Performing Through Privatization: An Ecological Natural Experiment of the Impact of the Swedish Free Choice Reform on Ambulatory Care Sensitive Conditions

In 2010, Sweden opened up for establishment of privately owned primary health care providers, as part of a national Free Choice in Primary Health Care reform. The reform has been highly debated, and evidence on its effects is scarce. The present study therefore sought to evaluate whether the reform have impacted on primary health care service performance. This ecological register-based study used a natural experimental approach through an interrupted time series design. Data comprised the total adult population of the 21 counties of Sweden 2001-2009 (pre-intervention period) and 2010-2016 (post-intervention period). Hospitalizations and emergency department visits for ambulatory care sensitive conditions (ACSC) were used as indicators of primary health care performance. Segmented regression analysis was used to assess the effects of the reform, in Sweden as a whole, as well as compared between counties grouped by (i) change in private provision pre- to post reform; (ii) the timing of the implementation; and (iii) sustained presence of private providers both pre- and post-reform. The results suggest that, following the introduction of the reform in Sweden as a whole, the trends in total hospitalizations rates were slowed down by 1.0% albeit acute emergency visits increased 1.1% more rapidly after the introduction of the reform. However, we found no evidence of more beneficial effects in counties where the reform had been implemented more ambitiously, specifically those with a larger increase in private primary care providers, or where the reform was introduced early and thus had longer time effects to emerge. Lastly, counties with a sustained high presence of private primary care providers displayed the least favorable development when it comes to ACSC. Taken together, the present study does not support that the Swedish Free Choice reform has improved performance of the primary care delivery system in Sweden, and suggests that high degree of private provision may involve worse performance and higher care burden for specialized health care. Further evaluations of the consequences of the reform are dire needed to provide a comprehensive picture of its intended and unintended impact on health care provision, delivery and results.