Population-Based Study Finds Higher Polysomnography Use but Similar PAP Initiation in Patients With Parkinson Disease

Ryan J. Gotfrit, MD, FRCPC

(Credit: Hawkesbury General Hospital)

Using health administrative data, a recent population-based study published in Neurology Clinical Practice revealed that patients with Parkinson disease (PD) had a slightly higher overall rate of undergoing polysomnography (PSG) compared with a nonPD population but showed no difference in positive airway pressure (PAP) initiation over a 10-year period. These findings suggest that sleep-medicine resource use in the PD population was similar to that of matched controls and followed the same upward trend over time observed in the general population.¹

The study analyzed data from 65,167 patients with PD and 11,460,672 controls who met the study’s inclusion criteria. Researchers reported a higher prevalence of PSG performed in the PD group compared with controls (8.2% vs 6.3%; RR, 1.30; 95% CI, 1.25–1.35; P <.001), regardless of sex or age. The same trend appeared for diagnostic PSGs (7.0% vs 5.0%; RR, 1.40; 95% CI, 1.34–1.46; P <.001). However, investigators observed no difference in the prevalence of therapeutic PSGs between groups, except among men, where rates were slightly higher in the PD group (4.2% vs 3.9%; RR, 1.08; 95% CI, 1.01–1.15; P = .04).

“The higher rate of PSGs performed in the PD group may be driven by the higher risk of REM sleep behavior disorder (RBD) or the combination of other sleep disorders that are frequently seen in the PD population, including sleep-maintenance insomnia, excessive daytime sleepiness, and circadian rhythm disorders. This is supported by the higher crude prevalence of diagnostic PSGs performed in the PD group, but with no difference in crude prevalence of therapeutic PSGs performed between the groups,” lead author Ryan J. Gotfrit, MD, FRCPC, neurologist at The Ottawa Hospital in Canada, and colleagues wrote.¹Previous studies suggest that the PD population may have a higher prevalence of obstructive sleep apnea (OSA)^2,3, which could explain the increased rate of PSGs performed, according to the authors.

This retrospective, longitudinal, population-based study analyzed health administrative databases from Ontario, Canada, between 2012 and 2021. Researchers compared overall and annual prevalence rates of PSGs performed, and PAP initiated in adult patients with PD to 1:1 propensity score–matched controls from the nonPD population. Authors matched controls using exact criteria for age, sex, and calendar year, along with caliper-matched propensity scores derived from a logistic regression model and validated health administrative definitions to identify PD cases and controls. Investigators hypothesized that patients with PD would have lower rates of PSG performance and PAP initiation than matched controls.

READ MORE: Several Preexisting Therapies Associated With Reduced Mortality Risk in Parkinson Disease, Study Shows

Findings showed no differences in the crude prevalence of any PAP initiation, except among individuals aged 20–49 years, where the PD group had a higher prevalence than the controls (7.3% vs 5.9%; RR, 1.24; 95% CI, 1.00–1.53; P = .04). Authors also observed no difference in CPAP initiation. Similarly, there was no overall difference in APAP initiation, except among women, where controls had a higher prevalence than the PD group (0.2% vs 0.1%; RR, 0.50; 95% CI, 0.33–0.77; P = .02).

All told, the prevalence of BiPAP spontaneous mode (BiPAP-S) initiation did not differ overall, but among women, patients with PD had a higher prevalence compared with the controls (0.3% vs 0.2%; RR, 1.50; 95% CI, 1.03–2.19; P = .01). Results also displayed that BiPAP spontaneous-timed mode (BiPAP-ST) initiation was higher in the PD group overall (0.2% vs 0.1%; RR, 2.00; 95% CI, 1.49–2.69; P = 0.02), particularly among women (0.2% vs 0.1%; RR, 2.00; 95% CI, 1.13–3.56; P <.001) and patients aged 50–64 years (0.3% vs 0.1%; RR, 3.00; 95% CI, 1.62–5.57; P = .03).

For both groups, the annual prevalence rates generally increased over time, whereas the annual incidence rates remained stable. Notably, researchers observed no change in these trends when stratifying by sex or age. The study reported no difference in the annual prevalence rate ratio of PSG performed between groups, regardless of sex or age, except among individuals older than 65 years. In this subgroup, controls had a higher annual prevalence rate ratio than patients with PD for diagnostic PSG (1.07; 95% CI, 1.06–1.08] vs 1.06; 95% CI, 1.05–1.06; P = .02) and therapeutic PSG (1.07; 95% CI, 1.06–1.08 vs 1.05; 95% CI, 1.04–1.06; P = .03).

Similarly, researchers noted no overall difference in the annual prevalence rate ratio of any PAP initiation between groups, except among patients older than 65 years, where controls had higher rates (1.11; 95% CI, 1.10–1.12 vs 1.09; 95% CI, 1.08–1.10; P = .003). Among this age group, the annual prevalence rate ratio of CPAP initiation was also higher in controls (1.15; 95% CI, 1.14–1.16 vs 1.10; 95% CI, 1.09–1.11; P <.0001), with no significant differences by sex. Authors also observed no significant differences in the annual prevalence rate ratios for APAP, BiPAP-S, or BiPAP-ST between groups, regardless of sex or age.

“Despite having a higher prevalence of any PSG performed, the prevalence of any PAP initiated in the PD population was similar to the control group. This may have been secondary to the higher prevalence of mood disorders in the PD group after matching, leading to a relatively lower use of PAP in this group, for example due to anxiety or apathy. Financial barriers could also play a role in reduced PAP use because PAP treatments are 70% covered by OHIP, while PSGs are 100% covered. If the prevalence of OSA is indeed higher in the PD population, then our results may reflect that OSA is undertreated in the PD population,” Gotfrit et al noted.¹

REFERENCES
1. Gotfrit R, Talarico R, Gros P, Kaminska M, Mestre TA, Kendzerska T. Sleep Medicine Resource Utilization in Individuals With Parkinson Disease: A Population Study of Health Administrative Data. Neurol Clin Pract. 2025;15(4):e200511. doi:10.1212/CPJ.0000000000200511
2. Gros P, Videnovic A. Overview of Sleep and Circadian Rhythm Disorders in Parkinson Disease. Clin Geriatr Med. 2020;36(1):119-130. doi:10.1016/j.cger.2019.09.005
3. Bollu PC, Sahota P. Sleep and Parkinson Disease. Mo Med. 2017;114(5):381-386.