic year of 2020, placements for remaining students.3 This left our cohort of medical students at the us with an unusual amount of free time yet EUniversity of Otago in Christchurch re- a strong desire to be part of our own coun- ceived a lecture on the developing COVID-19 try’s response. pandemic. While many felt concerned by the situation unfolding overseas, we found How were it difficult to conceptualise this virus in New Zealand. This changed when international students involved? classmates were prevented from return- Across the country many medical students ing from overseas and we soon had our volunteered their time and skills in a variety first case in late February. Without a clear of ways. During the pandemic students protocol for medical students, we were left helped with contact tracing, looking after wondering: what would our role be in this children of healthcare workers and at incoming pandemic? the national health call-line.4, 5 In Christ- Turning to news and social media we church, a medical-student initiative formed became aware that some overseas medical a ‘volunteer army’ for those who wanted students had been part of their country’s to help. This was utilised when a group response. In Italy, early graduation for of ten in our fifth year of medicine were around 10,000 final-year students helped asked to assist at an initial ‘pop-up’ surveil- 6 supplement their overburdened workforce.1 lance testing location. There we initially The United Kingdom also prioritised qual- worked to transcribe and document details ification for final-year students to help the alongside healthcare workers. As our National Health Service.2 We experienced an involvement progressed, we were trained early country-wide lockdown and, together to perform nasopharyngeal swabs and use with clear daily communication, avoided a primary protective equipment (PPE). This large-scale outbreak. During lockdown only gave us the ability to work on the frontline final-year medical students could remain and help assess symptomatic individuals at in clinical settings, a decision that resulted established locations known as communi- in the cancellation of hundreds of hours of ty-based assessment centres (CBACs). It also gave us insight to share when new protocols

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 135 www.nzma.org.nz/journal viewpoint

were being developed for the first time. We workers were over-represented in COVID-19 soon recognised that many members of the cases and deaths. Knowing that our assis- public were unable to commute to get tested. tance might increase the chance of exposure Mobile testing units were established and to us and others in our living situations, we became one of our main roles. We worked were at first apprehensive. Thankfully the out of vehicles alongside local general more controlled situation in New Zealand practitioners to help assess and test in the made our assistance less daunting over community. We found ourselves helping time. We found our involvement rewarding, design new protocols for the transportation whether it was on the frontline or not, and of swabs and use of PPE. Some of us were the level of risk balanced well with learning also involved in testing high-risk individuals, opportunities. If student roles are to be such as staff from the New Zealand Police, explored in future responses, we suggest a Fire and Emergency New Zealand and local range of options that can be adjusted to the rest homes. We found that, regardless of level of risk. our involvement, we developed a variety Barriers to testing of skills that built on our previous medical Mobile testing was a novel idea and training (Table 1). quickly became one of our key roles as students. This took us into the homes of Key learning points those unable to commute to COVID-19 Risks of involvement testing sites. We visited a wide range of As medical students, we often find locations, including quarantine facilities, ourselves in the unique situation of support housing residencies, juvenile observing clinical situations with reduced detention centres and even rural locations risk. Usually there are protocols that protect outside of Christchurch. We regularly visited us from the risk of harm to ourselves or individuals who lacked transport, were of others. In this pandemic, many of us felt an older age or had disabilities. As taught able to assist despite the unknown risk with at medical school, a key social determinant 7 a new virus. Our overseeing medical deans of health is access to healthcare. On many agreed that ‘immersive and experiential occasions our visit was the first contact with learning’ was crucial for student education, healthcare services for an issue that had not but only ‘in the right roles, and with appro- been previously addressed. Sometimes it priate supervision and support’.3 Initially it became the role of the more senior health was hard for us to conceive what this would professional to provide other cares for the entail and whether the risks really were patient in addition to COVID-19 testing. known. Overseas we saw that healthcare Many houses were damp and cold. In some

Table 1: Summary of medical student involvement in the COVID 19 response in Christchurch, New Zealand.

Student role Skills developed History taking and documenting • Acquiring patient details and establishing a symptom history • Screening for symptoms associated with COVID-19 and for those in high-risk populations • Communication with a range of healthcare workers COVID-19 testing • Explaining and performing a nasopharyngeal swab • Application of PPE and understanding of the aseptic tech- nique Clinical assessment • Recording and interpretation of vitals including tempera- ture, oxygen saturations, respiratory rate, heart rate and blood pressure. Pandemic planning • Helping set up and design clinical protocols • Providing feedback on the current response

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 136 www.nzma.org.nz/journal viewpoint

houses there were children with respi- We quickly appreciated that working in this ratory symptoms likely exacerbated by their type of environment required communi- environment. These visits let us view living cation and flexibility, skills much needed as situations from the centre of a patient’s future doctors. home. This left a lasting impact on many of us and cemented a strong reminder of how Future directions risk factors and living conditions can impact In light of our contribution, we suggest upon health. that preparing medical students for future Planning an evolving response disaster or pandemic situations may be We observed how rapidly a public health beneficial. So far the impact of COVID-19 response needed to change during a new within New Zealand has been manageable at pandemic. As the number of confirmed the hospital level. We make note that further cases increased, the protocols for events may occur where the health response community testing had to adjust from day to might not match the health demand. Our day. We helped set up pop-up surveillance involvement built on previously taught sites, which allowed us to think through medical knowledge and, along with ad hoc the finer details, like ensuring the layout training and support, we felt confident in was efficient yet safe. One important lesson the roles we were given. However, there was the importance of ongoing commu- was much initial uncertainty about our role, nication. Every morning each CBAC had a and only a small group of us were involved. briefing session about any recent changes to Some medical schools provide a curriculum the criteria for testing, self-isolation advice for pandemic or disaster situations to or protocols. This reflected the evolving allow faster and more effective student nature of the pandemic and response to involvement.8 We propose that disaster the latest guidelines, which accounted for and pandemic training tailored to medical local area challenges. We were impressed student experience would be beneficial. that whenever the Ministry of Health Furthermore, establishing local pathways requested the need for surveillance testing for student involvement could help ease an the relevant primary health organisation set overburden health system if, or inevitably up an assessment centre within 24 hours. when, future events occur.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 137 www.nzma.org.nz/journal viewpoint

Competing interests: Nil. Acknowledgements: Professor Lutz Beckert for prompting us to write up our experiences, providing feedback and supporting us throughout the manuscript process. Diane Bos and Dr Alex Shaw for pro- viding the opportunity and involving us with such enthusiasm and guidance. The doctors we worked alongside who made us feel so welcome and included, with special mention to Dr Satish Mistry, Dr Tom Davies and Dr John Irvine. Finally, to the fellow students we valued working with and to Dr Dali Fan and Professor David Murdoch for initially establishing the ‘student army’ in Christchurch. Author information: Matthew J Cowie: 6th year medical student, Department of Medicine, University of Otago, Christchurch. Cicely V M Barron: 6th year medical student, Department of Medicine, University of Otago, Christchurch. Anna G Bergin: 6th year medical student, Department of Medicine, University of Otago, Christchurch. Noella N Farrell: 6th year medical student, Department of Medicine, University of Otago, Christchurch. India G Hansen: 6th year medical student, Department of Medicine, University of Otago, Christchurch. Corresponding author: Matthew Cowie, Department of Medicine, University of Otago, 2 Riccarton Avenue, Christchurch, 8140 [email protected] URL: www.nzma.org.nz/journal-articles/how-were-medical-students-from-christchurch-new-zea- land-involved-in-their-covid-19-response-open-access

REFERENCES 1. Ivana Kottasová [Internet]. 2020. Available from: able from: https://www. Thousands of medical https://medicaldeans. nzdoctor.co.nz/article/news/ students are being fast- org.au/md/2020/03/2020- public-health-roles-medical- tracked into doctors to March-20_principles-for- students-during-covid-crisis help fight the coronavirus. medical-student-roles-in- 6. Hayward and Lines-Mack- CNN. March 2020. Available COVID-19-health-workforce. enzie [Internet]. from: https://edition.cnn. pdf?fbclid=IwAR1UhmcEd- Coronavirus: Wider com/2020/03/19/europe/ fixI11SwNK0h-7ptork- Covid-19 testing starts medical-students-coro- jsg7V3HBzgnlpS4m- Christchurch, Waikato. navirus-intl/index.html NUvBslyxzWCspqg Stuff. April 2020. Available 2. Medical Schools Council 4. Lisa Davies [Internet]. from: https://www.stuff. [Internet]. Advice from Otago University medical co.nz/national/health/ Medical Schools Council students become nannies coronavirus/121074955/coro- to UK medical schools for children of senior navirus-wider-covid19-test- on actions surrounding colleagues amid Covid-19 ing-starts-in-christ- covid-19. March 2020. lockdown. TVNZ. April 2020. church-community. Available from: https:// Available from: tvnz.co.nz/ 7. Marmot and Allen. Social www.medschools. one-news/new-zealand/ determinants of health ac.uk/media/2620/ otago-university-medi- equity. American Journal of msc-covid-19-advice-for- cal-students-become-nan- Public Health. September uk-medical-schools.pdf nies-children-senior-col- 2014. https://doi.org/10.2105/ 3. Medical Deans Australia leagues-amid-covid-19-lock- AJPH.2014.302200 down and New Zealand [Inter- 8. Kaji, Coates, and Fung. net]. Principles to support 5. Martin Johnston [Internet]. A disaster medicine medical students’ safe Public health roles for curriculum for medical and useful roles in the medical students during students. Teaching and COVID‐19 health workforce. COVID crisis. New Zealand learning in medicine. April Sydney: MDANZ. March doctor. March 2020. Avail- 2010. 2;22(2):116-22.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 138 www.nzma.org.nz/journal viewpoint

Teleophthalmology in the post-coronavirus era Francesc March de Ribot, Anna March de Ribot, Kelechi Ogbuehi, Ruth Large

ABSTRACT 173,766 New Zealanders suffer from visual impairment. The associated health-system costs are $523 million in total, or $3,008 per person. Yet eighty percent (80%) of blindness is avoidable if detected on time. Public health services have an increasing workload but are also limited by material and technical resources. Optometry practices continually increase referrals (up to 100% in one year) that cannot be solved on time, reducing the efficiency of the service. Teleophthalmology works by improving the efficiency of screening and monitoring and integrating eye healthcare and by decreasing referrals by up to 40%.

ffering excellent quality care to the risk-of-vision-loss conditions. This situation population via our public healthcare is particularly problematic in elderly people, Osystem is one of our national priori- diabetics with comorbidities and specific ties. To seek consultation from an ophthal- collectives, such as the Māori population, mologist, patients usually make an appoint- who face greater barriers to accessing the ment with an optometrist (or a general healthcare system. practitioner (GP)) before being referred. The The extra work inadvertently causes continuously growing number of referrals delays in clinic appointments and leads to from optometry practices (more than 100% vision loss in patients with chronic ocular 1 in one year ) increases pressure on the public diseases.3 Our public healthcare system service. Telemedicine systems are an op- is facing increasing financial expenses to portunity to implement new solutions and deliver quality eye care.23 Telemedicine optimise eye care in the post-coronavirus era. offers one opportunity to implement new New Zealand patients receive govern- solutions.6 ment-funded care from ophthalmologists in the public sector. Patients typically visit Coronavirus pandemic an optometrist or GP to be referred to an A large portion of ophthalmology patients ophthalmologist for further consultation. In are at high risk for coronavirus mortality the private sector, optometry practices are (patients with macular degeneration aged increasing the pressure of the public service 85 years or more, people with diabetes with a continuously growing number of and patients who are immunocompro- referrals.1 In some cases, referrals increased mised). Consequently, ophthalmic services 100%1 in a year, representing, in the reduced in-person consultations to minimise Auckland District Health Board (DHB), more potential exposure to the disease for the than 1,000 patients per month. There are no eye care providers and at-risk populations. practical solutions for this situation, and the A recent international survey suggests that sustainability of the system is at risk. Because ophthalmology has been the most affected there are many referrals, it is a challenge to medical specialty, with a decrease of 79% review all patients in a timely manner.23 in the number of visits.7 For these reasons, In 2020, before the coronavirus pandemic, delays in ophthalmology appointments more than 14,250 people within the Counties significantly worsened, probably resulting 8 Manukau DHB were waiting to see an in vision loss. It is a priority to restart our ophthalmologist.4 The worst scenario is services as soon as possible. In New Zealand, when patients are lost to follow-up, which because of the coronavirus pandemic, an happens in up to 38%5 of cases, even in high- estimate of more than 40,000 patients have had been their appointments postponed.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 139 www.nzma.org.nz/journal viewpoint

(The estimate comes from extrapolating conferencing, telemonitoring and mobile the effect of lockdown in the Southern DHB healthcare.12 across the New Zealand population). There are three main modalities of teleophthalmology. (1) Store and forward Inequalities in health and remote technology is the most common. Images and patients documents are taken and forwarded to the New Zealand has a vast territory with a ophthalmologist for review. For example, an significant number of patients in rural and ophthalmologist reviews slit lamp or retina isolated areas. These people have limited images. (2) Real-time telemedicine offers access to health services, medical shortages video consultation in real time, similarly to and constrained financial resources. an in-person office visit. (3) Remote moni- Receiving medical care implies travel to toring allows a distant follow-up. urban areas, which costs time and money.9 Teleophthalmology helps with screening, Isolated communities suffer complications diagnosing and monitoring patients, and of eye diseases more than urban popula- includes, among others: (1) acute services: tions9 due to inequitable access to public triage, diagnosis and treatment; (2) moni- healthcare resources. Furthermore, there are toring and follow-up on some conditions; gaps in the efficiency and quality of eye care (3) offering explanations to patients; (4) eye services,9 creating dissatisfaction with the screening services. provision of care.9 Disadvantaged patients in urban areas are also isolated due to poverty, A web-based application allows the user social conditions, ethnic discrimination, prej- to remotely send information (eg, retinal udice, cultural ethics and religion,9 especially images and patient data). Telehealth options Māori, Pacific people and people with lower must be implemented properly to minimise socioeconomic status, who experience highly security risks. The software communication inequitable health outcomes. structure guarantees the protection of the data, which are secured and encrypted. The New Zealand ophthalmology data are anonymised and they are reviewed challenges by an ophthalmologist who looks for signs of pathology and decides the management. Our eye healthcare is challenged with Analysed images and reports are forwarded offering equitable and efficient eye care at a to the user. Teleophthalmology can screen, sustainable cost. The Eye Health Workforce refer, and monitor patient eye care while Service Review10 established the need to focusing on patient needs. There are create a clinical eye network for integrating different platforms available that can New Zealand’s clinical health services at the integrate with existing platforms for coordi- primary, secondary and tertiary levels. To nation with the hospital. increase access to health services, optimise workforces and develop innovative eye Teleophthalmology evidence health roles, eye health screening should be There is a high level of agreement between standardised. optometrists who use teleophthalmology The way to improve our eye and vision and conventional examination.13,14,15,16 The healthcare is to integrate healthcare profes- agreement can be as high as 100% in retina sionals’ services into the ophthalmologists’ disorders such as diabetic retinopathy and practices. By offering devices to obtain age-related macular degeneration,17 and and share clinical data between eye health agreement can be above 87% in cases of professionals, we can deliver an optimised glaucoma.14,16 The implementation of teleop- service to the community. New Zealand hthalmology services may decrease the is continuously improving, and has some number of referrals to a hospital-based eye 11 pioneer experiences. services by 40%,18,19 and the actual benefits may vary depending on a large number of What is teleophthalmology? factors. Teleophthalmology is the use of tele- Teleophthalmology could lead to signif- communication and information services icant savings in time and travel expenses. to remotely deliver eye healthcare. This In a pilot study connecting hospital includes telephone consultation, live video- ophthalmologists with optometrists, 96%

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 140 www.nzma.org.nz/journal viewpoint

of patients were satisfied, mainly because tions.29 The cost of ophthalmic imaging of a reduction in travel (96%), cost (92%) equipment and hardware can be prohib- and time (92%).20 In addition to the eye care itive, and the spending may be inefficient. application, advantages of digital imaging Also, teleophthalmology in the outpatient systems include short examination time, setting relies on the coordination of services electronic medical images and the ability to benefit from the evaluation. The currents of non-ophthalmologists to screen for eye structures are already overburdened to diseases.6 Teleophthalmology provides perform additional tasks and ensure patient secondary specialist advice in the diag- compliance with recommendations. 21 nosis and management of difficult cases. It Medical perspectives must adapt to also supports real-time tele-mentoring and teleophthalmology services. The evaluations -teaching. The economic analysis supports can be difficult; 59% of ophthalmologists the evidence of the cost-effectiveness of reported low confidence when making deci- teleophthalmology for diabetic retinopathy sions based only on images. The follow-up of 22 and glaucoma. Increasing screening could chronic conditions can be complicated. Also, improve accessibility, especially for rural medical liability is a reason for concern. and remote populations.22 Successful application of teleophthal- The impact of eye diseases mology requires the development of efficient structures and consistent training Loss of vision has a devastating effect on of the involved personnel. Improving image daily life.52 It is difficult to perform basic processing and integration of patient’s physical and social tasks24 and compli- medical care teleophthalmology may cates the management of other conditions. improve results.30 Depression is 3.5 times higher25 in people with moderate to severe vision loss, and Conclusions dementia progresses more rapidly. Loss of Telemedicine has the potential to increase vision increases the risk of falls, multiples access to care, decrease costs and improve the risk of fractures by 2.5 times, creates adherence to evidence-based protocols. dependence and causes 4.23 times more Teleophthalmology may optimise the admissions in nurse homes.26 referrals and help to offer a more efficient In New Zealand, we have 173,766 New service on time. Properly implemented, Zealanders suffering from vision loss. we anticipate perhaps a 40% decrease in The disability’s healthcare costs are $523 the number of referrals to public ophthal- 27 million in total, or $3,008 per person. Most mology services in New Zealand, which vision loss is preventable or treatable. The would improve the workflow in ophthal- most common causes of blindness in New mology departments of public hospitals by Zealand are macular degeneration (48%) about 20%. Connectivity solutions can help 27 and glaucoma (16%). Eighty percent (80%) offer better eye healthcare by exploiting the of blindness is treatable and/or preventable, benefits of teleophthalmology for remotely according to the World Health Organi- screening, referring and monitoring patients 28 zation. Strategies for restoring sight and integrating the services provided by eye and preventing blindness are among the health professionals. Now is the moment to most cost-effective healthcare interventions implement innovations so as not to leave 28 worldwide. anyone behind. As remote consultations become available, healthcare organisa- Limitations tions and clinicians will have to re-evaluate Improvements in technology-based traditional care delivery models while still services have decreased connectivity ensuring that evidence-based protocols problems and storage limitations. improve outcomes. Nevertheless, there are important limita-

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 141 www.nzma.org.nz/journal viewpoint

Competing interests: Ruth Large declares: I am the Chair of the New Zealand Telehealth Leadership Group, which is advisory to the Ministry of Health unrelated to decisions regarding funding. I am the Chief Clin- ical Officer for Whakarongorau Aotearoa (formerly known as Homecare Medical), which sup- plies the Ministry of Health contracted services. such as Healthline, COVID line and 1737. There are no direct relationships to ophthalmology service supply in the either of these two roles. Author information: Francesc March de Ribot: Dunedin Hospital, University of Otago, Dunedin School of Medicine. Anna March de Ribot: Dunedin Hospital, University of Otago, Dunedin School of Medicine. Kelechi Ogbuehi: Dunedin Hospital, University of Otago, Dunedin School of Medicine. Ruth Large: Information services and virtual health care. Corresponding author: Francesc March de Ribot [email protected] URL: www.nzma.org.nz/journal-articles/teleophthalmology-in-the-post-coronavirus-era-open-access

REFERENCES 1. Ministry of Health - 5. Obeid A, Gao X, Ali FS, 2004;10(3):184-185. Ophthalmology Service Talcott KE, Aderman 10. Ministry of Health - Eye Demand. Official Responses CM, Hyman L, Ho AC, Health Workforce H201903332 Https:// Hsu J. Loss to Follow- Service Review Https:// Www.Health.Govt.Nz/ Up in Patients with Www.Health.Govt.Nz/ System/Files/Documents/ Proliferative Diabetic System/Files/Documents/ Information-Release/ Retinopathy after Panret- Pages/Eye-Health-Re- H201903332.Pdf. inal Photocoagulation or view-May-2011.Pdf. Intravitreal Anti-VEGF 2. Hannah Martin. Thousands 11. Singh A, Cheyne K, Wilson Injections. Ophthalmology. of Aucklanders Wait for G, Sime MJ, Hong SC. On the 2018 Sep;125(9):1386-1392. Overdue Specialist Eye Use of a New Monocular-In- Appointments Feb 26 6. Yogesan K, Kumar S, direct Ophthalmoscope for 2020. Press Https://Www. Goldschmidt L, Cuadros Retinal Photography in a Stuff.Co.Nz/National/ J. Teleophthalmology. Primary Care Setting. N Z Health/119779366/ Germany:Springer; 2006. Med J. 2020;133(1512):31- Thousands-of-Aucklanders- 7. Ateev Mehrotra et al., 38. Published 2020 Apr 3. Wait-for-Overdue-Special- “What Impact Has COVID- 12. RANZCO Position ist-Eye-Appointments. 19 Had on Outpatient Statement – Teleophthal- 3. Dr James Stewart, Kate Visits?,” To the Point (Blog), mology in New Zealand MacIntyre. Review of 34 Commonwealth Fund. Https://Ranzco.Edu/ Ophthalmic (Eye) Incidents April 23, 2020. Https://Doi. Policies_and_guideli/Ranz- in the Southern DHB Org/10.26099/Ds9e-Jm36. co-Position-Statement-Te- (SDHB) Identified in the 8. Paul Mitchel. Press. leophthalmology-in-New-Zea- Period 1 July 2015 to 30 4 April 2020 Https:// land. September 2016. Https:// Www.Miragenews.Com/ 13. Ang GS, Ng WS, Www.Southerndhb.Govt. Australia-Faces-Risk-of-In- Azuara-Blanco A. The Nz/Files/20058_2017 creased-Blind- Influence of the New 0517140858- ness-amid-Corona- General Ophthalmic 1494986938.Pdf. virus-Measures/. Services (GOS) Contract in 4. Amy Wiggins. More than 9. Labiris G, Petounis A. Optometrist Referrals for 20,000 People Overdue for A Framework to Assess Glaucoma in Scotland. Eye Eye Treatment. 21 Aug, the Readiness for (Lond) 2009;23(2):351-355. 2017. Press Https://Www. Teleophthalmology of 14. Wright HR, Diamond JP. Nzherald.Co.Nz/Nz/News/ Glaucoma Patients Living Service Innovation in Article.Cfm?C_id=1&- in Isolated Communities. Glaucoma Management: objectid=11908350. J Telemed Telecare Using a Web-Based

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 142 www.nzma.org.nz/journal viewpoint

Electronic Patient Record 20. Tuulonen A, Ohinmaa T, M, Lamoureux EL. to Facilitate Virtual Alanko HI, Hyytinen P, Vision-Specific Distress Specialist Supervision of Juutinen A, Toppinen E. and Depressive Symptoms a Shared Care Glaucoma The Application of Teleop- in People with Vision Programme. Br J Ophthal- thalmology in Examining Impairment. Invest mol 2015;99(3):313-317. Patients with Glaucoma: Ophthalmol Vis Sci. 15. Keenan J, Shahid H, Bourne A Pilot Study. J Glaucoma 2010 Jun;51(6):2891-6. RR, White AJ, Martin KR. 1999;8(6):367-373. 26. Klein BE, Moss SE, Klein Cambridge Community 21. Kennedy C, Van Heerden A, R, Lee KE, Cruickshanks Optometry Glaucoma Cook C, Murdoch I. Utiliza- KJ. Associations of Scheme. Clin Exp Ophthal- tion and Practical Aspects Visual Function with mol 2015; 43(3):221-227. of Tele-Ophthalmology Physical Outcomes and 16. De Mul M, de Bont AA, between South Africa and Limitations 5 Years Later Reus NJ, Lemij HG, the UK. J Telemed Telecare in an Older Population: Berg M. Improving the 2001;7 Suppl 1:20-22. The Beaver Dam Eye Quality of Eye Care with 22. Sharafeldin N, Kawa- Study. Ophthalmology. Tele-Ophthalmology: guchi A, Sundaram A, 2003;110(4):644–650. Shared-Care Glaucoma Campbell S, Rudnisky 27. Blind Foundation - Clear Screening. J Telemed Tele- C, Weis E, Tennant MTS, Focus - The Economic care 2004;10(6):331-336. Damji KF. Review of Impact of Vision Loss 17. Kawaguchi A, Sharafeldin Economic Evaluations in New Zealand Https:// N, Sundaram A, et al. of Teleophthalmology as Blindlowvision.Org.Nz/. Tele-Ophthalmology for a Screening Strategy for 28. World Health Organi- Age-Related Macular Chronic Eye Disease in zation - Blindness and Degeneration and Diabetic Adults. Br J Ophthalmol. Vision Impairment Retinopathy Screening: 2018 Nov;102(11):1485-1491. Https://Www.Who.Int/ A Systematic Review and 23. Scott AW, Bressler NM, News-Room/Fact-Sheets/ Meta-Analysis. Telemed Ffolkes S, Wittenborn Detail/Blindness-and-Vi- J E Health 2017. JS, Jorkasky J. Public sual-Impairment. 18. Henson DB, Spencer AF, Attitudes About Eye and 29. Rathi S, Tsui E, Mehta N, Harper R, Cadman EJ. Vision Health. JAMA Zahid S, Schuman JS. The Community Refinement of Ophthalmol. 2016 Oct Current State of Teleoph- Glaucoma Referrals. Eye 1;134(10):1111-1118. thalmology in the United (Lond) 2003;17(1):21-26. 24. Crews JE, Jones GC, Kim States. Ophthalmology. 19. Haymes SA, Johnston AW, JH. Double Jeopardy: The 2017 Dec;124(12):1729-1734. Heyes AD. Relationship Effects of Comorbid Condi- 30. Sommer AC, Blumenthal between Vision Impairment tions among Older People EZ. Telemedicine in and Ability to Perform with Vision Loss. Journal Ophthalmology in View Activities of Daily Living. of Visual Impairment and of the Emerging COVID-19 Ophthalmic and Physiolog- Blindness. 2006;100:824. Outbreak. Graefes Arch ic Optics. 2002;22(2):79–91. 25. Rees G, Tee HW, Marella Clin Exp Ophthalmol. 2020 M, Fenwick E, Dirani Nov;258(11):2341-2352.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 143 www.nzma.org.nz/journal viewpoint

COVID-19 serology: use and interpretation in New Zealand Gary N McAuliffe, Timothy K Blackmore

ABSTRACT Serology is now a well-established diagnostic tool for the diagnosis of COVID-19 infections in New Zealand. Using local and international experience, we provide an overview of serological response to infection and vaccination as well as the use and interpretation of antibody tests in our setting. We also discuss the potential future role of post-vaccination serology testing as a correlate of immunity. We conclude that, given the pitfalls of testing, clinical microbiologist advice is necessary for interpretation of high-consequence cases.

ucleic acid amplification tests as IgG and may persist for months, it is not (NAAT), such as reverse transcrip- reliable as a marker of acute versus past Ntase PCR or transcription- mediated infection.3,4 amplification, are the most widely used tests Antibody tests used in New Zealand’s diag- in New Zealand for the diagnosis of acute nostic laboratories generally perform well COVID-19 infections. But serology, which for the detection of past infection, demon- detects an immune response to past SARS- strating sensitivities >90% by 14 days after CoV-2 infections or vaccination, is also now onset of symptoms and specificities >99%.5 well established as a diagnostic tool. Antibodies wane over time and appear to Several types of tests are used in New decline more rapidly to different antigens, Zealand diagnostic laboratories, including with only 54% of individuals still positive for plate-based enzyme-linked immunoassays anti-N compared with 96% to anti-S in a New (ELISAs), which are labour intensive but Zealand cohort at or more than 125 days suit low-throughput testing scenarios, after infection.3 The strength and duration and chemiluminescent assays, which are of the immune response is quite variable performed on higher-throughput machines between individuals and also differs with a faster sample to answer time. according to severity of disease. New Zealand has a very low prevalence Depending on the antigen to which the of COVID-19,1 and therefore, in contrast to antibodies bind, they may be either neutral- serology tests performed in diagnostic labo- ising or non-neutralising. In the case of ratories, point-of-care or lateral flow tests SARS-CoV-2, antibodies to the RBD/spike are have suboptimal sensitivity and specificity most likely to be neutralising and protective in our setting; since April 2020, importation against subsequent symptomatic infection. 2 of these devices has been restricted. In New Zealand, COVID-19 serology is not centrally funded but may be funded for Serological response to infection selected patients by district health boards. Following SARS-CoV-2 infection, antibodies Approval for testing is usually determined are produced against various viral proteins by the clinical microbiologist overseeing the including the receptor-binding domain testing laboratory. Use is primarily as part (RBD) of the spike (S) protein and to nucleo- of public health investigations, where it can capsid (N) (Figure 1). This antibody response confirm past infections in NAAT-negative involves IgA, IgM and IgG, which are individuals or when using paired acute and detectable concurrently, and in some indi- convalescent sera to differentiate acute from viduals as early as 0–5 days after symptom past infection in NAAT-positive/antibody-neg- onset. As IgM appears around the same time ative individuals. There are some countries

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 144 www.nzma.org.nz/journal viewpoint

that require pre-departure serology testing as nological response to vaccination can only part of their entry requirements; this testing be detected if laboratories test for these is undertaken as a fee for service by certain specific antibodies. For example, an indi- New Zealand laboratories. vidual receiving the Pfizer-BioNTech vaccine Given the very low prevalence of past produces neutralising antibodies directed COVID-19 infection in the general population against the RBD of the S protein and will test outside managed isolation facilities (MIF), positive for anti-S but negative for anti-N the positive predictive value of a positive unless the individual has also had natural antibody test varies widely between these infection (Table 1). Some inactivated virus populations. For example, a positive pre-de- vaccines in use outside New Zealand (eg, parture IgM has a positive predictive value Sinopharm) are expected to give a response approaching zero, whereas a positive total that may include anti-N. antibody from MIF is highly likely to be a Preliminary studies post Pfizer-BioNTech true positive. Consequently, confirmation vaccination have shown a crude correlation of a positive antibody result by testing on between positive results for anti-RBD or a second assay is warranted in some situa- anti-S and neutralising antibody production. tions but unnecessary in others. However, at this stage, the level of antibody The pitfalls of interpretation of serology required for immunity is not known; in the New Zealand context mean that quantitative values between different 6 discussion with a clinical microbiologist is manufacturers are highly variable; and required prior to testing for other indica- the longevity of antibody response (and tions, such as diagnosis of complications protection) is unclear. of COVID-19 (eg, myocarditis), and also for any positive or negative results that Role of serology in assessing may have individual or public health immunity consequence. With these factors in mind, serology is not currently recommended to assess for Serological response to vaccination immunity to COVID-19 in a vaccinated Vaccines produce an immune response person, or to assess the need for vaccination against specific viral proteins, and an immu- in an unvaccinated person4 (for those with

Figure 1: Schematic representation of SARS-CoV-2 structure.

RNA

Spike protein (S)

Envelope

Nucleocapsid (N)

Receptor binding domain (RBD)

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 145 www.nzma.org.nz/journal viewpoint

prior COVID-19 infection, vaccination is is associated with the development of recommended). This may change in the cellular immunity, but the relative contribu- future as correlates of immunity become tions of humoral and cellular immunity are clearer; studies looking at this are underway. unclear at present.7 New quantitative assays to detect anti-S antibodies have become available and may Summary be useful for measuring contemporaneous COVID-19 serology is currently available immunity, but formal studies of correlates in diagnostic laboratories in New Zealand as of protection are awaited.6 Dependent a test that requires microbiologist approval. on these findings, future applications of Testing may provide useful information serology may include testing antibody levels in public health investigations or select in certain groups after vaccination, after a cases of post-infectious complications and COVID-19 exposure event in others, or for is necessary for overseas travel to some allowing risk-stratified quarantine decisions destinations. However, test reliability to be made by confirming vaccination or varies substantially according to the testing immunity status in travellers. scenario. Depending on the available tests, vaccine response and natural infection can Other immune responses to be differentiated, but the role of post-vac- vaccination or infection cination serology testing as a correlate of T-cell immunity is likely to play an immunity has not yet been determined. We important role, but at present there are no conclude that, given the pitfalls of testing, commercially available or easily standard- clinical microbiologist advice is necessary for isable assays. It appears that seroconversion interpretation of high-consequence cases.

Table 1: Antibody response to infection or Pfizer-BioNTech vaccine.

Anti-S Anti-N Natural infection + +

Pfizer-BioNtech vaccine response + -

Prior natural infection and post-Pfizer-BioNtech vaccine + + (-)*

Legend: *anti-N may wane faster than anti-S, and with time some infected individuals may test negative.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 146 www.nzma.org.nz/journal viewpoint

Competing interests: Nil. Acknowledgements: Thanks to all the clinicians, scientists and researchers in New Zealand who have contributed to knowledge and discussion on this topic. Author information: Gary N McAuliffe: Clinical Microbiologist and Virologist, LabPlus, Auckland hospital, Auckland. Timothy K Blackmore: Infectious Diseases Physician and Microbiologist, Wellington SCL and Capital & Coast District Health Board, Wellington. Corresponding author: Dr Gary McAuliffe, Clinical Microbiologist and Virologist, Department of Virology and Immunology, LabPlus, Auckland hospital, Park Road, Grafton, Auckland, 021 02157069 [email protected] URL: www.nzma.org.nz/journal-articles/covid-19-serology-use-and-interpretation-in-new-zea- land-open-access

REFERENCES 1. New Zealand Ministry of 3. Whitcombe, A L, McGre- comparison of SARS-CoV-2 Health [Internet]. COVID-19 gor, R, Craigie, A, et al. antibody assays evaluated current cases; 2021 [cited (2021). Comprehensive in Auckland, New Zealand. 2021 20 Mar]. Available analysis of SARS-CoV-2 N Z Med J. 2020;133:127-131. from: https://www. antibody dynamics in 6. Perkmann T, Perk- health.govt.nz/our-work/ New Zealand. Clinical mann-Nagele N, Koller T, diseases-and-conditions/ & translational immu- et al. Anti-Spike protein covid-19-novel-coronavirus/ nology, 10(3), e1261. assays to determine covid-19-data-and-statistics/ 4. Centers for Disease Control post-vaccination antibody covid-19-current-cases and Prevention [Inter- levels: a head-to-head 2. Medsafe [Internet]. COVID- net]. Interim guidelines comparison of five quan- 19 point of care test kits for COVID-19 Antibody titative assays. medRxiv restriction on importation, testing; 2021 [cited 2021 2021.03.05.21252977 manufacture, supply and 20 Mar]. Available from: 7. Flanagan KL, Best E, use; 2020 [cited 2021 20 https://www.cdc.gov/ Crawford NW, et al. Mar]. Available from: coronavirus/2019-ncov/ Progress and Pitfalls in https://www.medsafe. lab/resources/anti- the Quest for Effective govt.nz/Medicines/ body-tests-guidelines.html SARS-CoV-2 (COVID-19) policy-statements/COVID- 5. Fox-Lewis S, Whitcombe Vaccines. Front Immunol. 19PointOfCareTestKits.asp A, McGregor R, et al. A 2020 Oct 2;11:579250.

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 147 www.nzma.org.nz/journal clinical correspondence

Dangers of a single pellet Jeong Ha, Olga Korduke, Megan Rodney, Peter Stiven

21-year-old man was brought in with a partial left bladder wall injury, near the a shotgun wound centred at the right entry site. The small bowel perforation sites A upper thigh. He was agitated with and the bladder were suture repaired. cool peripheries. His clothing and a sheet were soaked with blood. Although his heart Discussion rate was elevated at 110bpm, he was nor- Selective non-operative management for motensive with systolic blood pressure of penetrating abdominal gunshot wounds 115mmHg. There was an open wound over is becoming an accepted alternative to the the right medial thigh with ongoing exsan- more traditional mandatory laparotomy.1 guination. A tight compression bandage was applied to the right thigh for haemostasis. Figure 1: X-ray of pelvis and hip. The right dorsalis pedis pulse was palpable, though there appeared to be no movement and only patchy sensation below the knee. The abdomen was soft and seemingly non-tender. FAST scan of the abdomen was negative. X-ray of the hip and pelvis showed multiple gunshot pellets in the right thigh, four pellets in the left thigh and a single pellet in the pelvis. CT angiogram showed no major vascular injury of lower limbs. No comment was made regarding the intra-abdominal shotgun pellet, though this is visible in retro- spect. The patient was taken to the operating theatre for exploratory and damage control surgery in co-operation with orthopaedic colleagues. The right thigh was debrided of all non-viable tissue. The path of damage Figure 2: A coronal view CT angiogram of lower extended to subcutaneous tissues on the abdomen and thigh. contralateral side. Extensive injury to the adductor muscles was noted. The sciatic nerve was observed and intact, though this was difficult to verify. We then performed a diagnostic lapa- roscopy in light of the single pellet evident on imaging. This demonstrated gas bubbles under the omentum, fibrin deposits and turbid fluids in the pelvis. A small amount of blood was evident in the left iliac fossa, suggesting the site of penetration. We converted to laparotomy to further examine the intra-abdominal contents. Five perfo- ration sites were found in the small bowel; two loops of bowel had through-and-through perforations and the fifth perforation had a shotgun pellet still lodged within. There was

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 148 www.nzma.org.nz/journal clinical correspondence

Patients with hemodynamic stability and trajectory alone is a recommended indi- benign abdominal findings may be kept for cation for laparotomy.3 Although the trauma serial observation and have similar overall guideline from Victoria, Australia recognises mortality outcome.2 However, these studies that serial physical examination may be were mostly conducted in level 1–2 trauma a reliable approach to detect a significant centres in North America with high volumes injury, this is only in an alert and orientated of penetrating injuries related to firearms. patient without distracting injuries.4 In the rural New Zealand hospital setting, the infrequency of these cases and lack of Conclusion resources after hours may lead to worse While it remains an uncommon presen- outcomes with a conservative approach. tation, recent figures suggest gun violence in Furthermore, in our case a significant New Zealand may be on the rise.5 The chal- distracting injury was present, with a degree lenge specific to managing shotgun injury is of hypovolaemic shock, making a conser- the wide field of damage caused by multiple vative approach inappropriate. Australasian pellets. It is easy to overlook a single pellet recommendations continue to lean towards when assessing a patient with distracting laparotomy. The rate of significant intra-ab- injuries. However, a single pellet is capable dominal injury can be as high as 98% of causing significant injuries. A clinician when peritoneal penetration is present should be vigilant and wary of a stray and gunshot wound with a transperitoneal innocuous appearing single pellet.

Competing interests: Nil. Acknowledgements: Gisborne Hospital Surgical Department. Author information: Jeong Ha: MBChB; Surgical Registrar, Gisborne Hospital. Olga Korduke: MBChB; Surgical Registrar, Gisborne Hospital. Megan Rodney: MBChB; House Officer, Gisborne Hospital. Peter Stiven: MBChB, FRACS; Consultant General Surgeon, Gisborne Hospital. Corresponding author: Corresponding author: Jeong Ha, Department of Surgery, Gisborne Hospital, Gisborne Hospital, 421 Ormond Road, Riverdale, Gisborne 4010, New Zealand, 0211836016 [email protected] URL: www.nzma.org.nz/journal-articles/dangers-of-a-single-pellet

REFERENCES 1. Peponis T, Kasotakis ment of Abdominal Trauma. Victoria. [Updated G, Yu J et al. Selective Gunshot Wounds in the 15 January 2021; cited 2 Nonoperative Manage- United States. J Surg Res. March 2021]. Available ment of Abdominal 2020; 253: 224-231. from: https://trauma. Gunshot Wounds from 3. American College of reach.vic.gov.au/guide- Heresy to Adoption: A Surgeons. Committee of lines/abdominal-trauma/ Multicenter Study of the Trauma. Advanced Trauma introduction Research Consortium of Life Support: ATLS Student 5. New Zealand Police New England Centers Course Manual. 12th [Internet]. Firearms for Trauma (ReCoNECT). edn. Chicago: American Information Summary. J Am Coll Surg. 2017; College of Surgeons; 2018. [Updated March 2021, cited 224: 1036-1045. 4. Victorian State Trauma 2 March 2021]. Available 2. Masjedi A, Asmar S, Bible System [Internet]. Major from: https://www. L et al. The Evolution of Trauma Guidelines and police.govt.nz/about-us/ Nonoperative Manage- Education, Abdominal publication/firearms-data

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 149 www.nzma.org.nz/journal 100 years ago

Notes on a Case of “Brodie’s Abscess” or “Circumscribed Abscess” of the Lower End of the Left Femur 1921

he patient, R.A., a boy of fifteen years, the previous operation. The muscles being was admitted to Dr. Cooper’s private retracted, the tissues near the bone were Thospital, Eltham, complaining of pain very œdematous, and a thin serous fluid in the lower end of the left femur. The pain exuded from the periosteum when it was was “boring” in character, worse at night, incised. There was no subperiosteal abscess, and latterly prevented him from getting and the periosteum separated easily. With sleep. a gouge the medullary cavity was opened He gave the following history :—Was an over the site of the greatest œdema, and a in-patient of the Wanganui Hospital for circular opening, half an inch in diameter, about twelve months (1916–1917), with an made down into the medulla of the bone. acute osteo-myelitis of the left femur, in the There was free oozing of blood-stained fluid lower third of the shaft. He had a metas- through this cavity, but no pus was found. tasis in the right humerus and still has a On introducing a probe bent at a right angle, sinus there. No sinus was present over the it could be easily passed down the medulla femur—only a four-inch operation scar on towards the epiphyseal line, but its upward its outer aspect, about a hand’sbreath above passage was barred by a bridge of sclerosed the knee-joint. bone. Bearing in mind that localised abscess is usually surrounded by dense bone, While under observation here he ran no another opening was then made into the temperature, and his pulse was normal. The shaft of the bone, about one inch above the examination showed definite tenderness previous one. A small circumscribed abscess on palpation all round the lower end of was opened, which contained about half a the left femur, and the shaft was expanded drachm of thick, yellowish pus. The abscess and thickened in the same situation. Part cavity was completely surrounded by of this expansion may have been due to the dense, sclerosed bone, and lined with a thin previous osteo-myelitis, but as the pain was granulation tissue, which was removed by becoming much worse an abscess of the curetting. No sequestrum was found in the bone was suspected. There was no disco- cavity. The abscess cavity was swabbed out louration of the skin, and no history of a with ungt. bipp. and packed with iodoform recent injury. gauze, and the wound closed in the usual Operation under general anæsthesia (C.E. manner below the drain. When the patient and open ether).—A four-inch incision, a was seen later on in the day, he said that hand’sbreadth above the knee-joint, was he was absolutely free from pain. No X-ray made on the outer aspect of the limb over plate was taken before the operation. the site of the former operation, and the This case illustrates well the condition of scar excised. There was some little difficulty relapsing osteo­myelitis, with chronic abscess in defining the interval between the vastus formation, and shows how easy it would externus and the biceps, owing to the scar of be to miss a small abscess, unless the bone

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 150 www.nzma.org.nz/journal 100 years ago

suspected is explored in all directions. fully exploring the bone when symptoms I have to thank Dr. H. A. Cooper, of point to a bone infection, more especially Eltham, who performed the operation, for when an acute periostitis is exposed. In permission to publish the notes of this case. this particular case the abscess could quite easily have been missed, the case diagnosed Dr. Cooper adds the following points of as one of acute localised periostitis, and the interest:—(1) The sudden exacerbation patient sent back to bed without any relief of the patient’s symptoms after the long from his symptoms. When a probe could not quiescent period of five years. (2) The be passed up the shaft of the femur, in the small amount of pus in its confined space, medullary canal, it was surmised that the giving rise to such acute symptoms. (3) opening had been made below the abscess. The complete obliteration of the medullary The second opening proved this to be so. (5) canal, with a considerable increase in the The almost immediate relief of pain when girth of the bone. (4) The importance of care- the abscess was opened.

URL www.nzma.org.nz/journal-articles/notes-on-a-case-of-brodies-abscess-or-circumscribed-ab- scess-of-the-lower-end-of-the-left-femur

NZMJ 9 July 2021, Vol 134 No 1538 ISSN 1175-8716 © NZMA 151 www.nzma.org.nz/journal published by the New Zealand Medical Association

NZMJDigest http://www.nzma.org.nz/publications/nzmjdigest

The NZMA publishes the e-magazine NZMJDigest 10 times a year. It contains news and views from the profession and the NZMA, including the NZMA Chair's editorial, along with highlights from and links to the New Zealand Medical Journal. Click on the image above to view the latest issue. We welcome contributions from members and readers. To contribute to the NZMJDigest, please email [email protected]