Polysomnographic Measurement of Sleep Duration and Bodily Pain Perception in the Sleep Heart Health Study.

STUDY OBJECTIVES

To determine whether total sleep time (TST) and specific sleep stage duration are associated with bodily pain perception and whether sex, age, or subjective sleepiness modifies this relationship.

METHODS

Data from adults ages 39-90 y (n = 5,199) who took part in the Sleep Heart Health Study Exam 1 were analyzed. TST, rapid eye movement (REM) sleep time, and slow wave sleep (SWS) time were measured by unattended, in-home nocturnal polysomnography. Bodily pain perception was measured via the Short Form-36 questionnaire bodily pain component. We used logistic regression to examine associations between total and individual sleep stage durations and bodily pain perception controlling for age, sex, race, body mass index, apnea-hypopnea index, antidepressant use, and important cardiovascular conditions (smoking [pack-years], history of diabetes, and history of percutaneous coronary intervention and/or coronary artery bypass graft).

RESULTS

In the fully adjusted model, REM sleep time and SWS time were not associated with "moderate to severe pain," whereas TST was: Each 1-h decrement in TST was associated with a 7% increased odds of "moderate to severe pain" (odds ratio 1.07, 95% confidence interval 1.002, 1.14). Due to modification of the association between SWS time and "moderate to severe pain" by sex (P for interaction = 0.01), we performed analyses stratified by sex: Each 1-h decrement in SWS time was associated with a 20% higher odds of "moderate to severe pain" among men (odds ratio 1.20, 95% confidence interval 1.03-1.42) whereas an association was not observed among women.

CONCLUSIONS

Shorter TST among all subjects and shorter SWS time in men was associated with "moderate to severe pain." REM sleep time was not associated with bodily pain perception in this cohort.