The effect of health insurance on health care utilisation among people with non-communicable diseases in Zambia

Abstract

There is a growing burden of burden of non-communicable diseases (NCDs) in low and middle income countries (LMICs), with inequalities in NCDs and access to NCDs care persisting. Evidence suggests that these inequalities can be reduced by an effective health insurance system. Health insurance may also improve quality of care through improved finance but also by reducing out of pocket payments. Thus health insurance may be critical to achieving universal health coverage (UHC). While studies have shown the benefits of health insurance, most of these do not control for health status. In addition, we do not know whether individuals with NCDs who have health insurance have higher healthcare utilisation than their counterparts without insurance. If this inequality exists, expanding coverage may reduce inequalities in access to NCD care and policy makers may design the benefits package to address this concern. This study seeks to contribute in filling this gap by examining if insurance increases the use of health services. Specifically, the study examines people who have NCDs and health insurance to see if they utilise health services more than their colleagues without insurance. The study design was cross sectional and used secondary data from the Zambia Household Health Expenditure Utilisation Survey conducted in 2013. The sample size was 13, 150, which was the number of people who reported visiting a health facility one month prior to the survey. The outcome variable was an indicator variable equal to one if an individual utilised health services and zero if not. There were three variables of interest. The first was an indicator variable equal to one if somebody had an NCD and another indicator variable equal to one if somebody had health insurance. The third was an interaction of the two indicator variables. A linear probability model was estimated in order to examine the association between health care utilisation on one hand, and insurance status, having an NCD, and the interaction between insurance and NCD status on the other hand. After controlling for other covariates and need factors such as chronic disease and self-assessed health, people without health insurance used health services 6.4 percentage points (pp) less than those without insurance, and the effect was significant at less than 5% level (p=0.048, 95% CI -0.13,0.0005). People with NCDs were 5.4pp more likely to use health services (p=<0.0001 95% CI -0.28, 0.087) than those without NCDs. Health insurance increased the likelihood of using health services among people with NCDs by 6.4pp (p=0.285, 95% CI -0.18, 0.05). Although large, this effect was not significant. This study suggests that expanding health insurance coverage in Zambia may substantially increase the use of health services, an important step in improving population health. Importantly, there is suggestive evidence that improving insurance coverage among people with NCDs may increase the use of care. This is important given the growing burden of NCDs. This study recommends that social health insurance be implemented and scaled up to the wider population and a benefit package for chronic illness be created. This has the potential to contribute towards achieving the goal of UHC in Zambia. However, the influence of other barriers to health care use such as perception of health will need to be evaluated. Key Words: NCDs, insurance, health care, utilisation, Zambia.