Abstract
Purpose
Accuracy of patient-perceived sleeping position has implications for the acquisition and interpretation of sleep studies, implementation of positive airway pressure (PAP) therapies, and determining efficacy of positional therapies at home. This study aimed to compare self-reported supine sleep to observed supine sleep on polysomnography (PSG) and identify patient-related factors that contribute to supine sleep misperception.
Methods
A retrospective review of clinical PSG records from a public sleep service was performed. Body position was measured with a Grael position sensor with manual editing based on digital video if discrepant. Each patient completed a sleep questionnaire including questions regarding sleep position on the PSG. Self-reported supine sleep was divided into categories of “None/Some/Half/Most/All”, with PSG-measured supine sleep as a percentage of total sleep time (%TST) classified into these categories based on the cut-points of 0, 2.5, 33.3, 66.6, 97.5 and 100%. Absence of supine sleep on PSG was defined by the cut-point of “None” (≤ 2.5%TST). Chance-adjusted agreement between objective and subjective measures was assessed using the Cohen’s kappa statistic (linear weighting).
Results
A total of 956 patient records were available for analysis. 93% of patients who self-reported supine sleep on the PSG were correct, whereas only 56% of those who denied supine sleep accurately reported the absence of supine sleep on PSG. Patients who reported not sleeping on their back during the PSG had 11.0 ((95% CI: 7.0, 17.3), p < .001) times the odds of sleep position misperception than those who did report sleeping on their back. Agreement with self-reported categories of “None/Some/Half/Most/All was only moderate (kappa 0.45 [95% CI 0.43, 0.48]). Younger patients (< 55 years) and females were more likely to under-report supine sleep.
Conclusions
Patients denying supine sleep in the laboratory were more likely to misperceive sleep position on PSG, with younger females under-reporting supine sleep. Understanding contributors to misperception may influence decisions regarding supine sleep sampling in the laboratory, as well clinical decision making.
