Department of Epidemiology, University of North Carolina, Chapel Hill, NC, USA.
Lineberger Comprehensive Cancer Center, University of North Carolina, Chapel Hill, NC, USA.
Pharmacoepidemiol Drug Saf. 2019 May;28(5):584-592. doi: 10.1002/pds.4769. Epub 2019 Mar 19.
De-implementation of low-value services among patients with limited life expectancy is challenging. Robust mortality prediction models using routinely collected health care data can enhance health care stakeholders' ability to identify populations with limited life expectancy. We developed and validated a claims-based prediction model for 5-year mortality using regularized regression methods.
Medicare beneficiaries age 66 or older with an office visit and at least 12 months of pre-visit continuous Medicare A/B enrollment were identified in 2008. Five-year mortality was assessed through 2013. Secondary outcomes included 30-, 90-, and 180-day and 1-year mortality. Claims-based predictors, including comorbidities and indicators of disability, frailty, and functional impairment, were selected using regularized logistic regression, applying the least absolute shrinkage and selection operator (LASSO) in a random 80% training sample. Model performance was assessed and compared with the Gagne comorbidity score in the 20% validation sample.
Overall, 183 204 (24%) individuals died. In addition to demographics, 161 indicators of comorbidity and function were included in the final model. In the validation sample, the c-statistic was 0.825 (0.823-0.828). Median-predicted probability of 5-year mortality was 14%; almost 4% of the cohort had a predicted probability greater than 80%. Compared with the Gagne score, the LASSO model led to improved 5-year mortality classification (net reclassification index = 9.9%; integrated discrimination index = 5.2%).
Our claims-based model predicting 5-year mortality showed excellent discrimination and calibration, similar to the Gagne score model, but resulted in improved mortality classification. Regularized regression is a feasible approach for developing prediction tools that could enhance health care research and evaluation of care quality.
在预期寿命有限的患者中取消低价值服务具有挑战性。使用常规收集的医疗保健数据构建稳健的死亡率预测模型可以增强医疗保健利益相关者识别预期寿命有限人群的能力。我们使用正则化回归方法开发和验证了一种基于索赔的 5 年死亡率预测模型。
从 2008 年开始，确定了年龄在 66 岁或以上、有门诊就诊记录且至少有 12 个月的就诊前连续 Medicare A/B 注册记录的 Medicare 受保人。通过 2013 年评估 5 年死亡率。次要结果包括 30 天、90 天、180 天和 1 年死亡率。使用正则化逻辑回归选择基于索赔的预测因子，包括合并症以及残疾、衰弱和功能障碍的指标，在 80%的随机训练样本中应用最小绝对值收缩和选择算子（LASSO）。在 20%的验证样本中评估模型性能并与 Gagne 合并症评分进行比较。
共有 183204 人（24%）死亡。除了人口统计学数据外，最终模型还纳入了 161 个合并症和功能指标。在验证样本中，c 统计量为 0.825（0.823-0.828）。中位预测 5 年死亡率为 14%；队列中几乎有 4%的人预测概率大于 80%。与 Gagne 评分相比，LASSO 模型导致 5 年死亡率分类得到改善（净重新分类指数为 9.9%；综合鉴别指数为 5.2%）。
我们基于索赔的预测 5 年死亡率模型显示出优异的区分度和校准度，与 Gagne 评分模型相似，但导致死亡率分类得到改善。正则化回归是开发预测工具的一种可行方法，可增强医疗保健研究和评估护理质量。
