An individual's leisure time (pastime) engagements are in a way important for society. Irrespective of whether leisure time activities are causal determinants of health or health is a prerequisite for taking full part in society, the interaction is a challenge for Public Health. The first question is whether the sum of their influence results in coherence between enjoying good health and having leisure time activity. The aim of this study was to estimate their covariance.A random sample was drawn from the adult population of Stockholm County, Sweden. The residents were mailed a questionnaire regarding their social circumstances, their health complaints, the social repercussions of the complaints, and if they had any leisure time activity. Altogether, 7252 (about 93%) individuals responding to the questionnaire constituted our cohort. The individuals aged 18-65 y in November 1969 were followed up to the age of 65 y, or to 1996 as to mortality. The main outcome measure was mortality irrespective of cause of death. More than two-thirds of the respondents (71%) reported that they had some leisure time activity. Leisure time activity was a determinant of survival in the statistical sense. The risk ratio was 0.77 and its 95% confidence interval was 0.68-0.87 for those reporting leisure time activity when age, sex, baseline health and baseline alcohol consumption was discounted. The conclusion was that having leisure time activity, unspecified, covaried with health. Further studies should specify the pastime activities.