There are considerable differences in the incidence of sickness, the pattern of diseases and the duration of episodes of sickness by age, gender, education and employment status. Employed have a significantly better health record than the unemployed. The question why this should be so is followed up in this paper. The data base is microcensus data, i.e., household health surveys, complemented by administrative data (morbidity rates of the employed and unemployed) and data from the European Community Household Panel (ECHP). The starting point of the analysis is the calculation of morbidity rates of different population groups, i.e., the number of days of sickness over a year, followed by the establishment of the linkage between the morbidity rate and an index of subjective feelings of wellbeing. The morbidity rate can be decomposed into the frequency of spells of sickness and the average duration of a spell of sickness. The age group which experiences the most frequent episodes of sickness are the elderly, followed by small children. However, the average duration of a spell of sickness is comparatively short in the case of children and quite long for older persons. Accordingly, the morbidity rate by age is slightly u-shaped, i.e., children under five have higher rates than youngsters between 5 and 15, the age groups with the lowest sickness rates. Thereafter the sickness rate rises exponentially with age. The gender pattern of morbidity rates by age has changed between the early 1970s and the late 1990s ...
