Department of Learning, Informatics, Management & Ethics (LIME), Karolinska Institutet, SE, 17177, Stockholm, Sweden; Stockholm Centre for Health Economics, Region Stockholm, Karolinska Institutet, Tomtebodavägen 18A, SE, 17177, Stockholm, Sweden.
School of Public Health and Community Medicine, Institute of Medicine, University of Gothenburg, Box 428, SE-40530, Gothenburg, Sweden.
Soc Sci Med. 2024 Apr;346:116711. doi: 10.1016/j.socscimed.2024.116711. Epub 2024 Feb 22.
Quasi-market reforms have been increasingly implemented in tax-funded health care, but their effects in terms of equity, quality and socioeconomic differentials in quality remain sparsely studied. We create a natural experiment setup exploiting the differential timing of a set of quasi-market reforms - including patient choice, free establishment of providers and changes in provider remuneration -, implemented in primary care in the two largest Swedish regions (Stockholm and Västra Götaland) in 2008-2009. Using a database with individual level data from 2005 to 2009, we construct a difference-in-difference-in-differences model that compares pre to post reform changes in avoidable hospitalizations (AHs) for low-income elders and a matched comparison group, in the region exposed to, versus unexposed to, reform (total N ∼ 200 000). The results show that for low-income elders - a group dominated by older women - reform led to higher AH rates, i.e., worse primary health care quality, than what would have been the case in absence of reform. Specifically, low-income elders exposed to reform missed out on improvements in AHs seen simultaneously in the unexposed region. At the same time, the reform had on average no effect for comparable, non-low-income, peers. The fact that this pattern was specific for avoidable hospitalizations - judged as amenable to interventions in primary care -, but not present for total hospitalizations, supports that it was driven by reform implementation rather than other factors. The study contributes with high-quality empirical evidence to a policy relevant but sparsely researched area and highlights the necessity to consider differential effects of organizational changes across socioeconomic groups.
准市场改革已越来越多地应用于税收资助的医疗保健中，但对于这些改革在公平性、质量以及社会经济差异方面对质量的影响，相关研究仍然很少。我们利用一组准市场改革（包括患者选择、提供者自由设立以及提供者报酬变化）在瑞典两个最大的地区（斯德哥尔摩和西约塔兰）于 2008-2009 年实施的时间差异，创建了一个自然实验设置。我们使用一个包含 2005 年至 2009 年个人层面数据的数据库，构建了一个差异中的双重差分模型，该模型比较了在改革前和改革后低收入老年人的可避免住院率（AHs）的变化，以及在改革地区（总 N∼200000）与未改革地区的匹配对照组。结果表明，对于低收入老年人（主要由老年女性组成的群体），改革导致 AH 率上升，即初级卫生保健质量恶化，而如果没有改革，情况可能会有所不同。具体来说，暴露于改革的低收入老年人错过了同时在未暴露地区观察到的 AH 改善。与此同时，改革对可比的非低收入同龄人平均没有影响。这种模式仅针对可避免的住院治疗（被认为可通过初级保健干预），而不是总住院治疗，这一事实支持了它是由改革实施而不是其他因素驱动的。该研究为政策相关但研究甚少的领域提供了高质量的经验证据，并强调了考虑组织变革对社会经济群体的差异化影响的必要性。
