Sleep Apnea Research Group

  Measures of OSA Burden and Outcomes

Polysomnography vs. Self-reported Measures in Patients with Sleep Apnea

Principal Investigator: Edward M. Weaver, MD, MPH

Co-Investigators: Vishesh Kapur, MD, MPH; Bevan Yueh, MD, MPH

Archives of Otolaryngology-Head and Neck Surgery 130: 453-458, April 2004

The objective of this research was to determine the correlation between polysomnography (PSG) indices and self-reported sleepiness, mental health status, and general health in patients with sleep apnea.

This was a cross-sectional study including ninety-six consecutive patients with sleep apnea confirmed by PSG with an apnea-hypopnea index (AHI) of 5 or more. Patients completed a questionnaire on the evening of diagnostic PSG which included Epworth Sleepiness Scale, Medical Outcomes Study 36-Item short-form health survey mental health domain (SF-36), and self-rated health. The associations between PSG indices and self-reported measures were adjusted for age, sex, body mass index, comoribity, and PSG type. The sample had 85% power to detect a correlation of 0.3 or greater.

Results: Overall, the PSG parameters correlated poorly with self-reported measures. AHI was not associated with self-reported sleepiness or general health. AHI was associated with SF-36 Health status mental health domain on multiple linear regression (p=.04) but not on multiple logistic regression (aOR=1.02, 95%CI 1.00-1.04, p=.09).

Conclusions: In this clinical sleep laboratory sample, PSG measures and AHI in particular, correlated poorly with self-reported measures. These findings suggest that sleep apnea disease burden should be quantified with both physiologic and subjective measures. Future research should be done to determine whether a change in physiologic measures after treatment correlates with a change in symptoms and quality of life.

Article: Polysomnography vs Self-reported Measures in Patients With Sleep Apnea

Polysomnography Indexes are Discordant with Quality of Life, Symptoms, and Reaction Time in Sleep Apnea Patients

Principal Investigator: Edward M. Weaver, MD, MPH

Co-Investigators: B. Tucker Woodson, MD; David L. Steward, MD

Otolaryngology-Head and Neck Surgery 135: 255-262, Feb 2005

This research represents a subgroup analyses of data from a randomized controlled trial testing whether polysomnography (PSG) indexes are associated with sleepiness, quality of life, and reaction times before and 8 weeks after surgical or sham treatment for mild to moderate obstructive sleep apnea syndrome (OSAS). These associations were examined with Spearman correlations and multivariate linear regression.

Results: Correlations between baseline PSG and non-PSG measures showed one significant association between arousal index and SF36 Mental Component Summary (r=0.25, p=0.03). Correlations between change in PSG and non-PSG measures showed no significant positive association among the 56 tested. Regression analyses confirmed these results. More specifically, the apnea-hypopnea index (AHI) was analyzed with multivariate regression adjusting for age, sex, and body mass index. There was no association between AHI and each of the non-PSG variables, nor between AHI and change in any non-PSG measure.

Conclusions: Standard baseline PSG indexes correlate poorly with other measures of OSAS burden. Moreover, the changes in PSG indices correlate poorly with changes in other measures important to OSAS, including subjective sleepiness, OSAS-quality of life, general health status and reaction times.

Article: Polysomnography Indexes are Discordant with Quality of Life, Symptoms, and Reaction Times in Sleep Apnea Patients

Sleep Study and Quality of Life Correlation in Sleep Apnea (abstract)

Principal Investigator: Edward M. Weaver, MD, MPH

Co-Investigators: Donald Patrick, PhD, MSPH; Richard Deyo, MD, MPH; Bevan Yueh, MD, MPH

This is an ongoing study assessing the relationship between objective and subjective outcomes of obstructive sleep apnea (OSA) in adults. The polysomnography (PSG), is considered the gold standard for assessing severity of OSA and outcomes of treatment. The physiologic metrics are assumed to estimate symptoms and quality of life (QOL) deficits associated with OSA.

This ongoing cross-sectional study involves patients referred for a full night, semi-monitored, in-lab diagnostic partial PSG at a VA facility for suspicion of sleep apnea. Respiratory Disturbance Index (RDI), Apnea Index (AI), lowest oxygen saturation (LSAT), and hypoxemic burden (% sleep time with oxygen saturation < 90%) were recorded. Immediately before the PSG, each patient completed a battery of self-administered validated instruments measuring functional status (SF36), general QOL (PQOL and Health Desirability), OSA-specific QOL (SNORE), and sleepiness (ESS). Spearman rank correlation coefficients were computed for the associations between PSG and subjective measures.

Results: Our sample (n = 31) consists of middle-aged (mean age 51 years), obese (mean body mass index 32 kg/m2), predominantly male (97% males) veterans. PSG parameters correlate poorly with the subjective measures (mean |rs| = 0.16 ± 0.22, range 0.005 0.35). 0/20 possible associations between PSG parameters (RDI, AI, LSAT, hypoxemic burden) and subjective measures (SF36, PQOL, Health Desirability, SNORE, ESS) were statistically significantly different from 0 (mean p = 0.44, range 0.054 0.98). Among the PSG parameters, RDI correlates least with the subjective parameters (mean |rs| = 0.10).

Conclusions: In our sample, PSG parameters are not associated with subjective measures of OSA severity. In particular, RDI, which is the most commonly reported parameter of OSA severity and currently defines our clinical categories of OSA, is not associated with subjective measures of OSA severity.

Abstract: Sleep Study and Quality of Life Correlation in Sleep Apnea

4333 Brooklyn Ave. NE
Box 359455, 14th Floor
Seattle, WA 98105
© 2002-2016
Comparative Effectiveness, Cost and Outcomes Research Center