Abstract
CPAP effectively treats obstructive sleep apnea (OSA), but as is widely recognized, adherence with the treatment remains a significantly less than desirable in the population. Recent work has begun to explore the association between insomnia and CPAP adherence and indicates that those with co-morbid OSA and insomnia are less likely to adhere to PAP therapy. Less is known the meditational link of CPAP adherence between nighttime symptoms (i.e., insomnia, OSA) and subjective daytime sleepiness in co-morbid insomnia/OSA. We hypothesized that OSA would be positively associated with adherence, insomnia would be negatively associated with adherence and the relationship between nighttime symptoms and daytime sleepiness would be fully mediated by adherence to CPAP.
Participants were 237 veterans who attended follow-up CPAP Clinic at the Miami VA. At that visit, participants had a CPAP download, completed a battery of questionnaires including the Epworth Sleepiness Scale, and the Insomnia Severity Index.Information regarding medical/psychiatric conditions as well as PSG results was extracted from the medical record.A structural equation model with latent variables was constructed and a full meditational model was tested in mPlus.The model hypothesized that the nighttime symptoms (i.e., insomnia, OSA) effect on subjective daytime sleepiness were fully mediated by adherence to CPAP.
Adequate statistical support was provided for the hypothesized model (ChiSq=270.44, df=143; RMSEA=0.061; CFI=0.93,SRMR=0.075). However, improvement in overall model fit could be made.A second model was hypothesized in which partial mediation was tested and direct paths from nighttime symptoms to daytime sleepiness were added.This model showed a significant improvement in fit (ChiSq change=15.17, df=2, p<.001). Overall model fit parameters are: ChiSq=255.27, df=141; RMSEA=0.058; CFI=0.94,SRMR=0.056. Path coefficients in the final model indicated that CPAP adherence, counter to our hypothesis, was not a significant mediator between nighttime symptoms and daytime sleepiness. Only nighttime insomnia symptoms were related to daytime sleepiness.
Insomnia appears to be a particularly troublesome symptom in those with OSA. First, it is negatively associated with adherence, and second, it is positively associated with subjective daytime sleepiness. These results indicated that neither the AHI or adherence were related to daytime sleepiness. More attention should be given to insomnia symptoms in those with OSA.
No Acknowledgements.