Norwegian Study Shows CPAP Use Linked to Improving Insomnia in Patients With Sleep Apnea

A longitudinal study from Norway published in Frontiers in Psychology has shown the positive effect that adherence to use of continuous positive airway pressure (CPAP) can have on reducing the prevalence and symptoms of insomnia in patients with obstructive sleep apnea (OSA).

Ultimately, a large decrease in insomnia symptoms and a significant reduction in prevalence of insomnia from baseline to follow-up were determined to be meaningful outcomes after CPAP treatment. The research suggests more broadly that CPAP has the potential to reduce insomnia in patients with OSA.

OSA affects at least 8% to 10% of the Norwegian population, and rates of insomnia are potentially even greater at 7% to 16%. The authors found a “clearly higher prevalence” of chronic insomnia (51.1%) among Norwegian patients with OSA compared with means of 8% to 20% in the general population.

All told, the sample used in data collection included 442 patients diagnosed with OSA (mean [SD] age, 54.9 [12.1] years; age range, 21-82 years; 74.4% men) who started treatment with CPAP at a university hospital between 2011 and 2018. Follow-up lasted between 6.0 and 51.9 weeks (mean, 23.1; median, 19.9 weeks), with the large time frame range explained as being due to patient appointment rescheduling.

For the study’s parameters, OSA was diagnosed according to standard respiratory polygraphy. Mean (SD) apnea-hypopnea index (AHI) was 30.1 (21.1) at baseline. Insomnia was assessed prior to CPAP treatment (baseline) and at follow-up with the Bergen Insomnia Scale (BIS). CPAP adherence was defined as an average use of at least 4 hours per night, whereas nonadherence was defined as use less than 4 hours per night.

Subsequently, mixed between-within subjects analyses of variance (group × time) were carried out by the researchers to determine the effect of CPAP adherence at follow-up (adherence vs nonadherence), insomnia diagnosis at baseline (chronic insomnia vs not chronic insomnia), and OSA severity at baseline (AHI ≥ 30 vs AHI < 30) on BIS scores (baseline vs follow-up).

Multiple linear regressions with difference in BIS from baseline to follow-up as a continuous dependent variable were then completed, with OSA severity, age, sex, and CPAP adherence as covariates. Then, a binary (crude) logistic regression analysis with CPAP adherence (0, adherent; 1, nonadherent) as the dependent variable and BIS scores at baseline as predictor was performed, followed by an adjusted logistic regression analysis.

The investigators observed a significant decrease in BIS scores from baseline (mean [SD], 18.8 [9.8]) to follow-up (12.9 [9.9]) (P < .001). The percentage of patients with chronic insomnia was significantly reduced from 51.1% at baseline to 33.0% at follow-up (p < 0.001).

In terms of CPAP adherence, there was a significant group × time interaction effect (P < .001), showing that the reduction in BIS scores was larger in the adherent group than in the nonadherent group. Later analyses of simple effects revealed a significant reduction in BIS scores for the adherent group with a large effect size (P < .001), whereas the nonadherent group had a small to moderate effect size, although it was also a significant reduction (P < .001).

Notably, 29 of the 216 patients who did not have chronic insomnia at baseline fulfilled the diagnostic criteria at follow-up. The authors noted this may be because problems with CPAP during use, such as mask usage, noise from the machine, and leakage, could have induced chronic insomnia.

The authors pointed to the relatively large sample size examined with a well validated insomnia questionnaire. The scale showed acceptable test-retest reliability and good convergent and discriminative validity among patients in a clinical setting. They thus believe their sample is representative of the overall OSA population in Norway.

There were also some possible limitations. The time frame in BIS changed from measuring the past 1 month to the past 3 months in 2018 due to new criteria in DSM-5, meaning 369 of 442 patients answered using the old criteria. The authors noted that prior research has showed that the risk of insomnia continuing after 1 year in patients who have had it longer than 1 month is 74%, and thus this change in the BIS should not affect the interpretation of their results.

The large range of time intervals between initial assessment and follow-up was cited as a limitation, as well as the lack of information on patients who completed follow-up within less than 1 month or more than 1 year after baseline. No significant interaction between follow-up time (<20 vs ≥20 weeks) and decrease in BIS scores was found.

Additionally, there was not enough statistical power to perform a separate analysis of the low percentage of patients with isolated sleep maintenance insomnia (3.4%) to compare them with those with other insomnia subtypes. Another limitation was that polygraph sleep testing can underestimate the total amount of obstructive respiratory events per hour compared with other measurements. A control group not receiving CPAP therapy was not included in the study.

The authors noted that other studies conducted in Nordic countries have showed differences in insomnia rates, potentially because of the use of alternative scales. To account for this, the authors say future research should try to develop a validated scale measuring insomnia symptoms and severity in patients with comorbid insomnia and sleep apnea that specifically tracks nonoverlapping symptoms as well as daytime impairments and nocturnal symptoms.

The authors concluded that the decrease from 51.1% at baseline to 33.0% at follow-up supports the efficacy of CPAP treatment in patients with OSA, and these findings should be noted by clinicians and researchers in the field for future decision-making.

Reference
Lundetræ RS, Saxvig IW, Aurlien H, Lehmann S, Bjorvatn B. Effect of continuous positive airway pressure on symptoms and prevalence of insomnia in patients with obstructive sleep apnea: a longitudinal study. Front Psychol. 2021;12:691495. doi:10.3389/fpsyg.2021.691495