Period- and cohort factors and shifts in self-perceived health of 55-64-year-olds between 1992 and 2002

Dorly J. H. Deeg, Vrije Universiteit Amsterdam
Karline Van de Kamp, Vrije Universiteit Amsterdam
Arjan W. Braam, Vrije Universiteit Amsterdam

Self-perceived health describes how a person perceives his or her own health. It is a widely used measure of health status and an important predictor of mortality. The aim of this study is to investigate possible differences in self-perceived health between 1992-93 and 2002-03 of men and women aged 55-64 years, and the explanation of these differences by factors that differ between birth-cohorts (cohort-factors) and by secular changes (period-factors). Two age-, sex- and region-stratified population-based samples are used from the Longitudinal Aging Study Amsterdam (LASA), in the Netherlands. The first sample ('oldest cohort') was first interviewed in 1992-92; the second sample ('most recent cohort'), in 2002-02. Self-perceived health is defined by asking the following question: How would you rate your health in general? Multinomial regression analyses are used to examine the differences in prevalence of excellent, good, fair and poor self-perceived health, and the cohort- and period-factors that are responsible for the cohort-difference. The most recent cohort rated their health more often as excellent and poor than the oldest (2 and 5 percent points, respectively), and lessoften as good and fair (2 and 4 percent respectively). The most recent cohort had a higher prevalence of chronic illness, functional limitations and depression, which negatively affected self-perceived health. After adjusting for health, the sign of the difference between the samples changed, in that the most recent cohort perceived their health less often as poor. Cohort- and period-factors did not explain the cohort difference. In conclusion, there is a shift in self-perceived health over time. The most recent cohort rate their health more favourably than the oldest cohort, given their poorer objective health. This finding suggests that the observed shift in self-perceived health disguises the true change of health of the population.