Measuring Efficiency in the Swedish Health Care Sector – Levels, Trade-offs and Determinants

This thesis measures cost efficiency in the production of health care and social services in Sweden. The object of analysis is not provider efficiency, but instead the relative efficiency of political organizations or different contractual arrangements. The thesis focuses on: 1) the measurement of cost efficiency, 2) the investigation of potential trade-offs between quantity of health care services and the quality of care, and 3) the exploration of determinants of cost efficiency. The first essay investigates the relative cost efficiency of the 21 Swedish county councils - how the councils fulfil their responsibility of providing and financing health care for their residents on the basis of need. Two models are estimated: one focusing on a traditional productivity measure and the other on quality in terms of health-related outcomes. Efficiency is estimated using non-parametric data envelopment analysis (DEA). The population weighted DEA-scores are 0.951 and 0.953 in the two models. The results also indicate that quantity and quality of care are complements in the production process. Councils which are net receivers in the equalization grant system have lower efficiency scores in both models. The second essay investigates the cost efficiency of Swedish municipalities in the delivery of care for the elderly. The first objective is to obtain a measure of cost efficiency, the second is to quantify the sources of cost efficiency and the third is to explore the relationship between cost efficiency and indicators of quality of care. It is based on a cross-sectional dataset consisting of 272 municipalities. Cost efficiency is estimated with a stochastic frontier cost function, using the modelling approach of Battese and Coelli (1995). The estimated average cost inefficiency is as high as 46 percent. Our empirical results also show that the share of private provision, rural location and small scale of operations are associated with higher efficiency. Furthermore, more educated staff is associated with greater efficiency. Finally, favourable results concerning outcome quality indicators, such as a lower prevalence of fall injuries, are associated with higher efficiency. This seems to indicate that there is no trade-off between cost efficiency and quality (outcome indicators) in Swedish care of the elderly. The third essay provides an analysis of the evolution of cost efficiency in public contracting of primary care during the first year after the primary care reform in Stockholm county council. A key element of the 2008 reform is the introduction of active individual listing and free choice of suppliers, where reimbursement follows the patient. The analysis is based on data from all the 147 primary care centres that delivered services during the period 2007-2008 and performed by stochastic frontier analysis (SFA). We analyze whether cost efficiency is influenced by different market characteristics and contractual arrangements. The relationship between cost efficiency and quality of care is investigated by including measures of patient satisfaction. The results show that cost efficiency in quantitative terms improved substantially. In addition, it seems that the increase in cost efficiency was not at the expense of patients’ experiences, nor did patients utilize other forms of care to a greater extent. A positive association between quality proxies and cost efficiency suggests that patient-driven competition may have worked well in this respect