Self-reported sleep in postmenopausal women : Menopause (original) (raw)

Articles

Regestein, Quentin R. MD1; Friebely, Joan EdD1; Shifren, Jan L. MD2; Scharf, Martin B. MD3; Wiita, Brinda PhD4; Carver, Judith BA4; Schiff, Isaac MD2

From 1Brigham & Women's Hospital and 2Massachusetts General Hospital, Boston, MA; 3Tri-State Sleep Disorders Center, Cleveland, OH; and 4Johnson & Johnson, Skillman, NJ.

Received February 26, 2002; revised and accepted August 19, 2003.

This study received support from Johnson and Johnson Inc.

Address correspondence to: Quentin R. Regestein, MD, 75 Francis St., Boston, MA 02115. E-mail: [email protected].

Abstract

Objective

We aimed to find how self-reported sleep (measured by the St. Mary's Hospital Sleep Questionnaire) in postmenopausal women having hot flash activity was related to objective sleep (actigraphy), psychological and somatic symptoms [Women's Health Questionnaire (WHQ)], and cognitive test performance (computerized tests). A secondary aim was to find if self-reported sleep showed expected correlations with hyperarousal (Hyperarousal Scale).

Design

Drug trial baseline data from 88 healthy, postmenopausal women were retrospectively analyzed. Multivariate regression was used to adjust for confounder variables and test whether differences in self-reported sleep measures were systematically associated with differences in objective sleep, WHQ symptom measures, or cognitive test performance scores.

Results

Increased self-report scores for low sleep quality were associated with an increased risk of WHQ symptoms and reduced cognitive test performance. Self-reported sleep measures showed little correlation with their analogous objective measures. Self-reported low sleep quality proved most closely associated with the WHQ symptoms of tiredness, clumsiness, and difficulty concentrating. Women whose self-reported sleep-onset latency times were longer than the median overestimated their objective sleep onset time by 30 min, whereas the other women underestimated theirs by 15 min (P < 0.0001). Women whose self-reported total sleep was longer or shorter than the median, respectively, underestimated objective sleep times by 9 and 71 min (P < 0.0001). High hyperarousal scores were associated with underestimations of objective sleep.

Conclusion

Self-reports of lower sleep quality were associated with increased WHQ psychological and somatic symptom measures and decreased cognitive test performance more than with differences in objective sleep. Self-reported trouble sleeping may signal problems independent from objectively low sleep quality, such as subjective distress or diminished cognitive function.