NBER Working Papers · 2026 · Melissa McInerney、Lorena Moreno、Jennifer M. Mellor
中文摘要
本文比较可能患有痴呆症者（PWPD）与匹配对照组在痴呆症发病前后的Medicare补充保险覆盖情况。本研究使用倾向得分权重和2000—2016年健康与退休研究（HRS）数据，对保险覆盖变化进行回顾性二次数据分析。样本包括3,001名参加Medicare、年龄65岁及以上的可能患有痴呆症者，以及14,055名对照者。可能痴呆症的判定采用Langa-Weir方法，代理访谈则采用访员评估。发病前四年，与匹配对照组相比，可能患有痴呆症者拥有Medicare但无补充保险的可能性更高，高出2.97个百分点（p=0.0016）。这反映了以下因素的净效应：可能患有痴呆症者拥有Medicare Advantage、或拥有附带补充保险的传统Medicare的可能性较低，而拥有Medicaid的可能性较高。发病前两年，可能患有痴呆症者符合Medicaid资格的可能性更高，高出4.95个百分点（p<0.001），但参保率相当。在痴呆症发病前后提供参保协助，可能改善可能患有痴呆症者的补充保险覆盖，从而提高其获得医疗服务的经济可及性。
Abstract
In this paper, we compare supplemental Medicare insurance coverage among people with probable dementia (PWPD) and matched controls, before and after onset. This retrospective secondary data analysis of insurance coverage changes using propensity score weights and the 2000-2016 Health and Retirement Study (HRS). The sample consists of 3,001 Medicare-enrolled adults age 65 and older with PWPD, and 14,055 controls. Probable dementia is assessed with the Langa-Weir approach and interviewer assessments for proxy interviews. Four years before onset, PWPD become more likely to have Medicare without supplemental coverage (2.97 percentage points; p=0.0016) than matched controls. This reflects the net effect of PWPD being less likely to have Medicare Advantage or traditional Medicare with supplemental coverage, and more likely to have Medicaid. Two years before onset, PWPD were more likely to be Medicaid-eligible (4.95 percentage points, p<0.001), but take-up rates were comparable. Enrollment assistance may improve supplemental coverage for PWPD around dementia onset, potentially increasing financial access to care.