Patient-predicted sleep position vs. HST data: a tendency to underestimate supine sleep

The purpose of this study is to measure people's accuracy when they estimate what proportion of their nightly sleep at home is supine vs. non-supine.

A series of patients referred for obstructive sleep apnea (OSA) evaluation were asked if they "knew with confidence" how they slept with regard to their body position. "Yes" responders were then asked to estimate what percentage of their sleep was supine vs. non-supine. This value was compared with the actual proportion of supine vs. non-supine sleep that they exhibited in a home sleep test (HST) that followed.

We obtained data from 49 subjects who expressed that they "knew with confidence" how they sleep in terms of body position. Subjects in aggregate underestimated their proportion of supine sleep by 21.6% (p < .001). Thirty-nine subjects (80%) slept supine more in the HST whereas 8 (16%) slept supine less compared to their pre-test estimates. Using a common classification of OSA severity, 9 subjects (18%) demonstrated a more severe degree of OSA than would have occurred had they slept as they had predicted.

Subjects in this study frequently underestimated their proportion of supine sleep compared to values measured in an HST. Because of the increased supine sleep they exhibited, the severity of their OSA was often greater in the test than it would have been had the subjects slept as they predicted. Sleep physicians should take into account the tendency of people to underestimate supine sleep. If patients with positional sleep apnea assert that they "always sleep laterally" when at home, they may be underestimating their true night-by-night OSA disease burden.