MEDCIDS-Department of Community Medicine, Information and Health Decision Sciences, Faculty of Medicine, University of Porto, Porto, Portugal
CINTESIS-Centre for Health Technology and Services Research, Porto, Portugal.
J Epidemiol Community Health. 2021 Aug;75(8):764-771. doi: 10.1136/jech-2020-213791. Epub 2021 Jan 15.
Healthy life expectancy (HLE) is a population health measure that combines mortality and morbidity, which can be calculated using different methods. In this study, we aimed to assess the correlation, reliability and (dis)agreement between two estimates monitored in the European Union (EU), that is, the European Commission's HLE based on self-perceived health (SPH-HLE) and the Institute for Health Metrics and Evaluation's HLE based on disability weight (DW-HLE), by sex, and comparing these results with LE and proportion of life spent in good health (%GH).
We performed a retrospective study in the EU28 countries, between 2010 and 2017. The HLE methods differ in definition, measurement and valuation of health states. While SPH-HLE relies directly on one question, DW-HLE relies on epidemiological data adjusted for DW. Spearman's r, intraclass correlation coefficient, information-based measure of disagreement and Bland-Altman plots were used to assess reliability, correlation and disagreement in HLE resulting from both methods and in LE or %GH measured by both institutions.
Correlation and reliability between SPH-HLE and DW-HLE were good (better for males), with low disagreement, and were even better for LE between both institutions. The HLE Bland-Altman plots suggest a variability range of approximately 6 years for both sexes, higher for females. There was also an increasing HLE difference between methods with higher average HLE for both sexes.
We showed wide variations between both methods with a clear and different high impact on female and male HLE, showing a tendency for countries with higher health expectancies to yield larger gaps between SPH-HLE and DW-HLE.
健康期望寿命（HLE）是一项结合了死亡率和发病率的人口健康指标，可使用不同方法进行计算。本研究旨在评估欧洲联盟（EU）监测的两种 HLE 估算值之间的相关性、可靠性和（差异）一致性，即欧盟委员会基于自我感知健康的 HLE（SPH-HLE）和基于残疾权重的 HLE（DW-HLE），并按性别进行比较，同时将这些结果与 LE 和良好健康状态下的生命比例（%GH）进行比较。
我们在 EU28 国家进行了一项回顾性研究，时间范围为 2010 年至 2017 年。HLE 方法在健康状况的定义、测量和评估方面存在差异。SPH-HLE 直接依赖于一个问题，而 DW-HLE 则依赖于调整了残疾权重的流行病学数据。我们使用 Spearman's r、组内相关系数、基于信息的差异度量和 Bland-Altman 图来评估两种方法得出的 HLE 以及两个机构测量的 LE 或%GH 的可靠性、相关性和差异。
SPH-HLE 和 DW-HLE 之间的相关性和可靠性都很好（男性更好），差异较小，两个机构之间的 LE 也更好。HLE Bland-Altman 图表明，两种性别之间的 HLE 差异约为 6 年，女性的差异更大。两种方法之间的 HLE 差异也在增加，男女的平均 HLE 越高，差异越大。
我们发现两种方法之间存在广泛的差异，对女性和男性的 HLE 有明显且不同的影响，表明健康期望寿命较高的国家之间的 SPH-HLE 和 DW-HLE 之间的差距更大。
