Depression is commonly associated with abnormal sleep architecture. This thesis undertook to assess sleep architecture as a biological correlate of self and observer-rated depressive state, and consists of three studies. The first used a categorical approach to examine the association of sleep architecture with subjective mood in a community sample of 74 preoperative patients, and found no association between high depression scores and hypothesized sleep markers. The second followed 16 patients with Major Depression who were treated with the antidepressant mirtazapine in an 8 week longitudinal study during which they underwent polysomnography on 6 occasions. It was found that classes of sleep markers (REM latency or REM, arousal index, and slow wave sleep) tend to predict response when analyzed concurrently. The third study was methodological in nature, and found that commercially available software for automating eye movement counts did not show strong correspondence with visually scored polysomnographic data.

Depression is commonly associated with abnormal sleep architecture. This thesis undertook to assess sleep architecture as a biological correlate of self and observer-rated depressive state, and consists of three studies. The first used a categorical approach to examine the association of sleep architecture with subjective mood in a community sample of 74 preoperative patients, and found no association between high depression scores and hypothesized sleep markers. The second followed 16 patients with Major Depression who were treated with the antidepressant mirtazapine in an 8 week longitudinal study during which they underwent polysomnography on 6 occasions. It was found that classes of sleep markers (REM latency or REM, arousal index, and slow wave sleep) tend to predict response when analyzed concurrently. The third study was methodological in nature, and found that commercially available software for automating eye movement counts did not show strong correspondence with visually scored polysomnographic data.