User choice in Swedish primary health care - geographical establishment of providers and distribution of care resources
In January 2010 it was made compulsory for local care providers in Sweden to introduce user choice in the primary care sector. The reform means that health care providers who meet certain basic requirements are free to establish themselves in the local community. In practice, this implies that the provision of care services, which has hitherto been predominately public, will gradually be privatized. Patients will be offered a free choice between public and private providers, who will then compete for these. In effect, a quasi-market for primary health services has thus been set up. Critics have argued that private care providers will establish themselves in prosperous areas and thereby attract high-income and better-educated patients. These are also likely to have lesser care needs and therefore be "profitable". This selection on part of providers has been referred to in the literature as cream skimming.
If this happens, the Swedish central political goal of needs-based allocation of health care resources is undermined. In this paper, we empirically investigate whether this selection on part of private care providers actually occurs; in other words, whether they establish foremost themselves in areas where the better-off live. This question also has theoretical significance. It often argued in the literature on choice models and health care, that cream-skimming is a big problem which will undermine universalist principles in tax-based health systems like the Swedish. Methodologically, the paper will use data from all Swedish local health care districts, the county councils. Data on the geographical distribution of newly established private health clinics will be linked to data on the socio-economic status of inhabitants in different residential areas. This unique combination of national data bases, which covers the entire Swedish population and all primary health care providers in the county, makes it possible to address a question which is often asked in the literature; e g whether choice and privatization in health care leads to a more unequal distribution of health resources. If so this turns out to be the case, well-documented differences in health status between different income groups will be further increased. The introduction of a user choice in Swedish health care also implies that the previous model of political governance of the system –based on the ideals of local democracy- is replaced by a more marked-based ideal, where individual choices, rather than democratic values, determine the allocation of resources.
