Journal of the New Zealand Medical Association Vol 131 | No 1473 | 13 April 2018 Marketing unhealthy commodities on social media

Volume, nature and potential impact of advertisements on Facebook and YouTube by food brands popular in New Zealand Regulatory chills: tobacco industry legal threats and the politics of tobacco standardized packaging in New Zealand

Poisoning due to tutin If only Teina Pora The Otago Medical School Anatomy in honey—a report of an had a MedicAlert Museum Collection: Taonga for outbreak in New Zealand bracelet learning in the 21st Century Publication Information published by the New Zealand Medical Association

NZMA Chair To contribute to the NZMJ, fi rst read: Dr Kate Baddock www.nzma.org.nz/journal/contribute

NZMJ Editor Other enquiries to: Professor Frank Frizelle NZMA PO Box 156 NZMA Communications Manager The Terrace Sharon Cuzens Wellington 6140 Phone: (04) 472 4741 NZMJ Production Editor Rory Stewart © NZMA 2018

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 10 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 2018 New Zealand subscription rates Overseas subscription rates Individuals* $306 Individual $426 Institutions $530 Institutions $571 Individual article $25 Individual article $25

*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 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 2 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]

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 3 www.nzma.org.nz/journal CONTENTS

EDITORIALS VIEWPOINT 7 72 Marketing unhealthy commodities on The Otago Medical School Anatomy social media Museum Collection: Taonga for learning Antonia C Lyons, Timothy McCreanor in the 21st Century 11 Louisa JM Baillie, Christopher L Smith Tobacco targets—doomed to fail? CLINICAL CORRESPONDENCE Mike Daube 78 ARTICLES Hypercalcemia in a bodybuilder with cosmetic silicone injections 14 Majdi Hamadeh, Jawad Fares, Khalil Maatouk, Volume, nature and potential impact Mohamad Darwish of advertisements on Facebook and YouTube by food brands popular in New 82 Zealand Purulent facial lesion Stefanie Vandevijvere, Charlotte Aitken, Pallav Patni, Vinod Bhandari, Mahak Bhandari, Boyd Swinburn Swadhin Raghuwanshi 25 RESEARCH LETTERS Regulatory chills: tobacco industry legal threats and the politics of tobacco 85 If only Teina Pora had a MedicAlert standardized packaging in New Zealand Eric Crosbie, George Thomson bracelet Anita Gibbs 42 Pertussis vaccination uptake in 88 Permanent pacemaker implantation pregnancy: lessons to be learned from a“ er cardiac surgery: rates, predictors an integrated healthcare approach Emma J Deverall, Benjamin Gilmore, Sam Illing, and a novel risk score Tom Kai Ming Wang, Diego Arroyo, Roshini Peiris-John Andrew Martin, Alastair McGeorge, 48 Michael Gillham Diagnosis of abdominal tuberculosis in LETTER Christchurch New Zealand: a case series James MM Bevin, Simon C Dalton, 92 Chris J Wakeman, Will RG Perry Gnome medicine: what does genomic medicine mean to our patients and us? 53 Sara Filoche, Kevin Dew, Anthony Dowell Metronidazole stewardship initiative at Christchurch hospitals—achievable with METHUSELAH immediate beneŠ ts Sharon J Gardiner, Sarah CL Metcalf, 94 Paul KL Chin, Matthew P Doogue, Aspirin or rivaroxaban for VTE Simon C Dalton, Stephen T Chambers prophylaxis a“ er hip or knee 59 arthroplasty? Poisoning due to tutin in honey—a report 100 YEARS AGO of an outbreak in New Zealand Michael Beasley, Dell Hood, Philippa Anderson, 95 John Reeve, Robin J Slaughter Innocent Intra-Orbital Growth By Edgar Whitaker, M.R.C.S., Ophthalmic Surgeon, Palmerston North Hospital

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 4 www.nzma.org.nz/journal SUMMARIES

Volume, nature and potential impact of advertisements on Facebook and YouTube by food brands popular in New Zealand Stefanie Vandevijvere, Charlotte Aitken, Boyd Swinburn Posts on Facebook pages of 45 popular packaged food, beverage and fast food company brands over two months and YouTube channels of 15 popular brands over two years were analysed for nutritional quality and use of activities, promotional strategies (eg, cartoons) and premium offers (eg, competitions). Social media is an important medium for food marketers in New Zealand and promotional strategies and premium offers are frequently used. Regulatory chills: tobacco industry legal threats and the politics of tobacco standardized packaging in New Zealand Eric Crosbie, George Thomson Political delays slow the diffusion of best practices. In this case the delays due to tobacco industry legal threats delayed standardised packaging, slowing smoking cessation and the reduction of initiation. They also delayed the reduction of government health expenditures and tobacco industry profi ts. Other countries introducing or implementing similar policies should learn from these experiences and take steps to proactively avoid unnecessary political delays that have a profound impact on public health. Pertussis vaccination uptake in pregnancy: lessons to be learned from an integrated healthcare approach Emma J Deverall, Benjamin Gilmore, Sam Illing, Roshini Peiris-John Pertussis (or Whooping cough) is a serious, but vaccine preventable disease. Infants are the most at risk of complication and death, and are best protected by giving a free vaccine to their mother during pregnancy. An audit in Lakes District Health Board found the immuni- sation rate was much higher in Taupo (76%) than in Rotorua (45%). This paper explores the differences in the two regions. Coordinated efforts between the Primary Care Organisation, Midwives and the District Health Board in Taupo helps to bring vaccination to women in their community and improve vaccination rates. Diagnosis of abdominal tuberculosis in Christchurch New Zealand: a case series James MM Bevin, Simon C Dalton, Chris J Wakeman, Will RG Perry This is a series of the 20 most recent cases of abdominal tuberculosis in Christchurch. We investigated as to how these patients presented and what investigations were used to diagnose these patients with abdominal tuberculosis. We found that patients with abdominal tuber- culosis have very non-specifi c symptoms and as such the method of investigation you use is crucial. The literature suggests laparoscopy followed by a positive microbiological result has the best diagnostic utility—we found this was the most common and effective method to diagnose abdominal tuberculosis in Christchurch as well.

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 5 www.nzma.org.nz/journal SUMMARIES

Metronidazole stewardship initiative at Christchurch hospitals—achievable with immediate beneŠ ts Sharon J Gardiner, Sarah CL Metcalf, Paul KL Chin, Matthew P Doogue, Simon C Dalton, Stephen T Chambers Metronidazole is a commonly used antibiotic. In hospitals, it is traditionally dosed three times daily, and often intravenously (IV). In October 2015, a campaign was started in Christchurch public hospitals to change to less frequent dosing (twice daily), and to prioritise the oral route for dosing. The initiative was successful with a 43% decrease in metronidazole IV use and savings of around $100,000 annually. Benefi ts include avoidance of bloodstream infections associated with IV access, and increased nursing ‘time to care’. Poisoning due to tutin in honey—a report of an outbreak in New Zealand Michael Beasley, Dell Hood, Philippa Anderson, John Reeve, Robin J Slaughter In autumn 2008, an outbreak of toxic honey poisoning was identifi ed in Thames, Waikato, New Zealand. This study aimed to investigate these cases of honey poisoning and determine which toxin was involved and describe the effects of the. After testing the honey, the causative toxin was identifi ed as tutin, which comes from the New Zealand native plant tutu. Twen- ty-two people were potentially poisoned with 11 of these cases being confi rmed as ingesting toxic honey. There were eight cases of seizures in poisoned patients. Food safety standards have since been enhanced to minimise the risk of toxic honey. The Otago Medical School Anatomy Museum Collection: Taonga for learning in the 21st Century Louisa JM Baillie, Christopher L Smith The WD Trotter Anatomy Museum has a large and diverse collection of teaching models and specimens. The Louis Auzoux paper mâché ear model has been in high demand for hands-on teaching since the 1880s. It features clastic parts, large scale (10:1) and clear, intricate anatomical details. Reproduction of this valuable model was urgently required as it was becoming worn and damaged from handling, and needed to be ‘retired’. A replica model has been successfully produced in the Anatomy Workshop using digital scanning, 3D printing and skilled modelling and hand painting effects.

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 6 www.nzma.org.nz/journal EDITORIAL

Marketing unhealthy commodities on social media Antonia C Lyons, Timothy McCreanor

he marketing of commodities on social profi t-streams from selling such data to media sites is increasing as these com- interested businesses.6 Facebook has over mercial platforms adapt to evolving 2.1 billion users, over half of whom access T 7 mobile technologies. New Zealand market- the platform on their phones daily, where ing regulations are frequently ineffective in every one of their keystrokes is recorded this environment, as they either do not ap- as data that is owned by the corporation. ply to foreign-domiciled sites or are evaded Social media marketing is also effective as corporations make use of ‘under-the-ra- because algorithms can identify and recruit dar’ techniques such as user-generated con- new generations of consumers, effectively tent, mimicry of online ‘friendship’ practices grooming children to become consumers as and networked sharing. Current ‘controls’ they gather disposable income or cross key rely on industry self-regulation. This means age-barriers such as those around alcohol that companies are not publicly accountable and tobacco. Given such platform affor- around their social media marketing activi- dances, it is not surprising that a recent ties, a business-world situation that parallels review concluded that digital marketing the political scandals involving social media has signifi cant detrimental effects on “the currently occurring globally.1 intended use and actual consumption of 8 The business models of social media plat- unhealthy commodities”. forms are structured around exploiting the In this volume, Vandevijvere, Aitken and warehoused personal data of the masses Swinburn9 provide much-needed insights from which algorithms can identify highly into the marketing of common unhealthy specifi c group and individual profi les, (highly-processed, energy dense and allowing micro-targeted advertising, often nutrient poor) food and beverage products via sophisticated psychological profi ling on Facebook and YouTube in New Zealand. tools.2 These algorithms draw on users’ Their study tracked Facebook activity values and meanings, embedding brands over two months, and YouTube presence within people’s everyday practices and iden- over two years, of snack food, beverage tities,3 catering for specifi c national contexts and fast food companies. It examined the and cultural settings.4,5 This type of highly nature and extent of promotional activities targeted marketing encourages peer-to-peer and their potential reach, estimating that transmission of messages and content, 10% of adolescents in New Zealand could enabling electronic ‘word-of-mouth’ viral be exposed to these infl uences. Diverse marketing. Such sharing of ‘information’ marketing techniques were employed, with blurs the boundaries between commercial companies encouraging users to engage with messages and private activity, making it posts and online content to ensure products diffi cult for users to identify marketing were seen by users’ friends and online content that often morphs as it travels networks. In this way, online brands amplify through the network. the reach and relevance of marketing The massive audience reach of online messages, appropriating users’ labour with 10 networks means these approaches are little incentive or reward. highly effective and cheap, in part because According to a recent OECD report, they work with the free ‘immaterial labour’ obesity rates in New Zealand are third of their numerous users, creating vast worst in the world,11 with one in three

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 7 www.nzma.org.nz/journal EDITORIAL

adults and over 12% of children classifi ed Extending existing regulatory codes to as obese.12 Overconsumption of unhealthy social media is not going to be effective as products is related to chronic non-commu- they do not take into account the particular nicable diseases and premature mortality.13 characteristics of the social media envi- Evidence demonstrates that marketing ronment,18 including its interactive, such products is related to adverse health immersive and personal nature (aspects outcomes,13 and conversely, regulating which appeal to children and young people food advertising to children is an effective particularly). Our exploratory research strategy to tackle eating behaviours closely with young adults engaging with alcohol linked to obesity.14 Promoting unhealthy marketing on Facebook found that the products on social media is an effective alcohol industry embeds their marketing marketing strategy, but is ethically prob- activities within online friendship activities. lematic, particularly when it is targeted at In this way the marketing became obscured, children and young people who are high but also normalised, effectively hidden in users of social media platforms. plain sight as part of the routine fl ow of 19 New Zealand children are over-exposed material on Facebook. We concluded that to unhealthy food marketing in everyday “social media marketing of alcohol is likely settings. A recent study of 11–13 year-olds to encourage consumption through new using wearable cameras found that they forms of promotion (particularly through were exposed to such marketing on average alcohol venues online), the exploitation more than 27 times per day (twice the of networked peer group friendship prac- rate than for healthy foods) and that most tices and [favourable] perceptions of the products they saw were sugary drinks, behaviour of others”. In this platform, fast food, confectionary and snack foods.15 algorithms target those users who are more The convergence with targeted, interactive likely to consume specifi c products such as marketing (often keyed to human neuro- alcohol or unhealthy foods; in other words, 3 logical processes) within social media those who may be most vulnerable. platforms means that children’s environ- Social media platforms are now signif- ments are potentially inundated with highly icant commercial institutions and as such engaging promotions of a wide range of should attend to the ethical responsibil- health-demoting products. The ubiquity of ities of constructive citizenship,3 including unhealthy food marketing within children’s accountability for brand activity on their environments, both offl ine and online, platforms.16 The vast databases that social normalises brands and has detrimental media corporations have assembled could health effects.8 Reducing children’s exposure also be made available for public scrutiny to the marketing of unhealthy foods is and research for the public good,17 showing essential and must include consideration of for instance the impact of marketing social media. activities in terms of overconsumption of Public health recommendations to protect unhealthy foods and other commodities. young people within social media envi- The infrastructure itself—usurped as it is ronments have been largely ignored by from the digital commons—could be defi ned governments and corporations.16,17 The and regulated, including making all use policy implications of marketing unhealthy of personal data transparent so users can commodities online have been viewed as too clearly see the fl ow of content, where their diffi cult. For example, the report of the New data goes, who receives it and how it is 3 Zealand Ministerial Forum on Alcohol Adver- subsequently used. tising and Sponsorship 2014 began: “As a Counter-marketing can also expose the forum we believe protecting the young from motives and tactics of companies, and alcohol-related harm is paramount” (p1) but highlight the ways in which users are subsequently placed social media marketing manipulated by industry20 and social media outside the remit of the review. We need to companies themselves. Social movements grapple with marketing on social media if may help to demonstrate the invisible (and we want to safeguard our young people from at times underhanded) corporate strategies the promotion of unhealthy products.8 and forces that are involved in encouraging

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 8 www.nzma.org.nz/journal EDITORIAL

consumption of products that lead to poorer tens of thousands of apps operative on the population health outcomes. Globally, platform.22 In contrast, very little attention or research suggests that companies vary their critique is directed towards the entrenched online marketing techniques by country, marketing on Facebook (and other social marketing healthier products in wealthier media platforms) that use the same tactics to countries, while showcasing their philan- build individuated consumer cultures, while thropic activities in less wealthy countries.21 making huge profi ts for corporations. The recent disclosures around Cambridge Social media are built on commercial Analytica and the unauthorised use of imperatives that have nothing to do with Facebook users’ data to infl uence demo- public health. The marketing of unhealthy cratic processes has brought the use of foods, alcohol, tobacco, gambling and other social media data into the public spotlight.1 products is rife on social media platforms Much attention is being given to the political and has been linked to poorer health and dimensions of what Steve Bannon (previous wellbeing outcomes for youth and adults. executive chairman of the “alt-right” It is daunting but crucial to document news network Breitbart) has referred to what is occurring in this ephemeral and as the cultural warfare necessary to drive evolving persuasion space, to gather particular political agendas. Whistleblowers evidence that shows how marketing creates have discussed how Facebook routinely health-demoting perceptions, attitudes harvested data for political purposes, data and behaviours, how they can be resisted that were “private and personally identi- and what public policy interventions are fi able”, belonging to ‘friends’ of users of indicated.

Competing interests: Nil. Author information: Antonia C Lyons, Professor of Health Psychology, School of Health, Victoria University of Wellington, Wellington; Timothy McCreanor, Professor of Public Health, SHORE and Whariki Research Centre, Massey University, Auckland. Corresponding author: Professor Antonia Lyons, School of Health, Victoria University of Wellington, PO Box 600, Wellington 6140. [email protected] URL: http://www.nzma.org.nz/journal/read-the-journal/all-issues/2010-2019/2018/vol-131-no-1473- 13-april-2018/7537

REFERENCES: 1. Cadwalladr, C. The 2. Kosinski M, Stillwell D, cultures in a digital Cambridge Analytica Files. Graepel T. Private traits world: Alcohol, social ‘I made Steve Bannon’s and attributes are predict- media and cultures of psychological warfare able from digital records intoxication Abingdon UK: tool’: meet the data war of human behavior. Proc Routledge; 2017; 115–131. whistleblower. [Internet]. Natl Acad Sci U S A. 2013 4. Goodwin I, Griffi n C, Lyons London, England: The Apr 9; 110(15):5802–5. doi: A, et al. Precarious popu- Guardian; 2018 [updated 10.1073/pnas.1218772110. larity: Facebook drinking 2018 18 March; cited 2018 Epub 2013 Mar 11. photos, the attention econ- 18 March]. Available from: 3. Carah, N. Alcohol corpo- omy, and the regime of the http://www.theguardian. rations and marketing in branded self. Soc Med Soci- com/news/2018/mar/17/ social media. In Lyons AC, ety. 2016 Feb 3; 2(1). doi: data-war-whistleblower- McCreanor T, Goodwin 10.1177/2056305116628889. christopher-wylie-faceook- I, Moewaka Barnes H, 5. Gupta H, Lam T, Petti- nix-bannon-trump editors. Youth drinking grew S, Tait RJ. Alcohol

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 9 www.nzma.org.nz/journal EDITORIAL

marketing on YouTube: Health Survey. Wellington editors. Youth drinking exploratory analysis of NZ: Ministry of Health; cultures in a digital content adaptation to 2017. Available from: world: Alcohol, social enhance user engage- http://www.health.govt.nz/ media and cultures of ment in different publication/annual-update- intoxication Abingdon UK: national contexts. BMC key-results-2016-17-new- Routledge; 2017; 230–241. Public Health. 2018 Jan zealand-health-survey 18. Dunlop S, Freeman B, Jones 16; 18(1):141. doi: 10.1186/ 13. Cairns G, Angus K, SC. Marketing to youth s12889-018-5035-3. Hastings G, Caraher M. in the digital age: The http://bmcpublichealth. Systematic reviews of the promotion of unhealthy biomedcentral.com/ evidence on the nature, products and health articles/10.1186/s12889- extent and effects of food promoting behaviours 018-5035-3 marketing to children. A on social media. Med 6. Fuchs C. Social media: retrospective summary. Commun. 2016; 4(3). doi: A critical introduction. Appetite. 2013 Mar; 10.17645/mac.v4i3.522. London; Sage; 2017. 400 p. 62:209–15. doi: 10.1016/j. 19. Niland P, McCreanor T, 7. Zephoria Digital Marketing. appet.2012.04.017. Lyons AC, Griffi n C. Alcohol The Top 20 Valuable Face- Epub 2012 May 2. marketing on social media: book Statistics – Updated 14. Cecchini M, Sassi F, Lauer Young adults engage with April 2018 [Internet]. JA, et al. Tackling of alcohol marketing on Face- USA; 2018 [updated April unhealthy diets, physical book. Addiction Res Theory. 2018; cited 2018 April 5]. inactivity, and obesity: 2017 Jul 4; 25(4):273–84. Available from: http:// health effects and cost-ef- doi: 10.1080/16066359.2016. zephoria.com/top-15-valu- fectiveness. Lancet. 2010 1245293. able-facebook-statistics/ Nov 20; 376(9754):1775–84. 20. Palmedo PC, Dorfman L, Retrieved 5 April. doi: 10.1016/S0140- Garza S, et al. Counter- 8. Buchanan L, Kelly B, Yeat- 6736(10)61514-0. marketing alcohol and man H, Kariippanon K. The Epub 2010 Nov 10. unhealthy food: An effec- effects of digital marketing 15. Signal LN, Stanley J, tive strategy for preventing of unhealthy commodities Smith M, et al. Children’s noncommunicable diseas- on young people: a system- everyday exposure to food es? Lessons from tobacco. atic review. Nutrients. marketing: an objective Annu Rev Public Health. 2018 Jan 29; 10(2):148. doi: analysis using wearable 2017 Mar 20; 38:119–44. 10.3390/nu10020148. cameras. Int J Behav doi: 10.1146/annurev- 9. Vandevijvere S, Aitken Nutr Phys Act. 2017 Oct publhealth-031816-044303. C, Swinburn B. Volume, 8; 14(1):137. doi: 10.1186/ 21. Bragg MA, Eby M, nature and potential s12966-017-0570-3. Arshonsky J, et al. Compar- impact of advertisements 16. Mart S. New marketing, ison of online marketing on Facebook and YouTube new policy? Emerging techniques on food and by food brands popular debates over regulating beverage companies’ in New Zealand. N Z Med alcohol campaigns in social websites in six countries. J 2018; 131(1473):14–24. media. In Lyons AC, Global Health. 2017 Oct 10. Freeman B, Kelly B, McCreanor T, Goodwin 26; 13(1):79. doi: 10.1186/ Baur L, et al. Digital I, Moewaka Barnes H, s12992-017-0303-z. junk: food and beverage editors. Youth drinking 22. Lewis, P. ‘Utterly horrify- marketing on Facebook. cultures in a digital ing’: ex-Facebook insider Am J Public Health. 2014 world: Alcohol, social says covert data harvesting Dec; 104(12):e56–64. doi: media and cultures of was routine [Internet]. 10.2105/AJPH.2014.302167. intoxication Abingdon UK: London England: The Epub 2014 Oct 16. Routledge; 2017; 218–229. Guardian, 20 March 2018. 11. OECD. Obesity Update. 17. McCreanor T, Moewaka Available from: http:// France: OECD; 2017. 16 p. Barnes H, Lyons AC, www.theguardian.com/ Available from: http://www. Goodwin I. Digital alcohol news/2018/mar/20/face- oecd.org/els/health-systems/ marketing and the public book-data-cambridge-an- Obesity-Update-2017.pdf good: Industry, research alytica-sandy-parakilas and ethics. In Lyons AC, 12. Ministry of Health. Annual McCreanor T, Goodwin Update of Key Results I, Moewaka Barnes H, 2016/17: New Zealand

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 10 www.nzma.org.nz/journal EDITORIAL

Tobacco targets— doomed to fail? Mike Daube

here has been incontrovertible evi- quite deliberately, a health epidemic. And dence that smoking is lethal for nearly moreover, a health epidemic which imposes seven decades, since publication in vast negative health and other costs upon T 7 1950 of authoritative papers by Doll and Hill the state”. The tobacco industry, with a in the British Medical Journal and Wynder well-justifi ed reputation for lies and deceit, and Graham in the Journal of the Ameri- has from the outset done everything in its can Medical Association.1,2 Vast numbers of power to oppose and delay any action that papers and reports since then have left no might reduce smoking, while continuing doubt that globally and nationally this is to promote its products wherever possible, the largest preventable cause of death and often with a special focus on vulnerable and disease. Each year according to the World disadvantaged populations. Health Organization, more than seven mil- Millions of once-confi dential industry lion deaths around the world, and, accord- documents, complemented by many media ing to the Ministry of Health, 5,000 deaths in exposes, demonstrate beyond doubt the 3,4 New Zealand, are caused by smoking. ruthless, cynical and dishonest approach The word “preventable” is critical. We taken by this lethal industry in its efforts to know the extent of the problem, but we promote and sell a product it knows will kill have also known for several decades the key at least half of its regular consumers. approaches needed to prevent the deaths The introduction of standardised (or and disease caused by smoking—no single “plain”) packaging in Australia was achieved magic bullet, but a comprehensive approach despite the most ferocious industry oppo- comprising regulation, public education, sition this author has seen in 45 years cessation supports and continuing advocacy. working on tobacco—testament to the There has been progress, and New impacts feared by the tobacco companies Zealand has indeed at times been one of the from a measure that would reduce their leading countries in terms of implementing last legitimate form of direct promotion. some of the actions needed, but it has been Few measures have so well illustrated the all too slow. tobacco “Scream Test” that health advocates The New Zealand government’s 2011 goal often cite—the louder the tobacco industry of “a smokefree New Zealand by 2025”5 was screams, the more effective a measure is laudable, albeit optimistic, but on the basis likely to be. It was therefore not surprising of current trends in terms of both preva- that following implementation of stan- lence and governmental action it is doomed dardised packaging the Imperial Tobacco to fail. In order to meet the smokefree goal company gloomily concluded that “Australia by 2025, the government recognised that is the darkest market in the world”. daily smoking rates would need to be at In an important paper8 that will be a 10% by 2018. The national data from 2015 valuable addition to the international liter- show daily smoking rates at about 15%.6 ature on tobacco control processes, Crosbie Smokefree by 2025 would require a massive and Thomson have shown not only that decline, well beyond optimism. the companies had similar fears in New As we have known the extent of the Zealand, but also that they learned from problem for decades, so we have known the their defeat in Australia by focusing much vector—the international tobacco industry of their lobbying effort on legal threats that which, in the words of a recent UK High successfully intimidated governments into Court judgement, “facilitates and furthers, delaying introduction of an evidence-based

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 11 www.nzma.org.nz/journal EDITORIAL

measure recommended by health author- companies themselves or through associated ities as part of a comprehensive approach. organisations, “think tanks” and front groups. While standardised packaging is now Legal action and threats against being implemented in New Zealand (as it government, of the nature outlined by is elsewhere—by the end of 2018 it should Crosbie and Thomson (and even, as in the be either in place or on the way in close to UK and Australia, against organisations 20 countries), the process has been slow, such as universities and cancer charities), and the Australian experience shows that provide yet further evidence that tobacco tobacco companies will do everything companies will do whatever it takes to they can and spend as much as it takes further their interests and cause more to undermine and counter its impacts. preventable deaths. As in Australia, these Meantime, the New Zealand Government companies have no interest in the health or has signifi cantly reduced its already modest wellbeing of New Zealanders: they are based support for tobacco control advocacy, which in London and New York, and their only means that further much-needed comple- interest is the fi nancial health and wellbeing mentary action is far less likely to eventuate, of their shareholders. and there will inevitably be less public Future generations will wonder why, exposure of the industry’s lobbying and nearly 70 years after incontrovertible other promotional activities. evidence about its harms, smoking was still New Zealand is a signatory to the WHO the largest preventable cause of death and Framework Convention on Tobacco Control, disease, and tobacco companies were still which came into force in 2005 and is legally allowed to fl ourish. The answer lies partly binding in 180 countries.9 The Convention in the sheer power of the tobacco industry, and its Guidelines are clear and explicit that and partly in the inability of health groups “There is a fundamental and irreconcilable to match its resources and ruthlessness. confl ict between the tobacco industry’s But even more than this, responsibility interests and public health policy interests”, rests with governments that have failed to and Article 5.3 of the Convention requires act despite clear evidence about the action that “In setting and implementing their needed and absolute certainty that their public health policies with respect to tobacco inaction would result in tens of thousands of control, Parties shall act to protect these unnecessary deaths. policies from commercial and other vested “Smokefree New Zealand by 2025” may interests of the tobacco industry…”. Govern- have seemed a laudable aspiration in ments have a responsibility to ensure that all 2011, but the loss of momentum, further their agencies (not only health departments) tobacco industry lobbying, promotional and are aware of and committed to Article 5.3. distraction activities, a focus on approaches This is especially important at a time when that will not offend tobacco interests and the tobacco industry is more aggressive than reduced support for public health advocacy ever in seeking to prevent any actions by mean that it now looks completely unre- governments that might reduce smoking. alistic. That is a triumph for Big Tobacco, The industry has always devoted substantial an indictment of the government’s failure resources to lobbying and public relations to take prevention seriously, and a public activities; in recent years these efforts have health tragedy. been stepped up, whether through the

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 12 www.nzma.org.nz/journal EDITORIAL

Competing interests: Nil. Author information: Mike Daube, Professor of Health Policy, Curtin University, Australia. Corresponding author: Professor Mike Daube AO, Professor of Health Policy, Curtin University, Australia. [email protected] URL: http://www.nzma.org.nz/journal/read-the-journal/all-issues/2010-2019/2018/vol-131-no-1473- 13-april-2018/7538

REFERENCES: 1. Doll R, Hill Bradford A. www.health.govt.nz/ 7. British American Tobacco Smoking and carcinoma your-health/healthy-living/ & others v Secretary of of the lung: preliminary addictions/smoking/ State for Health [2016] report. British Medical health-effects-smoking EWHC 1169 (Admin). Journal, 1950; 2:739–748. (accessed 9 Apr 2018). 8. Crosbie E, Thomson 2. Wynder EL, Graham EA. 5. Ministry of Health Manatū G. Tobacco industry Tobacco smoking as a Hauora. Smokefree legal threats and the possible etiologic factor in 2025. 2018. http://www. politics of tobacco stan- bronchogenic carcinoma. health.govt.nz/our-work/ dardized packaging in Journal of the American preventative-health-well- New Zealand. N Z Med J Medical Association, ness/tobacco-control/ 2018; 131(1473):25–41. 1950; 143:329–336. smokefree-2025 9. World Health Organization 3. World Health Organization. (accessed 9 Apr 2018). Framework Convention on Tobacco Fact Sheet. 2018. 6. Tobacco Control Data Tobacco Control. The WHO http://www.who.int/medi- Repository. Unadjusted Framework Convention acentre/factsheets/fs339/ changes in the prevalence on Tobacco Control: an en/ (accessed 9 Apr 2018). of daily smokers over overview, 2015. http:// 4. Ministry of Health Manatū time. http://www.tcdata. www.who.int/fctc/ Hauora. Health effects org.nz/Health%20 WHO_FCTC_summary_ of smoking. 2017. https:// Survey/Health_02.aspx January2015_EN.pdf?ua=1 (accessed 9 Apr 2018). (accessed 9 Apr 2018).

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 13 www.nzma.org.nz/journal ARTICLE

Volume, nature and potential impact of advertisements on Facebook and YouTube by food brands popular in New Zealand Stefanie Vandevijvere, Charlotte Aitken, Boyd Swinburn

ABSTRACT AIM: To analyse extent, nature and potential impact of marketing by food and beverage brands popular in New Zealand on Facebook and YouTube. METHOD: Popular food and beverage brands in New Zealand were selected from Socialbakers. Posts on Facebook pages of 45 packaged food, beverage and fast food companies over two months and YouTube channels of 15 companies over two years were analysed for nutritional quality and use of activities, promotional strategies (eg, cartoons) and premium o– ers (eg, competitions). RESULTS: The 45 brands selected made 762 Facebook posts during October–November 2016. About 28% of posts were videos and 2/3 (63%) contained at least one occasional (ie, unhealthy) food. Promotional strategies were used in 41% of posts, with a famous sportsperson/team being the most frequently used. Premium o– ers were used in 34% of posts, with competitions being the most frequently used. It was estimated some posts could potentially reach 10% of New Zealand adolescents. The 15 food brands selected posted about 300 videos on their YouTube channels during 2015–2016. About 84% of videos contained food marketing and 77% of products marketed were occasional. Promotional strategies and premium o– ers were used in 61% and 24% of videos respectively, and the most common marketing techniques were the same as on Facebook. CONCLUSION: Social media is an important medium for food marketers in New Zealand and promotional strategies and premium o– ers are frequently used. Methodology needs to be developed to monitor actual exposure to such advertisements.

esults from the recent New Zealand A review undertaken of New Zealand Health Survey revealed that one in studies to date indicates that advertisements three New Zealand children are either through a wide range of media platforms R 1 7 overweight or obese, and one of the fac- are predominantly for unhealthy foods. tors known to infl uence the unhealthy food Previous research on the extent and nature and beverage choices related to childhood of unhealthy food and beverage adver- obesity is food and beverage marketing.2–4 tising in New Zealand, however, has mainly Nowadays there are countless platforms that focused on traditional media platforms like companies can use to target children and television and magazines.8–10 Traditional adolescents and consumers more broadly, media is still important for marketers as with social media sites like Facebook and a recent study on television advertising in YouTube being very popular.5 In 2012, 93% New Zealand showed that the average rate of New Zealanders aged 15–24 years used of unhealthy food advertising was 9.1±5.2 the Internet, with this number likely to have per hour and about 88% of unhealthy food grown since. In addition, 88% of Internet advertisements were shown during chil- users engage in social media, with Facebook dren’s peak viewing times.11 However, new and YouTube being the most popular sites.6 media, such as online and social media

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 14 www.nzma.org.nz/journal ARTICLE

allows marketers to engage more deeply unhealthy food and beverage marketing,7,17 with their audiences and even magnify and a critical review of the new Code by 77 effects of traditional media.5 A previous health professors indicates that the impact New Zealand study identifi ed a wide range of the new Code on reducing exposure of of marketing techniques and features on children and adolescents to unhealthy food food brand websites, including adver- marketing is uncertain.18 cation (87%), viral marketing (64%), cookies The aim of this study is to determine the (54%), free downloadable items (43%), extent, nature and potential impact of inter- promotional characters (39%), designated net-based food and beverage marketing by children’s sections (19%) and advergaming analysis of the marketing of popular food (13%). Most techniques appeared more brands in New Zealand on Facebook and frequently on websites specifi cally targeting YouTube. children and adolescents, than on other websites targeting the general public.12 Food marketing on social media has not yet Methods been analysed in New Zealand and there Selection of food and beverage are very few studies internationally. A brands recent Australian study analysing content Using Socialbakers,19 a global social media of Facebook pages of the most popular food analytics company, the most popular food brands found that competitions based on and beverage brands on Facebook and user-generated content, interactive games YouTube in New Zealand were identifi ed. and apps were the most common tech- Numbers of page ‘likes’ by New Zealanders niques used to engage with consumers, on Facebook and numbers of New Zealand adolescents in particular.13 In addition, the channel subscribers on YouTube (including study found that adolescents and young all age groups) were used as a measure of adults appeared to be the users engaging popularity. 13 the most with the marketing content. We selected the 15 most popular Facebook Another Australian study also found that pages and the fi ve most popular YouTube the majority of promotional activities that channels for each of the packaged food, fast selected food brands are using to promote food and beverage categories, giving us 45 unhealthy food and beverages are targeted Facebook pages and 15 YouTube channels 14 at adolescents. In general, adolescents are to analyse. Brands tended to post less considered an important target group for frequently on YouTube than Facebook, so 15 food and beverage marketers. Currently in two years of YouTube videos (2015–2016) New Zealand, advertising is self-regulated and two months of Facebook posts (October– by the industry-led Advertising Standards November 2016) were analysed for each Authority (ASA). The ASA recently reviewed brand selected. its advertising codes. The Children’s and Young People’s Advertising Code (the new Data collection Code)16 went into full effect beginning Screenshots of each post on a popular October 2017. It strengthens previous brand’s Facebook page were captured with restrictions on advertising of occasional (ie, Evernote and the following information was unhealthy) food and beverages, specifi cally entered into a database: the type of post for children younger than 14 years. For (video, image or text alone), the number adolescents 14–18 years, these same restric- of likes, shares, comments and views (if tions do not apply.16 post was a video), the presence of a food or beverage product, the nutritional quality of In addition, the new Code includes defi ni- the product if one was advertised, the use tions about targeting children. As for social of activities, premium offers and/or promo- media in particular, the new Code implies tional strategies. Similar information was that brands and companies cannot target collected for each YouTube video, including any occasional food advertisements to the number of video views, the presence of children and young people aged less than 14 a food or beverage product, the nutritional years old.16 However, research has consis- quality of the food or beverage product if tently shown that self-regulation does not one was advertised, and the use of activ- signifi cantly reduce children’s exposure to ities, premium offers and/or promotional

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 15 www.nzma.org.nz/journal ARTICLE

strategies. For three food brand Facebook ad” feature, which defi nes a target Facebook pages and three Youtube Channels (one of audience for potential advertisers. When each food industry sector), two researchers logged in to Facebook as a Page Adminis- coded the advertisements independently trator, Create Adverts could be selected from and compared results. Since there were the dropdown menu. In Create a Custom no discrepancies, one researcher then Audience, we specifi ed ‘New Zealand’ and continued coding the remaining pages and ‘13–18 years’. The Interests fi lter then listed channels. the size of specifi c audiences by looking To assess the nutritional quality of at their interests, activities, the pages they advertised food and beverage products, have liked and closely related topics, ie, the Ministry of Health Food and Beverage Facebook assesses a potential audience who Classifi cation System (updated in 2016 and have ‘liked’ or engaged with named pages now under the auspices of the Heart Foun- or similar ones. We entered the food and dation and called “Fuelled For Life”)20–22 was drink brands that were most popular by the used, which classifi es products as ‘everyday’, general New Zealand population as derived ‘sometimes’ or ‘occasional’. This is the clas- from Socialbakers as interests and iden- sifi cation system used by the new Code. tifi ed which of these food and drink brands Everyday foods and drinks are the healthiest Facebook defi ned as generating the greatest choices and are from the four food groups potential reach among 13–18 year-olds in (vegetables and fruit, breads and cereals, New Zealand. However, we were unable to milk and milk products, meat and alter- obtain similar estimates for YouTube. natives). Sometimes foods and drinks are mostly processed foods with some added fat, Results salt or sugar. Occasional foods and drinks are Facebook high in saturated fat, salt or sugar and should The most popular Facebook page among not be provided or sold to children (eg, packaged food brands was Whittaker’s confectionery, sugar-sweetened beverages). Chocolate Lovers, however, the page with Activities for consumers included games, the highest potential reach among 13–18 recipe ideas, voting, commenting, tagging year-olds was Chupa Chups. Among popular friends, liking and sharing posts, following fast food companies, the most popular page the brand on other media forms (eg, and the page with the highest potential Snapchat, Instagram, Twitter), arts and reach among 13–18 year-olds was McDon- crafts, registering for an event or down- ald’s. For popular beverage brands, Coca loading apps. Cola was the most popular Facebook page Promotional strategies included the use and had the highest potential reach among of cartoons/company owned characters, 13–18 year-olds (Table 1). licensed characters, amateur sportspersons, Overall, 762 Facebook posts were made by famous sportspersons/teams, sports events, the 45 brands during October and November non-sports celebrities, movie tie-ins, 2016, and 28% of these posts were videos non-sporting events/festivals, claims of (Table 2). The number of posts made awards won, a child consuming a product over the two-month period varied largely and advercation (where information about between brands, ranging from zero posts the company or its foods is presented as (zero posts per day) by Skittles to 62 posts educational material). Premium offers (one post per day) by Lewis Road Creamery included the use of competitions, buy one get (Table 1), giving an overall average of 17 one free (or similar), 20% extra product (or posts (0.3 posts per day) (Table 2). similar), limited time only/limited edition, Of the Facebook posts, 64% contained donation to charity, free item with purchase a food or beverage product and the and limited time product discounts. remainder posts were marketing the brand Reach or company without depicting a specifi c On Facebook (not possible through food or beverage product. About 99% of Socialbakers) we estimated the potential food and beverage products marketed reach of posts made by each food brand were classifi ed as being for occasional among 13–18 year-olds using the “Create consumption only. Of the 45 pages

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 16 www.nzma.org.nz/journal ARTICLE

Table 1: Extent of advertising by food brands popular in New Zealand on their Facebook pages (October–November 2016).

Popularity Food brand Number of Potential likes Number of posts rank page likes1 of page among on page during 13–18 year olds2 two-month period PACKAGED FOOD 1 Whittaker’s Chocolate 394,359 14,000 19 Lovers

2 Gri– in’s 242,072 12,000 13

3 Cadbury Dairy Milk 234,456 21,000 9

4 Tip Top Ice Cream 191,569 12,000 18

5 Skittles 170,602 8,200 0

6 Chupa Chups 161,619 56,000 8

7 Lewis Road Creamery 158,798 5,300 62

8 Nutri-Grain NZ 113,953 11,000 6

9 Kiwi Bacon 106,834 4,100 8

10 Pringles 106,231 35,000 2

11 Tasti NZ 98,997 N.A3 7

12 Puhoi Valley 85,604 N.A3 45

13 Ferrero Rocher 84,033 7,100 4

14 KitKat 81,181 14,000 13

15 Marmite NZ 79,538 2,900 11

FAST FOOD 1 McDonald’s 423,818 100,000 10

2 KFC 372,312 81,000 26

3 Domino’s NZ 234,054 23,000 34

4 Pizza Hut 212,390 41,000 22

5 Subway NZ 212,390 35,000 19

6 Burger King NZ 185,295 27,000 53

7 Carl’s Jr. NZ 171,591 34,000 26

8 Pita Pit NZ 129,097 29,000 15

9 BurgerFuel 100,023 29,000 10

10 Starbucks 80,149 42,000 4

11 Hell Pizza 79,043 6,600 26

12 Subway 68,478 35,000 23

13 Nando’s NZ 68,014 8,100 18

14 Mexico NZ 56,969 N.A3 244

15 Wendy’s NZ 55,612 9,200 35

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 17 www.nzma.org.nz/journal ARTICLE

Table 1: Extent of advertising by food brands popular in New Zealand on their Facebook pages (October–November 2016) (continued).

BEVERAGES 1 Coca Cola NZ 213,369 42,000 6

2 Red Bull 190,654 41,000 14

3 Lemon & Paeroa 189,313 17,000 4

4 V Energy NZ 144,141 7,800 11

5 Monster Energy 115,654 9,500 54

6 Mountain Dew – NZ 76,787 17,000 10

7 Powerade NZ 69,491 8,800 12

8 V Energy Drink Australia 66,846 2,100 6

9 Pepsi NZ 60,274 12,000 6

10 Charlie’s Drinks 52,014 <1,000 7

11 Dr Pepper 27,253 1,800 21

12 Lipton Ice Tea 20,714 2,300 3

13 Phoenix Drinks 19,309 N.A3 105

14 Gatorade NZ 18,817 6,700 21

15 Sprite Australia and NZ 18,286 N.A3 7

1. Number of New Zealanders who like the Facebook page. Data obtained from the Social bakers website in mid- November 2016. 2. Estimated using the ‘Create Ads’ feature on Facebook for the particular food brand. 3. Not applicable. There was no information on potential likes by adolescents for these company pages. 4. Six posts were excluded because they contained alcohol. 5. One post was excluded because it contained alcohol.

Table 2: Comparison between popular packaged food, beverage and fast food brands of extent, nature and nutritional quality of posts made on their Facebook pages (October–November 2016).

Packaged Fast food Beverage Total food brands3 brands3 (n=45 brands (n=15 (n=15 brands) (n=15 brands) brands) brands) Volume and type of posts Total number of posts on all pages (n) 225 345 192 762

Average number of posts per page (n) 15 23 13 17

Average number of posts per day per page 0.2 0.4 0.2 0.3 (n)

Posts that were videos (n (%)) 45 (20) 76 (22) 94 (49) 215 (28)

Level of consumer interaction with posts1 Likes per post (mean ± SD) 830 ± 1,916 ± 8,526 ± 3,261 ± 1,408 9,503 26,791 15,228

Shares per post (mean ± SD) 71 ± 222 481 ± 3,463 989 ± 2,729 488 ± 2,727

Comments per post (mean ± SD) 294 ± 680 268 ± 796 294 ± 832 282 ± 773

Views per video (mean ± SD) 79,021 ± 437,088 ± 782,817 ± 514,908 ± 75,152 988,319 2,053,039 1,509,284

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 18 www.nzma.org.nz/journal ARTICLE

Table 2: Comparison between popular packaged food, beverage and fast food brands of extent, nature and nutritional quality of posts made on their Facebook pages (October–November 2016) (continued).

Nutritional quality of food and/or beverage products in posts Posts containing a food and/or beverage 187 (83) 231 (67) 71 (37) 489 (64) product (n (%))

Food and/or beverage products classified 205 (91) 208 (90) 71 (100) 484 (99) as occasional2 (n (%))

Facebook pages with 100% of products 11 (73) 8 (53) 13 (87) 32 (71) classified as occasional (n (%))

Use of marketing techniques in posts Posts with an activity for consumers (n (%)) 128 (57) 105 (30) 44 (23) 276 (36)

Posts with a promotional strategy (n (%)) 52 (23) 121 (35) 136 (71) 309 (41)

Posts with a premium o– er (n (%)) 81 (36) 145 (42) 35 (18) 261 (34)

1. The number of likes, comments, shares and views on each post may include non-New Zealanders who have also liked the brands Facebook page. 2. Classified according to the Ministry of Health food and beverage classification system. 3. Six Posts by fast food brands and one post by a beverage brand were excluded because they included an alcohol product.

Table 3: Extent of advertising by food brands popular in New Zealand on their Youtube channels (2015–2016).

Popularity Brand YouTube Channel Number of channel Number of videos during ranking subscribers1 two-year period (n) PACKAGED FOOD 1 KitKat Australia & New Zealand 2,786 16

2 Tic Tac ANZ 2,419 6

3 Whittaker’s Chocolate 872 11

4 Weetbix NZ 637 25

5 Anchor NZ 348 62

FAST FOOD 1 Hell Pizza NZ 35,577 1

2 Maccas NZ 2,620 20

3 KFC NZ 905 24

4 Carl’s Jr. NZ 549 23

5 Domino’s NZ 433 29

BEVERAGES 1 V Energy NZ 2,893 15

2 Coke Happiness NZ 841 10

3 Nescafe Australia & NZ 810 19

4 Sprite Australia & NZ 684 14

5 Mountain Dew NZ 394 25

1. Data obtained from the Socialbakers website in mid-November 2016.

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 19 www.nzma.org.nz/journal ARTICLE

Table 4: Comparison between popular packaged food, beverage and fast food brands of extent, nature and nutritional quality of posts made on their Youtube channel (2015–2016).

Packaged foods Fast food outlets Beverages (n=5 Total (n=15 (n=5 brands) (n=5 brands) brands) brands) Quantity and views of videos Total number of videos by all channels (n) 120 97 83 300

Average number of videos per channel (n) 24 19 17 20

Average number of videos per channel each month (n) 1.0 0.8 0.7 0.8

Average number of views per video (mean ± sd) 87,879 ± 279,754 27,576 ± 62,654 124,395 ± 247,355 78,408 ± 225,732

Nutritional quality of foods and/or beverages in videos Videos containing a food and/or beverage product (n (%)) 94 (78) 82 (85) 76 (90) 252 (84)

Food and/or beverage products that were classified as 36 (38) 82 (100) 76 (100) 194 (77) occasional1 (n (%))

YouTube channels with 100% of products classified as 3 (60) 5 (100) 5 (100) 13 (87) occasional (n (%))

Use of marketing techniques Videos containing an activity for consumers (n (%)) 42 (35) 29 (30) 29 (35) 100 (33)

Videos containing a promotional strategy (n (%)) 104 (87) 38 (39) 40 (48) 182 (61)

Videos containing a premium o– er (n (%)) 13 (11) 35 (36) 25 (30) 73 (24)

1. Classified according to the Ministry of Health food and beverage classification system.

Figure 1: Examples of activities, premium offers and promotional strategies on food brands’ Facebook pages.

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 20 www.nzma.org.nz/journal ARTICLE

analysed, 32 had 100% of their advertised products classifi ed as occasional (Table 2). Discussion The pages with the largest potential reach The extent of unhealthy food advertising among 13–18 year-olds in each of the three by popular food and beverage brands on categories (Coca Cola, McDonald’s and Facebook is substantial in New Zealand, Chupa Chups) all had 100% of their adver- with food brands posting on average every tised food or beverage products classifi ed three days, but some brands more than once as occasional. a day. In comparison, advertising was lower on YouTube, with brands posting videos Activities for consumers were used in 36% less than once a month on average on their of Facebook posts, with asking consumers to respective channels. The total volume of like, comment, tag and share posts being the advertising by those food brands is likely most commonly used activity. Promotional much larger as this study only looked at strategies were used in 41% of posts, and the posts on food brand pages and channels, most frequently used strategy was having and not all advertising by those brands on a famous sportsperson/team in the post. Facebook and YouTube is listed on their Premium offers were used in 34% of posts, pages and channels. with competitions being the most common offer used (Table 2 and Figure 1). On both Facebook and YouTube, the food and beverage products advertised by brands It was estimated some posts for the most were nearly all classifi ed as occasional popular brands could potentially reach 10% according to the Ministry of Health food of New Zealand adolescents (Table 1). and beverage classifi cation system.20 The YouTube potential exposure to such ads is important The most popular YouTube channels in as for the most popular brands it was esti- the packaged foods, fast food outlet and mated that about 10% of adolescents could beverage categories were KitKat, Hell Pizza potentially notice them in their Facebook and V Energy NZ respectively (Table 3). newsfeeds. Social media advertisements use Overall, 300 YouTube videos were made marketing techniques extensively. by the 15 YouTube channels during 2015– Nearly every brand asked followers 2016 (Table 4). The volume of videos made to like, comment, tag friends and share over the two-year period varied largely their posts, ensuring that their product between brands, ranging from one video was seen not only by their followers but (0.0 videos per month) by Hell Pizza to 62 also by the followers Facebook ‘friends’. videos (2.6 videos per month) by Anchor Famous sportspersons and teams, such (Table 3), giving an overall average of 20 as the All Blacks, were most frequently videos over the two-year period (0.8 videos used to promote products. By using these per month) (Table 4). techniques, brands attract the consumer’s Of the YouTube videos, 84% contained a attention, increase their brand loyalty and food or beverage product and 77% of these make them more likely to go and buy their products were classifi ed as being for occa- product.23,24 An Australian study similarly sional consumption only. Of the 15 YouTube found that food brand pages widely used channels analysed, 13 had 100% of their marketing features such as competitions advertised products classifi ed as for occa- based on user-generated content, interactive sional consumption only (Table 4). games and apps, and that adolescent and Activities for consumers were used in young adult Facebook users appeared most 13 33% of YouTube videos, with arts and crafts receptive to engaging with this content. being the most commonly used activity. The World Health Organization (WHO) Promotional strategies were used in 61% has published a recent report25 recognising of videos and premium offers were used in that digital marketing amplifi es marketing 24% of videos. Similar to Facebook, the most in traditional media, achieving greater common promotional strategy used was ad attention and recall, greater brand having a famous sportsperson/team in the awareness and more positive brand atti- video and the most common premium offer tudes, and greater intent to purchase.25 In used was having a competition (Table 4). addition, social media platforms collect

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 21 www.nzma.org.nz/journal ARTICLE

extensive personal data from users to companies cannot target children and deliver advertising and targeting children adolescents on social media with occa- and adolescents, without effective regulation sional food advertising, informing young to protect children from this practice.25 people and parents about the harmful Researchers do not have access to the effects of unhealthy food marketing and same data that marketers have, which makes identifying international options to deal it diffi cult to gain insights into real exposure with this particular form of food marketing. of children and adolescents to advertising The World Health Organization, in its 25 on social media. While this study shows that recent report, encourages governments to social media is an important medium for acknowledge their duty to protect children New Zealand food marketers to engage with online through regulations that extend the users and adolescents in particular, it is not protection they offer children offl ine to 25 yet possible to assess exposure of children online areas. The WHO proposes the devel- and adolescents to advertising on social opment of a rights-based framework for media or compliance of marketers with the regulation of digital food marketing to the new Code. From October 2017 onwards children based on the rights to participation according to the new Code, marketers and protection accorded to children under cannot anymore target any advertisements the United Nations Convention on the Rights with occasional foods to adolescents younger of the Child, which recognises the duty of than 14 years on social media. New method- states to protect the rights of children online, ologies have to be developed for monitoring including their right to health. Article social media food marketing, a need clearly 13 of the WHO Framework Convention 26 recognised in the recent WHO report.25 on Tobacco Control calls for Parties to Further research also needs to investigate recognise that “a comprehensive ban on food advertising across all media platforms advertising, promotion and sponsorship and interactivity between the different would reduce the consumption of tobacco media and how food and beverage adver- products” and “to undertake a compre- tising is perceived by teenagers. hensive ban of all tobacco advertising, promotion and sponsorship”, including Limitations of this research include that “a comprehensive ban on cross-border Facebook and YouTube posts were analysed advertising, promotion and sponsorship retrospectively, and because some brands originating from its territory”. Progress delete posts after a certain period of time, has been made by States in committing to we may have missed some posts leading to eliminate cross-border marketing of tobacco an underestimation of the volume of posts products within the European Union. A made by brands. In addition, brands do not similar Framework Convention to Protect post all advertisements on their page or and Promote Healthy Diets, as previously channel as they need to pay to have their proposed,27 could include similar provisions posts seen and they can pay Facebook to for advertising unhealthy foods to children. distribute advertisements on newsfeeds of a wide range of targeted users (including users who do not like the particular brands) Conclusion instead. For YouTube we only analysed food Social media is an important medium for brand channels, but food brands can pay to food marketers in New Zealand, and promo- advertise in a wide range of other popular tional strategies and premium offers are non-food YouTube channels. In addition, frequently used. Additional methodology the likes for the posts made by food brands needs to be developed to monitor exposure include all likes, not just New Zealand likes. of children and adolescents to such adver- Policy implications of this research tisements and to assess compliance of food include ensuring that food brands and marketers with the new Advertising Stan- dards Authority self-regulatory Code.

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 22 www.nzma.org.nz/journal ARTICLE

Competing interests: All authors report grants from Health Research Council of New Zealand during the conduct of the study. Author information: Stefanie Vandevijvere, Senior Research Fellow, School of Population Health, The University of Auckland, Auckland; Charlotte Aitken, Summer Student, School of Population Health, The University of Auckland, Auckland; Boyd Swinburn, Professor, School of Population Health, The University of Auckland, Auckland. Corresponding author: Stefanie Vandevijvere, Senior Research Fellow, School of Population Health, The University of Auckland, Auckland. [email protected] URL: http://www.nzma.org.nz/journal/read-the-journal/all-issues/2010-2019/2018/vol-131-no-1473- 13-april-2018/7539

REFERENCES: 1. Ministry of Health. Annual 6. Gibson A, Miller M, Smith P, implications. Public Update of Key Results Bell AC, Crothers C. World Health Nutr 2017:1–12. 2015/16: New Zealand Internet Project: Internet 12. Vandevijvere S, Sagar Health Survey. 2016. http:// trends in New Zealand K, Kelly B, Swinburn B. www.health.govt.nz/ (2007–2013). Auckland: Unhealthy food marketing publication/annual-update- AUT University, 2013. to New Zealand children key-results-2015-16-new- 7. Vandevijvere S, Swinburn and adolescents through zealand-health-survey B. Getting serious about the internet. N Z Med J 2. Boyland EJ, Nolan S, Kelly protecting New Zealand 2017; 130(1450):32–43. B, et al. Advertising as a cue children against unhealthy 13. Freeman B, Kelly B, Baur to consume: a systematic food marketing. N Z Med L, et al. Digital Junk: Food review and meta-analysis J 2015; 128(1417):36–40. and Beverage Marketing of the effects of acute 8. Wilson N, Signal L, on Facebook. Am J Public exposure to unhealthy Nicholls S, Thomson G. Health 2014: e1–e9. food and nonalcoholic Marketing fat and sugar 14. Boelsen-Robinson T, beverage advertising on to children on New Backholer K, Peeters intake in children and Zealand television. Prev A. Digital marketing adults. Am J Clin Nutr Med 2006; 42(2):96–101. of unhealthy foods to 2016; 103(2):519–33. 9. Jenkin G, Wilson N, Australian children and 3. Cairns G, Angus K, Hermanson N. Identifying adolescents. Health Promot Hastings G, Caraher M. ‘unhealthy’ food advertis- Int 2016; 31(3):523–33. Systematic reviews of the ing on television: a case 15. Dunlop S, Freeman B, Jones evidence on the nature, study applying the UK SC. Marketing to Youth extent and effects of food Nutrient Profi le model. in the Digital Age: The marketing to children. A Public Health Nutrition Promotion of Unhealthy retrospective summary. 2009; 12(05):614–23. Products and Health Appetite 2013; 62:209–15. 10. No E, Kelly B, Devi A, Promoting Behaviours 4. Cairns G, Angus K, Swinburn B, Vandevijvere on Social Media. Media Hastings G, Caraher M. S. Food references and Commun 2016; 4(3):35–49. Systematic reviews of the marketing in popular 16. Advertising Standards evidence on the nature, magazines for children Authority. Children and extent and effects of food and adolescents in New Young People’s Advertising marketing to children. A Zealand: A content analy- Code. 2017. http://www. retrospective summary. sis. Appetite 2014; 83:75–81. asa.co.nz/codes/codes/ Appetite 2012; (0). 11. Vandevijvere S, Soupen new-children-young-peo- 5. Kelly B, Vandevijvere S, A, Swinburn B. Unhealthy ples-advertising-code/ Freeman B, Jenkin G. New food advertising directed (accessed 09/03/2017). media but same old tricks: to children on New 17. Galbraith-Emami S, food marketing to children Zealand television: extent, Lobstein T. The impact in the digital age. Current nature, impact and policy of initiatives to limit obesity reports 2015.

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 23 www.nzma.org.nz/journal ARTICLE

the advertising of food 22. Heart Foundation. Food 2016. http://www.euro. and beverage products and Beverage Classifi - who.int/en/health-topics/ to children: a system- cation System Nutrient disease-prevention/nutri- atic review. Obes Rev Framework for Schools. tion/publications/2016/ 2013; 14(12):960–74. 2016. http://www.asa. tackling-food-marketing- 18. Swinburn B, Vandevijvere co.nz/wp-content/ to-children-in-a-digital- S, Woodward A, et al. uploads/2017/02/FBCS-Nu- world-trans-disciplinary- Proposed new industry trient-Criteria-March-2016. perspectives-2016 code on unhealthy food pdf (accessed 12/12/2017). (accessed 8/11/2016). marketing to children and 23. Kraak VI, Story M. Infl u- 26. World Health Organiza- young people: will it make ence of food companies’ tion. WHO Framework a difference? N Z Med J brand mascots and Convention on Tobacco 2017; 130(1450):94–101. entertainment companies’ Control2003. http:// 19. Socialbakers. http:// cartoon media characters whqlibdoc.who.int/publi- www.socialbakers.com/ on children’s diet and cations/2003/9241591013. (accessed 11/11/2013). health: a systematic review pdf (accessed April 2011). and research needs. Obes 20. Ministry of Health. 27. World Obesity Federation, Rev 2015; 16(2):107–26. Food and beverage Consumers International. classifi cation system 24. Jenkin G, Madhvani N, Recommendations towards for years 1–13., 2007. Signal L, Bowers S. A a Global Convention to systematic review of protect and promote 21. Heart Foundation. persuasive marketing tech- healthy diets. 2014. Fuelled4Life school niques to promote food to http://www.worldobe- canteen catering guide. children on television. Obes sity.org/what-we-do/ 2012. http://s3-ap-south- Rev 2014; 15(4):281–93. policy-prevention/advo- east-2.amazonaws.com/ cacy/global-convention/ hfweb-fuelled-for-life/ 25. World Health Organization (accessed 15/07/2014). uploads/documents/ Europe. Tackling food School_Catering_Guide. marketing to children in pdf?mtime=1464168002 a digital world: trans-dis- (accessed 06/12/2017). ciplinary perspectives

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 24 www.nzma.org.nz/journal ARTICLE

Regulatory chills: tobacco industry legal threats and the politics of tobacco standardised packaging in New Zealand Eric Crosbie, George Thomson

ABSTRACT AIMS: To describe the process of enacting tobacco standardised packaging (SP) amidst tobacco industry legal threats in New Zealand. METHODS: Relevant government and NGO documents, and media items were reviewed. Policymakers and health advocates in New Zealand were interviewed. The data were triangulated and thematically analysed. RESULTS: In 2011, the New Zealand Government announced the goal of becoming a smokefree country (reducing smoking prevalence to 5%) by the year 2025, and considered adopting SP. In April 2012, the Government announced it would introduce SP, but tobacco companies threatened the Government with litigation in international courts for violating investment and intellectual property rights. In response, the Government adopted a ‘wait and see’ approach, waiting until two legal challenges against Australia’s SP law were resolved before it enacted its legislation in September 2016. Health advocates, limited due to funding constraints, attempted to alter the Government’s approach to the legal threats without success. Interviews with policymakers and health advocates confirmed these threats helped produce a regulatory chill, delaying the policymaking process by three years. CONCLUSION: The New Zealand case illustrates how the threat of a potential international lawsuit can create a chilling e– ect by helping delay the implementation of public health policies.

he constantly evolving nature of trade on tobacco control policies.8 This emerging governance in the 21st century is literature highlights how tobacco companies increasingly having a profound impact have used trade agreements to try and T 9,10 on public health and the development of block innovative public health proposals, public health policies globally.1 New rules and how the threat of legal action can help governing international trade are increas- dissuade governments from implementing ingly affecting access to medicines,2 alcohol progressive policies.11,12 3 4 control and nutrition regulations. In partic- The ‘regulatory chill’ hypothesis suggests ular, these issues are highly contentious in that governments may weaken or withdraw 5 trade and tobacco control. public policies due to concerns over While much of the focus of the trade and capital fl ight or potentially costly trade tobacco control literature has concentrated and investment lawsuits.13 Some critics of on commercial aspects of trade liberali- this hypothesis argue policymakers are sation on tobacco consumption6 and legal unaware of trade and investment law, and implications of trade agreements on public consequently do not take these laws into health,7 recent studies have examined the consideration when making legislative and political implications of trade agreements regulatory decisions.14 Other critics argue

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 25 www.nzma.org.nz/journal ARTICLE

it is diffi cult to prove that legal threats in accordance with a protocol approved alter the regulatory process and produce by the University of California, Santa Cruz a chilling effect.15 However, recent studies Committee on Human Research. We also illustrate policymakers are increasingly interviewed one offi cial from the Ministry aware of trade and investment law and of Foreign Affairs and Trade who requested incorporate these understandings into the anonymity. In the Results below, inter- policymaking process.16 viewees are cited by initials, with a key This article examines the politics of to the initials given in the online supple- regulatory chill and the reactions of a mentary material (Supplementary Table nation-state to industry legal threats, by 1). Results from these sources were trian- analysing the policy process for tobacco gulated and thematically analysed through 19 standardised packaging (SP) in New Zealand standard process tracing frameworks. (2010–16). Instead of testing the regulatory chill hypothesis (whether a policy is enacted Results or weakened), this paper investigates the A series of delays to SP occurred from extent to which industry legal threats signifi - November 2010 to September 2016. All 23 cantly contributed to political delays that interviewees (policymakers and advocates) have a profound impact on public health. confi rmed that industry legal threats were In September 2009, the Māori Affairs either the most signifi cant or a primary Select Committee of the New Zealand reason for delays (Supplementary Table parliament started an inquiry into the 1).AI, BS, DS, LW, MF, FT, ILG, JS, KW, SS, SO, tobacco industry and the disproportionate TUF, CB, EC, GL, JK, LR, PS, RB, SB, SE, ME, LD. harm of tobacco use to the Māori popu- Introduction of standardised lation. In November 2010, the committee recommended New Zealand become packaging (SP) From November 2010, the Cabinet ‘smokefree’ (interpreted as reducing considered the SP recommendation and in smoking prevalence to 5%) by 2025.17 The April 2012 agreed in principle to introduce committee recommended several policies to SP20 (Figure 1). However, the New Zealand enable this goal, including SP. Over the next Government was divided on the issue of six years, tobacco companies pressured the introducing SP. Associate Health Minister New Zealand Government with legal threats and Māori Party (an indigenous rights party to the SP proposal. We examine the effects of within government) Member of Parliament this pressure. (MP), Tariana Turia, was highly supportive of introducing SP, stating it was “a powerful Methods tool” to reduce the appeal of tobacco We reviewed New Zealand Government products and smoking in general. Turia and health group documents, and New also stated SP would fulfi ll New Zealand’s Zealand media items from offi cial and commitment to the World Health Organiza- media websites using standard snowball tion’s Framework Convention on Tobacco searches,18 beginning with search terms Control treaty (FCTC).20 The National ‘plain packaging’, ‘standardised packaging’, Party-led coalition Government had dele- ‘international trade’, ‘intellectual property’, gated responsibility for tobacco control to ‘tobacco companies’, ‘threats’, as well as Turia, who was a Minister outside Cabinet. using key dates and specifi c actors. Between In New Zealand, an Associate Minister’s January and June 2015, we attempted to delegated authority is constrained, as policy recruit via email and telephone 38 New decisions are controlled by the Health Zealand interviewees closely involved in Minister and Cabinet. the SP process. Twenty-three agreed to be Meanwhile some members of Cabinet interviewed, 10 declined, and fi ve never were less optimistic and more cautious responded to multiple requests. Of the 23 about SP, including Health Minister interviewees, six were tobacco control Tony Ryall, who lacked a demonstrated advocates, four were academics, 11 were commitment to the Smokefree 2025 goal.21 members of parliament (MPs) and one Prime Minister Key told reporters that was a Health Ministry offi cial. The inter- Government was “likely” able to introduce viewees agreed to waive their anonymity

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 26 www.nzma.org.nz/journal ARTICLE

SP legally, but that it was “not absolutely disagree.co.nz), which reiterated industry clear cut” and no “slam dunk”22 (Supple- arguments submitted during the consultation mentary Table 2). National Party Trade period. Their television advertising achieved Minister Tim Groser also said there were very high reach and frequency of exposure, “some complexities” concerning the and presented arguments that were either proposal’s legality needing to be addressed.23 unsound or demonstrated fallacies.28 The Health Ministry, which is respon- Health group support sible for drafting and implementing the SP The main public health groups in New legislation, held a 60-day public consultation Zealand, and international health groups, period on the draft (the interim version is lawyers, activists and academic scholars called a ‘Bill’) between August and October submitted comments supporting the 2012. Bill.29 They argued SP reduced the appeal Tobacco industry initial opposition of tobacco products by removing the British American Tobacco (BAT) (68.3% glamorisation and contributed to the de-nor- market share in New Zealand), Imperial malisation of tobacco, especially for youth Tobacco (20.0%) and Philip Morris Interna- and vulnerable populations. Health groups tional (PMI) (7.1%) opposed SP. As with the also identifi ed there may be domestic and opposition against SP in other countries, international legal implications associated including Australia, Ireland, France and with SP, but did not see these as a reason the UK,24 their comments centred around not to proceed. Instead, they argued SP was arguments that SP: 1) would not work, 2) a justifi ed public health action and met the would increase illicit tobacco trade, 3) would Government’s commitments to the FCTC. create unnecessary problems for retailers Ministry of Health regulatory and 4) would violate domestic laws and impact statement and reports to international treaties governing intellectual property and investment.25–27 Cabinet In November 2012, the Health Ministry The industry claimed the proposal would: presented their report on the submissions,29 a) deprive them of intellectual property mentioning that several submitters argued rights by degrading the value of their SP violated several treaties. After consulting trademarks and expropriate their invest- with other ministries, the Health Ministry ments, b) constitute an unnecessary barrier also issued a regulatory impact statement to trade and c) did not accord them fair (RIS) addressing potential impacts and risks 25–27 and equitable treatment. In particular, of SP.30 The RIS stated that the Ministry of they argued the proposal violated New Foreign Affairs and Trade (MFAT) warned Zealand’s Bill of Rights and its obligations there was “a reasonably high risk of trade under various trade and investment agree- litigation,”30 (Supplementary Table 2). ments, including the Paris Convention MFAT also noted that Australia, the fi rst for the Protection of Industrial Property, country to introduce SP, was facing two and the World Trade Organization (WTO) international legal challenges (one through Agreements on Trade-Related Aspects of WTO). MFAT estimated a 1.5–2 million NZD the Intellectual Property Rights (TRIPS) and cost to defend a WTO case and potentially Technical Barriers to Trade (TBT). If SP was substantially more to defend an investment enacted, tobacco companies threatened arbitration lawsuit.30 to sue the Government for compensation, As part of the executive policy process, which they claimed would amount to Associate Health Minister Turia sent a billions of dollars. report on SP on November 27, 2012 to BAT media campaign the Cabinet Social Policy Committee. In August 2012, BAT ran a multi-media It expressed concerns over a potential campaign titled “Agree Disagree” opposing legal challenge as a signifi cant risk that the SP Bill with the slogan “We agree that “would require signifi cant investment of tobacco is harmful. We disagree that plain resources”,31 estimating $3–$6 million for packaging will work.” The campaign ran investment arbitration. The report acknowl- media advertising on television, radio and edged that risks would be signifi cantly print and had its own website (www.agree- mitigated upon conclusion of the Australian

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 27 www.nzma.org.nz/journal ARTICLE

legal disputes. It therefore proposed devel- waiting for the Australian legal challenges oping “policy details to enable legislation to before enacting legislation36 (Supplementary be considered for introduction by August/ Table 2). Despite the Government’s cautious September 2013.”31 Consequently, the approach, the Bill was allowed to proceed to Government developed a ‘wait and see’ the House Health Select Committee. approach reliant upon the two Australian legal challenges for “greater legal certainty” Health House Select Committee 2014 In February and March 2014, the Health before proceeding with SP in New Zealand.32 Select Committee received 191 substantive Resuming the process of SP submissions on the Bill. These included legislation some from international legal experts who On 21 August 2013, Minister Turia issued argued that the SP proposal was consistent another report proposing the Bill be intro- with international law, including WTO obli- duced in parliament and sent to the Health gations. Both tobacco companies25–27 and Select Committee. In proposing SP, the report health groups37 reiterated their positions on again took a cautious approach, stating the the Bill. In particular, health groups urged Bill “could be delayed if necessary”, reiter- the Government to pass the Bill immediately ating the “uncertainty” of the timetable for without further delays. the Australian legal challenges.33 Health Ministry response to Health Although the Cabinet continued to Committee consider legal challenges as a “high risk”, On 18 June 2014, the Health Ministry the August 2013 report addressed for the gave their submission to the Health Select fi rst time the issue of intellectual property Committee in consultation with other minis- rights. It acknowledged that the right to tries, especially MFAT.38 The Health Ministry “register” trademarks did not grant tobacco accepted the international legal advice companies the right to “use” the trade- on the Bill’s legal standing, stating it was marks,33 as acknowledged by domestic consistent with New Zealand’s WTO obliga- rulings outside New Zealand,34 international tions concerning intellectual property rights rulings8 and the tobacco industry’s internal and was non-discriminatory under trade legal counsel.12 Following the August 2013 and investment agreements (Supplementary report, Minister Turia introduced the SP Bill Table 2). The submission stated the legal to parliament in December 2013. analysis provided by opponents of the Bill SP first reading was “incomplete”, “selective” and did not On 11 February 2014, the parliament had “provide credible evidence to support their their fi rst reading of the SP Bill.35 MPs in claims”38 (Supplementary Table 2). opposition to the Bill (from the small New Health Select Committee report Zealand First and ACT parties) reiterated On 5 August 2014, the Health Select the industry’s arguments by discussing Committee submitted its report, which the potential risk and cost to tax payers rejected the industry’s trademarks associated with a legal challenge. MPs in argument but did not address other legal support of the Bill (from the Labour and issues pertaining to the Bill. The Health Green parties) condemned these industry Select Committee also did not address intimidation tactics and emphasised the whether the Bill should be delayed and importance of public health. The Bill was instead recommended the Bill to the House sent to the House Health Select Committee for a second reading.39 following a 142–1 vote. Following the fi rst reading, Associate Two more years of cautious delay On 20 September 2014, following the Health Minister Turia and the National recommendation for a second reading, New Party Cabinet disagreed on how the Bill Zealand held a general election. The National should proceed. Turia congratulated the Party remained in government, however, MPs for moving the Bill forward, and said Turia retired and the position of Associate that the Government should not be intim- Health Minister ‘responsible’ for tobacco idated by tobacco companies or delay the control was fi lled by new Cabinet Minister legislation.36 Prime Minister Key, however, Peseta Sam Lotu-liga, from the National stated the Government would continue

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 28 www.nzma.org.nz/journal ARTICLE

Party. The Māori Party lost two seats in semi-carve out for tobacco, meaning that parliament and became less infl uential as a governments would have the opportunity support party for the National Party. to prevent tobacco companies from using Health advocates interviewed for this the investor-state dispute settlement (ISDS) study felt this change represented a drastic mechanism to directly challenge public shift in leadership (Supplementary Table health policies under the TPP in future 1).CB,EC,GL,JK,LR,PS,RB,SB,SE Along with disputes. This prompted Labour Party several MPs also interviewed for this study MP Annette King to ask Lotu-liga when (Supplementary Table 1),DS,LW,MF,FT,IL- the Bill would be brought up for a second 47 G,KH,SS,SO,TUF they commented that reading. Minister Lotu-liga responded, former Minister Turia was a “bold” and stating the Government had a busy legis- 48 “courageous” leader who had dedicated her lative programme for 2015. Some health political career to advancing Māori rights. advocates felt this was a poor excuse, since While Lotu-liga did need agreement from the National Party was in its third term senior ministers to prioritise the Bill, as did of government and its legislative agenda his predecessor, health advocates stated was the least busy it had been in 30 years that, unlike Turia, he followed the caution (Supplementary Table 1). PS,RB,SE of the National Party instead of challenging On 18 December 2015, the international this approach. tribunal examining the Australia-PMI legal In 2014, two former tobacco industry dispute dismissed PMI’s challenge, ruling lobbyists, Christopher Bishop (Corporate that it did not have jurisdiction to hear PMI’s 49 Affairs Manager PMI, 2011–2013) and Todd claim. This tribunal ruled that initiation Barclay (Corporate Affairs Manager PMI, of the arbitration constituted an “abuse of 2013–2014) became National Party MPs in right” because Philip Morris Asia did not the new parliament. In 2012, Bishop was have any relevant investment in Australia PMI’s lead New Zealand spokesperson, when the SP Bill was announced in April appearing on television programmes to 2010, because PMI moved ownership oppose the SP proposal.40 Health advocates of its Australian operations from Swit- noted a confl ict of interest between their zerland to Hong Kong in February 2011, previous jobs and their duties as MPs, and 10 months after the Australian Govern- worried they were contributing to delaying ment’s SP announcement. Since one of the Bill. Although there is no evidence to the two Australian trade law disputes suggest these new MPs helped delay the had been settled and the Australian High process, their presence in government Court had ruled SP did not constitute refl ected long-term alliances between the acquiring the property (trademark), but National Party and the industry,41,42 and was merely restricting the use of the other strong contemporary links.43 trademark on the packaging and presen- tation of tobacco products,50 Prime Minister In February 2015, the UK and Ireland Key was questioned about the progress passed legislation requiring SP by May of SP. On 15 February 2016, he stated that 2016.44 In response, health advocates and the Government was “feeling a lot more MPs began calling on the New Zealand confi dent.”51 Government to call the Bill for its second reading and not continue waiting on the Although Prime Minister Key did not Australian legal disputes.44,45 In March identify a date for the Bill’s second reading 2015, Associate Health Minister Peseta Sam in parliament, he stated he expected it to 51 Lotu-liga replied to the demands, echoing become law by the end of the year. On the Government’s cautious ‘wait and see’ 19 February 2016, MP Annette King again approach by stating it was “prudent to await asked Minister Lotu-liga to explain the the World Trade Organization decision”46 on Bill’s delay in the House. Minister Lotu-liga Australian SP. responded that the bill would progress as priorities permitted.52 In September 2015, the text of the Trans-Pacifi c Partnership Agreement (TPP), Three of the four MPs from the National a regional trade agreement between New Party and NZ First Party interviewed for this Zealand and 11 countries in the Asia-Pa- study felt it prudent and pragmatic to await cifi c region, was fi nalised. It included a the result of the Australian legal challenges

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 29 www.nzma.org.nz/journal ARTICLE

before enacting SP (Supplementary Table 1). pressuring the Government to adopt a Bill SS,SO,BS They felt New Zealand, especially that had high public approval and strong as a small nation, could learn from those supporting evidence (Supplementary Table cases and adopt the necessary adjustments 1).PS,EC,LR,SE to avoid any unnecessary and protracted legal battles. These MPs were concerned SP second reading On 31 May 2016 (World No Tobacco about the legal costs associated with trade Day), Associate Health Minister Lotu-liga disputes, which they argued required announced the SP Bill would fi nally have spending taxpayers’ dollars (Supplementary its second reading in parliament in June Table 1).SS 2016. Prime Minister Key also confi rmed On the other hand, MPs from the Labour that tobacco industry legal threats were Party, Green Party and Māori Party argued the primary reason for delaying the Bill, this ‘wait and see’ approach ignored health stating it had been prudent to wait. He priorities and undermined New Zealand’s said, however, that his offi cials were now sovereignty (Supplementary Table 1).DS,L- advising him that it was safer to proceed W,MF,ILG,KH,TUF These MPs emphasised (Supplementary Table 2). New Zealand’s sovereign right to implement On 23 August 2016, parliament held its public health measures and said that it was second reading of the SP Bill, followed by alarming that a corporation could directly a discussion in the committee of the whole sue a government over attempts to advance House. A few MPs addressed the tobacco public health. However, some MPs opposing industry’s legal threats and accused the the ‘wait and see’ approach empathised Government of unnecessarily delaying the with the Government’s desire to avoid risk, process of SP.55 MPs then voted on the Bill, considering the legal uncertainty (Supple- which passed 108–13. The Bill had its third mentary Table 1).LW,ILG reading on 8 September 2016 and was given Constrained health groups (Health royal assent on 14 September 2016. Ministry realignment) Throughout 2015 and 2016, health Discussion advocacy groups urged the Government The case of SP in New Zealand illus- to move forward with the second reading, trates how an industry can use trade and but were reluctant to pressure the Health investment agreements to delay the policy- Ministry from whom they received the making process for public health measures. majority of their funding (Supplementary Other possible contributing factors for Table 1).PS,EC,LR,SE In 2014, the Ministry political delays include the closeness of the of Health announced they would ‘realign’ Government to the tobacco industry, and the their tobacco control services and prior- Government’s reluctance to be seen regu- ities between April 2015 and June 2016 with lating industry and acting against foreign more focus on tobacco use cessation rather investors.41 The government could also have 53 than prevention. been using the legal threats as an excuse More importantly, by June 2016, the not to pass SP for other reasons. While government had cut funding for national these factors may have contributed to the tobacco control advocacy by 79% (from delay, litigation threats were prominent in $1.7 million to $450,000) closing or largely government explanations for the delays of curtailing the operations of several health SP, indicating the integral role played by groups.54 These included the two most concerns over legal risks. active and experienced advocacy groups, The six-year period for this legislation the Smokefree Coalition and ASH New compares with less than one year for the law Zealand. Advocates interviewed for this banning tobacco advertising in New Zealand study in June 2015 were concerned the (1990) and less than three years (2001–2003) Health Ministry’s realignment would for the next major tobacco legislation. affect their funding to perform adequate While the trade and health literature has advocacy operations (Supplementary Table primarily focused on the direct effects of 1).PS,EC,LR,SE Some interviewees argued trade agreements on public health policies,7 this realignment prevented and in some this paper demonstrates that trade agree- sense ‘silenced’ the public health voices

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 30 www.nzma.org.nz/journal ARTICLE

ments can also have indirect effects by outset, and emphasised the public health disrupting the policymaking process. The importance of SP. This boldness can be ability of transnational corporations to use attributed to strong leadership in these two the mere threat of challenging a public countries, but research in both cases indi- health proposal in international court forces cates that there were signifi cant differences governments at a minimum to consider across party lines.8,57 trade and investment law into the deci- Policy implications sion-making process.11 This can create legal The tobacco companies’ threats to New risk and uncertainty for governments. Zealand highlight the industry’s long- This paper broadens the regulatory chill standing fear of New Zealand’s tendency literature by examining the chilling effect to adopt policies similar to Australia’s,12 in terms of delay (time elapsed between and its fear of diffusion of best practices introducing and enacting legislation) and globally.59 Internationally, such threats help its impact on public health. In comparison explain the slow diffusion of pictorial health to the timeframes of other early adopters warnings exceeding 50% of the package, and of SP—Australia (18 months), Ireland (22 of SP. months), the UK (35 months) and France As of April 2018, other countries, including (20 months)—New Zealand was by far the Georgia, Hungary, Norway, Romania, slowest (53 months). These political delays Slovenia and Thailand have enacted SP, and have substantial public health effects, Canada, South Africa, Malaysia, Turkey, including slowing the diffusion of best India, Panama, Brazil and Chile have practices and delaying the effect of SP on announced plans to introduce SP.60,61 Health smoking cessation and initiation. They also advocates and lawyers should inform delay the reduction of government health governments, especially those that may have expenditures and tobacco industry sales. similar legal concerns, about the growing In February 2016, the Australian legal certainty of implementing these Department of Health released its Post-Im- policies to avoid unnecessary delays in the plementation Review of SP, reporting regulatory process. Advocates can highlight signifi cant health gains after two years of the Australian,62 UK,63 French64 and Indian65 implementation, including delaying youth High Court decisions to uphold strong smoking initiation from 15.4 years to 15.9 packaging and labeling policies. Advocates years. Within the reduction of smoking can also highlight an international trade prevalence over the two years from 19.4% tribunal ruling in favor of Uruguay’s strong to 17.2%, 0.55 of the drop was attributed to tobacco packaging and labeling laws, also SP. 56 The Department of Health estimated SP based on similar intellectual property, would generate health costs savings of $273 expropriation of trademark property and million over 10 years.56 trade law, which can be applicable to SP The New Zealand case also provides and to wider public health policies.8 On 5 insight into how various political parties May 2017, news sources reported that the may react to industry legal threats and be WTO dispute panel’s interim report upheld more susceptible to regulatory chill. As in the Australian SP laws.66 Although the fi nal Australia,57 tobacco regulations may share ruling is expected in 2017, as of April 2018, bi-partisan support, but the intersections it has not been made public. Given these of trade and health, and trade and tobacco favorable legal rulings governments should can be a dividing issue along party lines. ignore any ‘wait and see’ arguments by the Some MPs within centre-right and right industry and their allies. parties evoked industry legal concerns tied The New Zealand case also points to the to SP, and those in centre-left and left parties importance of non-government funding rejected these arguments. The case of New resources in supporting tobacco control Zealand was similar to the UK,58 where the advocacy efforts,67–69 including legal centre-right party leadership was cautious, expertise and support to help shape the and delayed enacting SP. In contrast, the Government’s reaction and response to the centre-left and left leadership in Australia industry legal threats. Due to their reliance and Uruguay respectively was bold in on government funding, primarily from the rejecting the industry legal threats from the Health Ministry, some New Zealand health

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 31 www.nzma.org.nz/journal ARTICLE

advocacy groups may have been constrained undue delays, the Government runs the risk in challenging the Government’s decision of not fulfi lling its commitment to becoming to delay SP. Although New Zealand has a smokefree by 2025. As of April 2017, it well-established tobacco control network, it appears that the requirement for SP will not lacked the extent of domestic legal expertise be implemented until June 2018.74 or international legal support found in other contexts to counter the industry’s Limitations pressure.8,57 Sustainable funding for inde- Cabinet politicians and some offi cials from pendent advocacy and legal assistance can MFAT declined requests to be interviewed possibly help minimise legal fears and limit for this study. Also, some information the effects of regulatory chill. was withheld under Offi cial Information Although New Zealand has fi nally Act provisions in the Regulatory Impact enacted SP, health advocates should expect Statement and the reports to Cabinet, continued industry interference during limiting a complete understanding of how 70,71 the implementation phase. Delays to the Key administration responded to the progress in Georgia are reported to be due tobacco industry legal threats. to industry interference,72 and in Thailand the Ministry of Commerce has raised trade concerns and it appears the Government is Conclusions waiting until the results of the WTO dispute The New Zealand case illustrates how the with Australia before moving forward.73 threat of a potential international lawsuit In Uruguay8 and Australia,57 following the can create a chilling effect by delaying enactment of strong packaging and labeling public health policies. Other countries intro- regulations, tobacco companies sued in ducing or implementing similar policies international courts, almost forcing the should learn from these experiences and Uruguay Government to weaken its regula- take steps to proactively avoid unnecessary tions. Unless the New Zealand Government political delays that have a profound impact issues the necessary regulations without on public health.

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 32 www.nzma.org.nz/journal ARTICLE

Appendix

Supplementary Table 1: Key for interviewees interviewed in New Zealand in June 2015.

Name of Interviewee Initials for Date of Location of interview interviewee interview Anonymous interviewee at Ministry AI 18 June 2015 Wellington, New Zealand of Foreign A– airs and Trade

Barbara Stewart BS 18 June 2015 Wellington, New Zealand

Chris Bullen CB 11 June 2015 Auckland, New Zealand

David Shearer DS 18 June 2015 Wellington, New Zealand

Louise Delany LD 16 June 2015 Wellington, New Zealand

Edward Cowley EC 10 June 2015 Auckland, New Zealand

Fletcher Tabuteau FT 17 June 2015 Wellington, New Zealand

George Laking GL 12 June 2015 Auckland, New Zealand

Ian Lees-Galloway ILG 17 June 2015 Wellington, New Zealand

James Shaw JS 16 June 2015 Wellington, New Zealand

Jane Kelsey JK 12 June 2015 Auckland, New Zealand

Kevin Hague KH 17 June 2015 Wellington, New Zealand

Louisa Ryan LR 10 June 2015 Auckland, New Zealand

Louisa Wall LW 18 June 2015 Wellington, New Zealand

Marama Fox MF 16 June 2015 Wellington, New Zealand

Matthew Everett ME 18 June 2015 Wellington, New Zealand

Prudence Stone PS 16 June 2015 Wellington, New Zealand

Robert Beaglehole RB 11 June 2015 Auckland, New Zealand

Scott Simpson SS 18 June 2015 Wellington, New Zealand

Shane Bradbrook SB 17 June 2015 Wellington, New Zealand

Simon O’Connor SO 16 June 2015 Wellington, New Zealand

Stephanie Erick SE 10 June 2015 Auckland, New Zealand

Te Ururoa-Flavell TUF 17 June 2015 Wellington, New Zealand

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 33 www.nzma.org.nz/journal ARTICLE

Supplementary Table 2: Standardised plain packaging (SPP) policy process in New Zealand (2010–2016).

Event Time- Response Key statements frame to tobacco industry trade threats

Māori Af- November Acknowledged Committee: “Tobacco companies have indicated they will legally challenge the plain packaging proposal. Imperial Tobacco told us that banning branded fairs Select 2010 the threat but packaging was an infringement of their intellectual property, and they along with two other tobacco companies in New Zealand, opposed the move.”17 Committee did not address Recom- the risks mendation

Cabinet November Same Prime Minister John Key: “There are lots of things we need to consider—I wouldn’t say it’s a slam dunk by any chance that plain packaging will take place review and 2010 but nor would I rule it out. It really is, genuinely, a true consultation period. As the National Party, we haven’t made the decision yet about whether we would proposal -April 2012 support that any further.”22 (17 Trade Minister Tim Groser: “I think it’s getting a bit ahead of the play here because there are some complexities around this. Plain packaging could remove months) the tobacco companies’ intellectual property. We need to listen carefully, especially to other companies that would be very concerned if we were setting a precedent on this. That might actually go against our own interests. We know what the real target is, but we need to consult the public and then we’ll need to have some very careful decisions to make sure that if we are going to move forward with legislation in this area, is properly designed to deal with those legitimate concerns. I’m really thinking outside tobacco.”23

Ministry of July 2012– Same Health Ministry consultation report: “Areas that submitters considered required attention in the RIS [Regulatory Impact Statement] included the need Health con- November to…assess the actual impact of a WTO challenge, and that this should be focused broadly on the impacts for all of New Zealand’s traded products (not just sultation 2012 tobacco).”29 (4 months)

Cabinet November Acknowledged Regulatory Impact Statement (11/24/12): “The Ministry of Foreign A– airs and Trade (MFAT) considers that there is a reasonably high risk that if New Zealand reports 2012–De- high risk of po- implements plain packaging legislation, a World Trade Organization (WTO) dispute settlement case or investment arbitration may be brought against New and formal cember tential litigation, Zealand. There is also the potential for challenges to be brought under regional or bilateral trade and investment agreements, particularly those containing introduc- 2013 and estimates of investor-state dispute settlement clause. If a legal challenge was mounted against New Zealand by a tobacco company in relation to alleged breaches of tion (13 trade challenges international investment agreements, the remedy sought would include payment of compensation. Any claim for compensation would be based on the loss months) in value of the company’s investments including its trademarks. The potential loss to tobacco companies, if any, is presently unable to be quantified and the consultation process was not able to shed any further light on this matter. However, it is expected that data will emerge from Australian disputes that will be useful in quantifying any potential losses.”30 Cabinet paper (11/27/12): “There is a further risk of an international arbitration challenge from tobacco companies under bilateral investment treaties, such as that faced by Australia from Philip Morris Asia under Australia’s bilateral investment treaty with Hong Kong. Regardless of the strength of New Zealand’s case, the possibility of international dispute proceedings are a risk for New Zealand and defending them would require significant investment of resources. However, these risks will be significantly mitigated if the Australia disputes conclude prior to the enactment of New Zealand’s legislation. In that regard, it is possible that the WTO cases will conclude in time but the investment arbitration is likely to take a longer period of time…There will also be financial impli- cations for the Government if New Zealand is forced to defend a WTO challenge or international investment arbitration, as happened in Australia’s case. The cost of defending such legal challenges is not known at this stage, but has been estimated to be in the order of $1.5 million–$2 million for a WTO challenge and $3–6 million for an investment arbitration…If necessary, New Zealand could delay the making of regulations until the Australia cases conclude and certainty regarding WTO legal implications is obtained.”31 Cabinet paper (8/21/13): “Once the bill is introduced, its passage through the House can adhere to standard timelines. This allows time for greater legal cer- tainty over Australia’s plain packaging disputes at the World Trade Organization to emerge. As previously agreed, enacting the legislation, or at least bringing it into force through the subsequent regulations, could be delayed if necessary…Cabinet also noted that: the risk of international legal proceedings being brought against New Zealand under trade and investment agreements remains, but that greater legal certainty may be evident by the time that legislation is enacted in New Zealand if World Trade Organization (WTO) disputes against Australia advance in good time. If necessary, the enactment of the legislation or the making of regulations could be delayed until the Australian cases conclude and certainty regarding WTO legal implications is obtained.”33

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 34 www.nzma.org.nz/journal ARTICLE

Supplementary Table 2: Standardised plain packaging (SPP) policy process in New Zealand (2010–2016) (continued).

Parliament 11 Febru- A majority of Hon. Tariana Turia (Māori): “We are convinced that plain packaging is a really important step on our path to being a smoke-free country by 2025, and that it First Read- ary 2014 MPs rejected ar- will stack up against our World Trade Organization obligations. That is why we are pushing forward to take the legislation through the parliamentary processes ing of Bill guments against without delay.” SPP and a few “New Zealand takes all of its international obligations seriously. Our plain packaging regime has been developed to be consistent with our trade obligations, and MPs addressed our approach to negotiating new trade agreements continues to protect our ability to take public health measures such as plain packaging. The agreements and concerns treaties can, and should, work together to boost both international trade and public health, and this is a good example of where we can achieve both objectives.” “Although the tobacco industry may have laid down a threat if this legislation is passed, my message to it is that our country has a sovereign right and a legal right to protect its citizens. I am firmly of the opinion that it is not for any tobacco company to be telling us what we should be doing in our own land. Five thousand New Zealanders die from smoking a year, and that death toll places a responsibility on every politician to pass legislation in our land that will help save lives and increase well-being—legislation that makes a tangible, enduring impact on the lives of the people of this country. I commend this bill to the House for its first reading.”35 Ian Lees-Galloway (Labour): “E– ectively, the Government gets to decide when this legislation comes into force. The reason for that, of course, is concerns around being sued by the tobacco industry as a result of a potential breach of trade agreements. The real concern is that the Trans-Pacific Partnership will foist upon New Zealand rules and regulations that stop us from doing exactly this, which is to legislate in the best interests of the public health of New Zealanders. We must be vigilant. We must be vigilant and ensure that any trade agreements we sign up to do not allow us to fall into that trap. We are watching Australia closely, but I want New Zealanders to understand that the agreement that Australia has with Hong Kong was poorly dra´ ed in this area and le´ Australia exposed to the type of litigation that it is facing. New Zealand’s trade agreements, generally speaking…our right—our sovereign right—to legislate in the interests of the public health of New Zealanders. New Zealand is a sovereign nation that ought to be able to say that we do not accept that 5,000 of our citizens are killed every year by tobacco.”35 Dr Paul Hutchinson (National): “The issues around the World Trade Organization (WTO) are that every country has the sovereign right to protect the health of its people. I do not believe the problem is so much about free trade and the WTO; I believe it is much more about scurrilous tobacco companies colluding with tobacco-producing countries to bring in expensive, delaying court action.”35 Hon. Annette King (Labour): “I would have to say that I am a little disappointed that we have to wait for the passage of this legislation and that we are waiting to see what happens in the Australian court case. I think it is good on the Australians for having the courage to say to those big tobacco companies: ‘Bring it on.’ I am glad that they have got the money to be able to fund their legal—interruption—Well, that is a good point, Mr Banks. He just asked why we are waiting. It is a question that you need to put to the Prime Minister. The Prime Minister wants to wait to see what happens in the court case in Australia. I think the fear is probably that the tobacco companies might then take us to court. Well, I would give them the two-finger salute and say ‘Bring it on.’, because we as a Parliament will first of all want to protect the health of New Zealanders.”35 Kevin Hague (Green): “It is deeply disturbing, therefore, that the Government is proposing to delay the implementation of this bill until such time as the various court cases and actions against the Australian Government are settled…I agree with Dr Paul Hutchison, who said that every nation has the sovereign right to protect the health of its people. I agree with that, and the Greens say that if that sovereign right is threatened, then there is all the more reason for the Government to stand up and protect that sovereign right…Delaying the implementation of this legislation is caving in to the threats, extortion and delaying tactics of an evil industry.”35 Metiria Turei (Green): “We—the country, the Government, the community—are being threatened by the tobacco industry. We saw in today’s paper that there are further threats by the tobacco industry for the consequences of this policy. We are quite right in saying, so be it, bring it on. We are in the job of making good policy for the health and well-being of our country, and none of us make any apologies for that whatsoever.”35 Barbara Stewart (NZ First): “New Zealand would be the second country in the world to approve plain packaging, a´ er Australia, and we are likely to meet the same legal challenges. I know that the New Zealand Herald article in December last year outlined it clearly: ‘New Zealand was also likely to face legal challenges if it followed Australia’s lead, and o– icials have estimated the cost of a legal dispute as between $2 million to $6 million, not including compensation if a case was lost.’ So that is something else to consider.”35 Clare Curran (Labour): “I want to say that the argument that is used by big tobacco—the apologists who pretend that this is a debate about intellectual property rights or removing barriers to trade—is wrong and that that has been proven…the companies decided to fight plain packaging on trade grounds because it provided them a more solid footing than allowing health issues to enter the debate. For this reason, they focused their energies on the Intellectual Property agreements governed by WIPO and the investment protection contained in NAFTA agreements…Despite being told repeatedly by WIPO—that they had no legal basis for their arguments, that there was no legal basis for any of those arguments, and—that their analysis was flawed, the companies persisted in telling the government—and this was Canada—and the public that plain packaging would be inconsistent with international intellectual property protections. Following the industry’s misrepresentation of international trade law, new health ministers in Canada and Australia forsook plain packaging as a tobacco control measure they mistakenly believed to be contrary to their countries’ obligations under international trade agreements. Finally, this battle is moving towards a conclusion. We are seeing it in Australia. We should not be taking notice of big tobacco’s argument that this is an intellectual property argument, because it is not. There is no basis in law for that argument.”35 Hon. Phil Go› (Labour): “I think that the Philip Morris case against the Australian Government is a disgrace. The Australian authorities tell me that they will suc- ceed in that case. We should not lack the courage to confront the vested interests that promote for their own material benefit the peddling of tobacco as a lethal product. We should not be frightened to confront them. We should not be frightened to bring in this legislation on the date that we consider appropriate and to take on those corporates, because we would have the support of the World Health Organization. We would be aligned with the Framework Convention on Tobacco Control. That has been passed internationally by a responsible body, and I do not believe for a moment that another international body, the World Trade Organi- zation, would in the end defend the right of companies to kill people with their products. It just does not stack up. It is not credible. I support this bill. I commend those with the courage to vote for this bill now, and I urge the Government to bring it into e– ect as soon as possible so we can stop that last bastion of promotion of a lethal product by the vested interests of big tobacco…They may pretend that the debate is about intellectual property. They may pretend that the debate is about removing barriers to trade. I am a believer in reasonable protection for intellectual property and I am a strong believer that we should remove barriers to trade, but neither argument stacks up to defend the promotion of a product that kills people if used as the manufacturer intends.”35 Hon. John Banks (National): “I ask my Māori Party and National Party colleagues to carefully consider the precedent they will set with this bill. This bill guts the intellectual property rights of tobacco companies. Some will ask: well, who cares? But do we want to gut the intellectual property rights of KFC or Red Bull sugar drinks? KFC and Red Bull sugar drinks are putting this country’s level of obesity up at the top of the OECD. They help to contribute to that. It may be seen as a long bow, but the removal of intellectual property rights to the names and brandings of their products from tobacco companies without compensation is wrong, because which international company selling products that are bad for our health will be the next target? The State is e– ectively seizing their property because it does not like the health e– ects of their still lawful business. It is still a lawful business.”35

Media 11 Febru- Some MPs Hon. Tariana Turia (Māori): “While the tobacco industry may have laid down a threat that if this legislation is passed [it will be challenged] my message to statements ary 2014 rejected argu- them is that our country has a sovereign right and a legal right to protect its citizens. I am firmly of the opinion that it is not for any tobacco company to be in response ments against telling us what we should be doing in our own land.”36 to First SPP and some John Banks (ACT): “I don’t believe the State should seize property rights from legitimate companies selling legitimate products.”36 Reading MPs addressed Prime Minister John Key: “I don’t really see the point in us finally passing the legislation until we see exactly what happens in the Australian court case. We concerns have a slightly di– erent system, but there might just be some learnings and if there are learnings out of that, it would be sensible to potentially incorporate those in either our legislation or avoid significant costs.”36 David Cunli› e (Labour): “If we have a legitimate health regulatory policy step, then we should pursue it in the public interest. The Government should not be running scared of tobacco interests because they’re worried about being sued.”36

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 35 www.nzma.org.nz/journal ARTICLE

Supplementary Table 2: Standardised plain packaging (SPP) policy process in New Zealand (2010–2016) (continued).

Ministry 18 June • Rejected P.7 “The Government has announced that it wishes to take account of the implications of Australia’s legal cases at the WTO before deciding to pass the Bill. The of Health 2014 industry Government is confident that tobacco plain packaging can be implemented in a way that is consistent with trade agreement obligations, and New Zealand report to legal ar- is supportive of Australia’s defense of the challenges it is facing at the WTO. However, the timing of these international legal processes is beyond the Govern- Health guments ment’s control. The Bill is now likely to become a matter for the next Parliament to consider. If the WTO process progresses su– iciently or if the international Select • Acknowl- litigation risks are reassessed, it is possible the Bill could be passed early in the term of the new Parliament. Equally the passage of the Bill may be significantly Committee edged delayed, if that is found to be necessary.”38 govern- P.14 “The weight of expert legal opinion was that the international legal challenges against plain packaging were unlikely to succeed.”38 ment P.19 “The Bill provides for regulations to be promulgated that will significantly limit tobacco companies’ ability to use their trade marks on tobacco packaging. may However, any such restrictions would be in accordance with domestic and international law.”38 delay P.20 “O– icials consider that plain packaging will be shown to meet its intended objective and that it does not impair freedom any more than necessary for the passage achievement of public health objectives. This view has been confirmed by the Ministry of Justice.”38 of the P.25 “O– icials agree with the submissions of the academic and NGO legal experts that the Bill is consistent with New Zealand’s WTO obligations. Tobacco Bill company Philip Morris’s submission, which annexes a report by Professor Christopher Gibson, is the only submission by an opponent of the Bill that attempts a thorough analysis of the compatibility of tobacco plain packaging measures with New Zealand’s WTO obligations. It is noteworthy that Professor Gibson’s report does not conclude that the Bill violates the TRIPS Agreement or the TBT Agreement…The WTO analysis submitted by the other opponents of the Bill is incomplete and appears selective, and those submitters did not provide clear evidence to support their claims. Those submissions failed to refer the Commit- tee to relevant WTO jurisprudence that does not support their interpretation of New Zealand’s WTO obligations.”38 P.26 “O– icials agree with the academic legal expert submissions that indicate tobacco plain packaging is consistent with New Zealand’s investment obliga- tions under the trade and investment agreements that New Zealand is a party to. Tobacco plain packaging is nondiscriminatory and is a legitimate exercise of sovereign regulatory power that restricts certain uses of trade marks in order to protect public welfare, namely public health. Philip Morris’s two paragraph submission on this issue focuses on the risk of litigation rather than providing analysis or evidence to support their view. BAT alleges plain packaging violates New Zealand’s investment obligations, but the analysis is incomplete and BAT does not provide credible evidence to support their claims.”38

Parliament 5 August Rejected “The bill would not have any e– ect on intellectual property rights to register, own and enforce trademarks and copyright in designs; it is only the use of trade- Health 2014 industry legal marks and copyrighted designs as promotional devices on tobacco products and packaging that would be controlled.”39 Select arguments Committee report

Parliament November A few MPs Prime Minister John Key: It was waiting, and I think the view I initially took was, given Australia was in the middle of this court case it probably didn’t make Second 2014– complained the sense for us to embark on that, and then potentially face exactly the same costs for the taxpayer in defending another legal action. Last year I asked for advice Reading June 2016 threats delayed on that matter, and the advice I got back was that they felt we were on very firm ground and didn’t feel there was really any issues. A number of others have of Bill and the process moved on plain packaging and were doing so without court cases being brought against them. We’re feeling a lot more confident about that and the bill’s now statements progressing through and it’s my expectation it will become law at some point.51 to the Hon. Peseta Sam Lotu-Liga (National): Our stance remains the same that it is prudent to await the World Trade Organization decision.46 media Hon. Annette King (Labour): What I will be critical of is the time that it has taken to get this bill here. We are talking almost two years—two years waiting to pass a piece of legislation that tightens the screws on tobacco control in New Zealand. Why did we wait two years? We waited because the Government refused to be a leader in the fight for tobacco control, with this measure. It wanted to wait to see what happened in Australia, because Australia had the guts to put in place plain packaging. It said: “We are an independent sovereign nation. We will make our own decisions about what we have in public health law.” And they went ahead, they passed their legislation, they brought in plain packaging, and they were sued by the tobacco companies. So rather than say “We are a sovereign nation. We are prepared to stand up for New Zealanders and pass our legislation.”, we sat there wringing our hands and saying: “We need to wait and see.”55 Simon O’Connor (National): I am conscious, too, of some of the counterarguments that have been put forward around intellectual and property rights. Although I can sort of understand that from one point of view, I think it is really important to make the distinction that, in this case, tobacco product owners still own the property rights, or the intellectual rights; they are just not allowed to use them for, I think, very good reason because, ultimately, the public good overrides it.55 Ian Lees-Galloway (Labour): Has that Government dragged the chain on this legislation and taken every single possible opportunity to slow the process down. So more than three years—more than three years—a´ er the Government first decided it was going to introduce plain packaging, here we are, not passing the bill, not passing it into law, not actually beginning the regime of plain packaging—but here we are at the second reading… I have to say, of course, all of this is tied up in the business of the Trans-Pacific Partnership and free trade and investor-State dispute clauses, and the threat that the tobacco industry continues to make that it will try to sue the New Zealand Government if this Parliament enacts legislation designed to protect the health of New Zealanders. I have to say that if I was an exponent of the Trans-Pacific Partnership—which I certainly am not. But if I was someone who was a backer of the Trans-Pacific Partnership, and if I was someone who was an enthusiast for investor-State dispute settlement clauses, I would be really hacked o– with the tobacco industry right now. This is because the tobacco industry, with its threat to sue the New Zealand Government if this legislation is passed, is actually playing into all the fears people have about what will happen to this country if we sign up to the Trans-Pacific Partnership and if we pass the legislation enabling the Trans-Pacific Partnership. The tobacco industry—the Minister looks confused. The Minister is trying to figure out how this is associated with this bill. He does not do his reading if he does not understand it. This legislation, if it passes, will trigger the tobacco industry’s suing the New Zealand Government under investor-State dispute settlement clauses. That is what people fear—this Parliament not having the sovereign right to legislate in the interests of the public health of New Zealanders because we fear being sued by the tobacco industry, or by any other industry for that matter. Sam Lotu-Iiga, Associate Minister of Health, this legislation—the tobacco industry is threatening to sue the New Zealand Government if this legislation is passed. I am sure the tobacco industry will sue the New Zealand Government if this legislation is passed. That is why the Government has said that it wanted to put this legislation on the back-burner. It is more concerned about business interests. It is more concerned about the tobacco lobby than it is about the public health of New Zealanders.55 David Seymour (ACT): What is in dispute is whether or not smoking cessation is the only value that New Zealand holds. I think that there are a number of other values that are important to New Zealand, including property rights and the right of a business to employ its brand… Nobody wants to defend the tobacco industry, but the principles behind New Zealand’s tradition of property rights, freedoms of trade and the freedom to do as you damn well please so long as you are not harming anyone else are also very important. That is why I am opposed to this bill, which will have a minimal e– ect on smoking behaviour, as demonstrated in Australia. But it is a major step in eroding our tradition of property rights and freedom to trade. That is something that every legitimate business in New Zealand and every business person listening tonight should be very, very concerned about.55 Poto Williams (Labour): This bill also looks at ensuring that the property rights issues that the tobacco companies raised during the submissions are continued, so that the tobacco companies can still have their logos on their tobacco products, but the idea is that they will be standard across all products. All tobacco companies will have the criteria for their logos on packaging determined by this legislation, so that in terms of the percentage or the proportion of the packaging itself the logos of the tobacco companies will be reduced.55

Final September N/A N/A Vote and 2016 Approval in Full House

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 36 www.nzma.org.nz/journal ARTICLE

Competing interests: Eric Crosbie reports grants from National Cancer Institute Training Grant 2T32 CA113710- 11 and Tobacco-Related Disease Research Program Dissertation Research Award 24DT-0003 during the conduct of the study. Author information: Eric Crosbie, Center for Tobacco Control Research and Education, University of California, San Francisco, CA; George Thomson, Department of Public Health, University of Otago, Wellington. Corresponding author: Eric Crosbie, Postdoctoral Fellow, Center for Tobacco Control Research and Education, Room 366 Library, 530 Parnassus, San Francisco, CA. [email protected] URL: http://www.nzma.org.nz/journal/read-the-journal/all-issues/2010-2019/2018/vol-131-no-1473- 13-april-2018/7540

REFERENCES: 1. Mitchell A, Voon T. Impli- 6. Gilmore AB, McKee M. 11. Tienhaara K. Regulatory cations of the World Trade Exploring the impact of chill and the threat of Organization in combating foreign direct investment arbitration: A view from non-communicable on tobacco consumption political science. In: Brown diseases. Public health. Dec in the former Soviet C, Miles K, eds. Evolution in 2011; 125(12):832–839. Union. Tobacco control. Investment Treaty Law and 2. Gleeson D, Lopert R, Reid Feb 2005; 14(1):13–21. Arbitration. Cambridge: P. How the Trans Pacifi c 7. Stumberg R. Safeguards for Cambridge University Partnership Agreement tobacco control: options for Press; 2011: 606–628. could undermine PHAR- the TPPA. Am J Law Med. 12. Crosbie E, Glantz SA. MAC and threaten access 2013; 39(2–3):382–441. Tobacco industry argues to affordable medicines 8. Crosbie E, Sosa P, Glantz domestic trademark laws and health equity in New SA. Defending strong and international treaties Zealand. Health Policy. tobacco packaging and preclude cigarette health Oct 2013; 112(3):227–233. labelling regulations in warning labels, despite 3. Zeigler DW. The alcohol Uruguay: transnational consistent legal advice industry and trade tobacco control network that the argument is agreements: a preliminary versus Philip Morris invalid. Tobacco control. assessment. Addiction. Feb International. Tobacco May 2014; 23(3):e7. 2009; 104 Suppl 1:13–26. control. Mar 23 2017;Online 13. Neumayer E. Do countries 4. Thow AM, McGrady B. First: doi: 10.1136-053690. fail to raise environmental Protecting policy space 9. Alemanno A. The legality, standards? An evaluation for public health nutrition rationale and science of policy options address- in an era of international of tobacco display bans ing “regulatory chill”. investment agreements. after the Philip Morris LSE Research Online. Bulletin of the World judgment. European 2001; 4(3):231–244. Health Organization. Feb Journal of Risk Regulation. 14. Coe J, Rubins N. Regulatory 1 2014; 92(2):139–145. 2011; 2(1):591–599. expropriation and the 5. Crosbie E, Gonzalez 10. Jarman H. Attack on Tecmed case: context and M, Glantz SA. Health Australia: tobacco indus- contributions. In: Weiler T, preemption behind closed try challenges to plain ed. International Invest- doors: trade agreements packaging. Journal of ment Law and Arbitration: and fast-track authority. public health policy. Aug Leading Cases from the Am J Public Health. Sep 2013; 34(3):375–387. ICSID, NAFTA, Bilateral 2014; 104(9):e7–e13. Treaties and Customary

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 37 www.nzma.org.nz/journal ARTICLE

International Law. London: 22. Hartevelt J. Plain-packaged 27. Imperial Tobacco Cameron; 2005:597–667. tobacco no sure thing. New Zealand Limited. 15. Lydgate EB. Biofuels, Dominion Post. 24 July Submission on Smokefree Sustainability, and 2012. Available at: http:// Environments (Tobacco Trade-Related Regula- www.stuff.co.nz/national/ Plain Packaging) Amend- tory Chill. Journal of health/7333603/Plain-pack- ment Bill. Wellington, Int Econ Law. 15 March aged-tobacco-no-sure-thing New Zealand: House 2012; 15(1):157–180. Accessed 23 July 2016. of Represenatives Standing Committee on 16. Harten GV, Scott DN. 23. Hartevelt J. Govern- Health, 9 April 2013. Investment Treaties and ment Not in a Hurry on Available at: http://www. the Internal Vetting of Plain Packaging Scoop parliament.nz/resource/ Regulatory Proposals: A Independent News. 8 en-NZ/50SCHE_EVI_00DB- Case Study from Canada. July 2012. Available at: HOH_BILL12969_1_ Osgoode Legal Studies. 7 http://www.scoop.co.nz/ A385113/6e51b2632a December 2015; 12(6):1–27. stories/PO1207/S00087/ government-not-in-a- 575e49502d37dfcb2f6 17. Māori Affairs Committee. hurry-on-plain-packaging. dd2b9abf68a Inquiry into the tobacco htm Accessed 8 July. 28. Waa AM, Hoek J, Edwards industry in Aotearoa R, Maclaurin J. Analysis of and the consequences of 24. Crosbie E, Eckford R, the logic and framing of a tobacco use for Māori. Bialous SA. Containing tobacco industry campaign Wellington, New Zealand: diffusion: The tobacco opposing standardised House of Representa- industry’s multi-pronged packaging legislation in tives, November 2010. trade strategy to block New Zealand. Tobacco Available at: http:// tobacco standardized control. Sep 30 2016. www.parliament.nz/ packaging. Tobacco control. resource/en-NZ/49DBSCH_ 2017;Under review. 29. Allen and Clarke Policy SCR4900_1/2f- 25. Philip Morris (New and Regulatory Specialists c4d36b0fbdfed73f- Zealand) Limited. Submis- Limited. Submissions 3b4694e084a5935cf967bb sion on the Smoke-free analysis on the proposal to introduce plain packaging 18. Malone RE, Balbach ED. Environments (Tobacco of tobacco products in Tobacco industry docu- Plain Packaging) Amend- New Zealand. Wellington, ments: treasure trove or ment Bill. Wellington, New Zealand: Ministry of quagmire? Tobacco control. New Zealand: House of Health, 21 November 2012. Sep 2000; 9(3):334–338. Represenatives Standing Committee on Health, 28 Available at: http://www. 19. King G, Keohane RO, Verba March 2013. Available at: health.govt.nz/system/ S. Designing Social Inquiry: http://www.parliament.nz/ fi les/documents/pages/ Scientifi c Inference in Qual- resource/en-NZ/50SCHE_ plain-packaging-submis- itative Research. Princeton, EVI_00DBHOH_ sion-analysis-a-c.doc New Jersey: Princeton BILL12969_1_A385101/ 30. Ministry of Health. Regu- University Press; 1994. f049ad52ef7a7391762e5d- latory Impact Statement: 20. Turia T. Speech: Tobacco c770a174403ada5e91 Plain Packaging of Tobacco free Aotearoa Conference. 26. British American Tobacco Products. Wellington, New Zealand Govern- New Zealand. Submission New Zealand: Ministry of ment. 8 November 2012. opposing the Smoke-free Health, 24 November 2012. Available at: http://www. Environments (Tobacco Available at: http://www. parliament.nz/en-NZ/PB/ Plain Packaging) Amend- health.govt.nz/system/ Presented/Papers/8/6/0/49D- ment Bill. Wellington, fi les/documents/pages/ BHOH_PAP21175_1-Gov- New Zealand: House of plain-packaging-ris.pdf ernment-Final-Response- Represenatives Standing 31. New Zealand Cabinet. to-Report-of-the-M-ori.htm Committee on Health, 28 Plain Packaging of tobacco 21. Edwards R, Hoek J, March 2013. Available at: products proposal. Welling- Thomson G. Smokefree http://www.parliament.nz/ ton, New Zealand: Cabinet 2025: patterns and trends resource/en-NZ/50SCHE_ Social Policy Committee, 27 in references to the EVI_00DBHOH_ November 2012. Available smokefree goal in polit- BILL12969_1_A385351/ at: http://www.health.govt. ical speeches and press c1fb0ae0f98dce9f518f832e- nz/system/fi les/documents/ releases. N Z Med J. Jul 18 24c1529a62ebb567 pages/plain-packag- 2014; 127(1398):122–125. ing-cabinet-paper.pdf

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 38 www.nzma.org.nz/journal ARTICLE

32. New Zealand Cabinet. of Representatives 43. Hager N. Dirty Politics: Minute of Decision CAB Standing Committee on How attack politics is Min (13) 4/16: Plain Pack- Health, 27 March 2014. poisoning New Zealand’s aging of Tobacco Products. Available at: http://www. government political Wellington, New Zealand: parliament.nz/resource/ environment. Nelson, New Zealand Government, en-NZ/50SCHE_EVI_00D- New Zealand: Craig 13 February 2013. Available BHOH_BILL12969_1_ Potton Publishing; 2014. at: http://www.health.govt. A384980/99bdba7b- 44. Voxy. Māori Party congrat- nz/system/fi les/documents/ 243cf05a8d3ed- ulates England on plain pages/plain-packaging-cab- 48099b39e961851c94e cigarette packaging. 12 inet-paper-minutes.doc 38. New Zealand Ministry of March 2015. Available at: 33. New Zealand Cabinet. Health. Smoke-Free Envi- http://www.voxy.co.nz/ Tobacco Plain Packaging: ronments (Tobacco Plain politics/m%C3%A Approval for Drafting. Packaging) Amendment 4%C2%81ori-party-congrat- Wellington, New Zealand: Bill Departmental Report. ulates-england-plain-ciga- Cabinet Social Policy Wellington, New Zealand: rette-packaging/5/216091 Committee, 21 August 2013. House of Representatives Accessed 22 March 2015. Available at: http://www. Standing Committee on 45. The Maori Party. Ireland health.govt.nz/system/ Health, 18 June 2014. shows the way with fi les/documents/pages/ Available at: http://www. plain packaging tobacco cabpaper-21aug13.pdf parliament.nz/en/pb/sc/ law. Scoop Independent 34. Steele SL, Gilmore AB, submissions-and-advice/docu- News. 28 February 2015. McKee M, Stuckler D. The ment/50SCHE_ADV_00DB- Available at: http://www. role of public law-based HOH_BILL12969_1_A399457/ scoop.co.nz/stories/PA1502/ litigation in tobacco ministry-of-health-de- S00380/ireland-shows- companies’ strategies in partmental-report the-way-with-plain-pack- high-income, FCTC ratifying 39. New Zealand Parliament. aging-tobacco-law.htm countries, 2004-14. Journal Smoke-free Environments 46. Johnston M. Pressure to of public health. Jun 1 2015. (Tobacco Plain Packaging) bring in tobacco plain-pack- 35. New Zealand Parliament. Amendment Bill: As aging. New Zealand Smoke-free Environments reported from the Health Herald. 2 March 2015. (Tobacco Plain Packaging) Committee. Wellington, Available at: http://www. Amendment Bill-First New Zealand: House of nzherald.co.nz/business/ Reading. Wellington, New Representatives Standing news/article.cfm?c_id=3&- Zealand: House of Repre- Committee on Health, 5 objectid=11410127 August 2014. Available at: sentatives, 11 February 47. Ryan S. Plain packaging http://www.parliament.nz/ 2014. Available at: http:// for cigarettes: ‘It should be resource/en-nz/50DBSCH_ www.parliament.nz/en/pb/ no problem’. New Zealand SCR57000_1/3b- hansard-debates/rhr/docu- Herald. 6 October 2015. 7b2a74d0e52515fe202f- ment/50HansD_20140211_ Available at: http://www. 3cdfaac3c15fd156d4 00000028/ nzherald.co.nz/business/ smoke-free-environ- 40. One News. Q+A: Transcript news/article.cfm?c_id=3&- ments-tobacco-plain-pack- interview with Christopher objectid=11524706 aging-amendment Bishop. Television. 10 June 48. New Zealand Parliament. 2012. Available at: http:// 36. Radio News New Zealand. Hon. Annette King to tvnz.co.nz/q-and-a-news/ Plain packaging bill passes the Associate Minister of transcript-interview-chris- fi rst hurdle. 11 February Health. Wellington, New topher-bishop-4921995 2014. Available at: http:// Zealand, 16 September www.radionz.co.nz/ 41. Thomson GW, Wilson 2015. Available at: http:// news/political/235845/ NA. Lost in the smoke: www.parliament.nz/en/pb/ plain-packaging-bill- tobacco control in New order-paper-questions/writ- passes-fi rst-hurdle Zealand during the 1990s. ten-questions/document/ Accessed 10 March 2015. N Z Med J. Apr 14 2000; QWA_11145_2015/11145- 37. Action on Smoking and 113(1107):122–124. 2015-hon-annette-king-to- Health New Zealand. 42. Wilson N, Thomson G. the-associate-minister Submission on Smoke- Politicians and tobacco 49. Taylor R. Philip Morris free Environments control in New Zealand: Loses Latest Case Against (Tobacco Plain Packaging) a brief three-year audit. Australia Cigarette-Pack Amendment. Wellington, N Z Med J. Aug 26 2005; Laws. Wall Street New Zealand: House 118(1221):U1645.

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 39 www.nzma.org.nz/journal ARTICLE

Journal. 18 December news/article.cfm?c_id=1&- mailonline.com/malaysia/ 2015. Available at: http:// objectid=11663855 article/health-ministry- www.wsj.com/articles/ 55. New Zealand Parliament. hits-pause-on-plain-to- philip-morris-loses-latest- Smoke-free Environments bacco-packaging-plan case-against-australia-ciga- (Tobacco Plain Packaging) Accessed 25 May 2016. rette-pack-laws-1450415295 Amendment Bill-Second 61. Kozak R. Cigarette Maker to Accessed 20 January 2016. Reading. Wellington, Cut Operations as Congress 50. High Court of Australia. New Zealand: House of Debates Tobacco Law. Wall Japan International SA Representatives, 23 August Street Journal. 9 July 2015. and British American 2016. Available at: http:// Available at: http://www. Tobacco Australiasia v. the www.parliament.nz/en/ wsj.com/articles/cigarette- Commonwealth of Austra- pb/hansard-debates/rhr/ maker-to-cut-chile-opera- lia. Canberra, Australia, combined/HansDeb_20160 tions-as-congress-debates- 5 October 2011. Available 630_20160630_20 tobacco-law-1436475357 at: http://www.tobacco- 56. Australia Department of Accessed 30 May 2016. controllaws.org/litigation/ Health. Tobacco Plain 62. High Court of Australia. decisions/au-20121005-jt- Packaging Post-Implemen- Japan Tobacco Internation- intl.-and-bat-australasia-l tation Review. Canberra, al SA v Commonwealth of Accessed 25 May 2016. Australia, 26 Februrary Australia; British American 51. Kirk S. Tobacco plain 2016. Available at: http://ris. Tobacco v Commonwealth packaging likely to be pmc.gov.au/sites/default/ of Australia [2012] HCA law by end of year-John fi les/posts/2016/02/Tobac- 43. Judgement given on 5 Key. Dominion Post. 15 co-Plain-Packaging-PIR.pdf October 2012. Available February 2016. Available Accessed 14 April 2016. at: http://www.tobacco- at: http://www.stuff.co.nz/ 57. Crosbie E, Thomson G, controllaws.org/litigation/ national/politics/76917027/ Freeman B, Bialous S. decisions/au-20121005-jt- tobacco-plain-packag- Advancing progressive intl.-and-bat-australasia-l ing-likely-to-be-law-by- health policy to reduce Accessed 25 May 2016. end-of-year--john-key NCDs amidst international 63. High Court of England Accessed 20 June 2016. commercial opposition: and Wales. R (British 52. New Zealand Parliament. Tobacco standardised American Tobacco and Hon. Annette King to packaging in Australia. others) v Secretary of the Associate Minister of Glob Public Health. Health [2016] EWHC 1169. Health. Wellington, New 2018 Feb 27:1-14. DOI: Judgement given on 19 Zealand, 19 February 10.1080/17441692.2018 May 2016. Available at: 2016. Available at: http:// .1443485 http://www.judiciary. www.parliament.nz/en/pb/ 58. James W. UK government gov.uk/judgments/brit- order-paper-questions/writ- delays decision on cigarette ish-american-tobacco-oth- ten-questions/document/ branding ban. Reuters. ers-v-department-of-health/ QWA_01145_2016/1145- 12 July 2013. Available Accessed 25 August 2016. 2016-hon-annette-king-to- at: http://www.reuters. 64. Le Conseil d’État F. CE, the-associate-minister com/article/us-britain-cig- Japan Tobacco Internation- 53. New Zealand Ministry arettes-packaging-idUS- al SA, Philip Morris France of Health. Realignment BRE96B0P820130712 SA, Republic Technologies of tobacco control 59. Hiilamo H, Crosbie E, France, National Confed- services, 1 March 2016. Glantz SA. The evolution eration of Tobacconists Available at: http://www. of health warning labels and British American health.govt.nz/our-work/ on cigarette packs: the Tobacco France SA. Nos preventative-health-well- role of precedents, and 399117, 399789, 399790, ness/tobacco-control/ tobacco industry strategies 399824, 3998883, 399938, smokefree-2025/ to block diffusion. Tobacco 399997, 402883, 403472, realignment-tobac- control. Jan 2014; 23(1):e2. 403823, 404174, 404381, co-control-services 404394, 23 December 60. Lin MM. Health Ministry Accessed 20 July 2016. 2016. Available at: http:// hits pause on plain www.tobaccocontrollaws. 54. Collins S. Smokefree tobacco packaging plan. org/litigation/decisions/ lobbyists face chop. New Malay Mail Online. 21 fr-20161223-japan-to- Zealand Herald. 27 June March 2016. Available bacco-international-an 2016. Available at: http:// at: http://www.themalay- www.nzherald.co.nz/nz/ Accessed 25 August 2016.

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 40 www.nzma.org.nz/journal ARTICLE

65. Rana P. Inda’s Supreme outside funding, and Not Weaken Them with Court Orders Tobacco business support. Salud Plain Packaging. Forbes. Companies to Comply With publica de Mexico. 30 July 2017 2017. Health Warning Rules. 2017; 59(2):128–136. Available at: http://www. Wall Street Journal. 4 May 69. Crosbie E, Sebrie EM, forbes.com/sites/loren- 2016. Available at: http:// Glantz SA. Strong advocacy zomontanari/2017/07/30/ www.wsj.com/articles/ led to successful implemen- georgia-should-protect-ip- indias-supreme-court- tation of smokefree Mexico rights-not-weaken-them- orders-tobacco-companies- City. Tobacco control. with-plain-packaging/ to-comply-with-health- Jan 2011; 20(1):64–72. - 2ddea2eb6b84. Accessed warning-rules-1462366934 30 August 2017. 70. Uang R, Crosbie E, Glantz Accessed 25 May 2016. SA. Tobacco control 73. Sancelme N. The era of 66. Miles T, Geller M. Australia law implementation plain packaging is coming wins landmark WTO tobac- in a middle-income in Thailand. Vidon. 20 co packaging case. Reuters. country: Transnational February 2017. Available 5 May 2017. Available at: tobacco control network at: http://www.vidon.com/ http://www.reuters.com/ overcoming tobacco en/news/249-the-era-of- article/us-wto-tobacco-aus- industry opposition in plain-packaging-is-coming- tralia-idUSKBN1801S9 Colombia. Glob Public in-thailand.html Accessed Accessed 6 July 2017. Health. Aug 17 2017:1–15. 11 December 2017. 67. Crosbie E, Sosa P, Glantz 71. Crosbie E, Sosa P, Glantz 74. New Zealand Cabinet. S. Costa Rica’s successful SA. The importance of Standardised Tobacco implementation of the continued engagement Product Packaging: Update Framework Convention during the implementation & Next Steps. Wellington, on Tobacco Control: phase of tobacco control New Zealand: Cabinet Overcoming decades of policies in a middle-income Social Policy Committee, industry dominance. country: the case of Costa February 2016. Available Salud publica de Mexico. Rica. Tobacco control. at: http://www.health.govt. 2016; 58(1):62–70. Jan 2017; 26(1):60–68. nz/system/fi les/documents/ pages/standardised-tobac- 68. Uang R, Crosbie E, Glantz S. 72. Montanari L. Georgia co-product-packaging-up- Smokefree implementation Should Protect IP Rights in Colombia: Monitoring, date-next-steps-redacted.pdf

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 41 www.nzma.org.nz/journal ARTICLE

Pertussis vaccination uptake in pregnancy: lessons to be learned from an integrated healthcare approach Emma J Deverall, Benjamin Gilmore, Sam Illing, Roshini Peiris-John

ABSTRACT AIM: To determine the proportion of pregnant women vaccinated with the pertussis booster in the third trimester of their pregnancy, and explore factors influencing coverage. METHODS: A clinical audit was undertaken at Rotorua hospital using electronic databases to determine pertussis immunisation among women who birthed from 25 March to 25 April, 2017 (n=111). Lead maternity carers (LMCs) were surveyed to assess knowledge of the vaccine and explore suggestions to increase vaccination coverage. RESULTS: Only 44% (n=49) of women were vaccinated in 2017. Women 25 years and under, and women from Rotorua were less likely to be vaccinated. A woman not being recalled to the GP for vaccination was the biggest reason for not being vaccinated (n=27). Every woman in Taupo/Turangi was recalled in pregnancy, leading to greater vaccine uptake compared to women in Rotorua. CONCLUSION: Overall, the proportion of pregnant women vaccinated for pertussis continue to be low with coverage being disproportionally lower for younger women. The integrated healthcare approach in Taupo/Turangi has resulted in improved vaccine uptake. Interventions that allow general practitioners, LMCs and primary health organisations to work together can improve vaccination rates of pregnant mothers in New Zealand.

espite being a vaccine-preventable are evident. Māori have the lowest coverage disease, pertussis causes signifi cant rate (62%) with an incidence rate of 734 per Dmorbidity and mortality worldwide, 100,000 in under-one-year-olds, while Pacifi c with 140,000–250,000 cases reported an- peoples have a coverage rate of 73% and the nually from 2012–2016.1 In New Zealand, highest incidence rate in under-one-year- pertussis epidemics occur every 3–5 years. olds (934 cases per 100,000). Although immunisation rates have been Infants are the most at risk of serious increasing, pertussis remains a major public disease complications and death from health issue in New Zealand. During the pertussis, and are best protected by recent outbreak in 2012, over 5,500 cas- maternal immunisation during pregnancy es were reported with over 350 cases per and on-time immunisation during infancy.3,4 month reported from mid-2011 to late-2013. Maternal immunisation is 91−93% effective This outbreak resulted in hundreds of at preventing spread to infants aged under hospitalisations and three deaths of infants, eight weeks old.5 Prenatal immunisation has including two who were too young to be led to a 50% reduction in pertussis among 2 immunised. Ethnic disparities in pertussis infants under age 13 weeks in the UK, and disease rates and vaccination coverage rates has been estimated to reduce total annual

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 42 www.nzma.org.nz/journal ARTICLE

infant cases by up to a third, hospitalisa- If the general practitioner (GP) was listed tions by 39% and deaths by 51% in the US.6,7 as ‘not found’ on hospital records, this was Immunisation during pregnancy is now confi rmed by accessing patient records recommended by national health organisa- within the patient management system tions in several countries, including the US, through the Rotorua Area Primary Health UK, Belgium and New Zealand.5,8,9 Services (RAPHS) or by calling GP practices. In 2013, New Zealand’s Ministry of Health If no GP was identifi ed then the mother was introduced fully funded pertussis immu- excluded from the audit. Women who were nisation for women between 28–38 weeks vaccinated in the DAU were identifi ed by of gestation. However, the Ministry esti- accessing DAU records of LMCs. mates maternal immunisation coverage to The LMCs of the women who gave birth be as low as 13%, with considerable vari- during the study period were surveyed to ation between district health boards.2 A assess views and knowledge of the vaccine recent letter in the New Zealand Medical as well as to explore suggestions to increase Journal10 showed that mothers of over 80% immunisation rates. Information collected of children with pertussis aged less than included knowledge of subsidy, safety, 20 weeks in Auckland had not received appropriate trimester, use of the secure maternal vaccination. Their suggestion was emails provided as an intervention after to promote the effectiveness of the maternal the last audit and reasons for not use, and vaccine to would-be parents, lead maternity recommendations for improving coverage. carers (LMCs) and general practitioners The questionnaire was interviewer adminis- (GPs). tered at the birthing unit at Rotorua hospital An audit conducted at Rotorua hospital or by a telephone call. in 2015 indicated a 35% immunisation Ethical approval for the study was uptake. To improve the vaccination rate received from the Rotorua Hospital Ethics for women in the community, a package of Committee. care was developed to support and educate GPs in their provision of primary care to Results pregnant women. LMCs were offered secure A total of 113 women gave birth within the email addresses to strengthen communi- audit period. Two women were excluded as cation between them and other primary they were enrolled in GP practices outside care providers. Rotorua Maternity Day of the Lakes District Health Board (the study Assessment Unit (DAU) also offered immu- location). This left 111 women eligible for nisation to women using hospital services. the audit. The study reported here aimed to determine the proportion of pregnant women vacci- The overall proportion of pregnant nated for pertussis in the third trimester of women vaccinated for pertussis from their pregnancy following the steps taken to March–April 2017 was 44% (n=49). Those improve coverage in Rotorua, examine vari- 25 years and under, and living outside of ation by selected demographic variables and Taupo/Turangi, were less likely to be vacci- explore factors infl uencing coverage. nated (Table 1). Māori (cf. non-Māori) and nulliparous women (cf. multiparous women) were less likely to be vaccinated, although Methods these differences were not statistically All women who birthed in Rotorua signifi cant. Almost half (49%) of the women hospital from 25 March 2017 to 25 April 2017 (n=45) were vaccinated by a GP. Twelve were included (n=113). The birth registry women (11%) had unknown immunisation (available on the Rotorua electronic health status as they were not enrolled at a practice system) was accessed to extract information and their notes could not be accessed. about the variables of interest: whether the Women in Taupo/Turangi were more likely patient had received pertussis vaccination to be vaccinated at a rate of 76% compared in this pregnancy, age, ethnicity, parity, to 45% in Rotorua. Taupo women have GP and LMC. If they had not received the access to a greater number of interventions vaccine, it was asked if the women had been within an integrated healthcare approach. seen in pregnancy, seen but not recalled, Rather than arrange secure emails for LMCs, declined the vaccine or lost to follow up.

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 43 www.nzma.org.nz/journal ARTICLE

Table 1: Characteristics of study population by vaccination status.a

Total Yes No p-valueb (n=111) (n=49) (n=50) Agec ≤25 years 50 15 (30.0%) 26 (52.0%) 0.006

>25 years 58 34 (58.6%) 21 (36.2%)

Ethnicity Māori 54 20 (37.0%) 28 (51.9%) 0.224

NZ European 32 15 (46.9%) 14 (43.8%)

Other 25 14 (56.0%) 8 (32.0%)

Parity PO 46 19 (41.3%) 22 (47.8%) 0.598

P1+ 65 30 (46.2%) 28 (43.1%)

Locationd Taupo/Turangi 21 16 (76.2%) 5 (22.7%) 0.012

Rotorua 69 31 (44.9%) 38 (55.1%)

aExcludes women with unknown vaccination status (n=12). bChi-square test. cExcludes women with age unknown (n=3), all were not vaccinated. dExcludes women from areas outside of Taupo/Turangi and Rotorua (n=9), only two of whom were vaccinated.

Table 2: Reasons for not being vaccinated.

Not seen in the Seen; no Vaccination Seen; recalled, Unknown pregnancy recall declined (n=3) lost to follow-up (n=3) (n=16) (n=27) (n=1) Age ≤25 years (n=26) 8 15 2 1 -

>25 years (n=21) 8 12 1 - -

Unknown (n=3) - -- - 3

Ethnicity Māori (n=28) 9 14 3 1 1

NZ European (n=14) 2 10 - - 2

Other (n=8) 5 3 - - -

Parity PO (n=22) 6 13 2 - 1

P1+ (n=28) 10 14 1 1 2

Location Taupo/Turangi 2 -- - 3 (n=5)

Rotorua (n=38) 14 21 2 1 1

Other (n=7) - 6 1 - -

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 44 www.nzma.org.nz/journal ARTICLE

Taupo maternity unit takes responsibility for largely unlikely, there may also have been notifying GPs of their patient’s pregnancy. incorrect documentation of immunisation GPs are then able to enter patients for recall by providers who opportunistically vacci- to discuss vaccination, and a nurse attends nated mothers in locations away from their antenatal classes for opportunistic immuni- GP clinics. Despite these limitations, this sation. After-hours vaccination is available study provides useful insights into current at the pharmacy. In Turangi the Community practice and explores opportunities for Child Health Nurse from Pinnacle Midlands improving uptake. Health Network vaccinates pregnant women Previous studies in New Zealand have at a monthly clinic. shown pertussis vaccination uptake to be A woman not being recalled to the GP for infl uenced by mothers’ desire to protect vaccination was the biggest reason for not their baby (96%), following health profes- being vaccinated (Table 2). Every woman in sional advice (84%), awareness of pertussis Taupo/Turangi was recalled in pregnancy. in the community (50%) and being funded Only three women declined the vaccine and (43%).11 Those who did not accept the only one was lost to follow up after recall. vaccine were either unaware of it (73%), There were 24 LMCs responsible for births had safety concerns (68%) or were doubtful during the study period; one was on leave at of its effectiveness (56%). Pregnant mothers the time of surveying, leaving 23 eligible. whose midwife recommended vaccination are more likely to be immunised.2 While While all midwives routinely discussed we found all midwives routinely discussed vaccination with their patients, and under- vaccination with their patients, one in four stood it was subsidised, four (17.3%) were were unsure of its safety and effectiveness. unsure the vaccine was safe, and two (8.7%) Programmes to update midwives on current were unsure if it was effective. knowledge on vaccine safety and effec- None of the midwives used the secure tiveness would be useful. emails provided by Rotorua, as an inter- Clearly, immunisation uptake could vention was put in place after the last audit. be improved by increasing awareness of Reasons for this included: hard to access funded pertussis immunisation during (n=8); not set up (n=7); women referred from pregnancy and by increasing opportunistic GPs (n=3); laborious (n=2); problems with immunisation. Although we found LMCs account (n=2); and cost (n=1). Midwives felt are relatively well informed, it appears that that advertising through local and social the message for all pregnant women to be media (n=5), along with patient education offered pertussis vaccination in pregnancy resources (n=7) may help improve vacci- is not being comprehensively translated nation rates. into practice with variations found between localities. Discussion In Rotorua, unvaccinated women Overall, the proportion of pregnant were commonly seen by their GP, yet not women receiving pertussis booster vacci- recalled for vaccination. It is not possible to nation within Rotorua DHB continues to determine if this was a missed opportunity be low, with coverage disproportionally for vaccination, as data on gestational age lower for younger women, and little when seen or the purpose of the visit was improvement in coverage from 2015 to not recorded. 2017. The proportion of pregnant women Overall, there is a need for more focus receiving pertussis vaccination in Taupo/ on improving delivery of vaccines to young Turangi, however, was greater than women pregnant women. The signifi cantly higher in Rotorua. rate of pertussis vaccination for women This was an audit consisting of a small living in Taupo may be due to the way sample from a secondary hospital, and care in which multiple healthcare providers must be taken when interpreting the results. bring vaccination to the community, rather The main limitation of the audit is that 11% than any attempt to bring the women of the original population could not have to the vaccine. Key components of this their immunisation status identifi ed due to programme are: not being registered at a GP clinic. Although

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 45 www.nzma.org.nz/journal ARTICLE

1. Notifying GPs of pregnancy: fi nding The availability of a greater number new ways to promote communication of interventions within an integrated between GPs and LMCs, such as the healthcare approach in Taupo/Turangi has Taupo-maternity unit letter, could also led to a higher uptake of vaccination among work; pregnant women compared to women in 2. Vaccinations available in antenatal Rotorua. It is essential that we continue to clinics and; develop and foster the relationships and resources that allow GPs, LMCs and primary 3. Opportunistic vaccination at phar- health organisations to work together to macies and parent education sessions. improve the vaccination rate of pregnant Community-based opportunistic vacci- mothers. Adoption of the integrated nation in Rotorua could include local healthcare approach, as found in Taupo/ maraes. Turangi, could help improve the vaccination rate of all pregnant mothers in New Zealand.

Competing interests: Nil. Acknowledgements: The authors would like to thank Sue Taft, Child Health and Immunisation Coordinator, Rotorua Area Primary Health Services (RAPHS) for assistance with data collection and Fiona Lafferty, Community Child Health Nurse, Pinnacle Midlands Health for her contribution on how the process works in Taupo. We also acknowledge the lead maternity carers and GP practices in the Lakes DHB region for their support. Author information: Emma J Deverall, Obstetrician, Lakes District Health Board, Rotorua; Honorary Senior Lecturer, Department of Obstetrics and Gynaecology Faculty of Medical and Health Science, University of Auckland, Auckland; Benjamin Gilmore, Medical Student, Faculty of Medical and Health Sciences, University of Auckland, Auckland; Sam Illing, Rural Hospital Trainee, Bay of Plenty District Health Board, Whakatane; Roshini Peiris-John, Senior Lecturer, Faculty of Medical and Health Sciences, University of Auckland, Auckland. Corresponding author: Emma Deverall, Women Child and Family Corridor, Rotorua Hospital, Private Bag 3023, Lakes District Health Board. [email protected] URL: http://www.nzma.org.nz/journal/read-the-journal/all-issues/2010-2019/2018/vol-131-no-1473- 13-april-2018/7541

REFERENCES: 1. World Health Organisa- 3. Nowlan M, Turner N, 5. Dabrera G, Amirthalingam tion. Vaccine-preventable Kiedrzynski T, et al. G, Andrews N, et al. A disease monitoring system Pertussis control strategies: Case-Control Study to 2017. Available at: A consistent approach for Estimate the Effectiveness http://www.who.int/ New Zealand: Synopsis of of Maternal Pertussis immunization/monitor- Ministry of Health Work- Vaccination in Protect- ing_surveillance/data/ shop, April 2015. NZ Med ing Newborn Infants gs_gloprofi le.pdf?ua=1 J. 2016; 129(1433):78–85. in England and Wales, Accessed: 8 August 2017. 4. Healy CM, Munoz FM, 2012–2013. Clin Infect 2. Ministry of Health. Pertus- Rench MA, et al. Prevalence Dis. 2015; 60(3):333–7. sis Control Strategies 2015: of pertussis antibodies in 6. Amirthalingam G, Andrews A consistent approach for maternal delivery, cord, N, Campbell H, et al. New Zealand. Wellington: and infant serum. J Infect Effectiveness of maternal Ministry of Health, 2015. Dis. 2004; 190(2):335–40. pertussis vaccination

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 46 www.nzma.org.nz/journal ARTICLE

in England: an obser- at: http://apps.who.int/ 10. Reynolds G, Grant N, vational study. Lancet. immunization_monitoring/ Thornley S, Hale M. 2014; 384(9953):1521–8. globalsummary/coun- Low uptake of maternal 7. Terranella A, Asay GRB, tries?countrycriteria%5B- vaccination in notifi ed Messonnier ML, et al. country%5D%5B%5D=N- pertussis cases aged less Pregnancy Dose Tdap ZL&commit=OK# than 20 weeks. NZ Med and Postpartum Cocoon- Accessed: 7 August 2017. J. 2017; 130(1449):72–4. ing to Prevent Infant 9. World Health Organisa- 11. Hill L. Factors Infl uencing Pertussis: A Decision tion. Number of reported Women’s Decisions about Analysis. Pediatrics. mumps cases, New Zealand Having the Pertussis 2013; 131(6):E1748–E56. 2016. Available at: http:// Containing Vaccine during 8. World Health apps.who.int/immu- Pregnancy [Masters of Organisation. WHO nization_monitoring/ Health Science]. Christ- vaccine-preventable diseas- globalsummary/JPG/ church, New Zealand: es: monitoring system. 2017 NZLMUMPS_Cases.jpg University of Otago; 2015. global summary. Available Accessed: 7 August 2017.

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 47 www.nzma.org.nz/journal ARTICLE

Diagnosis of abdominal tuberculosis in Christchurch New Zealand: a case series James MM Bevin, Simon C Dalton, Chris J Wakeman, Will RG Perry

ABSTRACT AIM: Abdominal tuberculosis presents with non-specific symptoms, including generalised abdominal pain. Prompt and accurate diagnosis is critical to improving outcomes and avoiding complications. We conducted a retrospective review of cases of abdominal tuberculosis presenting to Christchurch Hospital to explore the epidemiology, clinical features and diagnostic modalities used. METHOD: Cases were identified by searching for relevant ICD discharge codes from January 1996 to January 2016. Data on age, clinical presentation, investigations and microbiological results were obtained. RESULTS: There were 20 patients diagnosed with abdominal tuberculosis over the study period. The median age was 34. Thirteen patients were male (65%), seven female (35%). The majority (11) were from Asia (predominantly India), five were African, and three were New Zealand Europeans. Abdominal pain was the most common presenting symptom (70%) followed by fevers (50%) and night sweats (50%). The C-reactive protein was elevated in 15 patients (75%), anaemia was found in 11 (55%) and nine had abnormal liver function tests (45%). Abdominal ultrasound (US) and computed tomography (CT) showed generic inflammatory change in all patients in this series (100%). Laparoscopy was undertaken in 10 (50%) patients, all of which had positive laparoscopic biopsies. Ascitic fluid was obtained in nine, with stains for acid-fast bacilli uniformly negative, however three (33%) had mycobacterial growth from culture. Six colonoscopies were performed: in three (50%) culture and/or histology was positive. Three lymph node biopsies and two formal laparotomies were the remaining diagnostic techniques employed with two biopsies and one laparotomy yielding positive results. Overall, of the 20 cases, 15 (75%) were able to be definitively confirmed, with the remaining five treated presumptively for probable abdominal tuberculosis. CONCLUSION: Abdominal tuberculosis is an uncommon presentation at our institution, with an average of one case each year. The typical patient was a young immigrant from Asia or Africa. Diagnostic laparoscopy was the most common and uniformly reliable means of obtaining a definitive diagnosis.

rior to the 1960s, tuberculosis (TB) non-specifi c abdominal pain and a constel- was more commonly found in New lation of systemic features. It lends itself to PZealand than it is today.1 This decline a wide ranging differential that can include in incidence was bought about by increas- infection, infl ammatory bowel disease and ing standards of living, control of bovine malignancy.3 Prompt and accurate diagno- tuberculosis through slaughter of reactive sis leads to earlier tuberculosis treatment, animals, pasteurisation of milk and the which creates advantages not only with re- introduction of anti-tuberculous medica- gard to patient prognosis but also in savings tions. However, it has not been eradicated to the health system. As such, major institu- as anticipated, largely due to rising migrant tions should be well-equipped to deal with populations, poverty, inadequate access abdominal tuberculosis. 2 to healthcare and overcrowded housing. This series aims to establish how cases of In comparison to the pulmonary form of abdominal tuberculosis have both presented the disease, abdominal manifestations of and been diagnosed at Christchurch Public tuberculosis are infrequent and the clinical Hospital and whether this is in line with presentation tends to be ambiguous with current practice elsewhere.

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 48 www.nzma.org.nz/journal ARTICLE

On abdominal ultrasound, TB is suggested Method by the presence of one or more of the A retrospective study of patients admitted following features: lymphadenopathy, to Christchurch Public Hospital from ascites, free fl uid or thickened small bowel January 1996 to January 2016 was carried loops.4 An ultrasound scan of the abdomen out. Patients were identifi ed by code for was performed in 10 of the patients and all discharge diagnosis A183: “Tuberculosis had at least one of the fi ndings consistent of intestines, peritoneum or mesenteric with abdominal tuberculosis. glands”. Data on age, clinical presentation, Abdominal CT was performed in every investigations and treatment were obtained patient in this series and all had some or all from a combination of clinical records and of the following positive fi ndings suggested electronic patient information databases. in the literature: retroperitoneal or mesen- Ethical approval for this study was obtained teric lymphadenopathy, thickened small from the New Zealand Health and Disability bowel, free fl uid or mesenteric stranding.4 Ethics Committees (HDEC). Of the 20 CT scans, only 12 reports contained TB in the differential diagnosis. These 12 Results cases were the patients with known or Twenty patients were diagnosed with potential TB contact in the history. The abdominal TB between 1996 and 2016. other eight cases of abdominal TB (without Thirteen patients were male, seven were reported TB exposure) had only malig- female. The median age was 34. Six patients nancy and/or infl ammatory bowel disease were of Indian origin, fi ve Asian, fi ve considered in the radiological reports. African, three New Zealand European and Ascitic fl uid samples were taken in one Melanesian. The three New Zealand nine cases; Ziehl-Neelsen (ZN) stains were Europeans were the only members of our uniformly negative and culture was positive cohort who had not recently immigrated. in only three. Colonoscopy was performed Presenting features in six patients. In four of these infl ammation Abdominal pain was the predominant and ulceration was found. Histology showed presenting feature. Fourteen patients granulomatous infl ammation. Culture of reported a history of generalised abdominal the biopsy was positive for Mycobacterium pain. Fevers and night sweats were the next tuberculosis in three and negative in one. most common complaint in 10 patients, The other two colonoscopies were reported followed by associated weight loss in eight, as normal. diarrhoea in fi ve and anorexia in four Typical macroscopic fi ndings of abdominal patients. One patient presented with urinary TB include diffuse involvement of the incontinence, which was later found to be visceral and parietal peritoneum, white from an infl ammatory mass making contact ‘miliary’ nodules, mesenteric lymph nodes, with the bladder. ‘violin string’ fi brinous strands, omental On examination, only 10 of the 20 patients thickening and small bowel lesions.5 Lapa- had a generalised tender abdomen. Of these, roscopy was performed in 10 of the 20 six abdomens were also noted to be grossly cases and nine patients had these macro- distended. Only three patients had other scopic fi ndings. The remaining laparoscopy examination fi ndings, including a mass was converted to laparotomy due to small found on palpation of the abdomen. bowel injury from insertion of the trocar. In this case the appearance of the bowel was Six patients gave a history of pulmonary suggestive of tuberculosis. All 10 patients TB, however no information was available were later diagnosed with tuberculosis; regarding prior treatment for this. Another confi rmed on histology and culture from six reported possible contact with TB in the intra-abdominal lymph nodes or omental/ past. peritoneal tuberculous nodules. Investigations Three cases had image-guided lymph A CRP over 5mg/L (75%) was the most node biopsies. Central low attenuation and consistent fi nding. Other fi ndings including peripheral enhancement were consistent anaemia (55%), elevated transaminases imaging fi ndings. Two cases were of (45%) and elevated white cell count (10%) abdominal nodes; the fi rst demonstrated were less frequent. granulomatous infl ammation on cytology

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 49 www.nzma.org.nz/journal ARTICLE

and was both culture and PCR positive for M. tuberculosis, whereas the second had Discussion limited cellular material and was culture Tuberculosis remains a global scourge negative. The fi nal image-guided lymph and unfortunately the symptoms can be node biopsy was from a cervical node non-specifi c.2 with granulomatous infl ammation seen on The clinical presentations and positive cytology and subsequent growth of M. tuber- fi ndings of investigations from various culosis from culture. The latter two cases published case series are summarised in were treated as presumed abdominal TB Table 1. The most common symptom of based on consistent imaging fi ndings in the abdominal pain in our series was also the terminal ileum and peritoneum. leading symptom in two other series from All 20 cases had specimens sent for sent the UK and Saudi Arabia.6,8 The number of for histology/cytology—all but one of these presentations with abdominal pain in every demonstrated granulomatous infl ammation. series reviewed was similar to the 70% of Furthermore, all cases had TB culture patients in our series—the only exception to requested—M. tuberculosis was isolated this being a publication from India in which from culture in 14 cases. In one other case only 29% had presented with abdominal M. tuberculosis was detected by PCR with pain.7 A Turkish study listed weight loss no subsequent growth from culture. All 14 as the most common presentation fi nding positive cultures were uniformly susceptible in 81% of cases—nearly double our expe- to standard anti-tuberculous medications rience.5 Furthermore, the series from the UK with no drug resistance identifi ed. by Rai et al had a high proportion of weight loss, whereas Muneef et al in Saudi Arabia Overall, of the 20 cases, 15 were able to be found a lower proportion more in line with defi nitively confi rmed, with the remaining our fi ndings.6,8 fi ve treated presumptively for probable abdominal tuberculosis. Night sweats and fevers were never the most frequent presenting symptom at any

Table 1: Comparison of case series for positive fi ndings on investigations and clinical presentations.

S Rai Krishnan et al Uzunkoy et Muneef et Islam et al Bevin et al ’95–’016 ’99–’0577 al ’96–’035 al ’84–’978 ’08–’104 CPH ’96–‘15 Number of cases 24 41 11 46 81 20

Country UK India Turkey Saudi Arabia South Africa New Zealand

Weight loss 87% 81% 68% 40%

Abdominal pain 88% 29% 72% 70% 70%

Night sweats/fever 55% 36% 70% 50%

Elevated CRP/ESR >90% 78% 75%

Elevated white cell count 33% 10%

Anaemia >90% 63% 55%

Mantoux 22% 18% 27%

Abdomen x-ray 21% 0%

USS abdomen 32% 100% 100% 100%

CT scan abdomen 55% 100% 100% 90% 100%

Culture ascitic fluid 0% 0% 100% 7% 35% 33%

Laparoscopy/laparotomy 92% 95% 100% 96.4% 94% 100%

Colonoscopy 50%

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 50 www.nzma.org.nz/journal ARTICLE

institution. Half of our patients experi- a defi nitive diagnosis and was successful in enced night sweats and/or fevers and this every case. proportion is similar to all other case series One case at Christchurch Hospital resulted 5,6,8 reviewed. in a small bowel injury during a diag- Overall, the presenting symptoms found nostic laparoscopy–this is the only reported in this series are broadly similar to those adverse outcome from the 10 diagnostic identifi ed overseas. Likewise, other studies laparoscopies in this series. In the liter- have also found the majority of patients to ature, of 208 total laparoscopies reviewed have an elevated infl ammatory marker.6,8 there were 10 adverse outcomes leading to No other study found any diagnostic utility conversions to laparotomy. Two of these in other blood tests, however Rai et al also were for bowel perforation during trocar found over 90% of their cases had anaemia.6 insertions, one for omental haematoma All studies agreed that the blood derange- during trocar insertion, three for intra-op- ments commonly found could not reliably erative bleeding, two for adhesions and two make a rapid diagnosis of abdominal for ‘technical diffi culties’ with laparascopy.4–8 tuberculosis. Laparoscopy in these patients is challenging This is also true of imaging—over 90% and if used indiscriminately could result of cases in three other series had generic in a far greater number of surgical compli- infl ammatory changes on CT or ultrasound cations, particularly as TB infection in the that could not distinguish between differ- abdomen will produce highly vascularised ential diagnoses.4,5,8 adhesions, which are friable and diffi cult to manipulate restricting mobilisation of the Rapid distinction from the major alternate bowel. As such, the presence of adhesions diagnoses of infl ammatory bowel disease will usually indicate conversion to lapa- and malignancy is critical as there is rotomy. Surgical expertise is required when evidence that delaying treatment, even by continuing this operation via laparotomy as as little as 30 days, can have a detrimental these patients are chronically ill and compli- effect on the patient’s prognosis.9 A delay cations are best avoided. of up to eight weeks for microbiological confi rmation from culture before initi- The Canterbury Infl ammatory Bowel ating treatment is therefore undesirable. Disease (IBD) project has previously shown As such a more effi cient alternative means that that Canterbury has high levels of IBD 11 of diagnosis is required without losing the compared to the rest of the New Zealand. specifi city of microbiology. The typical IBD patient is also approxi- mately the same age as someone likely to be The use of laparoscopy and culture of a suffering from abdominal TB. The median surgical specimen is widely accepted as the age of Ulcerative Colitis and Crohn’s was investigation of choice for abdominal tuber- found to be 43.7 and 39.9 respectively.11 culosis.6 The appearance of the bowel in They both present with similar vague laparoscopy is one potential way to presump- complaints and infl ammatory changes tively diagnose tuberculosis. To recognise on imaging. Our data revealed 13 out of the appearance it is important to understand 20 admission notes and radiology reports the mechanism of spread of this disease. The listing abdominal TB in the differential diag- main methods of gastrointestinal infection nosis; in comparison, IBD was listed in 17 of of M. tuberculosis are swallowing infected the 20 cases. It may be that the large burden sputum from active lung disease and local of IBD in our region has led healthcare staff invasion from adjacent viscera.2 As shown in into aggressive and early diagnosis of IBD Table 1, laparoscopy has excellent diagnostic while not fully exploring other potential yield. It is suggested that this is because differential diagnoses. As colonoscopy is a direct observation of the entire peritoneal commonly used diagnostic tool in IBD, this space is feasible and biopsies of specifi c reasoning could also explain why our series nodules can be easily carried out. A review was the only series to use colonoscopy as a of the literature suggests rates ranging from diagnostic method. 92–100% presumed diagnosis on laparoscopy being confi rmed later by microbiology. Our Overall, Christchurch Public Hospital has series showed that diagnostic laparoscopy had a low volume case load of abdominal was the most common means of obtaining tuberculosis with one case per year on

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 51 www.nzma.org.nz/journal ARTICLE

average. Patients have vague presenting Due to there being only one discharge symptoms and generic infl ammatory diagnosis code applying to this set of changes on imaging which makes effi cient patients, it is reasonable to believe we diagnosis of TB diffi cult. The literature indi- have underestimated the sample size by cates that laparoscopy followed by a positive not including those who were not coded microbiological result has the best diag- correctly and those simply misdiagnosed. nostic utility and we found that diagnostic This study is further limited by a small laparoscopy was the most common means sample size and technological advances over of obtaining a diagnosis of abdominal TB in the 20 years of the series which has changed Christchurch with excellent diagnostic yield the accuracy of most microbiological and similar to other published series. biochemical tests.

Competing interests: Nil. Author information: James Bevin, House Offi cer, Auckland City Hospital, Auckland; Simon Dalton, Infectious Diseases Physician, Christchurch Hospital, Christchurch; Chris Wakeman, General Surgeon and Senior Lecturer, University of Otago, Christchurch; Will Perry, General Surgical Registrar, Middlemore Hospital, Auckland. Corresponding author: Dr James Bevin, House Offi cer, Auckland City Hospital, Auckland. [email protected] URL: http://www.nzma.org.nz/journal/read-the-journal/all-issues/2010-2019/2018/vol-131-no-1473- 13-april-2018/7542.

REFERENCES: 1. Das D, Baker M, Calder 5. Uzunkoy A. Diagnosis of Scandinavian Journal of L. Tuberculosis epidemi- abdominal tuberculosis: Gastroenterology. 2001; ology in New Zealand: Experience from 11 36(5):528–32. 1995–2004. The New cases and review of the 9. Voigt M, Trey C, Lombard Zealand Medical Journal literature. World Journal C, et al. Diagnostic value (Online). 2006; 119(1243). of Gastroenterology. of ascites adenosine 2. Horvath K, Whelan R. 2004; 10(24):3647–9. deaminase in tuberculous Intestinal tuberculosis: 6. Rai S, Thomas W. peritonitis. The Lancet. return of an old disease. Diagnosis of abdominal 1989; 333(8641):751–4. The American Journal tuberculosis: the impor- 10. Rasheed S, Zinicola of Gastroenterology. tance of laparoscopy. R, Watson D, et al. 1998; 93(5):692–6. JRSM. 2003; 96(12):586–8. Intra-abdominal and 3. Marshall J. Tuberculosis 7. Krishnan P, Vayoth SO, gastrointestinal tuber- of the gastrointestinal Dhar P, et al. Laparoscopy culosis. Colorect Dis. tract and peritoneum. The in suspected abdominal 2007; 9(9):773–83. Pediatric Infectious Disease tuberculosis is useful as an 11. Gearry RB, Richardson Journal. 1994; 13(4):341. early diagnostic method. A, Frampton CM, et al. 4. Islam J, Clarke D, Thomson ANZ journal of surgery. High incidence of Crohn’s S, et al. A prospective 2008; 78(11):987–9. disease in Canterbury, audit of the use of diag- 8. Al Muneef M, Memish Z, New Zealand: results of nostic laparoscopy to Al Mahm S. Tuberculosis an epidemiologic study. establish the diagnosis of in the Belly: A Review of Infl ammatory bowel diseas- abdominal tuberculosis. Forty-six Cases Involving es. 2006; 12(10):936– 43. Surgical Endoscopy. the Gastrointestinal 2014; 28(6):1895–901. Tract and Peritoneum.

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 52 www.nzma.org.nz/journal ARTICLE

Metronidazole stewardship initiative at Christchurch hospitals—achievable with immediate bene’ ts Sharon J Gardiner, Sarah CL Metcalf, Paul KL Chin, Matthew P Doogue, Simon C Dalton, Stephen T Chambers

ABSTRACT AIMS: To evaluate an antimicrobial stewardship (AMS) initiative to change hospital prescribing practice for metronidazole. METHODS: In October 2015, the Canterbury District Health Board (CDHB) AMS committee changed advice for metronidazole to promote two times daily dosing for most indications, prioritisation of the oral route and avoidance of double anaerobic cover. Adoption of the initiative was facilitated via change in prescribing guidelines, education and ongoing pharmacy support. Usage and expenditure on metronidazole for adult inpatients were compared for the five years pre- and two years post-change. Other district health boards (DHBs) were surveyed to determine their dosing recommendation for metronidazole IV. RESULTS: Mean annual metronidazole IV use, as defined daily doses per 1,000 occupied bed days, decreased by 43% post-initiative. Use of non-IV (oral or rectal) formulations increased by 104%. Total savings associated with the initiative were approximately $33,400 in drug costs plus $78,200 per annum in IV giving sets and post-dose flushes. Twelve of 20 (60%) DHBs (including CDHB) endorse twice daily IV dosing. CONCLUSIONS: In addition to financial savings, reduction in IV doses has potential benefits, including avoidance of IV catheter-associated complications such as bloodstream infections. Approaches to metronidazole dosing vary across DHBs and could benefit from national coordination.

etronidazole is a synthetic nitroim- the treatment of anaerobic infections, there idazole developed in the 1950s to is no consensus on the ideal dosing strategy treat urogenital infections caused for metronidazole administered intrave- M 1 by the parasite, Trichomonas vaginalis. Its nously (IV). Indeed, international guidelines activity against anaerobic bacteria was later on the treatment of intra-abdominal infec- discovered serendipitously in 19622 and now tions in adults endorse a two-fold variation forms the basis for most of its use in hospi- in daily dose (1,000–2,000mg) administered talised patients. at four different dose intervals (6-, 8-, 12- or 7–12 Metronidazole has a unique phar- 24-hourly) (Table 1). An antimicrobial macological profi le that includes rapid stewardship (AMS) perspective is needed to concentration-dependent bactericidal action rationalise these regimens, which are not against susceptible anaerobic bacteria3,4 equivalent in terms of cost or administration and low resistance rates within these complexity, and may differ in both effi cacy organisms.5 It also has an excellent oral and adverse effects. bioavailability (>90%), favourable pene- Canterbury District Health Board (CDHB) tration to the site of infection and a long had a long history of dosing metroni- half-life (by antimicrobial standards) of dazole IV as 500mg every eight hours, and eight hours.6 However, despite more than orally (PO) as 400mg three times daily, for 50 years of use and an established role in treatment of anaerobic bacterial infections.

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 53 www.nzma.org.nz/journal ARTICLE

Table 1: Recommended metronidazole IV dosing strategies for intra-abdominal infections.

Source 1,000mg per 24h 1,500mg per 24h 2,000mg per 24h 500mg 1,000mg 500mg 1,500mg 500mg q6h q12h q24h q8h q24h

Infectious Diseases Society of America7 Y YY

Surgical Infection Society (USA)8 Y* Y*

World Society of Emergency Surgery9 Y

Therapeutic Guidelines of Australia10 Y

New Zealand Formulary11 Y

Metronidazole datasheet12 YYY

*with a loading dose of 1,000mg IV.

In 2015, our AMS committee considered literature recommendations for metroni- Methods dazole dosing together with contemporary After consultation with senior medical knowledge of pharmacokinetic-pharmaco- offi cers, and nursing and pharmacy staff dynamic relationships, risk of bloodstream from relevant areas such as general surgery, infections with IV access, administration the following initiative was implemented in issues and cost. A multifaceted AMS October 2015: initiative was developed for adult inpatients 1. Updating online CDHB antimicrobial at CDHB and comprised: guidelines for intra-abdominal infec- 1. Prioritisation of the oral route, with tions, sepsis, pelvic infl ammatory the IV route only to be used in the disease and deep neck space infec- presence of compromised gastro- tions to refl ect the twice daily dosing intestinal absorption, or a patient strategy and prioritisation of the oral designated ‘nil by mouth’, route, 2. Twice daily dosing for most indications, 2. Updating the e-prescribing and admin- as PO 600mg twice daily or IV 500mg istration system used at CDHB to 12-hourly, include the new dosing guidelines, 3. Avoidance of unnecessary dupli- 3. Verbal education sessions for medical, cation of anaerobic cover, by not nursing and pharmacy staff, co-administering metronidazole 4. Written bulletins and a poster dissem- with amoxicillin+clavulanic acid, inated to clinical staff, carbapenems (eg, meropenem), 5. Pharmacist support of the initiative clindamycin, moxifl oxacin or via ongoing education on the wards, piperacillin+tazobactam. 6. Ward drug stocks were altered to Clostridium diffi cile-associated diarrhoea, improve access to metronidazole H. pylori eradication regimens and sexual/ 200mg oral tablets for the 600mg oral reproductive health indications other than dose. pelvic infl ammatory disease were excluded Metronidazole usage and expenditure from this initiative. from 1 October 2010 until 30 September The purpose of this paper is to describe 2017 (fi ve years pre- and two years post-ini- the multipronged AMS initiative undertaken tiative) were assessed with data extracted to change prescribing practices at CDHB and from the hospital pharmacy dispensing the resultant impact on metronidazole usage software (ePharmacy, v1.7, DXC Tech- and expenditure. nology, Virginia) into Microsoft Excel (2013).

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 54 www.nzma.org.nz/journal ARTICLE

Systemic metronidazole use was determined boards (DHBs) were surveyed electronically from issues to ward imprests and to indi- and via telephone to determine the dosing vidual patients for adult inpatients to refl ect strategy used for metronidazole IV and, the population targeted by the initiative. where relevant, the date when a twice daily Adult inpatients were those admitted to dosing strategy was adopted. Christchurch, Christchurch Women’s, Burwood or the Princess Margaret Hospitals Results after exclusion of psychiatric, paediatric This AMS initiative was associated with a and day stay areas as per established local 43% decrease in metronidazole IV use (Table method.13 2 and Figure 1), which translates to around All formulations administered to 11,800 avoided IV doses annually (data not treat systemic infections were included: shown). By contrast, non-IV administration 500mg/100mL IV infusion bags, 200mg of metronidazole increased by 104%, largely tablets, 400mg tablets, 200mg/5mL due to a 339% increase in use of the 200mg suspension and 500mg rectal suppositories. tablets. Thus, the proportion of metronida- Topical and vaginal formulations were zole-administered IV decreased from 80% excluded. Use was expressed as defi ned prior to the initiative to 52% post-initiative. daily doses (DDD) per quarter and per year Mean annual expenditure on metroni- normalised to 1,000 bed days. This was calcu- dazole in adult inpatients decreased by lated for the individual formulations, for IV 59% following the initiative, with addi- versus non-IV routes of administration, and tional savings in consumables and nursing for total systemic metronidazole use. Expen- time. The total saving was approximately diture was determined using the pharmacy $111,600 per year, comprising $33,400 and purchasing price per unit. This was essen- $78,200 in drug cost and consumables, tially static (less than 10% variation) over the respectively. seven-year study period for the 200mg tablet (NZ$0.10), 400mg tablet ($0.18), 200mg/5mL Eleven of the 19 remaining DHBs reported suspension 100mL (~NZ$25.63) and 500mg a 12-hourly dosing strategy for metroni- suppository (~NZ$2.60). The metronidazole dazole IV. Hence, 60% (12 of 20) of all DHBs IV 500mg infusion bag changed from a price (including CDHB) now recommend twice of $2.46 to $1.39 in February 2015. Costs daily dosing. All of these advise a dose of of administration consumables was taken 500 mg 12-hourly while one DHB recom- simplistically as $6.63 per IV dose for the mended 15mg/kg 12-hourly, a two-fold cost of a giving set (Alaris secondary set, greater dose (ie, 1,000mg 12-hourly for a CareFusion, Switzerland) and 100mL sodium 70kg person). The earliest adopter reported chloride 0.9% infusion bag (post-dose fl ush). that twice daily dosing was included in their guidelines since the 1990s. The remaining Between September and November 2017, 11 DHBs (including CDHB) changed to the other 19 New Zealand district health 12-hourly between 2011 and 2017.

Table 2: Mean annual metronidazole usage and expenditure for adult inpatients for the fi ve years before and after commencement of the initiative in October 2015.

DDDs per 1,000 occupied bed Expenditure days (mean usage per year) (unadjusted mean cost per year) Before A ž e r % Change Before Cost % Change* 200mg tablets 3.1 13.6 ↑ 339% $899 $3,635 ↑ 304%

400mg tablets 3.9 1.3 ↓ 67% $999 $315 ↓ 68%

200mg/5mL susp 0.5 0.5 ↔ $1,667 $1,585 ↔

500mg suppository 0.05 0.07 ↑ 40% $125 $191 ↑ 53%

Total non-IV 7.6 15.5 ↑ 104% $3,690 $5,726 ↑ 55%

500mg IV 29.9 16.9 ↓ 43% $53,925 $18,104 ↓ 66%

Total non-IV + IV 37.5 32.4 ↓ 14% $57,615 $23,830 ↓ 59%

*Changes ≤5% are arbitrarily reported as unchanged Note: susp = suspension, IV = intravenous

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 55 www.nzma.org.nz/journal ARTICLE

Figure 1: Metronidazole use (DDDs per 1,000 occupied bed days) per quarter in adult inpatients at CDHB public hospitals 1 October 2010–30 September 2017.

Discussion area under the concentration-time curve (AUC0-24h) /MIC ratio ≥ 70 was ≥ 99.8% for This AMS initiative to avoid unnec- 1,000mg and 1,500mg daily for organisms essary metronidazole IV doses in CDHB with an MIC <2mg/L. However, for an MIC public hospitals was associated with a of 4mg/L, this decreased to 28.5% and 80.0%, 43% decrease in use of the IV formulation respectively, showing an advantage for the and a 104% increase in use of the non-IV higher dose. Almost all (213/218) of the B. route. Despite overall metronidazole use fragilis isolates studied had MICs <1mg/L, decreasing minimally (14%), the shifts and the remaining fi ve isolates had MICs in route of administration resulted in of 2mg/L.15 This is in keeping with New substantial healthcare savings. Drug costs Zealand research.5 Collectively, these studies were 59% lower post-initiative resulting in support an IV dose of 1,000mg per 24 hours savings of $33,400 per annum. An additional as achieving satisfactory concentrations in $78,200 was saved in consumables through most circumstances. Given the formulation avoidance of 11,800 IV doses annually. Addi- available in New Zealand (a 100mL infusion tional benefi ts and savings that we have not bag containing 500mg), we elected to give quantifi ed relate to nursing time, avoided this in two divided doses, in line with the complications of IV access such as blood- Therapeutic Guidelines of Australia.10 In the stream infections and facilitated discharge absence of a 500mg oral dose formulation, from hospital. we chose an oral dose of 600mg to achieve While there is a lack of clinical trials similar concentrations to the IV dose and for comparing outcomes with different dosing clinicians to replace IV with oral metroni- strategies, the concept of giving metroni- dazole with confi dence in similar outcomes. dazole twice daily is not new. In 1989, Earl The data used in this analysis has limita- et al14 stated that IV 500mg or PO 400mg tions as it was derived from pharmacy twice daily had been used at their hospital dispensing software. It refl ects ‘mass’ shifts for “many years” with “apparent success”. in stock from pharmacy to clinical areas They reported that this regimen produced rather than dispensings, prescriptions adequate metronidazole serum concentra- or administrations to patients. While the tions in 48 surgical patients, defi ned as a assumption is that this stock movement trough concentration above both 2mg/L and refl ects usage in patients it is undoubtedly the minimum inhibitory concentration (MIC) less accurate than information obtained of most relevant anaerobes.14 Nearly three from electronic prescribing and adminis- decades later, Sprandel et al15 determined tration software. This has only been used for that the probability of attaining a target a short time at Canterbury DHB and cannot,

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 56 www.nzma.org.nz/journal ARTICLE

therefore, be used to assess changes in from national guidance and a coordinated usage over a long period of time. However, approach. All of our DHBs are under-re- e-prescribing data for the six months to sourced for AMS, and only half employ September 2017, shows that 81% of metroni- dedicated staff to manage AMS in New dazole IV is prescribed as 12-hourly on our Zealand public hospitals.16 While we believe general surgical wards (data not shown). that AMS programmes should be driven The slow change observed in New Zealand with a quality rather than fi nancial focus, DHBs and variability in guidelines suggests it is clear that this initiative provides a that AMS in New Zealand would benefi t compelling economic case for employing staff dedicated to AMS.

Competing interests: Nil. Acknowledgements: QianYi Chuah, Sheryl Priest, Margaret Simpson, Holly Wang, Kenny Daly and Andrew Villazon for assistance with data extraction on hospital pharmacy issues, MedChart™ prescribing, and occupied bed days. Author information: Sharon J Gardiner, Antimicrobial Stewardship Pharmacist, Departments of Infectious Diseases, Clinical Pharmacology and Pharmacy, Canterbury District Health Board, Christchurch; Sarah CL Metcalf, Infectious Diseases Physician, Department of Infectious Diseases, Canterbury District Health Board, Christchurch; Paul KL Chin, Clinical Pharmacologist and Senior Lecturer, Department of Clinical Pharmacology, Canterbury District Health Board, Christchurch; Department of Medicine, University of Otago, Christchurch; Matthew P Doogue, Clinical Pharmacologist and Associate Professor, Department of Clinical Pharmacology, Canterbury District Health Board, Christchurch; Department of Medicine, University of Otago, Christchurch; Simon C Dalton, Infectious Diseases Physician, Department of Infectious Diseases, Canterbury District Health Board, Christchurch; Stephen T Chambers, Infectious Diseases Physician and Professor, Department of Infectious Diseases, Canterbury District Health Board, Christchurch; Department of Pathology, Canterbury District Health Board, Christchurch. Corresponding author: Sharon J Gardiner, Department of Infectious Diseases, Christchurch Hospital, PO Box 4710, Christchurch. [email protected] URL: http://www.nzma.org.nz/journal/read-the-journal/all-issues/2010-2019/2018/vol-131-no-1473- 13-april-2018/7543

REFERENCES: 1. Durel P, Roiron V, Siboulet 3. Tally FP, Goldin BR, Sullivan mice and men. Clin Infect A, Borel LJ. [Trial of an N, et al. Antimicrobial Dis. 1998; 26:1–12. antitrichomonal drug activity of metronidazole 5. Roberts SA, Shore KP, derived from imidazole: in anaerobic bacteria. Paviour SD, Holland D, RP-8823]. C R Soc Fr Antimicrob Agents Chemo- Morris AJ. Antimicrobial Gyncol. 1959; 29:36–45. ther. 1978; 13:460–5. susceptibility of anaerobic 2. Shinn DL. Metronida- 4. Craig WA. Pharmacoki- bacteria in New Zealand: zole in acute ulcerative netic/pharmacodynamic 1999–2003. J Antimicrob gingivitis. Lancet 1962; parameters: rationale for Chemother. 2006; 57:992–8. 279(7240):1191. antibacterial dosing of

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 57 www.nzma.org.nz/journal ARTICLE

6. Lau AH, Lam NP, Piscitelli from a global perspective: of antimicrobial use in SC, et al. Clinical pharma- 2017 WSES guidelines for New Zealand hospitals – a cokinetics of metronidazole management of intra-ab- comparison to Australian and other nitroimidazole dominal infections. World and English data. NZ anti-infectives. Clin Phar- J Emerg Surg. 2017; 12:29. Med J. 2015; 128(1421). macokinet. 1992; 23:328–64. 10. Antibiotic Expert Groups. 14. Earl P, Sisson PR, Ingham 7. Solomkin JS, Mazuski JE, Therapeutic guidelines: HR. Twelve-hourly dosage Bradley JS, et al., Diagno- antibiotic. Version 15. schedule for oral and sis and management of Melbourne: Therapeutic intravenous metronidazole. complicated intra-abdomi- Guidelines Limited, 2014. J Antimicrob Chemother. nal infection in adults and 11. New Zealand Formulary 1989; 23:619–21. children: guidelines by the (NZF). NZF v63, 2017. 15. Sprandel KA, Drusano GL, Surgical Infection Society Available from: www. Hecht DW, et al. Population and the Infectious Diseases nzf.org.nz (accessed pharmacokinetic modeling Society of America. Clin 04 September 2017). and Monte Carlo simula- Infect Dis. 2010; 50:133–64. 12. Metronidazole-Claris tion of varying doses of 8. Mazuski JE, Tessier (metronidazole intrave- intravenous metronidazole. JM, May AK, et al. The nous infusion 500 mg/100 Diagn Microbiol Infect Surgical Infection Soci- mL), AFT Pharmaceuticals Dis. 2006; 55:303–9. ety revised guidelines Limited, Takapuna: 16. Gardiner SJ, Pryer JA, on the management of Auckland. 30 March 2006 Duffy EJ. Survey of intra-abdominal infection. (revised 15 March 2017). antimicrobial steward- Surg Infect (Larchmt). Found at: www.medsafe. ship practices in public 2017; 18:1–76. govt.nz/profs/Datasheet/m/ hospitals in New Zealand 9. Sartelli M, Chichom-Mefi re MetronidazoleAFTinf.pdf district health boards. NZ A, Labricciosa FM, et 13. Duffy E, Gardiner S, du Med J. 2017; 130:27–41. al. The management of Plessis T, et al. A snapshot intra-abdominal infections

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 58 www.nzma.org.nz/journal ARTICLE

Poisoning due to tutin in honey—a report of an outbreak in New Zealand Michael Beasley, Dell Hood, Philippa Anderson, John Reeve, Robin J Slaughter

ABSTRACT AIM: In autumn 2008, an outbreak of toxic honey poisoning was identified. The outbreak was not recognised initially until three cases from one family group presented to hospital, with a common factor of recent consumption of locally produced honey. The aim of this study was to investigate potential cases of this honey poisoning and determine which toxin was involved. METHOD: The incident was investigated retrospectively by Waikato District Health Board’s Population Health unit and the New Zealand Food Safety Authority (NZFSA). Identified patients were followed up by questionnaire to gather case information. HortResearch (now Plant and Food Research) tested honey samples for toxins. RESULTS: The causative agent was identified as tutin, which comes from the New Zealand native plant tutu (Coriaria arborea) which has long been known as a potential source of contamination of honey produced in the warmer parts of New Zealand. Retrospective case investigation identified a total of 22 possible or probable cases, based on a clinical case definition. The spectrum of toxic e– ects reported were broadly similar to those previously described for tutin, derived either directly from the plant itself or indirectly from honey. There were 13 samples of honey, linked to symptomatic individuals, which were available for testing. Of these, 10 were positive for tutin and its hydroxy metabolite hyenanchin (hydroxytutin) and one was positive for hyenanchin alone. CONCLUSION: Toxic honey production is a significant risk in parts of New Zealand. Beekeepers and health professionals need to be informed of this risk and know how best to manage it. Due to this poisoning incident, public and professional awareness of honey poisoning has been substantially enhanced. This incident led to development of new food safety standards for New Zealand honey.

oxic honey is a well-defi ned phenome- In a detailed review of honey poisoning non that has been described since the in New Zealand published in 1965, Palm- Ttime of the ancient Greeks.1 Poisoning er-Jones records both the history of clinical from toxic honey occurs when phytotoxins recognition of the syndrome and the from plant species are present in honey, sequence of events which fi nally solved the usually via direct transfer in nectar or pol- mystery of how tutin, the toxin known to be len. Because none of New Zealand’s native present in the native plant Tutu (Coriaria bees produce honey, the eating of locally arborea), was transferred to honey where it produced honey did not begin until after the could then produce human toxicity.2 introduction of the honeybee Apis mellifera Tutu is a native shrub or small tree which by missionaries in Northland in 1839. How- is common throughout New Zealand. It ever, it was not until some decades later, in has distinctive fl owers, berries and foliage the late 1880s, that vomiting, headache and (Figures 1 and 2). It is an early coloniser of delirium was reported after eating locally cleared land, with seeds spread by birds. All produced honey from both managed hives parts of the plant are toxic, other than the 2 and wild honey bees. Outbreaks of toxic fl eshy petals around the seeds. These were honey have continued to occur sporadically used as a sweet fl avouring and medicine in New Zealand over the last 130 years. by Māori, who knew that eating the entire

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 59 www.nzma.org.nz/journal ARTICLE

berry resulted in poisoning and sometimes Figure 1: Tutu berry. © Used with permission. death. Serious and sometimes fatal human Photo Credit: Trevor James. poisoning from eating the berries was formally documented by European settlers.3 The fi rst attempt to isolate the toxin in tutu was around 1870,3 and experimental studies were underway by ~1890,4 and continued through the early 1900s to better characterise its toxicity.5–7 Tutin is also found in Coriaria species in South America, and the related toxin coriamyrtin in species in the western Mediterranean,8 mainly in France, Spain and Morocco. However, there appears to be little evidence to suggest that these species have been responsible for honey poisoning, which is more explicitly reported in association with New Zealand.8–10 The symptom spectrum of toxic honey poisoning was noted to be similar to tutu poisoning,2 so this plant became a major suspected source. However, tutu fl owers do not contain nectar and the toxin tutin is not present in pollen, which bees take for its protein. Therefore a mechanism of transfer was unclear. The toxin was known to be present in the sap, leaves and seeds.2,3 The transfer of tutin to honey was noted to occur only where an insect, Scolypopa (‘honeydew’) which it excretes contain australis, the passion vine hopper or ‘lace tutin.2 This insect is found only in warm wing’ was also present. This pest insect, regions, and this explains why honey which was accidentally introduced from poisoning has only been described in the Australia around 1870, sucks sap from warmer parts of New Zealand (particularly young green shoots. When it is feeding on the northern half of the North Island and tutu plants, the sticky intestinal secretions Marlborough), even though tutu grows throughout the country.

Figure 2: Tutu mature foliage. © Used with permission. Photo Credit: Trevor James.

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 60 www.nzma.org.nz/journal ARTICLE

Figure 3: Chemical structures of tutin (a), hyenanchin (b), picrotoxinin (c) and picrotin (d).

At the time of the incident, it was believed that bees took honeydew only when their Method: outbreak normal foods were in short supply, but investigation subsequent investigation has revealed that On 21 March 2008, a small rural hospital bees take some honeydew at any time. In in Waikato, New Zealand reported that three the areas in New Zealand from where toxic people from a six-person family group had honey poisoning has been reported, fl owers presented to their emergency department on of native vegetation usually provide suffi - two separate occasions with acute onset of cient nectar and pollen, but it is likely that vomiting and headache after eating locally bees consume more of this toxic honeydew produced comb honey. Two of the three had 2 in drought conditions. grand mal seizures. The association of the The major risk to human health is the illness with the ingestion of comb honey consumption of honey directly from the was recognised because the illness affected comb, which avoids the usual dilution when only those who ate the honey, and because honey is separated from the comb and large the third case presented some hours later, volumes from different sources are mixed. having no other food intake in common and As well as tutin, the less toxic hyenanchin having eaten the implicated honey for the (also known as hydroxytutin or mellitoxin) fi rst time at a later meal. The outbreak was is typically also present in toxic honey. Tutin notifi ed to the public health service within and hyenanchin are similar compounds; hours of the presentation of the third family their chemical structures are presented in member. Figure 3. Hyenanchin is produced by the The beekeeper who produced the honey vine hopper’s metabolism of tutin.2,11 was contacted for information about the

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 61 www.nzma.org.nz/journal ARTICLE

source of his product, and the amount The questionnaire used to gather case which had been distributed for sale. It then information was hastily devised and was emerged that the beekeeper had himself based on the few relevant published papers experienced a very similar illness with which could be obtained in a short time. gastrointestinal symptoms and seizures (See Appendix) It was administered around about fi ve weeks earlier. Symptom onset the country by various public health staff, was around three hours after he had eaten many without clinical training. The inter- around 20ml of the same comb honey which pretation of some questions may have been he had harvested the previous day. He had variable. eaten “a taste” of the honey at the time of Based on the information collected the harvest without incident. Although he had fi nal case defi nitions used were: received care from the same hospital, the • Possible case = vomiting OR any association with ingestion of comb honey had neurological symptom within 24 not been recognised by patient or clinicians. hours of eating comb honey. The working diagnosis was that the honey • Probable case = vomiting AND any was contaminated with tutin, a phytotoxin neurological symptom within 24 present in the native shrub tutu. Urgent hours of eating comb honey. retrieval of the implicated honey was initiated from the three outlets where it had • Confi rmed case = Possible or been sold. Because the notifi cation occurred probable case where tutin or at the beginning of the Easter holiday, media hyenanchin was detected in honey assistance was sought to advise the public consumed by the case. not to eat any comb honey, and to return The fl ow chart used to categorise cases is any uneaten product from the implicated depicted below (Figure 4). producer. Publicity was used to identify The questionnaire included questions further cases. about the source and quantity of honey The New Zealand Food Safety Authority consumed. Unsold honey packages and any (NZFSA), at that time the statutory body remaining honey were collected from cases’ responsible for food safety, managed food homes. Analysis for tutin and hyenanchin safety issues arising from this incident while (hydroxytutin) was carried out later at Waikato District Health Board’s Population HortResearch, then a government-owned Health managed the outbreak investigation laboratory in Hamilton (HortResearch has and control measures, in an inter-agency since been incorporated into a new agency, collaborative emergency response. NZFSA Plant and Food Research). The method has subsequently implemented longer-term of analysis for tutin and hyenanchin was measures to ensure the safety of commer- liquid chromatography–mass spectrometry cially produced honey in New Zealand. (LC-MS/MS) after aqueous extraction and NZFSA has since merged into the Ministry partitioning into solvent. for Primary Industries. Fifty-three of the 300-gram individual Media coverage of this poisoning incident comb packages of honey were sold, under allowed retrospective identifi cation of 22 two different product names identifying the people who had consumed the implicated two hive sites from where the honey had honey. As no other method of case fi nding been harvested. Subsequent toxin analysis was possible, additional cases may have revealed that the honey from only one of occurred outside New Zealand, as retailers these locations contained tutin. All but seven recalled selling some of the contaminated of the sold packages were traced, although honey to international visitors returning not all could be recovered. home.

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 62 www.nzma.org.nz/journal ARTICLE

Figure 4: Flow chart used to categorise cases.

Two of these sample tested were confi rmed Results of outbreak as containing toxins. Overall, this gave 11 investigation confi rmed cases, nine probable cases and Of the 22 “probable” or “possible” cases two possible cases (Figure 4). initially defi ned on clinical grounds, 11 fi tted The signs and symptoms which developed the defi nition of a confi rmed case after the in the probable and confi rmed cases are relevant honey samples had been tested presented in Figure 5. The onset of the (Table 1). Initially, 18 cases were defi ned as clinical effects ranged from 0.5 to 17h with a probable cases and there were 10 samples of median of 7.5h. Nausea and vomiting were honey from this group available for testing. commonly observed and there were eight Of these samples, nine were subsequently cases of seizures. The amount consumed confi rmed as testing positive for tutin ranged from a “smear” to up to 200g of and hyenanchin. Four cases were initially honey. The majority of these patients were classed as possible cases and three of these male (70%) with an age range of between 3 had honey samples available for testing. and 76 years (Table 1).

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 63 www.nzma.org.nz/journal ARTICLE

Table 1: The key features of confi rmed and probable cases.

Confirmed cases Probable cases Number of cases 11 9

Age 3–74 12–76

Percent male 64% 77%

Ingestion—symptom onset interval Median 7.2 hours Median 8 hours Range 0.5–17 hours Range 0.5–17 hours

Honey consumed “smear” – ~70 grams “smear” - ~200 grams

Hospitalised/hospital visit 5 4

Saw GP only 1 3

No medical attention 5 2

Seizures 4 4

Figure 5: Signs and symptoms that developed in probable and confi rmed cases.

the body.13 This antagonising of inhib- Discussion itory processes results in central nervous Both tutin and hyenanchin are struc- system stimulatory effects which charac- turally similar to picrotoxinin (see Figure teristically include seizures, but can also 3), the active principle of picrotoxin, found involve increased activity of the vasomotor, in plants of the Order Menispermacea.10,12 respiratory and autonomic centres in the These compounds block the inhibitory medulla and/or cortex.13 Early researchers actions of the neurotransmitter GABA had concluded that tutin increased the (γ-aminobutyric acid) that are mediated excitability of the “medullary centres”.5–7 via GABAA receptors at various sites in

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 64 www.nzma.org.nz/journal ARTICLE

Central nervous system stimulation was a spasm.19 Indeed, some experimental simi- prominent feature among this series of cases larities had been earlier noted with tutin with 40% (8/20) of probable and confi rmed itself with Fitchett and Malcolm commenting cases suffering from one or more seizures. on tachycardia, arteriolar constriction and The effects observed correlate well with increases (or decreases) in blood pressure.5–7 the observed symptoms of human tutin The respiratory symptoms in the current poisoning; after a characteristic delay of cases were not classifi ed in detail; however, three to six hours, nausea and vomiting increased respiratory rate and effort is develop, often followed by tremor and/or well described with picrotoxin, where it grand mal seizures, which may be severe is linked to stimulation of the respiratory and recurrent. Nausea and vomiting can centre.20 The same effect is described with persist for many hours. Other effects can tutu (and coriamyrtin) poisoning, followed include tachycardia, tachypnoea, diffi culty however in severe cases by respiratory breathing, delirium, blurred vision, anxiety, depression including apnoea, also noted agitation, excitement, weakness, dizziness, experimentally.5,6 amnesia, stupor and coma. Death is usually There are considerable similarities to the due to respiratory arrest.3,14–16 There were effects noted from poisonings from closely no prolonged periods of unconsciousness or related plants in Chile and the Western deaths associated with this cluster. Mediterranean, containing tutin and cori- Drug- or toxin-induced seizures are typi- amyrtin respectively,8,21 but the constellation cally of relatively short duration, usually of factors leading to honey poisoning from manageable with standard treatments, tutu in New Zealand does not appear to including benzodiazepines as fi rst line agents occur with similar species elsewhere. (which as GABA receptor agonists, would At the time of this outbreak, the kinetics seem particularly appropriate for tutin and of tutin (and hyenanchin) in humans were hyenanchin).13,14 With most toxins, one would largely unknown. Even the experimental not generally expect any major long-term kinetic data was limited,3,5,6 though it (and neurological sequelae in the great majority of that regarding picrotoxin)22 suggested these patients. However, seizures are a “hallmark” compounds have a relatively short half-life, effect of picrotoxin-like compounds and probably a few hours at most. The data might be more protracted in this context from this outbreak also suggest the duration than with many other toxic causes. The two of most adverse effects is relatively short, cases that had more than one seizure in this being a matter of a few days only, but it series were managed conservatively with appears some effects can last for longer. A initial restrictions on their driving. There pharmacokinetic study performed following are accounts of incomplete neurological this outbreak in six healthy males given recovery following tutu poisoning.3,5,6 The tutin-contaminated honey (tutin dose 1.8ug/ observation of amnesia in this series of cases kg) showed two peak plasma concentrations; is not new. It has been previously noted that the fi rst at 0.9 hours and a higher peak at tutu poisoning may result in loss of memory 15 hours. This double-peak effect may help and “incapacity for work”.2,3 explain the range of onset times for clinical Tachycardia was also commonly reported effects (0.5 to 17 hours) found in this case in this series of cases, and likely arose in series. Limited adverse effects were noted part from increased central sympathetic in this low-dose pharmacokinetic study with outfl ow. Experimental studies of the related only mild transient headache developing in picrotoxin found it could produce an initial one subject and mild light-headedness devel- short-lived phase of decreased heart rate oping in another. The half-life of tutin was and blood pressure, attributed to activation determined to be 5.4 hours.11 of central parasympathetic centres, followed Immediately after this episode, the then by a second phase of sinus tachycardia and/ New Zealand Food Safety Authority (NZFSA) or increased blood pressure, attributed to began reviewing policy and procedures central sympathetic activation.17,18 At very related to beekeeping in New Zealand to high doses, early bradyarhythmias and later refi ne existing safety measures in relation ventricular tachyarrhythmias were noted, to tutu-related risks. Detailed discussion the latter thought due to coronary artery

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 65 www.nzma.org.nz/journal ARTICLE

of these is beyond the scope of this report. The policy and food standards related to Limitations managing the risk of tutin in honey are This investigation was based on retro- available online at http://www.foodsafety. spective information, obtained in most govt.nz/industry/sectors/honey-bee/tutin/ cases by telephone interview, and some- index.htm. To support the review, NZFSA times undertaken by non-clinical staff. It is commissioned the purifi cation of tutin and likely that the interpretation of the infor- then several toxicity studies. These experi- mation provided by the cases varied. Where mental studies identifi ed the doses of tutin possible this was validated with clinical that cause no observable adverse effects, information, but only 13 of 22 cases had and that hyenanchin was essentially not sought medical advice, and the treating clini- toxic at amounts where tutin would cause cians had recognised the likely cause of the death. Experimentally, the median oral illness only in the last case to present. Addi- tionally, the publicity given to this outbreak lethal dose (LD50) of tutin in mice is 4.7mg/ kg body-weight.23 The acute toxicity study of would have augmented existing recall bias. tutin demonstrated a “no observed adverse However, the symptoms reported by the effect level” (NOAEL) of tutin in test animals confi rmed and probable cases are generally of 0.25mg tutin/kg body weight23 from consistent with other New Zealand cases which NZFSA derived an interim upper reports of toxic honey ingestion, and from acceptable concentration of tutin in honey24 “direct” poisoning from ingestion of the tutu (as mentioned NZFSA is now within the plant itself.2,14,15 Ministry for Primary Industries). Subsequent Dry summers in the area of distribution to this interim standard, the human phar- of the passion vine hopper in northern macokinetic study11 enabled a more robust New Zealand are not uncommon, and maximum residue limit to be established given incontrovertible evidence that the and the maximum concentration of tutin in risk management strategies for honey honey set in the Joint Australia New Zealand production in place until the summer of Food Standards Code is now 0.7mg/kg. 2007–2008 were inadequate, it is highly A 2016 New Zealand standard has been likely that there have been other outbreaks produced to ensure honey producers comply of tutin poisoning. Dispersed outbreaks with the Food Standards Code. This 2016 arising from a widely distributed common standard requires that all honey be shown source are diffi cult to identify and may go to comply with the Food Standards Code unnoticed. Tutin poisoning may explain when packed, and there are fi ve options to other cases of isolated seizures in adults do this. Either the honey has to be tested without identifi ed neurological abnormality. and demonstrated to comply, or shown to be of low risk of non-compliance. The four Conclusions options for showing that the honey is low Toxic honey production is a signifi cant risk are either the honey was proven to risk in parts of New Zealand if honey safety have been harvested outside of the time is not managed effectively. Poisoning can period when scolypopa are present, or it has result from the ingestion of processed as been shown that the honey was from hives well as comb honey. With amateur and stationed where there is no tutu present professional beekeeping now burgeoning, within bee foraging range, or from hives the risk of consumption of toxic freshly stationed south of 42 degrees south latitude, harvested comb honey may increase. It is or after three years of targeted monitoring, critical that beekeepers are informed of this the concentration of tutin in the extracted risk and know how best to manage it. honey is always not more than 0.035mg/kg, It is also important that clinicians or targeted honey has tutin concentrations consider the diagnosis of toxic honey not more than 0.01mg/kg if the honey is ingestion in patients who present with from comb honey for sale.24 nausea, vomiting and symptoms of central

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 66 www.nzma.org.nz/journal ARTICLE

nervous system agitation. Extrinsic toxins for many decades. As well as a generational had understandably not been considered loss of knowledge about tutin toxicity in in the differential diagnosis of any case New Zealand honey (given that the last seen by a medical practitioner prior to previously documented outbreak occurred the presentation of the family cluster, in 1974), many healthcare workers in this although in retrospect, the association of country are overseas trained, and thus less CNS stimulation with nausea and vomiting likely to be familiar with the toxicity of New is characteristic. Foodborne toxin ingestion Zealand native plants. needs to be considered as a possibility, One of the undoubted outcomes from particularly where prior risk of seizures was this episode is that public and professional remote. This, however, may be diffi cult as awareness of this form of honey poisoning tutin poisoning had had little or no publicity has been substantially enhanced.

Appendix Questionnaire used in investigation HOUSEHOLD HEADER INFORMATION TOXIC HONEY POISONING OUTBREAK

NOTIFIER IDENTIFICATION

Name of reporting source ______Contact ph: ______Date: ______/ ______/ _____

CASE IDENTIFICATION Name of Case: ______Address of Case: ______Phone numbers: home ______Work: ______Other: ______NHI Number: ______

CASE DEMOGRAPHY Date of Birth: ______/______/______OR Age: ______(days/months/years) Sex: M/F Approx Weight*______Ethnicity: NZ Māori / NZ European-Pakeha / Pacifi c Island / Other European/ Other: ______Occupation: ______Place of work/school/pre-school: ______*Either weight in kg or use descriptor – large adult etc

HONEY PRODUCT DETAILS Form of Honey: Comb / Liquid / Creamed

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 67 www.nzma.org.nz/journal ARTICLE

PACK DETAILS: Trade or Brand Name: ______Product size or weight: ______Any Date Mark or Batch Number: ______

PURCHASE DETAILS Where purchased: ______Date purchased: ______Where/How the honey been stored since purchase: ______

Is there any additional product: Y/N (ADVISE CUTOMER TO SEAL THE LEFT-OVER PRODUCT AND HOLD. WE WILL ARRANGE COLLECTION) What address this currently at: ______

RELATED CASES/CONTACTS Anyone else been unwell following honey consumption: Y/N Anyone consumed the honey and not been sick: Y/N Aware of anyone else may have received this honey (whether or not consumed): Y/N N.B. Interviewer - If Y please complete the contact list

CONSUMPTION DETAILS Date and time honey eaten: ______

Amount eaten: ______

SYMPTOMS Date and time became ill: ______Visit Doctor: Y/N Name and address: ______Hospital Y/N

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 68 www.nzma.org.nz/journal ARTICLE

SYMPTOM ONSET DATE/TIME DURATION Dry mouth

Nausea

Vomiting

Diarrhoea

Muscle spasms

Muscle weakness

Blurred vision

Red eyes

Headache

Respiratory problems

Rapid heart beat

Delirium

Giddiness

Increased Excitability

Stupor

Coma

Convulsions

Amnesia (memory loss)

Other (describe):

Case is happy for information to go to NZFSA Y/N ANY ADDITIONAL COMMENTS/INFORMATION:

NAME OF INTERVIEWER: ______DATE AND TIME OF INTERVIEW: ______

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 69 www.nzma.org.nz/journal ARTICLE

Competing interests: Nil. Author information: Michael Beasley, National Poisons Centre, University of Otago, Dunedin; Dell Hood, Formerly Waikato District Health Board, Hamilton; Philippa Anderson, Counties Manukau District Health Board, Auckland; John Reeve, New Zealand Ministry for Primary Industries, Wellington; Robin J Slaughter, National Poisons Centre, University of Otago, Dunedin. Corresponding author: Michael Beasley, National Poisons Centre, University of Otago, PO Box 56, Dunedin 9054. [email protected] URL: http://www.nzma.org.nz/journal/read-the-journal/all-issues/2010-2019/2018/vol-131-no-1473- 13-april-2018/7544

REFERENCES: 1. Gunduz A, Turedi S, Russell review of the literature. reports of Coriaria arborea RM, Ayaz FA. Clinical Toxicon. 2005; 46:600–3. ingestion, review of review of grayanotoxin/ 9. Bruneton J. Toxic plants literature and recommen- mad honey poisoning past dangerous to humans and dations for management. N and present. Clin Toxicol animals. Paris: Lavoisier Z Med J. 2013; 126:103–9. (Phila). 2008; 46:437–42. Publishing, 1999. 16. Chilvers CD. Tutu poisoning 2. Palmer-Jones T. Poisonous 10. Burrows GE, Tyrl RJ. in an elderly Maori lady. honey overseas and in Toxic plants of North N Z Med J. 1972; 75:85–6. New Zealand. N Z Med America. Ames (IA): 17. DiMicco JA, Hamilton BL, J. 1965; 64:631–7. Iowa State Press, 2001. Gillis RA. Central nervous 3. Connor HE. The poisonous 11. Fields BA, Reeve J, system sites involved the plants in New Zealand. Bartholomaeus A, Mueller cardiovascular effects of Wellington: Crown U. Human pharmacokinetic picrotoxin. J Pharmacol Copyright, 1977. study of tutin in honey; a Exp Ther. 1977; 203:64–71. 4. Christie WL. On the plant-derived neurotox- 18. DiMicco JA, Prestel T, Pearle physiological action of in. Food Chem Toxicol. DL, Gillis RA. Mechanism Coriaria Ruscifolia, the tutu 2014; 72:234–41. of cardiovascular changes poison of New Zealand. N 12. Momtaz S, Lall N, produced in cats by activa- Z Med J. 1889-90; 3:211–24. Hussein A, et al. Inves- tion of the central nervous 5. Fitchett F. A contribution tigation of the possible system with picrotoxin. to our knowledge of biological activities of a Circ Res. 1977; 41:446–51. the physiological action poisonous South African 19. Segal SA, Pearle DL, Gillis of tutin. Trans NZ Inst. plant; Hyaenanche globosa RA. Coronary spasm 1908; 41:286–366. (Euphorbiaceae). Pharma- produced by picrotoxin 6. Fitchett F, Malcolm J. cogn Mag. 2010; 6:34–41. in cats. Eur J Pharmacol. On the physiological 13. Nistri A, Constanti A, 1981; 76:447–51. action of tutin. Q J Exp Quilliam JP. Letter: Central 20. Hirsh K, Wang SC. Respi- Physiol. 1908; 2. inhibition, G.A.B.A., and ratory stimulant effects of 7. Malcolm J. Some experi- tutin. Lancet. 1974; 1:996–7. ethamivan and picrotoxin. ments on tutin and tutu 14. TOXINZ poisons infor- J Pharmacol Exp Ther. poisoning. Trans NZ mation. Coriaria arborea 1975; 193:657–63. Inst. 1913; 46:248–54. Datasheet. New Zealand, 21. Fuentealba J, Guzman L, 8. de Haro L, Pommier 2017, [cited 1 August Manriquez-Navarro P, et P, Tichadou L, et al. 2017]. Available from: al. Inhibitory effects of Poisoning by Coriaria http://www.toxinz.com tutin on glycine receptors myrtifolia Linnaeus: 15. Belcher SF, Morton TR. in spinal neurons. Eur J a new case report and Tutu toxicity: three case Pharmacol. 2007; 559:61–4.

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 70 www.nzma.org.nz/journal ARTICLE

22. Duff DM, Dille JM. Distribu- Zealand beekeeping 24. Food (Tutin in Honey) tion and rate of elimination industry and consumers. Standard 2016. Wellington, of picrotoxin. J Pharmacol Hamilton, New Zealand: New Zealand: Ministry Exp Ther. 1939; 67:353–7. HortResearch, 2008, [cited for Primary Industries 23. McNaughton DE, Good- 2017 August 1.Available (MPI), 2016, [cited 2017 win RM. Reducing the from: http://maxa.maf. August 1.Available from: threat Tutu toxic tutu govt.nz/sff/about-proj- http://www.foodsafety. honey poses to the New ects/search/L07-041/ govt.nz/elibrary/indus- technical-report.pdf try/dv11137.htm

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 71 www.nzma.org.nz/journal VIEWPOINT

The Otago Medical School Anatomy Museum Collection: Taonga for learning in the 21st Century Louisa JM Baillie, Christopher L Smith

lumni of The Otago Medical School, Anatomy Museum for active use in teaching and many others, will remember has been against the international trend. Atheir time in the W.D. Trotter Anato- However, the retention of all models my Museum. Its rich and diverse collection including those from the 19th and 20th of now over 3,000 catalogued anatomy spec- century is now placing the Museum in an imens and models for learning and research enviable position. Not only are many now has been amassed by staff over nearly 140 appreciated for their artistic merit and irre- years. Included are wax models by the placeable value, they are being refocused Ziegler and Tramond studios, 77 authentic on as powerful anatomy learning aids that painted plaster models by the Leipzig fi rm complement and deepen learning experi- 1 of Steger, clastique papier mâché models ences from other pedagogy, including digital by Louis Auzoux’s factory, as well as many methods.4–9 Indeed, Drake and Pawlina in-house wet and plastinated specimens and recently recommended to “remove those models including fi berglass, wax and hand- anatomical models from the glass cabinets. 2 carved wooden examples. There are even Take the plastic fi gures down from shelves, replicas of classical statues by D Brucciani dust them off and teach the students to learn and Co. of London. The display is enhanced them well”.4 Reasons why learning from 3D by being housed in an elegant, tall space objects is so effective are corroborated by with a mezzanine fl oor that borders three recent research on sensory input from touch sides, and warm aesthetics with wooden and kinesthetics.10–12 framed glass cabinets, rimu stair railings Dr Louis Auzoux’s large (scale 10:1) and diffuse natural lighting. Yet this col- clastique papier mâché ear model is an lection is vulnerable to attrition, because iconic and endearing item of this museum’s of space stress at an institution that has a collection. A feature of Auzoux’s human squeeze on every square metre, and surety anatomy models is that they were designed of ongoing funding required to maintain it. to be pulled apart, imitating dissection of This vulnerability has been further height- the organs and structures presented. The ened by the trend at some schools to replace, very nature of their design therefore means rather than augment, this type of learning heavy use and handling over an extended with digital resources. In Britain alone, over period of time. This particular ear model the past 30 years, many medical schools was introduced to the W.D. Trotter museum have modernised their collections in favour in the 1880s, and has been continuously of digital learning methods, and in the used for hands-on teaching ever since process have discarded the old 3D models.3 (Figure 1). Therefore, retention of all resources in this

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 72 www.nzma.org.nz/journal VIEWPOINT

Figure 1: Professor George Dias showing pathway of 7th cervial nerve to a student.

This ear model continues to be in high follow and understand the pathways of the demand as its clastic design means pieces intracranial nerves. It wasn’t until I arrived can be removed to see deeper structures, in the Department in 1995 and laid eyes and the anatomical structures themselves are hands on this model that I truly understood very detailed, and all can be clearly viewed the pathway of the 7th nerve. It is so beauti- because of its large scale. Professor George fully demonstrated on this model” (Figure 2). Dias (Head and Neck Lecturer, Otago The following teaching explanation is by Medical School) particularly appreciates Professor George Dias: it when teaching postgraduate surgical 1. 7th nerve, with Nervous Intermedius and ophthalmology students. He says he adjacent, enters the external auditory always feels guilty handling such a delicate meatus museum piece, but no other images or th diagrams from textbooks or digital media 2. 7 nerve courses through the facial quite show the structures he likes to focus canal in the petrous region of the on with such clarity and intricate detail. In facial bone. It is running above the fact, he says this particular model is the only vestible of the inner ear one he has come across which so clearly 3. 7th nerve exits the stylomastoid demonstrates the branches and pathways of foramen as a pure motor nerve, the facial nerves “As a student I struggled to having shed all the Nervous Inter- medius fi bres.

Figure 2: Detail of pathway of 7th cervical nerve.

Dotted lines indicate borders between parts of the clastic model

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 73 www.nzma.org.nz/journal VIEWPOINT

Figure 3: Arrows indicate interference artefact from wire substructure of vessels in printed reproduction.

In 2015, conscious of the value of this in 1:1 ratio using ABS plastic. The 3D print model, Chris Smith (curator, Anatomy included internal hollow regions in the Museum) investigated ways of reproducing large parts, an advantage as their weight to it to allow the retirement of the original, hold was lightened to be approximately the using the replica in its place for teaching. same as the originals. However, what wasn’t Reproducing this also afforded an oppor- anticipated was signifi cant surface shape tunity to trial the potential of scanning and distortion, due to ‘noise’, or interference, 3D printing for other Museum items (the because metal wire within the internal struc- Auzoux collection alone comprises of 15 tures interfered with radiation from the CT pieces, from a 90-year purchase period from scanner (Figure 3). This was an unexpected the 1880s to 1970s). CT scanning was chosen outcome as it had been hoped that the fi rm as the most appropriate approach given the outsourced to print the scan fi les would edit complex 3D nature of the model, including and ‘clean up’ the fi les before printing. many concave surfaces (undercuts). All eight One of the structures, the cochlea parts of the clastic ear were CT scanned presented as two parts, was so poorly repro- using a Siemens SOMATOM Emotion CT duced that it could not be salvaged by hand scanner. The DICOM digital images obtained sculpting (Figure 4). were then converted to .stl fi les and printed

Figure 4: Cochlea print (left) next to original cochlea (right).

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 74 www.nzma.org.nz/journal VIEWPOINT

Figure 5: Original (left) and copy (right) of Auzoux ear.

Dr Louisa Baillie, working as an hand-sanded pieces to 400 grit smoothness, anatomical scientifi c artist, salvaged sprayed them with plastic adhesion and modifi ed the remaining six struc- promoter, and then intricately painted them tures to become as anatomically similar using high gloss enamels colour matched to to the original pieces as possible. This the original model (Figure 5). was achieved in the Anatomy Museum She reproduced a timpanic membrane, workshop, using dentistry grinding burrs, similar in translucency to the original, using razor blades and sandpaper to subtract tissue paper, shellac, teased red cotton and a (carve back) and epoxy fi ller and Kneadite clear plastic spray (Figure 6). “green stuff” to add on. Finally, Dr Baillie

Figure 6: Timpanic membrane, a translucent ochre colour, viewed from the middle ear.

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 75 www.nzma.org.nz/journal VIEWPOINT

Figure 7: Cilia in the semi-circular canals represented by pipe cleaner.

Additionally, quirky details that showed the expertise, knowledge and network creative problem solving by Dr Auzoux were of people to recreate further anatomical also reproduced. These include pipe cleaner taonga in this collection, thus allowing their to describe the cilia hairs in the semi-cir- continued use in teaching. It may be that cular canals (Figure 7), 0.22 rifl e shells as resources from our enormously popular shafts to receive long pins from the external museum, so carefully collected and cared for ear part, and the long pins themselves by past and present staff, can be reproduced are presented in the same material as the and disseminated beyond in-house use, original—No. 8 wire. including back to some of the countries from While a lengthy and costly process, we where they originated, to aid the teaching of hope that this project will provide us with new generations of learners.

Competing interests: Nil. Acknowledgements: Grateful thanks to Steve Swindells, technical manager of the Dental Laboratory, Otago University Dental School. He showed generous enthusiasm for this project and kindly loaned grinding burrs and a base machine so that carving of the ear parts could be achieved. Author information: Louisa JM Baillie, Contracted Researcher and Sculptor, Department of Anatomy, University of Otago, Dunedin; Christopher L Smith, Curator of Anatomy Museum, Department of Anatomy, University of Otago, Dunedin. Corresponding author: Dr Louisa JM Baillie, Department of Anatomy, University of Otago, PO Box 56, Dunedin 9054. [email protected] URL: http://www.nzma.org.nz/journal/read-the-journal/all-issues/2010-2019/2018/vol-131-no-1473- 13-april-2018/7545

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 76 www.nzma.org.nz/journal VIEWPOINT

REFERENCES: 1. Cornwall J, Smith C. Laboratory yet. Anat Sci utilizing a novel cyclical Anatomical models by Educ. 2013; 6:209–210. “Observe-Refl ect-Draw-Ed- F.J. Steger (1845–1938): 5. Khot Z. The relative it-Repeat” learning process. the University of Otago effectiveness of comput- American Association of Collection. Eur. J. Anat. er-based and traditional Anatomists. 2017; 10:7–22. 2014; 18:209–211. resources for education doi:10.1002/ase.1616 2. Neuman F. Pots and in Anatomy. Anat Sci 9. Motoike HK, O’KaneRL, Pieces: The Anatomy Educ. 2013; 6:211–215. Lenchner E, Haspel C. Clay Museum of the Otago 6. Preece DM, William SB, modeling as a method to Medical School and how it Lam R, Weller R. “Let’s learn human muscles: A came to be. NZ Museums get physical”: Advantages community college study. Journal. 1993; 23:17–22. of a physical model over Anat Sci Ed. 2009; 2:19–23 3. Marreez YMA-H, Willems 3D computer models and doi:10.1002/ase.61 LNA. The Use of Medical Textbooks in learning 10. Linden D. Touch: The School Museums in imaging anatomy. Ant. Sci. Science of the Hand, Teaching “Anatomy” Educ. 2013; 6:216–224. Heart and Mind. London: Within an Integrated 7. Van Nuland SE, Rogers KA. Penguin Books, 2016. Medical Curriculum. Ch. 30. The skeletons in our closet: 11. Adam Gopnik. ‘Feel Me: Pp 267–275. In: Teaching E-learning tools and what What the new science Anatomy: A Practical happens when one side of touch says about Guide, DOI 10.1007/978-3- does not fi t all. American ourselves’. New Yorker. 319-08930-0_30. Springer Association of Anatomists. 16 May 2016; pp 56–57. International Publishing 2017; doi:10.1002/ase.1708 12. Minogue J, Jones G. Haptics Switzerland. 2015. 8. Backhouse M, Fitzpatrick in Education: Exploring 4. Drake R, Pawlina W. M, Hutchinson J, et al. an Untapped Sensory Anatomical Models: Improvements in anat- Modality. Review Educ Don’t Banish Them omy knowledge when Resea rch. 2006; 76:317–348. From the Anatomy

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 77 www.nzma.org.nz/journal CLINICAL CORRESPONDENCE

Hypercalcemia in a bodybuilder with cosmetic silicone injections Majdi Hamadeh, Jawad Fares, Khalil Maatouk, Mohamad Darwish

ABSTRACT Granulomatous hypercalcemia due to silicone injections is a rare disease with scarce literature. We present a case of a 35-year-old Caucasian male bodybuilder with multiple silicone injections in his upper extremities who developed hypercalcemia and urinary symptoms. He necessitated two sessions of dialysis. A biopsy of the upper arm showed granulomatous tissue. Corticosteroids were administered to relieve symptoms and reverse laboratory abnormalities. Silicone-induced hypercalcemia should be on high alert because of the increasing trend of body contour enhancements with injections, implants and fillers.

ypercalcemia is often due to unpre- On inspection, the large size of the upper dicted illness. When hyperparathy- limbs can be noted. The physical exam- Hroidism and malignancy are ruled ination was positive for mild left fl ank out, rare causes of hypercalcemia need tenderness, gynecomastia and bilateral mild to be investigated. Hypercalcemia due to testicular atrophy. The patient was admitted silicone injections has rarely been reported; for a suspected urinary tract infection. His however, there is concern that there will be initial laboratory fi ndings showed numerous more cases in the future as the popularity of WBCs on urine analysis, creatinine of 2.3mg/ cosmetic silicone is growing. dl (normal range 0.7–1.36), calcium of 13.1mg/dl (normal range 8.6–10.3), and uric Case report acid of 13.3mg/dl (normal range 3.6–7.7). The patient was primarily diagnosed with A 35-year-old Caucasian male bodybuilder a UTI associated with acute renal failure, was referred to the clinic with left fl ank hypercalcemia and hyperuricemia. Urine pain and dysuria. Symptoms started three culture was taken, and the patient was months ago with recurrent attacks of left started on antibiotics with ceftriaxone 2g by fl ank pain that is non-radiating, exacerbated intravenous-drip daily. He was also hydrated by movement and associated with dysuria with 1L normal saline every eight hours, and and intermittency. The patient’s history is started on allopurinol 300mg orally daily. positive for peptic ulcer disease, and he has had multiple injections of vitamins, testos- Pan CT scan revealed bilateral nephro- terone and growth hormones for increasing calcinosis and the presence of mesenteric body mass during the past 17 years. Injec- and retroperitoneal ganglions; no other tions were accompanied with supplements signifi cant fi ndings were noted. Despite and diuretics. Ten years before presentation, initial management, calcium levels in serum the patient had a session of multiple injec- remained high. The patient underwent two tions of silicone in the shoulder, arms and sessions of hemodialysis to restore calcium forearms, which was complicated later on back to normal. Further laboratory workup and necessitated a sub-mucosal excision of ruled out hyperparathyroidism, vitamin D a silicone mass from his right forearm. The intoxication, hyperthyroidism, malignancy, patient had no allergies and had no family sarcoidosis and multiple myeloma (Table 1). history of disease.

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 78 www.nzma.org.nz/journal CLINICAL CORRESPONDENCE

Table 1: Important lab fi ndings.

Test Result Unit Normal range

Parathyroid hormone (PTH) 3.95 pg/ml 15–65

Parathyroid hormone related-protein (PTH-rP) <0.8 pmol/L <1.3

Thyroid-stimulating hormone (TSH) 2.30 µIU/ml 0.27–4.2

Testosterone 2.40 ng/ml 2.8–8

Vitamin D2+D3 19.29 ng/ml 30–70

Calcium (urine) 462 mg/24hr 100–320

Ionized calcium 1.31 mMol/l 0.95–1.3

Angiotensin-converting enzyme (ACE) 60 ACE Units 20–70

Erythrocyte sedimentation rate (ESR) 30 mm/hr 0–15

A biopsy from the right triceps tendon showed active granulomas with giant cells, Discussion fi brous backgrounds and histiocytes (Figure Silicone injections have been used 1). Magnifi cation showed persistent silicone widely over the past 40 years for soft tissue particles in the tissue (Figure 2). The diag- enhancement. The most common of the nosis of silicone-induced granulomatous silicone polymers, the biologically inert hypercalcemia was made. The patient was medical fl uid 360, has been implicated in started on oral corticosteroids, 40mg daily a variety of adverse reactions, including for three weeks, and was tapered by 5mg granulomas, disfi guring nodules, and lymph- weekly afterwards. Calcium and creatinine edema, with latent periods ranging from levels gradually returned to normal, and three weeks to 20 years.1 symptoms resolved. A repeat blood test, The pathogenesis of granuloma formation one-month post treatment showed a calcium in similar cases is still not well established. level of 9.1mg/dl. T-cell activation triggered by infection,

Figure 1: A biopsy from the right triceps tendon showed fi brosed and sclerosed granulomas (1), active granulomas with giant cells (2) and fi brous backgrounds with histiocytes (3).

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 79 www.nzma.org.nz/journal CLINICAL CORRESPONDENCE

Figure 2: Magnifi cation with the polariser showed birefringent bodies corresponding to persistent silicone particles in the tissue (arrows).

trauma, adulterants added to the silicone, optimal effect. Therefore, dialysis can be or denatured host proteins has been lifesaving. proposed.2 Once activated, T-cells release cytokines, which promote granuloma Conclusion formation. Although granulomas represent Silicone-induced hypercalcemia should an adverse effect of silicone injections inde- be on high alert because of the increasing pendent of the purity of silicone used, they trend of body contour enhancements with have rarely been considered as a cause of injections, implants and fi llers. Dialysis can hypercalcemia.3,4 be lifesaving in resistant cases of silicone-in- This patient necessitated two sessions duced hypercalcemia. It is advised that of dialysis to reverse his persistent silicone injections be performed by trained hypercalcemia. It is vital to note that physicians using medical-grade silicone. bisphosphonates need 48 hours to reach

Competing interests: Nil. Author information: Majdi Hamadeh, Nephrology and Hypertension, Department of Nephrology and Hypertension, Al-Zahraa University Hospital, Beirut, Lebanon; Jawad Fares, Medicine, Faculty of Medicine, American University of Beirut, Beirut, Lebanon; Khalil Maatouk, Nephrology and Hypertension, Department of Nephrology and Hypertension, Al-Zahraa University Hospital, Beirut, Lebanon; Mohamad Darwish, Nephrology and Hypertension, Department of Nephrology and Hypertension, Al-Zahraa University Hospital, Beirut, Lebanon. Corresponding author: Jawad Fares, Faculty of Medicine, American University of Beirut, Beirut, Lebanon. [email protected] URL: http://www.nzma.org.nz/journal/read-the-journal/all-issues/2010-2019/2018/vol-131-no-1473- 13-april-2018/7546

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 80 www.nzma.org.nz/journal CLINICAL CORRESPONDENCE

REFERENCES: 1. Bigatà X, Ribera M, Bielsa uloma. Journal of drugs hypercalcemia. Case Rep I, Ferrándiz C. Adverse in dermatology. J Drugs Med 2013; 2013:807292. granulomatous reaction Dermatol 2006; 5(9):894–6. 4. Hamadeh M, Fares J. after cosmetic dermal 3. Agrawal N, Altiner S, Diagnosis and Management silicone injection. Dermatol Mezitis NH, Helbig S. of Hypercalcemia Associ- Surg 2001; 27(2):198–200. Silicone-induced granu- ated with Silicone-Induced 2. Desai AM, Browning loma after injection for Granuloma. Rev Assoc Med J, Rosen T. Etanercept cosmetic purposes: a rare Bras (1992) 2018. In press. therapy for silicone gran- entity of calcitriol-mediated

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 81 www.nzma.org.nz/journal CLINICAL CORRESPONDENCE

Purulent facial lesion Pallav Patni, Vinod Bhandari, Mahak Bhandari, Swadhin Raghuwanshi

50-year-old man was referred to A Cone Beam Computed Tomography dental hospital from the general (CBCT) after injecting a radio-opaque A surgery department in January 2012, contrast in the lesion was performed (Figure with a 10-month history of draining lesion 2). CBCT showed that contrast travelled from on left cheek (Figure 1). He had seen many cutaneous surface to roots of molar tooth different doctors, including physicians, in maxilla. The condition was diagnosed as dermatologists, general surgeons and plastic chronic suppurative odontogenic infection surgeons before reporting to our hospital. with facial cutaneous sinus tract. He had been given antibacterial, antifungal We treated our patient with root canal and topical corticosteroid treatment. He had treatment (removal of infected pulp of also undergone surgical intervention twice. tooth). At a six-year follow-up in January He also reported history of pain with molar 2018, the patient is doing well, with no cuta- tooth on the same side few years back. Phys- neous drainage. The lesion has healed with a ical examination was normal apart from minor scar formation (Figure 3). skin lesion.

Figure 1: Cutaneous lesion on the face.

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 82 www.nzma.org.nz/journal CLINICAL CORRESPONDENCE

Figure 2: Reconstructed image of CBCT confi rming dental involvement.

correct diagnosis of such lesions should be Discussion suspected by gross appearance of lesion. Cases of facial lesions of dental origin They present as erythematous, smooth and have been commonly reported in medical, non-tender lesions of 1mm to 20mm in dental and dermatology literature.1 They diameter, with crusty and periodic drainage are frequently misdiagnosed due to wide in most cases. Patient may present with differential diagnosis. Differential diag- history of dental pain or trauma few years nosis includes actinomycosis, pustule, back. Of all such cases, approximately 80% osteomyelitis, neoplasms, carbuncle, are mandibular and 20% are maxillary in infected epidermoid cyst, pyogenic gran- origin.3 A simple dental procedure like root uloma, chronic tuberculosis, salivary gland canal therapy or extraction of involved tooth fi stula and gumma of tertiary syphilis.2 The may lead to successful outcome.

Figure 3: At a six-year follow-up, the cutaneous lesion healed with a minimal scar formation.

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 83 www.nzma.org.nz/journal CLINICAL CORRESPONDENCE

Competing interests: Nil. Acknowledgements: The authors acknowledge Dr Manjushree Bhandari, Chairperson, SAIMS towards their guidance in preparation of this manuscript. Author information: Pallav Mahesh Patni, Department of Conservative Dentistry and Endodontics, Sri Aurobindo College of Dentistry (SAIMS) and Hospital, Madhya Pradesh Medical Science University, Indore, MP, India; Vinod Bhandari, Department of General Surgery, Sri Aurobindo Institute of Medical Sciences (SAIMS) and Hospital, Madhya Pradesh Medical Science University, Indore, MP, India; Mahak Bhandari, Department of General Surgery, Sri Aurobindo Institute of Medical Sciences (SAIMS) and Hospital, Madhya Pradesh Medical Science University, Indore, MP, India; Bhandari Hospital & Research Center, Indore, MP, India; Swadhin Raghuwanshi, Department of Conservative Dentistry and Endodontics, Sri Aurobindo College of Dentistry (SAIMS) and Hospital, Madhya Pradesh Medical Science University, Indore, MP, India. Corresponding author: Dr Pallav Mahesh Patni, Professor, Department of Conservative Dentistry and Endodontics, Sri Aurobindo College of Dentistry (SAIMS) and Hospital, Madhya Pradesh Medical Science University, Indore, MP, India. [email protected] URL: http://www.nzma.org.nz/journal/read-the-journal/all-issues/2010-2019/2018/vol-131-no-1473- 13-april-2018/7547

REFERENCES: 1. Patni PM, Jain P, Patni MJ. Kobayashi N, Asada H. 3. Giménez-García R, Marti- Cutaneous sinus tract of Usefulness of ultrasonogra- nez-Vera F, Fuentes-Vera L. dental origin. Postgrad phy for rapidly diagnosing Cutaneous Sinus Tracts of Med J. 2016; 92:625. cutaneous sinus tracts Odontogenic Origin: Two 2. Shobatake C, Miyagawa of dental origin. Eur J Case Reports. J Am Board F, Fukumoto T, Hirai T, Dermatol. 2014; 24:683–7. Fam Med. 2015; 28:838–40.

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 84 www.nzma.org.nz/journal RESEARCH LETTER

If only Teina Pora had a MedicAlert bracelet Anita Gibbs

eina Pora was an 18-year-old Māori of vulnerable groups— children in care, male who was convicted in 1994 of the adopted children, offenders, indigenous rape and murder of Susan Burdett in populations and those with other disorders T1 2,7 1992. Pora did not commit these crimes but or mental health issues. FASD rates are he spent 21 years in prison before his case higher than rates of autism or ADHD.3,7 The was heard by the UK Privy Council in Lon- productivity losses due to FASD in New don and his convictions were quashed.1 Cru- Zealand are estimated to range from $49 cial to his release and acquittal was the dis- million to $200 million and the social and covery that Pora had Fetal Alcohol Spectrum emotional burden on families supporting Disorder (FASD), a neuro-disability arising those with FASD is immense.8 Given these from prenatal alcohol exposure with lifelong wide ranging costs, surely it is time to pay impacts, most noticeably impairments of more attention to the scale of FASD as both a executive functioning leading to impulsive- health issue and social problem.9 ness, poor reasoning, problem-solving skills In the last 10 years, New Zealand has and memory defi cits. The primary impair- begun to implement measures to assess and ments of FASD include physical and neuro- diagnose FASD, as well as develop action for logical impacts and they often co-occur with intervention, but this is currently happening conduct and language disorders, and mental in only a few places.9 FASD assessments need health issues, and lead many with FASD to to be conducted by a multidisciplinary team poor educational outcomes, school exclu- and due to their complexity can cost up to sion, relationship problems, addictions and $8,000. The medical and allied health profes- 2–4 arrests. Teina Pora admitted to crimes he sions are in need of training and upskilling did not commit and he had a disability that to screen, understand and help individuals 1 made him vulnerable to doing this. I won- and families where FASD is an issue as well der if his story would have been different as organisational support that enables them had he been diagnosed with FASD as a child to sustain these important services. The and offered the protection of early interven- New Zealand FASD Action Plan9 has led to tion as a well as a raft of others supports, a few initiatives to pilot training materials for example, wearing a MedicAlert bracelet and hopefully increase clinical diagnostic as young adults with FASD in Canada now and assessment capacity. The action plan, 5 do. By being diagnosed and wearing a while detailed and positive, has only $12 MedicAlert device, perhaps Pora might have million from the public purse allocated to its received a more appropriate response from implementation over four years, compared arresting police offi cers and medical profes- for example to Alberta, Canada (population sionals involved with his case. similar to New Zealand), which spends A study just out from Western Australia around $18 million per annum on FASD confi rms that up to 90% of young people services and research to cost-effectively incarcerated have a severe neuro-im- prevent and treat FASD. Also sobering is the pairment, many of whom will have FASD, fact that the New Zealand Treasury levied but more often not than most of these almost one billion dollars in alcohol excise will be undiagnosed.6 In New Zealand, we duties last year,10 and this has not been used currently have no prevalence data for FASD to fund action to increase training, services but robust international evidence tells us or research in this now well-established that up to 5% of the general population will area of neuro-disability. have FASD, and a much greater proportion

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 85 www.nzma.org.nz/journal RESEARCH LETTER

In a recent UK Lancet publication, philos- guidelines available for a range of health opher Havi Carel points out the extensive professionals and under the New Zealand everyday struggle that children with FASD FASD Plan more training is coming.2,9 have and that we all need to understand this People with FASD can fl ourish with experience and accommodate the ‘virtue in targeted help and support. We need to at 11 defi cit’ that having this disability entails. the very least ensure that their rights as Each year New Zealand has up to 2,000 disabled citizens are upheld.9 They should children born with likely FASD, and with have access to good-quality health provision, 1,000s of other undiagnosed children, youth especially early screening and assessment, and adults, it is time to offer more support. but they should also be acknowledged as Even a small intervention like MedicAlert having a neuro-disability that warrants life- bracelets can offset some of the negative course provision of support, and the right to impacts of school exclusion or arrest. The be the defi ners of what that support might evidence-based research, mostly out of entail to enable them to live good lives in the North America, highlights the importance of same way as people with other impairments early diagnosis and intervention, having a can live good lives. Teina Pora lost 21 years stable home life and ensuring that a range of of his life in prison, yet he showed great appropriate accommodations and targeted ‘virtue in defi cit’ and was poorly compen- interventions that help build self-control are sated, but his story compels us to intervene 2,3,12 implemented. Medical and allied health and ensure the next generation of Pora’s are professionals can play an important role treated better and helped before the damage in both the prevention of FASD and inter- is done. vention when FASD is suspected. There are

Competing interests: Nil. Author information: Anita Gibbs, Associate Professor, Department of Sociology, Gender and Social Work, University of Otago, Dunedin. Corresponding author: Anita Gibbs, Department of Sociology, Gender and Social Work, PO Box 56, University of Otago, Dunedin 9054. [email protected] URL: http://www.nzma.org.nz/journal/read-the-journal/all-issues/2010-2019/2018/vol-131-no-1473- 13-april-2018/7548

REFERENCES: 1. Freckelton I. Fetal Alcohol Journal of Psychology and orders-could-be-a-hid- Spectrum Disorders, Psychiatry. 2017; 1(1):1–3. den-epidemic-52835 Expert Evidence and the Available from: http://www. 4. Popova S, Lange S, Shield Unreliability of Admissions oatext.com/fetal-alcohol- K, et al. Comorbidity of during Police Interviews. spectrum-disorder-fasd- fetal alcohol spectrum Psychiatry, Psychology and a-call-on-mental-health- disorder: a systematic Law. 2016; 23(2):173–183. treatment-professionals- review and meta-analysis. Available from: http:// to-become-informed.php Lancet. 2016; 387:978–987. doi.org/10.1080/1321 3. Cook P, Mukherjee R. How Available from: https:// 8719.2016.1149898 foetal alcohol spectrum doi.org/10.1016/ 2. Brown J, Harr D, Morgan disorders could be a S0140-6736(15)01345-8 S, et al. Fetal Alcohol hidden epidemic. The 5. Malone, G. How a bracelet Spectrum Disorder (FASD): Conversation, January 13, is helping fi rst responders A call on mental health 2016. Available from: http:// understand fetal alcohol treatment professionals theconversation.com/how-foe- spectrum disorder. TVO, to become informed. tal-alcohol-spectrum-dis-

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 86 www.nzma.org.nz/journal RESEARCH LETTER

Jan 12, 2017. Available ders in 4 US Communities. Update 2017. 14 December from: http://tvo.org/ JAMA. 2018; 319(5):474– 2017. Available from: http:// article/current-affairs/ 482. doi:10.1001/ www.treasury.govt.nz/ shared-values/how-a-brace- jama.2017.21896 budget/forecasts/hyefu2017 let-is-helping-fi rst-respond- 8. Easton B, Burd L, Rehm J, 11. Carel H. Virtue in defi cit: ers-understand-fetal-alco- et al. Productivity losses the 9-year-old hero. Lancet. hol-spectrum-disorder associated with fetal 2017; 389:1094–1095. 6. Bower C, Watkins R, alcohol spectrum disorder Available from: http:// Mutch R, et al. Fetal in New Zealand. NZMJ. www.thelancet.com/ alcohol spectrum disor- 2016; 129(1440):72–83. journals/lancet/article/ der and youth justice: a Available from: http:// PIIS0140-6736(17)30709-2 prevalence study among www.nzma.org.nz/ 12. Petrenko C, Alto M. Inter- young people sentenced journal/read- the-journal/ ventions in fetal alcohol to detention in Western all-issues/2010-2019/2016/ spectrum disorders: An Australia. BMJ Open. 2018; vol-129-no-1440/6974 international perspective 8:e019605. doi:10.1136/ 9. Author and European Journal of bmjopen-2017-019605 colleagues 2017. Medical Genetics. 2017; 7. May P, Chambers C, Kalberg 10. Treasury. Half Year 60(1):79–91. Available from: W, et al. Prevalence of Fetal Economic and Fiscal http://doi.org/10.1016/j. Alcohol Spectrum Disor- ejmg.2016 .10.005

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 87 www.nzma.org.nz/journal RESEARCH LETTER

Permanent pacemaker implantation a• er cardiac surgery: rates, predictors and a novel risk score Tom Kai Ming Wang, Diego Arroyo, Andrew Martin, Alastair McGeorge, Michael Gillham

isturbances in native cardiac electric types were divided into category 1: isolated conduction after cardiac surgery coronary, 2: 1 valve+/-coronary, 3: 2 valves- Dmanifest as transient or persistent +/-coronary and 4: other (such as aortic bradyarrhythmia, the latter necessitating surgery, congenital heart surgery, triple permanent pacemaker (PPM) implantation valve surgery). in a minority of patients after observation Mean+/-standard deviation and 1 for 5–7 days. Accurate pre-operative risk percentage (frequency) were used to present stratifi cation for PPM requirement may quantitative and qualitative variables, reduce bradyarrhythmic complications and Hosmer-Lemeshow and Fisher’s Exact and length of hospital stay, but this is done Tests used for univariable analysis of these suboptimally in the clinical setting, and respectively. Logistic regression was used studies have reported mixed fi ndings of to identify independent predictors of PPM 2–6 predictors. We aim to review and identify implantation, and construct an additive predictors of PPM implantation after cardiac model for predicting this. Its accuracy is surgery at our centre. assessed using the area under the receiv- er-operator characteristics curve c-statistic. Methods P-values <0.05 were deemed statistically All patients who had cardiac surgery at signifi cant and all tests were two-tailed. Auckland City Hospital during June 2014– SPSS (Version 17.0, SPSS Inc., Chicago, IL, October 2016 who had PPM implantation USA) was used for statistical analyses. during the same admission and after cardiac surgery were reviewed. A random sample Results of the identical number of patients of other Of 2,446 cardiac surgeries undertaken, cardiac surgery patients were obtained as 4.0% (98) received a post-operative PPM controls. Clinical characteristics including during the study period, so 98 controls were demographics, past history, presentation, selected for comparison, with characteristics pre-operative ECG, surgery type and times presented in Table 1. PPM patients were were extracted from computerised records on average older, had higher prevalence 7 and EuroSCORE II was calculated. of previous surgery and lower prevalence Pre-operative ECGs were divided into of previous myocardial infarction, higher 4 categories defi ned as 1: normal, 2: EuroSCORE II, higher proportion with arrhythmia (in the absence of any heart higher grade block and valvular surgery block), 3: low-grade block (not a standalone in univariate analysis. In particular, PPM pacemaker indication such as fi rst degree, incidence was 0.6% (7/1165) for coronary left or right bundle branch, bifasiscular surgery, 7.1% (45/636) for single valve+/-cor- block, Wenckebach), and 4: high grade block onary surgery, 9.0% (12/133) for double (that is pacemaker indication, eg, 2nd degree valve+/-coronary surgery and 7.0% (34/489) type 2 or complete heart block). Surgery for other cardiac surgeries.

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 88 www.nzma.org.nz/journal RESEARCH LETTER

Table 1: Clinical characteristics and outcomes of cardiac surgery receiving PPM or not (controls).

No PPM PPM P N 98 98

Demographics

Age (years) 63.7+/-12.8 68.1+/-12.0 0.007 Female 29.6% (29) 28.6% (28) 1.000

Body mass index 28.2+/-6.0 28.6+/-5.9 0.263

Past history

Previous cardiac surgery 7.1% (7) 17.3% (17) 0.048 Hypertension 76.5% (75) 71.4% (70) 0.515

Diabetes mellitus 25.5% (25) 29.6% (29) 0.632

Myocardial infarction 23.5% (23) 11.2% (11) 0.037 Creatinine clearance (mL/min) 76+/-29 83+/-44 0.490

Presentation NYHA class 2.1+/-0.7 2.2+/-0.7 0.380

Elective surgery 39.8% (39) 41.8% (41) 0.799

Urgent surgery 54.1% (53) 54.1% (53)

Emergency surgery 6.1% (6) 4.1% (4)

EuroSCORE II (%) 3.7+/-7.8% 6.6+/-8.3% <0.001

Pre-operative ECG <0.001 Normal 64.3% (63) 30.6% (30)

Arrhythmia 14.3% (14) 24.5% (24)

Low-grade block 21.4% (21) 36.7% (36) 0.027

High-grade block 0.0% (0) 8.2% (8) 0.007

Broad complex (QRS>120ms) 7.1% (7) 21.4% (21) <0.001 Surgery

Types <0.001 Coronary only 50.0% (49) 7.1% (7)

1 Valve+/-coronary 27.6% (27) 45.9% (45)

2 Valve+/-coronary 6.1% (6) 12.2% (12)

Other 16.3% (16) 34.7% (34)

CPB time 115+/-63 144+/-66 <0.001

Cross clamp time 77+/-43 102+/-58 0.001 Pacemaker indications Arrhythmia with bradycardia 20.4% (20)

Low-grade block 6.1% (6)

High-grade block 73.4% (72)

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 89 www.nzma.org.nz/journal RESEARCH LETTER

Table 2: Area under the curve (c-statistic) for predicting PPM implantation after cardiac surgery.

Predictor C-statistic 95% confidence interval P

Age 0.61 0.53–0.69 0.007

EuroSCORE II 0.65 0.57–0.73 <0.001

Pre-operative ECG 0.68 0.61–0.76 <0.001 Operation type 0.72 0.64–0.79 <0.001

PPM risk model* 0.78 0.72–0.85 <0.001

*Change to risk model is based on two parameters and total score of 7: a) pre-operative ECG score 0 for normal, 1 for arrhythmia, 2 for low-grade block and 3 for high-grade block; and b) type of surgery score 0 for isolated coronary, 2 for 1 valve+/-coronary or other, 4 for 2 valve+/-coronary.

Multivariable analysis performed based knowledge, we have developed the fi rst risk on the above univariable analysis param- score for predicting PPM requirement after eters, and combining category 4 and 2 of cardiac surgery, based on the two important operation type as one group, identifi ed predictors of pre-operative ECG and surgery pre-operative ECG and operation type as type, which performed moderately well with independent predictors of PPM implan- c=0.78. The score out of 7 can be clinically tation after cardiac surgery. Odds ratios utilised as 0–1 as low risk, 2–3 as moderate (95% confi dence intervals) per category risk and 4–7 as high risk. The clinical impli- were 1.65 (1.08–2.53) and 5.2 (2.47–10.9) cation is that in low-risk patients, temporary respectively. As such, a simple additive risk epicardial wires could be turned off and model out of seven of these two variables removed earlier, and those moderate and was created: a) pre-operative ECG score 0 for particularly high-risk patients would need normal, 1 for arrhythmia, 2 for low-grade close monitoring early on for, and ongoing block and 3 for high-grade block; and b) monitoring with low threshold of PPM type of surgery score 0 for isolated coronary, implantation of, any conduction disturbance 2 for 1 valve+/-coronary or other, 4 for 2 post-operatively. The simplicity of the score valve+/-coronary. C-statistics of this model should aid rather than hinder utility even if and individual parameters for PPM implan- performance maybe somewhat hindered. tation predictor are shown in Table 2. The This study has several limitations. It is a risk model had the highest c-statistic of 0.78, single-centre retrospective observational and if the score cutpoint was set at 3, gave study. The study sample size was small so sensitivity of 70.4% and specifi city of 77.6%. power was limited. It was a case-control rather than cohort study design, and the Discussion controls, although randomly selected, PPM was implanted in 4.0% of our cardiac weren't matched to the pacemaker group, surgery cohort, comparable to the 1.4–8.5% however this allowed us to assess and incidence reported in contemporary highlight the differences in characteristics studies.2–6 Pre-operative ECG fi ndings of between the two groups. We did not specifi - bundle branch block or other heart block,2–5 cally examine long-term outcomes, including and valvular heart surgery4,5 were also survival, pacing burden and late pacemaker found to be predictors in other studies like dependency, but focused only on what ours, although we found that multi-valve factors predicted patients getting PPM in the surgery had independently higher risk than fi rst place. The score has not been exter- single-valve surgery. Other predictors previ- nally validated, which would be a next step. ously identifi ed include redo operation and Nevertheless, our analysis identifi ed two pulmonary hypertension,5 age, emergency important parameters of pre-operative ECG admission, diabetes, renal impairment and type of cardiac surgery and constructed and heart failure,6 which are common a user-friendly additive risk model to predict risk factors to operative mortality.7 To our PPM after cardiac surgery with moderately good accuracy.

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 90 www.nzma.org.nz/journal RESEARCH LETTER

Competing interests: Nil. Author information: Tom Kai Ming Wang, TKMW Cardiology Advanced Trainee Registrar, Cardiovascular Intensive Care Unit, Auckland City Hospital, Auckland; Green Lane Cardiovascular Service, Auckland City Hospital, Auckland; Diego Arroyoa, DA Cardiologist, Department of Cardiology, University and Hospital Fribourg, Fribourg, Switzerland; Andrew Martin, Electrophysiology Cardiologist, Green Lane Cardiovascular Service, Auckland City Hospital, Auckland; Alastair McGeorge, Cardiovascular Intensivist and Anaesthetist, Cardiovascular Intensive Care Unit, Auckland City Hospital, Auckland; Michael Gillham, Cardiovascular Intensivist and Anaesthetist, Cardiovascular Intensive Care Unit, Auckland City Hospital, Auckland. Corresponding author: Dr Tom Kai Ming Wang, Auckland City Hospital, 2 Grafton Road, Grafton, Auckland. [email protected] URL: http://www.nzma.org.nz/journal/read-the-journal/all-issues/2010-2019/2018/vol-131-no-1473- 13-april-2018/7551

REFERENCES: 1. Brignole M, Auricchio valve replacement. J Card AlRuwaili N, Shoukri M, et A, Baron-Esquivias G, Surg 2006; 21:211–215. al. Predictors of Permanent Bordachar P, Boriani 3. Dawkins S, Hobson Pacemaker Implantation G, Breithardt OA, et al. AR, Kalra PR, Tang AT, After Coronary Artery 2013 ESC Guidelines on Monro JL, Dawkins KD. Bypass Grafting and Valve cardiac pacing and cardiac Permanent pacemaker Surgery in Adult Patients in resynchronization therapy: implantation after isolated Current Surgical Era. Cardi- the Task Force on cardiac aortic valve replacement: ol Res 2016; 7:123–129. pacing and resynchro- incidence, indications, and 6. Leyva F, Qiu T, McNulty nization therapy of the predictors. Ann Thorac D, Evison F, Marshall H, European Society of Cardi- Surg. 2008; 85:108–12. Gasparini M. Long-term ology (ESC). Developed 4. Merin O, Ilan M, Oren A, requirement for pace- in collaboration with the Fink D, Deeb M, Bitran D, et maker implantation after European Heart Rhythm al. Permanent pacemaker cardiac valve replacement Association (EHRA). Eur implantation following surgery. Heart Rhythm. Heart J. 2013; 34:2281–329. cardiac surgery: indications 2017; 14:529–534. 2. Erdogan HB, Kayalar N, and long-term follow-up. 7. Nashef SA, Roques F, Ardal H, Omeroglu SN, Pacing Clin Electrophysi- Sharples LD, Nilsson Kirali K, Guler M, et al. Risk ol 2009; 32:7–12. J, Smith C, Goldstone factors for requirement 5. Al-Ghamdi B, Mallawi Y, AR, et al. EuroSCORE of permanent pacemaker Shafquat A, Ledesma A, II. Eur J Cardiothorac implantation after aortic Surg 2012; 41:734–44.

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 91 www.nzma.org.nz/journal LETTER

Gnome medicine: what does genomic medicine mean to our patients and us? Sara Filoche, Kevin Dew, Anthony Dowell

“Gnome medicine did you say?” With large numbers of individuals’ genomic more than a hint of incredulity, this was DNA, linked with other health, lifestyle and the response from my (SF) mum when I administrative data6,7—which raises signif- talked to her about my work. We then went icant, and as yet unresolved, consideration on to laugh about the misunderstanding. around governance, caretaking, informed I use this example, not to humiliate my consent and sharing (or not) fi ndings with mum (she has given permission for me family members.8,9 to use her response) but to highlight the Genomic medicine is poised to transform divide between clinicians, scientists and patient care, and it will become more 1,2 the everyday person. My mum is not common as genomic medicine is being alone in her response. In the UK there is mainstreamed (ie, no longer a specialist an engagement project “Socialising the service).3 But as one of the investigators Genome” funded by Genomics England, the from the socialising the genome project Wellcome Trust and the Wellcome Trust stated, “We don’t yet know how to make Sanger Institute, which aims to “explore genomics an everyday conversation for what people already understand about DNA people currently unconnected to it”. If the and genomics—even if they think they know people that genomic medicine is intending nothing—and how they are currently talking to benefi t are excluded from the dialogue about it.” as this technology advances, the divide will In their letter, Parry and Middleton only widen and it will remain inaccessible discussed how some genomics professionals (and distrusted) on many different levels. “argue fi ercely that the public should be How about health providers? The uptake educated to use and understand technically of genomic technology into clinical practice 1 precise genomics terminology.” However, will depend on providers’ perspectives Parry and Middleton went on to say that of its utility in patient care. Currently we they felt that insisting on such an approach don’t have a good handle on what these are, would present “a marked barrier to commu- nor on the educational needs.10 If we look nication and also creates an unhelpful overseas, evidence barriers to adopting power differential of expert versus other.” genomic medicine are cited as many, with Genomic medicine, sometimes also “variable knowledge and comfort with known as personalised medicine, is a way genetic concepts” being leading concerns.10 to customise medical care to your body’s The rise of direct-to-consumer testing unique genetic makeup—where treatment (DTC) with kits such as 23andMe will likely 3 plans can be tailored to the individual. increasingly reach our practices, as patients Rather than looking at one gene, genomic may turn up for help interpreting their medicine looks at all of the genes, using tech- health report, as an example, because they niques such as next-generation sequencing. have been found to have the ε4 variant in It takes into account family health history the APOE gene associated with late onset and environmental factors. A key aspect of Alzheimer’s disease.11 the success of genomic medicine is related As health researchers, medical sociologists to public acceptance;1–5 in particular around and practitioners, we would like to highlight collecting family health histories and the a gap in our engagement and education development of biobanks that contain

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 92 www.nzma.org.nz/journal LETTER

around genomic medicine in Aotearoa what practitioners want and need to deliver New Zealand with both our patients and it. Are we genome ready? We would contend providers. We don’t know what messages not really, and we would welcome the about genomics are meaningful to people opportunity to work with health providers, (our patients) in Aotearoa, nor do we know stakeholders and researchers to become so.

Competing interests: Nil. Author information: Sara Filoche, Department of Obstetrics and Gynaecology and Department of Pathology and Molecular Medicine, University of Otago, Wellington; Kevin Dew, Department of Obstetrics and Gynaecology and Department of Pathology and Molecular Medicine, University of Otago, Wellington; Anthony Dowell, Department of Obstetrics and Gynaecology and Department of Pathology and Molecular Medicine, University of Otago, Wellington. Corresponding author: Sara Filoche, Department of Obstetrics and Gynaecology and Department of Pathology and Molecular Medicine, University of Otago, Wellington. sara.fi [email protected] URL: http://www.nzma.org.nz/journal/read-the-journal/all-issues/2010-2019/2018/vol-131-no-1473- 13-april-2018/7552

REFERENCES: 1. Parry V, Middleton A. 5. Etchegary H, Green J, in clinical practice and Socialising the genome. Parfrey P, et al. Community research? Clinical Ethics Lancet (London, England) engagement with genet- 2017; 12(3):150–65. doi: 2017; 389(10079):1603–04. ics: public perceptions 10.1177/1477750917704156 doi: 10.1016/s0140- and expectations about 9. Gilbar R. Communicating 6736(17)31011-5 [published genetics research. Health genetic information in Online First: 2017/04/27]. Expectations: an interna- the family: the familial 2. Middleton A. Your DNA, tional journal of public relationship as the Your Say. The New participation in health care forgotten factor. Journal Bioethics: a multidis- and health policy 2015; of Medical Ethics 2007; ciplinary journal of 18(5):1413–25. doi: 10.1111/ 33(7):390–93. doi: 10.1136/ biotechnology and the hex.12122 [published jme.2006.017467 Online First: 2013/08/24]. body 2017; 23(1):74–80. doi: 10. Raghavan S, Vassy JL. Do 10.1080/20502877 6. Hudson M, Beaton A, Milne physicians think genomic .2017.1314890 [published M, et al. Te Mata Ira Guide- medicine will be useful for Online First: 2017/05/19]. lines for Genomic Research patient care? Personalized 3. Chief Medical Offi cer. with Māori: University Medicine 2014; 11(4):424– Annual Report of the Chief of Waikato, 2016. 33. doi: 10.2217/pme.14.25 Medical Offi cer 2016: 7. Dheensa S, Fenwick A, 11. Guerreiro RJ, Gustafson Generation Genome, 2016. Lucassen A. Approaching DR, Hardy J. THE GENET- 4. Etchegary H, Green J, confi dentiality at a familial IC ARCHITECTURE OF Dicks E, et al. Consulting level in genomic medicine: ALZHEIMER’S DISEASE: the community: public a focus group study with BEYOND APP, PSENs expectations and attitudes healthcare professionals. AND APOE. Neurobi- about genetics research. BMJ Open 2017; 7(2) doi: ology of Aging 2012; European Journal of e012443. doi:10.1136/ 33(3):437– 56. doi: 10.1016/j. Human Genetics : EJHG bmjopen-2016-012443. neurobiolaging.2010.03.025 2013; 21(12):1338–43. 8. Dove ES, Kelly SE, Lucivero doi: 10.1038/ejhg.2013.64 F, et al. Beyond individ- [published Online ualism: Is there a place First: 2013/04/18]. for relational autonomy

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 93 www.nzma.org.nz/journal METHUSELAH

Aspirin or rivaroxaban for VTE prophylaxis after hip or knee arthroplasty? Clinical trials and meta-analyses have suggested that aspirin may be effective for the prevention of venous thromboembolism (proximal deep-vein thrombosis or pulmonary embolism) after total hip or total knee arthroplasty, but comparisons with direct oral anticoag- ulants are lacking for prophylaxis beyond hospital discharge. This report is of a trial that addresses this issue. A total of 3,424 patients (1,804 undergoing total hip arthroplasty and 1,620 undergoing total knee arthroplasty) were enrolled in the trial. All received oral rivaroxaban (10mg) until the fi fth postoperative day. The patients were then randomised to continue rivaroxaban or switch to aspirin (81mg) daily for nine days (knee patients) or 30 days for the hip patients. It was concluded that among patients who received fi ve days of rivaroxaban prophy- laxis after total hip or total knee arthroplasty, extended prophylaxis with aspirin was not signifi cantly different from rivaroxaban in the prevention of symptomatic venous thrombo- embolism. Clinically important bleeding risks were similar in both groups of patients. N Engl J Med 2018; 378: 699–707 Adjunctive rifampicin for Staphylococcus aureus bacteraemia Staphylococcus aureus bacteraemia is a common cause of severe community-acquired and hospital-acquired infection worldwide. Adjunctive rifampicin has been hypothesised to improve results when used in addition to standard treatment. This report is of a randomised trial which aimed to clarify this issue. Seven hundred and fi fty-eight patients were involved. All received treatment with an anti-staphylococcal drug—penicillin, fl ucloxacillin or methicillin. Those with methicillin-re- sistant organisms were treated with a glycopeptide (vancomycin). Half of the patients were also treated with rifampicin and the other half a placebo. The result of this trial was that adjunctive rifampicin provided no overall benefi t over standard antibiotic therapy in adults with S aureus bacteraemia. Lancet 2018; 391:668–78 Burden of atrial fi brillation in Māori and Pacifi c people in New Zealand Atrial fi brillation (AF) is a major risk factor for stroke and cardiovascular events. Previous studies suggest that Māori and Pacifi c people have a higher incidence of AF than people of European ancestry. Data obtained from 37 New Zealand general practices is reviewed in this study, which involved 135,840 subjects, including 19,918 Māori and 43,634 Pacifi c people. The incidence of AF was found to be similar to that noted in other countries, and strongly age-related. However, it was discovered that AF was diagnosed 10 years earlier in the Polynesian patients compared with their European counterparts— Māori 60 years, Pacifi c 61 years and European 71 years. In view of these fi ndings it was suggested that AF screening and stroke thromboprophylaxis in Māori and Pacifi c people could start below the age of 65 years in New Zealand. Internal Medicine Journal 2018; 48:301–309

URL: http://www.nzma.org.nz/journal/read-the-journal/all-issues/2010-2019/2018/vol-131-no-1473- 13-april-2018/7549

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 94 www.nzma.org.nz/journal 100 YEARS AGO

Innocent Intra-Orbital Growth By Edgar Whitaker, M.R.C.S., Ophthalmic Surgeon, Palmerston North Hospital

he following is a rare case of innocent swelling was twice as big as fi fteen months intra-orbital new growth having for- previously and the patient thought that it Tmidable effects upon the eyesight:— had grown rapidly lately. Patient, a married woman, just suckling I advised removal of the whole eye and her fourth child, came under my notice in tumour and sent her down to Dr. Harty for November, 1916. She then had prominence an opinion. He was away, but Dr. Webster of the left eye for twelve years, but “the last was kind enough to see her and had the seven months much more so.” This period orbit skiagraphed. It did not show very coincided, in her opinion, with the bearing much, save that the orbital cavity appeared and suckling of her child. No pain in the eye, less distinct than usual. He thought the very little discomfort. tumour non-malignant and advised removal The condition then was: Left eye as completely as possible. proptosed half an inch in front of level of Patient was admitted to hospital on her the right eye, which is normal. The eyelid return and I removed the eyeball. A smooth, can completely cover the ball and the round, blue-grey walled cyst was exposed, movements are not restricted. There is no about the size of a large walnut, unattached perception or projection of light. There is no to the orbital wall save to the optic foramen. pulsation in the ball and it cannot be pushed The lengthened optic nerve was intimately back into orbit. attached to one wall of the cyst, which The ophthalmoscope shows a small patch contained pale yellow clear fl uid. of pigment on the upper inner side of the The whole cyst was removed and sent to disc, the disc being very white. The veins Professor Murray Drennan, whose report is dilated double normal. Arteries about as attached. Recovery was uneventful. usual. The fundus was quite intact, there Report by Professor Murray Drennan:— was no irregularity of surface of any kind “The eyeball shows of little note. Cyst such as would be shown by pressure of presents considerable diffi culty. The optic growth or invasion. The media were clear. nerve enters on one side and then is lost in The diagnosis was, I thought, clear that the cyst. The wall of cyst consists of fi brous it was not a malignant tumour, and as the and granulation tissue, in parts of which are patient had a large goitre and was suckling many vessels with thick hyaline walls. Many there was a possibility of the swelling going fat-laden cells are present in cyst wall and down if suckling and stress ceased. in optic nerve; also cells with blood pigment. The swelling did go down considerably. The appearances suggest that a tumour, probably myxomatous, has been originally In February, 1918, the patient came with present on optic nerve. Haemorrhage into it the history that the last few months the eye has occurred with organisation, so that now had become very much more prominent. only the latter appearances remain.” On examination there was still no invasion of eyeball, no pulsation, no pain, but the

URL: http://www.nzma.org.nz/journal/read-the-journal/all-issues/2010-2019/2018/vol-131-no-1473- 13-april-2018/7550

NZMJ 13 April 2018, Vol 131 No 1473 ISSN 1175-8716 © NZMA 95 www.nzma.org.nz/journal