Department of Biostatistics, Epidemiology and Informatics, University of Pennsylvania, Philadelphia, Pennsylvania, USA.
Division of Public Health Sciences, Washington University School of Medicine in St. Louis, St Louis, Missouri, USA.
J Am Med Inform Assoc. 2022 Jul 12;29(8):1366-1371. doi: 10.1093/jamia/ocac067.
To develop a lossless distributed algorithm for generalized linear mixed model (GLMM) with application to privacy-preserving hospital profiling.
The GLMM is often fitted to implement hospital profiling, using clinical or administrative claims data. Due to individual patient data (IPD) privacy regulations and the computational complexity of GLMM, a distributed algorithm for hospital profiling is needed. We develop a novel distributed penalized quasi-likelihood (dPQL) algorithm to fit GLMM when only aggregated data, rather than IPD, can be shared across hospitals. We also show that the standardized mortality rates, which are often reported as the results of hospital profiling, can also be calculated distributively without sharing IPD. We demonstrate the applicability of the proposed dPQL algorithm by ranking 929 hospitals for coronavirus disease 2019 (COVID-19) mortality or referral to hospice that have been previously studied.
The proposed dPQL algorithm is mathematically proven to be lossless, that is, it obtains identical results as if IPD were pooled from all hospitals. In the example of hospital profiling regarding COVID-19 mortality, the dPQL algorithm reached convergence with only 5 iterations, and the estimation of fixed effects, random effects, and mortality rates were identical to that of the PQL from pooled data.
The dPQL algorithm is lossless, privacy-preserving and fast-converging for fitting GLMM. It provides an extremely suitable and convenient distributed approach for hospital profiling.
开发一种用于广义线性混合模型（GLMM）的无损分布式算法，应用于隐私保护的医院概况分析。
GLMM 常用于实施医院概况分析，使用临床或行政索赔数据。由于个体患者数据（IPD）隐私法规和 GLMM 的计算复杂性，需要一种用于医院概况分析的分布式算法。我们开发了一种新颖的分布式惩罚拟似然（dPQL）算法，当只能在医院之间共享汇总数据而不是 IPD 时，该算法可用于拟合 GLMM。我们还表明，标准化死亡率（通常作为医院概况分析的结果报告）也可以在不共享 IPD 的情况下分布式计算。我们通过对先前研究过的 929 家因 2019 年冠状病毒病（COVID-19）死亡率或转介临终关怀而排名的医院进行展示，证明了所提出的 dPQL 算法的适用性。
所提出的 dPQL 算法在数学上被证明是无损的，也就是说，它与从所有医院汇总 IPD 获得的结果完全相同。在 COVID-19 死亡率的医院概况分析示例中，dPQL 算法仅需 5 次迭代即可收敛，并且固定效应、随机效应和死亡率的估计值与来自汇总数据的 PQL 相同。
dPQL 算法是无损的、隐私保护的和快速收敛的，适用于拟合 GLMM。它为医院概况分析提供了一种非常合适和方便的分布式方法。
