Measuring elders' symptoms with daily diaries and retrospective reports.

The current study investigated the sufficiency of symptom data collected for 7 days from a daily diary and the comparability and validity of data obtained from daily diaries and retrospective reports. Three hundred and five older persons completed a daily symptom diary, a retrospective symptom questionnaire, and a measure of health status. The number of symptoms endorsed each day on the diary declined, F(6, 214) = 13.51, p< .0001. New symptoms were endorsed 15%, and previous symptoms were no longer endorsed 1.9% of the time. More symptoms were endorsed on retrospective reports than on daily diaries, t(304) = 8.48, p < .0001. Symptoms from both methods were significantly correlated with health status (p <.0001) and differences in the correlations were not statistically significant, t(304) = 1.62, p = ns . Because of comparable criterion validity, less burdensome retrospective reports should be used, unless the focus is new symptoms or how symptoms unfold over time.