Low physical activity is a plausible risk factor for obstructive sleep apnea (OSA), but whether exercise can prevent OSA incidence has mainly been studied cross-sectionally.

Using longitudinal data from three large cohorts, Yue Liu, MD, a research fellow in the Channing Division of Network Medicine at Brigham and Women’s Hospital, Tianyi Huang, ScD, an associate epidemiologist in the Department of Medicine, and colleagues recently conducted the largest prospective study of this question to date.

In the European Respiratory Journal, they link physical inactivity and sedentary behavior with OSA incidence, suggesting an active lifestyle may have a role in prevention.

Methods

The data included in this analysis came from participants in the following studies who did not have OSA diagnosed at baseline:

• 50,332 in the Nurses’ Health Study (women)
• 68,265 in the Nurses’ Health Study II (women)
• 19,320 in the Health Professionals Follow-up Study (men)

Throughout follow-up, participants reported their average time spent per week on walking, jogging, running, bicycling (including stationary machine), lap swimming, tennis, squash/racket ball, calisthenics/rowing, outdoor work and weightlifting.

They were also regularly asked about their usual walking pace, the number of flights of stairs climbed per day, and hours per day spent watching TV, sitting at work and sitting away from home.

The researchers assigned metabolic equivalent of task (MET) scores to each type of physical activity and calculated MET-hours/week of total activity. They pooled the results across the three cohorts.

Physical Activity

Higher levels of physical activity were associated with a lower incidence of OSA after adjustment for potential confounders. Compared with <6.0 MET-hours/week of physical activity:

• 6.0–11.9 MET-hours/week—HR for OSA, 0.78
• 12.0–20.9 MET-hours/week—HR, 0.67
• 21.0–35.9 MET-hours/week—HR, 0.57
• ≥36 MET-hours/week—HR, 0.46 (P for trend < 0.001)

The association between physical activity and OSA was moderately weakened after further adjustment for diabetes, hypertension, BMI and waist circumference, but it remained statistically significant.

Both moderate and vigorous physical activity were strongly inversely associated with OSA incidence.

Physical activity was more protective for women than men, adults <65 years old than older adults and individuals with BMI ≥25 kg/m² than those with BMI <25 kg/m².

Sedentary Behavior

Conversely, sedentary behavior was associated with a higher incidence of OSA incidence. Compared with <4.0 hours/week sitting watching TV:

• 4.0–6.9 hours/week—HR, 1.10
• 7.0–13.9 hours/week—HR, 1.25
• 14.0–27.9 hours/week—HR, 1.38
• ≥28.0 hours/week—HR, 1.78 (P for trend < 0.001)

The corresponding HRs for hours sitting at work or away from home were 1.02, 1.13, 1.27 and 1.49 (P for trend < 0.001).

After adjustment for the metabolic variables, the associations between OSA and sitting at work/away from home were weaker but statistically significant, whereas TV viewing time was no longer associated with OSA.

Commentary

This study focused exclusively on recreational exercise. Research that measures physical activity objectively (e.g., with wearable devices) could capture other types of activity (e.g., occupational), as well as more types of sedentary behavior. Future studies also need to involve populations that are more diverse in race/ethnicity, profession and socioeconomic status.