Healthcare utilisation in Indonesia : determinants and projections

Abstract

This study explored the roles played by the demographic and non-demographic characteristics of users and provider's characteristics in determining outpatient healthcare utilisation in Indonesia, and simulated the effects of these demographic, health insurance subscription and chronic disease factors on outpatient healthcare utilisation in the future. The study is expected to contribute to public policy by providing empirical evidence of the determinants of healthcare utilisation in Indonesia, filling a gap in research of roles of provider characteristics, and providing insights into how to incorporate non-demographic factors into utilisation projections. The study employed various statistical and mathematical tools, including discrete choice models for determinant analysis, cohort component methods for population projection and propensity methods for the projection of healthcare utilisation. Three main sets of data are used: the 2007 National Social Economic Survey (Susenas), the 2007 Indonesia Family Life Survey (IFLS) and the 2007 Basic Health Survey (Riskesdas). For the population projections, the base population is drawn from the post-enumeration-adjusted 2010 Population Census. The study demonstrated that healthcare utilisation and choice of providers in Indonesia are determined by both demographic and non-demographic factors and, in general, this is in line with Andersen's behaviour framework of health service use. The study findings also support the notion that the effect of age on utilisation is not monotonic. The difference in association and the degree of influence of independent variables on the decision to visit and on choice of provider indicates that the decision to use healthcare and the decision to choose a provider is not simultaneous, but rather a two-stage process. High price of service and low numbers of doctors deters the use of health services. Distance and drug availability are also associated with the choice of providers. Response to provider characteristics is not uniform, but varies across the population. This study also indicates the presence of a 'bypassing phenomenon', which occurs when patients bypass nearer healthcare providers to seek a higher-quality provider. The size of the Indonesian population is projected to increase by 19.6% during 2010-2025. Due to this population growth, total healthcare utilisation is expected to increase by 25.7% over the same period, with the pattern of utilisation to follow the pattern of population increase. Compared to the effects of health insurance and chronic disease, demographic change will continue as the major driver of increased healthcare utilisation in the future. Further, for some age groups (for example, children and adult), the effect of health insurance and chronic disease will also be very significant. In the policy context, the study indicates that income-related inequity in access to healthcare services will not be a significant issue; however, inequity in access to high-quality providers will. Access to healthcare among children and the elderly from lower social economic backgrounds is substantially lower than for other cohorts, and it was also found that urban dwellers are more sensitive to price of service and less sensitive to distance to healthcare providers, while rural dwellers are the opposite.