Shen Yu-Chu, Eggleston Karen, Lau Joseph, Schmid Christopher H
Graduate School of Business and Public Policy, Naval Postgraduate School, Code GB, 555 Dyer Road, Monterey, CA 93943, USA.
Inquiry. 2007 Spring;44(1):41-68. doi: 10.5034/inquiryjrnl_44.1.41.
This study applies meta-analytic methods to conduct a quantitative review of the empirical literature on hospital ownership since 1990. We examine four financial outcomes across 40 studies: cost, revenue, profit margin, and efficiency. We find that variation in the magnitudes of ownership effects can be explained by a study's research focus and methodology. Studies using empirical methods that control for few confounding factors tend to find larger differences between for-profit and not-for-profit hospitals than studies that control for a wider range of confounding factors. Functional form and sample size also matter. Failure to apply log transformation to highly skewed expenditure data yields misleadingly large estimated differences between for-profits and not-for-profits. Studies with fewer than 200 observations also produce larger point estimates and wide confidence intervals.
本研究运用元分析方法，对自1990年以来有关医院所有权的实证文献进行定量综述。我们考察了40项研究中的四项财务结果：成本、收入、利润率和效率。我们发现，所有权效应大小的差异可以通过研究的重点和方法来解释。与控制了更广泛混杂因素的研究相比，使用控制较少混杂因素的实证方法的研究往往发现营利性医院和非营利性医院之间的差异更大。函数形式和样本量也很重要。对高度偏态的支出数据未进行对数变换会导致营利性和非营利性医院之间的估计差异大得具有误导性。观察值少于200的研究也会产生更大的点估计值和更宽的置信区间。
