Objectives This paper aims to examine ethnic differences in financial barriers to access to prescription medication in New Zealand.
Methods Data from SoFIE-Health (wave 3), which is an add-on to the Statistics New Zealand-led longitudinal Survey of Family, Income and Employment (SoFIE) (N=18 320), were analysed using logistic regression, adjusting for demographic, socioeconomic, health behaviour and health variables. Financial barriers to access to prescription items were measured by the following question: ‘In the past 12 months, have there been any times when a doctor gave you a prescription, but you didn't collect one or more of these items because you could not afford the cost?’.
Results The odds of deferring buying a prescription at least once during the preceding 12 months because they could not afford the cost of the prescription were greater for Māori and Pacific people than for NZ Europeans (OR 2.98, 95% CI 2.56 to 3.47 vs OR 3.52, 95% CI 2.85 to 4.35). Adjusting for potential confounders attenuated the ORs to 1.31 (95% CI 1.08 to 1.58) for Māori people and to 2.17 (95% CI 1.68 to 2.81) for Pacific people. Deferring buying medications because of cost was also associated with increased odds of poor self-reported health status, high/very high psychological stress and the presence of two or more comorbid conditions.
Conclusion Ethnicity plays a critical role in facilitating or impeding access to primary health care. This suggests that policy measures to further reduce financial barriers to buying medication may improve access to care for everyone including Māori and Pacific people and may have positive health implications.
- Primary healthcare
- prescription cost
- access barriers
- New Zealand
- Access to HLTH SERV
Statistics from Altmetric.com
A growing body of researchers has consistently found differences in health and access to health services by ethnicity.1–4 However, less attention has been paid to ethnic differences in barriers to access to primary healthcare due to cost. Even less is known about the association between ethnicity and cost barriers to collecting prescription medication. However, such barriers lead to decreased prescription medication use5–9 or stopping of medication.10 This is of concern for many reasons. First, there is strong evidence that cost-related decreases in medication are not only for treating ‘less essential’ health problems such as allergies, but also for treating disabling and potentially life-threatening illnesses such as hypertension, hyperlipidaemia, depression, osteoporosis, prevention of stroke, asthma and diabetes.11 Second, if cost barriers result in partial or discontinued drug therapy, they are also likely to result in unfavourable health outcomes.7 12–14 Third, cost barriers to drugs are associated with increased rates of non-elective hospitalisations, visits to the emergency departments and death which, in turn, increase healthcare costs.14–17 This has substantial economic consequences for society,18 especially as healthcare cost containment becomes an increasingly important policy issue.
Research, mainly from the USA, has focused on general determinants of drug affordability and prescription use and has identified that elderly subjects, women, non-white and low-income populations are most likely to report restricting medications because of cost.7 9 19–24 Given the disparity in health and health needs among ethnic groups and the critical importance of prescription medications for chronic conditions and preventing further deterioration in health status, it is important to understand ethnic differences in cost barriers to prescription medication. This paper sets out to examine ethnic differences in financial barriers to prescription medication in New Zealand, specifically the extent to which there is ethnic variation in financial barriers to prescription medication. If there are differences, do they remain even after controlling for socioeconomic, health behaviour and health factors? Examining ethnic differences in access to prescription medication is particularly important in the New Zealand context because ensuring equal access to healthcare irrespective of socioeconomic position is one of the seven fundamental principles guiding the 2001 New Zealand Health Strategy.25
Access to prescription drugs in New Zealand
PHARMAC is the government agency which decides which prescription drugs will be subsidised, negotiates prices for medicines, sets subsidy levels and conditions and ensures spending stays within the budget. Before drugs are subsidised, drug companies need to demonstrate not only that their products are effective but also that they are more cost-effective than similar products. For many medicines, a process of reference pricing is used.26 After negotiating with drug companies for the lowest possible price, PHARMAC sets a dollar amount for the subsidy of all medicines in ‘therapeutic subgroups’. Usually this amount is set at the manufacturer's price for the least expensive drug in a given category (eg, ACE inhibitors). If the cost of another drug within the group is higher than the subsidy, the patient is required to pay the difference. This is referred to as a ‘part charge’. At the time of the study the final price paid by patients for a drug also depended on whether they held a community service card, a high user health card or a prescription subsidy card (for patients who receive a large number of prescriptions per year) which entitled them to a higher subsidy and a lower co-payment on partly or fully subsidised medicines.
At the time of this study (2004), the maximum prescription fee for fully subsidised medicines was $15 for those without a community service card, prescription subsidy card or a high user health card and $3 for those with any of these cards. In addition, ‘part charges’ were made for partially subsidised drugs. Cost barriers to prescription medications have further reduced since 2004.27
This research used SoFIE-Health data, which is an add-on to the Statistics New Zealand-led Survey of Family, Income and Employment (SoFIE). SoFIE is a single fixed panel and is the largest longitudinal survey ever run in New Zealand. It is a nationally representative study of 22 000 adults drawn by random sampling of households, interviewed face-to-face. All adults in the original sample will be followed for a maximum duration of 8 years starting from October 2002, even if their household or family circumstances change. Information is collected yearly on income levels, sources and changes, and on the major influences on income such as employment and education experiences, household and family status and changes, demographic factors and self-rated health status. Additional information on assets and liabilities is collected every 2 years.
The SoFIE-Health add-on comprises 20 min of questionnaire time in waves 3 (2004–5), 5 (2006–7) and 7 (2008–9) in the following health-related domains: Short-Form health survey (SF-36), Kessler-10 (K-10), perceived stress, chronic conditions (heart disease, diabetes, and injury-related disability), tobacco smoking, alcohol consumption, healthcare utilisation, access and continuity of primary healthcare and an individual deprivation score. The health module was administered to the original sample members.
Financial barriers to collection of prescription items within the past year were measured by the following questions: ‘In the past 12 months, have there been any times when a doctor gave you a prescription, but you didn't collect one or more of these items because you could not afford the cost?’ ‘If yes, how many times have you done this in the last 12 months?’. We dichotomised the responses into one or more versus no deferred collection of prescription items. Although the question is limited in the sense that it does not tell about the medical necessity of the medication, it is found to be a powerful measure for broad comparisons of drug access between population groups.28
The main exposure variable for this analysis is ethnicity. Each respondent was assigned to a mutually exclusive ethnic group by means of a prioritisation system commonly used in New Zealand: Māori, if any of the responses to self-identified ethnicity was Māori; Pacific, if any one response was Pacific but not Māori; Asian, if any one response was Asian but not Māori/Pacific; and the remainder were non-Māori non-Pacific non-Asian (nMnPnA). The nMnPnA category mostly comprises New Zealanders of European descent but, strictly speaking, is not an ethnic group. The nMnPnA group is the reference group.
Other covariates included sociodemographic variables, health risk behaviour and health status. Sociodemographic variables in this analysis were affiliation with a primary care provider, age, gender, marital status, ethnicity, family structure, being born in or outside of New Zealand, household equalised income, working status, highest level of education achieved, NZDep (area deprivation) and NZiDep (individual deprivation). Health behaviour and health included current smoking status, self-assessed health, K-10 and number of chronic conditions. Categories for the various measures are shown in table 1. Detailed descriptions of the creation of various variables have been published elsewhere.29
This paper provides cross-sectional analyses of wave 3. The population used in the analyses consisted of 18 320 adult (age 15 years and above) original sample members at wave 3. We first estimated the bivariate association between deferred prescription and other independent variables. Using univariate and multivariate logistic regression models, the independent effect of ethnicity was evaluated for deferring prescription medication because of cost, while controlling for other covariates. We also conducted separate logistic regression analyses for all four ethnic groups in order to examine whether the determinants of financial barriers to access to prescription drugs were different according to the ethnicity of the respondent (results not shown but discussed). The population used in the regression analyses consisted of 17 035 adults (age 15 years and above) original sample members at wave 3 who had complete information on all the socioeconomic, health behaviour and health characteristics. All analyses were performed using SAS Version 8.2.
Table 1 shows the characteristics of the sample population and bivariate associations between the demographic, socioeconomic, health behaviour and health predictors and the outcome measure. Of the total of 18 320 respondents, 6.4% reported that they had deferred collecting a prescription at least once during the preceding 12 months because they could not afford the cost of a visit or prescription (table 1). Younger adults aged 15–24 and 25–44 years, female gender, those who never married, Māori and Pacific, those in the lowest income tertiles, people living in the most deprived areas, those with more individual deprivation characteristics (5+), current smokers, those reporting high and very high levels of psychological distress and more than two co-morbid diseases were all more likely to put off buying prescription drugs because of cost barriers than their counterparts.
Table 2 indicates that ethnicity was significantly associated with deferring buying a prescription, with the odds of postponing buying a prescription drug 2.9 and 3.5 times higher for Māori and Pacific subjects, respectively, than for NZ European people in model 1. In contrast, Asians had an odds of 0.68 for deferring buying prescription drugs compared with NZ Europeans. The addition of demographic factors to the model reduced the ethnic effect OR to 2.15 for Māori subjects, to 2.9 for Pacific subjects (table 2 model 2) and to 0.65 for Asians; the association between ethnicity and deferring buying a prescription remained statistically significant. After controlling for the demographic and socioeconomic factors in model 3 of table 2, the ethnic OR declined further to 1.35 for Māori subjects, 1.91 for Pacific subjects and 0.59 for Asians; however, the effect of ethnicity still remained statistically significant. The final model (model 4, table 2) indicates that adding health behaviour and health variables to model 3 either did not change the ethnicity OR (in the case of Māori subjects) or increased the ethnicity effect OR slightly from 1.91 to 2.17 in the case of Pacific subjects, and the association between ethnicity and the outcome variable remained statistically significant. However, in the final model there was no statistically significant difference in the odds of deferring prescription drugs between Asians and NZ Europeans.
The final model (model 4, table 2) also indicates that, after adjusting for demographic, socioeconomic, health behaviour and health characteristics of the respondents in multivariate analyses, having an affiliation with a primary care provider, younger age, female gender, low income tertile, having more individual deprivation characteristics (5+), being a current smoker, reporting poor self-assessed health, high and very high levels of psychological distress and more than two comorbid diseases were all significantly associated with increased odds of deferring buying a prescription. However, sole parent family structure was associated with lower odds of deferring collecting medications. One of the reasons that sole parent families are more likely to fill prescriptions is that family type may have affected co-payment levels—that is, sole parent families might have more frequently met the income criteria for lower co-payments.
Logistic regression analyses were carried out separately for all four ethnic groups in order to identify ethnic differences in the predictors of cost-related barriers to medications (results not shown but available from the corresponding author on request). For all ethnic groups, younger age, having more individual deprivation characteristics (5+), current smokers and reporting more than two comorbid diseases were all significantly associated with increased odds of deferring prescription medications because of cost. While poor self-assessed health was significantly associated with increased odds of deferring prescription medications for Māori subjects but not for other ethnic groups, high and very high levels of psychological distress were significantly associated with increased odds of deferring prescription medications for all other ethnic groups but for Māori people. Being of female gender and affiliation with a primary care provider were significantly associated with increased odds of deferring buying prescription medications for both NZ European and Pacific people but not for Asians and Māori people.
We also computed interactions between ethnicity and important predictors (income, family type, NZiDep and comorbidity) to determine the significant ethnic differences in the coefficients. Interaction terms between ethnicity, family structure, NZiDep and presence of comorbidity were statistically significant (table 3). The relationship for family structure was significant for Asian and Pacific ethnic groups (the reference category is couple with children). The results in table 3 show that sole parents of Asian ethnicity were four times more likely to defer buying prescription medications than Asian couples with children, while their Pacific counterparts were 50% less likely to defer buying prescription medications. Furthermore, while having one or more individual deprivation characteristics was linked to deferred buying of prescriptions for all ethnic groups (the reference category is no individual deprivation characteristics), the size of the effect was greater for the European and Pacific groups followed by the Māori and Asian ethnic groups. For example, Pacific and European people with one or more individual deprivation characteristics were six times more likely to defer buying prescriptions, while Māori and Asian people with one of more individual deprivation characteristics were, respectively, four and two times more likely to defer buying a prescription.
With regard to the interaction between ethnicity and comorbid conditions, while the presence of one or more comorbid conditions is an equally important determinant of deferring prescriptions for all ethnic groups (ie, presence of comorbid conditions contributes to deferred prescriptions), the presence of one or more comorbid condition is more significant for Asians. For example, Asians with one of more comorbid conditions were eight times more likely to defer buying prescriptions, while Pacific and Māori people were twice as likely to defer buying prescriptions. Europeans with no comorbid conditions were 1.6 times more likely to defer buying prescriptions than Europeans with no comorbid conditions. It is possible that there is a cultural influence on medication-taking in the presence of other illnesses; however, exploring the cultural influence on medication-taking in the presence of other illnesses and non-filling of prescriptions in the different ethnic groups and the rationality of failure to purchase prescribed drugs is beyond the scope of this paper.
This study explored the impact of ethnicity on cost-related barriers to medications. We also examined systematically the extent to which demographic, socioeconomic, health behaviour and health-related factors might explain the ethnic differences in cost-related barriers to medications. Findings from this study reveal that ethnic disparities in cost-related barriers to prescription medication in New Zealand exist. Consistent with previously published literature, ethnicity independently predicted deferred prescription medication because of cost in the multivariate regression analysis,21 30 31 with Māori and Pacific people having higher odds of deferring medication purchase than NZ Europeans. Given the fact that Māori and Pacific people have a high unmet need because of cost, even after controlling for sociodemographic and other known confounders, primary healthcare policies targeting Māori and Pacific people are warranted. High rates of deferral of buying prescription medication by Māori and Pacific people are of concern, not only because they have lower access to resources to pay out-of-pocket costs for medication and other healthcare services but also because they have high health needs32 33 and may also have greater structural barriers to access to healthcare services than NZ Europeans.34
This work has also identified both distinct and common predictors of deferring buying prescriptions for all the major ethnic groups. While some factors associated with deferred buying of prescriptions were significant for all ethnic groups, other factors were significant only for one. These findings underscore the complexity of examining the nexus of the ethnicity and the deferred buying of prescriptions and the determinants of that nexus. Some factors such as individual deprivation characteristics determine deferred prescriptions regardless of ethnicity, while other factors such as family structure determine deferred prescriptions depending on the ethnicity of the respondent. Moreover, the effect size of some predictors (eg, presence of one or more comorbid condition) vary by ethnicity. Our finding that those who reported poor self-assessed health, high or very high psychological stress and two or more comorbid conditions were more likely to defer buying medications due to cost were in line with previous research which reported that those with a greater number of health problems or poorer health were most vulnerable to medication non-use due to cost.9 10 35 36
Individual deprivation, which is a measure of poverty, and belonging to the low and middle income tertile were highly associated with postponing prescription medications. While cost is certainly the major barrier to buying prescription medication for those with high levels of individual deprivation and low income, other barriers such as transportation problems or cultural and language barriers also weigh more heavily on the poor. Since SoFIE-Health asked only about financial barriers to postponing prescription medication, our analyses probably underestimate the number of people unable to obtain needed prescription drugs. The middle income group has the lowest rate of prescription filling which could be because they are not eligible for community service cards and therefore face higher costs. This study also found that, after adjusting for other factors, respondents with an affiliation with a primary care provider were consistently more likely to defer buying prescription medications. Although it is beyond the scope of this study, one explanation for this result may be the perception that a GP consultation will often result in a medicine being prescribed and a cost incurred (in addition to the cost of the consultation itself), thus deterring prospective patients affiliated with a primary care provider from seeking timely medication.
Several caveats to this study are worth mentioning. First, the study reports cross-sectional analyses which prohibit drawing causal conclusions. Follow-up data (wave 5) will allow conclusions regarding the direction of effects, enabling causal inferences to be drawn more confidently. Second, as with other self-reported surveys, this study relied on the respondents' ability to recall information accurately. If the reporting of deferred primary care among ethnic groups differed in some systematic way from each other, this may bias the results. The magnitude and direction of such bias is unknown. Third, although we have adjusted for many confounding variables, it is possible that the ethnic differences we found in deferred prescription medication could be the result of other factors associated with unmet primary care that we did not measure. Fourth, we did not ask about the perceived need or type of medication that was deferred because of cost. Fifth, information on the number of drugs prescribed is lacking. If cost were a barrier, it should have resulted in greater barriers with an increasing number of medications. Sixth, information on experience with medications is lacking. Given the known association of medications with adverse effects, even more so with an increasing number of drugs, it would be important to understand whether people's prior experiences with medications is a confounding (or at least an interacting) factor in failure to fill prescriptions. Seventh, belief in the efficacy of drugs may vary across cultural groups and should be ascertained in future studies. Another important question for future studies is whether people would have filled prescriptions if they did not have to pay for them. And finally, since 2004, co-payment levels have decreased for most of the population. However, the data presented in this paper do not allow us to investigate the impact of these changes.
Despite these limitations, the results presented here are important in several ways. This study uses a large original national survey on financial barriers to prescription medications. Few previous studies have considered cost as a factor in delaying supplementary healthcare services such as prescription drugs. Even fewer have focused on ethnic differentials in access to prescription drugs. The study findings increase understanding of the importance of ethnicity in the context of addressing inequalities in access to primary healthcare including prescription drugs. Given the importance of prescription medication in maintaining health and treatment of both acute and chronic illness, it is important to reduce co-payments for buying medication in order to improve access to care for everyone including Māori and Pacific people. While the policy issues regarding prescription drug coverage are complex, the public health message is simple and straightforward: it is important to remove barriers to drug access in order to improve population health.
Previous studies have identified a range of strategies adopted by patients to lower prescription costs. These include decrease in the use of medications, skipping or splitting doses, discontinuing prescribed medications, calling pharmacies to find the best price, switching to lower-cost generic medications, using free samples, not starting new medications, spending less on basic needs such as food or electricity and/or obtaining lower cost versions from other countries.7 9 20 23 24 Some of the strategies may put patients at risk of disease progression, leading to increased risk of hospitalisation and mortality. It remains to be seen what strategies patients in New Zealand adopt to save prescription costs. This is an important issue for further research.
The data available from SoPHIE cannot tell us which aspects of prescription charges are causing the differences found between ethnic groups. The basic prescription charge (usually $3 per item) may be the main barrier for individuals and families with high health needs and low incomes. Families are eligible for a Prescription Subsidy Card after they receive 20 items between 1 February and 31 January in any year, but the initial charge of up to $60 for the first 20 items may be prohibitive for those on low incomes. If this is underlying the differences observed in the study, then charges and/or ceilings need to be changed for the groups affected.
Another possibility is that Māori and Pacific people are exposed to partially subsidised or unsubsidised medicines more than other people. This could be because (1) they receive more prescriptions and therefore the chances of receiving partially subsidised or unsubsidised medicines is higher; (2) the medicines they are likely to take are more likely to be only partly subsidised or unsubsidised; or (3) prescribers and pharmacists do not make the decisions required to select a fully subsidised medicine for Māori and Pacific patients at the same rate as they do for other patients. For most drugs that are partially subsidised there is a fully subsidised alternative. Usually prescribers take this into account in their prescribing and, if a partially subsidised medicine is prescribed and the patient is unhappy about this, pharmacists often ring the prescriber to ask if the prescription can be changed.
Each of these possible explanations suggests different policy implications, so further research is needed to explore which, if any, explain the observed differences. If option (2) is correct, this would suggest that PHARMAC's decisions are having a negative effect on health disparities. Option (3) suggests that either prescribers and/or pharmacists need to be targeted to address this problem.
It is also possible that direct-to-consumer advertising of prescription medicines, which is allowed in New Zealand, has different effects in different population groups. Many prescription drugs that are advertised direct to consumers are not subsidised—that is, patients have to pay the full cost themselves. There are no data currently available about use of these medicines by ethnicity, but if Māori and Pacific people were more likely to be influenced by advertisements, they would face larger prescription costs for unsubsidised drugs and therefore could be more likely to report not picking up a prescription because of cost. As mentioned above, another possible explanation for the existence of an ethnic effect (Māori and Pacific people compared with others) after controlling for all other factors including income could be that Māori and Pacific people receive more prescriptions (because they experience more ill health) and therefore have to pay more in co-payments, increasing the chance that at some stage during the year they will find co-payments difficult to pay.
If cost as a barrier is not overcome, many people—particularly Māori and Pacific—will remain at risk of receiving less timely prescription medication which, in turn, may have negative repercussions for their health. Without addressing the cost barriers to collecting prescription medication, achieving equitable access to services may not be realised. Effective intervention programmes to reduce disparities in health outcomes among Māori and Pacific people should also focus on identifying other cultural and structural barriers to receipt of prescription medication. Cultural competence training for pharmacists, as is already occurring, may also assist in reducing barriers for Mâori and Pacific patients.
What is already known on this subject
Disparities in health status and use of healthcare services by ethnicity have been shown in a number of populations. Whether these disparities persist in collecting prescription medication because of cost is less well known. Such cost-related barriers, however, could lead people to restrict their use of medication, thus resulting in adverse health outcomes.
What this paper adds
This study found that ethnic disparities in buying prescription medication in New Zealand remain, while controlling for confounding demographic and socioeconomic variables. Deferring collection of prescription medications because of cost is disproportionately high among Māori and Pacific people. This study also found significant interaction between ethnicity, family structure, NZiDep and the presence of co-morbidity.
The authors are grateful for the contribution of Ken Richardson and Kristie Carter in preparing the data set and thank the reviewers for their insightful comments on this paper.
Access to the data used in this study was provided by Statistics New Zealand in a secure environment designed to give effect to the confidentiality provisions of the Statistics Act, 1975. The results in this study and any errors contained therein are those of the authors, not Statistics New Zealand.
Funding SoFIE-Health is primarily funded by the Health Research Council of New Zealand as part of the University of Otago's Health Inequalities Research Programme. Establishment funding was also received from the University of Otago, Accident Compensation Corporation of New Zealand (ACC) and the Alcohol Liquor Advisory Council (ALAC).
Competing interests None.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.