Determinants of self-reported medicine underuse due to cost: a comparison of seven countries

The University of Western Australia, Perth, WA 6009, Australia.
Journal of Health Services Research & Policy (Impact Factor: 1.73). 03/2010; 15(2):106-14. DOI: 10.1258/jhsrp.2009.009059
Source: PubMed

ABSTRACT To compare the predictors of self-reported medicine underuse due to cost across countries with different pharmaceutical subsidy systems and co-payments.
We analysed data from a 2007 survey of adults in Australia, Canada, Germany, the Netherlands, New Zealand (NZ), the United Kingdom (UK) and the United States (US). The predictors of underuse were calculated separately for each country using multivariate poisson regression.
Reports of underuse due to cost varied from 3% in the Netherlands to 20% in the US. In Australia, Canada, NZ, the UK and the US, cost-related underuse was predicted by high out-of-pocket costs (RR range 2.0-4.6), below average income (RR range 1.9-3.1), and younger age (RR range 3.9-16.4). In all countries except Australia and the UK, history of depression was associated with cost-related underuse (RR range 1.2-4.1). In Australia, Canada, Germany, the UK and the US lack of patient involvement in treatment decisions was associated with cost-related underuse (RR range 1.2-1.4). In Australia, Canada and NZ, indigenous persons more commonly reported underuse due to cost (RR range 2.1-2.9).
Cost-related underuse of medicines was least commonly reported in countries with the lowest out-of-pocket costs, the Netherlands and the UK. Countries with reduced co-payments or cost ceilings for low income patients showed the least disparity in rates of underuse between income groups. Despite differences in health insurance systems in these countries, age, ethnicity, depression, and involvement with treatment decisions were consistently predictive of underuse. There are opportunities for policy makers and clinicians to support medicine use in vulnerable groups.

  • [Show abstract] [Hide abstract]
    ABSTRACT: Aim To evaluate the associations between socio-economical factors and the use of medications in the elderly. Methods The data was collected in a cross-sectional study in 2009. We received 624 completed questionnaires (response rate — 48.9%) from elderly people aged 60–84 years living in Kaunas (Lithuania). For an evaluation of the impact of explanatory variables on the analyzed events (binary dependent variable), an Enter model of logistic regression was used. Results Our findings suggest that 50.8% (n=317) of respondents used at least one drug daily. 18.3% (n=114) of respondents indicated that they use medications regularly, but not on a day-by-day basis. One quarter (25.6%; n=160) used medication only on an “at need” basis. Only 5.3% (n=33) of older persons did not use any medications at all. Logistic regression showed that being male (OR=0.67; 95%CI:0.45–0.98) was associated with using medications “regularly + daily.” For the use of “daily” medications, older age (OR=1.33; 95%CI:1.15–1.53) was associated with using medications daily. An opposite association was observed for respondents having no paid work (OR=0.48; 95%CI:0.26–0.82). Conclusions Our study suggests that more than half of older persons in Lithuania use medications every day. Use was associated with socioeconomic factors (gender, age, and employment status).
    Central European Journal of Medicine 12/2013; 8(6):855-860. DOI:10.2478/s11536-013-0237-7 · 0.21 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Background This study investigated the household economic burden of eating disorders and cost-related non-adherence to treatment in Australia.Methods Multi-centre prospective observational study using a structured questionnaire. Ninety participants were recruited from two clinic settings in New South Wales, Australia and from the community using social media. The primary outcome measures were household economic burden of illness measured in terms of out-of-pocket expenditure, household economic hardship and cost-related non-adherence.ResultsThe pattern of out-of-pocket expenditure varied by diagnosis, with Bulimia Nervosa associated with the highest total mean expenditure (per three months). Economic hardship was reported in 96.7% of participants and 17.8% reported cost-related non-adherence. Those most likely to report cost-related non-adherence had a longer time since diagnosis. Cost-related non-adherence and higher out-of-pocket expenditure were associated with poorer quality of life, a more threatening perception of the impact of the illness and poor self-reported health.Conclusions This study is the first to empirically and quantitatively examine the household economic burden of eating disorders from the patient perspective. Results indicate that households experience a substantial burden associated with the treatment and management of an eating disorder. This burden may contribute to maintaining the illness for those who experience cost-related non-adherence and by negatively influencing health outcomes. Current initiatives to implement sustainable and integrated models of care for eating disorders should strive to minimise the economic impact of treatment on families.
    BMC Psychiatry 11/2014; 14(1):338. DOI:10.1186/s12888-014-0338-0 · 2.24 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Polypill-based strategies have improved patient use of preventive cardiovascular disease (CVD) medications in clinical trials. Continued use in real-world settings relies on patients preferring a polypill over current treatment. Within a clinical trial assessing a CVD polypill-based strategy on patient adherence (Kanyini Guidelines Adherence with the Polypill study [Kanyini GAP]), we used discrete choice experiment (DCE) to assess the influence of polypill-based treatment attributes and patient characteristics on preferences for CVD preventive treatment. A DCE survey was administered to Kanyini GAP participants, involving choices between 2 hypothetical treatment options and no treatment for CVD prevention. Attributes delineating a polypill from current treatment were assessed: out-of-pocket costs, tablet number, administration, and prescriber visit frequency. The odds ratios (ORs) for preferring treatment, trade-off between treatment-related attributes, and willingness to pay against other attributes were estimated. In all, 332 of 487 (68%) participants completed the survey. Active treatment, compared with no treatment, was chosen by 93%. Treatment preference decreased with increasing out-of-pocket cost (OR = 0.04; 95% CI = 0.03-0.05) and tablet number (OR = 0.69; 95% CI = 0.59-0.81). Out-of-pocket cost was the most important attribute. Respondents were willing to pay $3.45 per month for each tablet reduction. Education and household income significantly influenced treatment preference. Assuming equivalent efficacy and safety of treatment options, the treatment-specific attributes that were assessed and influenced patient preference strongly accord with the posited advantages of the cardiovascular polypill. The study provides promising evidence that improvements in treatment adherence observed in CVD polypill trials may translate to the real world and potentially close treatment gaps in CVD prevention. © The Author(s) 2015.


Available from
Dec 12, 2014