Childhood asthma in New Zealand: the impact of ongoing socioeconomic disadvantage (2010-2019)
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
article Childhood asthma in New Zealand: the impact of ongoing socioeconomic disadvantage (2010–2019) Deborah Schlichting, Tayaza Fadason, Cameron C Grant, Justin O’Sullivan ABSTRACT AIM: To document trends in number and cost of asthma hospital admissions and asthma prescriptions in children (0–14 years) from 2010–2019 in New Zealand. METHOD: A retrospective analysis of public hospital admission and pharmaceutical prescription data. RESULTS: The dataset included 39,731 hospitalisations with asthma as a discharge diagnosis and 5,512,856 prescriptions for asthma medication in children ≤14 years old. From 2010 to 2019, there was a 45% reduction in the number of asthma hospitalisations and an 18% reduction in prescriptions attributable to asthma. Declines were evident for both Māori and non-Māori children. However, Māori children were hospitalised with asthma at twice the rate of non-Māori children (7.2/1,000 versus 3.5/1,000, p
article 41/100,000.6 The prevalence of childhood asthma in New Zealand in 2019 was Methods estimated to be 13% by the New Zealand Ethics Health Survey, which used a case definition Permission to use the data was provided dependent on each child’s parents or care- by the Ministry of Health. This study did givers having ever been told by a doctor not require submission to the Health and that the child has asthma and whether Disability Ethics Commission (HDEC) as it is the child is currently taking treatments not within the scope of HDEC review. Confir- for asthma (inhalers, medicine, tablets mation of this was received from HDEC on 9 or pills).7 The asthma prevalence among August 2019. Māori children (16%), New Zealand’s indig- New Zealand administrative health enous population, was higher than that observed for children of non-Māori ethnic- data De-identified New Zealand administrative ities. That Māori children have higher health data from January 2010 to December asthma morbidity8 and mortality9 rates 2019 were accessed through the Ministry of was already documented over thirty years Health. All paediatric administrative data ago. Critically, in 2019, the Global Burden from across New Zealand were analysed for of Disease10 reported the rate of deaths due patterns of hospitalisation and medication to childhood asthma in New Zealand was prescription for asthma among children nearly four times higher than the global aged 0–14 years. All demographic infor- rate for children in the 10–14-year age mation was extracted using the national group, at 3.06% compared to 0.70%.10 For health index number (NHI), a unique iden- children aged 5–9 years, the proportion of tifier assigned to each individual in New disability-adjusted life years (DALY) due Zealand at birth or on their first contact with to asthma in New Zealand (8.95% of total the healthcare system. These demographic DALYs) is 3.6 times higher than the global data included an individual’s sex, date of rate for children aged 5–9 years.10 birth, self-reported prioritised ethnicity and There exist various efforts to assess the NZDep2013,15 an indicator of neighbourhood evidence on resource use and costs among socioeconomic deprivation derived from people with asthma in New Zealand.11–14 A variables collected at the New Zealand 2013 1989 estimate11 put the combined cost of national census. Neighbourhood socioeco- asthma hospitalisations and medications in nomic deprivation was categorised into New Zealand at $102.3m per annum. The quintiles. annual cost of asthma hospitalisations and De-identified child data were stratified by medications in New Zealand was estimated age and ethnic group. Age groups were ≤4 to be $349m per annum in 2001,14 $799m12 in years, 5–9 years and 10–14 years. Neonates 2015 and over $1 billion in 2018.13 Therefore, (age
article (NMDS)16 and prescription data from the Cost of illness Pharmaceutical Collection.17. The Pharma- We use a cost-of-illness approach to ceutical Collection records all subsidised estimate the cost of paediatric asthma in prescription medications dispensed from New Zealand. Consistent with published community pharmacies in New Zealand. In studies on the cost of asthma in New New Zealand, all prescription asthma medi- Zealand,11 we included only the major cations are subsidised for children. direct costs of prescription medication and Asthma hospitalisations were defined as hospital admission. In contrast to others,21 any admission to a public hospital with a we did not include indirect or intangible discharge diagnosis of asthma. Asthma was costs. defined using ICD-10-AM18 system clinical Hospitalisation costs, adjusted for inflation codes (Table 1). Admissions with a diagnosis and reported in 2019/20 financial year terms of wheezing (R06.2) were not included in ($FY2019/20), were derived by multiplying asthma admissions. the case weight of each admission (supplied Asthma-medication prescriptions were in the NMDS) by the annual purchase unit identified in the pharmaceutical claims price for 2019/20 (Appendix Table 2).22 data using the PHARMAC schedule19 and The Ministry of Health pharmaceutical the New Zealand Formulary for Children claims data includes information on the cost (NZFC)20 (Appendix Table 1). PHARMAC, New of each prescription. The cost value used in Zealand’s national drug-buying agency, clas- this analysis is the GST-exclusive subsidy, sifies all medications by therapeutic group. as listed in the Pharmaceutical Schedule, We used a three-stage identification process: that district health boards reimburse (1) prescriptions were selected first from pharmacies. These values were inflation ‘Respiratory system and allergies’ in Ther- adjusted, and all prescription costs are apeutic Group 1; (2) medications specific to reported in real values for the financial year paediatric asthma were identified in Thera- 2019/20 ($FY2019/20). We retrieved inflation peutic Group 2; (3) asthma medication was data from Stats NZ23 using the first quarter stratified into preventer and reliever groups of 2010 as the index base period. based on classification in the NZFC.20 Not all medications used in the treatment of asthma Statistical methods are listed under the respiratory systems For conventional statistical tests, therapeutic group. Therefore, additional continuous variables were compared medications (ie, prednisone and prednis- using Mann–Whitney tests, and cate- olone (from therapeutic group ‘Hormone gorical data were compared using Fisher’s preparations’) and omalizumab (from exact tests and chi-squared tests. Simple therapeutic group ‘Oncology agents’)) were linear regression and a nonparametric included in this analysis to capture the full loess regression were performed to assess spectrum of asthma medication used to treat whether changes in best practice and paediatric asthma. public health policy measures may have Table 1: ICD-10-AM codes for identifying asthma admissions, including primary and secondary diagno- ses, for children aged 0–14 years in New Zealand, 2010–2019. ICD10 code Definition Number of admissions 2010–2019 J45.0 Predominantly allergic asthma 307 J45.1 Non-allergic asthma 389 J45.8 Mixed asthma 3 J45.9 Asthma, unspecified 37,681 J46 Status asthmaticus 1,351 (including acute severe asthma) Total 39,731 82 NZMJ 16 April 2021, Vol 134 No 1533 ISSN 1175-8716 © NZMA www.nzma.org.nz/journal
article contributed to the observed reductions was greater than 1.5:1.0 in all age groups in childhood-asthma hospitalisation and in all years (Figure 2B). This is despite an prescription. Statistical significance was aggregate 47% reduction in asthma hospi- set at p≤0.05. All analysis for this study talisation rates, across all groups, between was generated using SAS Enterprise Guide 2010 and 2019. software v7.15.24 Poverty is a known risk factor for respi- ratory illness and asthma.25 Consistent with Results this, asthma admission rates for children Discharge codes identified 32% of all from families living in the most deprived childhood hospitalisations (2010–2019) quintile of neighbourhoods were, on as having ICD-10-AM codes related to the average, 2.8 times higher than for children respiratory system. Asthma accounted for living in the least deprived areas (Figure 4% of paediatric hospitalisations and 11% 3). The difference between admissions for of paediatric respiratory hospitalisations. children in families from the least and most There were 39,731 paediatric admissions to deprived areas deprivation scores remained hospital with an asthma discharge diagnosis constant between 2010 and 2019. (2010–2019). Seasonal fluctuations in number of asthma Asthma admissions as a proportion of admissions were larger for children living in all respiratory admissions declined from the most versus the least deprived quintile 16% (n=5,104, 2010) to 8% (n=2,811, 2019). of neighbourhoods and were evident from At the population level, there was a 62% 2010–2019 (Figure 4). The difference in reduction in asthma admissions for the number of asthma admissions between youngest children (those age
article Figure 2: Asthma hospitalisation rates in New Zealand children by ethnic group and age group. (A) Asthma hospitalisation rates for children of Māori and non-Māori ethnicities age 0–14 years, 2011–2019. (B) Ratio of Māori to non-Māori for asthma hospitalisation rates by age group, 2010–2019. Figure 3: Number of admissions for asthma, children aged 0–14 years, by neighbourhood deprivation, 2010–2019. 84 NZMJ 16 April 2021, Vol 134 No 1533 ISSN 1175-8716 © NZMA www.nzma.org.nz/journal
article of their first admission was larger for children acute management (reliever medications), of Māori (18%) compared with non-Māori accounted for 33% (n=2,431,048) of all respi- (14%) ethnicity (p
article Figure 5: (A) Length of hospital stay for children (0–14 years) admitted to hospital with asthma by ad- mission year (2010–2019). (B) Comparison of length of hospital stay for children of Māori and non-Māori ethnicities from 2010–2019. (C) Proportions of children readmitted with asthma within 90 days of first hospital discharge, by admission year. 86 NZMJ 16 April 2021, Vol 134 No 1533 ISSN 1175-8716 © NZMA www.nzma.org.nz/journal
article Figure 6: Corticosteroids as a proportion of child asthma prescriptions by ethnic group from 2010 to 2019. Figure 7:Prescriptions for asthma-preventer and -reliever medications: children of Māori versus non- Māori ethnicity, in most and least deprived neighbourhood quintile of deprivation areas, 2010–2019. (Panel A) Māori children in the least deprived quintile. (Panel B) Māori children in the most deprived quintile. (Panel C) Non-Māori children in the least deprived quintile. (Panel D) Non-Māori children in the most deprived quintile. Figure 8: Asthma-preventer and -reliever medication costs, prescriptions for children aged 0–14 years, 2010–2019. 87 NZMJ 16 April 2021, Vol 134 No 1533 ISSN 1175-8716 © NZMA www.nzma.org.nz/journal
article reliever to preventer prescriptions (1.97). asthma treatment, made up of hospital Asthma-medication treatment costs for admissions and community pharmacy children in the most deprived quintile of prescriptions, was $165m. Of this, $103m neighbourhoods were double those for was for hospital admissions and $62m for children living in the least deprived quintile community pharmacy prescriptions (Table of neighbourhoods (Figure 10), suggesting 2). Notably, there was a 32% reduction in that their asthma is more symptomatic and the annual hospital admission costs from they require more asthma medications and/ 2010 to 2019. or for longer. At a per capita level of spending on The total hospitalisation cost for children asthma prescriptions, there was little (aged 0 to 14 years) admitted for all respi- difference between prescription spending ratory conditions was $1.2bn between 2010 for children of Māori versus non-Māori and 2010. Asthma hospitalisation costs ethnicity. However, asthma hospitalisation ranged from $2,436/admission (2010) to costs for Māori children were typically $2,798 (2015), for an average annual spend double that of a non-Māori child (Figure 11). of $10.3m from 2010–2019 (Figure 9). This indicates that Māori children require In total, we estimate that from 2010 longer and/or more complex care for their to 2019 the direct costs of paediatric asthma. Figure 9: Average cost per asthma hospitalisation, children aged 0–14 years, 2010–2019. Figure 10: Total combined direct costs of asthma treatment (hospital admissions and community pharmacy prescriptions) for children age 0–14 years living in the most (NZDep Q5) and least (NZDep Q1) quintile of neighbourhoods, 2010–2019. 88 NZMJ 16 April 2021, Vol 134 No 1533 ISSN 1175-8716 © NZMA www.nzma.org.nz/journal
article Policy influence on asthma asthma declined by 18% over the same period. There are clear inequities in the incidence health outcomes of New Zealand children Various health-related policy measures with asthma. To the extent that prescription were implemented over the period 2010– patterns reflect primary care patterns 2019, some of which may potentially have and hospital admissions reflect shortfalls exerted an influence on paediatric asthma in primary care,6 our analyses suggest hospitalisations. For example, lowering that many New Zealand children are not the age for free general practitioner (GP) receiving sufficient quality healthcare to care and changes in the asthma diagnosis optimise management of their asthma. protocol for children. Best practice guide- Māori children and children living in the lines for children aged
article Figure 11: Per capita real costs of asthma treatment: hospitalisations and prescriptions for children of Māori and non-Māori ethnicity aged 0-14 years, 2010–2019. Figure 12: (Panel A) Paediatric population, children age 0–14 years. (Panel B) Monthly admissions to hospital with an asthma diagnosis, children age 0–14 years, and public health policy measures, 2010– 2019. The scatter plot is fitted with a LOESS and linear regression curves. 90 NZMJ 16 April 2021, Vol 134 No 1533 ISSN 1175-8716 © NZMA www.nzma.org.nz/journal
article It is difficult to make comparisons across underuse of inhaled corticosteroids are cost-of-illness studies, primarily because of problem areas of management.29 In addition, differences in the types of costs included in inhaled corticosteroids are sometimes inap- the analysis. Telfar Barnard et al12 estimated propriately prescribed for respiratory tract that childhood asthma cost $24.4m in 2011, infections.30,31 Therefore, it is difficult to of which $7.8m was for prescriptions and interpret medication prescribing data with $7.6m for hospitalisations (total $15.4m). absolute certainty. (4) Further, this study has In calculating these costs, Telfar Barnard not specifically addressed the role of combi- et al included (1) ‘wheeze’ in the asthma nation therapy. A more detailed analysis classification, (2) estimates for work-days that covers combination therapy may yield lost and (3) GP visits. A 2015 repeat of this further important insights. (5) The pharma- study13 estimated that childhood asthma cost ceutical cost data do not include many of $78.3m, of which prescriptions accounted the dispensing costs and other payments to for $2.1m and hospitalisations for $7.4m pharmacies and therefore underestimate (total $9.5m). Our estimates are broadly the cost of supplying medicine to patients. comparable to these studies for the 2011 (6) The actual price that PHARMAC pays for treatment year but, in comparison to 2015, pharmaceuticals is confidential. PHARMAC are $3.86m higher for prescriptions and negotiates rebates with manufacturers, and $1.2m higher for hospitalisations. We therefore the list prices in the pharmaceu- contend that this is because Telfar Barnard tical schedule will overestimate the cost of et al12,13 estimated and deducted the rebate supplying medicine. given to pharmaceutical companies. As this Despite these limitations, there is a clear information is not publicly available, we and obvious disparity in asthma rates and were unable to account for the rebate. prescriptions and hospitalisation for asth- This study, while being comprehensive, matic children in New Zealand. Speculating suffers from a number of limitations. Specif- on the causes of these disparities would ically: (1) The administrative health data are almost certainly point to socioeconomic not well-positioned to inform on sociode- differences, inequalities in accessibility mographic factors beyond neighbourhood and perhaps unconscious bias in treatment deprivation. (2) The pharmaceutical data approaches. However, speculation is insuffi- captures prescriptions that were dispensed cient. Although the causes will undoubtedly but cannot determine whether these medi- be various, they must be identified. Only cations were filled or taken. (3) Inhaled then will we be able to tailor our response corticosteroids are front-line medica- and address the equity problem that exists tions for asthma, and non-adherence to in the care for childhood asthma. treatment, overuse of bronchodilators and 91 NZMJ 16 April 2021, Vol 134 No 1533 ISSN 1175-8716 © NZMA www.nzma.org.nz/journal
article Appendix Appendix Table 1: Classification of asthma medication in the PHARMAC pharmaceutical schedule. Therapeutic Group 1 Reliever or Preventer Therapeutic Group 2 Chemical name Respiratory System Reliever Beta-adrenoceptor Salbutamol and Allergies agonists Terbutaline Methylxanthines Theophylline Anticholinergic Agents Salbutamol with ipra- tropium bromide Ipratropium bromide Hormone Preparations Corticosteroids and Prednisone related Prednisolone Respiratory System Preventer Inhaled corticosteroids Fluticasone and Allergies Beclomethasone dipropionate Budesonide Inhaled Long-acting Salmeterol Beta-adrenoceptor Agonists Eformoterol fumarate Inhaled Corticoste- Fluticasone with sal- roids with Long-Acting meterol Beta-Adrenoceptor Fluticasone furoate Agonists with vilanterol Anticholinergic Agents Tiotropium Leukotriene receptor Montelukast antagonist Mast cell stabiliser Sodium cromoglycate Nedocromil Combination Drugs Combination drugs Budesonide with efor- moterol Oncology Agents Immunosuppressants Omalizumab 92 NZMJ 16 April 2021, Vol 134 No 1533 ISSN 1175-8716 © NZMA www.nzma.org.nz/journal
article Appendix Table 2: Purchase Units for events included in casemix funding, adapted from Ministry of Health.20 Financial Year Medical & Surgical 1998/99 2,433.62 1999/00 2,399.22 2000/01 2,487.16 2001/02 2,479.01 2002/03 2,617.72 2003/04 2,728.55 2004/05 2,854.88 2005/06 2,949.09 2006/07 3,151.01 2007/08 3,740.38 2008/09 3,985.32 2009/10 4,315.48 2010/11 4,410.38 2011/12 4,567.49 2012/13 4,614.36 2013/14 4,655.43 2014/15 4,681.97 2015/16 4, 751.58 2016/17 4,824.67 2017/18 4,921.16 2018/19 5,068.12 2019/20 5,216.21 2020/21 5,545.26 93 NZMJ 16 April 2021, Vol 134 No 1533 ISSN 1175-8716 © NZMA www.nzma.org.nz/journal
article Competing interests: Nil. Acknowledgements: The authors would like to thank the Liggins Māori advisory group for their input and discussion into this work. We would also like to thank Frank Bloomfield for comments on the manuscript and the Ministry of Health for data access and help. DS and TF were funded by an HRC Explorer Grant (HRC 19/774). Author information: Deborah Schlichting, Research Fellow, Liggins Institute, The University of Auckland, Auckland. Tayaza Fadason, Research Fellow, Liggins Institute, The University of Auckland, Auckland. Cameron C. Grant, Professor, Head of Department of Department of Paediatrics: Child and Youth Health, The University of Auckland, Auckland. Justin M. O’Sullivan, Associate Professor, Deputy Director, Liggins Institute, The University of Auckland, Auckland. Corresponding author: Deborah Schlichting, Research Fellow, Liggins Institute, The University of Auckland, Auckland d.schlichting@auckland.ac.nz Justin M O’Sullivan, Associate Professor, Deputy Director, Liggins Institute, The University of Auckland, Auckland j.m.osullivan@auckland.ac.nz URL: www.nzma.org.nz/journal-articles/childhood-asthma-in-new-zealand-the-impact-of-on-go- ing-socioeconomic-disadvantage-2010-2019 REFERENCES 1. Asher I, Pearce N. Global Respiratory medicine 8. Mitchell EA. Racial burden of asthma 2016;116:34-40. doi: inequalities in childhood among children. The 10.1016/j.rmed.2016.05.008 asthma. Social Science & international journal of [published Online Medicine 1991;32(7):831- tuberculosis and lung disease First: 2016/06/15] 36. doi: https://doi. : the official journal of the 5. Health Quality and Safety org/10.1016/0277- International Union against Commission New Zealand. 9536(91)90309-Z Tuberculosis and Lung Atlas of Healthcare Varia- 9. Sears MR, Rea HH, Fenwick Disease 2014;18(11):1269- tion Methodology | Asthma J, et al. Deaths from 78. doi: 10.5588/ Wellingotn2019 [Available asthma in New Zealand. ijtld.14.0170 [published from: https://www.hqsc. Archives of disease in Online First: 2014/10/10] govt.nz/assets/Health-Quali- childhood 1986;61(1):6. 2. Serebrisky D, Wiznia ty-Evaluation/Atlas/Asthma/ doi: 10.1136/adc.61.1.6 A. Pediatric Asthma: Methodology-asthma-up- 10. Institute for Health Metrics A Global Epidemic. date-2020.pdf accessed and Evaluation. GBD Annals of Global Health 07 August 2020 2019. Compare. Seattle, WA: 2019;85(1):6. doi: http://doi. 6. OECD. Health at a Glance: IHME, University of Wash- org/10.5334/aogh.2416 2019. 2019 doi: https://doi. ington; 2015 [Available 3. Hsu J, Qin X, Beavers SF, org/10.1787/4dd50c09-en from: http://vizhub.health- et al. Asthma-Related 7. Ministry of Health. data.org/gbd-compare School Absenteeism, Annual Update of Key accessed November 8 2019. Morbidity, and Modifiable Results 2018/19: New 11. Mitchell E. Asthma costs. Factors. American Journal Zealand Health Survey. New Zealand Medical Jour- of Preventive Medicine Wellington2019 [Available nal 1989;102(865):171-2. 2016;51(1):23-32. doi: from: health.govt.nz/ 12. Telfar Barnard L, Baker M, https://doi.org/10.1016/j. publication/annual-update- Pierse N, et al. The impact amepre.2015.12.012 key-results-2018-19-new- of respiratory disease 4. Mirabelli MC, Hsu J, Gower zealand-health-survey in New Zealand: 2014 WA. Comorbidities of accessed April 2020. update. Wellington: The asthma in U.S. children. Asthma Foundation, 2015. 94 NZMJ 16 April 2021, Vol 134 No 1533 ISSN 1175-8716 © NZMA www.nzma.org.nz/journal
article 13. Telfar Barnard L, Zhang J. nz/wwwtrs/Sched- 28. Mills C, Reid P, Vaithiana- The impact of respiratory uleOnline.php accessed than R. The cost of child disease in New Zealand: September 26 2020. health inequalities in 2018 update. Wellington: 20. New Zealand Formulary Aotearoa New Zealand: Asthma + Respiratory for Children (NZFC). a preliminary scoping Foundation NZ, University NZFC v89 2019 [Avail- study. BMC Public Health of Otago, Wellington, 2018. able from: https://www. 2012;12(1):384. doi: 14. Wilson N. The cost burden nzfchildren.org.nz/ 10.1186/1471-2458-12-384 of asthma in New Zealand. accessed February 2020. 29. Bush A, Fleming L. New Zealand Medical 21. Holt S, Beasley R. The Diagnosis and management Journal 2001;114(1128):148. Burden of Asthma in of asthma in children. 15. Atkinson J, Salmond C, New Zealand. Wellington: 2015;350:h996. doi: 10.1136/ Crampton P. NZDep13 Asthma and Respiratory bmj.h996 %J BMJ : Brit- Index of Deprivation. Foundation of New ish Medical Journal Wellington: University Zealand, 2001. 30. Poulos LM, Ampon RD, of Otago, 2014. 22. The NCCP Casemix – Cost Marks GB, et al. Inappropri- 16. Ministry of Health. Weights Project Group. ate prescribing of inhaled National Minimum Dataset New Zealand Casemix corticosteroids: are they (hospital events) Welling- Framework For Public- being prescribed for respi- ton2019 [Available from: ly Funded Hospitals ratory tract infections? A https://www.health.govt.nz/ - WIESNZ19 2019/20. retrospective cohort study. nz-health-statistics/nation- Wellington: Ministry Primary care respiratory al-collections-and-surveys/ of Health, 2019. journal : journal of the collections/ General Practice Airways 23. Statistics New Zealand. national-minimum-data- Group 2013;22(2):201-8. doi: Infoshare. In: Zealand SN, set-hospital-events 10.4104/pcrj.2013.00036 ed. Wellington, 2020. accessed August 2019. [published Online 24. SAS Enterprise Guide First: 2013/04/26] 17. Ministry of Health. [program]. 7.15 Pharmaceutical Collec- 31. Teichert M, Schermer T, version. Carey, NC: SAS tion Wellington2019 van den Nieuwenhof L, et Institute Inc., 2017. [Available from: https:// al. Prevalence of inap- 25. Environmental Health propriate prescribing of www.health.govt.nz/ Indicators New Zealand. inhaled corticosteroids for nz-health-statistics/nation- Asthma hospitalisations respiratory tract infections al-collections-and-surveys/ (0–14 years) Wellington: in the Netherlands: a collections/pharma- Massey University; 2020 retrospective cohort study. ceutical-collection [Available from: www. NPJ Prim Care Respir Med accessed August 2019. ehinz.ac.nz accessed 2014;24:14086-86. doi: 18. Australian Consortium August 07 2020. 10.1038/npjpcrm.2014.86 for Classification Develop- 26. Maurer M, Rosén K, Hsieh ment. The international H-J, et al. Omalizumab for statistical classification of the Treatment of Chronic diseases and related health Idiopathic or Spontaneous problems, tenth revision, Urticaria. New England Australian modification Journal of Medicine (ICD-10-AM/ACHI/ACS) 2013;368(10):924-35. doi: (Tenth ed.). Darlinghurst, 10.1056/NEJMoa1215372 NSW: Independent Hospital Pricing Authority, 2017. 27. Diagnosing and manag- ing asthma in children. 19. PHARMAC. Online Best Practice Journal Pharmaceutical Sched- 2012;42 doi: https://bpac. ule Wellington2020 org.nz/BPJ/2012/Febru- [Available from: https:// ary/asthma.aspx www.pharmac.govt. 95 NZMJ 16 April 2021, Vol 134 No 1533 ISSN 1175-8716 © NZMA www.nzma.org.nz/journal
You can also read