Challenge for COVID-19 Vaccines to Protect the New Zealand Population

Te ara tika o te hauora hapori

Journal of the New Zealand Medical Association Vol 134 | No 1534 | 30 April 2021

Challenge for COVID-19 vaccines to protect the New Zealand population

Making the Convention on the Rights of Persons with Disabilities real: our word is our bond

Why is it so hard for general practice to provide COVID-19 vaccines?

Parental use of physical What does abortion law Predictors of medicinal cannabis users’ punishment in a birth reform mean for primary care willingness to utilise a new prescription cohort practitioners in New Zealand? Medicinal Cannabis Scheme in 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 Kate Baddock

NZMJ Production Editor NZMA Communications Manager Richard Beer Madeleine Boles de Boers (Acting)

Other enquiries to: To contribute to the NZMJ, first read: NZMA www.nzma.org.nz/journal/contribute PO Box 156 The Terrace © NZMA 2021 Wellington 6140 Phone: (04) 472 4741 Cover photo 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 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 2 www.nzma.org.nz/journal CONTENTS

EDITORIALS 66 8 Predictors of medicinal cannabis Making the Convention on the users’ willingness to utilise a new Rights of Persons with Disabilities prescription medicinal cannabis real: our word is our bond scheme in New Zealand James Cooney, Michael Daly, Marta Rychert, Chris Wilkins, Karl Parker, Giles Newton-Howes, Kris Gledhill, Thomas Graydon-Guy Sarah Gordon REVIEW ARTICLES 11 Challenge for COVID-19 76 vaccines to protect the Methoxyflurane toxicity: New Zealand population historical determination and Stephen T Chambers, Nigel J Raymond lessons for modern patient and occupational exposure ARTICLES Serah J Allison, Paul D Docherty, 17 Dirk Pons, J Geoffrey Chase Parental use of physical VIEWPOINTS punishment in a birth cohort Geraldine FH McLeod, John Horwood, 91 Joseph M Boden, Lianne J Woodward What does abortion law reform mean for primary care 31 practitioners in New Zealand? Demographic differences in Emma Macfarlane, Michael Stitely, the initiation and maintenance Helen Paterson of statins in the first year post ACS in New Zealand: a data linkage 99 Official Information study (ANZACS-QI 57) Aravindra Muniandy, Mildred Lee, Act investigation of the Ministry Corina Grey, Katherine Ferrier, of Health’s process to assess Andrew J Kerr the Southern District Health Board’s readiness to join the 46 National Bowel Screening The outcomes of patients Programme in 2018 with newly diagnosed exudative Philip Bagshaw, Paula Goodman, age-related macular degeneration Brian Cox in Palmerston North Aaron Yap, Adeline Kho, John Ah-Chan 114 Why is it so hard for 51 general practice to provide Comparison of physical-activity COVID-19 vaccines? patterns across large, Vanessa Weenink medium and small urban areas and rural settings in the Otago Region, New Zealand Brittany White, Enrique García Bengoechea, John C Spence, Kirsten J Coppell, Sandra Mandic

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 3 www.nzma.org.nz/journal contents

CLINICAL CORRESPONDENCE LETTERS 118 146 Intravenous iron infusion and Letter to the editor newer non-dextran formulations Katherine Rich Tim Aung, Justin Coleman, Peter W Davidson, David J Hetzel, 148 Sandy T Aung Letter to the editor: response to Katherine Rich 128 Jennie Connor Granular parakeratosis secondary to benzalkonium OBITUARIES chloride exposure from common 150 household laundry rinse aids Dr William Alexander Fraser Catherine JL Tian, Diana Purvis, Harriet S Cheng 100 YEARS AGO 143 152 Cerebrospinal fluid rhinorrhoea Treatment of Mental following nasal packing for and Nervous Disorders epistaxis: case series from 1921 a single centre Alice Flavell-Birch, Campbell Baguley, Alice Stringer

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 4 www.nzma.org.nz/journal summaries

Parental use of physical punishment in a birth cohort Geraldine FH McLeod, John Horwood, Joseph M Boden, Lianne J Woodward In 2007, as part of wider initiatives to improve child health outcomes, the New Zealand Government introduced important anti-smacking legislation that prohibited the use of physical punishment with children. This analysis examines the extent to which the prevalence of child physical punishment by New Zealand parents may have changed over a 15-year period (2002–2017) spanning before and after this legislation. Findings showed a clear downward trend in the use of child physical punishment, with minor assaults reduced by almost a half (77% to 42%), and severe assaults by two-thirds (12% to 4%), between 2002 and 2017. Parents who were more likely to use physical punishment tended to be younger, caring for more children and in a violent partner relationship. A history of personal mental health problems and concurrent family socioeconomic disadvantage were also influential. Demographic differences in the initiation and maintenance of statins in the first year post ACS in New Zealand: a data linkage study (ANZACS-QI 57) Aravindra Muniandy, Mildred Lee, Corina Grey, Katherine Ferrier, Andrew J Kerr This study captured virtually all New Zealand patients discharged from hospital after a heart attack or unstable angina who had been considered appropriate for invasive heart treatments. A key pillar of management for these patients is ongoing cholesterol lowering with statin medication. We found that in the year after discharge only 60% of people received an optimal supply of daily statin medication. Māori and Pacific people were less likely to receive optimal medication than other ethnic groups. Suboptimal statin medication provision is likely to be a marker of less optimal medical management and of other healthcare more generally. We discuss the opportunities, both in hospital and in the community, to improve survival and quality of life by improving the use of these evidence-based prevention medications. The outcomes of patients with newly diagnosed exudative age-related macular degeneration in Palmerston North Aaron Yap, Adeline Kho, John Ah-Chan The Palmerston North Eye Department has employed innovative strategies to effectively treat the growing number of patients with wet macular degeneration. The information systems in place provide insight into the time it takes for patients with newly diagnosed wet macular degeneration to access sight-preserving treatment and their visual outcome. So far, it has shown that a good proportion of patients experience stabilisation and improvement of their vision on treatment. Furthermore, the time to access treatment has decreased over the past two years.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 5 www.nzma.org.nz/journal summaries

Comparison of physical-activity patterns across large, medium and small urban areas and rural settings in the Otago Region, New Zealand Brittany White, Enrique García Bengoechea, John C Spence, Kirsten J Coppell, Sandra Mandic Less than half of Otago adolescents meet physical-activity guidelines. Higher proportion of males, those participating in school sports and using active transport to school met physical-activity guidelines compared to their counterparts. Overall, adolescents are less physically active on weekends compared to school days. Adolescents from large urban areas spent more time being stationary compared to their rural peers. Continuous encouragement of physical is needed for adolescents across urban and rural settings. Predictors of medicinal cannabis users’ willingness to utilise a new prescription medicinal cannabis scheme in New Zealand Marta Rychert, Chris Wilkins, Karl Parker, Thomas Graydon-Guy The online research survey of 3,634 medicinal cannabis users found majority (66%) were interested in transitioning to the new prescription Medicinal Cannabis Scheme (yet to be fully implemented). Medicinal-cannabis users with higher incomes, who are younger and who administered cannabis via smoking were more likely to express intention to use the scheme. Conversely, medicinal-cannabis users who identified as Māori and those who grew their own cannabis were less likely to intend to engage with the new prescription scheme. The lower intended engagement with the Medicinal Cannabis Scheme by Māori, lower income groups and those who home-grow cannabis may reflect their perceptions of the Medical Cannabis Scheme as restrictive and expensive. As more products become approved under the scheme, engagement with those groups of medicinal cannabis users will be important. Methoxyflurane toxicity: historical determination and lessons for modern patient and occupational exposure Serah J Allison, Paul D Docherty, Dirk Pons, J Geoffrey Chase Anaesthesia is the medical process of temporarily ensuring unconsciousness for a patient, for surgery or other medical procedures. Analgesia is the medical term for ‘pain relief’. This paper starts by investigating historical anaesthetic use (high doses used to create the state of unconsciousness), and then proceeds to discuss the relevance of that historical research in considering the safety of modern analgesic use (lower doses used to relieve pain and sometimes also to cause a small amount of sedation/relaxation without causing unconsciousness). What does abortion law reform mean for primary care practitioners in New Zealand? Emma Macfarlane, Michael Stitely, Helen Paterson Introduction of the Abortion Legislation Act 2020 has significantly changed the way abortion care can be provided in New Zealand, with the potential to improve access, reduce inequities and transform the abortion experience for those people who choose to end their pregnancy. The primary care sector stands to be key players in the provision of first-trimester abortion care. However, the health sector is not yet ready to provide best practice abortion care within the new legislative framework, with issues relating to funding, training and access to medications yet to be resolved.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 6 www.nzma.org.nz/journal summaries

Official Information Act investigation of the Ministry of Health’s process to assess the Southern District Health Board’s readiness to join the National Bowel Screening Programme in 2018 Philip Bagshaw, Paula Goodman, Brian Cox The Ministry of Health (MoH) gave permission for the Southern DHB (SDHB) to start doing bowel cancer screening. But the process used by the MoH is seriously flawed and should be immediately revised before the suitability of other DHB to join the screening programme is assessed. This investigation was done using Official Information Act (OIA) requests for data. This was met with delays and obfuscated by the MoH and only completed when intervention by the ombudsman was anticipated. The OIA Act needs urgent revision in order to achieve its objectives of providing the public and the media with transparency on the functioning of public agencies. Why is it so hard for general practice to provide COVID-19 vaccines? Vanessa Weenink The COVID-19 vaccination campaign is the best chance New Zealand has to protect our citizens' health and will be a crucial activity for the health sector in 2021.1 The procurement strategy for New Zealand started early and has been a dynamic process. However, there has been no evidence of large-scale and detailed operational planning at the level of district health boards (DHBs), drawing political comments that plans are “half-baked” and prompting an Office of the Auditor-General review. To have a successful vaccination program, New Zealand will need to learn from international experience, and it must include the participation of primary care.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 7 www.nzma.org.nz/journal EDITORIAL

Making the Convention on the Rights of Persons with Disabilities real: our word is our bond James Cooney, Michael Daly, Giles Newton-Howes, Kris Gledhill, Sarah Gordon

he Convention on the Rights of Persons law and practice in regards to compliance with Disabilities (CRPD) was adopted by with the CRPD. One of the most significant Tthe United Nations (UN) General Assem- areas of non-compliance is the failure to bly on 13 December 2006. Recognising that implement supported decision-making persons with experience of disability were in any part of New Zealand’s healthcare not being accorded their rights on an equal system, despite being bound to do so for basis, its goal was to clarify how equality of more than a decade. New Zealand is not rights was to be secured.1 As such, it provides alone—no state party reviewed to date has a comprehensive international framework been found fully compliant—but there is to guide states as to what the human rights clear room for improvement, including by and full inclusion and participation of people supporting those who experience psycho- with experience of disability entails, for logical distress to participate in their which political, legal, social and physical medical care. Supported decision-making measures will be needed for implementation. is significantly different from substituted As of January 2021, 181 states plus the decision-making. For the latter, which is European Union have agreed to be bound by the common model, if a person fails in the CRPD. These parties to the Convention one of the domains of capacity—to under- are required to promote, protect and stand, retain or weigh information, or to ensure the full enjoyment of human rights communicate a decision—clinicians take by persons with disabilities, including full over responsibility for decision-making. The equality under the law. New Zealand ratified failure is usually decided by the treating 3 the CRPD on 25 September 2008, and as such clinician. Substituted decision-making is it should guide how we act in the medical disempowering. It often leads to an erro- arena when working with those with neous conflation between difficulties in disability, including persons engaged with making decisions (mental capacity) and the mental health services and those experi- right to make decisions and exercise rights encing mental distress. Implementation of (legal capacity). Supported decision-making, the CRPD is supervised by an expert CRPD with its focus on understanding the will and Committee, which recommends steps to preferences of people with impairments, ensure compliance in practice. Key to this emphasises the full enjoyment of rights and is the transition from substituted deci- so maximises individual empowerment. sion-making to supported decision-making. It also recognises the social elements of disability, by having a focus not on a diag- Does New Zealand adhere to nosis but on how society responds. For the CRPD? example, a patient in psychosocial distress, brought on by low mood and suicidality, Although the CRPD has been widely may struggle to make decisions. Substitute adopted, its good intentions are insuffi- decision-making may consider her to fail to cient for change.2 New Zealand’s reporting ‘weigh in the balance’ and dictate an acute to the CRPD Committee has identified a treatment response, with the shadow of number of limitations and omissions in its

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 8 www.nzma.org.nz/journal EDITORIAL

detention under mental health legislation if a resource that was universally accessible the response is not agreed to. This is vastly and promotes the principles of the CRPD, different from supporting her to understand but that also provides the nuanced infor- her options, consider the social supports mation needed to guide mental health available to her and contemplate a variety services through the changes necessary of treatment decisions for her to choose to be compliant with the CRPD. To enact from. Although examples such as this are true progress, the WHO identified that it obvious in a mental health arena, they are was crucial that these resources are used present in many specialties such as emer- to educate medical professionals to a high gency medicine, old-age care and pediatrics, standard in their understanding of clinical although the content of the support will practice that is CRPD compliant, both at differ by specialty and disability. an early stage in their training and also throughout their specialty training. Despite Can a supported decision-making the promotion of this training material, a theory led to practice change? review of the available online curricula The core ethos of the CRPD is to shift for the majority of medical schools in away from a traditional model of disability, Australasia indicates that it has not been one whereby persons with disabilities incorporated into education programmes. are viewed as ‘objects’ of charity, medical Hence it seems likely that the important treatment and social protection. This concept and practice of supported deci- denies people with disabilities “their full sion-making has yet to be incorporated into and effective participation in society on medical school teaching, which is likely a an equal basis with others”.4 The aim is to necessary step in the development of change move towards a human rights model built in clinical practice. on the understanding that disability is the interaction between the individual and Concluding comments their external environment (composed of Although state parties that have ratified attitudinal and structural barriers). Part the CRPD have demonstrated intent to of the difficulty of switching from models support persons people with disability of disability based on biomedicine to a to make and manage their own medical human rights model is the variable nature of decisions, the majority fail to comply with ‘disability’ that cannot be rigid in definition this commitment. The CPRD Committee’s but rather depends on the predominant ongoing review of state parties show efforts environment. It is this conceptualisation of beyond ratification and legal change are disability that will be new for many doctors needed to develop and roll out supported and may lead to a tension as to how to decision-making into standard clinical re-orient services to meet this need while practice. To this end a focus on medical delivering risk-averse, socially acceptable professional education is required to help care. Legal amendments in other juris- shift from traditional models of care to a dictions (such as many Australian states) human rights model of support. A clear have not led to this change.5 Increasingly, commitment from deans of medical schools it is becoming apparent that education and directors of psychiatric training in in medical professionals’ training and particular is needed to move clinical practice careers may be needed to facilitate changes more closely towards the goal of facilitating in perception in relation to supported supported decision-making. We have a proud decision-making.6 tradition of leading the field in many areas of equality and rights, including proposing the What can we do to encourage appointment of Sir Robert Martin to the CRPD medical school student and trainee Committee, the first person with an intel- lectual disability to be elected to a UN treaty doctor training? body. We need to continue this tradition in In order to facilitate this paradigm shift, our medical training. the World Health Organization (WHO) To provide services in accord with the developed the comprehensive training obligations of the CRPD, there needs to modules QualityRights in 2010. The goal in be significant advances in the teaching of creating such a programme was to provide

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 9 www.nzma.org.nz/journal EDITORIAL

disability rights in medical and psychiatric of medical students need to progress. We education. It is our word as medical profes- have a clear framework, guidelines and sionals that are our bond to our patients. resources to support us to do this—now we, As the requirements of society shift, from as a professional body, need to ensure those a practice model of paternalism to one are utilised to address this currently unmet that encourages autonomy and empowers but critical area of medical education and human rights,7 so too does the teaching development.

Competing interests: Nil. Author information: Dr James Cooney: Senior Psychiatry Registrar, Royal Australian and New Zealand College of Psychiatry. Dr Michael Daly: Senior Psychiatry Registrar, Royal Australian and New Zealand College of Psychiatry Associate Professor Giles Newton-Howes: Department of Psychological Medicine, University of Otago, Wellington. Professor Kris Gledhill: Professor of Law, AUT Law School. Dr Sarah Gordon: Senior Lecturer – Service User Academic, Department of Psychological Medicine, University of Otago, Wellington. Corresponding author: Giles Newton-Howes, Department of Psychological Medicine, University of Otago, Wellington [email protected] URL: www.nzma.org.nz/journal-articles/making-the-convention-on-the-rights-of-persons-with-disabilities-real-our-word-is-our-bond

REFERENCES 1. SZMUKLER, G. 2015. UN sion-making capacity in 6. GORDON, S., ELLIS, P., CRPD: equal recognition hard medical cases. Clinical GALLAGHER, P. & PURDIE, before the law. The Lancet Ethics, 1477750919876248. G. 2014. Service users Psychiatry, 2, e29. 4. Article 1 – Purpose | United teaching the recovery para- 2. STAVERT, J. 2018. Paradigm Nations Enable [Internet]. digm to final year medical Shift or Paradigm Paraly- Available from: https:// students. A New Zealand sis? National Mental Health www.un.org/develop- approach. Health Issues, 15. and Capacity Law and ment/desa/disabilities/ 7. BREEZE, J. 1998. Can Implementing the CRPD convention-on-the-rights- paternalism be justified in Scotland. Laws, 7, 26. of-persons-with-disabilities/ in mental health care? 3. NEWTON-HOWES, G., article-1-purpose.html Journal of Advanced PICKERING, N. & YOUNG, 5. Personal correspon- Nursing, 28, 260-265. G. 2019. Authentic deci- dence, C Ryan

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 10 www.nzma.org.nz/journal EDITORIAL

Challenge for COVID-19 vaccines to protect the New Zealand population Stephen T Chambers, Nigel J Raymond

ew Zealand faces a major challenge vaccine, the relatively good effectiveness this year to use the COVID-19 (SARS- of influenza vaccine and the current NCoV2 virus) vaccination programme public acceptance of social distancing to optimise protective immunity, in order and improved personal hygiene. No one protect our largely non-immune population. strategy alone will deliver all of the health The practical goal is to achieve a high uptake benefits we need. The health professional of vaccination while not leaving any parts of community and the public both have the community unprotected. Never before important roles if our country is to make the has the country had such a programme of most of this opportunity. vaccinating so many people within such a short time frame. COVID-19 vaccine response This challenge in made more acute as New Zealand is well placed to roll out the the New Zealand health system has a long COVID-19 vaccination programme as there and challenging experience of responding are purchase agreements for sufficient Pfizer to seasonal winter increases in respiratory vaccine (PfizerBioNTech) to give two doses to viruses, particularly influenza. The uptake the population. This is a mRNA-based vaccine of publicly funded influenza vaccine has that stimulates cells to make spike protein remained modest, with ongoing equity gaps antigen, but does not incorporate itself into for important high-needs populations (eg, human DNA.2,3 The vaccine vials require an Māori) despite the well-publicised effects on ultra-cold chain for transportation to New the health system. The immunisation rate Zealand where they are held in -80°C freezers. for influenza was reported as 63% overall These facilities are already in place across for 2019 despite ongoing improvement the country. The vaccine can then be stored from previous years and some reductions at 2–8°C for up to five days after thawing for in equity gaps.1 These infections routinely distribution to use at vaccine administration cause unseemly bed shortages and compro- sites around the country. Use of standard mised delivery of healthcare services across pharmaceutical freezer temperatures (-20°C) the country. At present we do not know what for up to two weeks, recently approved in shape future possible winter outbreaks of New Zealand by MedSafe, should consid- COVID-19 infections will take, but COVID-19 erably aid distribution.4 Vials have sufficient added to the normal pattern of winter viral content for 5–6 doses and must be used infections creates a daunting prospect. The within a few hours of opening. The second combination of a winter increase in influenza dose should then be administered at least cases plus COVID-19 cases in similar numbers three weeks (or longer) after the first dose. would undoubtedly lead to a substantial The Pfizer vaccine has been found to be increase morbidity and mortality from these both highly efficacious in adults of all ages infections, because of the overall burden and children as young as 12 years from on the health system and the severity of evidence of large clinical trials, and highly COVID-19 infections. effective in subsequent large-population We now have tools to mitigate the experience.2,5 The vaccine has been found epidemic curve of both COVID-19 and to be highly immunogenic and produces influenza, as well as other viral infections, higher levels of neutralising antibodies than given the effectiveness of the COVID-19 following wild-type COVID-19 infection,

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 11 www.nzma.org.nz/journal EDITORIAL

even in older adults (55–85 years). The for excellent efficacy and safety of the Pfizer most important phase 3 clinical trial was vaccine in subjects aged 12–16 years.5 This conducted across multiple sites in the is welcome news. While children are less United States, Argentina, Brazil, South prone to both asymptomatic and symp- Africa, Germany and Turkey.2 The study tomatic COVID-19 infections and play a lesser enrolled 43,448 people and was conducted role in transmission within families to more in accordance with rigorous FDA standards. vulnerable members, there has been of Failure was defined as being symptomatic lingering concern that this could provide an and having laboratory confirmed COVID-19 under-recognised pathway for transmission. seven days after the second dose was As yet there is little information on how administered. In those with no evidence of long protection is likely to last and how prior infection the vaccine efficacy was 95% effective the Pfizer vaccine will be against (95% CI 90–98%). Vaccine efficacy among COVID-19 variants. Studies using a rapid subgroups (age, sex, race, ethnicity, obesity neutralising antibody assay of serum taken and comorbidities) was consistent with from patients who have had a natural that observed in the overall population. infection or two doses of the Pfizer vaccine There are limited published studies as yet found that the faster spreading COVID-19 on the effectiveness in the ‘real world’, but variants acquired a partial resistance to reports we have so far show impressive the neutralising antibody.11 This was most results in preventing hospitalisation and marked for the B1.351 variant, suggesting death. For example, Public Health England that the vaccine may be less effective against has reported that a single dose of either the this organism. The poorly controlled spread Pfizer or AtraZeneca vaccines cut the risk of of COVID-19 in large populations in Europe, hospital admission by 80% in people over South America and South Asia is almost 6 80 years. In Israel, two doses of the Pfizer certain to allow more variants to emerge. vaccine administered to a cohort of 596,681 In addition, piecemeal or poorly rolled out people reduced hospitalisations by 87% (95% vaccination programmes are likely to create CI 55–100) and severe disease by 95% (95% further evolutionary pressure towards the CI 75–100) compared with unvaccinated selection of escape variants of COVID-19. It 7 controls. Experience in Scotland found a is likely that booster vaccines will be needed vaccine efficacy of 85% (95% CI 76–91) for with new antigens added to the suite of COVID-19 related hospitalisation at 28–34 vaccines against COVID-19 over time.12 days post vaccination.7 More recently, a study showed a similarly high vaccine 5 Community immunity efficacy in children age 12–16 years. The implementation of the public health A growing body of evidence also demon- measures in New Zealand to ‘suppress the strates that fully vaccinated people are less curve’ during the first half of 2020 was likely to have asymptomatic infection and remarkably effective and showed that potentially much less likely to transmit elimination was possible. Most vaccination SARS-CoV-2.8 The few cases of infection programmes aim at disease reduction to following vaccination that have been some ‘tolerable’ level, and it remains to be reported are associated with a shorter mean seen whether prevention of endemic trans- duration of symptoms and often lower viral mission by ‘herd protection’ is possible.5,13–16 load than infection in unvaccinated people.9 Both elimination and control strategies aim For example, a study of 3,950 front-line at protecting the maximum number of indi- health workers, most of whom had been viduals at risk. Following completion of the vaccinated with either the Pfizer and current vaccination programme, is seems Moderna vaccine (another RNA vaccine), unlikely that we will be going straight back found a 90% reduction of symptomatic to ‘normal’, and we may need to navigate a and asymptomatic infections following the pathway to a hopefully low level of endemic second vaccine dose.10 infection. It is doubtful that aiming for an There are clearly areas of uncertainty arbitrary target, such as 70% of the adult that need to be clarified. These include the population being vaccinated, would be suffi- effectiveness of the vaccine in children. cient to control COVID-19, as the degree of There is good evidence from a recent study asymptomatic spread reduction is still not

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 12 www.nzma.org.nz/journal EDITORIAL

established and immunity is rarely uniform Individuals who have been affected, across a population. So we will need to or whose families/whānau have been ensure subpopulations are adequately affected, by COVID-19 have an important protected. To do this we need to specifically role to play in advocating for an effective address some challenging factors. whole of community approach. This could Firstly, it is essential that all sections of include managed isolation and quarantine society are reached with vaccination, espe- (MIQ) staff who have been vaccinated and cially those with comorbidities, including presumably feel a significant measure of Māori and Pacific Island communities, reassurance from the protection this offers, advanced age (eg, rest-homes), and particu- healthy people who have been affected larly if their community may be more likely by suffering a serious COVID-19 illness, or to be exposed (eg, South Auckland). It will be those who have been affected by the ‘long a huge operational challenge to reach most COVID syndrome’ that compromises their of New Zealand’s population with COVID-19 enjoyment of life in the long term. vaccination during the year. The national COVID-19 immunisation register is a vital Benefits of an effective vaccination component of this strategy and will be programme needed for future vaccines delivery. The vaccination programme has Secondly there has been a major effort important implications. Many people will launched to overcome vaccine hesitancy. be keen to break out from the restrictions It is imperative that healthcare profes- of the infection prevention and control sionals take responsibility to provide (IPC) measures, such as social distancing independent, high-quality and clear infor- and use of masks on public transport. These mation to address legitimate concerns and measures have undoubtedly been effective counter misinformation. Health staff should in reducing the rate of transmission of be well informed so that the vaccination COVID-19 and also influenza and other programme can be discussed during hospital severe respiratory infections. The benefits admission and outpatient clinics, as well as of these measures should be pursued in the general practitioner visits. The importance short term until there is a reliably high level of taking into account behavioural and of community immunity from COVID-19, social drivers of vaccine uptake, including and then they should be seen as part of the during healthcare interactions, has also normal response to the annual winter respi- been emphasized by the World Health ratory virus epidemics.19 Organisation (WHO) in their BeSD model New Zealand has opened a ‘bubble’ with in endeavouring to achieve high uptake of Australia, and other countries will soon be COVID-19 vaccines (Figure 1).17,18 added to this. But there will be an ongoing

Figure 1: The BeSD of Covid-19 model.

Source: Based on the “increasing vaccination” model (Brewer et al., 2017).18

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 13 www.nzma.org.nz/journal EDITORIAL

need for the MIQ border protections. These vector) does not seem to be a problem with have worked well and limited community the Pfizer vaccine. The specific clotting incursions to outbreaks, which have been problems have produced very serious, if contained by intensive work by Public rare, clinical effects, including cavernous Health Units (PHUs). Extensive vaccination sinus thrombosis and deaths, which are of the staff manning these facilities will likely to be clinically recognised within offer a further layer of protection against OECD medical systems. Given the intense an importation, and high uptake of vacci- scrutiny engendered by the publicity of nation in communities who later become clotting problems with the other vaccines, it contacts of an infected traveller would seems very unlikely that these problems are greatly help this situation by lowering the associated with the Pfizer vaccine. Moni- effective transmission reproductive number toring of the primary outcomes from the (Ro) and limiting community spread to be trials will continue until August 2021, while containable. This would greatly lower the monitoring of the secondary outcomes will burden on PHU to contact trace and shut continue until January 2023.22 down outbreaks and minimise the need for lockdowns. Leadership from healthcare providers Vaccine safety evidence and New Zealand owes a great debt of grat- monitoring (high) itude to the leadership provided by the A comprehensive discussion of vaccine government and health leaders who have safety is beyond the scope of this paper. communicated so effectively with the public. However, the evidence from controlled The next hurdle is to roll out the vaccine trials has found that the Pfizer vaccine is in a coordinated and equitable way. There extremely safe.10,20 The most common side are already several effective vehicles for effects include mild to moderate pain at providing information on the ongoing the injection site in about 80% of subjects, evolution and response to the COVID-19 fatigue in about 60% and headache in epidemic. It is important that healthcare about 50% of subjects.21 All were short providers support the strategy by helping lived. Reports of serious side effects, such to build confidence in the outcomes as we as allergic reactions, have been very rare, mediate between the overall policy and the and no long-term complications have been concerns of our patients. Key ingredients confirmed. The clotting problems that have needed in this roll-out are clarity of purpose, been associated with the Oxford Astra- knowledge of the benefits and safety of the Zeneca and possibly with Johnson & Johnson vaccine and optimism that a scientific and vaccines (both of which use an adenovirus humane response will be effective.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 14 www.nzma.org.nz/journal EDITORIAL

Competing interests: Nil. Acknowledgements: The authors thank Dr Nikki Turner and Prof David Murdoch, for reviewing and comments on the manuscript. Author information: Nigel J Raymond: Infectious Diseases & General Physician, Infection Service, Capital & Coast District Health Board; Department of Medicine, University of Otago, Wellington (Honorary CSL). Stephen T Chambers: Department of Pathology, University of Otago, Christchurch. Infectious Diseases Physician Christchurch Hospital. Corresponding author: Dr NJ Raymond, Infection Service, Lv6 GNB, Wellington Hospital, Private Bag, Wellington, +64-4-385-5999 [email protected] URL: www.nzma.org.nz/journal-articles/challenge-for-covid-19-vaccines-to-protect-the-new-zealand-population-open-access

REFERENCES 1. Influenza. NZ Ministry reports 100% vaccine of a mRNA SARS-CoV-2 of Health. https://www. efficacy in children aged 12 vaccine is associated with health.govt.nz/our-work/ to 15. BMJ 2021;373:n881. lower nasopharyngeal immunisation-hand- http://dx.doi.org/10.1136/ viral load among nursing book-2020/11-influenza bmj.n881 home residents with accessed 10 April 2021. 6. Lacobucci G. Covid-19: asymptomatic COVID-19. 2. Polack FP, Thomas SJ, Single dose of Pfizer Clin Infect Dis 2021 26 Mar Kitchin N, Absalon J, and Oxford vaccines (accepted; preprint) https:// Gurtman A, Lockhart cuts risk of hospital doi.org/10.1093/cid/ciab263 S, Perez JL, et al. Safety admission by 80% in over 10. Thompson MG, Burgess and Efficacy of the 80s, data suggest. BMJ JL, Naleway AL, Tyner BNT162b2 mRNA Covid- 2021;372:n612 http://dx.doi. HL, Yoon SK, Meece K, 19 Vaccine. N Eng J Med org/10.1136/bmj.n612 Olsho LEW, et al. Interim 2020; December 16, 1-13. 7. Dagan N, Barda N, Kepten estimates of vaccine https://doi.org/10.1056/ E, Miron O, Perchik S, Katz effectiveness of BNT162b2 NEJMoa2034577 MA, Hernán MA, et al. and mRNA-1273 COVID-19 3. Lamb YN. BNT162b2 BNT162b2 mRNA Covid-19 vaccines in preventing mRNA COVID19 vaccine: vaccine in a nationwide SARS-CoV-2 infection first approval. Drugs mass vaccination setting. among health care person- (2021) 81:495–501. N Engl J Med 24 Feb 2021. nel, first responders, and https://doi.org/10.1007/ https://doi.org/10.1056/ other essential and front- s40265-021-01480-7 NEJMoa2101765 line workers - eight U.S. locations, December 2020– 4. Coronavirus (COVID-19) 8. Edelstein S, Tannous S, March 2021. Morb Mort update: FDA allows more Jacobs MT, Ben-Amram H, Weekly Rep MMWR / April flexible storage, trans- Zarka S. BNT 13b2 Pfizer 2, 2021;70(13):495-500. portation conditions for vaccine protects against Pfizer-BioNTech COVID- SARS-CoV-2 respiratory 11. Planas D, Bruel T, Grzelak 19 vaccine. February mucosal colonization even L, Guivel-Benhassine 25, 2021. https://www. after prolonged exposure to F, Staropoli I, Porrot F, fda.gov/news-events/ positive family members. Planchais C, et al. Sensitivi- press-announcements/ J Hosp Infect (preprint, ty of infectious SARS-CoV-2 coronavirus-covid-19-up- accepted 24 March 2021). B.1.1.7 and B.1.351 variants date-fda-allows-more-flex- https://doi.org/10.1016/j. to neutralizing antibod- ible-storage-transporta- jhin.2021.03.023 ies. Nature Med 2021 https://doi.org/10.1038/ tion-conditions-pfizer 9. McEllistrem MC, Clancy s41591-021-01318-5 accessed 15 April 2021. CJ, Buehrle DJ, Lucas A, 5. Mahase E. Covid-19: Pfizer Decker BK. Single dose 12. Hodgson SH, Mansatta

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 15 www.nzma.org.nz/journal EDITORIAL

K, Mallett G, Harris V, 16. Mallory ML, Lindesmith LC, 19. Moore S, Hill EM, Tildesley Emary KRW, Pollard Baric RS. Vaccination-in- MJ, Dyson L, Keeling MJ. AJ. What defines an duced herd immunity: Vaccination and non-phar- efficacious COVID-19 successes and challenges. maceutical interventions vaccine? A review of J Allergy Clin Immunol. for COVID-19: a mathemati- the challenges assessing 2018 July ; 142(1): 64–66. cal modelling study. Lancet the clinical efficacy of https://doi.org/10.1016/j. Infect Dis, March 18, 2021 vaccines against SARS- jaci.2018.05.007 https://doi.org/10.1016/ CoV-2. Lancet Infect Dis 17. Data for action: achieving S1473-3099(21)00143-2 2021;21: e26–35. Published high uptake of COVID- 20. Medsafe. Adverse events online October 27, 2020. 19 vaccines. Gathering following immunisation https://doi.org/10.1016/ and using data on the with COVID-19 vaccines: S1473-3099(20)30773-8 behavioural and social Safety Report #1. 7 13. Fine P, Eames K, Heymann drivers of vaccination. A April 2021. https://www. DL. ‘‘Herd immunity’’: a guidebook for immuni- medsafe.govt.nz/COVID- rough guide. Clin Infect Dis zation programmes and 19/safety-report-1.asp 2011;52(7):911–916. https:// implementing partners. Accessed 15 April 2021. doi.org/10.1093/cid/cir007 Interim Guidance. W.H.O. 21. COMIRNATY™ COVID-19 14. Rashida H, Khanda- 3 February 2021. https:// Vaccine. New Zealand kera G, and Booy R. apps.who.int/iris/rest/ Datasheet; Medsafe. Vaccination and herd bitstreams/1331778/ https://www.medsafe. immunity: what more retrieve. Accessed govt.nz/profs/Data- do we know? Curr Opin 25 April 2021 sheet/c/comirnatyinj.pdf Infect Dis 2012, 25:243–249. 18. Brewer NT, Chapman Accessed 20 April 2021. https://doi.org/10.1097/ GB, Rothman AJ, Leask 22. Clinical trial number QCO.0b013e328352f727 J, Kempe A. Increasing NCT04368728 for 15. Zachreson C, Chang SL, vaccination: putting “NCT04368728: Study to Cliff OM, Prokopenko M. psychological science Describe the Safety, Tolera- How will mass-vaccination into action. Psychol bility, Immunogenicity, and change COVID-19 lockdown Sci Pub Interest 2017; Efficacy of RNA Vaccine requirements in Australia? 18(3) 149–207. https:// Candidates Against COVID- April 2, 2021 https://arxiv. doi/10.1177/1529 19 in Healthy Individuals” org/abs/2103.07061 100618760521 at ClinicalTrials.gov

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 16 www.nzma.org.nz/journal ARTICLE

Parental use of physical punishment in a birth cohort Geraldine FH McLeod, John Horwood, Joseph M Boden, Lianne J Woodward

ABSTRACT AIM: To document the prevalence of child physical punishment by parents and associated predictors in the Christchurch Health and Development Study (CHDS) birth cohort over a 15-year period. METHOD: A cohort of 1,265 CHDS individuals were followed from birth (1977) to age 40 years. At ages 25 (n=155), 30 (n=337), 35 (n=585) and 40 years (n=636), the cohort members with dependent children (<16 years of age) were interviewed about their use of child physical punishment in the past 12 months using the Parent-Child Conflict Tactics Scale. Parent, child and family predictors were also examined. RESULTS: The most common forms of physical punishment were smacking on bottom and slapping on hand, arm or leg. Rates of all forms of physical punishment declined with age, ranging from 77% reporting any physical punishment at age 25 to 42% at age 40. In multivariable models, significant predictors included parental age, numbers/ages of children in the household, childhood family socioeconomic status, parental history of adolescent mental health problems and concurrent intimate partner violence. CONCLUSION: Use of physical punishment remains a relatively common form of child discipline despite the 2007 anti-smacking legislation and reduced public tolerance for physical violence towards children. Implications for prevention/intervention are discussed.

ew Zealand has very high rates of punish their child.13,14 In addition, national child maltreatment compared to oth- child maltreatment statistics show that er developed countries.1–3 Research many children are still being maltreated by N 2 shows that childhood physical punishment/ their caregivers. Oranga Tamariki (New physical maltreatment has detrimental Zealand Ministry for Children) figures in effects on an individual’s later adult partner the 12 months to 31 December 2019 showed relationships,4 mental and physical health that there were 85,000 reports of concern (including suicidal behaviours),5,6,7 substance regarding 61,300 individual children.15 It use,8,9 educational achievement10 and crimi- has been estimated that almost one in four nal activity.11 Further, these impacts increase New Zealand children have been subject with the severity of physical maltreatment, to at least one report of abuse or neglect to but adverse effects are observable even authorities by age 17.2 Of these, around one at relatively minor levels of physical pun- in ten were subjected to substantiated abuse ishment.3,12 Societal approval of physical or neglect with a considerable proportion punishment has been declining over more of these complaints relating to parental recent decades.13,14 In 2007, the New Zealand use of physical punishment/assault.2 Using Government introduced “anti-smacking” these figures, the incidence of notifications legislation, which prohibited the use of child to child protective services in New Zealand physical punishment.3,13 was higher than the incidence of childhood Despite this legislation, a 2013 report medicated asthma and similar to the preva- 2 estimated that 40% of New Zealand adults lence of obesity. still agreed that there are certain circum- These findings tend to suggest that, despite stances when parents may physically the law change in 2007, violence towards

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 17 www.nzma.org.nz/journal ARTICLE

children is still relatively common.2,13,14 teacher report; and official record data.22,23 However, none of these analyses have All phases of the study have been subject to considered rates of physical punishment ethical approval by the Regional Health and over time within the New Zealand popu- Disabilities Ethics Committee. lation, and in particular how these rates may The current analysis is based on a have changed, and hopefully reduced, as sub-sample of 763 CHDS participants who a result of legislation and increased public were parenting a dependent child and had awareness. To help inform further public data recorded on their child-management health efforts, it is important to continue to practices at one or more assessments from monitor rates of child physical punishment/ age 25–40 years. The observed samples at abuse among New Zealand parents, ideally each assessment were: age 25 (n=155), 30 using a common measurement approach. (n=337), 35 (n=585) and 40 years (n=636). To Also important is the need to identify be included in the analysis, the participant modifiable predictors of child physical had to have at least one resident dependent punishment that can be targeted by public child under the age of 16 years. health and clinical interventions to reduce Parental use of physical rates of parental violence towards their children.13 A number of predictors of child punishment When participants were age 25, 30, 35 physical punishment have been identified and 40 years old, those who were parenting by previous research.16 These predictors a dependent child were questioned about mostly fall into four broad categories: family their use of different forms of physical and social environment factors; parental punishment/abuse using the physical assault factors; child characteristics; and socio-cul- sub-scale of the Parent-Child Conflict Tactics tural influences.1,16–21 Scale (CTS-PC).24 This 12-item subscale An analysis of the physical discipline assesses the extent to which parents had practices of 155 individuals within the used each of the physical punishment Christchurch Health and Development methods to discipline their child/children Study (CHDS) cohort who had become over the previous 12 months. Parents were parents before the age of 25 years found that questioned separately about their own 77% had physically punished and around behaviour and that of their partner (if 12% had severely physically assaulted a applicable). Items ranged in severity from 1 dependent child in the past year. However, minor to very severe assault. Minor assault this analysis was conducted prior to the items included: Smacked your child on 2007 law change and was based largely on the bottom with your bare hand; Slapped a subset of younger and generally higher your child on the hand, arm or leg; Pinched risk parents. The CHDS cohort has now been your child; Shook your child; Hit your assessed on three further occasions up to child on the bottom with something like a age 40: in 2007 (age 30), 2012 (age 35) and belt, hairbrush, a stick or some other hard 2017 (age 40). Thus, the aim of this study object. Severe assault items included: Hit 1 was to extend the 2002 study and report your child on some other part of the body rates and predictors of the use of physical besides the bottom with something like a punishment by cohort members towards belt, hairbrush, a stick or some other hard their children over time. object; Slapped your child on the face, head or ears; Hit your child with a fist or kicked Methods her/him hard; Threw or knocked your child Participants down. Very severe assault items included: Participants were members of the Christ- Grabbed your child around the neck and church Health and Development Study choked her/him; Hit your child over and (CHDS) birth cohort. The CHDS is a longitu- over as hard as you could; Burned or scalded dinal study of 1,265 children (630 females) your child on purpose. One item from the born in Christchurch over a four-month original CTS-PC, “threatened with a knife or period during 1977. This cohort has been gun,” was excluded due to its very low base studied regularly from birth to age 40 using rate in the cohort. Items were scored on a a combination of: interviews with parents 7-point scale from never (0) to 20+ times (6). and participants; standardised testing; For the purposes of the present analysis,

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 18 www.nzma.org.nz/journal ARTICLE

parents were classified at each assessment The primary outcome for the multi- as having used a particular form of physical variable analysis was a dichotomous punishment if they reported having used measure of any use of physical punishment it on any of their children. The CTS-PC is a at each age. Random effects logistic widely used and valid measure of parental regression methods were used to model the aggression towards their child/children.24,25 repeated measures of physical punishment Antecedent and concurrent over time as a function of age, household composition, individual characteristics predictors of child physical and other factors. An initial model was punishment fitted to explore the extent to which age-re- A range of candidate predictor variables lated variations in the use of physical were considered from the database as punishment could be explained by age-re- potential indicators of increased risk of lated variations in family composition, parental use of physical punishment. individual characteristics or childhood-back- These were selected on the basis of ground characteristics associated with the previous research and theory: timing of parenthood. This analysis was a. Measures of family structure/ then extended to consider the full set of composition: presence of a cohab- potential predictors. Predictor variables iting partner, numbers and ages of meeting a p≤0.30 criterion were tested in dependent children. a full model, which was then refined using b. Measures of individual characteristics: forward and backward selection methods sex, ethnicity, educational attainment, to identify a consistent and parsimonious adolescent antisocial behaviour set of predictors. A sensitivity analysis (conduct/oppositional disorders), was conducted using negative binomial adolescent mental health disorders regression to model the frequency (number (depression, anxiety, suicidal of instances) of physical punishment ideation). summed over all scale items, with frequency estimated at the midpoint of the relevant c. Measures of childhood family circum- response category for each item. stances: parental education, family socioeconomic status, family insta- Assuming an overall base rate of physical bility, exposure to child physical punishment in the region of 40%–50%, the or sexual abuse, family violence, study sample had 80% power at α=.05 to parental adjustment problems, detect odds ratios with a dichotomous risk parental bonding. factor in the range 1.5 to 2.3, depending on the base rate of exposure in the non-pun- d. Measures of concurrent family ishment group (5% to 50%), or a correlation context at the time of assessment: in excess of .12 with a continuous risk family income, welfare dependence, factor.26 Power is likely to be enhanced by parental mental health/substance use the repeated measures nature of the data. problems, adverse family life events, This suggests the study has adequate power intimate partner violence. to detect small to moderate effect size These measures are described in greater associations. detail in the Appendix. Statistical methods Results Tabular analyses summarised the % (n) of the items and summary measures Characteristics of parenting of physical punishment/assault at each samples assessment. Chi-square and repeated Appendix Table 1 summarises the measures analysis of variance were used to demographic and childhood-background examine variations in the sample charac- characteristics of the parenting samples teristics of the parenting samples over time. assessed at each of the four study waves Tests for age-related trends in the observed from ages 25–40 years. Consistent with rates/frequency of physical punishment general population demographics, women were conducted using random effects and those of Māori or Tagata Pasifika logistic or negative binomial regression. ethnicity were more likely to have made an earlier transition to parenthood. The earlier

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 19 www.nzma.org.nz/journal ARTICLE

parenting samples were less likely to have punishing their child/children. Extension of attained degree-level qualifications, more this analysis to incorporate use of physical likely to be sole parenting and had fewer punishment by a partner (if present) as well dependent children than later parenting as the responding parent revealed only very samples. They were also more likely to have small increases in the prevalence of all types been raised in families characterised by of physical punishment across households, socioeconomic disadvantage, family insta- though there was an increase in the median bility and exposure to child abuse/family frequency of reported incidents of physical violence. punishment when partner data were Parental use of physical included (Table 3). punishment Predictors of child physical Table 1 shows the reported rates of child punishment physical punishment in the past 12 months Consideration of a wide range of potential at each parental age (25 to 40 years) by predictors in a multivariable model (see type/severity of punishment. The most section Methods) predicting any use of common forms of physical punishment physical punishment identified the following were smacking on bottom and slapping on five factors as independent predictors. The hand, arm or leg. Only a very small minority two strongest predictors were parental of parents reported using severe or very age at assessment (p<.001), as reflected in severe forms of physical assault. There was the results above, and the numbers/ages a downward trend in the use of all forms of dependent children in the household of physical punishment with age. This is (p<.001). In particular, use of physical reflected in the overall rate of use of any punishment increased with the number of form of physical punishment. At age 25, over 2–4-year-old children (OR=4.6, 95%CI 3.4–6.2, three quarters of parents reported using p<.001) and the number of children aged physical punishment on their child/children, 5–10 years being cared for (OR=2.5, 95%CI with this rate declining to just over 40% 2.0–3.1, p<.001), but not with the number of parents at age 40. A similar downward of children aged <2 (OR=1.3, 95%CI 0.9–1.8) trend was observed in parental reports of or 11–15 years (OR=0.9, 95%CI 0.7–1.2). In the frequency of physical punishment use. addition, use of physical punishment was Specifically, among those parents using higher among parents who were themselves physical punishment, the median number of raised in lower-socioeconomic-status house- instances declined from 16 at age 25 to 4 at holds (OR=1.4, 95%CI 1.1–1.8, p=.01); who age 40. had experienced mental health problems The observed age-related decline in the (depression, anxiety, suicidal ideation) rate of physical punishment could not be during adolescence (OR=1.9, 95%CI 1.3–2.7, explained by age-related differences in p<.001); or who were in romantic relation- family composition, parental demographics ships characterised by intimate-partner or childhood-background characteristics physical violence (OR=2.3, 95%CI 1.5–3.5, associated with the timing of parenthood. p<.001). Reanalysis of the data using the After being adjusted for these differences, estimated number of instances of physical the overall rates of physical punishment punishment reported at each age as a proxy were negligibly different from the observed for severity of physical punishment/abuse rates at each age (marginal adjusted rates: identified the same set of predictors. 76.6% at age 25; 58.2% at age 30; 47.5% at To illustrate the level of prediction age 35; 41.7% at age 40). obtained when the identified predictors Stratifying by sex of parent showed strong were considered in combination, a gender similarities in the prevalence and composite risk score was constructed by frequency of physical punishment (Table summing five dichotomous indicators at 2). Younger female parents reported the each age. These included: (i) the participant highest overall rate of physical punishment, was raised in a low SES household; (ii) the with 82% reporting that they had physi- participant experienced mental health cally punished their child/children at age problems in adolescence; (iii) the participant 25. Whereas, at age 40, 39% of mothers was involved in a relationship characterised and 47% of fathers reported physically by intimate-partner physical violence in the

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 20 www.nzma.org.nz/journal ARTICLE

Table 1: Use of child physical punishment in the past 12 months for all parenting respondents at ages 25–40 years.

Item Age 25 Age 30 Age 35 Age 40 pb (N=155) (N=337) (N=585) (N=636) Assessment year 2002 2007 2012 2017

% (n) Minor assault 77.4 (120) 58.5 (197) 46.2 (270) 41.7 (265) <.001 Smacked child on bottom with 62.6 (97) 46.7 (155) 36.6 (214) 32.4 (206) bare hand

Slapped child on hand, arm or leg 66.5 (103) 43.4 (145) 28.6 (167) 24.4 (155)

Pinched child -a 5.1 (17) 1.7 (10) 1.6 (10)

Hit child on bottom with belt, hair- 11.0 (17) 3.3 (11) 3.6 (21) 3.5 (22) brush stick or other hard object

Shook child 6.5 (10) 3.3 (11) 1.2 (7) 1.6 (10)

% (n) Severe assault 12.3 (19) 4.5 (15) 4.8 (28) 3.9 (25) .005 Slapped child on face, head or ears 10.3 (16) 3.0 (10) 2.6 (15) 2.5 (16)

Hit child on body besides bottom 3.9 (6) 0.9 (3) 1.0 (6) 0.3 (2) with belt, hairbrush, stick or other hard object

Threw or knocked child down 1.9 (3) 0.9 (3) 0.9 (5) 0.8 (5)

Hit child with fist or kicked them 1.3 (2) 0.6 (2) 0.7 (4) 0.8 (5)

% (n) Very severe assault 2.6 (4) 1.2 (4) 0.7 (4) 0.3 (2) .007 Choked or grabbed child around 1.3 (2) 0.3 (1) 0.2 (1) 0.3 (2) the neck

Hit child over and over as hard as could 2.6 (4) 0.9 (3) 0.5 (3) 0.0 (0)

Burned or scalded child on purpose 0.7 (1) 0.0 (0) 0.0 (0) 0.0 (0)

% (n) Any form of physical pun- 77.4 (120) 58.5 (197) 47.4 (277) 42.5 (270) <.001 ishment

Median (IQR) number of instanc- 16 10 5 4 <.001 es (amongst those using physical (8–40) (4–27) (2–15) (2–10) punishment)

a Not assessed at age 25. b Tests for trend in major categories of physical punishment.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 21 www.nzma.org.nz/journal ARTICLE

Table 2: Use of child physical punishment in the past 12 months (age 25–40 years) by sex of parent.

Measure Age 25 Age 30 Age 35 Age 40 p Assessment year 2002 2007 2012 2017

Females (N=110) (N=204) (N=333) (N=343) % (n) Minor assault 81.8 (90) 60.3 (123) 43.5 (145) 38.5 (132) <.001

% (n) Severe assault 13.6 (15) 3.9 (8) 4.8 (16) 3.8 (13) .006

% (n) Very severe assault 2.7 (3) 1.0 (2) 0.6 (2) 0.0 (0) .01

% (n) Any form of physical 81.8 (90) 60.3 (123) 45.1 (150) 38.8 (133) <.001 punishment

Median (IQR) number of instances 16 14 4 4 <.001 (amongst those using physical (8-40) (4-27) (2-15) (2-8) punishment)

Males (N=45) (N=133) (N=252) (N=293) % (n) Minor assault 66.7 (30) 54.9 (73) 49.6 (125) 45.4 (133) .004

% (n) Severe assault 8.9 (4) 5.3 (7) 4.8 (12) 4.1 (12) .28

% (n) Very severe assault 2.2 (1) 1.5 (2) 0.8 (2) 0.7 (2) .26

% (n) Any form of physical 66.7 (30) 55.6 (74) 50.4 (127) 46.8 (137) .008 punishment

Median (IQR) number of instanc- 15 8 6 4 <.001 es (among those using physical (4–50) (4–28) (2–15) (2–10) punishment)

Table 3: Use of child physical punishment in the past 12 months for all parenting respondents and their partners at ages 25–40 years.

Measure Age 25 Age 30 Age 35 Age 40 p (N=155) (N=337) (N=585) (N=636) % (n) Minor assault 81.9 (127) 62.6 (211) 48.6 (284) 44.3 (282) <.001

% (n) Severe assault 15.5 (24) 6.5 (22) 5.0 (29) 5.0 (32) <.001

% (n) Very severe assault 5.2 (8) 1.2 (4) 0.9 (5) 0.3 (2) <.001

% (n) Any form of physical 81.9 (127) 62.9 (212) 49.9 (292) 45.4 (289) <.001 punishment

Median (IQR) number of instances 23 16 8 6 <.001 (among households using physical (8–56) (6–46) (4–24) (2–16) punishment)

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 22 www.nzma.org.nz/journal ARTICLE

past 12 months; (iv) the family had one or often they used it. This trend could not be more dependent children aged 2–4 years; explained by differences in family structure, (v) the family had one or more dependent personal characteristics or the family back- children aged 5–10 years. Table 4 describes grounds of parents at each age assessment. the relationship between this risk score Other possible explanations include: and the likelihood of a parent using any increasing maturity of the parenting sample form of physical punishment in the past 12 over time (less reactive, more experienced, months at each age. Results show strong older parents); a cultural shift towards monotonic associations, with parental use the unacceptability of violence towards of physical punishment increasing rapidly children over the period of the study; and with the number of risk factors present. the law change in 2007, which prohibited These trends were most marked at older physical punishment and violence towards ages. For example, at age 40, parents with children. Given the nature of its design, four or more of the above indicators were it is not possible for the current study to approximately 14 times more likely to have distinguish between these explanations. used physical punishment than those with a However, it does not seem unreasonable risk score of zero (80.0% vs 5.9%). Extension to conjecture that all three processes are of the model to test for age by risk factor likely to have played a role. For example, interactions showed no evidence that the there is good evidence from the wider effects of predictors varied with the age of literature suggesting that older parents use the parent. more effective management strategies for child misbehaviour.20 In addition, over the Discussion last two decades, there has been increasing recognition across society of the potential This study reports data gathered over harms associated with physical punishment a 15-year period (2002–2017) spanning and violence towards children and, in turn, the interval from five years prior to intro- a decline in the perceived acceptability of duction of the anti-smacking legislation physical punishment.13,14 in 2007 to 10 years after for a cohort of parenting-age New Zealand adults. Results Nonetheless, despite the downward trend demonstrate a clear downward trend in in the use of physical punishment within parental reported use of child physical this large cohort, it is clear that physical punishment over this period, both in terms punishment remains a fairly common form of the proportion of parents still relying of child discipline. Even at age 40, over on physical punishment, but also how 40% of parents reported using physical punishment on their child/children in the

Table 4: Rates (%) of physical punishment in the past 12 months by risk factor score and age of parent (25–40 years).

Number of risk 25 years 30 years 35 years 40 years factors 0 60.0 26.8 6.2 5.9 (n=5) (n=41) (n=65) (n=57)

1 58.5 55.2 37.6 33.6 (n=41) (n=105) (n=189) (n=217)

2 84.0 59.8 56.0 48.2 (n=50) (n=87) (n=207) (n=247)

3 83.3 72.2 63.0 58.4 (n=48) (n=79) (n=92) (n=101)

4+ 100.0 76.0 87.5 80.0 (n=11) (n=25) (n=32) (n=20)

p <.001 <.001 <.001 <.001

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 23 www.nzma.org.nz/journal ARTICLE

previous 12 months. Similar findings were responsive and strengths-based efforts to reported in a 2013 study that showed 40% support vulnerable groups of parents, such of New Zealand adults still agreed that there as those with high-risk personal back- were certain circumstances when parents grounds who are parenting very young may physically punish their child.13,14 These children, alongside (b) wider public health findings suggest that, despite both changing programmes to promote the use of alter- perceptions towards physical punishment native and more effective strategies for and violence toward children and the dealing with child behavioural issues in the 2007 legislation, a substantial minority of context of stressful family situations.20 New Zealanders may still view physical This analysis uses data from a well-studied punishment as an acceptable form of child New Zealand-based representative longitu- discipline. However, on a positive note, dinal birth cohort. Cohort members have these parents do appear to be engaging less been studied prospectively on 24 occasions frequently in these problematic parenting to age 40, with extensive assessment of behaviours, although this may not neces- potential risk/protective factors. However, sarily mitigate the extent of child harm. the study also needs to be considered in light Relatively few predictors of child physical of a number of limitations. In particular, punishment were identified in the analysis. due to social desirability bias, the chosen The strong association found with parental measure of child physical punishment (the intimate-partner physical violence is CTS-PC) may underestimate rates. This study consistent with the now well-established reports rates of physical punishment by link between wider family violence and parents who were aged 25 in 2002 (before child physical punishment/abuse,27 rein- the 2007 legislation to prohibit physical forcing the need for continued public health punishment), so it is unclear what rates of interventions to reduce family violence. physical punishment of children would be in The more modest links with socioeconomic studies of contemporary young parents. background and adolescent mental health Overall, the findings suggest that, while 18,28 are also consistent with existing research. both the number of parents using physical However, by far the strongest predictor punishment and the frequency of physical was the number of 2–10 year olds in the violence have decreased, a large number of household and, in particular, the number New Zealand parents are continuing to use of 2–4 year olds. This is consistent with physical punishment when disciplining their developmental data showing that children children. This is despite the New Zealand between the ages of 2–4 years can pose Government legislating against physical particular challenges for parents, given their punishment of children in 2007 and public rapidly increasing verbal and motor skills health efforts to increase awareness of but limited ability to regulate their emotions the potential harms of physical violence and behaviours. This developmental period towards children. There remains a need is often referred to as the “terrible twos” for continued public education on reducing or “terrible threes,” given that opposi- physical violence; for providing alternative tional behaviour and uncontrolled temper strategies to manage child behaviour; and tantrums are common and harder to deal for ongoing monitoring of parental use with, especially when a parent may be of physical punishment against changing stressed, tired or lack the resources to societal tolerance of violence toward 20,21 cope. Collectively, these findings suggest children. the need for (a) individualised, culturally

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 24 www.nzma.org.nz/journal ARTICLE

Appendix

Appendix Figure 1: Description of measures. The following measures were considered as potential candidate predictors of parental use of child physical punishment: • Measures of family structure/composition • Partnership status: Participants were classified as either sole parents or cohabiting with a resident partner based on reported partnership status at each assessment. • Numbers/ages of children: Variability in the age composition of dependent children (<16 years) in the household was represented in the analysis by counts of the numbers of children in the age ranges <2years, 2–4 years, 5–10 years and 11–15 years at each assessment. • Measures of individual characteristics • Sex: The biological sex of the participant was recorded at their birth. • Ethnicity: At birth, each cohort members’ ethnicity was recorded based on parent-reported ethnicity. • Educational attainment: This was classified into four levels based on the highest educational qualification attained by age 25: no formal qualifications; high school qualifications; tertiary qualifications below degree level; university degree or equivalent. • Adolescent adjustment (14–16 years): At ages 15 and 16 years, participants and their parents were interviewed about aspects of the young person’s mental health/ adjustment over the preceding 12 months. These interviews combined a range of standardised assessment instruments, including components of the relevant version (self- or parent-report) of the Diagnostic Interview Schedule for Children (DISC),29 together with custom-written survey items to assess DSM-III-R symptom criteria for a range of mental disorders as well as the occurrence of suicidal behaviours (suicidal ideation or attempt) in each interview period.30 For the purposes of the present analysis these data were combined over the two assessments to derive two dichotomous measures: (a) adolescent mental health disorders: whether the young person met diagnostic criteria on the basis of either parent- or self-report for a major depressive episode or an anxiety disorder (generalised anxiety disorder, overanxious disorder, phobias), or they were reported to have experienced suicidal behaviour at any time during the period from age 14–16 years; (b) adolescent conduct/oppositional disorders: whether the young person met diagnostic criteria on the basis of either parent- or self-report for a diagnosis of conduct or oppositional defiant disorder from age 14–16 years. • Measures of childhood family circumstances • Parental education: This was classified into three levels based on the highest level of educational attainment reported for either parent at the time of the participant’s birth (no formal qualifications, high school qualifications, tertiary qualifications). • Family socioeconomic status (birth): This was assessed on the basis of paternal occupation at the time of birth using the Elley and Irving scale of socioeconomic status,31 classified into three levels: professional/managerial; clerical/technical/ skilled; semiskilled/unskilled/unemployed. • Changes of parents (0–16 years): Childhood family instability was assessed on the basis of a count of the number of changes of parents experienced by the child from birth to age 16 years. Parental changes included separation/divorce, recon- ciliation, remarriage/cohabitation, fostering and any other changes of custodial parents.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 25 www.nzma.org.nz/journal ARTICLE

• Childhood sexual/physical abuse (<16 years): At ages 18 and 21, participants were questioned about their experience of physical or sexual abuse prior to age 16 years. Physical abuse was assessed based on participant reports of the extent to which their parents used physical punishment /maltreatment during their childhood. Separate ratings were obtained for each parent. These ratings were combined into a single four-point scale of parental physical punishment/ maltreatment, based on the most severe rating for either parent at either 18 or 21 years: 0=parents never used physical punishment; 1=parents seldom used physical punishment; 2=at least one parent regularly used physical punishment; 3=at least one parent used frequent or severe punishment or treated the participant in a harsh/abusive manner.32,33 For childhood sexual abuse, participants were questioned about their exposure to a range of unwanted sexual experiences including non-contact episodes (eg, indecent exposure, public masturbation); episodes involving sexual contact in the form of sexual fondling, genital contact or attempts to undress the participant; and episodes involving attempted or completed vaginal, oral or anal intercourse.32,33 Using these data participants were classified on a single four-point scale reflecting the most severe form of abuse exposure reported at either age 18 or 21: 0=no childhood sexual abuse; 1=non-contact childhood sexual abuse; 2=contact childhood sexual abuse not involving attempted or completed sexual penetration; and 3=severe childhood sexual abuse involving attempted or completed sexual penetration. • Family violence (<16 years): Witnessing parental intimate partner violence during childhood (prior to age 16 years) was assessed via participant self-report at age 18, through a series of eight items derived from the Conflict Tactics Scale.34 The eight items included: 1. threaten to hit or throw something; 2. push, grab or shove other parent; 3. slap, hit or punch other parent; 4. throw, hit, kick or smash something (in the other parent’s presence); 5. kick the other parent; 6. choke or strangle other parent; 7. threaten other parent with a knife, gun or other weapon; 8. call other parent names or criticize other parent (or put other parent down). An overall measure was created by summing the responses for both father and mother-ini- tiated violence (α=0.88). • Parental adjustment problems: At age 11 parents of cohort members were questioned about their history of illicit drug use. At age 15 parents of cohort members were questioned as to whether any parent had a history of alcohol problems/alco- holism or a history of criminality. • Parental bonding (16 years): At age 16, participants were questioned about the quality of their relationship with their parents during childhood using the Parental Bonding Instrument (PBI).35 This 25-item scale assessed two broad domains of parenting: parental over-protection and parental care. The parental over-protection domain measures variations in the extent to which a parent was perceived to be controlling and unwilling to allow the child autonomy. The parental care dimension assessed the extent to which the parent was perceived to be loving, caring and emotionally supportive. Separate assessments were obtained for mothers and fathers. For the purposes of the present analysis, maternal and paternal ratings were averaged to create overall scores for parental care (α=0.90) and over-protection (α=0.86). • Measures of concurrent family context • Gross family Income (past 12 months): At each assessment from age 25–40 years, participants were questioned about their gross annual income from all sources in the past 12 months and that of their partner (if any). Incomes reported in currencies other than New Zealand dollars (NZD) were converted into NZD using Purchasing Power Parities (Organisation for Economic Co-operation and Devel- opment (OECD)). Incomes were adjusted for inflation using the Consumers Price Index (CPI) to NZD 2017.36

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 26 www.nzma.org.nz/journal ARTICLE

• Welfare dependence (past 12 months): At each assessment, families reporting receipt of social welfare benefits including job seeker, sole parent or supported living payments in the previous 12 months were classified as welfare dependent. • Parental mental health/substance use problems (past 12 months): At each assessment from 25-40 years participants were interviewed using components of the Composite International Diagnostic Interview (CIDI)29 to assess DSM-IV37 diagnostic criteria for a range of mental disorders including major depression, anxiety disorders (generalised anxiety, panic, phobias), alcohol and illicit substance abuse/ dependence in the previous 12 months, together with custom written items to assess suicidal behaviours. For the purposes of this analysis a measure of the severity of the participant’s mental health/substance use problems in the previous 12 months was constructed based on a count of the number of problems reported. • Adverse family life events (past 12 months): At each assessment participants completed a life events checklist spanning a range of adverse life experiences including: relationship problems; employment/financial problems; serious illness, accident or death in the family; victimisation; and related issues. A measure of the extent of exposure to life course stress/adversity was constructed based on a count of the number of reported life events in the previous 12 months at each age. • Intimate partner violence (IPV): At each assessment at ages 25, 30, 35 and 40 years, participants who reported that they were in (or had been in) a married, cohabiting, romantic, intimate or close relationship lasting one month or longer at any time in the past 12 months were questioned about their most recent relationship. Physical IPV over the previous 12 months was assessed using the physical assault sub-scale of the Revised Conflict Tactics Scales (CTS2).38 The scale comprised a series of 25 questions regarding acts of verbal aggression, physical violence or threats ranging from incidents of minor verbal aggression through to severe physical assault. Separate questioning was conducted concerning violence perpetration and violence victimization. For the purposes of the present analysis only the physical assault sub-scale was used, and participants were classified on a dichotomous measure at each age reflecting whether they reported any incident of physical violence perpetration or victimisation in the previous 12 months.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 27 www.nzma.org.nz/journal ARTICLE

Appendix Table 1: Household demographics and childhood-background characteristics of parenting samples at age 25–40 years.

Measure 25 years 30 years 35 years 40 years p (N=155) (N=337) (N=585) (N=636)

Household composition Mean (SD) and range of number of dependent children (<16 years) 1.77 (0.91) 1.98 (1.09) 1.97 (0.96) 2.16 (0.97) Range 1–5 Range 1–8 Range 1–8 Range 1–8 <0.001 children children children children

% (n) Cohabiting partner 71.0 (110) 82.2 (277) 88.7 (519) 89.0 (566) 0.067

Individual Characteristics % (n) Female 71.0 (110) 60.5 (204) 56.9 (333) 53.9 (343) 0.001

% (n) Māori/Tagata Pasifika ethnicity 23.9 (37) 16.6 (56) 14.7 (86) 11.9 (75) 0.001

% (n) Attained university degree by age 25 3.9 (6) 13.5 (45) 25.8 (149) 29.1 (183) <0.001

Childhood-background characteristics % (n) Family of semiskilled/unskilled socioeconomic status 40.7 (63) 35.0 (118) 26.0 (152) 23.7 (151) <0.001

% (n) Mother lacked formal education qualifications 70.0 (109) 63.2 (213) 50.8 (297) 48.1 (306) <0.001

% (n) Entered single parent family at birth 15.5 (24) 8.9 (30) 7.2 (42) 5.8 (37) <0.001

% (n) Experienced 3+ changes of parents (<16 years) 36.8 (57) 26.7 (90) 20.0 (117) 17.1 (109) <0.001

% (n) Experienced childhood physical or sexual abuse (<16 years) 48.4 (75) 37.1 (125) 29.4 (172) 25.0 (159) <0.001

% (n) Witnessed substantial parental IPV (<16 years) 36.1 (56) 29.7 (100) 25.3 (148) 23.1 (147) 0.004

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 28 www.nzma.org.nz/journal ARTICLE

Competing interests: Professor Boden reports grants from Health Research Council during the conduct of the study. Dr McLeod reports grants from Health Research Council during the conduct of the study. Profes- sor John Horwood reports grants from Health Research Council during the conduct of the study. Author information: Geraldine FH McLeod: Senior Research Fellow, Christchurch Health and Development Study, Department of Psychological Medicine, University of Otago, Christchurch. John Horwood: Research Professor, Christchurch Health and Development Study, Depart- ment of Psychological Medicine, University of Otago, Christchurch. Joseph M Boden: Research Professor and Director, Christchurch Health and Development Study, Department of Psychological Medicine, University of Otago, Christchurch. Lianne J Woodward: Professor, School of Health Sciences & Child Wellbeing Research Insti- tute, University of Canterbury, Christchurch. Corresponding author: Dr Geraldine McLeod, Senior Research Fellow, Christchurch Health and Development Study, Department of Psychological Medicine, University of Otago, Christchurch. PO Box 4345, Christchurch 8140, New Zealand [email protected] URL: www.nzma.org.nz/journal-articles/parental-use-of-physical-punishment-in-a-birth-cohort

REFERENCES 1. Woodward LJ, Fergusson AE, Bethell JM, et al. of mediating mechanisms DM, Chesney A, Horwood Childhood physical abuse through school-related LJ. (2007). Punitive and suicide-related outcomes. Child Maltreat. parenting practices of behavior: A systematic 2020; doi/10.1177/10775595 contemporary young review. Vulnerable Child 19900806. parents. N Z Med J. Youth Stud. 2011; 6:1-7. 11. Farrington DP, Malvaso 2007; 120:U2866. 7. Sugaya L, Hasin DS, Olfson CG. Physical punishment 2. Rouland B, Vaithianathan M, Lin KH, Grant BF, Blanco and offending in two R. Cumulative prevalence C. Child physical abuse successive generations of of maltreatment among and adult mental health: males. Child Abuse Negl. New Zealand children, a national study. J Trauma 2019; doi.org/10.1016/ 1998–2015. Am J Public Stress. 2012; 25:384-92. B978-0-12-815344-4.00011-8. Health. 2018; 108:511-3. 8. Carlson BE, Shafer MS, 12. Durrant, JE, Enson, R. 3. Wood B, Hassall I, Hook G, Duffee DE. Traumatic histo- Twentyfive years of phys- Ludbrook R. Unreasonable ries and stressful life events ical punishment research: force. Wellington: Save of incarcerated parents What have we learned? J the Children. 2008. II: Gender and ethnic Korean Acad Child Adolesc 4. McLeod GF, Fergusson DM, differences in substance Psychiatry. 2017; 28:20-24. Horwood LJ. Childhood abuse and service needs. 13. D’Souza AJ, Russell M, physical punishment TPJ. 2010; 90:494-515. Wood B, Signal L, Elder or maltreatment and 9. Simpson TL, Miller WR. D. Attitudes to physical partnership outcomes at Concomitance between punishment of children age 30. Am J Orthopsy- childhood sexual and are changing. Arch Dis chiatry. 2014; 84:307 physical abuse and Child. 2016; 101:690-3. 5. Springer KW, Sheridan J, substance use problems: 14. EPOCH New Zealand. Kuo D, Carnes M. Long- A review. Clin Psychol Changing public atti- term physical and mental Rev. 2002; 22:27-77. tudes towards physical health consequences of 10. Ringle JL, Mason WA, punishment of children. childhood physical abuse: Herrenkohl TI, Smith End Physical Punishment Results from a large GL, Stevens AL, Jung H. of Children. Wellington: population-based sample Prospective associations EPOCH New Zealand, 2013. of men and women. Child of child maltreatment 15. Oranga Tamariki. Quar- Abuse Negl. 2007; 31:517-30. subtypes with adult educa- terly Report December 6. Mironova P, Rhodes tional attainment: Tests 2019. Wellington: Oranga

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 29 www.nzma.org.nz/journal ARTICLE

Tamariki. Available from: 23. Fergusson DM, Horwood 30. Fergusson DM, Horwood LJ, https://www.orangatam- LJ. The Christchurch Lynskey MT. The preva- ariki.govt.nz/statistics/ Health and Development lence and comorbidity of quarterly-reporting/quar- Study. In: Joyce P, Nicholls DSM-III-R diagnoses in terly-report-december-2019/ G, Thomas K, Wilkinson a birth cohort of 15 year care-and-protec- T, (eds) The Christchurch olds. J Am Acad Child tion-statistics-2/ Experience: 40 Years of Adolesc Psychiatry. 16. Holden GW. Why Do Research and Teaching. 1993; 32:1127-34. Parents Hit Their Chil- Christchurch: University 31. Elley WB, Irving JC. Revised dren? From Cultural to of Otago, 2013; 79-87. socio-economic index for Unconscious Determinants. 24. Straus MA, Hamby SL. New Zealand. New Zeal J Psychoanal Study Child. Measuring Physical & Educ Stud. 1976; 11:25-36. 2020; 73(1):10-29. Psychological Maltreatment 32. Fergusson DM, Boden JM, 17. Woodward LJ, Fergusson of Children with the Conflict Horwood LJ. Exposure to DM. Parent, child, and Tactics Scales. 1997. Paper childhood sexual and phys- contextual predictors presented at the Annual ical abuse and adjustment of childhood physical Meeting of the American in early adulthood. Child punishment. Infant Child Educational Research Abuse Negl. 2008; 3:607-619. Association (Chicago, Dev. 2002; 11:213-35. 33. Fergusson DM, Horwood IL, March 24-28,1997). 18. Montgomery E, Just-Øster- LJ, Boden JM. Exposure gaard E, Jervelund SS. 25. Straus MA, Hamby SL, to childhood sexual and Transmitting trauma: Finkelhor D, Moore DW, physical abuse and adjust- a systematic review of Runyan D. Identification ment in early adulthood. the risk of child abuse of child maltreatment with Int J Methods Psychiatry perpetrated by parents the Parent-Child Conflict Res. 2011; 20(2): 93-104. Tactics Scales: Develop- exposed to traumatic 34. Straus MA. Measuring ment and psychometric events. Int J Public intrafamily conflict and data for a national sample Health. 2019; 64:241-51. violence: The conflict of American parents. Child 19. Wissow LS. Ethnicity, tactics (CT) scale. J Marriage Abuse Negl. 1998; 22:249-70. income, and parenting Fam. 1979; 41: 75-88. 26. Cohen J. Statistical contexts of physical 35. Parker G, Tupling H, analysis for the behav- punishment in a national Brown LB. A parental ioral sciences. 2013. New sample of families with bonding instrument. Br York: Academic Press. young children. Child J Med. 1979;51:1-10. Maltreat. 2001; 6:118-29. 27. Anderson RE, Edwards 36. Organisation for Econom- L-J, Silver KE, Johnson 20. Perron JL, Lee CM, LaRoche ic Co-operation and DM. Intergenerational KJ, Ateah C, Clément M-È, Development (OECD) transmission of child Chan K. Child and parent 2017. Organisation for abuse: Predictors of child characteristics associated Economic Co-operation abuse potential among with Canadian parents’ and Development (OECD), racially diverse women reports of spanking. 2017 PPP Benchmark residing in domestic Can J Commun Ment results. Retrieved from violence shelters. Child Health. 2014; 33:31-45. https://www.oecd.org/sdd/ Abuse Negl. 2018; 85:80-90. 21. MacKenzie MJ, Nicklas E, prices-ppp/purchasingpow- Brooks-Gunn J, Waldfogel 28. Doidge JC, Higgins DJ, erparitiespppsdata.htm Delfabbro P, et al. Econom- J. Who spanks infants 37. American Psychiatric Asso- ic predictors of child and toddlers? Evidence ciation. (1994). Diagnostic maltreatment in an Austra- from the fragile families and Statistical Manual of lian population-based and child well-being Mental Disorders (4th ed.). birth cohort. Child Youth study. Child Youth Serv Washington DC: American Serv Rev. 2017; 72:14-25. Rev. 2011; 33:1364-1373. Psychiatric Association. 29. Costello A, Edelbrock C, 22. Fergusson DM, Horwood 38. Straus MA, Hamby SL, Kalas R, Kessler M, Klaric LJ. The Christchurch Boney-McCoy S, Sugarman SA. The National Institute Health and Development DB. The revised Conflict of Mental Health Diagnostic Study: Review of findings Tactics Scales (CTS2). Devel- Interview Schedule for on child and adolescent opment and preliminary Children (DISC). Rockville, mental health. Aust N Z J psychometric data. J Fam MD: National Institute of Psychiatry. 2001; 35:287-96. Issues. 1996; 17:283-316. Mental Health, 1982.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 30 www.nzma.org.nz/journal ARTICLE

Demographic differences in the initiation and maintenance of statins in the first year post-ACS in New Zealand: a data linkage study (ANZACS-QI 57) Aravindra Muniandy, Mildred Lee, Corina Grey, Katherine Ferrier, Andrew J Kerr

ABSTRACT INTRODUCTION: Prior New Zealand studies suggest that only approximately two-thirds of patients who present with an acute coronary syndrome (ACS) are adequately maintained on a statin post-discharge. This could be due to low initiation and/or poor longer-term adherence. AIM: To identify the pattern and adequacy of statin maintenance following ACS from initial prescription to one-year post-discharge. METHODS: All New Zealand Acute Coronary Syndrome Quality Improvement (ANZACS-QI) registry data for consecutive New Zealand residents (2015–2017) who were hospitalised with ACS and managed with coronary angiography were anonymously linked to national datasets to derive a medication possession ratio (MPR) to assess medication maintenance. An MPR ≥0.8 is considered adequate maintenance and ≥1 is considered optimal. RESULTS: Of the 16,557 patients who survived their ACS, 15,431 (93.2%) were prescribed a statin at discharge and 89.8% were dispensed a statin within three months. 79.8% (13,219/16,557) of patients had an MPR ≥0.8 during the first year, but only 61.0% (10,096/16,557) had optimal dispensing over this period. Regression analysis identified the independent predictors of sub-optimal maintenance over the first year as age <45 years, no prior statin and Māori and Pacific ethnicity. CONCLUSION: After ACS discharge, the gap between prescribing and dispensing rates was small with only minor demographic variation. One in ten patients were not initially dispensed a statin. Although eight in ten patients were adequately maintained, only six in ten had optimal maintenance with clear ethnic and age differences, which may reflect more general disparities in healthcare.

ver 12,000 patients are admitted to an ACS are adequately maintained on a sta- New Zealand hospitals with an acute tin in the three years post-discharge.5 Vari- coronary syndrome (ACS) every ables independently associated with poorer O1 year. There is robust evidence to support maintenance included younger age and early initiation and long-term use of statin Māori and Pacific ethnicity. However, it is medications to improve outcomes in pa- not known whether this was due to sub-op- tients after ACS,2, 3 and poor adherence is timal prescribing at the time of hospital dis- associated with increased risk of rehospi- charge, barriers to initial dispensing or poor talisation and mortality.4 In a previous New longer-term maintenance. Furthermore, Zealand national analysis we reported that consistent with prior studies, adequate sta- only two-thirds of patients who present with tin maintenance in that study was defined as

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 31 www.nzma.org.nz/journal ARTICLE

a medication possession ratio (MPR) of ≥0.8, For this study, at least one year of post-dis- meaning patients had enough statin supply charge follow-up time was required. The for 80% or more of the days during the study study cohort was therefore comprised of period. However, for low-density lipoprotein consecutive New Zealand residents aged (LDL) cholesterol lowering with a statin, 35 to 84 years who presented to public more optimal dispensing (MPR≥1) has been hospitals with a confirmed diagnosis of associated with greater LDL-cholesterol low- myocardial infarction (MI) or unstable ering.6 The levels of optimal statin mainte- angina between 1 January 2015 to 31 nance in New Zealand and the demographic December 2017. Only the first admission for factors associated with sub-optimal mainte- each patient in the time period was used. nance have not been studied. MI was defined according to the contem- 10,11 In New Zealand, the All New Zealand porary universal definition. Patients Acute Coronary Syndrome Quality were excluded if they died within 30 days of Improvement (ANZACS-QI) registry captures discharge or spent less than 30 days out of data, including discharge prescriptions, hospital over the follow-up period. Patients for all New Zealand ACS patients who are referred for a coronary artery bypass graft investigated with coronary angiography. In (CABG) were excluded, as the prescribing addition, dispensing data for all subsidised data are not available at hospital discharge medications, including statins, are captured for these patients. An encrypted version of in the national routine datasets, allowing the National Health Index (NHI) number, a assessment of post-discharge statin main- unique identifier assigned to everyone who tenance.7 The registry cohort can be linked uses health and disability support services 12 to the national datasets to compare initial (>98% of the population), was used to anon- prescribing and dispensing of statins. ymously link in-hospital ANZACS-QI patient records to national administrative datasets, The aims of this study were therefore to including the medication dispensing, hospi- describe the use of statin medication and talisation and mortality data, as previously its demographic and clinical determinants described.8 both early and in the first year post-ACS discharge. Specifically, we sought to describe Data and definitions: Variables used for the initial prescribing of statins post-ACS this study included age, sex, ethnicity, prior and the gap between the initial prescribing statin use, diabetes, smoking status, history and dispensing of statin medications and of cardiovascular disease (CVD), percu- the one-year maintenance on statins. We taneous coronary intervention (PCI) and also sought to assess the variables associated a global measure of ACS risk (the Global 13 with sub-optimal statin utilisation at each Registry of ACS (GRACE) score). Sociodemo- stage. graphic variables and residency status were derived from the linked national dataset. In accordance with health sector protocols, Methods the ethnicities of patients for whom more Cohort and data collection: The study than one ethnic group was recorded were cohort was identified from the ANZACS-QI prioritised in the following order: indig- registry, a web-based electronic database that enous Māori, Pacific, Indian and New captures a mandatory dataset for all patients Zealand European (NZEO).14 Socioeconomic admitted to New Zealand public hospitals deprivation was assessed by the NZDep13 with acute coronary syndrome and who are score, a Census-based, small-area 10-point investigated with coronary angiography. index of deprivation based on each person’s Data collected includes patient demo- domicile.15 Prior statin use was derived graphics, admission ACS risk stratification, from the national pharmaceutical claims cardiovascular risk factors, investigations dataset and was defined as having a statin and management, inpatient outcomes and dispensed during the 90-day period prior to medications prescribed at discharge. Details index admission. regarding data collection have previously Main outcome measures: Statin prescription been reported.8,9 The registry is subject at discharge is recorded in the ANZACS-QI to monthly auditing to ensure >99% of all registry. The registry also records known patients are captured, and annual auditing to intolerance/contraindication to statin. check the accuracy of data entry.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 32 www.nzma.org.nz/journal ARTICLE

Early dispensing was defined as vention (PCI), Granger risk and non-cardiac dispensing within three months of discharge Charlson comorbidity. All p-values reported after an acute coronary syndrome event were two-tailed and p-value <0.05 was recorded in the national Pharmaceutical considered significant. Data were analysed Collection. Statin medication included any of using SAS statistical package, version 9.4 the following publicly funded medications: (SAS Institute, Cary, NC). atorvastatin, fluvastatin, pravastatin or Ethics approval: ANZACS-QI is part of the simvastatin with or without ezetimibe. Rosu- wider Vascular Informatics Using Epidemi- vastatin is not a funded medication in New ology and the Web (VIEW) study. The VIEW Zealand and is infrequently prescribed. study was approved by the Northern Region Statin use over the year after discharge Ethics Committee Y in 2003 (AKY/03/12/314), was assessed by calculating a medication with subsequent amendments to include possession ratio (MPR).16 The MPR is the the ANZACS-QI registries, and with annual number of days the drug was assumed approvals by the National Multi-region to be in a patient’s possession (based on Ethics Committee since 2007 (MEC07/19/ dispensed drugs) divided by the number EXP). of days spent out of hospital from the date of hospital discharge through to the end of Results the follow-up period or the date of death, In the three-year study period there whichever came first. The possession were 16,557 patients (Table 1). The mean of medications during 80% or more of age of patients was 64.7 years, with 68.2% follow-up time (ie, MPR≥0.8) was used to male (11,288/16,557), 77.8% European/ classify those adequately maintained on Other (12,873/16,557), 10.9% Māori medications, as reported in other studies.4 (1,808/16,557), 4.3% Pacific (713/16,557), An MPR=0 indicates no dispensing at all. For 4.2% Indian (702/16,557) and 2.8% Other many cardiac drugs, even short periods of Asian (461/16,557). There was a socioeco- non-adherence may have adverse clinical nomic gradient with the highest proportion consequences. We therefore also reported of patients living in the most deprived areas. the proportions of patients with optimal 35.7% (5,908/16,557) had a history of prior one-year (MPR≥1) dispensing. The MPR can CVD, 40.8% (6,752/16,557) were receiving a be greater than one when more medication statin prior to the index admission, 22.5% is dispensed than is required to cover the (3,730/16,557) had diabetes mellitus and dispensing period. 23.8% (3,934/16,557) were current smokers. Statistical analysis 65.9% (10,910/16,557) underwent PCI during Categorical data were summarised as the index admission. frequency and percentage, and continuous Statin prescribing and dispensing (Table data were reported as mean and standard 1): 93.2% (15,431/16,557) patients were deviation (SD). Comparison of categorical prescribed a statin at discharge and 1.9% data between groups was performed by (316/16,557) had a known intolerance to Chi-squared test. For continuous data, statin. Initial prescribing of statin was comparison between groups was performed higher in men than women, similar for by the non-parametric Mann–Whitney Māori, Pacific and European/Other groups, U test, due to data not being normally but slightly higher for Indian patients. There distributed. was no difference based on NZDep13 status. Poisson regression modelling using a Statin prescribing was also higher in those robust variance estimator was used to with more severe ACS type, those treated estimate relative risks (RRs) with accom- with PCI and those with less non-cardiac panying 95% confidence intervals (CIs) of comorbidity. initial statin prescribing, one-year statin 89.8% (14,864/16,557) of patients were MPR≥1 and one-year statin MPR≥0.8. The dispensed a statin within three months; the models were adjusted for baseline age, mean decrement between prescribing and gender, ethnicity, NZDep13 score, smoking dispensing was 3.4%. This gap was slightly status, diabetes, prior CVD, history of greater for older versus younger, for women congestive heart failure (CHF), prior statin, versus men, for non-smokers, those with type of ACS, percutaneous coronary inter-

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 33 www.nzma.org.nz/journal ARTICLE

prior CVD or less severe ACS type, those Other variables associated with higher who did not receive PCI and those with rates of ideal dispensing were non-smoking, more non-cardiac comorbidity. However, all diabetes, no prior CVD, more severe ACS differences between levels of each variable type, coronary revascularisation and higher were small—no more than 4%. There was GRACE ACS risk scores. no significant difference in the ‘prescribed– Multiple regression analyses of initial dispensed’ gap across ethnic groups or level prescribing and one-year statin mainte- of socioeconomic deprivation. nance (Table 2): After adjustment, the Other secondary prevention medi- variables associated with higher statin cations: The differences between initial prescribing rates included male sex, prescribing and dispensing were similar age over 45, Indian ethnicity, prior statin for other secondary prevention medica- dispensing, current smoking, more severe tions: aspirin (91.7% prescribed, 88.6% ACS presentation, coronary intervention and dispensed, difference=3.1%), P2Y12 less non-cardiac comorbidity. All differ- inhibitor (85.6% prescribed, 84.2% ences in adjusted relative risks were small dispensed, difference=1.4%), angiotensin and under 10%. There was no difference converting enzyme inhibitor/aldosterone between Māori, Pacific and European receptor blocker (68.8% prescribed, 65.3% people. dispensed, difference=3.5%) and beta- One-year optimal statin maintenance: blocker (79.6 prescribed, 77.2% dispensed, After adjustment, the largest adjusted difference=2.4%). relative risks were prior statin therapy and MPR, medication possession ratio; LDL, PCI, which were associated with 61%- and low-density lipoprotein; CVD, cardiovas- 23%-higher ideal dispensing than those with cular disease; CHF, congestive heart failure; no prior statin or no PCI, respectively. The ACS, acute coronary syndrome; USA, oldest patients were approximately 20% unstable angina; NSTEMI, non-ST-elevation more likely to have ideal dispensing than myocardial infarction; STEMI, ST segment the youngest patients. Optimal dispensing elevation MI; PCI, percutaneous coronary did not vary by sex. European patients intervention. were 12% more likely than Māori or Maintenance over one-year post-discharge Pacific patients to have optimal dispensing. (Table 1 and Figures 1 and 2): Overall, 79.8% Patients without prior CVD and those with (13,219/16,557) of patients had a statin MPR STEMI were also over 10% more likely than ≥0.8 in the first year post discharge but only reference groups to have ideal dispensing. 61.0% (10,096/16,557) had an ideal MPR ≥1. These findings were mirrored in the multi- Figure 2 shows the distribution of MPRs by variable analysis (Appendix Table 1) using age, sex, ethnic group and NZDep13 level. the adequate MPR>=0.8 cut-off, but the The MPRs are divided into four groups: no adjusted risks were all less marked. statin dispensed (MPR=0), sub-optimal statin dispensing (MPR>0–<0.8), adequate but Discussion not ideal (MPR.8–<1) and ideal dispensing In this nationwide study there were high (MPR≥1). Women were more likely than men rates of statin prescribing at discharge for to not receive a statin, which is consistent all subgroups, with only slightly lower use in with their lower initial prescription rate. women than men and higher use in Indian Despite similar initial prescribing and than other ethnic groups. Reassuringly, nine dispensing rates, only 50.8% (918/1,808) out of ten patients were dispensed a statin; of Māori and 53.7% (383/713) of Pacific the percentage of people dispensed a statin patients had ideal MPRs, lower than Indian was only 3% lower than those prescribed, (457/702, 5.1%), Other Asian (297/461, 64.4%) with only small differences between demo- and European/Other (8,041/12,873, 62.5%). graphic and clinical subgroups. Less than Older patients were also more likely than 2% of patients had a known intolerance to a younger patients to have ideal MPRs. In statin. Statin maintenance over the first year contrast, Māori, Pacific and younger patients was lower, with eight out of ten patients are disproportionately represented in the at least adequately maintained on a statin sub-optimal MPR category. There is no clear but only six out of ten receiving optimal difference according to NZDep13 status. dispensing. There was clinically important

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 34 www.nzma.org.nz/journal ARTICLE

Table 1: Patient characteristics according to initial statin prescribing and dispensing and maintenance over the year post discharge.

Age Mean (SD) 64.7 (11.0) 69.3 (9.8) 64.6 (11.0) 64.5 (11.0) 66.4 (10.7) 64.9 (10.9) 65.6 (10.8) P-value - <.001 <.001 <.001 <.001 <.001 <.001 Age group 35–<45 667 (4.0) 4 (0.6) 616 (92.4) 602 (90.3) 14 (2.3) 488 (73.2) 328 (49.2) 45–<55 2676 (16.2) 27 (1.0) 2516 (94.0) 2449 (91.5) 67 (2.7) 2067 (77.2) 1431 (53.5) 55–<65 4385 (26.5) 56 (1.3) 4154 (94.7) 4010 (91.4) 144 (3.5) 3516 (80.2) 2580 (58.8) 65–<75 5182 (31.3) 122 (2.4) 4828 (93.2) 4632 (89.4) 196 (4.1) 4230 (81.6) 3335 (64.4) 75–<85 3647 (22.0) 107 (2.9) 3317 (91.0) 3171 (86.9) 146 (4.4) 2918 (80.0) 2422 (66.4) P-value - <.001 <.001 <.001 0.001 <.001 <.001 Sex Male 11288 (68.2) 185 (1.6) 10696 (94.8) 10344 (91.6) 352 (3.3) 9273 (82.2) 7016 (62.2) Female 5269 (31.8) 131 (2.5) 4735 (89.9) 4520 (85.8) 215 (4.5) 3946 (74.9) 3080 (58.5) P-value - <.001 <.001 <.001 <.001 <.001 <.001 Ethnicity Māori 1808 (10.9) 22 (1.2) 1680 (92.9) 1630 (90.2) 50 (3.0) 1325 (73.3) 918 (50.8) Pacific 713 (4.3) 8 (1.1) 672 (94.3) 653 (91.6) 19 (2.8) 532 (74.6) 383 (53.7) Indian 702 (4.2) 6 (0.9) 685 (97.6) 664 (94.6) 21 (3.1) 589 (83.9) 457 (65.1) Other Asian 461 (2.8) 3 (0.7) 431 (93.5) 422 (91.5) 9 (2.1) 382 (82.9) 297 (64.4) European/ 12873 (77.7) 277 (2.2) 11963 (92.9) 11495 (89.3) 468 (3.9) 10391 (80.7) 8041 (62.5) Other P-value - <.001 <.001 <.001 0.053 <.001 <.001 NZDep13 1–2 2841 (17.2) 46 (1.6) 2650 (93.3) 2534 (89.2) 116 (4.4) 2317 (81.6) 1774 (62.4) 3–4 2851 (17.2) 55 (1.9) 2649 (92.9) 2546 (89.3) 103 (3.9) 2274 (79.8) 1752 (61.5) 5–6 3247 (19.6) 61 (1.9) 3024 (93.1) 2916 (89.8) 108 (3.6) 2630 (81.0) 2024 (62.3) 7–8 3772 (22.8) 84 (2.2) 3505 (92.9) 3389 (89.8) 116 (3.3) 3016 (80.0) 2319 (61.5) 9–10 3798 (22.9) 69 (1.8) 3557 (93.7) 3441 (90.6) 116 (3.3) 2969 (78.2) 2218 (58.4) Missing 48 (0.3) 1 (2.1) 46 (95.8) 38 (79.2) 8 (17.4) 13 (27.1) 9 (18.8) P-value - 0.720 0.715 0.333 0.127 <.001 0.003 Smoking status Non-smoker 6980 (42.2) 152 (2.2) 6429 (92.1) 6147 (88.1) 282 (4.4) 5526 (79.2) 4288 (61.4) Ex-smoker 5643 (34.1) 113 (2.0) 5265 (93.3) 5077 (90.0) 188 (3.6) 4620 (81.9) 3665 (64.9) Current 3934 (23.8) 51 (1.3) 3737 (95.0) 3640 (92.5) 97 (2.6) 3073 (78.1) 2143 (54.5) smoker P-value - <.001 <.001 <.001 <.001 <.001 <.001 LDL <2 5298 (32.0) 80 (1.5) 4966 (93.7) 4792 (90.4) 174 (3.5) 4478 (84.5) 3531 (66.7) 2–<3 4828 (29.2) 80 (1.6) 4487 (92.9) 4334 (89.8) 153 (3.4) 3840 (79.5) 2975 (61.6) ≥3 6413 (38.7) 156 (2.4) 5960 (92.9) 5720 (89.2) 240 (4.0) 4889 (76.2) 3582 (55.9) Missing 18 (0.1) 0 (0) 18 (100) 18 (100) 0 (0) 12 (66.7) 8 (44.4) P-value - 0.001* 0.001* 0.083* 0.184* <.001* <.001*

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 35 www.nzma.org.nz/journal ARTICLE

Table 1: Patient characteristics according to initial statin prescribing and dispensing and maintenance over the year post discharge (continued).

Overall Statin Prescribed Prescribed Prescribed MPR 1y ≥ 0.8 MPR 1yr ≥ 1 contraindi- statin and but not cated or dispensed dispensed not tolerated N n (row %) n (row %) n (row %) n (row %) n (row %) n (row %) Diabetes Yes 3730 (22.5) 80 (2.1) 3494 (93.7) 3375 (90.5) 119 (3.4) 3027 (81.2) 2382 (63.9) No 12827 (77.5) 236 (1.8) 11937 (93.1) 11489 (89.6) 448 (3.8) 10192 (79.5) 7714 (60.1) P-value - 0.039 0.192 0.105 0.337 0.023 <.001 Prior CVD Yes 5908 (35.7) 183 (3.1) 5452 (92.3) 5112 (86.5) 340 (6.2) 4593 (77.7) 3494 (59.1) No 10649 (64.3) 133 (1.2) 9979 (93.7) 9752 (91.6) 227 (2.3) 8626 (81.0) 6602 (62.0) P-value - <.001 0.001 <.001 <.001 <.001 <.001 History of CHF Yes 626 (3.8) 25 (4.0) 547 (87.4) 524 (83.7) 23 (4.2) 492 (78.6) 399 (63.7) No 15931 (96.2) 291 (1.8) 14884 (93.4) 14340 (90.0) 544 (3.7) 12727 (79.9) 9697 (60.9) P-value - <.001 <.001 <.001 0.502 0.429 0.149 Prior statin Yes 6752 (40.8) 25 (4.0) 6592 (97.6) 6436 (95.3) 156 (2.4) 6157 (91.2) 5025 (74.4) No 9805 (59.2) 291 (1.8) 8839 (90.1) 8428 (86.0) 411 (4.6) 7062 (72.0) 5071 (51.7) P-value - <.001 <.001 <.001 <.001 <.001 <.001 Type of ACS USA 2767 (16.7) 76 (2.7) 2495 (90.2) 2343 (84.7) 152 (6.1) 2122 (76.7) 1540 (55.7) NSTEMI 9410 (56.8) 191 (2.0) 8719 (92.7) 8403 (89.3) 316 (3.6) 7418 (78.8) 5653 (60.1) STEMI 4380 (26.5) 49 (1.1) 4217 (96.3) 4118 (94.0) 99 (2.4) 3679 (84.0) 2903 (66.3) P-value - <.001 <.001 <.001 <.001 <.001 <.001 PCI Yes 10910 (65.9) 191 (1.8) 10443 (95.7) 10124 (92.8) 319 (3.1) 9155 (83.9) 7122 (65.3) No 5647 (34.1) 125 (2.2) 4988 (88.3) 4740 (83.9) 248 (5.0) 4064 (72.0) 2974 (52.7) P-value - <.001 <.001 <.001 <.001 <.001 <.001 Granger risk <1% 4587 (27.3) 70 (1.5) 4307 (93.9) 4151 (90.5) 156 (3.6) 3580 (78.1) 2530 (55.2) 1–3% 6813 (41.1) 138 (2.0) 6337 (93.0) 6094 (89.5) 243 (3.8) 5453 (80.0) 4121 (60.5) ≥3% 5157 (31.1) 108 (2.1) 4787 (92.8) 4619 (89.6) 168 (3.5) 4186 (81.2) 3445 (66.8) P-value - 0.156 0.081 0.163 0.650 0.001 <.001 Non-cardiac Charlson 0 12024 (72.6) 193 (1.6) 11321 (94.2) 10944 (91.0) 377 (3.3) 9705 (80.7) 7374 (61.3) 1 1666 (10.1) 37 (2.2) 1535 (92.1) 1458 (87.5) 77 (5.0) 1299 (78.0) 1012 (60.7) 2 1510 (9.1) 38 (2.5) 1359 (90.0) 1298 (86.0) 61 (4.5) 1171 (77.6) 896 (59.3) 3+ 1357 (8.2) 48 (3.5) 1216 (89.6) 1164 (85.8) 52 (4.3) 1044 (76.9) 814 (60.0) P-value - <.001 <.001 <.001 0.005 <.001 0.403

* P-value excluded missing LDL.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 36 www.nzma.org.nz/journal ARTICLE

Figure 1: Initial statin prescribing and dispensing and maintenance over the year post discharge.

Figure 2: Distribution of one-year MPRs for gender, age, ethnicity and NZDep13.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 37 www.nzma.org.nz/journal ARTICLE

Table 2: Multivariable models: predictors of initial prescribing and one-year MPR ≥1.

Initial prescribing One-year MPR ≥1

RR (95% CI) P-value RR (95% CI) P-value Age <45 1.00 - 1.00 - 45–<55 1.02 (1.00–1.05) 0.03 1.06 (0.97–1.15) 0.20 55–<65 1.04 (1.02–1.06) <.01 1.10 (1.02–1.20) 0.02 65–<75 1.04 (1.02–1.07) <.01 1.18 (1.09–1.28) <.01 75–<85 1.03 (1.01–1.06) 0.01 1.21 (1.11–2.32) <.01

Gender Female 0.97 (0.96–0.98) <.01 0.98 (0.95–1.00) 0.07 Male 1.00 - 1.00 -

Ethnicity Māori 1.00 (0.99–1.01) 0.79 0.88 (0.84–0.92) <.01 Pacific 1.01 (0.99–1.03) 0.28 0.88 (0.82–0.94) <.01 Indian 1.03 (1.01–1.04) <.01 1.00 (0.94–1.05) 0.91 Asian 0.99 (0.97–1.02) 0.59 1.01 (0.95–1.09) 0.67 European/Other 1.00 - 1.00 -

NZDep13 1–2 0.99 (0.98–1.00) 0.20 1.02 (0.98–1.06) 0.38 3–4 0.99 (0.98–1.00) 0.10 1.00 (0.96–1.04) 0.96 5–6 0.99 (0.98–1.00) 0.18 1.02 (0.98–1.06) 0.38 7–8 0.99 (0.98–1.00) 0.25 1.02 (0.98–1.05) 0.36 9–10 1.00 - 1.00 -

Smoking status Non-smoker 1.00 - 1.00 - Ex-smoker 1.01 (1.00–1.01) 0.25 1.03 (1.00–1.06) 0.03 Current smoker 1.02 (1.01–1.03) <.01 0.93 (0.90–0.97) <.01

Diabetes Yes 1.01 (1.00–1.02) 0.28 1.02 (0.98–1.05) 0.34 No 1.00 - 1.00 -

Prior CVD Yes 0.98 (0.97–0.99) <.01 0.81 (0.79–0.83) <.01 No 1.00 - 1.00 -

History of CHF Yes 0.97 (0.94–0.99) 0.01 1.08 (1.02–1.15) 0.01 No 1.00 - 1.00 -

Prior statin Yes 1.10 (1.09–1.11) <.01 1.61 (1.57–1.65) <.01 No 1.00 - 1.00 -

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 38 www.nzma.org.nz/journal ARTICLE

Table 2: Multivariable models: predictors of initial prescribing and one-year MPR ≥1 (continued).

Initial prescribing One-year MPR ≥1

RR (95% CI) P-value RR (95% CI) P-value Type of ACS USA 0.93 (0.92–0.95) <.01 0.80 (0.76–0.83) <.01 NSTEMI 0.97 (0.96–0.98) <.01 0.91 (0.88–0.94) <.01 STEMI 1.00 - 1.00 -

PCI Yes 1.07 (1.06–1.08) <.001 1.23 (1.20–1.27) <.01 No 1.00 - 1.00 -

Granger risk <1% 1.00 - 1.00 - 1-3% 0.98 (0.97–0.99) <.01 1.00 (0.97–1.04) 0.94 ≥3% 0.97 (0.96–0.98) <.01 1.02 (0.98–1.07) 0.28

Non-cardiac Charlson 0 1.00 - 1.00 - 1 0.97 (0.96–0.99) <.01 0.94 (0.90–0.98) <.01 2 0.96 (0.95–0.98) <.01 0.92 (0.88–0.96) <.01 3+ 0.96 (0.95–0.98) <.01 0.92 (0.88–0.96) <.01

MPR, medication possession ratio; LDL, low-density lipoprotein; CVD, cardiovascular disease; CHF, congestive heart failure; ACS, acute coronary syndrome; USA, unstable angina; NSTEMI, non-ST-elevation myocardial infarction; STEMI, ST segment elevation MI; PCI, percutaneous coronary intervention.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 39 www.nzma.org.nz/journal ARTICLE

variation, with younger patients, Māori and been associated with better LDL lowering.6 Pacific people less likely to receive optimal Of perhaps greater significance, the lower dispensing. MPRs for these patients may be a marker of Prior studies: The maintenance rates for less-optimal medical management and other statins in our cohort are slightly higher than healthcare more generally. This has previ- those reported using a 2007 New Zealand ously been postulated as the inverse of the 4, 21 national ACS cohort. In that study, which ‘healthy adherer’ effect. included all ACS patients, 69% had a statin Furthermore, although the observed MPR >=0.8 in the first year post-ACS.17 The differences in these two studies are rela- higher rate in our cohort is partially due to tively small, they add to other work the cohort only being comprised of patients demonstrating similar differences between receiving coronary angiography, who are ethnic groups at several points across the more likely to receive medications than pathway of care. These include Māori with those not referred for coronary angiography, STEMI having longer delays in initially although the result is better than the MPR calling for help, and both Māori and Pacific of 75% reported in a smaller, two-centre patients having slightly lower angiog- ANZACS-QI cohort from 2007 to 2011. The raphy and intervention rates even after prior studies also reported lower statin adjustment for covariates.22,23 Although the use in Māori, Pacific and younger people. difference between ethnicities may be small In the prior national study, there was no at each point, cumulatively they may add up information about how many patients were to a significant impact on outcome. prescribed a statin at discharge or recorded Clinical implications and what is as being intolerant of statins. In the current achievable: In this study some subgroups are and the prior studies, patients who were achieving adequate statin maintenance in dispensed a statin prior to the index over 90% of patients, which suggests a target admission were more likely than those level that is potentially achievable overall. without prior statin dispensing to continue Commencement and continuation of on a statin. These are presumably patients statins and other medications of proven who have previously tolerated statin therapy clinical benefit that have been started in and have accepted it use. The current study hospital requires optimal performance demonstrates that approximately 10% of across this care continuum. While the the sub-optimal longer-term maintenance of patient is in hospital, multidisciplinary statin is due to initial prescribing practice, involvement of the medical, nursing known intolerances or barriers to getting and cardiac-rehabilitation teams and medication dispensed early after discharge. pharmacists is required to both educate In this cohort, only another 10% had and support patients and their families inadequate (MPR<0.8) statin maintenance regarding the benefits and potential risks over the first year. This will include some of medications. Practical considerations patients who stopped due to experiencing include simplifying the dosing regimen side effects. Prior large observational studies and encouraging patients to blister pack have reported an up-to-10% intolerance due their medications or utilise a similar to the statin-specific side-effects of myalgia adherence-facilitating process.24 Financial and myopathy18,19 and real or perceived barriers have an impact on adherence intolerance to statins. Despite 80% receiving and should be addressed.24 The default adequate dispensing, only 60% of patients charge per item dispensed in New Zealand received optimal dispensing (MPR≥1), and is $5. Although there are programmes to for Māori, Pacific and younger patients this reduce this for high-needs populations, it is lower at around 50%. Māori and Pacific is likely that some patients defer medica- people have the worst ischaemic heart tions because of cost considerations. Ideally disease outcomes.1,20 The clinical significance there should be referral to culturally appro- of this difference between optimal and priate cardiac-rehabilitation programmes adequate dispensing is not known. The defi- after discharge to continue to educate and nition of adequate dispensing (MPR>0.8) is support patients and facilitate the tran- an accepted convention,4 but for LDL choles- sition to primary care.25 Beyond hospital, terol lowering, more optimal dispensing has primary-care-based self-monitoring and

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 40 www.nzma.org.nz/journal ARTICLE

self-management programmes have dispensing data for future research and demonstrably improved adherence to quality-improvement projects. medication regimes, and technology that Limitations ensures re-prescription of medications We used dispensing of statin as a marker does continue after discharge should be of maintenance, but not everyone who is adopted.24 In addition, the publication of dispensed a drug routinely is necessarily data is an important way to give feedback taking the medication. to clinicians and modify prescribing behaviour.26Since 2019, New Zealand’s ANZACS-QI programme has reported the Conclusions one-year post-ACS statin maintenance In this cohort of ACS patients managed within each of the 20 district health board with an invasive strategy, the rate of initial catchments. From 2020, this annual report statin prescribing and dispensing was high has been provided directly to the public but could be further improved. A high via the Heart Foundation website.27 This proportion of patients had adequately reporting is intended to prompt units to maintained dispensing over the year post review their local guidelines and will allow discharge, but optimal maintenance was progress to be tracked. less satisfactory and there are clear ethnic and age differences in the optimal use of In this study, the initial post-ACS-dis- medication, which may reflect more general charge prescribing and dispensing rates disparities in healthcare. were similar, which supports the use of

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 41 www.nzma.org.nz/journal ARTICLE

Appendix

Appendix Table 1: Multivariable models: predictors of one-year MPR ≥0.8.

One-year MPR ≥ 0.8

RR (95% CI) P-value Age <45 1.00 - 45–<55 1.04 (0.99–1.10) 0.08 55–<65 1.07 (1.02–1.12) 0.01 65–<75 1.09 (1.24–1.15) <.01 75–<85 1.09 (1.03–1.14) <.01

Gender Female 0.94 (0.92–0.96) <.01 Male 1.00 -

Ethnicity Māori 0.94 (0.91–0.97) <.01 Pacific 0.93 (0.89–0.97) <.01 Indian 1.00 (0.96–1.03) 0.82 Asian 1.01 (0.97–1.05) 0.54 European/Other 1.00 -

NZDep13 1–2 1.01 (0.99–1.04) 0.23 3–4 0.99 (0.97–1.02) 0.60 5–6 1.01 (0.99–1.03) 0.37 7–8 1.01 (0.98–1.03) 0.59 9–10 1.00 -

Smoking status Non-smoker 1.00 - Ex-smoker 1.01 (1.00–1.03) 0.09 Current smoker 1.00 (0.98–1.02) 0.86

Diabetes Yes 1.00 (0.98–1.02) 0.99 No 1.00 -

Prior CVD Yes 0.87 (0.85–0.89) <.01 No 1.00 -

History of CHF Yes 1.03 (0.99–1.07) 0.16 No 1.00 -

Prior statin Yes 1.38 (1.36–1.41) <.01 No 1.00 -

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 42 www.nzma.org.nz/journal ARTICLE

Appendix Table 1: Multivariable models: predictors of one-year MPR ≥0.8 (continued).

One-year MPR ≥ 0.8

RR (95% CI) P-value Type of ACS USA 0.88 (0.85–0.90) <.001 NSTEMI 0.94 (0.92–0.96) <.001 STEMI 1.00 -

PCI Yes 1.15 (1.13–1.17) <.001 No 1.00 -

Granger risk <1% 1.00 - 1–3% 0.99 (0.97–1.01) 0.19 ≥3% 0.97 (0.94–0.99) 0.01

Non-cardiac Charlson 0 1.00 - 1 0.94 (0.92–0.97) <.01 2 0.95 (0.92–0.97) <.01 3+ 0.94 (0.91–0.97) <.01

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 43 www.nzma.org.nz/journal ARTICLE

Competing interests: Dr Grey reports grants from Heart Foundation and Healthier Lives (National Science Chal- lenge) during the conduct of the study. Dr Kerr reports grants from Health Research Council during the conduct of the study. Acknowledgements: ANZACS-QI programme implementation, coordination and analysis: The ANZACS-QI software was developed and supported by Enigma Solutions. Programme implementation is coordinated by the National Institute for Health Innovation (NIHI) at the University of Auckland. The ANZACS-QI programme is funded by the New Zealand Ministry of Health. We thank the the National Health Board Analytic Services and PHARMAC for enabling use of the national datasets. We also thank the VIEW team at the School of Population Health, Univer- sity of Auckland, for the curation and linkage of the national data. ANZACS-QI Governance group: Andrew Kerr (chair), Dean Boddington, Gary Sutcliffe, Gerry Devlin, Harvey White, John Edmond, Jonathon Tisch, Kim Marshall, Mayanna Lund, Michael Williams (deputy chair), Nick Fisher, Seif El Jack and Sue Riddle. ANZACS-QI Project management: Kristin Sutherland (Project Manager), Charmaine Flynn (Northern coordinator) and Maxine Rhodes (Southern coordinator). Data analysis: Mildred Lee. Editorial assistant: Julia Kerr. Data management: Billy Wu (SOPH), Michelle Jenkins (NIHI) and John Faatui (NIHI). We acknowledge all the New Zealand cardiologists, physicians, nursing staff, radiographers and patients who have supported and contributed to ANZACS-QI. Author information: Aravindra Muniandy MBChB: Cardiology Advanced Trainee, Counties Manukau District Health Board, New Zealand. Mildred Lee MSc: Biostatistician, Counties Manukau District Health Board, New Zealand. Corina Grey PhD: Public Health Physician, Auckland District Health Board, New Zealand. Katherine Ferrier: MBChB, Cardiologist, Hutt Valley District Health Board, New Zealand. Andrew J Kerr MD: Cardiologist, Counties Manukau District Health Board; Adjunct Associate Professor of Medicine, University of Auckland, New Zealand. Corresponding author: Andrew Kerr, Cardiologist, Counties Manukau District Health Board; Adjunct Associate Professor of Medicine, University of Auckland, New Zealand [email protected] URL: www.nzma.org.nz/journal-articles/demographic-differences-in-the-initiation-and-maintenance-of-statins-in-the-first-year-post-acs-in-new-zealand-a-data-linkage-study-anzacs-qi-57

REFERENCES 1. Grey C, Jackson R, Wells S, 3. Cholesterol Treatment Marshall, Sue Crengle, Jeff Marshall R, Riddell T, Kerr Trialists’ (CTT) Collabora- Harrison, Tania Riddell, AJ. Twenty-eight day and tors. The effects of lowering Andrew Kerr. Maintenance one-year case fatality after LDL cholesterol with of statin use over 3 years hospitalisation with an statin therapy in people following acute coronary acute coronary syndrome: at low risk of vascular syndromes: a national data a nationwide data linkage disease: meta-analysis linkage study (ANZACS- study. Aust N Z J Public of individual data from QI-2). Heart 10.1136/ Health. 2014; 38:216-20. 27 randomised trials. heartjnl-2013-304960. 2. Cholesterol Treatment Lancet. 2012; 380:581-90. 6. Bryson CL, Au DH, Young Trialists’ (CTT) Collabora- 4. Ho PM, Bryson CL, B, McDonell MB, Fihn SD. A tors. Efficacy and safety Rumsfeld JS. Medication refill adherence algorithm of cholesterol-lowering Adherence: Its Impor- for multiple short intervals treatment: prospective tance in Cardiovascular to estimate refill compli- meta-analysis of data from Outcomes. Circulation. ance (ReComp). Medical 90,056 participants in 14 2009; 119:3028-35. Care. 2007; 45:497-504. randomised trials of statins. 5. Corina Grey RJ, Sue Wells, 7. Kerr AJ, Looi JL, Garofalo Lancet. 2005; 366:1267-78. Simon Thornley, Roger D, Wells S, McLachlan

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 44 www.nzma.org.nz/journal ARTICLE

A. Acute Predict: a clini- 15. Salmond C, Crampton P, S, et al. Ethnic Differences cian-led cardiovascular Atkinson J. NZDep2006 in Coronary Revasculari- disease quality improve- index of deprivation. sation following an Acute ment project (Predict-CVD Wellington: Department of Coronary Syndrome 12). Heart, Lung & Circu- Public Health, Wellington in New Zealand: A lation. 2010; 19:378-83. School of Medicine and National Data-linkage 8. Kerr A, Williams MJ, Health Sciences., 2007. Study (ANZACS-QI 12). White H, et al. The All 16. Grey C, Jackson R, Wells S, Heart, Lung & Circula- New Zealand Acute et al. Maintenance of statin tion. 2016; 25:820-8. Coronary Syndrome use over 3 years following 23. Kerr A, Lee M, Grey C, et Quality Improvement acute coronary syndromes: al. Acute reperfusion for Programme: Implemen- a national data linkage ST-elevation myocardial tation, Methodology and study (ANZACS-QI-2). infarction in New Zealand Cohorts (ANZACS-QI 9). Heart. 2014; 100:770-4. (2015-2017): patient and New Zealand Medical 17. Thornley S, Marshall R, system delay (ANZACS-QI Journal. 2016; 129:23-36. Chan WC, et al. Four out of 29). New Zealand Medical 9. Kerr AJ, Mustafa A, ten patients are not taking Journal. 2019; 132:41-59. Lee M, et al. Ethnicity statins regularly during the 24. Ryan R, Santesso N, Lowe and revascularisation 12 months after an acute D, et al. Interventions following acute coronary coronary event. European to improve safe and syndromes: a 5-year cohort Journal of Preventive effective medicines use by study (ANZACS-QI-3). Cardiology. 2012; 19:349-57. consumers: an overview New Zealand Medical 18. Bruckert E, Hayem G, of systematic reviews. Journal. 2014; 127:38-51. Dejager S, Yau C, Begaud B. Cochrane Database 10. Thygesen K, Alpert JS, Mild to moderate muscular Syst Rev. 2014. Jaffe AS, Simoons ML, symptoms with high-dosage 25. McLachlan A, Doolan-Noble Chaitman BR, White HD. statin therapy in hyperlip- F, Lee M, McLean K, Kerr Third universal definition idemic patients--the PRIMO AJ. The electronic tracking of myocardial infarction. study. Cardiovasc Drugs of referral and attendance European Heart Journal. Ther. 2005; 19:403-14. at cardiac rehabilitation in 2012; 33:2551-67. 19. Nichols GA, Koro CE. Does Counties Manukau Health: 11. Thygesen K, Alpert JS, statin therapy initiation a potential model for New White HD. Universal increase the risk for Zealand. New Zealand Medi- definition of myocardial myopathy? An observation- cal Journal. 2016; 129:64-71. infarction. European Heart al study of 32,225 diabetic 26. Hamblin R, Shuker C, Journal. 2007; 28:2525-38. and nondiabetic patients. Stolarek I, Wilson J, Merry 12. Ministry of Health. Clin Ther. 2007; 29:1761-70. AF. Public reporting of National Health Index data 20. Grey C, Jackson R, Wells S, health care performance dictionary (version 5.3). et al. Trends in ischaemic data: what we know Ministry of Health 2009. heart disease: patterns and what we should do. New Zealand Medical 13. Granger CB, Goldberg RJ, of hospitalisation and Journal. 2016; 129:7-17. Dabbous O, et al. Predictors mortality rates differ by of hospital mortality in ethnicity (ANZACS-QI 27. Kerr A, Shuker C, Devlin the global registry of acute 21). New Zealand Medical G. Transparency in the coronary events. Archives Journal. 2018; 131:21-31. year of COVID-19 means of Internal Medicine. 21. Simpson SH, Eurich DT, tracking and publishing 2003; 163:2345-53. Majumdar SR, et al. A performance in the whole of the health system: 14. Ministry of Health. meta-analysis of the associ- progress on the public Ethnicity data protocols ation between adherence to reporting of acute coronary for the Health and Disabil- drug therapy and mortality. syndrome data in New ity Sector. Wellington, BMJ. 2006; 333:15. Zealand. NZMJ. accepted New Zealand, 2004. 22. Grey C, Jackson R, Wells for publication; 2020.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 45 www.nzma.org.nz/journal ARTICLE

The outcomes of patients with newly diagnosed neovascular age-related macular degeneration in Palmerston North Aaron Yap, Adeline Kho, John Ah-Chan

ABSTRACT AIM: To describe the outcomes of patients with newly diagnosed neovascular age-related macular degeneration (nAMD) treated in Palmerston North over the past two years. METHOD: A large prospective database was developed to capture the treatment and visual outcomes of patients with newly diagnosed nAMD. Data were subsequently extracted and analysed according to the result-based accountability (RBA) framework. RESULTS: Fifty-three patients in 2018 and 40 patients in 2019 were identified as having newly diagnosed nAMD. On average, there was an improvement in duration between the date of triage and the first intravitreal injection by eight days (22.7 vs 14.3 days), thus meeting national guidelines to assess and treat new referrals within 14 days. The total number of injections for the 2018 cohort was 227 compared to 301 in 2019. The percentage of patients achieving stabilisation of vision (15 letters vision loss, 82.5% vs 93.2%) and improvement in vision (15 letters gain, 10.5% vs 31.8%) was higher in 2019 compared to 2018. The percentage of patients that retained driving standard (ie, visual acuity of 6/12) was similar across both years (58.3% vs 62.5%). CONCLUSION: Patients receiving treatment for newly diagnosed nAMD in Palmerston North were achieving high rates of stabilisation and improvements in visual acuity, with more than half maintaining the national driving standard. The locally developed prospective database allows for real-time analysis of patient outcomes and the evaluation of the effectiveness of quality-improvement strategies.

phthalmology services worldwide the service provider and patient. Landmark are facing a growing burden of trials, such as the Comparison of Age-relat- Ochronic eye conditions. Age-related ed Macular Degeneration Treatments Trials macular degeneration (AMD) is the most (CATT) and VEGF Trap-Eye: Investigation of common cause of visual impairment in Efficacy and Safety in Wet AMD trial (VIEW1 older adults in the developed nations, and and VIEW2), have demonstrated the safety New Zealand is no exception. Neovascular and efficacy of bevacizumab and afliber- age-related macular degeneration (nAMD) cept.1,2 Following a loading regimen of three results in rapid loss of central vision if left monthly doses, the medication can continue untreated. The advent of intravitreal an- to be administered pro-re-nata (PRN), or ti-vascular endothelial growth factor (VEGF) at gradually extending intervals (treat and therapy, such as bevacizumab (Avastin®) extend) in order to maintain vision gained and aflibercept (Eylea®), has revolution- from treatment. Hence, successfully treated ised the treatment of nAMD but comes at a patients require long-term monitoring and considerable treatment burden and cost for treatment.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 46 www.nzma.org.nz/journal ARTICLE

Delays in diagnosis and treatment can flow through clinic without the need for lead to irreversible scarring of the macula assessment. They also allow for same-day and permanent vision loss. Therefore, in urgent injections. The nurse-led macular 2019 the Ministry of Health (MOH) and the review clinic was introduced to facilitate Royal Australian and New Zealand College the timely review of stable patients with of Ophthalmologists (RANZCO) introduced a view to overseeing maintenance intrav- national guidelines for the management of itreal therapy in 2016. In 2019, these were nAMD. The guidelines recommended that modified into ‘hybrid clinics’ to incorporate suspected nAMD patients be assessed within same-day injections. Under the remote one week of referral and treated within one supervision of an ophthalmologist, a trained week of first assessment.3 nurse specialist reviews each patient’s Provincial ophthalmology centres in New medical chart, OCT macula scans and colour Zealand have less staffing and less resource fundus photographs and formulates the capacity compared with their urban patient’s ongoing treatment plan, as per counterparts and need to adopt creative the treat-and-extend protocol, before the strategies in order to meet the needs of an patient receives their injection that day. The increasingly older population and a rising interval between injections is then adjusted prevalence of AMD. Continuous review of according to their clinical response. The key measures of performance is critical for treat-and-extend protocol is associated with focused planning around service delivery fewer patient visits, fewer injections and and to adapt to the needs of patients under lower overall medical costs compared with our care. fixed monthly injections or the pro-renata (PRN) protocol.5 Palmerston North Eye Department (PNED) operates under the MidCentral District This article describes the outcomes Health Board (DHB) to deliver compre- of patients with nAMD in PNED for the hensive ophthalmic care to the central North past two years. Outcome measures were Island of New Zealand, servicing a popu- selected according to the principles of lation of 178,820 people during weekdays the results-based accountability (RBA) and 243,370 people during weekends framework. (including Whanganui DHB).4 Compared to other DHBs, MidCentral DHB caters to a Methods population that is older and more socioeco- A prospective, multi-user database was nomically deprived. developed in 2017 to capture the treatment Several initiatives have been implemented details and visual outcomes of patients by PNED in the past five years to improve with nAMD in Palmerston North. Dates, outcomes for patients with nAMD. Collab- diagnoses, best corrected visual acuities orative care with optometrists and nurses (BCVA) and treatments were entered by was actively fostered and encouraged. The the attending nurse or clinician into the Acute Macula Clinic, introduced in 2019, database following each visit. Data for prioritises nAMD referrals for review and the newly diagnosed nAMD patients from initial treatment within two weeks and thus January 2018 to December 2019 were avoids the usual delays to first specialist extracted from the database. Patients’ appointments. In order to increase accuracy medical charts and electronic records were and facilitate timeliness of reviews, only also retrospectively reviewed to corroborate optometrist referrals with supporting optical data. Visual acuity was entered in Snellen coherence topography (OCT) scans are format but converted to Logarithm of the accepted into this clinic. Minimum Angle of Resolution (LogMar) for The upskilling of nurse injectors to analysis. safely administer intravitreal injections The primary outcomes were BCVA at increased clinicians’ capacity for more baseline and at the last recorded visit. complex clinical duties and improved skill– Secondary outcomes included the number task alignment. PNED nurse injectors run of intravitreal injections administered over high-volume clinics in which patients are the treatment period and the time between booked for injections only, streamlining the the triage date and treatment initiation.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 47 www.nzma.org.nz/journal ARTICLE

Stabilisation of vision was defined as a drop of fewer than 15 letters on the Snellen Discussion chart during the treatment period, while The burden of chronic disease is an improvement of vision was defined as a gain emerging health issue in developed of more than 15 letters. countries with ageing populations. The proportion of the New Zealand popu- Statistical analysis was conducted on lation aged 65 years and over will double Microsoft Office Excel version 16.29.1 by the year 2040, and health expen- (19091700) and Apple Numbers (© Apple diture as a percentage of gross domestic Inc. version 10.2 (7028.0.88)). Ethical product is expected to increase from 6% approval was granted by the MidCentral to 9% within that timeframe.6 Age-related DHB Research Office. macular degeneration is the leading cause of vision loss in people aged over 50 years Results old in New Zealand. A national study has Fifty-three patients were diagnosed and projected that the prevalence of AMD will treated for nAMD in 2018, compared to 40 rise to 208,000 in 2026.7 Early diagnosis and patients in 2019. Despite fewer patients, the regular treatment in the form of intrav- total number of injections rose from 227 in itreal anti-vascular endothelial growth 2018 to 301 in 2019. There was an increase factor (anti-VEGF) injections reduces vision in the percentage of patients achieving stabi- loss but imposes a significant burden on lisation (82.5% vs 93.2%) and improvement healthcare resources. of vision (10.5% vs 31.8%) in 2019 compared Approximately 100 patients are seen daily to 2018. The percentage of patients that in the Palmerston North Eye Department. retained driving standard, defined as BCVA The clinic faces increasing demands, an better than or equal to 6/12, was similar ongoing workforce shortage and finite across both groups (58.3% vs 62.5%). resources. A collaborative, integrated Although the average number of days team-care approach is undertaken to between date of triage and date of first provide effective, comprehensive and appointment is similar between both years, coordinated care in the most efficient way, the difference between the average number making the best use of technology and of days between triage date and first ‘lean-principle improvement methodology’ injection reduced by approximately eight to continuously improve current systems days in 2019 compared to 2018. and processes.

Table 1: Outcome measures of newly referred nAMD patients 2018–2019.

Outcome measures 2018 2019 Number of patients receiving treatment 53 patients (57 eyes) 40 patients (44 eyes) for wet macular degeneration

Number of injections administered 227 injections 301 injections

Number of days between triage and ap- 12.7 days 14.6 days pointment date

Number of days between triage and date 22.7 days 14.3 days of first injection

Percentage of patients achieving stabilisa- 82.5% 93.2% tion of vision

Percentage of patients achieving improve- 10.5% 31.8% ment of vision

Percentage of patients that maintained 58.3% 62.5% driving standard (BCVA 6/12)

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 48 www.nzma.org.nz/journal ARTICLE

The outcome of this paper completes the interdependent networks with decen- audit cycle that began five years ago.8,9 The tralised control and non-linear cause and number of new patients referred with wet effect relationships.10 There exists leverage AMD has increased. There was a total of 48 points or ‘sweet spots’ within such rela- eyes commencing treatment for wet AMD tionships where small changes can have in 2013 and 2014 combined, compared to 93 disproportionately significant effects. The new patients in 2018 and 2019.9 Despite this, development of a robust database facilitates the number of intravitreal injections admin- regular monitoring and reporting using istered for this cohort has not increased.8 advanced data analytics that can identify The number of patients achieving stabili- these leverage points in real time. Such sation and improvement of vision in the past health-information systems are required to two years has improved compared to 2013, support decision-making, as unfortunately during which time the stabilisation rate was even interventions supported by robust 81.3% and improvement rate was 25%.9 The clinical trials do not always translate into rates in 2019 were similar to the Comparison similar clinical outcomes in the real-world of Age-related Macular Degeneration Treat- clinical setting. This is due to the fact that ments Trials, which showed a 95% rate of the empirical reductionist approach (clinical stabilisation and 34% rate of improvement trials) does not replicate the complex with monthly injections.1 adaptive nature of contemporary healthcare The Royal Australia and New Zealand delivery systems. College of Ophthalmologists best practice Results based accountability is a guidelines for management of neovascular framework that has been adopted by various AMD (nAMD) recommends that referred New Zealand organisations focusing on patients have their first appointment within outcomes to make a positive change in their one week and that they commence treatment community.11 Performance-accountability within two weeks of referral. 3 Although measures are centred around three key ques- our cohort received their first appoint- tions: How much did we do? How well did ments more than one week after being we do it? Is anyone better off? This approach referred, each patient still received their acknowledges the complexity of the current first treatment within the recommended system and provides real-time insights timeframe because the Acute Macular Clinic onto the added value and effectiveness of combines their first appointment with proposed quality-improvement strategies. treatment initiation. This initiative shortened Variable compliance to data entry and the duration between the date of triage and inconsistent coding poses the greatest risk the date of first injection in 2019 compared to the data quality. Robust data quality is to the previous year. Other factors that lead essential for effective data-driven deci- to delays include a lack of awareness of AMD sion-making. Data integrity, quality of symptoms within the community and delays information and health information systems in the referral pathways. The latter aspect result in sound clinical decision-making has been addressed by encouraging direct and improved quality of healthcare. Poor phone call referrals from optometrists for data quality and health information systems nAMD. contribute to inefficiencies, waste, variation Palmerston North Eye Department, like and harm.12 The two identifiable factors many healthcare services, is a complex that affect data quality are the busy clinical adaptive system. There has been a environment and regular staff turnover. ‘paradigm shift’ in the methodological Hence current efforts are directed at making approach of improving healthcare service the database more user-friendly and less delivery and performance. There has been time-consuming. a transition from the traditional ‘empirical In conclusion, the outcome measures scientific (reductionist) approach’ to a more generated by the AMD database and asso- ‘systems approach’. Instead of breaking ciated advanced data analytics allow up the system into its components and for data driven, evidence-informed performing an analysis of the individual decision-making. This has resulted in parts, the systems approach emphasises service-improvement initiatives and strat- the interactions between the component egies that have led to an improvement in parts of the system. Each complex system visually significant outcomes for patients is comprised of a collection of intersecting, being treated for nAMD and the community.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 49 www.nzma.org.nz/journal ARTICLE

Competing interests: Nil. Acknowledgements: We thank Greg Bolton, Mike Yang, Paul Greatorex and Rahul Alate (Data Analytics and Business Advisory) and Dr Grieg Russell (Digital Services) for their help with the AMD database and data analytics. We thank the staff of the Palmerston North Eye Department for maintaining the database. Author information: Aaron Yap, MBChB: Palmerston North Eye Department, MidCentral District Health Board. Adeline Kho, MBChB, PGDipBSOphth: Waikato Hospital Eye Clinic, Waikato District Health Board. John Ah-Chan, FRANZCO, AFRACMA: Palmerston North Eye Department, MidCentral District Health Board. Corresponding author: Aaron Yap, Palmerston North Eye Department, 50 Ruahine Street, Roslyn, Palmerston North 4414, 0211215528 (phone), 06-3508644 (fax) [email protected] URL: www.nzma.org.nz/journal-articles/the-outcomes-of-patients-with-newly-diagnosed-neovascular-age-related-macular-degeneration-in-palmerston-north

REFERENCES 1. Group CR. my-dhb/midcentral-dhb/ Endothelial Growth Factor Ranibizumab and beva- population-midcentral-dhb. Treatment through an Inte- cizumab for neovascular 5. Gupta OP, Shienbaum grated Collaborative Team age-related macular degen- G, Patel AH, Fecarotta C, Care Approach. Journal of eration. New England Kaiser RS, Regillo CD. A Clinical & Experimental journal of medicine. treat and extend regimen Ophthalmology. 2018;09. 2011;364(20):1897-908. using ranibizumab for 9. Botha VE, Ah-Chan 2. Heier JS, Brown DM, neovascular age-related JJ, Ramachandran N. Chong V, Korobelnik J-F, macular degeneration: Improving accessibility Kaiser PK, Nguyen QD, et clinical and economic to intravitreal anti-vas- al. Intravitreal aflibercept impact. Ophthalmology. cular endothelial growth (VEGF trap-eye) in wet 2010;117(11):2134-40. factor treatment for age-related macular 6. Frizelle F. Health expen- ophthalmic patients in a degeneration. Ophthalmol- diture and the ageing peripheral centre. NZ Med ogy. 2012;119(12):2537-48. population. The New J. 2016;129(1445):56-66. 3. RANZCO. Referral Pathway Zealand Medical Journal 10. Plsek PE, Wilson T. for AMD Management 2020 (Online). 2005;118(1208). Complexity, leadership, [Available from: https:// 7. Worsley D, Worsley A. and management in ranzco.edu/wp-content/ Prevalence predictions healthcare organisations. uploads/2020/01/080 for age-related macular BMJ. 2001;323(7315):746-9. 120-RANZCO-Referral-path- degeneration in New 11. Friedman M. Results-based way-for-AMD-manage- Zealand have implications Accountability: Producing ment-revised.pdf. for provision of healthcare Measurable Improve- 4. Ministry of Health. services. The New Zealand ments for Customers and Population of MidCentral Medical Journal (Online). Communities: OECD; 2009. DHB 2019 [Available 2015;128(1409):44. 12. Coiera E. Guide to from: https://www. 8. Kim J, Ah-Chan J, Russell G. health informatics: health.govt.nz/new-zea- Improving Patient Access to CRC press; 2015. land-health-system/ Intravitreal Anti-Vascular

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 50 www.nzma.org.nz/journal ARTICLE

Comparison of physical activity patterns across large, medium and small urban areas and rural settings in the Otago Region, New Zealand Brittany White, Enrique García Bengoechea, John C Spence, Kirsten J Coppell, Sandra Mandic

ABSTRACT AIM: This study compared accelerometer-measured physical activity (PA) patterns in adolescents living in diverse urban and rural areas of Otago, New Zealand. METHOD: Participants (n=377; age: 14.9±1.4 years; 66.8% female; 23 schools) completed an online school travel survey, anthropometry and seven-day PA accelerometer assessment. Participants resided in large (n=237), medium (n=45) and small (n=44) urban areas or rural settings (n=51). RESULTS: Overall, participants participated in 54.4±21.0 minutes of moderate-to-vigorous physical activity (MVPA) daily and 35.0% met PA guidelines (school day vs weekend day: 40.8% vs 26.0%; p<0.001) with no difference across geographical settings. A greater proportion of males (43.2% vs 31.9%; p=0.016), school sport participants (70.1% vs 54.0%; p=0.005) and active-transport-to-school users (40.2% vs 26.1%) met PA guidelines compared to their counterparts. Compared to rural adolescents, those from large urban areas accumulated more MVPA during the school commute time (before school: 8.3±6.7 vs 5.3±3.8 minutes, p<0.001; after school: 10.1±6.0 vs 7.7±4.3 min, p=0.003), but overall spent more time sedentary (584.9±84.7 vs 527.8±88.2 minutes/day; p<0.001). CONCLUSION: PA in Otago adolescents is low, with significant differences by gender, sport participation, mode of travel to school and geographical setting. Increased PA should be encouraged in both urban and rural adolescents.

dolescents’ physical activity (PA) physical activity (MVPA) daily.2 However, is influenced by an interaction of recent evidence suggests there are high rates individual, social and environmen- of physical inactivity and time spent station- A 1 3 tal factors. Societal changes over recent ary among adolescents globally, including decades have markedly reduced the need New Zealand.4 Worldwide, 81% of 11–17 and opportunities for PA in daily life and im- year olds did not meet current guidelines for posed multiple barriers to PA across all PA recommended levels of physical activity in domains: transport, leisure and recreation, 2016, while 89% of New Zealanders in this school/workplace and household. In adoles- age group did not achieve recommended cents, PA has been associated with improved levels.3 cardiorespiratory fitness, muscular endur- Most studies examining adolescent PA ance and strength and the prevention of have been conducted in urban areas. 2 non-communicable diseases. To gain health However, due to contextual differences (eg, benefits, adolescents need to accumulate at population density, access to facilities, social least 60 minutes of moderate-to-vigorous and cultural norms), PA patterns observed

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 51 www.nzma.org.nz/journal ARTICLE

among urban adolescents may not be gener- Participants alisable to rural adolescents. Studies that Briefly, each participating secondary compare device-measured PA in adolescents school invited adolescents from one to four living in urban versus rural settings show classes from years 9 to 13 (ages 13 to 18 inconsistent results. Although most report years) to participate in the study. In small that urban adolescents accumulate more schools of <100 students, all students were accelerometer-measured MVPA compared invited. Potential participants received 5,6 to rural adolescents, one study in the information packages with parent and United States found that PA was higher student information sheets and consents. In in adolescents living in rural compared the BEATS Study, parental opt-in or parental 7 to urban areas. Similarly, inconsistent opt-out consent was used depending on each results have also been reported for time school’s preference for adolescents aged 13 spent in sedentary pursuits. Compared to to 15 years, whereas for the BEATS-R Study rural adolescents, urban adolescents spent no parental consent was required. In both 8 more time being sedentary in Australia, studies, participants were required to sign 9 but less time in Kenya, and no difference an additional consent to wear an acceler- 6 was reported for Canadian and Portuguese ometer. This analysis included data from 5 adolescents. These inconsistent findings 377 adolescents with valid survey and accel- may reflect urban–rural differences in social erometer data (Figure 1). norms, recreational opportunities and infra- Given their school and home address structure related to PA, country-specific PA location, participants were classified into characteristics and/or differences in how four geographical setting categories using rurality is defined. Stats NZ definitions:10 Understanding PA patterns in different • large urban area (30,000–99,999 geographical settings is necessary to inform residents; ie, Dunedin city excluding country-specific initiatives and programmes Mosgiel) aimed at increasing the low levels of PA among adolescents. New Zealand adoles- • medium urban area (10,000–29,999 cents live in a range of settings,10 and this residents) study compared accelerometer-measured • small urban area (1,000–9,999 PA patterns in adolescents living in large, residents) medium and small urban areas and rural • rural setting (<1000 residents). settings in the Otago region. The BEATS and BEATS-R studies did not collect data from participants living in major Method urban areas (>100,000 residents).10 Only Study design participants residing and attending school in Secondary data were analysed from two the same geographical setting were included cross-sectional studies conducted in the in this analysis (ie, participants residing in urban and rural areas of the Otago region, rural areas but attending a boarding school New Zealand. Dunedin city is the only large were excluded) (Figure 1). urban area, and the wider Otago region Measurement procedures (population of approximately 225,186) consists of medium urban, small urban and Questionnaire Participants completed an online survey rural areas.10 The 2014–15 Built Environment during one school period (50–60 minutes) and Active Transport to School (BEATS) under the supervision of research staff. The Study was conducted in all 12 Dunedin survey included questions about sociode- secondary schools. The 2018 BEATS Rural mographic characteristics, home address, Study (BEATS-R) was conducted in 11 of 15 number of vehicles and bicycles at home, regional secondary schools. Both studies used transport-to-school habits and sport partic- the BEATS research methodology, which has ipation. Participants reported frequency been described in detail elsewhere,11 with of use of different modes of transport to data being collected throughout the school school. The mode of transport used ‘most year. Both studies were approved by the of the time’ and ‘all of the time’ informed University of Otago Human Ethics Committee their transport to school category: ‘active (BEATS: 13/203; BEATS-R: 17/178).

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 52 www.nzma.org.nz/journal ARTICLE

Figure 1: Flowchart of participant recruitment and selection for the final study sample.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 53 www.nzma.org.nz/journal ARTICLE

transport’, ‘motorised transport’ or ‘mixed before school (08:00–09:00h), early after transport’.11 They also reported whether school (15:00–16:00h) and late after school they participated in sports at school and (16:00–20:00h) on school days.13 outside school (eg, club sport) with ‘yes’/’no’ responses to each question. Neighbour- Data analysis Demographic characteristics were hood-level socioeconomic status was analysed using descriptive statistics. determined by the home address data Continuous variables were checked for being matched with address codes from the normality and showed skewness values ≤2 New Zealand Index of Deprivation Study.12 and kurtosis (excess) ≤4, which are indic- Geographic Information Science network ative of considerable normality for sample analysis was used to calculate the shortest sizes >300,15 with the exception of only distance to school from each participant’s five variables (distance to school in the home address.11 total sample; average daily sedentary time; Anthropometry average moderate PA before school; average Research staff performed anthropometry vigorous PA before and late after school). measurements in a screened-off area of For continuous variables, differences across the classroom using standard procedures.11 geographical settings were compared using Height was measured using a stadiometer ANOVA with Scheffe post-hoc multiple (BEATS: custom-made portable stadiometer; comparisons (or Tamahane’s T2 test, when BEATS-R: portable (SECA213 stadiometer, the assumption of homogeneity of variance SECA Corp)) and weight using an electronic was violated). Categorical variables were scale (A&D scale UC321, A&D Medical). compared using χ2-test. Differences between school days versus weekend days were Accelerometer-measured PA compared using paired t-test for continuous Participants wore an accelerometer (Acti- variables and McNemar test for cate- Graph, GT3XPlus, Pensacola, FL, USA) above gorical variables. Continuous variables are their right hip for seven consecutive days, reported as means ± standard deviation, as described elsewhere.11,13 Briefly, research whereas categorical variables are reported staff instructed them to wear their device as frequencies (n (%)). An alpha of less than for ≥12 hours each day for seven days, and 0.05 was considered statistically significant. to take it off for sleep, water-based activ- Data were analysed using SPSS software ities (eg, swimming) and contact sports (eg, (Version 24.0). rugby). To promote adherence, participants were given an activity log to record their accelerometer wear time, sent reminders Results by email or text and received a $10 book Table 1 shows the sociodemographic char- voucher. acteristics of study participants. Among the Accelerometer data were downloaded in 377 participants (age: 14.9±1.4 years), 66.8% 10-second epochs for the BEATS Study and were female and 75.5% were New Zealand 15-second epochs for the BEATS-R Study European. Average distance to school was using ActiGraph software and measured 4.5±5.1 km (median: 3.0km; interquartile in average counts per minute (cpm).13 The range: 4.3km). Age, gender, ethnicity and wear-time validity was set at ≥5 days, with sport participation rates were not statis- ≥10 h/day (inclusive of three school days tically different between geographical and one weekend day).13 Accelerometer settings. Neighbourhood-area deprivation data were analysed by the MeterPlus data level, car and bicycle ownership, home-to- analysis service in San Diego, USA, using school distance and rates of active transport MeterPlus software (MeterPlus, San Diego, to school varied across the geographical CA, USA) with Evenson’s cut-points.14 Twen- settings, with the longest median distance to ty-minute stationary bouts were classified as school and lowest rates of active transport non-active periods. Processed accelerometer observed in rural settings. data included time spent in light, moderate, Throughout the week, participants spent moderate-to-vigorous and vigorous intensity on average 9.5 hours (69.2%) of the daily PA and sedentary time for an average day, accelerometer wear time in stationary a weekend day and a school day, as well as pursuits, 3.5 hours (24.5%) in light intensity

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 54 www.nzma.org.nz/journal ARTICLE

Table 1: Sociodemographic characteristics of study participants.

Total Large Medium Small Rural p-value sample urban urban urban setting n=377 area area area n=237 n=45 n=44 Age (years) 14.9 ± 1.4 14.8 ± 1.5 15.1 ± 1.4 15.1 ± 1.1 15.1 ± 1.5 0.223

Female gender 252 (66.8) 164 (69.2) 28 (62.2) 30 (68.2) 30 (58.8) 0.468 [n(%)]

Ethnicity [n(%)] n=375 n=236 n=45 n=44 n=50 New Zealand 283 (75.5) 170 (72.0) 37 (82.2) 32 (72.7) 44 (88.0) European

Māori 27 (7.2) 17 (7.2) 2 (4.4) 3 (6.8) 5 (10.0)

Other 65 (17.3) 49 (20.8) 6 (13.3) 9 (20.5) 1 (2.0) 0.067

Neighbourhood n=363 n=235 n=33 n=44 n=51 deprivation score [n(%)] 1 (least deprived) 106 (29.2) 73 (31.1) 4 (12.1) 16 (36.4) 13 (25.5)

2 92 (25.3) 53 (22.6) 11 (33.3) 8 (18.2) 20 (39.2)

3 69 (19.0) 38 (16.2) 7 (21.2) 15 (34.1) 9 (17.6)

4 64 (17.6) 43 (18.3) 8 (24.2) 4 (9.1) 9 (17.6)

5 (most deprived) 32 (8.8) 28 (11.9) 3 (9.1) 1 (2.3) 0 (0.0) 0.004

Number of bikes available to use to get to school [n(%)] None 73 (19.4) 62 (26.2) 4 (8.9) 1 (2.3) 6 (11.8)

One 86 (22.8) 57 (24.1) 7 (15.6) 7 (15.9) 15 (29.4)

Two or more 218 (57.8) 118 (49.8) 34 (75.6) 36 (81.8) 30 (58.8) <0.001

Number of vehicles at home [n(%)] None 9 (2.4) 7 (3.0) 2 (4.4) 0 (0.0) 0 (0.0)

One 98 (26.0) 77 (32.5) 6 (13.3) 8 (18.2) 7 (13.7)

Two or more 270 (71.6) 153 (64.6) 37 (82.2) 36 (81.8) 44 (86.3) 0.005

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 55 www.nzma.org.nz/journal ARTICLE

Table 1: Sociodemographic characteristics of study participants (continued).

Total Large Medium Small Rural p-value sample urban urban urban setting n=377 area area area n=237 n=45 n=44

Sport participation At school 196 (59.8) 122 (62.2) 25 (59.5) 17 (42.5) 32 (64.0) 0.120

Outside school 170 (51.8) 100 (51.0) 17 (40.5) 26 (65.0) 27 (54.0) 0.165

Transport to school Motorised 201 (53.3) 140 (59.1) 18 (40.0) 16 (36.4) 27 (52.9)

Active 117 (31.0) 57 (24.1) 19 (42.2) 24 (54.5) 17 (33.3)

Mixed 59 (15.6) 40 (16.9) 8 (17.8) 4 (9.1) 7 (13.7) 0.002

Distance to 4.5 ± 5.1 4.4 ± 3.3 b,c,d 2.3 ± 1.7 a,d 2.1 ± 1.3 a,d 9.1 ± 10.5 a,b,c <0.001 school (km)

Meeting physical activity guidelines [n(%)] Average week 132 (35.0) 88 (37.1) 18 (40.0) 13 (29.5) 13 (25.5) 0.312

Average school 154 (40.8) 99 (41.8) 20 (44.4) 18 (40.9) 17 (33.3) 0.678 day

Average week- 98 (26.0) 64 (27.0) 10 (22.2) 12 (27.3) 12 (23.5) 0.883 end day

a p<0.05 vs large urban area; b p<0.05 vs medium urban area; c p<0.05 vs small urban area; d p<0.05 vs rural settlement or area.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 56 www.nzma.org.nz/journal ARTICLE

PA and 0.9 hours (6.6%) in MVPA per day (Table 2). Specifically, they participated in Discussion 54.4±21.0 minutes of MVPA daily (Table 2) Key findings of this study are: (1) Less than and 35.0% met PA guidelines (Table 1). A half of Otago adolescents meet PA guide- greater proportion of participants met PA lines overall; (2) A greater proportions of guidelines on school days (40.8%) versus adolescents meeting PA guidelines were weekend days (26.0%) (p<0.001). On school male, participated in school sports and used days, participants spent significantly more active transport to school; (3) Although the time in sedentary activities, moderate and proportion of adolescents meeting PA guide- vigorous PA as well as MVPA compared to lines was not significantly different across weekend days (all p<0.05). geographical settings, those in large urban areas spent more time being stationary but A greater proportion of males, school- accumulated more MVPA during the school sport participants and users of active commute time compared to their rural transport to school met PA guidelines on counterparts. average school days, compared to their counterparts (Table 3). However, on an Taken together, these findings show that average weekend day, no differences were engagement in PA in Otago adolescents is observed between participants who met PA lower than recommended for the majority guidelines and those who did not, except for and should be encouraged across all sports participation at school (Table 3). geographical settings, particularly during weekends. The average amount of daily MVPA (Table 3) or proportion of participants meeting PA Using device-measured PA, 35% of Otago guidelines (Table 1) did not differ across adolescents met the PA guidelines and on the four geographical setting categories. average engaged in 54.4±21.0 minutes of However, those living in rural areas spent MPVA per day with no significant differ- more time in light PA compared to their ences by geographical setting. These findings counterparts from large and medium urban are consistent with estimates of 27–33% of areas, and less time in stationary pursuits children and adolescents meeting recom- compared to those from large urban areas mended PA levels worldwide.4,16 However, (Table 2). Similar patterns were observed on if these New Zealand adolescents as a average school days. On weekend days, rural group were to increase their daily PA by six adolescents engaged in significantly more minutes (approximately 10% of their daily light PA compared to their urban counter- activity), they would meet the guidelines. parts, whereas no significant difference This seems like an achievable public health existed across geographical settings for target. the amount of time participants spent in Previous studies that compared PA across stationary pursuits. urbanisation settings reported higher levels PA patterns on school days also varied of MVPA in urban compared to rural adoles- across geographical settings. During the cents in Portugal,5 Canada17 and the United hour before school (8:00–9:00h), partici- States.18,19 Greater access to recreational pants living in large and medium urban facilities in urban compared to rural areas,20 areas spent significantly more time in and consequently better access to organised MVPA than those from small urban areas sports and recreational activities,21 may in and rural settings (Table 4). In addition, part explain these findings. However, in the participants from large and medium present study no difference in MVPA was urban areas spent significantly more observed among Otago adolescents living time in MVPA during early after school in the four geographical settings, although (15:00–16:00h) than those from rural areas; rural adolescents spent more time in light PA however, no difference was found for small compared to their urban peers. It is highly urban areas (Table 4). During the late after likely that rural adolescents in New Zealand school period (16:00–20:00h), participants have more access to open green spaces and living in urban areas spent significantly natural environments compared to their more time in stationary pursuits and less urban counterparts.22 Greenness exposure time in light PA compared to their rural has been associated with higher levels of counterparts (Table 4). MVPA among American children.23 Never-

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 57 www.nzma.org.nz/journal ARTICLE

Table 2: Physical activity levels throughout the week.

Total sample Large urban Medium urban Small urban Rural p-value n=377 area area area setting n=237 n=45 n=44 n=51

Average daily activity throughout the week per valid day Sedentary 572.0±85.1 584.9±84.7 d 570.7±70.3 554.8±80.1 527.8±88.2 a <0.001 activities (min)

Light PA (min) 199.2±52.1 187.9±47.8 d 202.4±42.2 d 211.2±53.2 238.4±57.7 a,b <0.001

Moderate PA 33.1±12.2 33.0±11.9 34.5±12.6 31.8±12.4 33.8±13.0 0.745 (min)

Vigorous PA 21.3±13.4 21.7±13.6 23.0±14.0 21.0±11.6 18.2±13.0 0.285 (min)

MVPA (min) 54.4±21.0 54.7±21.0 57.5±20.5 52.8±16.9 52.0±24.4 0.577

Meeting PA 132 (35.0) 88 (37.1) 18 (40.0) 13 (29.5) 13 (25.5) 0.312 guidelines [n(%)]

Average daily activity during a school day per valid day Sedentary 585.0±89.0 599.6±91.0c,d 584.3±72.1 d 563.1±71.9 a 537.0±87.9 a,b <0.001 activities (min)

Light PA (min) 197.1±51.4 187.5±49.2d 201.5±45.3 d 203.1±50.4 d 232.7±52.1 a,b,c <0.001

Moderate PA 34.9±12.3 34.7±11.9 36.6±14.2 34.2±13.5 35.0±11.7 0.786 (min)

Vigorous PA 22.9±14.4 23.2±14.4 25.0±15.9 22.6±11.5 19.9±15.1 0.355 (min)

MVPA (min) 57.8±21.4 57.8±21.1 61.6±22.8 56.8±17.0 54.9±24.7 0.489

Meeting PA 154 (40.8) 99 (41.8) 20 (44.4) 18 (40.9) 17 (33.3) 0.678 guidelines [n(%)]

Average daily activity during a weekend day per valid day Sedentary 533.5±113.9 544.4±106.6 523.1±97.3 523.5±148.2 500.8±113.9 0.068 activities (min)

Light PA (min) 207.5±77.3 189.9±63.4 c,d 207.3±62.2 d 241.9±92.0 a 259.6±100.1 a,b <0.001

Moderate PA 28.2±19.6 28.1±19.6 27.7±15.2 25.8±16.8 31.4±24.6 0.563 (min)

Vigorous PA 16.9±16.5 17.9±17.4 16.6±13.6 16.3±18.2 13.2±12.2 0.319 (min)

MVPA (min) 45.2±31.8 46.0±32.8 44.3±25.9 42.1±29.4 44.6±34.0 0.889

Meeting PA 98 (26.0) 64 (27.0) 10 (22.2) 12 (27.3) 12 (23.5) 0.883 guidelines [n(%)]

a p<0.05 vs. large urban area; b p<0.05 vs. medium urban area; c p<0.05 vs. small urban area; d p<0.05 vs. rural settlement or area. MVPA = moderate to vigorous physical activity; PA = physical activity.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 58 www.nzma.org.nz/journal ARTICLE

Table 3: Characteristics of participants who met and did not meet physical activity recommendations.

Average day p-value Average school day p-value Average weekend day p-value

Met PA Did not Met PA Did not Met PA Did not guide- meet PA guide- meet PA guide- meet PA lines guidelines guidelines guide- n=132 lines n=154 lines n=98 lines n=245 n=223 n=279

Age (years) 14.8±1.4 15.0±1.5 0.431 15.0±1.5 14.8±1.3 0.314 14.8±1.4 15.0±1.4 0.467

Gender [n(%)] Male 54 (43.2) 71 (56.8) 64 (51.2) 61 (48.8) 39 (31.2) 86 (68.8)

Female 78 (31.0) 174 (69.0) 0.019 90 (35.7) 162 (64.3) 0.004 59 (23.4) 193 (76.6) 0.105

Ethnicity n=132 n=243 n=153 n=222 n=98 n=277 [n(%)] New 97 (73.5) 186 (76.5) 112 (73.2) 171 (77.0) 73 (74.5) 210 (75.8) Zealand European

Māori 12 (9.1) 15 (6.2) 17 (11.1) 10 (4.5) 9 (9.2) 18 (6.5)

Other 23 (17.4) 42 (17.3) 0.572 24 (15.7) 41 (18.5) 0.048 16 (16.3) 49 (17.7) 0.664

Neigh- n=124 n=239 n=146 n=217 n=92 n=271 bourhood deprivation score [n(%)] 1 (least 25 (20.2) 81 (33.9) 37 (25.3) 69 (31.8) 19 (20.7) 87 (32.1) deprived)

2 37 (29.8) 55 (23.0) 39 (26.7) 53 (24.4) 32 (34.8) 60 (22.1)

3 29 (23.4) 40 (16.7) 30 (20.5) 39 (18.0) 19 (20.7) 50 (18.5)

4 20 (16.1) 44 (18.4) 27 (18.5) 37 (17.1) 16 (17.4) 48 (17.7)

5 (most 13 (10.5) 19 (7.9) 0.050 13 (8.9) 19 (8.8) 0.766 6 (6.5) 26 (9.6) 0.080 deprived)

Sport participation [n(%)] At school 82 (70.1) 114 (54.0) 0.005 94 (69.1) 102 (53.1) 0.004 60 (71.4) 136 (55.7) 0.011

Outside 69 (59.0) 101 (47.9) 0.054 77 (56.6) 93 (48.4) 0.144 51 (60.7) 119 (48.8) 0.059 school

Transport to school [n(%)] Motorised 55 (41.7) 146 (59.6) 60 (39.0) 141 (63.2) 51 (52.0) 150 (53.8) only

Active only 53 (40.2) 64 (26.1) 64 (41.6) 53 (23.8) 33 (33.7) 84 (30.1)

Mixed 24 (18.2) 35 (14.3) 0.003 30 (19.5) 29 (13.0) <0.001 14 (14.3) 45 (16.1) 0.781 modes

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 59 www.nzma.org.nz/journal ARTICLE

Table 4: Physical activity levels throughout an average school day in participants across geographical settings.

Total sample Large Medium Small Rural p-value n=377 urban area urban area urban area settings n=237 n=45 n=44 n=51

Average activity before school (8:00–9:00h) per valid day Sedentary 31.6±9.2 32.8±8.3 28.8±11.6 30.9±10.2 29.2±9.4 0.008 activities (min)

Light PA (min) 17.8±6.9 16.8±5.9 d 17.3±7.9 18.7±7.6 21.6±8.1 a <0.001

Moderate PA 4.4±3.7 4.7±3.9 4.6±3.8 3.7±3.3 3.5±2.4 0.125 (min)

Vigorous PA 3.1±4.2 3.7±4.6 c,d 3.7±4.5 c 1.5±1.6 a,b 1.8±2.5 a 0.001 (min)

MVPA (min) 7.5±6.2 8.3±6.7 c,d 8.3±6.2 c,d 5.2±3.7 a,b 5.3±3.8 a,b <0.001

Average activity early after school (15:00–15:00h) per valid day Sedentary 31.3±7.7 32.4±8.1 d 29.6±6.3 31.2±6.9 28.0±6.7 a 0.001 activities (min)

Light PA (min) 18.6±6.1 17.2±5.5 d 18.3±4.6 d 19.6±6.5 d 24.2±6.4 a,b,c <0.001

Moderate PA 5.8±3.8 5.7±3.9 6.8±4.4 5.5±3.1 5.2±2.9 0.209 (min)

Vigorous PA 4.1±3.8 4.4±3.8 d 5.1±4.3 d 3.7±3.8 2.5±2.5 a,b 0.002 (min)

MVPA (min) 9.9±5.8 10.1±6.0 d 11.9±6.1 d 9.2±4.5 7.7±4.3 a,b 0.003

Average activity late after school (16:00–20:00h) per valid day Sedentary 164.0±26.6 168.5±24.3 d 164.3±23.8 d 162.4±28.3 d 145.1±29.6 a,b,c <0.001 activities (min)

Light PA (min) 58.3±20.2 d 54.2±18.3 d 57.9±17.3 62.9±20.5 73.0±22.9 a,b <0.001

Moderate PA 8.4±5.4 8.0±5.0 8.5±4.8 8.0±5.6 10.2±7.2 0.072 (min)

Vigorous PA 6.0±6.4 5.9±6.1 6.5±6.7 5.4±6.1 7.1±7.6 0.529 (min)

MVPA (min) 14.4±10.6 13.9±9.8 15.0±10.3 13.4±10.7 17.3±14.0 0.186

a p<0.05 vs large urban area; b p<0.05 vs medium urban area; c p<0.05 vs small urban area; d p<0.05 vs rural settlement or area. MVPA = moderate to vigorous physical activity; PA = physical activity.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 60 www.nzma.org.nz/journal ARTICLE

theless, the PA levels in Otago adolescents with the time adolescents spent partici- are low in both urban and rural areas, rein- pating in sport.26 Thus, future initiatives forcing the need for effective interventions and programmes designed for promoting to encourage PA in all New Zealand adoles- adolescents’ PA through organised sport cents, irrespective of setting. In addition, participation should focus on providing given the limitations of self-reports,24,25 well-organised and supportive sport envi- more research employing direct measures ronments in both urban and rural areas of (eg, accelerometers, pedometers, inclinom- New Zealand. eters) of the PA and sedentary behaviour of Adolescents in this study spent on average New Zealand youth will further contribute 9.5 hours being stationary. Those from large to understanding how patterns of such urban areas spent a significantly higher behaviour interact in diverse geographical proportion of time in stationary pursuits locations across the country. (particularly in the time periods early after A greater proportion of adolescents who school and late after school) compared to met PA guidelines in this study were male, their peers from rural areas. Given that light used active transport to school and partic- PA is considered the antidote to sedentary ipated in school sports. These findings are behaviour,29 children in rural areas in consistent with those reported elsewhere.1,26 New Zealand may be avoiding sedentary In the Asia–Pacific region, rates of active behaviour by engaging in more active and transport to school vary by gender, with outdoor play than those in larger urban different patterns across countries.27 Differ- communities. Previous studies reported ences in active transport may also further that adolescents from developed coun- contribute to gender differences in overall tries spend approximately 6–10 hours PA among adolescents observed in this and per day being sedentary.30,31 Accelerome- other studies.27 As active-transport-to-school ter-based studies that compared sedentary rates are lower than motorised-transport time between urban and rural adolescents rates and continue to decline in New reported inconsistent results.5,6,8,9 Among Zealand,4 active transport should continue to Portuguese adolescents, urban female be promoted as a way to increase PA among adolescents spent significantly more time adolescents, particularly females. Possible in sedentary pursuits compared to their ways to encourage active transport to school rural counterparts.5 In the United Kingdom, in diverse geographic settings include strong technology-based sedentary behaviour social support, creating safe walking and was the most common after school activity cycling routes to school through built envi- among adolescents.32 Among 9–16-year-old ronment changes and encouraging mixed Australians, screen time was higher among transport for adolescents who live beyond urban males compared to rural males, but walking and/or cycling distance to their no difference was found among females.33 school. The discrepancies between findings from The finding that a greater proportion of studies could be due to different defini- 8,9 adolescents who met PA guidelines partici- tions of ‘rurality’: based on accessibility, 5,6 pated in organised sports compared to their or population size, or because of different counterparts is also consistent with other availability of services, facilities and recre- studies.26 Recent New Zealand data showed ation spaces within urban and rural areas that 81% of 5–17 year olds participated in in New Zealand compared to other coun- 34 organised sports and 54% of those aged tries. In addition, differences are likely 13–18 years engaged in school sports.4 Many to exist in what qualifies as sedentary factors contribute to adolescents’ sports behaviour. As Otago adolescents spend a participation, including individual charac- significant amount of time in stationary teristics, family socioeconomic status28 and pursuits, future interventions should focus a supportive sport environment.26 Previous on reducing sedentary time by reducing Otago research showed that factors related recreational screen time and encouraging 4 to a supportive sport environment, such as outdoor activities among New Zealand provision of sport grounds at school, quality adolescents in both urban and rural areas. of sport management and availability of For adolescents living in large urban areas, sports outside school, were associated such interventions should include efforts to

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 61 www.nzma.org.nz/journal ARTICLE

increase active-transport-to-school rates and studies; a relatively small sample size in to reduce education-based sedentary time some geographical settings; and a lack through the introduction of regular activity of data on reasons for adolescents not breaks.35 consenting to participate in this assessment. The strengths of this study include the Although the findings may not be general- device-based measurement of PA; the isable to other geographical regions within inclusion of participants living in different New Zealand or to other countries, our geographical settings with geographically results are consistent with international matched participants’ home and school findings and provide further insight into location; high school participation rates understanding PA among adolescents in across the region; and PA being analysed at different environments. different times throughout the school day The proportion of Otago adolescents and on weekend days. However, there are meeting recommended PA levels was low. limitations that should be acknowledged. But a small increase in daily PA levels (eg, Although accelerometers measure a range of six minutes) would see most adolescents movements in multiple planes, they cannot meeting current guidelines. Continuous measure upper-body movements, movement encouragement of PA is needed for adoles- on a graded terrain and movement in activ- cents across urban and rural settings. ities such as swimming and cycling, which New Zealand already provides substantial may limit the validity of PA data among support for PA within school, community participants who regularly participate in and environmental settings, as well as at the those physical activities. Future research government/policy level.4 Future initiatives should improve the utility of accelerometers to increase PA among New Zealand adoles- for assessing swimming, cycling, upper-body cents should focus on facilitating active and graded-terrain PA. Additional limita- transport to school (eg, through initiatives tions include the 3–4 year gap between the such as creating safe walking and cycling BEATS and BEATS-R studies data collection; routes to school),36 promoting inclusive sport the difference in length of epoch for accel- participation26 and increasing peer and erometer data collection between the two family support for PA.4

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 62 www.nzma.org.nz/journal ARTICLE

Competing interests: Nil. Acknowledgements: The BEATS Study was supported by the Health Research Council of New Zealand Emerging Researcher First Grant (14/565), National Heart Foundation of New Zealand (1602 and 1615), Lottery Health Research Grant (Applic 341129), University of Otago Research Grant (UORG 2014) and the Dunedin City Council. The BEATS Rural Study was supported by the University of Otago Research Grant (UORG 2018) and Otago Energy Research Centre Seed Grant. Britta- ny White was supported by the University of Otago Masters Scholarship. The authors would like to acknowledge BEATS investigators and Advisory Board members, research personnel (research assistants, students and volunteers) and all participating schools and adolescents. Author information: Brittany White: Master of Science Graduate, School of Physical Education, Sport and Exercise Sciences, University of Otago, Dunedin. Enrique García Bengoechea: Succeed & Lead Fellow, Physical Activity for Health, Health Research Institute, Department of Physical Education and Sport Sciences, University of Limerick, Limerick, Ireland. John C Spence: Professor, Faculty of Kinesiology, Sport, and Recreation, University of Alberta, Edmonton, Alberta, Canada. Kirsten J Coppell: Public Health Physician and Research Associate Professor, Department of Medicine, University of Otago, Dunedin. Sandra Mandic: Adjunct Professor, School of Sport and Recreation, Faculty of Health and Environmental Sciences, Auckland University of Technology, Auckland; Research Affiliate, Centre for Sustainability, University of Otago, Dunedin. Corresponding author: Sandra Mandic, School of Sport and Recreation, Faculty of Health and Environmental Sciences, Auckland University of Technology, Private Bag 92006, Auckland 1142, New Zealand, +64 21 0902 0025 [email protected] URL: www.nzma.org.nz/journal-articles/comparison-of-physical-activity-patterns-across-large-medium-and-small-urban-areas-and-rural-settings-in-the-otago-region-new-zealand

REFERENCES 1. Biddle SJH, Atkin AJ, Cavill Child Adolesc Health. 6. Comte M, Hobin E, N, Foster C. Correlates of 2020;4:23-35. doi:10.1016/ Majumdar SR, et al. physical activity in youth: S2352-4642(19)30323-2 Patterns of weekday and a review of quantitative 4. Smith M, Ikeda E, Hinck- weekend physical activity systematic reviews. Int son E, et al. Results from in youth in 2 Canadian Rev Sport Exerc Psychol. New Zealand’s 2018 provinces. Appl Physiol 2011;4:25-49. doi:10.1080 Report Card on Physical Nutr Metab. 2013;38:115- /1750984X.2010.548528 Activity for Children and 119. doi:10.1139/ 2. Roberts KC, Yao X, Youth. J Phys Act Health. apnm-2012-0100 Carson V, et al. Meeting 2018;15(S2):S390-S392. 7. Euler R, Jimenez EY, the Canadian 24-Hour doi:10.1123/jpah.2018-0463 Sanders S, et al. Rural-ur- Movement Guidelines 5. Machado-Rodrigues AM, ban differences in for Children and Youth. Coelho-e-Silva MJ, Mota J, et baseline dietary intake Health Rep. 2017;28:3-7. al. Cardiorespiratory fitness, and physical activity 3. Guthold R, Stevens GA, weight status and objec- levels of adolescents. Riley LM, Bull FC. Global tively measured sedentary Prev Chronic Dis. 2019;16 trends in insufficient behaviour and physical doi:10.5888/pcd16.180200 physical activity among activity in rural and urban 8. Hardy LL, Dobbins T, adolescents: a pooled Portuguese adolescents. J Child Booth ML, et al. Seden- analysis of 298 popula- Health Care. 2012;16:166-177. tary behaviours among tion-based surveys with 1·6 doi:10.1177/1367493511 Australian adolescents. million participants. Lancet 430676 Aust N Z J Public Health.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 63 www.nzma.org.nz/journal ARTICLE

2006;30:534-540. doi:10.1111/ Matrix 3.0 Physical Activity of community design, j.1467-842X.2006.tb00782.x Report Card Grades for greenness, and physical 9. Ojiambo RM, Easton C, Children and Youth: activity in children using Casajús JA, et al. Effect of Results and Analysis From satellite, GPS and acceler- urbanization on objectively 49 Countries. J Phys Act ometer data. Health Place. measured physical activity Health. 2018;15:S251-S273. 2012;18:46-54. doi:10.1016/j. levels, sedentary time, and doi:10.1123/jpah.2018-0472 healthplace.2011.09.003 indices of adiposity in 17. Rainham DG, Bates CJ, 24. Götte M, Seidel CC, Kenyan adolescents. J Phys Blanchard CM, et al. Spatial Kesting SV, et al. Objectively Act Health. 2012;9:115-123. Classification of Youth Phys- measured versus self-re- 10. Statistics New Zealand. ical Activity Patterns. Am ported physical activity ANZLIC Metadata J Prev Med. 2012;42:e87- in children and adoles- Urban Rural, 2018. e96. doi:10.1016/j. cents with cancer. PLOS www.stats.govt.nz amepre.2012.02.011 ONE. 2017;12:e0172216. doi:10.1371/journal. 11. Mandic S, Williams J, 18. Moore JB, Beets MW, pone.0172216 Moore A, et al. Built Morris SF, Kolbe MB. Environment and Active Comparison of Objectively 25. LeBlanc AGW, Janssen I. Transport to School Measured Physical Activity Difference Between Self-Re- (BEATS) Study: protocol Levels of Rural, Suburban, ported and Accelerometer for a cross-sectional study. and Urban Youth. Am J Measured Moderate-to-Vig- BMJ Open. 2016;6:e011196. Prev Med. 2014;46:289- orous Physical Activity doi:10.1136/ 292. doi:10.1016/j. in Youth. Pediatr Exerc bmjopen-2016-011196 amepre.2013.11.001 Sci. 2010;22:523-534. doi:10.1123/pes.22.4.523 12. Salmond C, Crampton 19. Moore JB, Brinkley J, P, King P, Waldegrave C. Crawford TW, et al. 26. Mandic S, Bengoechea EG, NZiDep: A New Zealand Association of the built Stevens E, et al. Getting index of socioeconomic environment with physical kids active by participat- deprivation for individuals. activity and adiposity in ing in sport and doing It Soc Sci Med. 2006;62:1474- rural and urban youth. more often: focusing on 1485. doi:10.1016/j. Prev Med. 2013;56:145- what matters. Int J Behav socscimed.2005.08.008 148. doi:10.1016/j. Nutr Phys Act. 2012;9:86. ypmed.2012.11.019 doi:10.1186/1479-5868-9-86 13. Kek CC, Bengoechea EG, Spence JC, Mandic S. The 20. Yousefian A, Ziller E, Swartz 27. Uddin R, Mandic S, Khan A. relationship between J, Hartley D. Active Living Active commuting to and transport-to-school habits for Rural Youth: Address- from school among 106,605 and physical activity in a ing Physical Inactivity adolescents in 27 Asia-Pa- sample of New Zealand in Rural Communities. J cific countries. J Transp adolescents. J Sport Public Health Manag Pract. Health. 2019;15:100637. Health Sci. 2019;8:463- 2009;15:223. doi:10.1097/ doi:10.1016/j. 470. doi:10.1016/j. PHH.0b013e3181a11822 jth.2019.100637 jshs.2019.02.006 21. Moore JB, Jilcott SB, Shores 28. Fernandes RA, Reichert 14. Evenson KR, Catellier DJ, KA, et al. A qualitative FF, Monteiro HL, et al. Gill K, et al. Calibration of examination of perceived Characteristics of family two objective measures barriers and facilitators of nucleus as correlates of of physical activity for physical activity for urban regular participation in children. J Sports Sci. and rural youth. Health sports among adolescents 2008;26:1557-1565. Educ Res. 2010;25:355-367. | SpringerLink. Int J Public doi:10.1080/0264 doi:10.1093/her/cyq004 Health. 2012;57:431-435. 0410802334196 22. Pearce J, Witten K, 29. Janssen I. Active play: 15. Mishra P, Pandey CM, Bartie P. Neighbour- An important physical Singh U, et al. Descriptive hoods and health: a GIS activity strategy in the Statistics and Normality approach to measuring fight against childhood Tests for Statistical Data. community resource obesity. Can J Public Ann Card Anaesth. accessibility. J Epidemiol Health. 2014;105:e22-e27. 2019;22:67-72. doi:10.4103/ Community Health. doi:10.17269/cjph.105.4154 aca.ACA_157_18 2006;60:389-395. doi:10.1136/ 30. Smith M, Berdel D, jech.2005.043281 16. Aubert S, Barnes JD, Nowak D, et al. Physical Abdeta C, et al. Global 23. Almanza E, Jerrett M, Activity Levels and Dunton G, et al. A study Domains Assessed by

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 64 www.nzma.org.nz/journal ARTICLE

Accelerometry in German doi:10.1123/pes.20.4.446 doi:10.1186/1476-072X-13-43 Adolescents from GINIplus 33. Dollman J, Maher C, Olds 35. Salmon J, Tremblay MS, and LISAplus. PLOS TS, Ridley K. Physical Marshall SJ, Hume C. ONE. 2016;11:e0152217. activity and screen time Health Risks, Correlates, doi:10.1371/journal. behaviour in metropol- and Interventions to pone.0152217 itan, regional and rural Reduce Sedentary Behavior 31. Ting JLC, Mukherjee S, adolescents: A -sectional in Young People. Am J Hwa MCY. Physical Activity study of Australians aged Prev Med. 2011;41:197- and Sedentary Behavior 9–16 years. J Sci Med Sport. 206. doi:10.1016/j. Patterns of Singaporean 2012;15:32-37. doi:10.1016/j. amepre.2011.05.001 Adolescents. J Phys Act jsams.2011.05.011 36. Rahman ML, Moore A, Health. 2015;12:1213-1220. 34. Adams MA, Frank LD, Smith M, et al. A Concep- doi:10.1123/jpah.2014-0207 Schipperijn J, et al. tual Framework for 32. Atkin AJ, Gorely T, International variation in Modelling Safe Walking Biddle SJH, et al. Crit- neighborhood walkability, and Cycling Routes to ical Hours: Physical transit, and recreation envi- High Schools. Int J Envi- Activity and Sedentary ronments using geographic ron Res Public Health. Behavior of Adolescents information systems: the 2020;17:3318. doi:10.3390/ after School. Pediatr Exerc IPEN adult study. Int J ijerph17093318 Sci. 2008;20:446-456. Health Geogr. 2014;13:43.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 65 www.nzma.org.nz/journal ARTICLE

Predictors of medicinal cannabis users’ willingness to utilise a new prescription Medicinal Sannabis Scheme in New Zealand Marta Rychert, Chris Wilkins, Karl Parker, Thomas Graydon-Guy

ABSTRACT AIM: To investigate medicinal cannabis users’ intentions to transition to the new prescription Medicinal Cannabis Scheme (MCS) in New Zealand. METHOD: An online survey of 3,634 past-year medicinal cannabis users completed prior to implementation of the MCS in New Zealand in April 2020. Logistic regression models were fitted to identify predictors of intended future engagement with the MCS. RESULTS: Seventy-eight percent of respondents were aware of the new MCS and 66% intended to use it. Higher income (OR=1.57), younger age (OR=1.02) and smoking cannabis (v. vaping (OR=2.0) or oral ingestion in edible form (OR=2.22)) predicted intention to engage with the MCS. Conversely, Māori (OR=0.63) and those who grew their own cannabis (OR=0.52) were less likely to intend to engage with the new prescription MCS. CONCLUSION: The lower intended engagement with the MCS by Māori, lower income groups and those who home-grow cannabis may reflect their perceptions of the MCS as restrictive and expensive.

any countries have liberalised route was limited, with barriers including access to cannabis for medical financial cost and lack of support from the Mpurposes in the past two decades medical profession.7 Understanding con- or so, with considerable heterogeneity of sumers’ intentions to transition to the new regulatory approaches.1–3 The new prescrip- medicinal cannabis schemes is critical to tion regimes aim to provide consumers with informing their regulatory design and im- the benefits of medical oversight and access plementation. to quality-assured products, while accepting New Zealand’s Medicinal Cannabis the scarcity of high-quality scientific evi- Scheme (MCS) came into force on 1 April dence for the efficacy of cannabis in medical 2020. Under the MCS, medical professionals 4,5 treatment. Some medicinal cannabis re- can prescribe cannabis-based products gimes, especially in the initial stages of their (including both cannabidiol (CBD) and implementation, have suffered from poor tetrahydrocannabinol (THC)) for any patient engagement by patients and prescribers. In suffering from any health complaint. Canada, for example, even 10 years after Products allowed under this regime are their scheme was originally established in accessed via pharmacies and must be either 2001, enrolments to the federal programme in pharmaceutical dosage forms (eg, pills, represented 5% of estimated medicinal oils, lozenges) or sold as dried cannabis 6 cannabis users. In Australia, two years after flower intended for vaporisation. Dried legalisation of medicinal cannabis in 2016, herbal cannabis for smoking is not allowed users’ engagement with the legal access under the MCS. Products available under the

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 66 www.nzma.org.nz/journal ARTICLE

MCS must first be assessed by the Ministry and experiences with different sources of of Health (MOH) as compliant with specified supply (eg, from a drug dealer, friend or “minimum quality standards”8 (yet, unlike family, growing your own) predict intended other approved medicines, efficacy and engagement with the MCS. safety data are not part of the pre-marketing assessment). At the time of writing, there Methods were no domestically produced cannabis An online survey of adults (16+) who products approved for pharmacy sale self-reported using cannabis for medicinal under the MCS (10 cannabis-based products purposes in the past 12 months in New were under Medicinal Cannabis Agency Zealand (NZ Medicinal Cannabis Survey) assessment as of September 2020). Three was promoted via paid advertising on products manufactured overseas (TilrayTM FacebookTM (targeting users living in New CBD and CBD:THC oral solutions) were Zealand aged 16 years or older) between approved under the MCS in March 2021. May and August 2019 (ie, approximately Although patients in New Zealand have 9–11 months before the MCS entered been able to access imported CBD products into force). The survey preamble defined on prescription since September 2017, their “medicinal cannabis” as the “use of cannabis financial cost has remained prohibitive. For or cannabis-based products to treat a example, the price for Tilray 25mL bottle medical condition or alleviate a symptom.” ranges from NZ$150 to NZ$350 (excluding The questionnaire was developed based on 9 pharmacy mark-up). The new MCS aims a number of overseas surveys of medicinal to improve access and reduce the price cannabis users,13,14 including the recent of cannabis-based products for medicinal study in Australia.15 We collected a range use in New Zealand by allowing domestic of data on patterns of medicinal cannabis cultivation and production, introducing use and supply, which have previously new product forms (ie, dry cannabis flower been reported, alongside a more detailed for vaping) and eliminating bureaucratic description of the method.11 Ethical approval processes for prescribing THC (including was obtained from Massey University MOH sign-off for individual patients). Human Ethics Committee (SOA19/19). It has been estimated that around 5% of Measures New Zealand’s adult population uses illegally Demographics: Age, gender, ethnicity, sourced cannabis for medicinal purposes, educational achievement, employment with pain, anxiety and depression being the status, household income, social welfare leading reasons for self-medicating.10,11 The status, region of residence and community extent to which these medicinal cannabis size (i.e., rural, small town or city). users will transition to the legal prescribed Medical conditions treated with cannabis medicinal cannabis regime will depend on in the past 12 months: A multiple choice a range of factors, including the perceived question listing 50 conditions grouped into affordability and availability of prescrip- seven categories: (1) pain, (2) sleep, (3) tions (dependent on regulatory design and mental health, (4) gastrointestinal, (5) neuro- policy communication), support and advice logical (6) cancer and (7) other. from medical professionals, connection to the illegal cannabis market and wider social Main source of medicinal cannabis supply norms about the legitimacy of medicinal in the past 12 months: Participants were cannabis use. The influence of these factors presented with a list of 15 sources of supply may vary significantly between different (Table 1). types of users.12 Time of using medicinal cannabis: Number This study aimed to explore existing of months. medicinal cannabis users’ intentions to Main route of administering (ROA) engage with the new prescription MCS in medicinal cannabis in the past 12 months: New Zealand, including what demographic Participants chose one answer from a list of characteristics predict users’ intended 13 ROAs (Table 1). engagement and whether different patterns Discussions with health provider and of use (eg, routes of administration, co-use access via prescription: Participants were of cannabis for recreational purposes) asked whether they had discussed the use

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 67 www.nzma.org.nz/journal ARTICLE

Table 1: Main sources of supply and routes of administering cannabis for medicinal reasons.

Main source of supply of cannabis in the past 12 months (n=2,825) % Purchase from an illegal recreational drug dealer 27.7

I grow my own 12.6

I buy it from friends or family 12.2

Gift from friends or family 10

Purchased from a green fairy 8.6

I buy it from a friend who grows it for me 7.4

From a green fairy (for free) 5.9

A friend grows it for me for free 5.6

Purchase from an online supplier 4.5

Prescribed by a New Zealand medical practitioner and dispensed from a pharmacy 2.5

Bought while living/travelling overseas 1.5

Other 1.4

From an informal club/cooperative 0.4

Free by participating in a clinical trial 0.2

Purchase from the darknet 0.1

Main route of administration of cannabis in the past 12 months (n=3,564) % Smoked through a water pipe/bong 27.9

Rolled into a joint 22.5

Smoked though a dry pipe (glass/metal/plastic) 15.7

Taken by mouth: as a liquid (eg, tincture, oil) 13.0

Inhaled through a vaporiser 6.5

Eaten as a cookie (or other baked form) 4.4

Applied to skin (eg, cream) 4.4

Tablet/capsule 2.6

Eaten in a raw form (fresh juice, smoothie) 0.8

Other 1.7

Taken by mouth: in a spray form 0.3

Gummy bears 0.1

Nasal application 0

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 68 www.nzma.org.nz/journal ARTICLE

of cannabis for medicinal reasons with and “likely” to engage in the new MCS a health provider and accessed canna- were grouped together, as were those who bis-based products via doctor’s prescription responded “very unlikely” and “unlikely”. in the past 12 months. The following predictors were included Use of recreational drugs in the past 12 in the model: the demographic variables months: Participants were asked whether (ethnicity was not prioritised, meaning they used alcohol, tobacco, cannabis for that primary ethnicity was included in the recreational purposes and other illegal drugs model); conditions treated with cannabis; in the past 12 months. main source of medicinal cannabis supply in the past year; main route of medici- Awareness of the new MCS and anticipated nal-cannabis administration; number of engagement with the new MCS: Partici- months of using medicinal cannabis; use pants were first asked: “Are you aware that of recreational drugs (eg, alcohol, tobacco, the government is currently developing a cannabis for recreational use, other illegal scheme to improve access to domestical- drugs) in the past year; discussions about ly-produced medicinal cannabis products, medicinal cannabis with health profes- which will be finalised by the end of 2019?”. sionals (yes/no); and accessing cannabis Those who answered “yes” were then via prescription in the past 12 months (yes/ asked, “How likely are you to use the new no). Models including all variables were medicinal cannabis scheme once it is imple- first run, followed by variable selection mented?” and asked to answer either “very using a stepwise method to pick variables, unlikely”, “unlikely”, “likely”, “very unlikely” and retaining those with p-values less than or “don’t know.” 0.1. All analysis was conducted in R version 4.016 and results were deemed significant at Analysis p≤0.05 For the purposes of analysis, sources of supply were grouped into six categories: (1) Results illegal recreational drug dealer; (2) friends A total of 3,634 people completed the or family (includes “gifts from friends survey. Fifty percent of the sample was or family” and “I buy it from friends or female, median age was 38 years, 18% were family”); (3) grow-your-own (“I grow my Māori (the indigenous population of New own cannabis,” “a friend grows it for me for Zealand), 31% had only achieved high-school free” and “I buy it from a friend who grows education, 28% lived a small town and 18% it for me”); (4) from a “green fairy,” which in a rural area, 17% were on a sickness includes free and paid-for cannabis (“green benefit and 35% received a combined fairy” is a term used to described compas- household income of $30,000 or less per sionate illegal suppliers of home-made year (Table 2). cannabis products focussed primarily on medicinal users; (5) online supplier; and (6) Seventy-eight per cent were aware the prescribed and dispensed from a pharmacy. government was developing a scheme to ROAs were grouped into five categories improve access to medicinal cannabis. (1) smoking (includes smoking in a joint, Sixty-six per cent of those said they were “smoking through a water pipe/bong” and “very likely” or “likely” to use the new MCS “smoking though a dry pipe”); (2) vaping; (Table 2). (3) oral administration (includes “taken as Statistically significant predictors of a liquid (eg, oils, tinctures)” and sprays); (4) reporting future engagement in the MCS eaten (ie, oral ingestion of edibles, which are included in Table 3. Higher household includes “eaten as a cookie,” “eaten in a raw income (OR=1.40–1.57) and younger age form (fresh juice, smoothie)”, gummy bears (OR=1.02) predicted greater intention to and tablets/capsules); and (5) other ROAs engage with the MCS. Conversely, Māori (includes topical, nasal application). (OR=0.63) and those who mainly accessed Logistic regression models were fitted to cannabis by growing their own (v. those who identify independent predictors of intended mainly bought cannabis from a recreational future engagement with the MCS. Partic- drug dealer, OR=0.52) were less likely to ipants who said they were “very likely” express intention to engage in the new MCS. Those who smoked cannabis were more

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 69 www.nzma.org.nz/journal ARTICLE

Table 2: Sample demographics and intentions to use the new MCS regime.

Age (n=3,634) Mean: 39.3 (SD 15.2), Median: 38, range: 16–90

Gender (n=3,613) Male 48.4% Female 50%

Gender diverse 1.6%

Ethnicity (n=3,557) New Zealand European 75.9% Māori 17.8%

Pacific 1.0%

Asian 1.8%

Middle Eastern/Latin American/African 1.5%

Other 2.1%

Highest level of education (n=3,508) None 1.4% Primary/intermediate 1.1%

High school 31.1%

Polytech/technical/trade school 38.3%

University 27.9%

Other 0.2%

Main occupation (n=3,514) Work (includes self-employed) 56.2% (41.4% full time, 14.8% part time) A student 9.3%

Retired 6.8%

On sickness benefit 17.3%

Unemployed 4.3%

Parenting/unpaid work 6.1%

Region of residence (n=3,590) North Island of New Zealand 72.9% South Island of New Zealand 27.1%

Type of residence location (n=3,508) City 54.0% Small town 28.3%

Rural area 17.7%

Household’s combined annual income $30,000 or less 34.7% (before tax) (n=2,519) $30,001 to 100,000 48.1%

Over $100,0000 17.1%

Financial benefit related to medical None 72.4% condition (n= 2,445) Yes (mostly Supported Living Payment 27.6% from Work and Income NZ)

Awareness of the new MCS regime Yes 78.2% (n=2,632) No 21.8%

Likelihood of using the new MCS Very unlikely 10.6% (n=2,056) Unlikely 6.5%

Likely 20.2%

Very likely 45.8%

Don’t know 16.9%

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 70 www.nzma.org.nz/journal ARTICLE

Table 3: Statistically significant p( ≤0.05) predictors of future engagement with the MCS

Model variables N p OR 95% confidence interval

Age (per additional year of age) 1,536 <0.0001 0.98 0.97–0.99

Ethnicity 0.0061 Other (includes New Zealand European) (reference) 1,195 - -

Māori 341 0.63 0.45–0.87

Income 0.0500 $30,000 or less (reference) 518 - -

$30,001–$10,000 726 1.40 1.02–1.93

over $100,000 292 1.57 1.03–2.42

Main sources of supply in the past 12 months 0.0007 Illegal recreational drug dealer (reference) 423 - -

Via friends or family 336 1.01 0.64–1.60

Via a green fairy 215 0.96 0.57–1.62

Growing your own 437 0.52 0.35–0.77

Online supplier 81 1.71 0.72–4.56

Prescribed by a doctor and dispensed from a pharmacy 44 0.72 0.30–1.93

Main route of administration in the past 12 months 0.0034 Smoked 988 - -

Smoked vs Eaten 135 2.22 1.41–3.45

Smoked vs inhaled through a vaporiser 91 2.00 1.11–3.45

Smoked vs taken by mouth (as a liquid) 234 1.11 0.68–1.82

Smoked vs other ROAs 88 1.54 0.84–2.70

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 71 www.nzma.org.nz/journal ARTICLE

likely to engage in the MCS than participants irritable bowel syndrome, depression and whose main route of administration was via social anxiety disorders, with research vaping (OR=2.0) or oral ingestion in edible and evidence on medical applications of form (OR=2.22). cannabis still in its founding stages.4,5 Studies Use of other recreational drugs in the past of medicinal cannabis users found patients year, including use of cannabis for recre- may conceal their cannabis use to avoid ational purposes, was not associated with moral judgements in the provider–patient 21 participants’ intention to engage in the MCS. relationship, and this may be a factor for Māori, who already face ethnic biases in the healthcare system.22 As a result of those Discussion tensions, prescribing in Australia (and often and conclusions in New Zealand) increasingly takes place via We found that some types of current specialised private “cannabis clinics”.23 While medicinal cannabis users are more likely access via specialised clinics overcomes the to want to transition to the new legal immediate barriers caused by the scepticism MCS. Demographic predictors of intended from the health profession, it also adds to the engagement were non-Māori ethnicity and cost of accessing medicinal cannabis legally higher household income. This suggests and results in separating patients from their that, despite the stated objective of broad- regular health providers, including both GPs ening access to cannabis-based products, and specialists. existing health inequities related to ethnicity We found that medicinal cannabis users’ and income are likely to remain a factor.17 sources of illegal supply have implications We previously reported that anticipation for their reported future engagement with of high financial costs for prescribed the MCS. Namely, patients who grew their cannabis was the leading reason for not own cannabis were less likely to intend intending to engage with the MCS, followed to engage with the new MCS. This may by the perception that prescriptions will reflect the financial costs of transitioning be difficult to obtain, even though the new to the prescription and pharmacy supply scheme does not restrict eligible health as well as these users’ general satisfaction conditions or prescriber types.18 Additional with existing sources of home supply and/ support and information on the advan- or their ideological view of cannabis as tages of transitioning to the new MCS (ie, herbal medicine separate from main- medical oversight and access to quality-as- stream medicine. Previous research found sured products) may be justified for Māori that those growing cannabis for medicinal and low socioeconomic groups to improve reasons form a heterogenous group covering their engagement with the legal regime as a a range of intentions, habits and norms.3,24 It source of supply of cannabis-based products may be that they view the MCS prescription for medicinal purposes. regime as too restrictive and narrowly The lower likelihood of engagement from focused to accommodate their needs. This Māori and lower socioeconomic groups suggests the MCS may need to be made may also reflect challenges in discussing more attractive to these patients, such as medicinal uses of cannabis in a clinical by allowing home cultivation (as allowed setting. Integration of cannabis into clinical under the Canadian medicinal regime and practice remains a highly contested issue, recommended in a recent Senate review primary due to the limited scientific evidence in Australia25) or allowing retail avail- on the therapeutic efficacy of cannabis as a ability of some products beyond pharmacy medicine.19 Indeed, there is “conclusive or and prescription (as is done in some US substantial evidence”5 for use of cannabis and European jurisdictions2,26). We previ- preparations in improving patient-reported ously suggested that non-intoxicating CBD multiple sclerosis spasticity symptoms, products could be sold over the counter, chemotherapy-induced nausea and vomiting provided an appropriate quality-assurance and chronic pain (with methodological framework is implemented.27 limitations in clinical trials noted20), and Administering cannabis via smoking was there is insufficient evidence for many other a strong predictor of likely engagement medical uses, including sleep disorders, with the new MCS, with medicinal cannabis

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 72 www.nzma.org.nz/journal ARTICLE

users who smoked twice as likely to intend convenience sample of current medicinal to engage in the MCS than those who vaped cannabis users. The financial cost of (OR=2.0) or ingested cannabis orally in recruiting a representative sample of a edible forms (OR=2.22). This may reflect small, hidden population of medicinal users’ awareness of the health risks of cannabis users is likely to be prohibitively smoking and that the new medicinal scheme high. In addition, representative household will offer lower-risk means of admin- surveys have their own issues contacting istration, such as oral preparations (eg, this hidden population, including low tinctures, oils and lozenges) and cannabis response rates, particularly with regard to vaping products. hard-to-reach, stigmatised groups involved Limitations in illegal activities.28 Although our online sample broadly resembles the demographic We acknowledge a number of limita- profile of the New Zealand population tions with this study. First, our question (particularly with regard to gender and referred to current medicinal cannabis Māori representation), there were also users’ future intentions to utilise the MCS, important differences (eg, our online sample and respondents may or may not act on was better educated and reported higher their intentions. Second, the survey was a support from government benefits).

Competing interests: The research was supported by the NZ Health Research Council grant (19/647). Author information: Marta Rychert, PhD: Senior Researcher, SHORE & Whāriki Research Centre, Massey University, New Zealand. Karl Parker, MSc: Research Statistician, SHORE & Whāriki Research Centre, Massey University, New Zealand. Chris Wilkins, PhD: Associate Professor, SHORE & Whāriki Research Centre, Massey University, New Zealand. Thomas Graydon-Guy: Technical Officer, SHORE & Whāriki Research Centre, Massey University, New Zealand. Corresponding author: Marta Rychert, PhD, SHORE & Whāriki Research Centre, PO Box 6137 Wellesley Street, Auckland 1141, New Zealand [email protected] URL: www.nzma.org.nz/journal-articles/predictors-of-medicinal-cannabis-users-willingness-to-utilise-a-new-prescription-medicinal-cannabis-scheme-in-new-zealand

REFERENCES 1. Abuhasira R, Shbiro L, 3. Hakkarainen P, Frank 5. National Academies of Landschaft Y. Medical use VA, Barratt MJ, Dahl HV, Sciences EaM. The Health of cannabis and cannabi- Decorte T, Karjalainen K, Effects of Cannabis and noids containing products et al. Growing medicine: Cannabinoids: The Current - Regulations in Europe Small-scale cannabis State of Evidence and and North America. Eur J cultivation for medical Recommendations for Intern Med. 2018;49:2-6. purposes in six different Research. Washington, DC: 2. Klieger SB, Gutman A, Allen countries. International The National Academies L, Pacula RL, Ibrahim JK, Journal of Drug Policy. Press; 2017. 486 p. Burris S. Mapping medical 2015;26(3):250-6. 6. Belle-Isle L, Walsh Z, marijuana: state laws 4. Whiting PF, Wolff RF, Callaway R, Lucas P, Capler regulating patients, product Deshpande S, et al. R, Kay R, et al. Barriers to safety, supply chains and Cannabinoids for medical access for Canadians who dispensaries, 2017. Addic- use: A systematic review use cannabis for thera- tion. 2017;112(12):2206-16. and meta-analysis. JAMA. peutic purposes. Int J Drug 2015;313(24):2456-73. Policy. 2014;25(4):691-9.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 73 www.nzma.org.nz/journal ARTICLE

7. Lintzeris N, Mills L, Suraev 13. Sexton M, Cuttler C, cannabis-based medicines A, Bravo M, Arkell T, Arnold Finnell JS, Mischley LK. for chronic pain manage- JC, et al. Medical cannabis A Cross-Sectional Survey ment – An overview use in the Australian of Medical Cannabis of systematic reviews. community following Users: Patterns of Use European Journal of introduction of legal access: and Perceived Efficacy. Pain. 2018;22(3):455-70. the 2018–2019 Online Cannabis Cannabinoid 21. Ryan J, Sharts-Hopko N. Cross-Sectional Cannabis as Res. 2016;1(1):131-8. The Experiences of Medical Medicine Survey (CAMS- 14. Hazekamp A, Ware MA, Marijuana Patients: 18). Harm Reduction Muller-Vahl KR, Abrams A Scoping Review of Journal. 2020;17(1):37. D, Grotenhermen F. the Qualitative Litera- 8. MOH. Medicinal Cannabis The Medicinal Use of ture. J Neurosci Nurs. Agency - Minimum quality Cannabis and Cannabi- 2017;49(3):185-90. standard 2020. Available noids—An International 22. Talamaivao N, Harris R, from: https://www. Cross-Sectional Survey on Cormack D, Paine SJ, King health.govt.nz/our-work/ Administration Forms. P. Racism and health in regulation-health-and-dis- Journal of Psychoactive Aotearoa New Zealand: ability-system/ Drugs. 2013;45(3):199-210. a systematic review of medicinal-cannabis-agency/ 15. Lintzeris N, Driels J, Elias quantitative studies. New medicinal-cannabis-agen- N, Arnold JC, McGregor Zealand Medical Journal. cy-information-industry/ IS, Allsop DJ. Medicinal 2020;133(1521):55-68. medicinal-canna- cannabis in Australia, 2016: 23. Gulbransen G, Xu W, Arroll bis-agency-working-me- the Cannabis as aMedicine B. Cannabidiol prescription dicinal-cannabis/ Survey (CAMS-16). Medical in clinical practice: an audit medicinal-cannabis-agen- Journal of Australia. on the first 400 patients in cy-minimum-qual- 2018;209(5):211-6. New Zealand. BJGP Open. ity-standard. 16. R Core Team. R: A language 2020;4(1):bjgpopen2 9. Canterbury District Health and environment for 0X101010. Board. Christchurch statistical computing. 24. Hakkarainen P, Decorte T, MEdicines Information R Foundation for Sznitman S, Karjalainen Service. Cannabis-based Statistical Computing, K, Barratt M, Frank VA, et products: cannabidiol. Vienna, Austria. 2020. al. Examining the blurred 2019. Available from: Available from: https:// boundaries between https://www.medicinesin- www.R-project.org/. medical and recreational formation.co.nz/bulletins/ 17. Goodyear-Smith F, Ashton cannabis – results from cannabis-based-prod- T. New Zealand health an international study ucts-part-3-can- system: universalism of small-scale cannabis nabidiol-cbd/. struggles with persisting cultivators. Drugs: Educa- 10. Pledger M, Martin G, inequities. The Lancet. tion, Prevention and Cumming J. New Zealand 2019;394(10196):432-42. Policy. 2019;26(3):250-8. Health Survey 2012/13: 18. Rychert M, Wilkins C, 25. Senate Standing Commit- characteristics of medicinal Parker K, Graydon-Guy tees on Community Affairs. cannabis users. N Z Med T. Exploring medicinal Current barriers to patient J. 2016;129(1433):25-36. use of cannabis in a access to medicinal canna- 11. Rychert M, Wilkins C, time of policy change in bis in Australia (Senate Parker K, Graydon-Guy T. New Zealand. N Z Med J. inquiry). 2020. Available Exploring medicinal use of 2020;133(1515):54-69. from: https://www.aph.gov. cannabis in a time of policy 19. Zolotov Y, Vulfsons S, au/Parliamentary_Busi- change in New Zealand. Zarhin D, Sznitman S. ness/Committees/Senate/ New Zealand Medical Jour- Medical cannabis: An Community_Affairs/ nal. 2020;133(1515):54-69. oxymoron? Physicians’ Medicinalcannabis. 12. Fataar F, Goodman S, perceptions of medical 26. McGregor IS, Cairns Wadsworth E, Hammond cannabis. International E, Abelev S, Cohen RS, D. Consumer perceptions Journal of Drug Policy. Henderson M, Couch D, et of ‘legal’ and ‘illegal’ 2018;57:4-10. al. Access to cannabidiol cannabis in US states with 20. Häuser W, Petzke F, without a prescription: A legal cannabis sales. Fitzcharles MA. Efficacy, cross-country comparison Addictive Behaviors. tolerability and safety of and analysis. International 2021;112:106563.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 74 www.nzma.org.nz/journal ARTICLE

Journal of Drug Policy. Zealand Medical Journal. ing trans-national 2020;85:102935. 2015;128(1421):69-70. Internet-mediated partic- 27. Rychert M, Wilkins C. 28. Barratt MJ, Potter GR, ipatory research with Did we have the wrong Wouters M, Wilkins C, hidden populations of debate about Elixinol™ and Werse B, Perälä J, et al. cannabis cultivators. Inter- medicinal cannabis? New Lessons from conduct- national Journal of Drug Policy. 2015;26(3):238-49.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 75 www.nzma.org.nz/journal REVIEW ARTICLE

Methoxyflurane toxicity: historical determination and lessons for modern patient and occupational exposure Serah J Allison, Paul D Docherty, Dirk Pons, J Geoffrey Chase

ABSTRACT AIM: Historically methoxyflurane was used for anaesthesia. Evidence of nephrotoxicity led to abandonment of this application. Subsequently, methoxyflurane, in lower doses, has re-emerged as an analgesic agent, typically used via the Penthrox inhaler in the ambulance setting. We review the literature to consider patient and occupational risks for methoxyflurane. METHOD: Articles were located via PubMed, ScienceDirect, Google Scholar, Anesthesiology journal and the Cochrane Library. RESULTS: Early studies investigated pharmacokinetics and considered the resulting effects to pose minimal risk. Pre-clinical rodent studies utilised a species not vulnerable to the nephrotoxic fluoride metabolite of methoxyflurane, so nephrotoxicity was not identified until almost a decade after its introduction, and was initially met with scepticism. Further evidence of nephrotoxicity led to abandonment of methoxyflurane use for anaesthesia. Subsequent research suggested there are additional risks potentially relevant to recurrent patient or occupational exposure. Specifically, greater than expected fluoride production after repeated low-dose exposure, increased fluoride production due to medication-caused hepatic enzyme induction, fluoride deposition in bone potentially acting as a slow-release fluoride compartment, which suggests a risk of skeletal fluorosis, and hepatotoxicity. Gestational risk is unclear. CONCLUSIONS: Methoxyflurane poses a potentially substantial health risk in high (anaesthetic) doses, and there are a number of pathways whereby repeated exposure to methoxyflurane in lower doses may pose a risk. Single analgesic doses in modern use generally appear safe for patients. However, the safety of recurrent patient or occupational healthcare-worker exposure has not been confirmed, and merits further investigation.

ethoxyflurane is a volatile organic Renal failure was identified in some patients liquid,1 a fluorinated hydrocarbon, anaesthetised with methoxyflurane.10,11 Mthat vaporises readily2 and has Methoxyflurane is estimated to have been historically been used as an anaesthetic responsible for clinical nephrotoxicity in agent from 1958.3 A decade after discovery, approximately 100 patients worldwide, and methoxyflurane was used frequently, mak- death in approximately 20 cases, before its ing up 10% of annual purchases of inhaled near universal discontinuation since the anaesthetics in the USA.3 Sedative and 1970s.12 4,5 analgesic effects were described, which However, methoxyflurane has been prompted an extension of its use outside reintroduced into the contemporary arma- of the operating room for indications such mentarium as an analgesic for emergency or 6–8 9 as labour pain and dressing changes. short procedures. Australasia,13,14 Europe,15

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 76 www.nzma.org.nz/journal REVIEW ARTICLE

Canada,16 South Africa17 and, more recently, the following sections of this paper. Only numerous other countries18 allow methoxy- English-language material was reviewed. flurane administration via the Penthrox Reference sections of relevant articles inhaler. This device has been manufactured were examined to identify further relevant by Medical Developments International material for inclusion. Documents included (formerly Medical Developments Australia) in this review range across case reports, since 1978, specifically for analgesic use,19 animal and human prospective observa- and Penthrox is currently the only widely tional studies, experimental trials and other commercially available methoxyflurane literature reviews. administration device. The Penthrox inhaler is a tube into which the methoxyflurane Historic use of medication is poured to soak a wick, with a whistle-like mouthpiece through which the methoxyflurane patient inhales methoxyflurane vapour.20 Early published animal studies were The device features a ‘dilution hole’, which conducted on dogs and determined allows the patient to control the concen- that, with regards to incidence of lethal tration delivered18,21 and can incorporate arrhythmias,29,30 electrolyte disturbances an activated carbon (AC) filter through or liver dysfunction,31 methoxyflurane which the patient is encouraged to exhale. compared favourably with alternative Methoxyflurane has an estimated atmo- inhaled anaesthetic agents. Reports were spheric lifetime of 54 days and a 100-year conflicted with regards to cardiovascular global warming potential four times that and respiratory effects,31–33 although these of carbon dioxide, although it compares risks are not generally discussed in later favourably in that regard against other investigations. Researchers of canine inhalational anaesthetics.22 Administration models noted a slow recovery from anaes- of methoxyflurane via the Penthrox inhaler thesia, with the dogs appearing sluggish has been demonstrated as more effective at until the next day.31 Anaesthetised human relieving pain than placebos and alternative subjects also exhibited slow onset and analgesia in a variety of settings including emergence from anaesthesia.4 Arterial pre-hospital,23,24 clinic,25 and emergency methoxyflurane was approximately double department.26,27 the venous level during early anaesthesia We reviewed the historic literature to (2 to 15 minutes), equilibrating after 100 34 describe the evolution in the understanding minutes. This suggests ready uptake into of the health risks associated with anaes- tissue, confirming laboratory studies that thetic methoxyflurane. Thus we identify had suggested methoxyflurane would have the potential patient and occupational risks high solubility in blood and high uptake in 28 of methoxyflurane in the modern setting, adipose tissue. Methoxyflurane concen- in order to help guide policymakers and tration in subcutaneous fat rose slowly, clinicians who might consider making peaking 5–8 hours after commencement methoxyflurane available in their clinical of anaesthesia and remaining elevated 34 environments. beyond 30 hours after cessation. Overall, these studies suggested that methoxyflurane had a large volume of distribution and Search strategy consequent delayed equilibration between To gain a comprehensive overview of the compartments. historical literature, we sought to locate As methoxyflurane was a new anaes- research and commentary on methoxy- thetic agent, a significant proportion of the flurane administration in any setting for historical animal and human research was any indication by any method other than dedicated to investigating the rate of onset the modern Penthrox inhaler. Articles were of and emergence from anaesthesia,4,31,35,36 located using PubMed, ScienceDirect and the concentration required to achieve Google Scholar and by searching the Anes- adequate anaesthesia1,20,37 and the threshold thesiology journal and Cochrane Library of sedation.36 Early medication safety investi- databases with the term ‘methoxyflurane’. gations included effects on respiration2,4,31,35,38 Articles were selected based on rele- and cardiovascular stability,4,35,38,39 with vance to health effects, as categorised in

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 77 www.nzma.org.nz/journal REVIEW ARTICLE

methoxyflurane demonstrating reasonable of methoxyflurane into fluoride. Only safety with regards to these concerns. Fischer 344 rats demonstrated biochemical Although many of these historical anaes- and pathological renal changes following thesia studies might not meet modern methoxyflurane anaesthesia.51 In susceptible scientific standards due to the poor quality rats, elevated fluoride was associated with and quantity of data, examination of this dose-related high-output renal failure.43,52,53 literature nonetheless allows identification Therefore, due to the use of a non-sus- and extraction of worthwhile findings. ceptible rat type, preclinical trials had Fluoride-associated health effects unfortunately failed to identify the nephrotoxic potential of methoxyflurane. Methoxyflurane is metabolised by lung and liver tissue into a variety of products,40 Early observations of anaesthetised including fluoride and oxalic acid.20 human subjects also suggested no renal Some medications are known to increase toxicity.35 Nephrotoxicity in clinical use (‘induce’) metabolism pathways in the liver. was suggested by a 1966 case series of 17 Pre-treatment of rat hepatic microsomes patients,10 although unfortunately this first in vitro with phenobarbital, an anticon- report of human methoxyflurane anaes- vulsant therapy,41 caused a 7- to 10-fold thesia-associated nephrotoxicity was met increase in fluoride production in response with scepticism.12 It was a further five to methoxyflurane exposure.40,42 Similarly, years until further incidents of high-output in vivo pre-treatment of rats with pheno- failure were described,54–56 at which time a barbital increased methoxyflurane uptake relationship was identified between serum and fluoride production.43–45 A patient fluoride following methoxyflurane anaes- who had secobarbital prior to receiving thesia and the degree of renal toxicity. In methoxyflurane had peak serum fluoride 1973, strong correlations were identified that was three times that of patients who between methoxyflurane anaesthetic dose, did not receive secobarbital, and the patient increased serum inorganic fluoride concen- subsequently exhibited a decrease in renal tration and the degree of nephrotoxicity,11 function. This suggests increased risk of with further biochemical evidence of renal elevated serum fluoride, and possibly dysfunction emerging the following year increased susceptibility to health effects from patients receiving methoxyflurane associated with elevated fluoride levels, for anaesthesia.39,60 Two types of methoxy- exposed individuals concomitantly using flurane-induced renal pathology were medications that affect methoxyflurane identified: low output failure exhibited metabolism. calcium oxalate crystals in the tubules of the renal cortex and the collecting tubules Prolonged exposure to methoxyflurane of the medulla, with cortex tubular inflam- also causes enzyme induction, altering matory changes;61 whereas high output methoxyflurane metabolism.46 Rat hepatic failure exhibited tubular necrosis,51 wide- microsomes in vitro exposed to low-dose spread tubular dilatation and calcium and methoxyflurane produced proportionally calcium oxalate crystals in the tubular more fluoride than with high-dose epithelium.62 In combination, these exposure,42 and the susceptible Fischer 344 findings provided “compelling scientific rat strain demonstrated prolonged low-dose evidence [which] led practitioners and exposure also resulting in increased fluoride the manufacturer to abandon methoxy- production.47 This nonlinear response flurane.”12 Methoxyflurane administration suggests extrapolation from single high-dose to human patients for both anaesthesia outcomes to repeated low-dose use or occu- and analgesia was generally discontinued pational exposure 48–50 cannot be undertaken around 1974.19,63 with simple linear assumptions. There appears to be inter-person variation Renal toxicity in response to serum fluoride resulting from Although preclinical trials conducted in methoxyflurane, with some studies demon- Sprague-Dawley rats found no evidence of strating serum fluoride levels above the 51 nephrotoxicity, over a decade later it was toxic threshold11 without identifying renal determined that different rat types exhibited dysfunction.65,66 This variability may be different degrees of hepatic conversion partially explained by additive nephrotox-

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 78 www.nzma.org.nz/journal REVIEW ARTICLE

icity from medications, such as tetracycline skeletal fluorosis risk,80,81 suggesting the or other antibiotics,67,68 that can alter typical potential for skeletal fluorosis in exposed methoxyflurane dose-responses, although persons or the foetuses of exposed pregnant variable susceptibility to the effects of persons. However, the quantity of risk is fluoride and/or other metabolites is also a unclear in the context of modern analgesic possibility. use and occupational exposure, and further Interestingly, the renal toxic serum study is needed. fluoride threshold is not necessarily Gestational effects consistent across fluorinated anaesthetic In rats, recurrent subanaesthetic methoxy- 71–73 agents. For example, sevoflurane, which flurane exposure does not appear to undergoes minimal renal defluorination significantly alter foetal loss, but it does compared with methoxyflurane, some- decrease foetal weight.82 In mice, recurrent times produces serum fluoride ≥50µmol/L low-dose exposure to methoxyflurane 71 without apparent renal dysfunction. provokes increased rates of foetal devel- Transient changes in renal function have opment variation, and frequent higher-dose been observed in some studies using rats (though still subanaesthetic) exposure and human subjects following sevoflurane causes decreased birth weight and increased anaesthesia. The renal changes are believed rates of death in utero.79 These findings to be due to a different compound that is suggest that there are potentially gestational formed by sevoflurane and unrelated to effects consequent of recurrent maternal 74 fluoride formation. However, differences exposure, although large human cohort in renal function following sevoflurane studies are needed to examine whether such are not statistically significant on meta- effects occur with modern use. analysis.75 It has been suggested that Women in labour receiving methoxy- the more pronounced renal metabolism flurane analgesia during labour produce exhibited by methoxyflurane is responsible fluoride that appears in their urine and for its comparatively greater renal toxic the urine of their newborns.83,84 The effect.70,71 Direct comparison between agents biological effect of maternal methoxy- is challenging. flurane analgesia on newborns has not Fluoride bone deposition been studied in humans. Likewise, human A 1973 study of mice and Wistar rats antenatal methoxyflurane use does not found 4.7–6.7% of the fluoride in methoxy- appear to have been explicitly studied.85 flurane was deposited in bone. Washout Furthermore, there appears to be an of bone fluoride was slow, returning to absence of research of the occupational control after 40–60 days. In concert with the safety of pregnant healthcare workers findings of enzyme induction studies,40,42 exposed to methoxyflurane. phenobarbital increased the amount of Hepatotoxicity fluoride deposited in bone.77 Subsequent Researchers in 1962 noted no clinical research confirmed fluoride deposition in indications of liver toxicity in two patient rat bone following methoxyflurane anaes- cohorts anaesthetised with methoxy- thesia, with preferential deposition in foetal flurane.35,38 However, subsequent case bone in the third trimester of pregnancy.78 reports of hepatitis emerged, which were Similarly, recurrent high-dose exposure associated with anaesthesia,86 delivery7,8 appears to cause decreased foetal ossifi- and medication abuse.87 The majority cation and minor skeletal abnormalities in of these reports occurred as or after mice.79 methoxyflurane was becoming used less These findings suggest that bone has the frequently worldwide. A further study in potential to act as a long-acting storage 1980 identified changes in hepatic function compartment of fluoride metabolite, which biomarkers following occupational exposure raises the possibility of fluoride accumu- of delivery-ward personnel.88 Hepatotox- lation with repeated exposure within an icity appears to have been an infrequent extended clearance time frame. Elevated but important effect that was unpredictably fluoride intake is associated with both associated with single or multiple doses of elevated serum fluoride and increased methoxyflurane.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 79 www.nzma.org.nz/journal REVIEW ARTICLE

Summary of historical use a control group, but nonetheless it is highly In the 1960s, methoxyflurane was a new unlikely that anaesthetic methoxyflurane anaesthetic agent. Research focused on studies will ever be repeated to ensure determining its pharmacokinetics, utility scientific rigour. as an anaesthetic agent, respiratory effects Methoxyflurane is experiencing a revival and cardiovascular changes. None of these in lower-dose analgesic use. The historical factors were ultimately determined to be experience has provided some insight into areas of significant risk. Methoxyflurane risk in the modern setting, with researchers appeared to be a useful agent for providing specifically investigating the possibility and stable anaesthesia with gradual emergence. finding no evidence of patient nephrotox- 27,89 Nephrotoxicity was only identified after icity and hepatotoxicity in current use. approximately a decade of use and was not However, the literature suggests some unre- formally associated as a methoxyflurane solved safety concerns. dose-response until approximately 15 years The Australian Therapeutic Goods Admin- after its introduction. Individual variation istration (TGA) reported 25 cases of adverse in nephrotoxic threshold added further reaction associated with Penthrox between complication, and it is possible scepticism 1971 and 19 July 2020. These cases included may have impeded research into toxic two deaths and one report of adverse effect effects. Hepatotoxicity occurred infre- due to occupational exposure.90 The New quently enough that it was not identified as Zealand Medicines and Medical Devices an important risk of methoxyflurane, and Safety Authority (Medsafe) reported six the combination of circumstances under cases of adverse reaction associated with which methoxyflurane causes hepatitis are Penthrox between 2000 and 19 October still unclear. 2020, with no deaths.91 These reports are Studies of hepatic induction of methoxy- aggregated in Table 1. The earliest report in flurane metabolism suggest that frequent the Australian TGA database is November low-dose uses or exposures, or some medica- 2005, and the earliest report in the New tions taken concomitantly, have the potential Zealand Medsafe database is January 2012. to produce proportionally greater toxicity Hence it is possible that any earlier events 90,91 than historical one-off high dose exposures. were unreported. It is important to Bone fluoride deposition as a result of acknowledge that association with adverse methoxyflurane exposure has been mini- events does not necessarily imply methoxy- mally studied. A small but not irrelevant flurane caused the event. However, these teratogenic risk is suggested by two animal reports suggest a non-zero risk requiring studies undertaken after methoxyflurane surveillance and investigation. use had generally ceased. In contemporary use in Australasia, patients are administered up to 6mL Relevance to the methoxyflurane per day and up to 15mL per week.13,14 Each 3mL dose lasts between 20 modern setting and 60 minutes depending on how intensely This review was undertaken with the goal the patient inhales.92 Methoxyflurane is used of including the widest possible range of by Australian and New Zealand ambulance historical literature in order to describe the services,23,24,48,89,92–99 by at least one Australian evolution of understanding of the health emergency department100 and in Austral- risks associated with methoxyflurane. asian in-hospital and clinic settings.25,101–108 Literature was located non-systematically, It has been administered to five million including extensive use of locating citations patients in Australia18 and a further one and a wide range of search terms. Thus, the million patients outside of Australia.112 Given review search is not completely systematic, this extensive use, the number of adverse but it is nonetheless comprehensive. Equally, events reported is reassuringly low, and due to near-global abandonment of studies suggests that modern use of methoxyflurane in the 1970s, research was restricted until may be safe for short-term administration. recently, which limits numbers and avail- Furthermore, the low rate of adverse effects ability of prior publications. Much of the implies relative safety for those who are historical data utilise small samples and lack occupationally exposed to methoxyflurane.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 80 www.nzma.org.nz/journal REVIEW ARTICLE

Table 1: Aggregated reports of adverse events involving methoxyflurane from the Australian Therapeutic Goods Administration (TGA)90 and the New Zealand Medicines and Medical Devices Safety Authority (Medsafe).91

Month Year Country Gender Age Adverse Other Other con- effects suspected comitant/not medications suspected medications Apr 1985 Australia Male 20 Hepatitis Chlorproma- None reported zine, fentanyl, halothane, flucoxacillin, pancuronium, suxamethoni- um, pethidine, thiopentone

Dec 2000 Australia Male 30 Malignant hy- Suxamethoni- None reported perthermia um, sevoflurane, propofol

Nov 2005 Australia Male 57 Hypoxia, medi- None reported None reported cation error

Dec 2005 Australia Female 34 Confusion, diz- None reported None reported ziness, hypoxia, somnolence

Jun 2006 Australia Female 20 Jaundice, None reported None reported abnormal liver function test result, vomiting

Mar 2008 Australia Male 26 Blood pressure None reported None reported fluctuation

Feb 2010 Australia Female 33 Hepatitis, None reported Sodium tetra- hepatomegaly, decyl sulphate, jaundice, liver fexofenadine, injury paracetamol

Feb 2010 Australia Female Not reported Hepatic failure, None reported None reported renal failure

Jul 2010 Australia Male 19 Affect lability, None reported None reported amnesia

May 2011 Australia Female 12 Lipase Box jellyfish Paracetamol, increased, antivenom, prednisolone pancreatitis morphine, fentanyl

Nov 2011 Australia Female 75 Altered state Morphine Not suspected of consciousness, nausea, vomiting

Jan 2012 New Zealand Male 64 Hepatic failure None reported None reported

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 81 www.nzma.org.nz/journal REVIEW ARTICLE

Month Year Country Gender Age Adverse Other Other con- effects suspected comitant/not medications suspected medications May 2014 Australia Female 86 Anaphylactic Fentanyl, tel- None reported reaction misartan

Apr 2015 Australia Female Not reported Liver function None reported None reported test increased

May 2015 Australia Male 64 Cardiac arrest, None reported None reported hypotension, nausea, syn- cope

Apr 2016 Australia Male 30 Depressed None reported None reported mood, feeling abnormal, nightmare

Jul 2016 Australia Female 10 Vomiting Oxycodone, Metronidazole, fentanyl, ampi- Augmentin, cillin paracetamol

Sep 2017 New Zealand Male 36 Dizziness, mal- None reported None reported aise, nausea

May 2018 Australia Female 47 Dizziness, None reported None reported muscle rigidity, nausea, unresponsive to stimuli

Aug 2018 Australia Male Not reported Depressed level None reported None reported of consciousness

Aug 2018 New Zealand Male 11 Abnormal None reported Morphine behaviour, depressed level of consciousness, dysphemia, hyperaesthesia, myoclonus

Oct 2018 Australia Male Not reported Drug abuse, None reported None reported hepatitis

Feb 2019 Australia Female Not reported Dizziness, None reported Not reported occupational exposure to product

May 2019 Australia Female 48 Aggression, None reported None reported amnesia

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 82 www.nzma.org.nz/journal REVIEW ARTICLE

Month Year Country Gender Age Adverse Other Other con- effects suspected comitant/not medications suspected medications Jun 2019 Australia Female 36 Abdominal None reported None reported pain, hepatitis, liver function test increased, malaise

Jul 2019 Australia Female Not reported Intentional None reported None reported product misuse

Aug 2019 Australia Female 75 Aphasia None reported None reported

Aug 2019 New Zealand Female 29 Anaphylactic Ibuprofen, None reported reaction paracetamol

Sep 2019 Australia Male Not reported Nephropathy None reported None reported

Feb 2020 New Zealand Female 44 Anaphylactic None reported Ethinylestradi- reaction ol, felodipine, enalapril, citalopram

May 2020 New Zealand Female 63 Hepatitis, hy- None reported Pantoprazole perbilirubinae- mia, myocardial infarction

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 83 www.nzma.org.nz/journal REVIEW ARTICLE

However, absence of evidence does not use or exposure. Skeletal fluorosis, gesta- necessarily mean an absence of harm, and it tional and hepatotoxic risk should be may be difficult to identify the true risk. The further investigated. Workplaces where identification of potential nephrotoxic and methoxyflurane vapour is present could other risks, combined with the low numbers institute monitoring of at-risk healthcare and high age of prior studies, provides workers’ urine fluoride against published a basis for consideration of research of guidelines114,115 as a general health measure. methoxyflurane safety in the modern Although the historical literature associates setting. The literature illustrates a paucity serum fluoride level with renal toxicity, Safe of clinical trials confirming the safety of Work Australia does not recommend work- methoxyflurane in cases other than one-off places institute serum fluoride testing, due analgesic administration. Therefore, despite to practical complexities.115 the apparent relative safety in modern use Although there has been some thus far, there is good cause for further primary108,116 and secondary48 reporting research to be undertaken to ensure the of occupational exposure ranges with the safety of patients and healthcare workers. Penthrox inhaler, the degree to which In summary, the risks of patient and patients exhale methoxyflurane into the occupational methoxyflurane exposure local environment beyond the duration of have been identified. The degree of risk for their treatment could be explained further patients and healthcare workers appears as this might cause exposure for other low, but nonetheless remains unquantified healthcare workers who subsequently in the following domains: receive and care for the patient. A recent • renal toxicity study supported by the Penthrox manufacturer derived an eight-hour Maximum • enzyme induction due to concomitant Exposure Limit by estimating the level medication use or repeated methoxy- at which there is a 10% increased risk of flurane exposure kidney toxicity from historical single-ex- • a possibility of fluoride bone depo- posure anaesthetic data.48 This provides a sition with unknown skeletal fluorosis useful measure against which to compare risk reported or predicted exposures. However, • hepatotoxicity. independent confirmation of this Maximum Additionally, there is a notable dearth of Exposure Limit would be ideal.48,117,118 research of gestational effects relating to Finally, the only recently published deter- repeated methoxyflurane exposure. Because minations of occupational health risk have of these currently unquantified risks, it been theoretical models and extrapolation to may be prudent for healthcare workers compare with a single-exposure nephrotoxic to minimise exposure through adequate threshold.48–50 Further research to determine environment ventilation and by directing the nephrotoxic and other health effect risk patients to exhale through the AC filter of associated with recurrent exposure would the Penthrox device. be valuable. The historical literature has been examined and the evolution of under- A number of areas for future research standing of health risks associated with are suggested. For example, prolonged methoxyflurane described. This stands as a compartmental equilibration28,34 suggests a case study of a medication enthusiastically potential for healthcare workers to accu- put into clinical practice without sufficient mulate methoxyflurane and metabolites confirmation of occupational safety. Policy- with repeated exposure over prolonged makers should take heed of the persistent periods. Whether such an effect occurs need for scientific confirmation of the has been investigated only by one pilot occupational safety of methoxyflurane study,113 and further research is warranted. administration, and also of the potential for The rate of use among both patients and similar oversights that could occur with the healthcare workers of relevant enzyme-in- utilisation of new medications or medica- ducing and nephrotoxic medications could tions with controversial characteristics. be considered in the context of the increased risk posed by concomitant methoxyflurane

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 84 www.nzma.org.nz/journal REVIEW ARTICLE

Competing interests: Ms Allison reports a grant from the NZ National Science Challenge, and a grant from the NZ Tertiary Education Commission, during the conduct of the study, and is a paramedic who utilises methoxyflurane for patient analgesia. Acknowledgements: Funding for the overall study was provided by the NZ National Science Challenge 7, Science for Technology and Innovation [grant number 2019-S3-CRS]; and the NZ Tertiary Education Com- mission (TEC) fund MedTech CoRE (Centre of Research Excellence) [grant number 3705718]. Author information: Serah J Allison: PhD candidate, Department of Mechanical Engineering, University of Canterbury, New Zealand and Intensive Care Paramedic, Wellington Free Ambulance, New Zealand. Paul D Docherty: Associate Professor, Department of Mechanical Engineering, University of Canterbury, New Zealand; Institute of Technical Medicine, Furtwangen University, Germany. Dirk Pons: Associate Professor, Department of Mechanical Engineering, University of Canterbury, New Zealand. J Geoffrey Chase: Distinguished Professor, Department of Mechanical Engineering, University of Canterbury, New Zealand. Corresponding author: Paul D Docherty, Department of Mechanical Engineering, Private bag 4800, Christchurch 8041, New Zealand [email protected] URL: www.nzma.org.nz/journal-articles/methoxyflurane-toxicity-historical-determination-and-lessons-for-modern-patient-and-occupational-exposure

REFERENCES 1. Heusler H. Quantitative analgesia during labour. response in man. JAMA. analysis of common Br Med J. 1972; 1:81-3. 1973; 225(13):1611-6. anaesthetic agents. J Chro- 7. Rubinger D, Davidson JT, 12. Mazze RI. Methoxyflu- matogr. 1985; 340:273-319. Melmed RN. Hepatitis rane revisited: tale of an 2. McIntyre JWR, Gain following the use of anesthetic from cradle EA. Methoxyflurane. methoxyflurane in obstetric to grave. Anesthesiology. Canad Anaesth Soc analgesia. Anesthesiology. 2006; 105(4):843-6. J. 1962; 9:319-24. 1975; 43(5):593-5. 13. Douglas Pharmaceuticals 3. Van Poznak A. Methoxy- 8. DeLia JE, Maxson WS, Ltd. Penthrox New Zealand flurane and teflurane. Breen JL. Methoxy- data sheet. Cited April In: Chenoweth MB, (ed) flurane hepatitis: two 29, 2020. Available from: Modern Inhalation Anaes- cases following obstetric https://medsafe.govt. thetics. Berlin, Germany: analgesia. Int J Gynaecol nz/profs/datasheet/p/ Springer, 1972; 77-92. Ob. 1983; 21(1):89-93. penthroxinh.pdf 4. Andersen N, Andersen 9. Toomath RJ, Morrison 14. Medical Developments EW. Methoxyflurane: a RB. Renal failure follow- International. Australian new volatile anaesthetic ing methoxyflurane product information - agent. Acta Anaesthesiol analgesia. NZ Med J. Penthrox (methoxyflurane) Scand. 1961; 5:179-89. 1987; 100(836):707-8. inhalation. Cited May 30, 5. Tomlin PJ, Jones BC, 10. Crandell WB, Pappas SG, 2020. Available from: Edwards R, Robin PE. MacDonald A. Nephro- https://www.ebs.tga.gov.au/ Subjective and objective toxicity associated with ebs/picmi/picmirepository. sensory responses to methoxyflurane anes- nsf/pdf?OpenAgent&id=CP- inhalation of nitrous oxide thesia. Anesthesiology. 2010-PI-02403- and methoxyflurane. Br J 1966; 27(5):591-607. 3&d=202005301016933 Anaesth. 1973; 45:719-25. 11. Cousins MJ, Mazze 15. European Medicines 6. Rosen M, Latto P, Asscher RI, Barr GA, Kosek JC. Agency. List of nationally AW. Kidney function Methoxyflurane nephro- authorised medicinal after methoxyflurane toxicity: a study of dose products - methoxyflu-

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 85 www.nzma.org.nz/journal REVIEW ARTICLE

rane. Cited May 30, 2020. Available from: http:// 28. Eger EI, Shargel R. The solu- Available from: https:// www.medsafe.govt.nz/ bility of methoxyflurane www.ema.europa.eu/ 22. Hass SA, Andersen ST, in human blood and tissue en/documents/psusa/ Andersen MPS, Nielsen homogenates. Anesthesi- methoxyflurane-list-na- OJ. Atmospheric Chem- ology. 1963; 24(5):625-7. tionally-authorised-me- istry of Methoxyflurane 29. Bamforth BJ, Siebecker dicinal-products-psu- (CH3OCF2CHCl2): Kinetics KL, Kraemer R, Orth OS. sa/00010484/201711_en.pdf of the gas-phase reactions Effect of epinephrine 16. Purdue Pharma (Canada). with OH radicals, Cl on the dog heart during Penthrox receives market- atoms and O3. Chem Phys methoxyflurane anes- ing authorization from Lett. 2019; 722:119-23. thesia. Anesthesiology. Health Canada for adult 23. Jennings PA, Lord B, Smith 1961; 22(2):169-73. patients requiring relief K. Clinically meaningful 30. Wong KC, Tseng CK, Puerto from moderate to severe reduction in pain severity BA, Puerto AX. Chronic acute pain associated with in children treated by hypokalemia on epineph- trauma or interventional paramedics: a retrospective rine-induced dysrhythmias medical procedures. Cited cohort study. Am J Emerg during halothane, enflu- May 30, 2020. Available Med. 2015; 33(11):1587-90. rane or methoxyflurane from: https://purdue.ca/ 24. Johnston S, Wilkes GJ, with N2O anesthesia. Anes- en/2018/04/10/penthrox-re- Thompson JA, Ziman M, thesiology. 1979; 9(51):S119. ceives-marketing-autho- Brightwell R. Inhaled 31. Dobkin AB, Fedoruk rization-from-health-can- methoxyflurane and S. Comparison of the ada-for-adult-pa- intranasal fentanyl for cadiovascular, respiratory tients-requiring-re- prehospital management and metabolic effects lief-from-moderate-to-se- of visceral pain in an of methoxyflurane and vere-acute-pain-as- australian ambulance halothane in dogs. Anesthe- sociated-with-trau- service. Emerg Med siology. 1961; 22:355-62. ma-or-intervention- J. 2011; 28:57-63. al-medical-procedures/ 32. Andrews IC, Orkin LR. 25. Nguyen NQ, Toscano Methoxyflurane and 17. The South African Society L, Lawrence M, et al. halothane anesthesia of Anaesthesiologists. South Patient-controlled during controlled bleeding African acute pain guide- analgesia with inhaled in dogs: effect on respi- lines. Cited May 30, 2020. methoxyflurane versus ration. Anesthesiology. Available from: https:// conventional endosco- 1968; 29(1):171-2. painsa.org.za/wp-con- pist-provided sedation for tent/uploads/2016/07/ 33. Bagwell EE, Woods EF. colonoscopy: a randomized SASA-Acute-Pain- Cardiovascular effects of multicenter trial. Gastro Guidelines_2015.pdf methoxyflurane. Anesthe- Endo. 2013; 78(6):892-901. siology. 1962; 23(1):51-7. 18. Ikeda S. The reincarna- 26. Mercadante S, Voza A, tion of methoxyflurane. 34. Chenoweth MB, Robertson Serra S, et al. Analgesic J Anesth Hist. 2019. DN, Erley DS, Golhke R. efficacy, practicality and Blood and tissue levels of 19. Dayan A. Analgesic use of safety of inhaled methoxy- ether, chloroform, halo- inhaled methoxyflurane: flurane versus standard thane and methoxyflurane evaluation of its potential analgesic treatment for in dogs. Anesthesiology. nephrotoxicity. Medical acute trauma pain in 1962; 23:101-6. Developments Interna- the emergency setting: 35. Campbell MW, Hvolboll AP, tional. Cited June 26, 2016. a randomised, open-la- Brechner VL. Penthrane: Available from: http:// bel, active-controlled, a clinical evaluation in www.medicaldev.com/ multicentre trial in Italy 50 cases. Anesth Analg. 20. Dayan AD. Analgesic use of (MEDITA). Adv Ther. 2019. 1962; 41(2):134-9. inhaled methoxyflurane: 27. Coffey F, Wright J, 36. Stoelting RK, Longnecker evaluation of its potential Hartsholm S, et al. STOP!: DE, Eger EI. Minimum nephrotoxicity. Hum Exp a randomised, double- alveolar concentrations Toxicol. 2016; 35(1):91-100. blind, placebo-controlled in man on awakening 21. Medsafe. Penthrox study of the efficacy and from methoxyflurane, (methoxyflurane) safety of methoxyflu- halothane, ether and inhalation – product rane for the treatment fluroxene anesthesia: information. 16/12/2013 of acute pain. Emerg MAC awake. Anesthesi- edition. Cited July 26, 2016. Med J. 2014; 31:613-8. ology. 1970; 33(1):5-9.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 86 www.nzma.org.nz/journal REVIEW ARTICLE

37. Saidman LJ, Eger EI, 46. Berman ML, Bochantin JF. 54. Mazze RI, Shue GL, Jackson Munson ES, Babad AA, Nonspecific stimulation of SH. Renal dysfunction Muallem M. Minimum drug metabolism in rats by associated with methoxy- alveolar concentrations methoxyflurane. Anesthe- flurane anesthesia: a of methoxyflurane, siology. 1970; 32(6):500-6. randomised, prospective halothane, ether and 47. White AE, Stevens WC, clinical evaluation. JAMA. cyclopropane in Man: Eger EI, Mazze RI, Hitt 1971; 216(2):278-88. correlation with theories of BA. Enflurane and 55. Merkle RB, McDonald FD, anesthesia. Anesthesiology. methoxyflurane metab- Waldman J, et al. Human 1967; 28(6):994-1002. olism at anesthetic and renal function following 38. Thomason R, Light G, at subanesthetic concen- methoxyflurane anesthesia. Holaday DA. Methoxy- trations. Anesth Analg. JAMA. 1971; 261(6):841-4. flurane anesthesia: a 1979; 58(3):221-4. 56. Proctor EA, Barton FL. clinical appraisal. Anesth 48. Frangos J, Mikkonnen A, Polyuric acute renal failure Analg. 1962; 41:225-9. Down C. Derivation of an after methoxyflurane 39. Desmond JW. Methoxy- occupational exposure and tetracycline. Br flurane nephrotoxicity. limit for an inhalation Med J. 1971; 4:661-2. Canad Anaesth Soc J. 1974; analgesic methoxyflurane 57. Bruce DL, Eide KA, Linde 21(3):294-307. (Penthrox). Regul Toxicol HW, Eckenhoff JE. Causes 40. Blitt CD, Brown BR, Wright Pharmacol. 2016; 80:210-25. of death among anes- BJ, Gandolfi AJ, Sipes IG. 49. Allison SJ, Docherty thesiologists: a 20-year Pulmonary biotransforma- PD, Pons D, Chase JG. survey. Anesthesiology. tion of methoxyflurane: A bootstrap approach 1968; 29(3):565-9. an in-vitro study in the for predicting methoxy- 58. Bruce DL, Eide KA, Smith rabbit. Anesthesiology. flurane occupational NJ, Seitzer F, Dykes MHM. 1979; 51(6):528-31. exposure in paramedicine. A prospective survey of 41. PSM Healthcare. Phenobar- IFAC-PapersOnLine. 2017; anesthesiologist mortality, bitone - New Zealand data 50(1):6666-71. 1967–1971. Anesthesiol- sheet. 04/2017 edition. Cited 50. Allison SJ, Docherty ogy. 1974; 41(1). April 7, 2018. Available PD, Pons D, Chase JG. 59. Mazze RI, Trudell JR, from: http://www.medsafe. A bootstrap approach Cousins MJ. Methoxy- govt.nz/profs/Datasheet/p/ for predicting fluoride flurane metabolism Phenobarbitonetab.pdf toxicity in paramedics after and renal dysfunction: 42. Adler L, Brown BR, occupational methoxy- clinical correlation in Thompson MF. Kinetics flurane exposure. IFAC Man. Anesthesiology. of methoxyflurane Journal of Systems and 1971; 35(3):247-52. Control. 2019; 9(30). biotransformation with 60. Wu AHB. Tietz clinical reference to substrate 51. Mazze RI, Cousins MJ, guide to laboratory tests. inhibition. Anesthesiol- Kosek JC. Strain differences 4th ed. St. Louis: Saun- ogy. 1976; 44(5):380-5. in metabolism and suscep- ders Elsevier, 2006. tibility to the nephrotoxic 43. Baden JM, Rice SA, Denson 61. Kuzucu EY. Methoxyflu- effects of methoxyflurane DD, Mazze RI. Deuterated rane, tetracycline, and in rats. J Pharmacol Exp methoxyflurane (d4-MOF) renal failure. JAMA. Ther. 1973; 184(2):481-8. anesthesia. Anesthesiol- 1970; 211(7):1162-4. ogy. 1979; 9(51):S264. 52. Cousins MJ, Mazze RI, Barr 62. Powell HC, Garret RS, GA, Kosek JC. A comparison 44. Baden JM, Rice SA, Mazze Bernstein L, Mazze of the renal effects of isoflu- RI. Deuterated methoxy- RI. Methoxyflurane rane and methoxyflurane flurane anesthesia and nephropathy. Hum Pathol. in Fischer 344 rats. Anesthe- renal function in Fischer 1974; 5(3):359-63. 344 rats. Anesthesiology. siology. 1973; 38(6):557-63. 63. Fletcher S. From the 1982; 56:203-6. 53. Mazze RI, Cousins MJ, Editor. RCA Bulletin, 2015. 45. Berman ML, Lowe HJ, Kosek JC. Dose-relat- 64. Andersen NB, Cascorbi HF. Bochantin J, Hagler K. ed methoxyflurane Effects of methoxyflurane Uptake and elimination of nephrotoxicity in rats: and two metabolites on methoxyflurane as influ- a biochemical and sodium transport in the enced by enzyme induction pathologic correla- toad bladder. Anesthesi- in the rat. Anesthesiology. tion. Anesthesiology. ology. 1974; 40(4):371-5. 1973; 38(4):352-7. 1972; 36(6):571-87.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 87 www.nzma.org.nz/journal REVIEW ARTICLE

65. Cousins MJ, Nishimura TG, enigma. Anesthesiology. 81. World Health Organization. Mazze RI. Renal effects of 1995; 82(3):607-8. Guidelines for drinking-wa- low-dose methoxyflurane 73. Brown BR, Jr. Sevoflu- ter quality. 4th edition. with cardiopulmonary rane, fluoride ion, and Cited August 12, 2020. bypass. Anesthesiology. renal toxicity. Anesthe- Available from: https:// 1972; 36(3):286-92. siology. 1995; 83(1). apps.who.int/iris/bitstream/ handle/10665/25463 66. Creasser CW, Stoelting MJ. 74. Gentz BA, Malan TP, Jr. 7/9789241549950-eng.pdf Serum inorganic fluoride Renal toxicity with sevoflu- concentrations during and rane: a storm in a teacup? 82. Pope WD, Halsey MJ, after halothane, fluroxene, Drugs. 2001; 61(15):2155-62. Lansdown AB, Simmonds and methoxyflurane A, Bateman PE. Fetotoxicity 75. Sondekoppam RV, Nars- anesthesia in man. Anesthe- in rats following chronic ingani KH, Schimmel TA, siology. 1973; 39(5):537-40. exposure to halothane, McConnell BM, Buro K, nitrous oxide, or methoxy- 67. Albers DD, Leverett CL, Ozelsel TJ. The impact of flurane. Anesthesiology. Sandin JH. Renal failure sevoflurane anesthesia 1978; 48(1):11-6. following prostatove- on postoperative renal siculectomy related to function: a systematic 83. Dahlgren BE. Urinary methoxyflurane anesthesia review and meta-analysis fluoride concentration in and tetracycline — compli- of randomized-controlled mothers and neonates after cated by candida infection. trials. Can J Anaesth. methoxyflurane-nitrous J Urol. 1971; 106:348-50. 2020; 67(11):1595-623. oxide analgesia during labour. Acta Pharm Suec. 68. Blair HA, Frampton 76. Churchill D, Yacoub JM, 1978; 15(3):211-7. JE. Methoxyflurane: a Siu KP, Symes A, Gault review in trauma pain. MH. Toxic nephropathy 84. Cuasay OS, Ramamurthy Clin Drug Investig. after low-dose methoxy- R, Salem MR, et al. Inor- 2016; 36(12):1067-73. flurane anesthesia: ganic fluoride levels in 69. Kharasch ED, Schroeder drug interaction with parturients and neonates JL, Liggitt HD, Park SB, secobarbital? Canad Med following methoxyflurane Whittington D, Sheffels Assoc J. 1976; 114:326-33. analgesia during labor and delivery. Anesth P. New insights into 77. Fiserova-Bergerova Analg. 1977; 65:646-9. the mechanism of V. Changes of fluoride methoxyflurane nephro- content in bone: an index 85. Allison SJ, Docherty PD, toxicity and implications of drug defluorination Pons D, Chase JG. Expo- for anesthetic development in vivo. Anesthesiology. sure to methoxyflurane: (part 1). Anesthesiology. 1973; 28:345-51. Low-dose analgesia and 2006; 105:737-45. occupational exposure. 78. Fiserova-Bergerova V. Aus J Paramed. 2020; 17. 70. Kharasch ED, Schroeder JL, Fluoride in bone of rats Liggitt HD, Ensign D, Whit- anesthetised during 86. Klein NC, Jeffries tington D. New insights gestation with enflurane or GH. Hepatotoxicity into the mechanism of methoxyflurane. Anesthe- after methoxyflurane methoxyflurane nephro- siology. 1976; 45(5):483-6. administration. JAMA. toxicity and implications 1966; 197:1037-9. 79. Wharton RS, Sieven- for anesthetic development piper TS, Mazze RI. 87. Okuno T, Takeda M, (part 2). Anesthesiology. Developmental toxicity Horishi M, Okanoue T, 2006; 105:737-45. of methoxyflurane in Takino T. Hepatitis due 71. Kharasch ED, Hankins mice. Anesth Analg. to repeated inhalation DC, Thummel KE. Human 1980; 59(6):421-5. of methoxyflurane in kidney methoxyflu- subanaesthetic concen- 80. World Health Organiza- rane and sevoflurane trations. Canad Anaesth tion. Inadequate or excess metabolism. Intrarenal Soc J. 1985; 32(1):53-5. fluoride: a major public fluoride production as a health concern. Geneva: 88. Dahlgren BE. Hepatic possible mechanism of World Health Organization. and renal effects of low methoxyflurane nephro- Cited August 19, 2020. concentrations of methoxy- toxicity. Anesthesiology. Available from: https:// flurane in exposed delivery 1995; 82(3):689-99. apps.who.int/iris/bitstream/ ward personnel. J Occup 72. Brown BR, Jr. Shibboleths handle/10665/329484/ Med. 1980; 22(12):817-9. and jigsaw puzzles. The WHO-CED-PHE-EPE- 89. Jacobs IG. Health fluoride nephrotoxicity 19.4.5-eng.pdf effects of patients given

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 88 www.nzma.org.nz/journal REVIEW ARTICLE

methoxyflurane in the methoxyflurane in pediat- 105. Nguyen NQ, Toscano L, pre-hospital setting: a data ric patients. Prehosp Emerg Lawrence M, et al. Portable linkage study. Op Emerg Care. 2011; 15(2):158-65. inhaled methoxyflurane Med J. 2010; 3:7-13. 97. Bendall JC, Simpson PM, is feasible and safe for 90. Therapeutic Goods Middleton PM. Prehos- colonoscopy in subjects Administration (Australia). pital analgesia in New with morbid obesity Database of adverse event South Wales, Australia. and/or obstructive sleep notifications - medicines. Prehosp Disaster Medi- apnea. Endosc Int Open. Cited October 19, 2020. cine. 2011; 26(2):422-6. 2015; 3(5):E487-93. Available from: http:// 98. Wright S. The time has 106. Huang S, Pepdjonovic L, apps.tga.gov.au/Prod/ come to consider an Konstantatos A, Frydenberg daen/daen-entry.aspx alternative to methoxy- M, Grummet J. Penthrox 91. Medicines and Medical flurane use by ambulance alone versus Penthrox plus Devices Safety Authority service paramedics. periprostatic infiltration (New Zealand). Suspected Response. 2013; 40(4):29-35. of local analgesia for analgesia in transrectal medicine adverse reaction 99. Babl FE, Jamison SR, ultrasound-guided search. Cited October 19, Spicer M, Bernard S. prostate biopsy. ANZ J 2020. Available from: Inhaled methoxyflurane Surg. 2016; 86(3):139-42. https://www.medsafe. as a prehospital analge- govt.nz/Projects/B1/ sic in children. Emerg 107. Gaskell AL, Jephcott CG, ADRSearch.asp Med Australas. 2006; Smithells JR, Sleigh JW. 92. Oxer HF. Effects of 18(4):404-10. Self-administered methoxy- ® flurane for procedural Penthrox (methoxy- 100. Babl F, Barnett P, Palmer analgesia: experience in flurane) as an analgesic G, Oakley E, Davidson A. a tertiary Australasian on cardiovascular and A pilot study of inhaled centre. Anaesthesia. respiratory functions in methoxyflurane for proce- 2016; 71(4):417-23. the pre-hospital setting. dural analgesia in children. J Mil Vet Health. 2016; Ped Anesth. 2007; 17:148-53. 108. Ruff R, Kerr S, Kerr D, 24(2):14-20. Zalcberg D, Stevens J. 101. O’Rourke KM, McMaster S, Occupational exposure to 93. St. John (NZ). Clinical Lust KMC. A case of hepati- methoxyflurane admin- procedures and guidelines tis attributable to repeated istered for procedural 2019. Cited May 30, 2020. exposure to methoxyflu- sedation: an observational Available from: https:// rane during its use for study of 40 exposures. Br J www.stjohn.org.nz/ procedural analgesia. Med Anaesth. 2018; 120(6):1435- globalassets/documents/ J Aus. 2011; 194(8):423-4. health-practitioners/clini- 7. 102. Spruyt O, Westerman D, cal-procedures-and-guide- 109. Umana E, Kelliher JH, Blom Milner A, Bressel M, Wein lines---comprehen- CJ, McNicholl B. Inhaled S. A randomised, double- sive-edition.pdf methoxyflurane for the blind, placebo-controlled reduction of acute anterior 94. Wellington Free Ambu- study to assess the safety shoulder dislocation in the lance. Clinical procedures and efficacy of methoxy- emergency department. and guidelines 2019 flurane for procedural CJEM. 2019; 21(4):468-72. - comprehensive edition. pain of a bone marrow Cited May 30, 2020. Avail- biopsy. BMJ Supp Pal 110. Viglino D, Termoz Masson able from: https://www.wfa. Care. 2014; 4(4):342-8. N, Verdetti A, et al. org.nz/assets/What-we-do/ Multimodal oral analgesia 103. Wasiak J, Mahar PD, Paul b8a3986cc7/WFA-CPG-Com- for non-severe trauma E, Menezes H, Spinks prehensive-2019-2022.pdf patients: evaluation of AB, Cleland H. Inhaled a triage-nurse directed 95. Buntine P, Thom O, Babl methoxyflurane for protocol combining F, Bailey M, Bernard S. pain and anxiety relief methoxyflurane, parac- Prehospital analgesia during burn wound care etamol and oxycodone. in adults using inhaled procedures: an Australian Intern Emerg Med. methoxyflurane. case series. Int Wound 2019; 14(7):1139-45. Emerg Med Austral. J. 2014; 11(1):74-8. 2007; 19:509-14. 111. Borobia AM, Collado SG, 104. Lee C, Woo HH. Penthrox Cardona CC, et al. Inhaled 96. Bendall JC, Simpson PM, inhaler analgesia in tran- methoxyflurane provides Middleton PM. Effec- srectal ultrasound-guided greater analgesia and faster tiveness of prehospital prostate biopsy. ANZ J onset of action versus morphine, fentanyl, and Surg. 2015; 85(6):433-7.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 89 www.nzma.org.nz/journal REVIEW ARTICLE

standard analgesia in 114. Worksafe (NZ). Workplace A, Dautheville S, et al. patients with trauma pain: exposure standards Non-interventional study InMEDIATE: a random- and biological exposure evaluating exposure to ized controlled trial in indices. 12th edition. inhaled, low-dose methoxy- Emergency Departments. Cited January 19, 2021. flurane experienced Ann Emerg Med. 2019. Available from: https:// by hospital emergency 112. Porter KM, Dayan AD, www.worksafe.govt.nz/ department personnel in Dickerson S, Middleton topic-and-industry/work-re- France. BMJ Open. 2020; PM. The role of inhaled lated-health/monitoring/ 10(2):e034647. methoxyflurane in acute exposure-standards-and-bi- 117. Stelfox HT, Chua G, pain management. ological-exposure-indices/ O’Rourke K, Detsky AS. Open Access Emerg 115. Safe Work Australia. Health Conflict of interest in the Med. 2018; 10:149-64. monitoring - guide for debate over calcium-chan- 113. Allison SJ, Docherty PD, fluorides. Cited January nel antagonists. New Engl Pons D, Chase JG. Serum 19, 2021. Available from: J Med. 1998; 338(2):101-6. fluoride levels following https://www.safeworkaus- 118. Lundh A, Lexchin J, commencement of tralia.gov.au/system/ Mintzes B, Schroll JB, Bero methoxyflurane for patient files/documents/2002/ L. Industry sponsorship analgesia in an ambulance health_monitoring_guid- and research outcome service. Br J Anaesth. ance_-_fluorides.pdf (review). Cochrane 2020; 125(6):e457-e8. 116. Frangos J, Belbachir Database Syst Rev. 2017; (2).

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 90 www.nzma.org.nz/journal VIEWPOINT

What does abortion law reform mean for primary care practitioners in New Zealand? Emma Macfarlane, Michael Stitely, Helen Paterson

ABSTRACT New Zealand achieved a major sexual reproductive health and rights milestone when abortion ceased to be a crime. Introduction of the Abortion Legislation Act 2020 has significantly changed the way abortion care can be provided in New Zealand, with the potential to improve access, reduce inequities and transform the abortion experience for those people who choose to end their pregnancy. The primary care sector stands to be a key player in the provision of first-trimester abortion care. However, with issues relating to funding, training and access to medications yet to be resolved, the health sector is not yet ready to provide best-practice abortion care within the new legislative framework.

n 23 March 2020 the Abortion Legis- abortion, abortion law reform, primary care, lation Act (AL) 2020 was passed into mid-level health providers. Any references law with the result that abortion in of papers identified through the literature O 1 New Zealand is no longer a crime. There is review that seemed relevant were located now scope for qualified health practitioners and considered for inclusion in the review. to provide abortion care that is evidence This review includes relevant New Zealand based, aimed at reducing inequities in ac- legislation and standards and international cess and more acceptable to pregnant people recommendations on provision of abortion and their whānau. However, in almost in primary care. The aim of this paper is to 12 months since the law reform, little has review the international literature to inform changed in the way that abortion services how optimal, first-trimester abortion care are configured, and we are yet to realise the can be provided in the primary care setting full extent of abortion care within a decrimi- post abortion law reform in Aotearoa nalised environment. New Zealand. A review of the literature was undertaken using the databases Medline and Google Background Scholar. The following key words were used: Abortion is one of the most common gynaecological procedures, and one in four women internationally will have an Figure 1: Key points of Abortion Law Reform in abortion in their lifetime.2 For the year New Zealand ending 2019 the general abortion rate in New Zealand was 13.5 abortions per 1,000 • No statutory test for abortions <20 weeks women (pregnant people) aged 15–44, and gestation 19% of all known pregnancies ended in • Abortions can be provided by a range of abortion.3 The term ‘woman’ is used by the health practitioners AL Act. We consider this to be inclusive of • Provision remains for conscientious objection transgender, gender-fluid, non-binary and • No requirement for licensed premises gender non-conforming people.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 91 www.nzma.org.nz/journal VIEWPOINT

The previous laws governing abortion The new law: there is no statutory in New Zealand were sections 182A–187A test for abortion before 20 weeks of the Crimes Act 1961 and sections 10–46 of the Contraception, Sterilisation and gestation Under the AL Act 2020, health-practitioner Abortion (CSA) Act 1977.4,5 With safe abor- approval is not required for an abortion tions readily available, the only rational before 20 weeks gestation.1 Under previous purpose for criminalising abortion is legislation, two certifying consultants had to deter, punish and place the rights of to agree that the person met the criteria for the fetus over the rights and autonomy having an abortion as specified by section of the pregnant person.6 New Zealand’s 187A of the Crimes Act.4 Most pregnant abortion laws were considered by some people less than 20 weeks had an abortion as a violation of human rights, and in on the grounds that continuing the preg- 2019 Abortion Law Reform Aotearoa New nancy posed a serious danger to life, or Zealand (ALRANZ) brought a case to the physical or mental health.4 The decision Office of Human Rights Proceedings. The to have an abortion was not the pregnant complaint was subsequently withdrawn person’s and there was the potential for an when the AL Act 2020 was passed, as the abortion request to be declined. Under the AL Act addressed the issues raised by AL Act, people less than 20 weeks pregnant ALRANZ.7 who seek an abortion can obtain an abortion 8 The intention of the AL Bill 2019 was to from a suitably qualified health practi- decriminalise abortion and bring the legal tioner.1 This means that up until 20 weeks’ framework for abortion in New Zealand gestation it is the pregnant person’s choice to in line with how other health services are have an abortion. For pregnant people over delivered, and in doing so treat abortion as 20 weeks, the suitably qualified health prac- a health issue rather than a criminal justice titioner must consult with another health 9 issue. Decriminalisation of abortion can practitioner, decide whether an abortion is be defined as not punishing anyone for clinically appropriate and “have regard to; providing or having an abortion, and not all relevant legal, professional, and ethical involving the criminal justice system in standards to which the qualified health prac- deciding who can have an abortion. Above titioner is subject; and the woman’s physical all it means treating abortion like any other health; and mental health; and overall well- 6 health procedure. being; and the gestational age of the fetus.”1 Implementation of the AL Act trans- This means that, once they have been ferred supervision of abortion services pregnant for more than 20 weeks, it is not from the Ministry of Justice to the Ministry solely the pregnant person’s choice to have of Health (MOH). The MOH is now respon- an abortion. The final decision remains with sible for ensuring that abortion care and the health practitioner. counselling are provided according to standards published by the Director-General.1 Abortion care can be provided by a Currently, these are the Interim Standards range of health practitioners for Abortion Services in New Zealand The AL Act 2020 states that abortions (the ‘Interim Abortion Standards’), which, can be provided by a “qualified health published in April 2020, are an amended practitioner,” as defined by the Health 10 version of the 2018 Abortion Standards. Practitioners Competence Assurance Act.1 Abortion has been included in the draft However, what this actually means for prac- Health and Disability Services Standards titioners requires clarification. Review.11 The World Health Organization (WHO) Recently, the MOH released a report from advocates the shifting and sharing of district health boards (DHBs) on abortion abortion care from specialist providers service provision post abortion-law reform. to mid-level providers, such as registered The results identify issues relating to work- nurses, nurse practitioners and midwives. force development, training and the way in The WHO recommends that these health which services are delivered. These issues practitioners can safely provide EMA and continue to negatively impact on equitable aspiration abortion in the primary care access and timeliness of abortion care.12

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 92 www.nzma.org.nz/journal VIEWPOINT

setting.13 The rationale for training mid-level pregnant people can be monitored as outpa- health providers in abortion care is to tients.16,17 EMA providers can be trained in increase access to abortion and reduce the manual vacuum aspiration as a back-up for burden of cost to the healthcare system, a failed EMA, and most EMA complications while not compromising on safety.14 can be safely managed in primary care.17 Health practitioners who are not abortion However, a few complications associated 17 providers can refer the pregnant person to with EMA require emergency referral. an abortion provider. However, section 13 Therefore it is important to have referral 14 of the AL Act 2020 states that provision of pathways in place to secondary care. abortion care is not contingent on a referral Primary health providers, particularly in from another health practitioner and that rural areas, have the potential to increase pregnant people can self-refer.1 access to EMA. Pregnant people living in rural or remote regions rely on primary What about conscientious objection? healthcare services to meet their basic Sections 14 and 15 of the Abortion health needs. Strengthening primary care Legislation Act 2020 maintains a health prac- is a way of reducing health inequities.17 titioner’s legal right to conscientiously object Furthermore, abortion may be more to provide or assist with abortion, sterili- acceptable when provided by a primary sation, contraception or advice regarding health practitioner. One study of pregnant pregnancy options.1 The key differences people attending either a primary care between the AL Act 2020 and the CSA Act university clinic or a free-standing abortion 1977 are that conscientious objectors must clinic found that most preferred to receive now disclose their stance as soon as possible early abortion care with their trusted and provide contact details of the closest primary provider. The authors suggest provider. However, if the conscientious integrating early abortion care into primary objection causes an unreasonable disruption care to improve access and health outcomes to the service, an employer can take steps, for people with an unintended pregnancy.18 including provision of less favourable terms The Interim Abortion Standards of employment, termination or retirement.1 recommend that people should not be It could be seen that the new legislation required to travel more than two hours does go some way in addressing the impact to access an abortion.10 Providing first-tri- of conscientious objection on abortion mester abortion in primary care may services. However, it is less likely to address significantly decrease the travel some people the impact in primary care where prac- presently undertake to have an abortion, titioners provide a range of services and and also allow for provision of EMA via whose patients may not be aware of their telemedicine. stance. Unless conscientious objectors are Under the previous law, administration of made to publically disclose their objection, abortion pills was defined as the abortion.19 there is always the potential for pregnant People were required to be observed taking people to be denied care. the medication on a licensed premise, and as the most effective regime is to take the Where abortions can be provided medications 36–48 hours apart, people had Before the introduction of the AL Act to return to the clinic for the second dose.15 2020, pregnant people were required to be Removing the requirement for abortion to referred to one of 27 licensed institutions.5 be provided on licensed premises makes The new legislation does not limit where self-managed abortion possible. abortions can be provided. Early abortion A recent Cochrane review shows that services provided in primary care are safe self-managed EMA is as effective as provid- 13 and effective. Early medical abortion er-administered EMA and is acceptable to (EMA) involves taking two medications women.20 A self-managed EMA allows a 36–48 hours apart to end the pregnancy. The pregnant person to take the abortion pills recommended medications are mifepristone at a time and place that suits them without 15 and misoprostol. EMA is well suited to supervision. It also means that the person primary care as it does not require proce- can self-assess completion of the procedure dural training or technical facilities, and

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 93 www.nzma.org.nz/journal VIEWPOINT

by using appropriate pregnancy test kits support to new providers.23 Specialist and symptom checklists.20 The aim of self- obstetricians in Nepal provided training, managed abortion is to increase access and oversight and support to the mid-level acceptability, and it has the potential to health providers who subsequently reduce demand on health services.20 The provided abortion care in the commu- role of the primary care clinician in self- nities.22 Specialist providers will always managed EMA is to provide assessment, be needed to provide emergency care, and information and advice to the pregnant although their roles in abortion care will person on how to take the medication be different, they will remain crucial to correctly, how they can self-assess progress the provision of safe and effective abortion and how they can access help and infor- care.23 20 mation as needed. Perhaps the most important factor in the success of expanding health-worker roles by What is required for successful task shifting is their willingness to provide implementation of abortion into abortion care. Willingness is influenced by primary care? a number of factors, such as personal views Clinicians should receive training on abortion, the method of abortion they are and support to competently provide asked to perform, gestation of the fetus and safe abortion care. Primary care should health-provider perceptions of their roles also have tool kits, which could include as preservers of life. One way that has been assessment templates, clinical deci- shown to be successful in increasing will- sion-making support tools and information ingness to provide abortion care is through 24 and summary-of-care templates, in case the values-clarification workshops. pregnant person presents acutely to another It will be important for health regulatory health provider.16 bodies to ensure that abortion care is clearly 23 Prior to the repeal of the Eighth included in appropriate scopes of practice. Amendment, Ireland essentially did not The Midwifery Council of New Zealand has provide abortions, and subsequent to provided a clear statement that abortion the repeal, practitioners in primary and is within the midwifery scope of practice. secondary care needed to rapidly acquire Midwives, as authorised prescribers, can the skills to provide the service. The prescribe the medications required for government was criticised for its lack of abortion and, with appropriate training, support and leadership.21 Nepal’s success in can perform surgical abortion to the extent 25 making safe, legal abortion widely available allowed by their scope. has been attributed to a number of factors, Essential to provision of EMA is the ability including commitment and leadership by to provide the required medications with the Nepalese government and a compre- as few barriers as possible. Nurses often hensive approach to implementation of provide medications under standing orders abortion services. Abortion programmes as per the Medicines (Standing Orders) and policies were based on international Regulations 2002, whereby non-prescribers evidence for best-practice abortion care can administer or supply specific medi- and the training of mid-level providers to cations according to a written instruction increase the number of clinicians able to issued by an authorised practitioner, provide care to a wider geographical area.22 nurse practitioner or optometrist.26 In New Zealand currently does not have a practice, standing orders can pose a signif- primary care-based training programme icant amount of work for an organisation for abortion, but the New Zealand College to comply with the requirements of the of Sexual and Reproductive Health is legislation. A scenario exists for a non-pre- developing an online learning module for scriber who wishes to provide EMA but provision of EMA. lacks a supportive prescriber to administer For task shifting of abortion care to a standing order. One solution would be to non-specialist providers to occur we need develop a national standing order for miso- the support and leadership of our current prostol and mifepristone and a network of abortion providers, who will be crucial in supportive prescribers to administer the leading training and providing ongoing standing order. A further solution would

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 94 www.nzma.org.nz/journal VIEWPOINT

be to ensure that the medications can be contracts to allow time for the Interim prescribed by all non-medical prescribers. Abortion Standards to be updated and the There are a range of non-medical establishment of national clinical guidelines. prescribers in New Zealand. Under the It is envisaged that these restrictions will interpretations of the Medicines Act 1981 then be removed, allowing access to funded authorised prescriber means “a nurse prac- EMA medications on PSO for all suitably 31 titioner; or an optometrist; or a practitioner; qualified prescribers. or a registered midwife; or a designated Currently, the brand of misoprostol prescriber,” where a “designated prescriber” imported into New Zealand is not approved is a registered health professional who has by Medsafe for use as an abortifacient. undertaken further education and training Therefore, it is prescribed by authorised in order to be able to prescribe.27 Currently, prescribers only under Section 25 of the there appears to be confusion regarding Medicines Act 1981.27 Ongoing it will be whether a designated prescriber is an autho- important to gain clarification regarding rised prescriber or not. For example, in an the legal status of designated prescribers as overview of non-medical prescribing in New authorised prescribers and for mifepristone Zealand, designated prescribers are clearly to be added to the schedule of medicines distinguished from authorised prescribers.28 they can prescribe. Without this move, However, this is in contrast to the interpre- designated prescribers will be dependent tations section of the Medicines Act 1981, on standing orders, which may impact on which lists designated prescribers among equity of access to abortifacients. 27 authorised prescribers. How abortion will be funded in primary There are two levels of registered care requires clarification. Presently the nurse (RN) designated prescribers in New Primary Maternity Services Notice Review Zealand: (1) prescribing in primary health 2021 specifically excludes funding of abortion and speciality teams, and (2) prescribing (termination of pregnancy). It also excludes in community health.29 RN prescribers in nurses from being maternity providers which primary health and specialty teams have poses a barrier to provision of autonomous undertaken further training, including a first trimester abortion care by nurses.32 Nursing Council approved postgraduate diploma in RN prescribing for long-term conditions. They can prescribe from a list of Figure 2: What is required for first trimester medications as per their area of practice and abortion to be provided in primary care. competency.30 The list includes misoprostol but not mifepristone. The second level of RN • Updated abortion standards prescribers, community prescribers, work • Appropriate training and support for within DHBs or other health organisations abortion providers and undergo a recertification programme • Access to EMA drugs on PSO in primary care to become prescribers. The medications list • Support for non-prescribers to provided EMA at this level of prescribing is more limited and does not include either misoprostol or mifepristone.29 A further way of reducing barriers to Conclusion One of the more pressing requirements EMA is for health practitioners to supply to providing safe abortion care in the the medications from a Practitioner Supply community is the provision of compre- Order (PSO). PHARMAC announced on 2 hensive and accessible training for health July 2020 that mifepristone and misoprostol practitioners. Unless there is adequate would be listed in Section B of the pharma- and appropriate funding for the primary ceutical schedule effective from1 August care sector, including for the training of 2020. This means that these medications midwives, nurses and nurse practitioners, can be supplied on PSO so that pregnant there will be no incentive to undertake the people are not required to go to a pharmacy training nor the provision of abortion care, for dispensing. However, this is currently with the result that access to safe abortion temporarily restricted to Family Planning will not be improved. Fundamental to equi- clinics and abortion providers with DHB table provision of EMA is access to funded

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 95 www.nzma.org.nz/journal VIEWPOINT

mifepristone and misoprostol by a range In passing the AL Act 2020, New Zealand of qualified health practitioners. We need achieved a major milestone in sexual and clarification regarding the legal status of reproductive health and human rights. With designated prescribers under the Medi- new legislative and regulatory frameworks, cines Act 1981 to prescribe misoprostol as health practitioners in primary care have a section 25 drug and, for misoprostol and the potential to be key players in shaping mifepristone to be added to the schedule of the future of abortion care. Further research medicines for both levels of RN designated is required to develop a framework for prescribing. Trained non-prescribers should provision of optimal first trimester abortion be supported to provide EMA via a national care in the primary sector in Aotearoa standing order. New Zealand.

Competing interests: Ms Macfarlane reports: I am a committee member of Abortion Providers Group Aotearoa New Zealand (APGANZ) and a member of The College of Sexual and Reproductive Health. I was actively involved in campaigning in support of abortion law reform. While undertaking this research I was the recipient of the University of Otago Dunbar Scholarship which paid for my PhD fees and a monthly stipend. Dr Paterson reports: I received professional fees from Southern District Health Board and professional fees from Ministry of Justice, outside the submitted work; I was the Chair of Abortion Providers Group Aotearoa and a member of the committee, sat on the Ministry of Health Abortion Standards working group and presently act in an un-payed advisory role for the Ministry of Health abortion guidelines development. I am a member of The College of Sexual and Reproductive Health and have been funded to attend an educational conference by Bayer. I and Michael Stitely hold a contraception-related patent. Acknowledgements: This paper was written while the author Emma Macfarlane was the holder of the University of Otago, Dunbar Scholarship. Author information: Emma Macfarlane: Lecturer, Department of Women’s and Children’s Health, Otago Medical School, University of Otago, Dunedin. Associate Professor Michael Stitely: Head of Department, Department of Women’s and Children’s Health, Otago Medical School, University of Otago, Dunedin. Dr Helen Paterson: Senior Lecturer, Department of Women’s and Children’s Health, Otago Medical School, University of Otago, Dunedin. Corresponding author: Emma Macfarlane, Department of Women’s and Children’s Health, Otago Medical School – Dunedin Campus, PO Box 56, Dunedin 9054, +64 3 470 9750 [email protected]. URL: www.nzma.org.nz/journal-articles/what-does-abortion-law-reform-mean-for-primary-care- practitioners-in-new-zealand

REFERENCES 1. Abortion Legislation Act 3. Stats NZ: Tatauranga from: http://www. 2020. Public Act. 2020 No Aotearoa. Abortion legislation.govt.nz/act/ 6. [Available from: http:// statistics: year ended public/1961/0043/149.0/ www.legislation.govt. December 2019 [Inter- DLM327382.html. nz/act/public/2020/0006/ net]. 2020 [cited 2020 5. Contraception, Sterilisa- latest/LMS237550.html. 26 Sep]. Available from: tion, and Abortion Act 2. Amnesty International. https://www.stats.govt. 1977. Public Act. 1977 Key facts on abortion nz/information-releases/ No 112. New Zealand [Internet]. 2020 [updated abortion-statistics-year-end- [Available from: http:// 4 Mar 2020; cited 2020 ed-december-2019. www.legislation.govt.nz/ 15 Aug]. Available from: 4. The Crimes Act 1961. act/public/1977/0112/10.0/ https://www.amnesty.org. Public Act. 1961. No 43. DLM17680.html. nz/key-facts-abortion. New Zealand [Available

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 96 www.nzma.org.nz/journal VIEWPOINT

6. Berer M. Abortion law and 11 Dec]. Available from: 18. Logsdon MB, Handler A, policy around the world: https://www.google.com/ Godfrey EM. Women’s in search of decriminaliza- url?sa=t&rct=j&q=&es- preferences for the location tion. Health and human rc=s&source=web&c- of abortion services: a rights. 2017;19(1):13. d=&ved=2ahUKEw- pilot study in two Chicago 7. Abortion Law Reform iSu8C-y8ftAhVWxDgGHRe- clinics. Matern Child New Zealand [ALRANZ]. 3C7YQFjAAegQIARAC&ur Health J. 2012;16(1):212-6. ALRANZ’s complaint to the l=https%3A%2F%2F- 19. Ministry of Justice. Human Rights Commission www.health.govt. Standards of care for [Internet]. 2017 [cited nz%2Fsystem%2Ffiles%2F- woman requesting 2020 Aug 15]. Available documents%2Fpages%- abortion in Aotearoa from: http://alranz.org/ 2Fabortion-services-sum- New Zealand [Internet]. human-rights-complaint/. mary-dhb-quarterly-sur- 2018 [cited 2020 15 Aug]. vey-responses-dec20. 8. New Zealand Parliament: Available from: https:// pdf&usg=AOvVaw1x-PE1w- Pāremata Aotearoa. www.justice.govt.nz/assets/ 9FzWUjwJEA7Fiwa. Abortion Legislation Bill Documents/Publications/ [Internet]. 2019 [cited 2020 13. World Health Organization Standards-of-Care-2018.pdf. 14 Aug]. Available from: DoRHaR. Safe abortion: 20. Gambir K, Kim C, Necastro https://www.parliament. technical and policy guid- KA, Ganatra B, Ngo TD. nz/en/pb/bills-and-laws/ ance for health systems Self‐administered versus bills-proposed-laws/ [Internet]: World Health provider‐administered document/BILL_89814/ Organisation; 2012 [cited medical abortion. Cochrane abortion-legislation-bill. 2020 25 April]. Available Database of Systematic from: https://apps.who.int/ 9. Te Aka Matua o te Ture: Reviews. 2020(3). iris/handle/10665/70914. Law Commission. Minis- 21. Taylor M, Spillane A, terial briefing: alternative 14. World Health Organization. Arulkumaran S. The approaches to abortion Health worker roles in Irish Journey: Removing law: ministerial brief providing safe abortion the shackles of abortion [Internet]. 2018 [cited 2020 care and post-abortion restrictions in Ireland. 15 Aug]. Available from: contraception [Internet]. Best Practice & Research https://www.lawcom. Geneva: World Health Clinical Obstetrics & govt.nz/abortion. Organisation; 2015 [cited Gynaecology. 2020;62:36-48. 2020 12 April]. Available 10. Ministry of Health: 22. Samandari G, Wolf M, from: https://www.who.int/ Manatū Hauora. Interim Basnett I, Hyman A, Ander- reproductivehealth/publi- Standards for Abortion sen K. Implementation of cations/unsafe_abortion/ Services in New Zealand legal abortion in Nepal: a abortion-task-shifting/en/. [Internet]. 2020 [cited model for rapid scale-up of 2020 15 April ]. Available 15. Istar Limited. Mifegyne high-quality care. Reprod from: https://www.health. 200mg tablets data sheet Health. 2012;9(1):7. 25 January 2021. [cited govt.nz/publication/ 23. Kim C, Sorhaindo A, 2021 23 Feb]. Available interim-standards-abor- Ganatra B. WHO guide- from: https://www.google. tion-services-new-zealand. lines and the role of the com/search?client=fire- 11. Hauora MoHM. Health physician in task sharing fox-b-d&q=medsafe+mife- and disability services in safe abortion care. pristone+data+sheet standards and review Best Pract Res Clin Obstet [Internet]. 2020 [202 16 16. Beaman J, Schillinger D. Gynaecol. 2020;63:56-66. Responding to Evolving December]. Available 24. Glenton C, Sorhaindo Abortion Regulations from: https://www. AM, Ganatra B, Lewin S. - The Critical Role of health.govt.nz/our-work/ Implementation consid- Primary Care. N Engl J regulation-health-and-dis- erations when expanding Med. 2019;380(18):e30. ability-system/ health worker roles to certification-health-care-ser- 17. Iyengar SD. Introducing include safe abortion care: vices/services-standards/ medical abortion within a five-country case study health-and-disability-ser- the primary health synthesis. BMC Public vices-standards-review. system: comparison with Health. 2017;17(1):730. other health interven- 12. Hauora MoHM. Abortion 25. Midwifery Council. tions and commodities. services: Summary of DHB The Midwifery scope Reprod Health Matters. quarterly survey responses of practice: Abortion 2005;13(26):13-9. [Internet]. 2020 [cited 2020 services (March 2020)

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 97 www.nzma.org.nz/journal VIEWPOINT

FAQs [Internet]. 2020 Zealand: an overview 2020 4 Aug]. Available [cited 2020 4 August]. of prescribing rights, from: https://gazette.govt. Available from: https:// service delivery models nz/notice/id/2017-gs2787. www.midwiferycouncil. and training. Therapeutic 31. PHARMAC. Modifying health.nz/midwives/ advances in drug safety. funded access to mife- practice-issues/midwifery- 2017;8(11):349-60. pristone and misoprostol scope-practice-abortion- 29. Nursing Council of New [Internet]. 2020 [cited 2020 services-march-2020-faqs. Zealand: Te Kaunihera 4 August ]. Available from: 26. Medicines (Standing Tapuhi o Aotearoa. Tūtohu https://www.pharmac. Order) Regulations 2002 Kua Rēhitatia: Registered govt.nz/news/notifica- New Zealand [Available Nurse Prescribing [Inter- tion-2020-07-02-mifepri- from: http://www.legisla- net]. [cited 2020 4 August]. stone/. tion.govt.nz/regulation/ Available from: https:// 32. Ministry of Health: public/2002/0373/10.0/ www.nursingcouncil. Manatū Hauora. Primary DLM170135.html. org.nz/Public/Nursing/ Maternity Services Notice: 27. Medicines Act 1981. Public Nurse_prescribing/ Draft For Consultation Act. 1981 No 118. New NCNZ/nursing-section/ [Internet]. 2020 [cited Zealand [Available from: Nurse_Prescribing.aspx. 2020 22 Sep]. Available http://www.legislation.govt. 30. New Zealand Gazette from: https://www.health. nz/act/public/1981/0118/ Te Kāhiti o Aotearoa. govt.nz/system/files/ latest/whole.html. Medicines (Designated documents/publications/ 28. Raghunandan R, Tordoff Prescriber) – Registered primary-maternity-ser- J, Smith A. Non-medical Nurses Prescribing in vices-notice-draft-con- prescribing in New Community Health Notice sultation-sep20.pdf. 2017 [Internet]. 2017 [cited

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 98 www.nzma.org.nz/journal VIEWPOINT

ABSTRACT We examined the documentation underlying the decision to permit the Southern District Health Board (SDHB) to join the National Bowel Screening Programme (NBSP) at a time when it was not providing an adequate colonoscopy service for symptomatic patients. A coordinated sequence of relevant Official Information Act 1982 (OIA) requests was lodged with the New Zealand Ministry of Health (MoH), which is responsible for determining the readiness of district health boards (DHBs) to join the NBSP. However, the MoH OIA process was obfuscating, unduly long and responded only after they anticipated imminent intervention by the Office of the Ombudsman. The amount of information provided was massive, partly irrelevant and presented in an inconvenient format. It revealed that the MoH readiness process was incomplete, and permission for the SDHB to join the NBSP was given prematurely without following due process and despite concerns expressed by some MoH staff. Subsequently, the MoH has failed to admit that they made errors in this case or have any weaknesses in their readiness assessment process. The MoH readiness process failed to determine that the SDHB was not ready to join the NBSP in 2018. Concerns have been expressed in the public media that such failures have occurred with the assessment of other DHBs. The process needs to be overhauled or replaced before further permissions are granted to DHBs. Requests for information under the OIA from the MoH, and similar public entities and agencies subject to the OIA, are too easily deferred, derailed or declined. The OIA is in need of revision.

hree independent external reviews Programme (NBSP). As a result, in April 2017, showed that between 2013 and 2018 following a request from the Ministry of Tthe Southern District Health Board Health (MoH) in December 2016, the SDHB (SDHB) colonoscopy service delayed or submitted a draft business case for inclusion denied colonoscopies to numerous symp- in the NBSP Business Case for 2017/2018 tomatic patients, with adverse clinical to Treasury for Budget 2017. Funding was outcomes for some.1–3 The SDHB Board has subsequently approved, and in April 2017 subsequently accepted responsibility for the SDHB began preparing for an April 2018 these unacceptably poor outcomes and has ‘go-live’ for the screening programme. 4 apologised to all those affected. The NBSP Business Case 2017/20185 During this same period, the SDHB identified six evaluation areas to be used wished to join the National Bowel Screening to assess district health board (DHB) read-

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 99 www.nzma.org.nz/journal VIEWPOINT

iness for delivery of the bowel screening Official Information Act 19826 (OIA) were programme (Table 1). lodged with the MoH (Appendix Table 1). To assess the SDHB’s readiness to begin These were refused on the grounds that the offering bowel screening to their population, requests, as worded, were for “a very large the MoH’s NBSP team monitored progress volume of information and may be refused and achievement against these criteria, via a under section 18(f) of the Official Infor- series of self-assessment reports, document mation Act 1982, because the information reviews and site visit-based assessments requested could not be made available by NBSP personnel. After going through without substantial collation or research.” this process, permission for the SDHB to It was suggested by the MoH that by commence delivery of the NBSP to its eligible narrowing the scope of the requests, either population was granted by the NBSP Lead- by selecting a shorter time frame and/or ership and Governance Groups on 18 April specifying a topic, the information may be 2018. The service went live in the SDHB provided. However, even with the requests region on 24 April 2018. Given the state of the narrowed, the MoH may still have required SDHB colonoscopy service and some related them to be narrowed further, extend the colorectal cancer (CRC) clinical management amount of time necessary to respond or services, the authors of this article feel it is even refuse the requests. their professional responsibility to ask how Accordingly, a series of refined requests permission was granted under the circum- was then made. This second series focussed stances that prevailed at the time. on two main areas of interest: (i) the introduction and operationalisation of Requesting information under the the Colonoscopy Waiting Times Indicator Official Information Act (CWTI), as the only metric of colonoscopy In order to systematically investigate performance management information the adequacy of the assessment process, a on which DHB readiness to assume bowel series of requests for information under the screening would be assessed; and (ii) the

Table 1: NBSP Readiness Assessment Evaluation Areas.

Criteria Assessment Area Governance, leadership, • Leadership, accountabilities, performance monitoring and management management • Production plan • Staffing • Quality standards

Promotion and equity • Promotion and equity

Preparing primary care • DHBs • PHOs • GPs

Referral and pre-assessment • Referral and pre-assessment

Information systems and • Information systems and resources resources

Colonoscopy • Staffing, training and supervision • Management of family history • Colonoscopy • Capacity to undertake investigations • Processes for continuous quality improvement • Facilities • Reporting

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 100 www.nzma.org.nz/journal VIEWPOINT

National Bowel Screening Programme at The documents obtained the SDHB, with particular emphasis on Despite earlier communication with the the identification, awareness and miti- MoH representative handling these requests gation of risk to colonoscopy access for regarding the file size of the aggregated symptomatic patients arising from either documents and the possibility of providing of these, as evidenced by reports, meeting them electronically, the entire document minutes, emails or discussions. The time- set was subsequently printed and then frame covered by the OIA requests was couriered. There are a number of remedies reduced by half as a result of re-scoping to the issue of email file size limitations, any and refining. Individual questions were of which would have preserved the elec- assigned an indicative priority to allow tronic format. Instead, the format in which the MoH to stratify the timing of their the documents were received defied mean- responses according to their resources ingful analysis. This required the entire (Appendix Table 2). document set to be scanned to portable The second series of refined requests was document format (.pdf), then converted also declined under section 18(f) of the OIA to optical character reader (OCR) format, 1982 because the information requested in order to become fully searchable docu- could not be made available without ments. Trial and error revealed the best substantial collation or research. However, resolution settings in the OCR conversion following intercession by the Office of the process to overcome the issue of pages Ombudsman, and after an understandable being watermarked with ‘Released under slowing of the process due to the necessity the Official Information Act 1982’. It is clear of dealing with the COVID-19 pandemic, that receiving documents of this volume and 3,516 pages of documents (in hard copy in hard-copy format precludes any useful only) were received from the MoH in August analysis for most people most of the time. 2020, relating to the OIA requests made to The documents were then searched the MoH. The lengthy process to obtain any for key words and phrases. The context information under the OIA is detailed else- in which the word or phrase occurred where (Appendix Table 3). could then be assessed and the containing The 3,516 pages received constitute document identified for further analysis a partial response to the information as required. After further analysis, a ‘once requested of the MoH. For example, over’ read of documents assessed as low no information was provided on the relevance was undertaken to ensure all recollections of ministers of health, direc- relevant information had been seen. Much tors-general of health or MoH staff of of the information provided was irrel- discussions relating to colonoscopy access evant to the question of the identification for symptomatic patients within SDHB, as and management of risks to provision of had been requested. It is worth noting that clinically appropriate levels of diagnostic in the time period covered by these requests colonoscopy to symptomatic patients, in (2016 through 2019) there were, within the either a pre- or post-bowel screening SDHB SDHB, ongoing concerns regarding access environment. Unsurprisingly, copious pages to colonoscopy raised on multiple occa- of National Bowel Screening communica- sions with the SDHB Chief Executive Officer tions/promotional material, which were (CEO) and Chief Medical Officer.1 Given included, failed to shine any light on risk the ‘no surprises’ expectations in the CEO– awareness or risk management strategy in MoH relationship, it seems unlikely that no this regard. such discussions occurred. A question as to whether monitoring of referral trends Deficiencies in the plan for the for symptomatic patients requiring colo- SDHB to join the NBSP noscopy was undertaken as preparation for The ‘onboarding’ process driven by SDHB implementation went unanswered. A the MoH included a requirement for the number of other questions were apparently SDHB to provide a colonoscopy production ignored, with no rationale for a non-re- plan with a six-month horizon. The only sponse provided. production plan included in the document

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 101 www.nzma.org.nz/journal VIEWPOINT

set is populated out to the end of June for colonoscopy. Also, in 2017 the SDHB 2018, just two months after go-live. The surveyed the views of their endoscopy template beyond that date is empty, yet it is services users. Here 32.4% of respondents precisely at this time that the first increase were aware of patients they thought came in demand for colonoscopy arising from to harm because of declined endoscopies. bowel screening participants with positive None of these concerns are evident in any results would be experienced. There is of the documentation provided under the no modelling of the increased volumes OIA. That the SDHB was already failing required by radiology, surgery or oncology to fully assess symptomatic patients in a arising from additional cancers detected clinically timely manner, before the bowel through the screening programme. The screening programme was introduced, was information obtained under the OIA was not revealed. devoid of evidence of capacity to support additional volumes in diagnostic and Concerns about the DHB capacity treatment services. This is a concern, as a to join the NBSP formal review by the University of Otago of From the emails provided it is apparent the SDHB Department of Surgical Sciences that there were significant concerns among in June 2017 was informed by staff that the MoH NBSP team in the week immedi- access to both elective and acute operating ately prior to NBSP granting permission for theatres on the Dunedin site resulted in the SDHB to go live with screening in their patient delays and suboptimal treatment region. These concerns related to the ability 7 experiences. of the SDHB to provide sufficient capacity The specific risk of reduced access to colo- for both NBSP participants and symptomatic noscopy for symptomatic patients as a result patients, in terms of colonoscopy, radiology, of the introduction of bowel screening was surgery, medical oncology treatment not obviously mitigated through the SDHB and radiation oncology treatment. These NBSP implementation process. Instead, there concerns arose after months of preparation appears to be reliance solely on the CWTI and site-readiness assessments undertaken target performance. It is now understood by the MoH NBSP team. In an email from 11 that the CWTI can easily be manipulated April 2018, the Clinical Lead for NBSP asked by a number of means, including applying NBSP staff to obtain some reassurance from direct access referral criteria inappropri- SDHB that “symptomatic patients will not be ately to specialist referrals, applying the disadvantaged in any way.” The documents criteria inconsistently and even choosing to provided did not include a response to that decline referrals using tone and language request. that discourages referrers from future Several MoH staff emails to SDHB referrals to the service. Pre-implementation personnel stressed the need for a letter referral management practices were not of assurance on these matters from the assessed, and the impact of these on the SDHB CEO, prior to seeking approval five apparent performance of the endoscopy days later for SDHB to go live with bowel service was therefore never considered nor screening. A letter was received by the MoH associated risks identified. NBSP team from the CEO of SDHB (dated Referral management practice was not 10 April 2018) in which he advised that audited or otherwise reviewed, though a surgery, medical oncology and radiation document provided suggests that a national oncology staff were committed to making change to data collection by the MoH the programme work and he was satisfied meant that the rate of declined colonoscopy they will be able to provide the additional referrals would have been known. It would volumes that the programme will generate. also have been known that the SDHB No CEO assurance was provided for colo- declined referral rate was high compared noscopy or radiology capacity. Subsequent to other DHBs.3 From 2016 there were to receiving this letter, the Clinical Lead for repeated formal written complaints to the NBSP at the MoH expressed her concerns SDHB CEO from senior surgeons concerned (in an email dated 12 April 2018 to the MoH at the difficulty they were experiencing NBSP Programme Director and the Imple- getting patients they referred accepted mentation Manager – Bowel Screening at the

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 102 www.nzma.org.nz/journal VIEWPOINT

National Screening Unit) that the SDHB CEO OIA. For many people wishing to use the letter did not say “without compromising the OIA mechanism to make enquiries of public waiting times for symptomatic patients— institutions, the net confounding effect of that is the ethical concern we have.” these tactics would be insurmountable. The authors conclude, along with others, The decision to join the NBSP including its principal sponsor, that the OIA In a memo requesting permission from 1982 is no longer fit-for-purpose and needs NBSP Governance and Leadership Groups to be revised.8–10 for SDHB to go live with screening, NBSP We, the authors, stress that the evidence personnel note having received a letter from for this investigation was obtained from the SDHB CEO giving assurance that there the MoH under the conditions of the OIA. will be sufficient colonoscopy capacity. The We assume that their answers were correct letter obtained under OIA offers no such and complete. On the evidence, we conclude assurance of sufficient colonoscopy capacity. the process to assess SDHB capacity to Despite this, permission was granted on assume NBSP did not ask critical questions. 17 April 2018—in our opinion inappropri- Further, the SDHB was not required by the ately—and screening commenced in the assessment process to disclose relevant SDHB region on 24 April 2018. issues, and apparently did not choose to The MoH was asked to confirm if the CEO disclose voluntarily. letter of assurance provided under OIA was the only written assurance as to capacity Analysis of the process to join availability obtained from the SDHB in the the NBSP process to commence bowel screening in It has become clear that the MoH process that region. A letter of reply from the MoH for assessing the readiness of a DHB to join (Population Health and Prevention) in the NBSP was flawed and should be made October 2020 (Appendix Figure 1) confirmed more open and transparent. There is a that: (i) there is no additional written corre- need for specific targets, and their method spondence the MoH received from SDHB of calculation, to be identified and consis- beyond what was sent to us in August 2020; tently met before screening is implemented. (ii) “the decisions relating to SDHB were Any additional clinical load on a DHB must taken by an earlier leadership team so [they be considered in the context of its current were] unable to provide further context”; ability to meet the needs of its population. and (iii) “the delivery by SDHB of a consis- In the case of the SDHB, uncovering the tently high-performing bowel screening true level of unmet need in symptomatic programme, exceeding national targets patients would have allowed an oppor- would suggest the confirmation of readiness tunity to ensure safeguarded access for was sound.” these patients to colonoscopy as clinically indicated. Instead, reviews have found that Problems encountered with the OIA their access and clinical outcomes were The authors have confidence in their jeopardised.3,4 analysis of the material provided under the Further, the authors believe that signif- OIA, and in the conclusions drawn. They are icant value and safety would be added to experienced in analysis of complex infor- the NBSP-onboarding assessment process mation and had access to the technology by ensuring that it includes benchmarking and skills required to ‘drill-down’ through (i) the rate of declined colonoscopy referrals what was provided, even though the format with other DHBs, and considering this rate was not analysis-friendly. The unduly long against the known CRC incidence rate for process required to obtain the requested the DHB population, and (ii) the proportion information fails to diminish the value of of new CRCs diagnosed via emergency analysis in this instance—in fact, it empha- department presentation. Additionally, DHBs sises necessity of the analysis. The authors should be required to disclose complaints are aware that the intent of the OIA 1982 and issues raised with DHB management is sometimes undermined by the obfus- pertaining to endoscopy service provision cation, delay and deliberate barriers used in the years preceding NBSP-onboarding by some entities and agencies subject to the assessment.

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 103 www.nzma.org.nz/journal VIEWPOINT

There should be recognition, at the level bowel screening programme suggests of government, of the potential for health that the MoH’s permission for screening targets to incentivise perverse behaviours to begin there was correct. That opinion by organisations and individuals. The ignored or ascribed no importance to any reliance on the CWTI as the sole metric of possible adverse clinical consequence that performance of colonoscopy services should the concomitant restrictions on colonoscopy cease, or at least be coupled with a more access had for numerous symptomatic transparent evaluation of performance, patients in the SDHB region. The deficiencies which looks for and reports on unmet clearly revealed in subsequent reviews1–3 need. This is particularly important given outline the serious consequences of the poor that some other DHBs are yet to commence decision-making process. bowel screening, and the same failings of The introduction of new health policy readiness assessment may produce similar and services must be accompanied by adverse outcomes to those seen within well-defined lines of responsibility and the symptomatic population of the SDHB. accountability, particularly in environments Concerns about these possible failings have of constrained resource. Where new policy 11 been recently aired in the public media. or service provision creates inequities in access, as this innovation has, clear lead- Need for better planning before ership and accountability for remediation is DHBs join the NBSP essential. There should be greater planning capa- The potential for harm to symptomatic bility within both the MoH and DHBs patients resulting from the additional when introducing new services. It is noted demand on already limited resources occa- that the preliminary groundwork for sioned by the introduction of the National bowel screening in New Zealand began Bowel Screening Programme warrants over twenty years ago, and workforce further research. development was considered a critical Good bowel screening programmes pre-requisite at that time. Despite this early and timely colonoscopy for symptomatic awareness, successive governments have patients are equally important in the battle failed to prioritise workforce development against CRC. This is particularly so in a in any meaningful way.1 The opportunity country such as New Zealand where CRC to use the intervening years to build a is so common. Indeed, screening should specialist workforce sufficient for the task have started here many years earlier, of bowel screening was wasted. The intro- should have a lower positive test threshold, duction of a national bowel screening and should provide coverage starting at a programme has not been effectively planned younger age, particularly for Māori. Both for in this regard. It is hoped that current screening and symptomatic colonoscopy plans to mitigate the resourcing deficiencies services must be adequately resourced will be successful. It is a serious concern, (probably the main underlying problem) however, that the final letter from the MoH and run separately so they are never (Appendix Figure 1) expresses the view that allowed to compete for resources—as they the high level of performance of the SDHB’s did in the SDHB.1

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 104 www.nzma.org.nz/journal VIEWPOINT

Appendix

Appendix Table 1: Original OIA requests to Ministry of Health, early September 2019.

2. All other electronic text or messaging communications between MoH employees, MoH employees and any independent contractors or external consultants engaged by the MoH, MoH employees and SDHB employees or SDHB Board Chairperson or Commissioners or independent contractors or external consultants engaged by the SDHB, which relate to or reference any of the following; • SDHB Colonoscopy referrals, numbers of such referrals, or management of such referrals • SDHB Colonoscopy access for symptomatic patients • SDHB Colonoscopy procedure volumes, funded and actual • Readiness, or preparations for readiness, to roll out the National Bowel Cancer Screening Programme at SDHB • Colonoscopy Waiting Time Indicator performance by SDHB in the period July 1 2013 to October 1 2019

3. The best recollections of Minister/s of Health, DGOH, National Bowel Cancer Screening Programme personnel, and MoH employees of any conversations or discussions with other MoH employees, independent contractors or external consultants engaged by the MoH, SDHB employees, SDHB Board Chairperson or Commissioners, or independent contractors or external consultants engaged by the SDHB, whether face-to-face, by telephone or any other means, which relate to or reference any of the following; • SDHB Colonoscopy referrals, numbers of such referrals, or management of such referrals • SDHB Colonoscopy access for symptomatic patients • SDHB Colonoscopy procedure volumes, funded and actual • Readiness, or preparations for readiness, to roll out the National Bowel Cancer Screening Programme at SDHB • Colonoscopy Waiting Time Indicator performance by SDHB in the period July 1 2013 to October 1 2019

Ministry of Health – Reports/Documentation

4. Do un-monitored ‘Opportunity Costs’ discussed in ‘2018 Ministry of Health National Bowel Screening Programme: Benefits Realisation Plan’ include reduced numbers of diagnostic colonoscopies for symptomatic patients as an accepted dis-benefit of the NBCSP?

5. Was any monitoring of referral trends for symptomatic colonoscopy in the preparation for readiness for SDHB BCSP roll out required by the MoH? • If not, why? • If yes, please provide any documentary evidence of such monitoring, whether meeting minutes, reports, briefing notes, emails, electronic text or messaging communications.

Appendix Table 1: Original OIA requests to Ministry of Health, early September 2019 (continued).

6. All reports (routine management, and interim, progress, briefing or final) prepared by MoH employees, or by any independent contractors or external consultants engaged by the MoH, which relate to or reference any of the following; • SDHB Colonoscopy referrals, numbers of such referrals, or management of such referrals • SDHB Colonoscopy access for symptomatic patients • SDHB Colonoscopy procedure volumes, funded and actual • Readiness, or preparations for readiness, to roll out the National Bowel Cancer Screening Programme at SDHB • Colonoscopy Waiting Time Indicator performance by SDHB • SDHB Gastroenterology Service in the period July 1 2013 to October 1 2019

7. The register of risks associated with the National Bowel Cancer Screening Programme roll out project, with planned risk management strategies

8. The Terms of Reference, Project Scope and Project Plan for the National Bowel Cancer Screening Programme roll out

9. Any MoH risk assessments and MoH held Risk Register entries which relate to or reference provision of diagnostic colonoscopy services to symptomatic patients at SDHB or to the roll out the National Bowel Cancer Screening Programme at SDHB, in the period July 1 2013 to October 1 2019

10. All agendas and minutes of any meetings involving MoH employees, or any independent contractors or external consultants engaged by the MoH, which relate to or reference any of the following; • SDHB Colonoscopy referrals, numbers of such referrals, or management of such referrals • SDHB Colonoscopy access for symptomatic patients • SDHB Colonoscopy procedure volumes, funded and actual • Readiness, or preparations for readiness, to roll out the National Bowel Cancer Screening Programme at SDHB • Colonoscopy Waiting Time Indicator performance by SDHB in the period July 1 2013 to October 1 2019

11. All complaints, disclosures (protected or otherwise), incident reports received by MoH , whether from internal or external sources, along with any subsequent investigations and findings, which relate to or reference any of the following; • SDHB Colonoscopy referrals, numbers of such referrals, or management of such referrals • SDHB Colonoscopy access for symptomatic patients • SDHB Colonoscopy procedure volumes, funded and actual • Readiness, or preparations for readiness, to roll out the National Bowel Cancer Screening Programme at SDHB • Colonoscopy Waiting Time Indicator performance by SDHB • SDHB Gastroenterology Service in the period July 1 2013 to October 1 2019

Ministry of Health - Data 12. The numbers of funded colonoscopies undertaken by SDHB, split by the associated procedure codes and purchase units, for each of the financial years July 2013 to June 2019 inclusive.

13. Any data, whether in table, chart or graph form, used by MoH to monitor, review, manage, model or predict SDHB referral volumes for diagnostic colonoscopy of symptomatic patients in the period July 1 2013 to October 1 2019.

14. All requests made to the MoH from any external source for data or information (and any responses provided by MoH) which relate to or reference any of the following; • SDHB Colonoscopy referrals, numbers of such referrals, or management of such referrals • SDHB Colonoscopy access for symptomatic patients • SDHB Colonoscopy procedure volumes, funded and actual • Readiness, or preparations for readiness, to roll out the National Bowel Cancer Screening Programme at SDHB • Colonoscopy Waiting Time Indicator performance by SDHB • SDHB Gastroenterology Service in the period July 1 2013 to October 1 2019

Appendix Table 2: Re-scoped OIA requests to Ministry of Health, late September 2019.

Ministry of Health – Emails and other communications 1. All emails between MoH employees, MoH employees and any independent contractors or external consultants engaged by the MoH, MoH employees and SDHB employees or SDHB Board Chairperson or Commissioners or independent contractors or external consultants engaged by the SDHB, which relate to or reference the following; • Colonoscopy access for symptomatic patients within the SDHB – particularly (but not exclusively) awareness of any risks (potential or actual) to ongoing provision of clinically appropriate levels of diagnostic colonoscopy to symptomatic patients; plans or advice to mitigate any such risks; in the period July 1 2016 to current date. NB: Lower priority request 2. All emails between MoH employees, MoH employees and any independent contractors or external consultants engaged by the MoH, MoH employees and SDHB employees or SDHB Board Chairperson or Commissioners or independent contractors or external consultants engaged by the SDHB, which relate to or reference the following; • Readiness, or preparations for readiness, to roll out the National Bowel Cancer Screening Programme at SDHB – particularly (but not exclusively) the indicators used by the MoH to determine progress of the SDHB to offer the NBCSP; assessments and monitoring of SDHB capacity and capability relating to provision of the NBCSP at SDHB in the planning, roll out or post roll out periods; awareness of any risks (potential or actual) to ongoing provision of clinically appropriate levels of diagnostic colonoscopy to symptomatic patients arising from provision of the NBCSP at SDHB; plans or advice to mitigate any such risks in the period July 1 2016 to current date. NB: Lower priority request 3. All emails between MoH employees, MoH employees and any independent contractors or external consultants engaged by the MoH, MoH employees and SDHB employees or SDHB Board Chairperson or Commissioners or independent contractors or external consultants engaged by the SDHB, which relate to or reference the following; • Colonoscopy Waiting Time Indicator performance by SDHB – particularly (but not exclusively) any guidance provided by the MoH to SDHB in the implementation of the CWTI; Any awareness of risks (potential or actual) to ongoing provision of clinically appropriate levels of diagnostic colonoscopy to symptomatic patients arising from implementation of the CWTI; plans or advice to mitigate any such risks in the period July 1 2016 to current date. NB: Lower priority request 4. All other electronic text or messaging communications between MoH employees, MoH employees and any independent contractors or external consultants engaged by the MoH, MoH employees and SDHB employees or SDHB Board Chairperson or Commissioners or independent contractors or external consultants engaged by the SDHB, which relate to or reference the following; • Colonoscopy access for symptomatic patients within the SDHB – particularly (but not exclusively) awareness of any risks (potential or actual) to ongoing provision of clinically appropriate levels of diagnostic colonoscopy to symptomatic patients; plans or advice to mitigate any such risks in the period July 1 2016 to current date. NB: Lower priority request

Appendix Table 2: Re-scoped OIA requests to Ministry of Health, late September 2019 (continued).

5. All other electronic text or messaging communications between MoH employees, MoH employees and any independent contractors or external consultants engaged by the MoH, MoH employees and SDHB employees or SDHB Board Chairperson or Commissioners or independent contractors or external consultants engaged by the SDHB, which relate to or reference the following; • Readiness, or preparations for readiness, to roll out the National Bowel Cancer Screening Programme at SDHB – particularly (but not exclusively) the indicators used by the MoH to determine progress of the SDHB to offer the NBCSP; assessments and monitoring of SDHB capacity and capability relating to provision of the NBCSP at SDHB in the planning, roll out or post roll out periods; awareness of any risks (potential or actual) to ongoing provision of clinically appropriate levels of diagnostic colonoscopy to symptomatic patients arising from provision of the NBCSP at SDHB; plans or advice to mitigate any such risks in the period July 1 2016 to current date. NB: Lower priority request 6. All other electronic text or messaging communications between MoH employees, MoH employees and any independent contractors or external consultants engaged by the MoH, MoH employees and SDHB employees or SDHB Board Chairperson or Commissioners or independent contractors or external consultants engaged by the SDHB, which relate to or reference the following; • Colonoscopy Waiting Time Indicator performance by SDHB – particularly (but not exclusively) any guidance provided by the MoH to SDHB in the implementation of the CWTI; Any awareness of risks (potential or actual) to ongoing provision of clinically appropriate levels of diagnostic colonoscopy to symptomatic patients arising from implementation of the CWTI; plans or advice to mitigate any such risks in the period July 1 2016 to current date. NB: Lower priority request 7. The best recollections of Minister/s of Health, DGOH, National Bowel Cancer Screening Programme personnel, and MoH employees of any conversations or discussions with other MoH employees, independent contractors or external consultants engaged by the MoH, SDHB employees, SDHB Board Chairperson or Commissioners, or independent contractors or external consultants engaged by the SDHB, whether face-to-face, by telephone or any other means, which relate to or reference the following; • Colonoscopy access for symptomatic patients within the SDHB – particularly (but not exclusively) awareness of any risks (potential or actual) to ongoing provision of clinically appropriate levels of diagnostic colonoscopy to symptomatic patients; plans or advice to mitigate any such risks; in the period July 1 2016 to current date NB: Lower priority request 8. The best recollections of Minister/s of Health, DGOH, National Bowel Cancer Screening Programme personnel, and MoH employees of any conversations or discussions with other MoH employees, independent contractors or external consultants engaged by the MoH, SDHB employees, SDHB Board Chairperson or Commissioners, or independent contractors or external consultants engaged by the SDHB, whether face-to-face, by telephone or any other means, which relate to or reference the following; • Readiness, or preparations for readiness, to roll out the National Bowel Cancer Screening Programme at SDHB – particularly (but not exclusively) the indicators used by the MoH to determine progress of the SDHB to offer the NBCSP; assessments and monitoring of SDHB capacity and capability relating to provision of the NBCSP at SDHB in the planning, roll out or post roll out periods; awareness of any risks (potential or actual) to ongoing provision of clinically appropriate levels of diagnostic colonoscopy to symptomatic patients arising from provision of the NBCSP at SDHB; plans or advice to mitigate any such risks in the period July 1 2016 to current date. NB: Lower priority request

Appendix Table 2: Re-scoped OIA requests to Ministry of Health, late September 2019 (continued).

9. The best recollections of Minister/s of Health, DGOH, National Bowel Cancer Screening Programme personnel, and MoH employees of any conversations or discussions with other MoH employees, independent contractors or external consultants engaged by the MoH, SDHB employees, SDHB Board Chairperson or Commissioners, or independent contractors or external consultants engaged by the SDHB, whether face-to-face, by telephone or any other means, which relate to or reference the following; • Colonoscopy Waiting Time Indicator performance by SDHB – particularly (but not exclusively) any guidance provided by the MoH to SDHB in the implementation of the CWTI; Any awareness of risks (potential or actual) to ongoing provision of clinically appropriate levels of diagnostic colonoscopy to symptomatic patients arising from implementation of the CWTI; plans or advice to mitigate any such risks in the period July 1 2016 to current date. NB: Lower priority request Ministry of Health – Reports/Documentation

10.. Do un-monitored ‘Opportunity Costs’ discussed in ‘2018 Ministry of Health National Bowel Screening Programme: Benefits Realisation Plan’ include reduced numbers of diagnostic colonoscopies for symptomatic patients as an accepted dis-benefit of the NBCSP? If Yes, please provide rationale and any documents used by the MoH to support this position. NB: Higher priority request 11.. Was any monitoring of referral trends for symptomatic colonoscopy in the preparation for readiness for SDHB BCSP roll out required by the MoH? • If not, please explain why not, include any documents relied upon to preclude such monitoring • If yes, please provide any documentary evidence of such monitoring, whether meeting minutes, reports, briefing notes, emails, electronic text or messaging communications. NB: Higher priority request 12.. All reports (routine management, and interim, progress, briefing or final) prepared by MoH employees, or by any independent contractors or external consultants engaged by the MoH, which relate to or reference the following; • Colonoscopy access for symptomatic patients within the SDHB – particularly (but not exclusively) awareness of any risks (potential or actual) to ongoing provision of clinically appropriate levels of diagnostic colonoscopy to symptomatic patients; plans or advice to mitigate any such risks in the period July 1 2016 to current date. NB: Lower priority request 13. All reports (routine management, and interim, progress, briefing or final) prepared by MoH employees, or by any independent contractors or external consultants engaged by the MoH, which relate to or reference the following; • Readiness, or preparations for readiness, to roll out the National Bowel Cancer Screening Programme at SDHB – particularly (but not exclusively) the indicators used by the MoH to determine progress of the SDHB to offer the NBCSP; assessments and monitoring of SDHB capacity and capability relating to provision of the NBCSP at SDHB in the planning, roll out or post roll out periods; awareness of any risks (potential or actual) to ongoing provision of clinically appropriate levels of diagnostic colonoscopy to symptomatic patients arising from provision of the NBCSP at SDHB; plans or advice to mitigate any such risks in the period July 1 2016 to current date. NB: Lower priority request

Appendix Table 2: Re-scoped OIA requests to Ministry of Health, late September 2019 (continued).

14. All reports (routine management, and interim, progress, briefing or final) prepared by MoH employees, or by any independent contractors or external consultants engaged by the MoH, which relate to or reference the following; • Colonoscopy Waiting Time Indicator performance by SDHB – particularly (but not exclusively) any guidance provided by the MoH to SDHB in the implementation of the CWTI; Any awareness of risks (potential or actual) to ongoing provision of clinically appropriate levels of diagnostic colonoscopy to symptomatic patients arising from implementation of the CWTI; plans or advice to mitigate any such risks in the period July 1 2016 to current date. NB: Lower priority request 15. Any MoH risk assessments and MoH held Risk Register entries which relate to or reference provision of diagnostic colonoscopy services to symptomatic patients at SDHB in the period July 1 2013 to October 1 2019

16. All agendas and minutes of any meetings (including MIF meetings) involving MoH employees, or any independent contractors or external consultants engaged by the MoH, which relate to or reference the following; • Colonoscopy access for symptomatic patients within the SDHB – particularly (but not exclusively) awareness of any risks (potential or actual) to ongoing provision of clinically appropriate levels of diagnostic colonoscopy to symptomatic patients; plans or advice to mitigate any such risks; awareness of any potential impacts on such access arising from any MoH initiatives or plans; the plans or advice to manage any such impacts in the period July 1 2016 to current date. NB: Higher priority request 17. All agendas and minutes of any meetings (including MIF meetings) involving MoH employees, or any independent contractors or external consultants engaged by the MoH, which relate to or reference the following; • Readiness, or preparations for readiness, to roll out the National Bowel Cancer Screening Programme at SDHB – particularly (but not exclusively) the indicators used by the MoH to determine progress of the SDHB to offer the NBCSP; assessments and monitoring of SDHB capacity and capability relating to provision of the NBCSP at SDHB in the planning, roll out or post roll out periods; awareness of any risks (potential or actual) to ongoing provision of clinically appropriate levels of diagnostic colonoscopy to symptomatic patients arising from provision of the NBCSP at SDHB; plans or advice to mitigate any such risks in the period July 1 2016 to current date. NB: Higher priority request 18. All agendas and minutes of any meetings (including MIF meetings) involving MoH employees, or any independent contractors or external consultants engaged by the MoH, which relate to or reference the following; • Colonoscopy Waiting Time Indicator performance by SDHB – particularly (but not exclusively) any guidance provided by the MoH to SDHB in the implementation of the CWTI; Any awareness of risks (potential or actual) to ongoing provision of clinically appropriate levels of diagnostic colonoscopy to symptomatic patients arising from implementation of the CWTI; plans or advice to mitigate any such risks in the period July 1 2016 to current date. NB: Higher priority request Ministry of Health - Data 19. Any data, whether in table, chart or graph form, used by MoH to monitor, review, manage, model or predict SDHB referral or intervention volumes for diagnostic colonoscopy of symptomatic patients in in the period July 1 2016 to current date. NB: Higher priority request 20. Any research papers, briefing papers, documents, evidence considered, risk assessments, advice used by the MoH in the development and introduction of the Colonoscopy Waiting Times Indicator NB: Higher priority request

Appendix Table 3: P Bagshaw OIA requests—progress timeline, September 2019 to August 2020.

12.09.19 OIA request sent to Ministry of Health by email

13.09.19 Email from MoH – “The request as it is currently worded is for a very large volume of information and may be refused under section 18(f) of the Official Information Act 1982, because the information requested cannot be made available without substantial collation or research.” MoH suggested refining scope of request.

17.09.19 Refined and re-scoped OIA requests sent to MoH (Some questions removed, coverage period reduced, requests prioritised to aid response stratification by MoH)

26.09.19 Email from MoH declining OIA request “because the information requested cannot be made available without substantial collation or research.” There was little point in further refining or reducing the scope again. The request was substantial because that was what was required to answer critical questions.

17.12.2019 Email to Office of The Ombudsman. Requested opinion of the Ombudsman as to the application of section 18(f) of the Official Information Act in view of the significant public interest considerations inherent in these matters

19.02.2020 Email to Office of The Ombudsman, enquiring as to status of the complaint re declined OIA requests to Ministry of Health.

19.02.2020 Response from Office of The Ombudsman, advising that complaint had not yet been allocated to an investigator

26.02.2020 Email from Senior Investigator at the Office of The Ombudsman, advising that he will be assisting the Ombudsman with the complaint and requesting all correspondence relating to the requests

27.02.2020 Email providing all requested correspondence sent to Senior Investigator at Office of The Ombudsman

30.03.2020 Email from MoH representative advising that she will be reviewing the 19 requests to the Ministry, following the lodge- ment of complaint with the Office of the Ombudsman.

30.03.2020 Email response to MoH representative advising her that requestor is happy for work to be suspended whilst resources are focussed on the response to Covid-19 in New Zealand.

30.04.2020 Email received from MoH representative advising that her work on the MoH response to the OIAs directed to the MoH is now largely complete.

10.06.2020 Email to MoH representative requesting that, as New Zealand is back at Level 1 Covid-19 response, I wish all the required information to be provided as soon as possible.

16.06.2020 Email from MoH representative, advising that electronic files will be created for sending to me, aiming for “no later than mid-July”.

18.07.2020 Email to MoH representative, asking for confirmation that information relating to the very longstanding OIA request will be sent to me by the end of the coming week (by Friday 24th July 2020)

23.07.2020 Email to Senior Investigator at the Office of The Ombudsman, requesting his intervention to ensure provision of information.

23.07.2020 Email from Senior Investigator at the Office of The Ombudsman, noting that no investigation or opinion by Ombuds- man had been necessary as the Ministry of Health had, upon receipt of notification of the complaint having been made, decided that it would provide the information requested after all. He offered to make inquiries with the MoH in order to expedite provision of the information. Offer duly accepted by return email.

3.08.2020 Email to Senior Investigator at the Office of The Ombudsman, asking what progress made.

3.08.2020 Email received from Senior Investigator at the Office of The Ombudsman, advising MoH have new deadline of 21st August to provide the information concerned.

13.08.2020 3516 documents (in hard copy only) received from Ministry of Health, relating to the OIA requests made to the Ministry of Health.

Appendix Figure 1: Letter from Ministry of Health, Population Health and Prevention, to P Bagshaw 30 October 2020.

Competing interests: Nil. Author information: Philip Bagshaw: Chair, Canterbury Charity Hospital Trust. Paula Goodman: Independent. Brian Cox: Hugh Adam Cancer Epidemiology Unit, Department of Preventive and Social Medicine, Dunedin School of Medicine. Corresponding author: Philip Bagshaw, Chair, Canterbury Charity Hospital Trust [email protected] URL: www.nzma.org.nz/journal-articles/official-information-act-investigation-of-the-ministry-of- healths-process-to-assess-the-southern-district-health-boards-readiness-to-join-the-national-bowel-screening-programme-in-2018

REFERENCES 1. Bagshaw P, Cox B. Adequa- health.nz/publications/ at the New Zealand cy of publicly funded statement-dave-cull-chair- experience with the colonoscopy services in sdhb-colonoscopy-services Official Information Act New Zealand. NZ Med J 5. National Bowel Screening after 25 years. Address to 2020;133(1526):7-11. Programme Business Case International Conference 2. Agenda Southern District 2017/2018. https://www. of Information Commis- th Health Board Meeting 3rd health.govt.nz/system/files/ sioners, Wellington 27 November 2020. https:// documents/pages/nbsp_ November 2007. https:// www.southernhealth.nz/ business_case_2017_18_v2_ www.lawcom.govt.nz/sites/ sites/default/files/2020- redacted-optimised.pdf default/files/audioFiles/ Palmer%20Speech%20 10/2020-11-03%20 6. Official Information on%20Official%20 SDHB%20Board%20 Act 1982. https://www. Information%20Act.pdf Agenda_Public.pdf legislation.govt.nz/ 3. Bissett I, Broome K. Colo- act/public/1982/0156/ 10. Ecclestone A. An updated noscopy Patient Review latest/whole.html Official Information Act must strengthen our Report for Southern District 7. University of Otago Review. right to know. 14th July Health Board September Department of Surgical 2020. https://www.stuff. 2020 https://www.south- Sciences. University of co.nz/national/politics/ ernhealth.nz/sites/default/ Otago; June, 2017. files/2020-10/2020-10- opinion/122114366/ 8. Amnesty International. 06%20SDHB%20Board%20 an-updated-official-infor- Joint call for overhaul of Agenda_public.pdf mation-act-must-strength- Official Information Act. en-our-right-to-know 4. Statement from Dave 9th October 2020. https:// 11. Broughton C. Screening Cull, Chair, SDHB - Colo- www.amnesty.org.nz/ despite demand failings. noscopy Services. Issue: joint-calls-overhaul-oia Tue 6th October 2020. The Press, Christchurch 9. Palmer G. A hard look rd https://www.southern- 23 January 2021.

Why is it so hard for general practice to provide COVID-19 vaccines? Vanessa Weenink

he COVID-19 vaccination campaign is crucial ‘last-mile logistics’ 7 of the vaccination the best chance New Zealand has to programme may be left to individual DHBs to protect our citizens’ health and will coordinate vaccination for the health work- T 2,3 be a crucial activity for the health sector force and the wider public. 1 in 2021. The procurement strategy for International evidence highlights both New Zealand started early and has been general practice’s (GP) and community 1,2 a dynamic process. However, there has pharmacy’s essential role in successful been no evidence of large-scale and detailed campaigns.8–13 In the UK, 75% of the vaccina- operational planning at the level of district tions delivered have been through primary health boards (DHBs), drawing political care networks (including GP clinics) and 3 comments that plans are ‘half-baked’ and only the remaining 25% in specially desig- prompting an Office of the Auditor-General nated vaccination centres.14 Israel has fully review. To have a successful vaccination vaccinated more than 50% of its resi- program, New Zealand will need to learn dents with the Pfizer vaccine in under six from international experience, and it must months,12 and this was coordinated through include the participation of primary care. the primary care health plans, where GPs New Zealand has been praised interna- commonly work in private clinics and tionally for its response to the pandemic, are usually salaried by the health plans.12 and New Zealand has escaped other New Zealand has limited primary care nations’ death rates.4 One of the factors in involvement in the initial phases as the our success has been strong leadership.5 vaccine is targeted toward the most likely to Communication has been effective, and we be infected at the border. 3 have had an excellent response from our New Zealand should consider the Israeli 1 team of five million. Regardless of that campaign as a benchmark of success. The result, the medium- to long-term success Israelis have had a few advantages that New of New Zealand’s response overall will be Zealand could not replicate. For example, determined by the success of the vaccine high population density. 12 Fortunately, campaign. Communication with primary New Zealand and Israel share a universal care will need to improve in New Zealand if approach to health, well organised primary we are to continue our excellent response. care, a national unique identifier system for The plans for the vaccination rollout seem patients, substantial investment in vaccine to be evolving fast in New Zealand. A recent procurement and a centralised (rather than letter in the New Zealand Medical Journal federalised) government. If the vaccine expressed that border workers be vaccinated campaign is centred around the community as a priority group, which the Ministry of through primary care providers, New Health (MOH) had not previously planned.6 Zealand may meet the projected targets that Yet, by the time the letter was published, the government has set.2 the vaccine campaign for border workers The National Health Service (NHS) in the 2 was underway. Detailed plans following the United Kingdom shares features with the border force are not transparent or publicly New Zealand health system.15 The NHS has available. The prioritisation of vaccination primary care networks (PCNs), analogous to is a small part of the planning required. our primary health organisations (PHOs).16 While that has now been established, the The NHS is also a universal system and is

providing vaccination free to the population. The New Zealand health system is not opti- In the NHS, mainly nurses (who are already mally structured for rapid deployment of qualified vaccinators) are doing the immu- a whole-population vaccination campaign. nisation.14 The NHS attempted to recruit a The semi-autonomous DHBs are centred sizeable ancillary vaccination workforce, around local hospitals, with variable access which has been less successful. Without to services across the country giving rise additional workers, GP clinics are deferring to ‘post-code lotteries’ regarding patient other work to concentrate on the vaccine.17 services and outcomes.16 Systemic racism The NHS also considers incentive payments and inequity exist in every facet of the to PCNs to target priority and hard to reach health and social support systems.22 The populations. 18 Meanwhile, the New Zealand New Zealand health system was under MOH is yet to indicate the payment structure review before the pandemic, with the final for vaccine delivery, leaving primary care report released mid-2020.23 The government providers uncertain of their required planned actions for the transition are not yet level of involvement. New Zealand GPs public, which is unsettling and will require are not protected by collective bargaining MOH and DHB attention and may dismantle as they are under the NHS and the Israeli the PHOs that would be important in coordi- health system, and funding agreements are nating vaccine programmes, like the PCNs in tortuous and often contested.22 the UK. Primary care will need to be well Despite the challenges in logistics, coor- supported to participate in the New Zealand dination and personnel outlined above, vaccine programme. The logistical consid- New Zealand can learn from international erations for transport to vaccination sites experience. Implementing those findings are daunting.19 However, in practice in in our system’s milieu will require dedi- the UK and the USA and Australia, one of cation and vaccination plans to be ‘fully the most significant challenges has been baked’ and transparent to everyone. Logis- coordinating patients to attend booked tical limitations are well explained19 and appointments.10,12,20 The coordination is anticipated. The vaccination programmes’ made difficult when there are supply administrative requirements in the UK and uncertainties, and the short shelf-life once Australia have been frustrating to GPs.10,11,21 defrosted adds to the complexity of the In contrast, Israel has efficiently mobilised Pfizer vaccine.10 GPs in Australia have the community health workforce through struggled with administration and logistical coordinated general practice and has problems.21 The logistical requirements are stream-lined all the steps.12 New Zealand such that a coordinated approach through should learn from those international expe- PHOs may be more efficient than individual riences and provide an efficient vaccination practices undertaking the work. campaign for our citizens.

Competing interests: Dr Vanessa Weenink is Chair of the General Practitioners Council of the NZMA, Deputy Chair (Elect) of the NZMA and on the Board of Directors of Pegasus Health Charitable limited (the Primary Health Organisation). Author information: Dr Vanessa Weenink Corresponding author: Dr Vanessa Weenink [email protected] URL: www.nzma.org.nz/journal-articles/why-is-it-so-hard-for-general-practice-to-provide-covid- 19-vaccines-open-access

REFERENCES 1. Bloomfield A. COVID-19, Scheduling of COVID-19 How the UK vaccine 20, 21: lessons from vaccination for at-risk rollout delivered success, New Zealand’s 2020 employees. NZMJ. 2021;134 so far. BMJ. 2021:n421. response for 2021 and 7. Lee TH, Chen AH. Last-Mile 15. Gauld R, Horsburgh S. beyond. NZMJ. 134 Logistics of Covid Vaccina- What motivates doctors to 2. Ministry of Health web tion — The Role of Health leave the UK NHS for a “life page [updated 10 March Care Organizations. New in the sun” in New Zealand; 2021. Available from: England Journal of Medi- and, once there, why don’t https://www.health. cine. 2021;384(8):685-7. they stay? Human Resourc- govt.nz/our-work/ 8. Freed GL. Actionable es for Health. 2015;13(1). diseases-and-conditions/ lessons for the US COVID 16. Gauld R. New Zealand’s covid-19-novel-corona- vaccine program. Isr post-2008 health system virus/covid-19-vaccines/ J Health Policy Res. reforms: Toward covid-19-vaccine-strate- 2021;10(1):14. re-centralisation of gy-and-planning#strategy. 9. Harnden A, Lim WS, organisational arrange- 3. Mansh T. Government Earnshaw A. COVID-19 ments. Health Policy. releases Covid-19 vaccine vaccination programme: 2012;106(2):110-3. delivery schedule: a central role for primary 17. Iacobucci G. Covid-19: elderly and people with care. Br J Gen Pract. GPs can limit routine pre-existing conditions 2021;71(703):52-3. work to focus on to be prioritised. Stuff. 10. Mahase E. Covid-19: vaccination, says NHS 2021 March 10th 2021. Logistical problems England. BMJ. 2021:n67. 4. Huang QS, Wood T, Jelley frustrate GPs ready to 18. Iacobucci G. Covid-19: L, Jennings T, Jefferies deliver vaccine in England. GPs could get extra S, Daniells K, et al. BMJ. 2020;371:m4849. funding to boost vaccine Impact of the COVID-19 11. Majeed A, Molokhia M. uptake in hard-to-reach nonpharmaceutical Vaccinating the UK against groups. BMJ. 2021:n548. interventions on influenza covid-19. BMJ. 2020:m4654. 19. Holm MR, Poland GA. Crit- and other respiratory ical aspects of packaging, viral infections in New 12. Rosen B, Waitzberg R, storage, preparation, and Zealand. Nature Commu- Israeli A. Israel’s rapid administration of mRNA nications. 2021;12(1). rollout of vaccinations for COVID-19. Isr J Health and adenovirus-vectored 5. Dada S, Ashworth HC, Policy Res. 2021;10(1):6. COVID-19 vaccines for Bewa MJ, Dhatt R. W ords optimal efficacy. Vaccine. 13. Volpp KG, Loewenstein matter: political and gender 2021;39(3):457-9. analysis of speeches made G, Buttenheim AM. 20. Mills MC, Salisbury by heads of Government Behaviorally Informed D. The challenges of during the COVID-19 Strategies for a National distributing COVID-19 Pandemic. BMJ Global COVID-19 Vaccine Promo- vaccinations. EClinicalMed- Health. 2021;6(1):e003910. tion Program. JAMA. 2021;325(2):125-6. icine. 2021;31:100674. 6. Walls C, Gavaghan S, 21. Mannix L. Victoria’s Gorman D, McBride D. 14. Baraniuk C. Covid-19: vaccines ready to roll -

but some teething issues 4DC0-9DD3-A961CC5360D- 22. Moewaka Barnes H, remain 2021 [22/3/21]. B&promote_channel=ed- McCreanor T. Colo- Available from: https:// mail&utm_campaign=am- nisation, hauora and www.theage.com.au/ theage&utm_medi- whenua in Aotearoa. national/victoria/victoria- um=email&utm_ Journal of the Royal s-vaccines-ready-to-roll- source=newsletter&utm_ Society of New Zealand. but-some-teething-issues- term=2021-03-22&mb- 2019;49(sup1):19-33. remain-20210321-p57cpf. nr=MjA0MjIwNzg&in- 23. Simpson H. Health and html?utm_content=NAR- stance=2021- Disability System Review RATIVE&list_ 03-22-06-38-AEDT&- Final Report. 2020. name=3C1D28D0-7E1E- jobid=29297386.

Intravenous iron infusion and newer non-dextran formulations Tim Aung, Justin Coleman, Peter W Davidson, David J Hetzel, Sandy T Aung

ABSTRACT There are several newer intravenous iron formulations to treat iron deficiency and its anaemia. Its use in the primary care setting has been infrequent compared to tertiary centres, due to historical concerns such as anaphylaxis. There is a lack of overall comparison among the intravenous formulations of iron. Compared to oral iron therapy, the newer intravenous formulations, which allow a complete or near-complete replacement in a single sitting of 15–30 minutes, have an improved safety profile with better tolerability, efficacy and effectiveness. They are suited for administration in the primary care setting. The four non-dextran formulations (ferric carboxymaltose, iron sucrose, iron isomaltoside and ferumoxytol) share an equal or near equal efficacy and safety profile. This article also outlines how to provide iron infusion safely and effectively in the community.

ron is an essential element and nu- setting. This article outlines how to provide trient for the human body functions. iron infusion safely and effectively, along with IIron deficiency (ID) and iron deficiency a brief comparison of the newer formulations anaemia (IDA) are common health problems (ferric carboxymaltose (FCM), iron sucrose worldwide. The World Health Organization (ISC), iron isomaltoside (IIM) and ferumoxytol (2011) estimated 34% of the global popula- (FOT)). tion (>2 billion) is affected by anaemia. The Although in most guidelines oral iron most common type of anaemia was ID (50% remains as a the first-line treatment for of total anaemia), which primarily affect- replacement of iron, its common side ed women of reproductive age and young effects (gastric upset and constipation and 1 children. the need to take it regularly for months Although the practice of intravenous (IV) to replenish iron stores) often result in iron to treat iron deficiency and its anaemia non-adherence. Because intramuscular iron has existed for more than six decades,2 its injection can cause pain and skin staining use in primary care centres has, compared to and require multiple injections with ques- tertiary centres, not been widespread, largely tionable absorption, it is therefore no longer due to historical concern of anaphylaxis and favourable. lack of remuneration. Administration of newer (non-dextran) iron formulations allows Indications and a complete or near-complete replacement in a single sitting of 15–30 minutes. These newer contraindications formulations have an improved safety profile Symptoms specific to ID are relatively and better tolerability, efficacy and effec- uncommon but can include pagophagia (ice tiveness compared with oral iron therapy. craving) or other forms of pica and rest- They are suited for administration in primary less-legs syndrome. Symptoms related to care or community practices in a proper IDA are non-specific: they include fatigue,

weakness, tiredness, dizziness, irritability Infusion offers an alternative route of and pale skin, and in severe cases chest administration for those in whom oral pain, palpitations or shortness of breath. iron is unsuitable due to intolerance, poor Hair and nail disorders such as koilonychia adherence, impracticality or contraindica- (spoon nails) can also occur in chronic IDA. tions. Indications for iron infusion include ID More importantly, the untreated ID and and IDA caused by the underlying conditions IDA can eventually impact on cognition, described in Figure 1.4-6 Figure 2 lists contra- academic achievement, exercise tolerance, indication for infusion.4-6 Precautions include work productivity and quality of life.3 ID individuals with acute infection, asthma, and IDA should be confirmed by blood count marked atopy, liver dysfunction or conditions and iron studies with correct interpretation. associated with low phosphate in the body. Although the details of investigation for There are insufficient data to support the IDA are beyond the scope of this article, it is safety of iron infusion in the first trimester always imperative to address the underlying of pregnancy and in children under 14 years cause while correcting the deficiency. of age. Nevertheless, trial infusions of ferric carboxymaltose, iron sucrose and ferumoxytol in children (<14 years) at tertiary Figure 1: Indications for iron infusion. centres have shown some promises and is going to have further evaluation. 9–15 • Malabsorption (coeliac disease, inflammatory bowel disease, weight-loss surgery, bowel Benefits resection) Intravenous iron rapidly restores iron and • Inadequate diet (vegan, vegetarian) expedites haemoglobin synthesis. Newer • Excess blood loss (chronic occult bleeding, formulations (Table 1) contain carbohydrate excess menstrual blood loss, childbirth) cores that more tightly bind elemental iron, • Increased body demand (pregnancy of allowing for a much slower release and beyond first trimester, rapid growth) providing marginal or fewer reactions.1,16 • Anaemia of chronic disease/inflammation They have also demonstrated greater (chronic kidney disease, cancers, rheuma- efficacy and effectiveness and, compared to oral therapy and previous parenteral toid arthritis, heart failure), in which func- formulations, can improve quality of life tional iron deficiency (FID)† can be present and productivity. 6,17,18 Newer formulations • Intolerance, poor adherence or unrespon- also have the potential save health costs by siveness to oral iron reducing the frequency of visits to hospital and healthcare providers.18–21 Some studies †FID denotes a state in which there is insufficient iron have indicated that the newer genera- mobilisation for erythropoiesis despite normal or high ferritin concentrations with a low transferrin saturation tions of intravenous iron are underutilised reflecting inadequate iron availability. Maintaining due to historical fears about anaphy- the ferritin level greater than 100–200 μg/L in some laxis that were far more common with chronic conditions, such as heart failure, is important high molecular weight iron dextrans (eg, to improve ventricular function, quality of life and to reduce hospitalisations.7,8 imferon, dexferrum), which are no longer available.16,19 Two narrative reviews and one meta-analysis have suggested a reconsid- Figure 2: Contraindications for iron infusion. eration of the current paradigm whereby oral iron treatment is considered a first-line • Anaemia not caused by iron deficiency therapy;16,19,22 it needs further evaluation • Known anaphylaxis to the specific iron with a broad consensus. There are greater product than 20 randomised studies in which IV administration of iron offers better tolera- • Iron overload conditions (haemochromato- bility, efficacy and effectiveness compared sis, hemosiderosis, thalassemia major) to oral iron.23–28 One randomised control • High-risk patients with serious comorbidi- trial in Australia described the cost of IV use ties (moderate to severe failure of the heart, as no more than the cost for oral therapy, liver or respiratory system); such individu- in addition to having superior tolerance, als could be managed at tertiary facilities efficacy and effectiveness.29

therapy could be considered.33 It Adverse effects may become a serious issue in young • Anaphylaxis is rare but can be women. See Figure 3 on how to avoid life threatening if not managed or minimise such an incidence and properly.21 It involves bronchospasm others. with dyspnoea, angioedema, tachy- • Delayed symptoms may occur one- cardia and hypotension. Rates vary or two-days post infusion: chills and according to iron formulation: FCM fever, headache, arthralgia, myalgia, and ISC (1–10/10,000 cases (≈0.1%– urticaria/rash, angioneurotic oedema. 0.01%), IPC (10–100/10,000 cases).6,30 Two review studies, one of which • Transient hypophosphatemia may 30,34 was a meta-analysis, reported that a also occur, particularly with FCM. properly defined serious anaphylaxis are even less than the above rate, if Dosage dextran is excluded and drug therapy The Ganzoni formula (created by Dr (adrenaline, antihistamine) is given Ganzoni of Zurich) usefully estimates a for minor reactions in the context of replacement dosage of parenteral iron practitioners’ fears for further deterio- required (Table 2).35 However, a simplified ration, which is not uncommon.31,32 method of calculation based on current Less severe or minor reactions include: haemoglobin level and body weight can • Facial flushing, urticaria, arthralgia, also be used if preferred (Table 2). Each myalgia, sensation of stiffness in face iron product has specific instructions for or limbs. dilution and duration of infusion, so refer to the specific instruction or local guide- • Dizziness, headache, nausea, lines for each product whenever possible. dysgeusia. Generally, FCM (1,000mg) and IIM (1,000– • Injection site reactions (pain, discol- 1,500mg) can be given as a single dose over oration of skin). Skin staining 15–30 minutes.36 As per manufacturer, FOT usually fades over time, but may is to be given 510mg at a time; two trials of be permanent, in which case laser 1,020mg infusion over 15–30mins reported

Table 1: History of parenteral iron products.1,18,19

Year Iron products Before 1954 Ferric hydroxide and iron saccharide (severe reactions and toxicity).

1954 Iron dextran (IDX): Imferon (high molecular weight) withdrawn in 1991.

1992–1997 Iron dextran (IDX): INFeD (low molecular weight), Dexferrum (high molecular weight, withdrawn in 2009), Infufer (low molecular weight). Unavailable in Australasia.

1990s Iron polymaltose complex (IPC): Ferrosig, Ferrum H. Note: the oral formulation of IPC has been available since 1978.

1999 Ferric gluconate (FGC): Ferrlecit. Newer formulations

2000 Iron sucrose (ISC):† Venofer.

2001 Iron dextran (IDX): Cosmofer (low molecular weight). Unavailable in Australasia.

2009–2010 Ferumoxytol (FOT): Feraheme. Iron isomaltoside (IIM):† Monofer. Ferric carboxymaltose (FCM):† Ferinject.

† = Approved by Medsafe and subsided by PHARMAC in New Zealand.

no safety concerns, but this needs further patients with any cause, such as diet, child- appraisal.37.38 ISC requires multiple fraction bearing and menorrhagia, gastrointestinal doses with 100–200mg maximum at a time, disorders, renal failure and cancer. The which is commonly used in renal dialysis overall analysis revealed that those four patients. formulations share a comparable safety profile and efficacy overall. However, Comparison products that can deliver a replacement with a single large dose rapidly increase A literature search of comparison blood parameters and offers a more conve- studies for four non-dextran formula- nient dosing regimen. Hypophosphatemia tions (FCM, ISC, IIM and FOT) yielded can be encountered infrequently with FCM a total of 22 head-to-head comparisons and its long-term clinical significance is (n=10,269) and three multi-comparisons unknown due to lack of study. Appendix (more than two products).39–63 The compar- Table 1 illustrates details of comparison isons included various types of studies, with total numbers, methods and subjects such as randomised and clinical trials, of studies. cohorts and reviews. The subjects were IDA

Figure 3: Preparation and administration of intravenous iron.

• The iron ampoules/vials should be stored below 25–30°Celcius and protected from light. Do not refrigerate nor freeze. • Obtain informed consent after discussing the pros and cons of the procedure. • The ideal solution for dilution is 0.9% isotonic saline, as not all formulations are compatible with dextrose. Dilution amount for drip infusion is 250ml–500ml, and 10ml–20ml for IV push/bolus. • Use aseptic technique (as for venepuncture). • Keep a resuscitation kit (including adrenaline) nearby in case of anaphylaxis. • Adverse reactions can be minimised with a slower infusion rate, particularly in the initial stages. Consider taking 30 minutes, instead of 15 minutes, and using a drip infusion instead of an IV bolus. • The injector should be seated in a comfortable position if elected for slow IV bolus. • Take appropriate observations during the infusion. • To avoid or minimise skin staining, you must prevent extravasation by ensuring the IV canula is inside the vein at the insertion site by initially running saline fluid and then, at the end of the infusion, flushing with 20–50mL of saline fluid and applying an instant compression pressure with cover. Do not rub or massage over the IV/infusion site. • Do not attempt for IV iron administration without being trained for IV canula insertion and proper setup.

Table 2: Ganzoni formula and simplified method of calculation for iron dosage.

Ganzoni formula Total iron deficit/requirement (mg) = {body weight (kg) x (target Hb – actual Hb in g/L)} x 0.24 + iron depot (500mg).

Example of calculation for a patient with an ideal body weight of 60 kg and Hb = 80g/L, and the target Hb is set to be 150 g/L. The total required iron dose = {60 x (150 - 80)} x 0.24 + 500 mg = 1508 mg (1500 mg). Simplified method of calculation for iron dosage Hb g/L Body weight 35 to <70Kg Body weight ≥70 kg

<100 g/L 1,500 mg 2,000 mg

≥100 g/L 1,000 mg 1,500 mg

Discussion Key points: • Modern intravenous iron formula- and conclusion tions offer an improved safety profile Apparently, iron infusion with those with better tolerability, efficacy and newer formulations has a better safety effectiveness compared to oral iron profile than old dextran iron and can therapy. provide a more convenient and effective • There is no strong current evidence alternative to adhering to months of oral to recommend a single best iron iron replacement. Overall, there is little product for infusion among the four current evidence to recommend a single non-dextran formulations. best iron product for infusion among the • Choice of iron product for infusion four non-dextran products. However, will depend on local availability/ formulations that can deliver a replacement guidelines, cost and convenience. with a single large dose would be more convenient and require fewer visits to • Two narrative reviews and one meta- healthcare providers. Although anaphylactic analysis suggested a reconsideration reactions are rare with newer non-dextran of the current paradigm whereby formulations, close monitoring during oral iron treatment is considered a administration is recommended for infusion first-line therapy; it needs further with all IV iron products. Choice of product evaluation. will be determined by local availability/ guidelines, cost and convenience.

Appendix

Appendix Table 1: Head-to-head comparison of non-dextran intravenous iron formulations.

Year of publication Type of study Sample number Note

FCM vs ISC 1. Lee et al (2019)39 Randomized open-label, multicentre 101 Menorrhagia patients

2. Laso-Morales et al (2019)40 RCT 48 Colorectal cancer patients

3. Mahey et al (2017)41 RCT 30 Gynaecological patients

4. Jose et al (2019)42 Randomized clinical trial 100 Pregnant women

5. Naqash et al (2018)43 Phase IV clinical trails 200 IDA in normal women

6. Christoph et al (2012)44 Cohort (retrospective) 206 Pregnant women

7. Sharma et al (2017)45 Cohort (prospective) 120 Post-partum anaemia.

8. Bisbe et al (2011)46 Cohort, multicentre comparative 76 Surgical patients

9. Hofman et al (2018)47 Cohort (retrospective) 221 Renal patients

Total (n) 1102

FCM vs IIM 1. Ehlken et al (2019)48 Cohort retrospective (FCM 309 + IIM 339) 748 Data extraction (2014-2017)

2. Mulder et al (2019)49 Cohort (single-centre) 1334 IDA of any cause

3. Wolf et al (2020)50 Randomized clinical trials 123 IDA of normal adult.

4. Pollock et al (2019)51 Systematic review on RCTs 19 Indirect comparison

2224

FCM vs FOT

1. Adkinson et al (2018)52 Randomized, multicenter, 1997 IDA of any cause double-blind clinical trial

Total (n) 1997

ISC vs IIM 1. Bhandari et al (2015)53 Randomized open-label 351 Renal patients

2. Auerbach et al (2019)54 Prospective, multi-center, randomized 1512 IDA in normal adult comparison

3. Derman et al (2017)55 Randomized open-label, comparative, 511 IDA of any cause multi-center trial

4. Derman et al (2018)56 Randomized trial 248 Gynaecology patients

5. Bhandari et al (2020)57 Randomized, open-label, 1538 Renal patients

Total (n) 4160

ISC vs FOT 1. Macdougall et al (2014)58 Randomized, open-label, multicentre 162 Renal patients

2. Hetzel et al (2014)59 Randomized, open-label 605 IDA of any cause

3. Strauss et al (2016)60 RCTs, post-hoc analysis 767 Renal patients + IDA of any cause

Total (n) 1534

Grand Total (n) 10269

ISC= iron sucrose, FCM= ferric carboxymaltose, IIM= iron isomaltoside (aks ferric derisomaltose), FOT= Ferumoxytol.

Competing interests: Nil. Author information: Dr Tim Aung, FRNZCGP FRACGP: Primary Care Practitioner. Brisbane, QLD. Dr Justin Coleman, MBBS FRACGP MPH: Primary Care Practitioner, Northern Territory. Dr Peter W Davidson, FRACP, PhD: Haematologist (Princess Alexandra Hospital), and Director of QML Pathology. Brisbane, QLD. Dr David J Hetzel, FRACP, PhD: Gastroenterologist, Adelaide, SA. Sandy T Aung, BPharm (UQ): Pharmacist, Logan Hospital, Queensland Health. Brisbane. QLD. Corresponding author: Dr Tim Aung, FRNZCGP FRACGP: Primary Care Practitioner. Brisbane, QLD [email protected] URL: www.nzma.org.nz/journal-articles/intravenous-iron-infusion-and-newer-non-dextran-formulations

REFERENCES 1. World Health Orga- Haehling S. Iron deficiency deficiency anemia. J nization. The global in heart failure. J Cardio- Pediatr Hematol Oncol. prevalence of anemia in vasc Med (Hagerstown). 2017 Jul;39(5):e259-e262. 2011. Geneva: WHO; 2015. 2018 Dec;19(12):706- doi: 10.1097/ ISBN 978 92 4 156496 0. 716. doi: 10.2459/ MPH.0000000000000879. 2. Auerbach M, Ballard JCM.0000000000000686. 12. Stein RE, Plantz K, Maxwell H. Clinical use of 8. Anand IS, Gupta P. Anemia EC, Mamula P, Baldassano intravenous iron: admin- and Iron Deficiency in RN. Intravenous iron istration, efficacy, and Heart Failure: Current sucrose for treatment of safety. Hematology Am Soc Concepts and Emerging iron deficiency anemia in Hematol Educ Program. Therapies. Circulation. pediatric inflammatory 2010;2010:338- 2018 Jul 3;138(1):80-98. bowel disease. J Pediatr 47. doi: 10.1182/ doi: 10.1161/CIRCULA- Gastroenterol Nutr. 2018 asheducation-2010.1.338. TIONAHA.118.030099. Feb;66(2):e51-e55. 3. Zhang E, Tattersal SJN. 9. Tan MLN, Windscheif PM, doi: 10.1097/ Iron deficiency anemia Thornton G, Gaynor E, MPG.0000000000001684. in adults: update. Medi- Rodrigues A, Howarth L. 13. Sabe R, Vatsayan A, Mahran cineToday (Aust). 2019 Retrospective review of A, Khalili AS, Ahuja S, Oct; Vol 20 (10): 27-34. effectiveness and safety of Sferra TJ. Safety and 4. Baird-Gunning J, Bromley intravenous ferric carboxy- efficacy of intravenous iron J. Correcting iron defi- maltose given to children sucrose for iron-deficiency ciency. Aust Prescriber with iron deficiency anemia in children and 2017;40:41. doi: 10.18773/ anemia in one UK tertiary adolescents with inflam- austprescr.2016.096. centre. Eur J Pediatr. 2017 matory bowel disease. Glob Oct;176(10):1419-1423. doi: Pediatr Health. 2019 Aug 5. Simpson J, Ross B, De 10.1007/s00431-017-2995-8. 18;6:2333794X19870981. Malmanche J. Iron doi: deficiency: how to detect it, 10. Powers JM, Shamoun 10.1177/2333794X19870981. how to correct it. Medicine M, McCavit TL, Adix L, Today 2017; 18(10): 40-50. Buchanan GR. Intravenous 14. Hassan N, Boville B, Ferric Carboxymaltose Reischmann D, Ndika 6. South Australian Perinatal in Children with Iron A, Sterken D, Kovey K. Practice Guidelines. Iron Deficiency Anemia Who Intravenous Ferumoxytol infusion. Approved SA Respond Poorly to Oral in Pediatric Patients with Health Safety & Quality Iron. J Pediatr. 2017 Iron Deficiency Anemia. Strategic Governance Jan;180:212-216. doi: Ann Pharmacother. 2017 Committee 19/04/2016. 10.1016/j.jpeds.2016.09.053. Jul;51(7):548-554. doi: www.sahealth.sa.gov. 10.1177/1060028017699429. au/wps/wcm/connect/. 11. Kaneva K, Chow E, Accessed 16/05/2020. Rosenfield CG, Kelly MJ. 15. Hassan N, Cahill J, Intravenous iron sucrose Rajasekaran S, Kovey K. 7. Dinatolo E, Dasseni N, for children with iron Ferumoxytol infusion in Metra M, Lombardi C, von

pediatric patients with Apr;7(4):e342-e350. 10.3109/14767058 gastrointestinal disor- doi: 10.1016/ .2012.746299. ders: first case series. S2352-3026(19)30264-9. 28. Vadhan-Raj S, Strauss Ann Pharmacother. 23. Qunibi WY, Martinez C, W, Ford D, Bernard K, 2011 Dec;45(12):e63. doi: Smith M, Benjamin J, Boccia R, Li J, Allen LF. 10.1345/aph.1Q283. Mangione A, Roger SD. A Efficacy and safety of IV 16. Auerbach M, Macdougall randomized controlled trial ferumoxytol for adults with IC. Safety of intravenous comparing intravenous iron deficiency anemia iron formulations: facts ferric carboxymaltose with previously unresponsive and folklore. Blood Trans- oral iron for treatment of to or unable to tolerate fus. 2014 Jul;12(3):296-300. iron deficiency anemia oral iron. Am J Hematol. doi: 10.2450/2014.0094-14. of non-dialysis-depen- 2014 Jan;89(1):7-12. doi: 17. Strauss WE, Auerbach dent chronic kidney 10.1002/ajh.23582. M. Health-related quality disease patients. Nephrol 29. Khalafallah AA, Hyppa A, of life in patients with Dial Transplant. 2011 Chuang A, Hanna F, Wilson iron deficiency anemia: May;26(5):1599-607. doi: E, Kwok C. A Prospective impact of treatment with 10.1093/ndt/gfq613. Randomised Controlled intravenous iron. Patient 24. Shim JY, Kim MY, Kim YJ, Trial of a Single Intrave- Relat Outcome Meas. 2018 Lee Y, Lee JJ, et al. Effi- nous Infusion of Ferric Aug 27;9:285-298. doi: cacy and safety of ferric Carboxymaltose vs Single 10.2147/PROM.S169653. carboxymaltose versus Intravenous Iron Polymalt- 18. Auerbach M, Macdougall I. ferrous sulfate for iron ose or Daily Oral Ferrous The available intravenous deficiency anemia during Sulphate in the Treatment iron formulations: History, pregnancy: subgroup of Iron Deficiency Anemia efficacy, and toxicology. analysis of Korean women. in Pregnancy. Semin Hema- Hemodial Int. 2017 Jun;21 BMC Pregnancy Childbirth. tol. 2018 Oct;55(4):223-234. Suppl 1:S83-S92. doi: 2018 Aug 28;18(1):349. doi: doi: 10.1053/j. 10.1111/hdi.12560. 10.1186/s12884-018-1817-y. seminhematol.2018.04.006. 19. Cançado RD, Muñoz M. 25. Breymann C, Milman N, 30. Hardy S, Vandemergel Intravenous iron therapy: Mezzacasa A, Bernard X. Intravenous iron how far have we come? R, Dudenhausen J, et al. administration and hypo- Rev Bras Hematol Hemoter. Ferric carboxymaltose vs. phosphatemia in clinical 2011;33(6):461-9. doi: oral iron in the treatment practice. Int J Rheumatol. 10.5581/1516-8484.20110123. of pregnant women with 2015;2015:468675. doi: iron deficiency anemia: an 10.1155/2015/468675. 20. Radhika AG, Sharma international, open-label, AK, Perumal V, et al. 31. Avni T, Bieber A, Grossman randomized controlled trial Parenteral Versus Oral A, Green H, Leibovici L, (FER-ASAP). J Perinat Med. Iron for Treatment of Iron Gafter-Gvili A. The safety of 2017 May 24;45(4):443-453. Deficiency Anemia During intravenous iron prepara- doi: 10.1515/jpm-2016-0050. Pregnancy and post-par- tions: systematic review tum: A Systematic Review. 26. Bhavi SB, Jaju PB. Intra- and meta-analysis. Mayo J Obstet Gynaecol India. venous iron sucrose v/s Clin Proc. 2015 Jan;90(1):12- 2019 Feb;69(1):13-24. doi: oral ferrous fumarate for 23. doi: 10.1016/j. 10.1007/s13224-018-1191-8. treatment of anemia in mayocp.2014.10.007. pregnancy. A randomized 21. Rampton D, Folkersen J, 32. DeLoughery TG. Safety controlled trial. BMC Fishbane S, et al. Hyper- of Oral and Intravenous Pregnancy Childbirth. sensitivity reactions to Iron. Acta Haematol. 2017 May 8;17(1):137. doi: intravenous iron: guidance 2019;142(1):8-12. doi: 10.1186/s12884-017-1313-9. for risk minimization and 10.1159/000496966. management. Haematolog- 27. Froessler B, Cocchiaro C, 33. Canning ML, Gilmore ica. 2014 Nov;99(11):1671-6. Saadat-Gilani K, Hodyl N, KA. Iron stain following doi: 10.3324/ Dekker G. Intravenous an intravenous iron haematol.2014.111492. iron sucrose versus oral infusion. Med J Aust. iron ferrous sulfate for 22. Auerbach M, Gafter-Gvili A, 2017 Jul 17;207(2):58. doi: antenatal and postpartum Macdougall IC. Intravenous 10.5694/mja17.00040. iron deficiency anemia: a iron: a framework for 34. Adkinson NF, Strauss WE, randomized trial. J Matern changing the manage- Macdougall IC, Bernard KE, Fetal Neonatal Med. 2013 ment of iron deficiency. Auerbach M,5, Kaper RF, et May;26(7):654-9. doi: Lancet Haematol. 2020 al. Comparative safety of

intravenous ferumoxytol treatment of postoperative 46. Bisbe E, García-Erce JA, versus ferric carboxy- anemia in colorectal cancer Díez-Lobo AI, Muñoz M; maltose in iron deficiency patients: study protocol for Anemia Working Group anemia: A randomized a randomised controlled España. A multicentre trial. Am J Hematol. 2018 trial. Trials. 2019 Jan comparative study on the May;93(5):683-690. doi: 7;20(1):23. doi: 10.1186/ efficacy of intravenous 10.1002/ajh.25060. s13063-018-3125-2. ferric carboxymaltose and 35. Ganzoni AM. Intravenous 41. Mahey R, Kriplani A, iron sucrose for correcting iron-dextran: therapeu- Mogili KD, Bhatla N, preoperative anemia tic and experimental Kachhawa G, Saxena R. in patients undergoing possibilities. Schweiz Randomized controlled major elective surgery. Med Wochenschr. 1970 trial comparing ferric Br J Anaesth. 2011 Feb 14;100(7):301-3. carboxymaltose and iron Sep;107(3):477-8. doi: 10.1093/bja/aer242. 36. National blood authority sucrose for treatment of (Australia). Iron products iron deficiency anemia due 47. Hofman JMG, Eisenga MF, and dose calculation choice to abnormal uterine bleed- Diepenbroek A, Nolte IM, for adults: guidance for ing. Int J Gynaecol Obstet. van Dam B, Westerhuis R, Australian health provid- 2016 Apr;133(1):43-8. doi: Bakker SJL, Franssen CFM, ers. 2016 March. www. 10.1016/j.ijgo.2015.09.007. Gaillard CAJM. Switching blood.gov.au/... 20052016. 42. Jose A, Mahey R, Sharma iron sucrose to ferric pdf. accessed 18/018/2020. JB, et al. Comparison of carboxymaltose associates to better control of iron 37. Auerbach M, Strauss W, ferric Carboxymaltose status in hemodialysis Auerbach S, Rineer S, and iron sucrose complex patients. BMC Nephrol. Bahrain H. Safety and for treatment of iron 2018 Sep 20;19(1):242. doi: efficacy of total dose deficiency anemia in 10.1186/s12882-018-1045-8. infusion of 1,020 mg of pregnancy- randomised ferumoxytol administered controlled trial. BMC 48. Ehlken B, Nathell L, Gohlke over 15 min. Am J Hematol. Pregnancy Childbirth. A, Bocuk D, Toussi M, 2013 Nov;88(11):944-7. 2019 Feb 4;19(1):54. doi: Wohlfeil S. Evaluation doi: 10.1002/ajh.23534. 10.1186/s12884-019-2200-3. of the Reported Rates of Severe Hypersensitivity 38. Karki NR, Auerbach M. 43. Naqash A, Ara R, Bader GN. Reactions Associated with Single total dose infusion Effectiveness and safety Ferric Carboxymaltose and of ferumoxytol (1020 mg in of ferric carboxymaltose Iron (III) Isomaltoside 1000 30 minutes) is an improved compared to iron sucrose in Europe Based on Data method of administration in women with iron from EudraVigilance and of intravenous iron. Am deficiency anemia: phase VigiBase™ between 2014 J Hematol. 2019 Jun 2. IV clinical trials. BMC and 2017. Drug Saf. 2019 doi: 10.1002/ajh.25548. Womens Health. 2018 Jan 5;18(1):6. doi: 10.1186/ Mar;42(3):463-471. doi: 39. Lee S, Ryu KJ, Lee ES, s12905-017-0506-8. 10.1007/s40264-018-0769-5. Lee KH, Lee JJ, Kim T. 49. Mulder MB, van den Comparative efficacy and 44. Christoph P, Schuller C, Hoek HL, Birnie E, van safety of intravenous ferric Studer H, Irion O, De Tejada Tilburg AJP, Westerman carboxymaltose and iron BM, Surbek D. Intravenous EM. Comparison of sucrose for the treatment iron treatment in pregnan- hypersensitivity reactions of preoperative anemia in cy: comparison of high-dose of intravenous iron: patients with menorrhagia: ferric carboxymaltose vs. iron isomaltoside-1000 An open-label, multicenter, iron sucrose. J Perinat Med. (Monofer® ) versus randomized study. J 2012 May 13;40(5):469-74. ferric carboxy-maltose Obstet Gynaecol Res. doi: 10.1515/jpm-2011-0231. (Ferinject® ). A single 2019 Apr;45(4):858-864. 45. Sharma N, Thiek JL, center, cohort study. Br doi: 10.1111/jog.13893. Natung T, Ahanthem SS. J Clin Pharmacol. 2019 Comparative Study of 40. Laso-Morales MJ, Vives R, Feb;85(2):385-392. doi: Efficacy and Safety of Vallejo-Tarrat A, Caló N, 10.1111/bcp.13805. Valle-Beltran A, Pontes C. Ferric Carboxymaltose 50. Wolf M, Rubin J, Achebe Single dose of intravenous Versus Iron Sucrose in M, Econs MJ, Peacock ferric carboxymaltose Post-partum Anemia. J M, Imel EA, et al. Effects infusion versus multiple Obstet Gynaecol India. 2017 of Iron Isomaltoside vs fractionated doses of intra- Aug;67(4):253-257. doi: Ferric Carboxymaltose venous iron sucrose in the 10.1007/s13224-017-0971-x.

on Hypophosphatemia in Sep;94(9):1007-1014. doi: with iron sucrose for Iron-Deficiency Anemia: 10.1002/ajh.25564. the treatment of iron Two Randomized Clinical 55. Derman R, Roman E, deficiency anemia in Trials. .JAMA. 2020 Feb Modiano MR, Achebe MM, patients with a history of 4;323(5):432-443. doi: Thomsen LL, Auerbach unsatisfactory oral iron 10.1001/jama.2019.22450. M. A randomized trial of therapy. Am J Hematol. 51. Pollock RF, Muduma iron isomaltoside versus 2014 Jun;89(6):646-50. G. A systematic liter- iron sucrose in patients doi: 10.1002/ajh.23712. ature review and with iron deficiency 60. Strauss WE, Dahl NV, Li Z, indirect comparison of anemia. Am J Hematol. Lau G, Allen LF. Ferumoxy- iron isomaltoside and 2017 Mar;92(3):286-291. tol versus iron sucrose ferric carboxymaltose in doi: 10.1002/ajh.24633. treatment: a post-hoc iron deficiency anemia 56. Derman R, Roman E, analysis of randomized after failure or intolerance Smith-Nguyen GN, Achebe controlled trials in patients of oral iron treatment. MM, Thomsen LL, Auer- with varying renal function Expert Rev Hematol. 2019 bach M. Iron isomaltoside and iron deficiency Feb;12(2):129-136. doi: is superior to iron sucrose anemia. BMC Hematol. 10.1080/17474086.201 in increasing hemoglobin 2016 Jul 26;16:20. doi: 9.1575202. in gynecological patients 10.1186/s12878-016-0060-x. 52. Adkinson NF, Strauss WE, with iron deficiency 61. Pollock RF, Biggar P. Indi- Macdougall IC, Bernard KE, anemia. Am J Hematol. rect methods of comparison Auerbach M,5, Kaper RF, et 2018 Jun;93(6):E148-E150. of the safety of ferric deri- al. Comparative safety of doi: 10.1002/ajh.25094. somaltose, iron sucrose and intravenous ferumoxytol 57. Bhandari S, Kalra PA, ferric carboxymaltose in the versus ferric carboxy- Berkowitz M, Belo D, treatment of iron deficiency maltose in iron deficiency Thomsen LL, Wolf M. anemia. Expert Rev Hema- anemia: A randomized Safety and efficacy of iron tol. 2020 Feb;13(2):187-195. trial. Am J Hematol. 2018 isomaltoside 1000/ferric doi: 10.1080/17474086.202 May;93(5):683-690. doi: derisomaltose versus iron 0.1709437. 10.1002/ajh.25060. sucrose in patients with 62. Aksan A, Işık H, Radeke 53. Bhandari S, Kalra PA, chronic kidney disease: HH, Dignass A, Stein J. Kothari J, Ambühl PM, the FERWON-NEPHRO Systematic review with Christensen JH, Essaian randomized, open-label, network meta-analysis: AM, et al. A randomized, comparative trial. Nephrol comparative efficacy and open-label trial of iron Dial Transplant. 2020 tolerability of different isomaltoside 1000 Feb 12. pii: gfaa011. doi: intravenous iron formula- (Monofer®) compared with 10.1093/ndt/gfaa011. tions for the treatment of iron sucrose (Venofer®) 58. Macdougall IC, Strauss iron deficiency anemia in as maintenance therapy WE, McLaughlin J, Li Z, patients with inflammatory in haemodialysis patients. Dellanna F, Hertel J. A bowel disease. Aliment Nephrol Dial Transplant. randomized comparison Pharmacol Ther. 2017 2015 Sep;30(9):1577-89. of ferumoxytol and iron May;45(10):1303-1318. doi: 10.1093/ndt/gfv096. sucrose for treating iron doi: 10.1111/apt.14043. 54. Auerbach M, Henry deficiency anemia in 63. Wetmore JB, Weinhandl D, Derman RJ, Achebe patients with CKD. Clin ED, Zhou J, Gilbertson MM, Thomsen LL, J Am Soc Nephrol. 2014 DT. Relative Incidence of Glaspy J. A prospective, Apr;9(4):705-12. doi: Acute Adverse Events with multi-center, random- 10.2215/CJN.05320513. Ferumoxytol Compared ized comparison of iron 59. Hetzel D, Strauss to Other Intravenous Iron isomaltoside 1000 versus W, Bernard K, Li Z, Compounds: A Matched iron sucrose in patients Urboniene A, Allen LF. Cohort Study. PLoS One. with iron deficiency A Phase III, random- 2017 Jan 30;12(1):e0171098. anemia; the FERWON-IDA ized, open-label trial of doi: 10.1371/journal. trial. Am J Hematol. 2019 ferumoxytol compared pone.0171098.

Granular parakeratosis secondary to benzalkonium chloride exposure from common household laundry rinse aids Catherine JL Tian, Diana Purvis, Harriet S Cheng

ABSTRACT AIM: Granular parakeratosis (GP) is a benign dermatosis characterised by a rash at intertriginous sites. The pathogenesis is uncertain although it is proposed to be an irritant contact reaction with cases related to benzalkonium chloride (BAC) reported. Our experience is that patients often have delayed diagnosis. This study aims to review the clinical presentation and histopathological features of GP. METHODS: This study is a retrospective case series of adult and paediatric patients seen at dermatology clinics in Auckland, New Zealand. Information was collected on patient demographics, presentation, investigations and management. RESULTS: Thirteen cases (seven adults; six children) are included. The typical presentation of GP was erythematous or brown, scaly papules and plaques with desquamation in a predominantly flexural distribution. All patients reported recent exposure to BAC in laundry rinse solution. Nine biopsies were taken from four patients. Psoriasiform and eczematous findings were common on histopathology. The mainstay of treatment was cessation of BAC exposure. CONCLUSION: GP has a distinct clinical pattern although histopathological findings are varied. Clinicians should have a high index of suspicion for GP in patients presenting with erythematous flexural eruptions and seek a history of BAC exposure, especially in the context of the COVID-19 pandemic and increased antiseptic use.

ranular parakeratosis (GP) is a benign understood. The process is possibly exacer- skin condition first described by bated by mechanical and chemical irritation, GNorthcutt et al in 1991 as pruritic, compounded by occlusive environments.2–4 red or brown, hyperkeratotic papules and In recent years, there has been a greater plaques confined to one or both axillae of recognition of the association between GP four patients.1 Since the initial description, and chemical irritants found in antiseptics, extra-axillary sites of involvement have also especially laundry rinse aids and deter- been reported. These include the inter- and gents containing benzalkonium chloride infra-mammary areas, inguinal, groin, peri- (BAC).5,6 BAC is a quaternary ammonium anal and genital skin, beneath the abdomi- cationic compound used as an antimicrobial nal pannus and in non-intertriginous sites preservative for a range of applications. It such as the lumbosacral area.2–4 is active against a wide range of bacteria, GP is thought to be a consequence of yeasts and fungi, but it is increasingly being 7–9 abnormal epidermal differentiation and recognised as a skin irritant. We have keratinocyte maturation from the stratum noticed a number of cases of GP in adults granulosum to the stratum corneum1–3 and children who share a common history through a pathogenesis that is not yet fully of recent exposure to BAC in laundry rinse

solutions commonly found on the New recorded for patients where available. Insti- Zealand market. tutional approval for the report was granted Patients with GP often pose as a diag- by the Auckland District Health Board nostic challenge for clinicians and can Research Office. have a long period of symptomatology with multiple presentations and investi- Results gations prior to correct diagnosis. In the Thirteen cases of GP (seven adults; six context of increasing use of antiseptics children) are included. Four patients (three and disinfectants during and beyond the adults; one child) were seen at the Auckland era of the novel coronavirus (SARS-CoV-2) District Health Board public dermatology COVID-19 pandemic, we also anticipate the outpatient clinic, and nine patients (four irritant effects of BAC to be heightened and adults; five children) were seen at a private that cases of GP will continue to rise. The dermatology clinic in Auckland. The aim of this study was to draw attention to duration from rash onset to GP diagnosis by the presentations and course of illness of the dermatologist ranged from two weeks patients with GP and examine the role of to 18 months. All patients reported recent BAC as a possible aetiology. exposure to Dettol laundry additive (Reckett Benckiser, United Kingdom) or Canesten Methods rinse solution (Bayer, New Zealand), both of We identified 13 cases of GP in adult and which are laundry rinse aids that contain paediatric patients who presented to public BAC (content described in Table 1). One child and private Auckland dermatological clinics (patient five) also had further exposure to between 2015 to 2020. All diagnoses were BAC in QV Flare Up Cream (Douglas Phar- made following specialist dermatologist maceuticals Ltd, New Zealand), which was consultation. Epidemiological data, clinical applied topically after his rash had started, presentation, investigation results and and this resulted in further exacerbation of treatment outcomes were collected from his condition. clinical records, and all patient information Clinical details of the 13 patients are was de-identified. Skin biopsy findings were summarised in Appendix Table 1. Patient

Table 1: Benzalkonium chloride concentrations in selected products available in New Zealand.

Brand Product name and concentration (selected products containing benzalkonium chloride) Dettol • Dettol Anti-bacterial Laundry Sanitiser Eucalyptus: 1.25 litre, con- (Reckett Benckiser, UK) tains benzalkonium chloride 70g per litre

• Dettol Washing Machine Cleaner Citrus Burst: 250ml, contains benzalkonium chloride 2.25%

• Dettol Washing Machine Cleaner Original: 250ml, contains benzalkonium chloride 2.25%

Canesten • Canesten Rinse Solution: 1 litre, contains benzalkonium chloride (Bayer, New Zealand) 70g per litre

QV series • QV Flare Up Cream: 100g tube, contains benzalkonium chloride 0.1% (Douglas Pharmaceuticals) • QV Flare Up Bath Oil: 200 and 500ml bottles, each contains benzalkonium 2.0%

Aveeno (Johnson • Aveeno Dermexa Daily Emollient Cream: 200ml, contains benzalko- & Johnson Ltd, UK) nium chloride (unspecified percentage) • Aveeno Baby Dermexa Emollient Cream: 200ml, contains benzalkonium chloride (unspecified percentage)

ages ranged from eight months to 72 years growth. Nine 4mm skin biopsies of the (median=31 years). Ten of the 13 patients affected areas were taken from four adults: were of full or partial Asian ethnicity. Six seven of the nine specimens showed typical patients had a personal (n=5) or family features of parakeratosis, acanthosis and (n=1) history of atopic or irritant contact mixed dermal inflammatory infiltrates. dermatitis. One patient also had active None of the nine biopsy specimens showed psoriasis at the time of diagnosis, which evidence of bacterial or fungal infection was confirmed histologically. The rashes on serial stains. The most frequent histo- shared a common theme of erythem- logical diagnoses were dermatitis (eczema) atous papules and plaques, accompanied or psoriasis. None of the paediatric patients by varying degrees of desquamation and had skin biopsies performed. Patch testing scaling (Figures 1 and 2). There was asso- was completed for two patients and both ciated lichenification (indicating chronicity) were negative for a series of allergens in eight patients. Four patients reported including BAC, thereby excluding allergic pruritus. The distribution included the contact dermatitis. axillae, groin, trunk, limbs, anterior neck Topical preparations (most commonly and natal cleft. Interestingly, in one patient emollients and mild to moderate potency the rash was noticed on the helices of the topical corticosteroids) had commonly been ears. In two patients, the rash was predomi- trialled prior to presentation to the derma- nantly in a pattern of distribution reflective tologist’s clinic, and systemic therapies, of areas of close contact with clothing or such as broad-spectrum antibiotics or anti- fabric (neckline, nappy and waistband). fungals, had also been prescribed in some Blood tests were done on six patients and cases. These yielded variable responses. did not reveal any systemic involvement. At the time of diagnosis, patients were Skin scrapings were taken from six patients advised to avoid further contact with BAC and all yielded skin flora of insignificant and rewash clothing without the BAC-con-

Figure 1: Granular parakeratosis in an adult. Close up of the back and left axilla, showing an erythematous rash with areas of hyperkeratosis at the peripheries of the eruption (a, b). Hyperkeratosis and fissuring in the left elbow flexure (c). Hyperpigmentation, fissuring and desquamation at the neckline (d).

Figure 2: Granular parakeratosis in a child. Clinical photography showing the morphology and distribution of the erythematous and brownish hyperkeratotic papules and plaques, around the umbilicus (e), in the front of the torso and axilla (f, g).

taining laundry rinses. Follow-up data were may be abnormal processing of profil- available for five cases, all of which reported aggrin to filaggrin (structures that maintain improvement or resolution of skin rash the keratohyalin granules in the stratum following cessation of BAC exposure. corneum during cornification).1–3 Physical factors such as hyperhidrosis, obesity and Discussion friction are also thought to contribute to the evolvement of GP via chemical and The clinical features in our 13 patients fit or mechanical irritation, resulting in well with the characteristic appearance of compromised epidermal maturation and GP described in published literature.1,3,9–12 barrier function.2,5,10 A humid and occlusive Hyperkeratotic papules on an erythem- environment, commonly found in cuta- atous or brown base, often coalescing neous folds, may exacerbate penetration into plaques in a predominantly flexural of irritants into the skin, which possibly distribution, are characteristic. Non-inter- explains the predominantly intertriginous triginous involvement of the lumbosacral distribution.2,4,13,14 Five of our patients also area, abdomen, limbs, face and neck was also had atopic dermatitis, which is a spec- noted in our cases. Involvement of the helices ulated risk factor for GP, as epidermal of the ears has not previously been reported barrier dysfunction in atopic dermatitis and we postulate that this may be related to may also facilitate increased penetration site of contact with traces of BAC on bedding. of irritants.13,15 The skin microbiome GP has been described in all ages and is also important in mediating various both sexes, although it is more commonly processes involving immune responses 9–11 reported in females. In our series, the age and epidermal development and differen- of our patients at time of rash onset ranged tiation.16 Previous studies have commented widely from eight months to 72 years and on how, in response to internal or external just over half of cases were in females (7/13, factors, altered skin microbial communities 54%). Interestingly, 10/13 (77%) of our cohort contribute to the disease pathology of a were of full or partial Asian ethnicity. To the number of cutaneous conditions including best of our knowledge, the occurrence of GP acne, atopic dermatitis and psoriasis.17 has not previously been described to have a Perhaps akin to the hypothesis of atopic particular geographical or racial predilection. dermatitis being associated with a loss of Our finding may reflect local cultural prac- microbiome diversity secondary to over- tices related to laundry and hygiene. abundance of cutaneous Staphylococcus There are many histopathologic variants aureus,18 Kumarasinghe et al postulate that of GP,11 with the most common findings flexural hyperkeratotic lesions such as GP being hyperkeratosis, parakeratosis, could be triggered by an overgrowth of flora epidermal acanthosis and hypergranu- with a predominance of anaerobes.19 In losis.2,3 Mild to moderate capillary dilatation, recent years GP has also been reported in proliferation of the upper papillary dermis association with chemotherapy for ovarian and scattered perivascular inflammatory and breast carcinoma, as well as several lymphohistiocytic infiltrate are also keratinocytic neoplasms.9,20,21 3,10,11 described. Some histological features Of the contact irritants that may may overlap with psoriasis, and many of our contribute to development of GP, BAC is most cases also had spongiosis in the epidermis, widely reported.5,7–9,22,23 BAC, an ammonium which is classically seen in dermatitis compound commercialised for more than 50 (eczema). Of note, eczema and psoriasis years, is used as an antiseptic and preser- were common histological misdiagnoses in vative7,23 across a wide range of applications our series. We propose that GP is unlikely commonly found on the New Zealand to be diagnosed on histopathology alone; in market, including but not limited to eye most cases, GP should be a clinical diagnosis. drops, bath oils, skin cleansers, sanitisers The precise aetiology and triggers of GP and laundry rinse aids. These are alternative remain uncertain at present. Microscopic names for BAC: studies showing decreased cytoplasmic • N-Alkyl-N-benzyl-N expression of filaggrin during cornification • N-dimethylammonium chloride suggests that the primary underlying cause

• Alkyldimethylbenzylammonium Australian Baseline Series (ABS). Prolonged chloride exposure to BAC can potentially predispose • ADBAC individuals to sensitisation and induce allergic contact dermatitis.5,7,10,24 • BC50 The Environmental Protection Authority • BC80 (EPA), a Crown Agent established in 2011, • Alklbenzyldimethyl regulates BAC in New Zealand. The EPA • Alkyl benzyldimethyl ammonium categorises BAC of different concentrations chloride (>33%; >5–25%; >1–5%) into classification • Ammonyx codes as per the Hazardous Substances and New Organisms Act 1996. All businesses • Barquat MB-50 selling goods containing chemicals must • Barquat MB080 apply for hazardous-substances approval • Benirol from the EPA, provide an accessible chem- • Bradophen ical-safety data sheet and comply with • BTC labelling in accordance with the hazards of the particular chemical concentration. • Cequartyl As per the EPA’s most recent reassessment • Drapolene of BAC in June 2020, BAC at all concentra- • Dropolex tions is not classified as a skin sensitiser. • Enuclene This means that in New Zealand there is no mandatory requirement for goods • Germitol containing BAC to display any labelling • Gesminol to warn consumers of potential irritant • Osuan or sensitisation effects. There is currently • Paralkan no limit on BAC concentration in cleaning products, although for products directly • Parasterol in contact with skin, the EPA limits BAC • Quaternary ammonium compounds content to 3% in hair products and 0.1% in rodalon other cosmetics, such as hand-sanitisers. In • Zephiran light of an increasing awareness of BAC’s • Zephiran chloride irritant properties and association with GP, we hence propose closer surveillance and • Zilkonium chloride an updated review of BAC at its various Robinson et al described GP in a suscep- commercially available concentrations, with tible subset of patients following exposure to a particular focus on cleaning products and 5 laundry wash containing BAC; and a paedi- other common household solutions and atric case study in Brazil reported six cases their potential role in dermal irritation or of GP following exposure to commonly used sensitisation. infant skincare products containing BAC.14 There is currently no optimal or stan- In our series, all patients reported use of a dardised approach to treatment for GP. laundry rinse containing BAC prior to devel- There is also a paucity of high-quality opment of the rash. BAC is thought to irritate evidence and controlled trials. According to the skin by disrupting cellular epidermal our experience and other reports, treatment lipid bilayers and promoting inflammatory in the form of topical and systemic corti- cell infiltration, leading to activation of costeroids, retinoids, vitamin D analogues leucocytes, granulocytes, inflammatory (eg, calcipotriene), keratolytics (such as proteins and cytokines (tumour necrosis salicyclic acid and ammonium lactate), anti- factor-alpha, prostaglandin E2, interleuk- fungals and antimycotics are all of variable ins-1a, -1b, -6, and -8).7,8,22 As an antiseptic, efficacy.15,25–27 Recently, the use of broad- BAC may also contribute to disease spectrum oral antibiotic therapy has been pathology by means of inducing changes reported, with the proposed mechanism to resident microbiome populations and being modification and correction of the disrupting skin homeostasis. Although less skin microbiome.19 However, this therapy frequently reported as a sensitiser, BAC was trialled in two of our cases without is also listed as a contact allergen on the

noticeable improvement. The use of neuro- or Letterer–Siwe disease.12,13 Clinical and toxin Clostridium botulinum type A has been investigative findings, histopathological described in patients who also suffer from subtleties, as well as limited or no response hyperhidrosis;29 and destructive procedural to targeted treatment(s) will help differen- treatments such as cryotherapy, YAG and tiate these from GP. 30 carbon dioxide lasers are also reported. This report is a retrospective case series, Many patients demonstrate spontaneous which makes it difficult to prove causality. improvement, although the course of the Only two of our patients received formal eruption varies from several weeks to skin patch testing to exclude an allergic 3,4 years. In general, we consider general contact dermatitis; however, we consider skin cares, withdrawal of BAC-containing patch testing to be unnecessary in most laundry rinses, and avoidance of irritants to cases where the clinical presentation is clas- be the mainstay of treatment. sical as GP arises from irritant rather than The variability in time from rash onset allergic mechanisms. Patch testing should be to diagnosis of GP in our cases (two weeks reserved for patients who do not respond to to 18 months) may be reflective of the lack avoidance of BAC and other irritants. of familiarity with this condition among clinicians locally and perhaps even inter- Conclusion nationally. Of the 363,343 skin biopsies Our case series describes the development from flexural dermatoses submitted by of GP in a cohort of patients exposed to BAC dermatologists at a large institute of derma- via laundry rinses and provides further topathology in New York, the frequency of evidence for the hypothesis that GP is an GP on histology was 0.005% (18 of 363,343), irritant contact reaction pattern with BAC with only one correct clinical diagnosis as a likely culprit. Our experience is that in the 18 histologically confirmed cases of patients often have delayed diagnosis. We GP. 11 Similarly, Braun Falco and colleagues advocate that healthcare professionals found histopathological features of GP in maintain clinical suspicion for GP when ten of 250,000 skin biopsies (0.004%), but encountering patients presenting with GP had not been considered as a clinical flexural dermatoses, and a history of recent diagnosis in any of the ten cases.10 Our cases exposure to BAC should be specifically demonstrated that histopathology is rarely sought. We anticipate this to be particularly definitive in GP, which may lead to alter- relevant in the context of the COVID-19 native diagnoses being sought, contributing pandemic and an anticipatory increased to further diagnostic delay. use of antiseptics. Heightened awareness The differential diagnoses of inter- of GP results in a more-timely diagnosis triginous rashes are broad and include with reduced medical visits, investigations seborrhoeic dermatitis, irritant or allergic and unnecessary therapies. Clinical diag- contact dermatitis, acanthosis nigricans, nosis is usually possible, and biopsy should Hailey–Hailey disease, Darier’s disease, only be considered to exclude differential pemphigus vegetans, candidiasis, derma- diagnoses in selected cases. We propose tophytosis, flexural drug eruptions and that the mainstay of treatment remains as 3,9–11 Dowling–Degos disease. In the paediatric identification and removal of skin irri- population, differentials also include napkin tants including BAC, in combination with dermatitis, acrodermatitis enteropathica, emollients.

Appendix

Appendix Table 1: Characteristics of patients presenting with granular parakeratosis.

Case Age Gender / Medical Presentation Investigations Histology Treatment Progress Ethnicity history (4mm skin biopsy) 1* ++ 8 Male, Asian Infantile 6-month history worsening • No investigations. Clinical No skin biopsy Emollient and No follow-up data months eczema, eczema on face and limbs since diagnosis potent topical available. seborrheic 2 months of age. More recently corticosteroid. dermatitis developing rash: red brown Advised to avoid plaques on upper thighs sparing BAC and rewash the folds. clothes.

2++ 10 Female, Irritant and Several months history of • PCR: Herpes simplex and No skin biopsy Topical pimecro- No response to months New periorificial erythematous, peeling rash Varicella zoster negative limus. Vaseline topical steroids. No Zealand dermatitis of in the groin mirroring contact • Skin swab right thigh: as barrier cream. complete follow-up European face, infantile with nappies, sparing inguinal normal flora Advised to avoid data available. haemangi- creases. Facial eruption with BAC and rewash oma papules around eyes, nose and nappies. mouth (concurrent diagnosis of periorificial dermatitis).

3++ 4 years Female, None Several months history of red/ • Skin scraping groin: no No skin biopsy Emollient and No response to Asian reported brown exfoliative symmetrical fungus moderate potency topical steroid or groin rash. topical steroid antifungals. De- and antifungal. spite advice contin- Advised to avoid ued to use laundry BAC and rewash rinse with BAC and clothes. rash persisted. Advice re-enforced on follow-up and rash resolved.

Appendix Table 1: Characteristics of patients presenting with granular parakeratosis (continued).

Case Age Gender / Medical Presentation Investigations Histology Treatment Progress Ethnicity history (4mm skin biopsy) 4+ 6 years Male, New Molluscum 16-month history bran-like scal- • Skin autoantibodies No skin biopsy Emollient, 10% Limited improve- Zealand contagiosum. ing on an erythematous base. negative. urea as kerato- ment with topical European Atopic ecze- Rash is pruritic and scattered • Plasma zinc; glucagon; lytic and mild to creams. Rash ma (mother) pustules noted. Originated from amino acids; ANA, iron moderate potency resolved at time and asthma neck, spreading to face, chest, studies; liver function, topical corticoste- of clinic follow up. (father) groin and legs. Rash later re- coeliac markers, thyroid roid. Diagnosis of GP curred as erythematous plaques function, immunoglobu- made retrospec- in the groin, with a well-defined lin, B12/folate normal, tively. peeling edge. Sister (9years) also • Urine: amino acid and had a milder scaly rash around organic acid screen neg- the neck and forearms. ative. • Skin scraping neck: no fungus • Skin prick and patch test: No reaction (house dust mite, cat, dog, Alternaria, aspergillus, mixed grass, perennial rye, plantain, birch mix, soyabean, cow’s milk, egg white, peanut, wheat, shrimp, fish mix). Open application to zinc oxide sunscreen negative.

Appendix Table 1: Characteristics of patients presenting with granular parakeratosis (continued).

Case Age Gender / Medical Presentation Investigations Histology Treatment Progress Ethnicity history (4mm skin biopsy) 5++ 9 years Male, New Infantile 3-month history dry, irritated • Skin scraping axilla: no No skin biopsy Emollient and Improved at 4 Zealand eczema eruption. Started in axillae, fungus advised to avoid weeks European spread down sides of torso, onto • Skin swab: normal flora BAC and rewash back and behind knees. No re- clothes. sponse to topical antifungals or topical corticosteroids. Exposed to Canestan laundry rinse for several years, QV flare up cream used after eruption started, making rash worse.

6++ 11 years Male, Childhood 12-month history worsening • No investigations. Clinical No skin biopsy Emollient and No follow-up data Chinese eczema eczema. Superficial brown, diagnosis potent topical available. European desquamating rash around corticosteroid. waistband of underwear and Advised to avoid trousers. BAC and rewash clothes.

7++ 31 years Female, None report- 4-week history reticulate erythe- • No investigations. Clinical No skin biopsy Emollient, potent Did not attend for Asian ed ma and scaling over lower ab- diagnosis topical cortico- planned follow up. dominal wall, confluent in groin steroid and oral and on buttocks. Later spreading loratadine for to face and helix of ears, then pruritus. Advised arms and hands. to avoid BAC and rewash clothes.

8++ 32 years Female, None report- 18-month history peeling skin • No investigations. Clinical No skin biopsy Emollients, Noticed im- Chinese ed with subtle hyperpigmentation diagnosis salicylic acid ker- provement after in the axillae, hips and forearms. atolytic and mild stopping laundry potency topical rinse. Recurrence corticosteroid. after wearing yoga Avoid soap and clothes which BAC and rewash had not been clothes. rewashed.

Appendix Table 1: Characteristics of patients presenting with granular parakeratosis (continued).

Case Age Gender / Medical Presentation Investigations Histology Treatment Progress Ethnicity history (4mm skin biopsy) 9++ 34 years Male, Asian None report- 2-month history progressive • Skin autoantibodies Skin biopsy left cubital Emollient and Rash resolved at ed eruption abdomen, groin, negative fossa and groin: Features advised to avoid follow up. antecubital and popliteal fossae, • Renal and liver function of dermatitis with ortho- BAC and rewash genitalia and axillae. Progressed normal keratosis, parakeratosis clothes. to involve trunk. Mild erythema • QuantiFERON TB Gold, and mild spongiosis. with brown bran-like scale. HIV and hepatitis screen Perivascular and inter- Widespread xerosis. No pruritus. negative / stitial chronic inflamma- • Skin scraping left thigh: no tory infiltrate in dermis, fungus including occasional • Skin swab left axilla: nor- eosinophils. mal flora • Patch testing: BAC negative

10+ 36 years Female, 25/40 preg- 6-week history rapidly evolving • QuantiFERON TB Gold, HIV Skin biopsy back, forearm Admitted as Improved after Chinese nant, pustu- peeling rash torso and groins antibody and trepone- and thigh: Psoriasiform inpatient given cessation of BAC. lar psoriasis associated with flare of pustular mal antibodies negative, features with parakerato- concurrent Discharged once psoriasis. Concurrent diagnosis skin autoantibodies not sis and neutrophils within diagnosis of stable and pustules of GP and pustular psoriasis of detected parakeratotic scale. pustular psoriasis. cleared. pregnancy. • Skin scraping right arm: Underlying superficial Treatment with no fungus perivascular lymphocytic moderate potency inflammatory infiltrate topical corticoste- with occasional eosino- roid, emollients phils. and ciclosporin. Advised to avoid soap and BAC and to rewash clothing.

Appendix Table 1: Characteristics of patients presenting with granular parakeratosis (continued).

Case Age Gender / Medical Presentation Investigations Histology Treatment Progress Ethnicity history (4mm skin biopsy) 11+ 42 years Female, None report- 2-week history itchy and painful • Iron, FBC, liver, renal and Skin biopsy left axil- Emollient and No follow-up data Chinese ed rash originating in groins. Ery- thyroid function normal. la, abdomen, breast: soap substitute. available. thematous, fissured and scaly QuantiFERON TB Gold, Features of impetiginised Advised to avoid papules coalescing into plaques. HIV antibody, hepatitis B dermatitis and/or subcor- BAC and rewash Rash later spread to become and C antibodies negative. neal pustulosis with mild clothes. symmetrical, well-demarcated ANA; ANCA; plasma zinc; acanthosis, parakeratosis areas of desquamation and and skin autoantibodies and spongiosis. Mounds hyperpigmentation on anterior negative / of parakeratosis contain and posterior trunk, including • Skin scraping groin: no numerous neutrophils. breasts. fungus Perivascular mixed • Skin swab left knee infiltrate in dermis. pustule: Staphylococcus Differentials include aureus impetiginized eczema- • Skin swab groin: normal tous reaction, a form of flora, no yeast isolated psoriasis or subcorneal neutrophilic dermatosis.

12++ 67 years Female, Childhood 6-month history painful dry rash ANA and RF negative. Skin biopsy left chest: Moderate potency Incomplete re- Indian eczema, originating from right axilla. B12, folate and FBC normal Features of dermatitis topical corticoste- sponse to topical allergic rhini- Asteototic plaques with erythe- with mild spongiosis and roid; oral erythro- steroid and anti- tis, salicylate ma and lichenification affecting compact parakeratosis in mycin 2 weeks (al- biotics. Follow-up sensitivity bilateral axillae, sub-mammary epidermis with super- lergy to penicillin). pending. and multiple fold and groin. Not pruritic. ficial perivascular lym- Advised to avoid drug allergies phohistiocytic infiltrate. BAC and rewash Scattered lymphocytes clothing. at dermoepidermal junc- tion without basement membrane destruction or keratinocyte vacuo- lation. Common pattern occurring as a cutaneous reaction to drugs

Appendix Table 1: Characteristics of patients presenting with granular parakeratosis (continued).

Case Age Gender / Medical Presentation Investigations Histology Treatment Progress Ethnicity history (4mm skin biopsy) 13+ 72 years Male. Asian Congestive 1-year history hyperpigmented, QuantiFERON TB Gold, HIV, No skin biopsy Topical and oral Incomplete re- heart failure, dry rash in axillae with general- hepatitis B&C screen negative steroid therapy, sponse to topical type 2 diabe- isation to involve the popliteal Skin autoantibodies negative antifungals, and or oral steroid ther- tes mellitus, and antecubital fossae; anterior course of roxithro- apies; antifungals; hypertension abdominal wall and natal cleaft/ mycin. Emollient or antibiotics. No groin. Well-demarcated, xerotic, and soap substi- complete follow-up hyperpigmented plaques with tute. Advised to data available. fissures, peeling and scale. Initial avoid bleach and diagnosis of seborrhoeic derma- BAC and rewash titis, treated with itraconazole clothes. which precipitated liver dysfunction and uncovered diagnosis of hepatocellular carcinoma.

*Telehealth consultation during COVID-19 lockdown. + Seen in public. ++ Seen in private. ANA (antinuclear antibodies); RF (rheumatoid factor); ANCA (antineutrophil cytoplasmic antibodies); HIV (human immunodeficiency virus); FBC (full blood count).

Competing interests: Nil. Author information: Dr Catherine JL Tian: House Officer (Doctor), Auckland District Health Board, Auckland, New Zealand. Dr Diana Purvis: Dermatologist, Department of Dermatology, Auckland District Health Board and University of Auckland, Auckland, New Zealand. Dr Harriet S Cheng: Dermatologist, Department of Dermatology, Auckland District Health Board and University of Auckland, Auckland, New Zealand. Corresponding author: Dr Harriet S Cheng, Department of Dermatology, Auckland District Health Board, Private bag 92189, Auckland Mail Centre, Auckland 1142, (09) 367 0000 [email protected] URL: www.nzma.org.nz/journal-articles/granular-parakeratosis-secondary-to-benzalkonium-chloride-exposure-from-common-household-laundry-rinse-aids

REFERENCES 1. Northcutt AD, Nelson 6. Tartari F, Vincenzi C, correlation of 10 cases. DM, Tschen JA. Axillary Di Altobrando A, et al. J Dtsch Dermatol Ges. granular parakeratosis. Allergic contact dermatitis 2009 Apr;7(4):340- J Am Acad Dermatol. to benzalkonium chloride 4. English, German. 1991 Apr;24(4):541-4. doi: with erythema multi- doi: 10.1111/j.1610- 10.1016/0190-9622(91)70078- forme-like reaction in a 0387.2008.06964.x. g. child. Contact Dermatitis. Epub 2008 Dec 3. 2. Mehregan DA, Thomas 2020 Jun;82(6):397-399. PMID: 19054421. JE, Mehregan DR. doi: 10.1111/cod.13481. 11. Scheinfeld NS, Mones J. Intertriginous granular Epub 2020 Apr 3. Granular parakeratosis: parakeratosis. J Am PMID: 32112429. pathologic and clinical Acad Dermatol. 1998 7. Basketter DA, Marriott M, correlation of 18 cases of Sep;39(3):495-6. doi: 10.1016/ Gilmour NJ, et al. Strong granular parakeratosis. J s0190-9622(98)70333- irritants masquerading as Am Acad Dermatol. 2005 0. PMID: 9738790. skin allergens: the case of May;52(5):863-7. doi: 3. Metze D, Rütten A. Granu- benzalkonium chloride. 10.1016/j.jaad.2004.12.031. lar parakeratosis - a unique Contact Dermatitis. 2004 PMID: 15858479. acquired disorder of kera- Apr;50(4):213-7. doi: 12. Pimentel DR, Michalany N, tinization. J Cutan Pathol. 10.1111/j.0105-1873.2004.00331.x. Morgado de Abreu MA, et 1999 Aug;26(7):339-52. doi: PMID: 15186375. al. Granular parakeratosis 10.1111/j.1600-0560.1999. 8. Wentworth AB, Yiannias in children: case report and tb01855.x. PMID: 10487291. JA, Davis MD, et al. review of the literature. 4. Wohlrab J, Lüftl M, Wolter Benzalkonium Chloride: Pediatr Dermatol. 2003 M. Submammary granular A Known Irritant and May-Jun;20(3):215-20. parakeratosis: an acquired Novel Allergen. Dermatitis. doi: 10.1046/j.1525- punctate hyperkeratosis of 2016 Jan-Feb;27(1):14-20. 1470.2003.20306.x. exogenic origin. J Am Acad doi: 10.1097/ PMID: 12787269. Dermatol. 1999 May;40(5 Pt DER.0000000000000160. 13. Patrizi A, Neri I, Misciali 2):813-4. PMID: 10321622. PMID: 26756511. C, et al. Granular paraker- 5. Robinson AJ, Foster 9. Wallace CA, Pichardo atosis: four paediatric RS, Halbert AR, et al. RO, Yosipovitch G, et al. cases. Br J Dermatol. 2002 Granular parakeratosis Granular parakeratosis: a Nov;147(5):1003-6. doi: induced by benzalkonium case report and literature 10.1046/j.1365-2133.2002.04953.x. chloride exposure from review. J Cutan Pathol. PMID: 12410715. laundry rinse aids. Austral- 2003 May;30(5):332-5. doi: 14. Giraldi, S, Abagge, KT, as J Dermatol. 2017 10.1034/j.1600-0560.2003.00066.x. Carvalho, VOd, et al. Para- Aug;58(3):e138-e140. PMID: 12753175. queratose granular: relato doi: 10.1111/ajd.12551. 10. Braun-Falco M, Laaff H. de seis casos em crianças. Epub 2016 Sep 19. Granular parakeratosis--a Anais Brasileiros de PMID: 27641714. clinical-pathological Dermatologia, 81(1), 59-64.

https://doi.org/10.1590/ treated with liposomal doi: 10.1001/arch- S0365-05962006000100008 doxorubicin for ovarian dermatol.2011.214. 15. Borok J, Matiz C, Gold- carcinoma. J Am Acad PMID: 21844474. enberg A, et al. Contact Dermatol. 2008 May;58(5 26. Contreras ME, Gottfried Dermatitis in Atopic Suppl 1):S84-7. doi: LC, Bang RH, et al. Axillary Dermatitis Children-Past, 10.1016/j.jaad.2007.05.031. intertriginous granular Present, and Future. Clin PMID: 18489055. parakeratosis responsive Rev Allergy Immunol. 21. Resnik KS, DiLeonardo to topical calcipotriene 2019 Feb;56(1):86-98. doi: M. Incidental granular and ammonium lactate. 10.1007/s12016-018-8711- parakeratotic cornifica- Int J Dermatol. 2003 2. PMID: 30225535. tion in carcinomas. Am May;42(5):382-3. doi: 16. Sanmiguel, Adam, Eliza- J Dermatopathol. 2007 10.1046/j.1365-4362.2003.01722.x. beth A. Grice. Interactions Jun;29(3):264-9. doi: 10.1097/ PMID: 12755978.roebrock between Host Factors DAD.0b013e3180465860. 27. Samrao A, Reis M, Niedt and the Skin Microbiome. PMID: 17519624. G, Rudikoff D. Granular Cellular and Molecular Life 22. Roebrock K, Wolf M, parakeratosis: response Sciences 72, no. 8. 2014: Bovens S, et al. Inhibition of to calcipotriene and brief 1499-515. doi:10.1007/ benzalkonium chloride-in- review of current thera- s00018-014-1812-z. duced skin inflammation peutic options. Skinmed. 17. Weyrich LS, Dixit S, in mice by an indol-1- 2010 Nov-Dec;8(6):357-9. Farrer AG et al. The skin ylpropan-2-one inhibitor PMID: 21413654. microbiome: associations of cytosolic phospholipase 28. Brown SK, Heilman between altered microbial A2 α. Br J Dermatol. 2012 ER. Granular paraker- communities and disease. Feb;166(2):306-16. doi: atosis: resolution with Australas. J. Dermatol. 10.1111/j.1365-2133.2011.10637.x. topical tretinoin. J Am Acad 2015; 56: 268–74. Epub 2011 Dec 6. Dermatol. 2002 Nov;47(5 PMID: 21929537. 18. Kong HH, Oh J, Deming C Suppl):S279-80. doi: et al. Temporal shifts in the 23. Sadakane K, Ichinose T. 10.1067/mjd.2002.109252. skin microbiome associated Effect of the hand antisep- PMID: 12399751. with disease flares and tic agents benzalkonium 29. Ravitskiy L, Heymann treatment in children with chloride, povidone-iodine, WR. Botulinum toxin-in- atopic dermatitis. Genome ethanol, and chlorhexidine duced resolution of Res. 2012; 22: 850–9. gluconate on atopic derma- axillary granular para- titis in NC/Nga mice. Int. J. 19. Kumarasinghe SPW, keratosis. Skinmed. 2005 Med. Sci. 2015; 12: 116–25. Chandran V, Raby E, et al. Mar-Apr;4(2):118-20. Hyperkeratotic flexural 24. Isaac J, Scheinman PL. doi: 10.1111/j.1540- erythema responding to Benzalkonium Chlo- 9740.2005.03700.x. amoxicillin-clavulanic acid ride: An Irritant and PMID: 15785142. therapy: Report of four Sensitizer. Dermatitis. 30. Laimer M, Emberger cases. Australas J Dermatol. 2017 Nov/Dec;28(6):346- M, Brunasso AM, et al. 2019 Nov;60(4):311-314. 352. doi: 10.1097/ Laser for the treatment of doi: 10.1111/ajd.13069. DER.0000000000000316. granular parakeratosis. Epub 2019 May 2. PMID: 28885313. Dermatol Surg. 2009 PMID: 31049927. 25. Patel U, Patel T, Skinner RB Feb;35(2):297-300. doi: 20. Jaconelli L, Doebelin B, Jr. Resolution of granular 10.1111/j.1524-4725.2008.01052.x. Kanitakis J, et al. Granular parakeratosis with topical Erratum in: Dermatol parakeratosis in a patient calcitriol. Arch Dermatol. Surg. 2009 Mar;35(3):557. 2011 Aug;147(8):997-8. PMID: 19215276.

Cerebrospinal fluid rhinorrhoea following nasal packing for epistaxis: case series from a single centre Alice Flavell-Birch, Campbell Baguley, Alice Stringer

pistaxis is one of the most common packing, with spontaneous resolution of ear, nose and throat (ENT) emergen- headache and rhinorrhea at three weeks Ecies, the majority of which are due to follow-up. bleeding from an anterior site that can be Case two: A 61-year-old woman, who managed with simple cautery measures. required anticoagulation for metallic Packing the nose is reserved for recalcitrant replacement heart valve, presented with bleeding or for bleeding from a posterior right-sided epistaxis. Upon admission to site that cannot be accessed. There is a trend the emergency department a Rapid Rhino towards immediate packing of the nose was placed. The pack was removed and no by first responders and emergency physi- specific bleeding point was identified, only cians with one of two commonly used nasal damaged mucosal ooze. NasoPore (Stryker, packing/tamponade devices in New Zealand: Kalamazoo, Michigan, USA) was placed Rapid Rhino (Smith & Nephew plc, Watford, over the mucosal tears for non-compressive England) and Merocel (Medtronic, Mystic, pro-coagulant effects. NasoPore is a soft Conneticut, USA). and absorbable porous packing material In recent years our department has that assists in achieving tamponade without encountered three cases of traumatic cere- firm pressure. It is often placed after endo- brospinal fluid (CSF) rhinorrhea following scopic nasal examination and allows the premature use of Rapid Rhino nasal packing addition of further topical medication, such devices for spontaneous anterior epistaxis. as tranexamic acid. It is not recommended as an alternative product to use in the Cases initial management of epistaxis. Bleeding restarted on the contralateral (left) side Case one: A medically frail, 85-year-old during admission requiring insertion of female presented with spontaneous right- Rapid Rhino pack in the left nostril. This was sided epistaxis that progressed to bilateral removed successfully, and further NasoPore bleeding. She had bilateral nasal packs was placed. The patient developed haemo- placed by first responders in the community. serous rhinorrhea on the right side that These were noted to be sitting vertically tested positive for beta-2 (tau) transferrin. and partially out of the nares, with severe CT imaging did not identify any skull base pain experienced during insertion and a defects. The patient was given prophylactic persisting headache. antibiotics as she had foreign material Following removal, rhinoscopy showed (NasoPore) in the nose and was observed nasal mucosal tearing from the nasal with resolution of rhinorrhoea within septum with an anterior bleeding site, fourteen days. followed by continuous bilateral rosé-co- Case three: A 56-year-old man with a loured rhinorrhea. This tested positive for background of well-controlled hypertension beta-2 (tau) transferrin. CT sinus series did presented with a first episode of sponta- not convincingly demonstrate any fractures. neous unilateral epistaxis. On transfer from She remained under observation without a peripheral emergency department he further active management or nasal

had a Rapid Rhino nasal pack in situ and to thoroughly soak the Rapid Rhino prior to reported intense pain during its insertion. insertion. By elevating the anterior nares This was removed and cautery was with one hand, the clinician can view the performed to the anterior bleeding point. He anterior nose and septum and ensure safe developed ipsilateral clear rhinorrhea that insertion along the floor of the nose (Figure persisted and was positive for beta-2 (tau) 2). transferrin. If CSF leak is suspected, a fluid sample CT sinuses showed a possible disruption should undergo electrophoresis testing for in the cribriform plate, so neurosurgical the presence of beta-2 (tau) transferrin.7 opinion was sought and surgical exploration Antibiotics are not routinely recommended was recommended. He proceeded to have as prophylaxis and resolution usually occurs an ipsilateral comprehensive functional with bed rest and observation.1,2 endoscopic sinus surgery, partial middle-tur- Recommendations binate resection and exploration of anterior It has been our experience that Rapid skull base. No obvious leak was found. Rhino packs are being used frequently Gelatin sponge pieces were applied to the without identification and treatment posterior olfactory cleft at the middle-tur- attempts to the bleeding sites. Our binate attachment point, because a subtle department is actively engaging with defect was suspected in this region due primary response and first-aid providers to previous displacement of the middle through education and training sessions turbinate. alongside the device manufacturers about The CSF rhinorrhea ceased over the following two days, a total of seven days after the onset. Figure 1: The patient is positioned upright but comfortable (approximately 45 degrees head ele- A follow-up CT-myelogram demonstrated vation). Physician is wearing personal protective no further leak and the patient remained equipment to avoid blood spray while performing symptom free at a clinic follow-up four anterior rhinoscopy and planning insertion of months later. nasal packing device. Discussion Nasal packing systems are more commonly placed by first aid providers, including paramedics and emergency medical officers in emergency settings, rather than ENT specialists.1,2 While they can be a useful adjunct in the management of epistaxis, they are not without their risks; the most notable complication is intense pain at the time of insertion.3 Watery rhinorrhea is also commonly reported, but confirmed CSF rhinorrhea is limited to case studies and series.1,2,4 Trauma to the skull base can occur by inserting the device in the wrong direction and displacing the middle turbinate that is attached to the skull base. Manufacturer instructions are clear. Devices should always be inserted horizontally in the direction of the floor of the nose, and insertion without adequate training is not recommended.6 Figure 1 demonstrates the technique of inserting the Rapid Rhino device. The minimum equipment needed is a headlight, personal protective equipment and saline

the appropriate use of these devices in bleeding site is not possible or fails, and addition to the initial management of if temporising measures such as topical epistaxis. adrenaline on gauze or cotton wool are Specific advice is being given regarding unsuccessful, a relatively firm nasal pack early involvement of ENT services to can be placed, but the correct technique is prevent unnecessary insertion, or if essential, as these cases demonstrate. insertion is necessary, to ensure they Our department recommends discussion are inserted in the appropriate direction with local ENT services prior to nasal with the appropriate amount of force and packing device insertions in the absence of inflation. Of course, if treatment of the life-threatening haemorrhage.

Competing interests: Nil. Author information: Alice Flavell-Birch: Resident Medical Officer, Department of Ear, Nose, Throat, Head & Neck surgery, Wellington Hospital, Capital & Coast District Health Board. Campbell Baguley: Consultant ENT Surgeon, Department of Ear, Nose, Throat, Head & Neck surgery, Wellington Hospital, Capital & Coast District Health Board. Alice Stringer: Consultant ENT Surgeon, Department of Ear, Nose, Throat, Head & Neck surgery, Wellington Hospital, Capital & Coast District Health Board. Corresponding author: Alice Flavell-Birch, Department of Ear, Nose, Throat, Head & Neck surgery, Wellington Hospital, Capital & Coast District Health Board, Riddiford Street, Newtown, Wellington, New Zealand. [email protected] URL: www.nzma.org.nz/journal-articles/cerebrospinal-fluid-rhinorrhoea-following-nasal-packing-for-epistaxis-case-series-from-a-single-centre

REFERENCES 1. Killick N, Malik V, Packing for Epistaxis. Device. RAPID RHINO Nirmal Kumar B. Nasal J Craniofac Surg. 2019 Arthrocare Product Usage packing for epistaxis: an Nov-Dec;30(8):2536-2538. Instructions For Epistaxis evidence-based review. Br 4. Edkins O, Nyamarebvu CT, Events. 2020. [online] Avail- J Hosp Med (Lond). 2014 Lubbe D. Cerebrospinal able at: . Hadi U. Changing Trends case report. J Laryngol 7. Skedros DG, Cass SP, in the Management of Otol. 2012 Apr;126(4):421-3. Hirsch BE, Kelly RH. Epistaxis. Int J Otolaryn- 5. Hollis GJ. Massive pneu- Beta-2 transferrin assay gol. 2015;2015:263987. mocephalus following in clinical management 3. Sireci F, Immordino V, Merocel nasal tamponade of cerebral spinal fluid Galletti F, Galletti B, Cimino for epistaxis. Acad Emerg and perilymphatic fluid G, Sbacchi C. Cerebrospinal Med. 2000 Sep;7(9):1073-4. leaks. J Otolaryngol. 1993 Oct;22(5):341-4. Fluid Leak Following Nasal 6. Rapid Rhino Epistaxis

Letter to the editor Katherine Rich

ear Professor Frizelle, Part of our concerns related to her Professor Connor’s editorial piece assertion that there was a conflict of Dof 26 March 2021 headed Why we interest with my appointment to the Health should be interested in the Whale Oil defa- Promotion Agency, which Professor Connor mation case does not fairly and accurately now repeats in her editorial. Professor present to your readers the reality of the Connor is entitled to hold a personal opinion defamation case as it relates to the New regarding what constitutes a conflict of Zealand Food & Grocery Council (NZFGC), interest, but in terms of the official gover- and myself as its chief executive. nance and public service definitions, there was none. What she again fails to mention Despite Professor Connor’s insinuations is that both the State Services Commission to the contrary, NZFGC did not pay anyone and the Office of the Auditor-General found to write any stories on its behalf on Whale there was no basis to the conflict of interest Oil or any other publication. Nor was NZFGC allegation that had been made by professors otherwise involved in any such stories. This Connor and Sellman at the time. was made abundantly clear to the three plaintiffs throughout the court process. The Professor Connor asserts that the identity claim of two of the plaintiffs, professors of those behind the posts in issue in the case Sellman and Swinburn, was settled on this had been revealed in Nicky Hager’s 2014 basis as far as NZFGC and myself were book Dirty Politics. Again what she fails to concerned. mention is that Professor Swinburn and Mr Bradbrook were not named or otherwise With regard to the other plaintiff, referred to in the book. The assertion that Mr Bradbrook, Professor Connor omits the NZFGC was behind the attacks on the important information regarding his part plaintiffs in the case is not correct. in the case against NZFGC and myself. It was made clear numerous times prior to A final comment. NZFGC is an industry and during the proceedings that we had association for all food, beverage and never heard of him, or his organisation Te grocery suppliers to supermarkets and Reo Marama, before the case. The plain- as such its membership is much greater tiffs accepted this and after some years Mr than that portrayed by Professor Connor. Bradbrook discontinued his claim. The Court While its members do include multi-na- ordered that he make a payment to NZFGC tionals, NZFGC is by far made up of small for significant Court costs. and medium local businesses, many of whom provide to supermarkets and other Professor Connor refers to so-called threats outlets the sort of foods public health to researchers “for publishing an account professionals advocate for. NZFGC is also of the scandal as a case study”. What is not transparent about the submissions it makes said is that Professor Connor was one of the to Parliament. They are readily available co-authors of the paper and a close colleague online. There are no hidden agendas or of Professor Sellman. The paper contained strategies to stifle public health initiatives or some errors of fact and in our view was not debate. Rather the focus is on working with a fair and accurate report of the proceedings. health and Government officials to achieve Rather it was seriously misleading in realistic and achievable goals through presenting only one side of the story and discussion and open debate involving all continued to attack my appointment to the who have a stake in the food and beverage Health Promotion Agency. We believed the industry. paper failed the test of academic objectivity and accuracy. Hence our letter. As a result of Yours sincerely, points made to the editor, Professor Connor’s Katherine Rich, paper was taken down and amended before Chief Executive, being republished. New Zealand Food & Grocery Council

Competing interests: The New Zealand Food & Grocery Council is an industry association and incorporated society for food and grocery manufacturers that supply supermarkets and other grocery-related channels. As CEO, Katherine Rich has no competing interests or conflicts in relation to the letter. Author information: Katherine Rich: Chief Executive, New Zealand Food & Grocery Council. Corresponding author: Katherine Rich, Chief Executive, New Zealand Food & Grocery Council [email protected] URL: www.nzma.org.nz/journal-articles/letter-to-the-editor

Letter to the editor: response to Katherine Rich Jennie Connor

he purpose of my recent editorial1 Second, the assertion that “both the State was to inform New Zealand Medical Services Commission and the Office of the TJournal readers about the conclu- Auditor-General found there was no basis sion of the recent Whale Oil defamation to the conflict of interest allegation that case and its connection to some broader had been made by professors Connor and issues for the health of New Zealanders. Sellman at the time” is incorrect on at least In common with many other parts of the two counts. The request for investigation world, the size of the public health prob- was not made by professors Connor and lems caused by production, promotion and Sellman, but by a group of 34 senior popu- sale of unhealthy commodities (eg, tobacco, lation health researchers and practitioners, unhealthy food, alcohol, environmental 33 of whom had first written to then Prime contaminants, oil) is staggering. I wanted to Minister John Key about their concerns, as remind health professionals that they are reported in the media at the time.4 More in a strong position to advocate for better importantly, the State Services Commission policy to reduce the harm, and to support (SSC) did not find there was “no basis” to the their colleagues who come under attack concerns raised about conflict of interest. while doing this work. The SSC did not conduct any investigation in Ms Rich’s response to the editorial2 has response to the request, because it did not pointed to things I didn’t include in it. I have authority to do so. didn’t give a comprehensive account of the The SSC advised that the Office of the Audi- trial, I didn’t claim to have written the case tor-General (OAG) had investigated conflicts study paper, I didn’t explain that two of of interest involving Crown entities several the plaintiffs were not mentioned in Dirty times in the past, so the group of health Politics, and so on. However, details of the professionals wrote to the Auditor-General. trial have been reported in a range of other The OAG response was more discursive, media including the articles referenced in but included this: “We have considered my editorial, and readers who are interested the issues raised, obtained and reviewed in Ms Rich’s own account can find a full and relevant documentation, and spoken with sympathetic interview reported in the New the chair and chief executive of the HPA. We Zealand Listener.3 have not found it necessary to interview Ms I would like to briefly address the factual Rich. We are satisfied there are no matters errors inadvertently included in Ms Rich’s we need to investigate further.” Thus, there letter. was no substantial investigation carried out and no judgment that the concerns First, the “case study” paper published were baseless, but the OAG letter (which in 2018 and subject to vigorous legal chal- is available online5) did make it clear that lenge from the New Zealand Food & Grocery New Zealand law provides few safeguards Council has been permanently withdrawn, to protect policymaking from commercial rather than “republished”. conflicts of interest.

Competing interests: Nil. Author informaton: Professor Jennie Connor: Department of Preventive and Social Medicine, University of Otago. Corresponding author: Professor Jennie Connor, Department of Preventive and Social Medicine, University of Otago [email protected] URL: www.nzma.org.nz/journal-articles/letter-to-the-editor-response-to-katherine-rich

REFERENCES 1. Connor J. Why we should Sugar & Spite. New ton.scoop.co.nz/?p=70745 be interested in the Whale Zealand Listener 5. Child, E. Health promo- Oil defamation case. New 2021;275(4118):22-6 tion Agency - Katherine Zealand Medical Journal 4. Rotherham F. Katherine Rich - possible conflicts of 2021;134(1532):11-3 Rich stands firm against interest. Office of the Audi- 2. Rich K. Letter to the editor. call to resign after alle- tor General, 14 May 2015. New Zealand Medical Jour- gations of conflict. Scoop Available at: https://oag. nal 2021;134(1534): 146-147 Wellington, 11 Sept 2014. parliament.nz/media/2015/ 3. Clifton J. Spin Scandal: Available at: http://welling- health-promotion-agency

Dr William Alexander Fraser 15 July 1924–24 November 2020

illiam Alexander Fraser (known as in association with Bob Gudex. A part-time Bill) was born in Hamilton on 15 position in O&G at Waikato Hospital WJuly 1924, the son of Dr T.C. (Tam) followed. The Gudex & Fraser team (“Bill & Madge Alexander Fraser. He was educated & Bob”) was a powerful unit. Both were at Southwell School, and at King’s College, highly energetic and productive, and inno- where he was a prefect and member of the vative surgeons, as well as attracting a First XV. He went to Otago medical school. large patient load. Both were popular prac- Despite his later lean build, he played as a titioners, and keen to teach and encourage prop in the university rugby team, gained their juniors. his blue, and represented South Island Uni- In addition to clinical work, Bill was versities XV. He graduated MB ChB in 1947, involved with the local O&G Society, a taking the Stanley Batchelor Memorial prize member of the Waikato Area Health Board for Clinical Surgery. service development committee, and a His junior doctor years were spent in member of a Medical Advisory Committee Auckland. Initially planning to specialise in formed to assist and advise the Medical Medicine, he was called back to Hamilton Superintendent of Waikato Hospital. He to help out in his father’s practice because was a member of the Maternity Benefits of the latter’s illness. Initially in general Tribunal. practice, with a high obstetric caseload, he He became FRCOG in 1977 and was a took his Diploma on Obstetrics in Auckland. Foundation Fellow of the New Zealand He was encouraged to specialise in O&G College of Obstetrics and Gynaecology (later with the return of his old friend Bob Gudex Royal) when it was formed in 1982. He was to Hamilton. He went first to National head of the O& G service at Waikato Hospital Women’s Hospital, and obtained his MRCOG from 1980 until 1988. A particular expertise in 1961. After some time in England he was developed in the surgery of urinary returned to his family and joined practice incontinence.

He had wide interests apart from medicine, In 1953 he and Judith Allan were married in particular horticulture, being prominent in Ashburton. They had four children. in camellia and rhododendron organisa- Alastair (a Taupō GP) Jane, Kate (both tions, and later cattle breeding. He was nurses), and David (chemical engineer, later knowledgeable in many subjects, from bees, finance). There are eleven grandchildren, to electronics to cars. He had a series of and two great grandchildren. A private exotic cars long before they became readily ceremony was held to commemorate his available in New Zealand. He was a hero, and long life after he died on 24 November 2020 role model to a generation of young doctors. aged 96.

Author information: Alastair Haslam. URL: www.nzma.org.nz/journal-articles/dr-william-alexander-fraser

Treatment of Mental and Nervous Disorders 1921

e have received a communication ings instead of one large one, but practically from a contributor, Dr. Driver, this is not possible on the score of expense Wwhich we hope later to be able to and adequate supervision. Such refuges as publish, but his purpose may be fulfilled for these are greatly to be commended and lead the present by calling the attention of our to excellent results, but we agree with our readers in the editorial columns to the im- correspondent that more is required, but not portance and difficulty of the task of provid- required of the Mental Hospital Department. ing adequate facilities for the treatment of These branches of the mental hospitals do incipient cases of mental disease. Dr. Driver not, and cannot remove the stigma of cer- thinks that our mental hospitals do not tification of insanity, and what that entails, provide for efficient classification generally, for they are intended mainly for patients and, in particular, not enough provision is who must be certified, because it is unsafe in made for preventive treatment in cases of these cases to allow liberty, but patients can neurasthenia and psychasthenia which may be admitted of their own free will without develop into definite insanity. He thinks that certification, and some are voluntary pa- a borderland case incarcerated in the main tients as the result of kindly persuasion. building of a mental hosptal is more likely But there are many patients with neuras- than not to progress in the wrong direction. thenia, psychasthenia, and various phobias, There is much to be said for this contention, never more than at the present time when but happily of late there has been a marked the whole world suffers in some degree improvement in the mental hospital build- from shellshock, who cannot attend regu- ings in New Zealand, and in the direction larly to their work, are unhappy at home of facilities for the treatment of patients and not suitable for admission to a mental whose minds are not very far astray and hospital, a general hospital, a private who give good hopes of recovery. We have hospital, or even to one of these special closely examined at Seacliff and at Porirua institutions in connection with a mental buildings set apart for this purpose. These hospital to which we have alluded. These buildings, it is true, are within the grounds patients are an economic loss, and the of the mental hospitals, but they are apart despair of the physician, particularly if the from the main buildings, and we understand physician has learned nothing and forgotten that the arrangements are similar at Sunny- nothing since the days before the Great side and at Auckland. Let us refer briefly to War. No honest doctor can now attempt this branch of the work at Porirua, where to cure such diseases with nerve tonics the building itself is bright and cheerful and sedatives, and visit the patient once or and the surroundings equally pleasant. The twice a week, because there is no medicine patients suffer no more restraint than is im- that will cure such diseases. Let us take the plied in being prohibited from mixing with treatment of these case out of the hands the outside world except through visits from of the Mental Hospital Department so that the friends and relations. They are provided we may take the patient’s mind away from with recreation and amusement, billiards insanity, and there will be fewer admissions and music and reading, and the lighteR to asylums. We think that a special ward in cases have private bedrooms such as a trav- a general hospital is not suitable for cases of eller has in a good hotel, and the rest of the psychasthenia, but special institutions are patients have dormitories not unlike small required. We must follow what has been wards in a general hospital. Theoretically it found most effective in the treatment of the would be well to have several smaller build- large number of soldiers whose nerves were

shattered in the war. Fortunately, there is suggestion, physiotherapy, recreational and an institution at Hanmer specially equipped vocational treatment, analysis and whatever for the treatment of the psychasthenics. else may be judiciously added to the advan- Probably soon the Hanmer Hospital can be tages of a country life. With provision of taken over by the hospital division of the this kind, in addition to what already exists, Health Department. An increasing number we think we shall be doing our duty in the of civilian male patients can be received, light of our present knowledge of an unfor- and it is a matter of urgency to make similar tunate class of patients for whom we have provision for female patients. The medical been able to do little in the past. Often the profession should encourage this form silver cord is loosed before the golden bowl of treatment by specialists who employ is broken.

URL: www.nzma.org.nz/journal-articles/treatment-of-mental-and-nervous-disorders

NZMJ 30 April 2021, Vol 134 No 1534 ISSN 1175-8716 © NZMA 153 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]