This article describes experience with the use of health diaries in a two-year prospective study of the influence of the psychosocial environment on the health status of 500 subjects in the Hamilton area. Three strategies were used to maintain compliance: 1) random sampling of three days within each two-week interval; 2) a lottery ticket incentive; and 3) telephone follow-up. The results of an initial telephone and lottery ticket study demonstrated a significant effect on compliance. The data from the two-year study were analyzed to characterize health change based on diary reports over successive six-month intervals and to determine the relationship of reported health to medical utilization data. The results indicated that subjects reported symptoms on about one third of the days surveyed but took no action on about one third of these symptom days. Only one quarter of the symptom days resulted in a change in usual activity, and fewer than one tenth of the symptom days resulted in time off from work or physician visits. The diary variables showed a low, positive correlation with health utilization.