Report

Teaser, summary, work performed and final results

Periodic Reporting for period 1 - LETHE (Levels and Trends of Health Expectancy: Understanding its Measurement and Estimation Sensitivity)

Teaser

The project aims at assessing the effects of the methodological features of “Health Expectancy” (HE). The general understanding is that this indicator simply extends the average “Life Expectancy” by one dimension. Technically, this is correct because the total number...

Summary

The project aims at assessing the effects of the methodological features of “Health Expectancy” (HE). The general understanding is that this indicator simply extends the average “Life Expectancy” by one dimension. Technically, this is correct because the total number of life years is divided into two quality dimensions: life years spent in good health and those spent in poor health. However, incorporating this additional dimension to the life table makes the indicator extremely sensitive to certain measurement and estimation issues. This is an important problem because HE is not only becoming more and more used in health research. It is also the most important structural indicator in health policy.

Our analyses of the HE indicators’ measurement and estimation sensitivity is conducted in the context of several major actual research topics, in particular the “compression versus expansion of morbidity” debate. Consequently, the outcome parameter is HE, estimated on the basis of different health measures, methods and estimation details. The basic model underlying all these estimates is the life table which will accordingly form the foundation of our analyses.

We perform the outlined research in a systematic manner, following a well-structured research plan. The tasks of the project are organized and conducted within nine “Work Packages” (WPs). Two of them (WPs 1: Database of health indicators; WP2: Overview of characteristics of common HE estimates) will provide the basic resources for the analyses and publications (data and literature). Three WPs 3-5 (WP3: Sensitivity of HE to reporting behaviour; WP4: CAHL as a new indicator for HE; WP5: Sensitivity of HE to measurement and estimation methods) will systematically analyse particular domains of the HE indicator’s measurement- and estimation-related sensitivity as well as the extent of the associated biases. The two WPs WP6 (Health conditions measured with self-reported indicators) and WP7 (Association between health measure(s) and happiness) will aim to better understand and improve the measurement of health on the basis of self-reported information from surveys. Finally, WP8 (European Health and Longevity Datasheet) and WP9 (Publication of a book about the project’s findings) will serve the dissemination of the project results among scientists, policy makers and other stakeholders who are interested in levels and trends in HE by producing useful materials.

Work performed

The main tasks so far (April 2019) were related to WP1 (Database of health indicators), WP2 (Overview of characteristics of common HE estimates), WP3 (Sensitivity of HE to reporting behaviour), WP4 (CAHL as a new indicator for HE), WP5 (Sensitivity of HE to measurement and estimation methods), and WP6 (Health conditions measured with self-reported indicators). The works were carried out as planned and the team was very successful in presenting the findings at several high impact conferences. During these presentations we noticed the strong recognition of our work. This confirmed our hypothesis that most users of the common health expectancy indicators are not aware the indicators’ sensitivities regarding various methodological and estimation issues. These can be problematic in practice and lead to misleading conclusions. This motivated us to give even more emphasis to our dissemination activities and we intend to strengthen these even stronger in the remaining project time. The first papers are already published or currently under review.

Final results

The probably most obvious example for HE’s estimation sensitivity is the definition of health. For example, when we calculate the proportions of European countries by level of change in HE from 2010 to 2013 on the basis of Global Activity Limitation Indicator (GALI), the health indicator used by the European Commission, we find that only a little bit more than half of the countries are on a good way with increasing HEs. However, when we evaluate the progress with the GBD as health measure, the health indicator used by the WHO, the situation looks much different: here, almost 95 percent of the populations are on a good or even very good track. This is an extreme difference to the GALI figures which leads to very different conclusions.