Patient choice, competition, and quality of primary health care

During the last two decades, patient choice reforms have been implemented in many countries. Such reforms commonly lower the costs of switching between providers, increase the availability of comparative quality information about providers, and encourage entry of new providers; all of which should in theory increase competition. Increasing competition could in turn improve the quality of health care by strengthening the incentives to provide high quality services. However, many dimensions of health care quality are difficult for patients to observe and high quality providers need therefore not be the ones most valued by patients. Competition may also induce cream-skimming among patients, and increase inequalities in for example access to health care services. Under such circumstances, there is no guarantee that increased competition leads to improved quality. Empirical evidence of the effects of patient choice reforms and competition on the quality of primary health care is however scant.

We examine a set of patient choice reforms implemented in Swedish primary health care during 2007-2010. Our empirical strategy has two steps: We first analyse how the reforms affected primary care markets in terms of a) new entry, and b) lowered switching costs and increased information about providers. The reforms substantially increased entry in markets that initially had a low per capita level of providers and low barriers to entry, but had little or no impact on entry in other areas. The market structures also provide variation in how lowered switching costs and increased information have affected competition. Some markets have only one provider, and competition should thus not be affected by more information or lower switching costs. There are also markets, which are competitive but not exposed to new entry.

In the second step, we use these market-based differences in exposure to new entry to analyse the effects on primary health care quality in a reduced form difference-in-differences framework. Our preliminary results show no significant effects of high exposure to entry on the objective quality of care, as measured by per capita avoidable inpatient care episodes and unplanned hospital visits.  Next, we will use that markets were affected differently by increased information and lower switching costs to examine the effect of these features on the objective quality measures. Thereafter, we plan to analyse subjective measures of health from three waves of a national patient satisfaction survey.