Construct validation of actigraphic sleep measures in hospitalized depressed patients.

This study validated wrist actigraphic-measured sleep in depressed patients using construct validity by experimental intervention methods. The experimental participants were 18 patients hospitalized for major depression. Control participants were hospital staff. A 2-between (depressed patients vs. controls) x 2-within (pre- vs. post-) experimental design was used. Sleep was evaluated for 1 week, 7 nights, using wrist actigraphy on hospitalization and for a second week just prior to discharge. Clinical improvement was corroborated by statistically significant changes in the Beck Depression Inventory and the Inventory to Diagnose Depression. Sleep-onset latency, number of nighttime awakenings, minutes awake after sleep onset, and sleep efficiency all improved significantly as hypothesized. Minutes of sleep changed in the predicted direction but not significantly. Significant differences from control participants remained at discharge regarding minutes awake after sleep onset and sleep efficiency. These findings extend practice guidelines for actigraphy established by the Standards of Practice Committee (1995) of the American Sleep Disorders Association.