Abstract
The analysis of health diary data has long relied on inferential statistical methods focusing on sample means and ad hoc methods to calculate each individual's variation in health outcomes. In this paper, an advanced statistical model is applied to daily diary self-reported health outcomes to simultaneously study an individual's likeliness to report an outcome, daily mean intensity level, and variability in daily measures. Using observational, secondary data from 782 adults, we analyzed self-report daily fatigue symptoms, distinguishing between whether an individual reported fatigue and its severity when reported. Self-reported depressed affect and participant characteristics were used as predictors of daily fatigue symptoms. A higher likeliness to report fatigue correlated with higher mean fatigue severity and greater stability in severity ratings. Higher mean severity correlated with greater stability in severity ratings. Females and those with high depressed affect were more likely to report fatigue. Females and those with high depressed affect reported greater mean severity. The model applied to daily measures allowed for the simultaneous study of an individual's likeliness to report a symptom, daily mean symptom severity, and variability in severity across days. An individual's daily variation in symptom severity was represented as a model parameter that did not contain measurement error that is present in ad hoc methods.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call