Centre for Health Economics, University of York, Alcuin A Block, Heslington, York YO10 5DD, United Kingdom.
Health Policy. 2013 Sep;112(1-2):88-99. doi: 10.1016/j.healthpol.2013.06.011. Epub 2013 Jul 24.
There is a fundamental gap in the evidence base on quantitative cross-country comparison of mental healthcare systems due to the challenges of comparative analysis in mental health including a paucity of good quality data. We explore whether existing limited data sources can potentially be exploited to examine technical efficiency of inpatient mental healthcare systems in 32 OECD countries in 2010. We use two analytical approaches: Data Envelopment Analysis (DEA) with bootstrapping to produce confidence intervals of efficiency scores and country rankings, and Cluster Analysis to group countries according to two broad efficiency groupings. We incorporate environmental variables using a two-stage truncated regression. We find slightly tighter confidence intervals for the less efficient countries which loosely corresponds with the 'inefficient' cluster grouping in the Cluster Analysis. However there is little stability in country rankings making it difficult with current data to draw any policy inferences. Environmental factors do not appear to significantly impact on efficiency scores. The most pressing pursuit remains the search for better national data in mental healthcare to underpin future analyses. Otherwise the use of any sophisticated analytic techniques will prove futile for establishing robust conclusions regarding international comparisons of the performance of mental healthcare systems.
由于精神卫生领域比较分析存在挑战，包括高质量数据的匮乏，因此在精神卫生保健系统的跨国定量比较方面，证据基础存在根本差距。我们探讨了现有有限的数据源是否可以被利用来检验 2010 年 32 个经合组织国家住院精神卫生保健系统的技术效率。我们使用两种分析方法：数据包络分析（DEA）和带引导的置信区间来生成效率得分和国家排名，以及聚类分析根据两个广泛的效率分组对国家进行分组。我们使用两阶段截断回归纳入环境变量。我们发现，对于效率较低的国家，置信区间稍微收紧，这与聚类分析中的“效率低下”聚类分组大致对应。然而，国家排名的稳定性很差，因此很难根据现有数据得出任何政策推论。环境因素似乎对效率得分没有显著影响。最紧迫的任务仍然是寻找更好的国家精神卫生保健数据，为未来的分析提供支持。否则，使用任何复杂的分析技术都将无法为精神卫生保健系统的国际比较性能建立可靠的结论。
