Prescription opioid use and physical inactivity: independent associations and no multiplicative interaction with...

Sleep disorders affect roughly one-third of adults worldwide. Both prescription opioid use and physical activity (PA) may influence sleep health.

We pooled four cycles of NHANES (2007-2014). Among 17,490 adults aged ≥20 years with complete data on prescription opioid use, PA, sleep disorders, and covariates, survey-weighted (complex-design) logistic regression estimated odds ratios (ORs) and 95 % confidence intervals (CIs) for sleep disorders by opioid use and PA status (inactive <150 min/week vs active ≥150 min/week). Models were progressively adjusted for sociodemographics, lifestyle factors, BMI, and chronic conditions; the primary model (Model 3) is reported, with an extended sensitivity model (Model 4) additionally adjusting for depression (PHQ-9) and sedative/hypnotic use. Restricted cubic splines (RCS) examined dose-response patterns for total and domain-specific PA (occupation, transportation, leisure-time). Subgroup and sensitivity analyses were performed.

Opioid use was associated with higher odds of self-reported clinician-diagnosed sleep disorders (adjusted OR 1.82; 95 % CI 1.52-2.20; P < 0.001). PA was also associated with higher odds (adjusted OR 1.33; 95 % CI 1.12-1.59; P = 0.002). Relative to adults with no opioid use and active PA, adjusted odds ratios were 2.21 (95 % CI 1.71-2.85) for opioid use only, 1.37 (95 % CI 1.14-1.64) for inactivity only, and 2.04 (95 % CI 1.52-2.74) for both exposures; the opioid*PA multiplicative interaction on the odds scale was not significant (P = 0.141). RCS suggested a monotonic inverse association of PA with sleep-disorder odds, with risk reductions most apparent up to around 260-300 min/week; formal tests of nonlinearity were largely non-significant. In extended adjustment including PHQ-9 depression and sedative/hypnotic use, estimates were attenuated but remained directionally consistent.

Prescription opioid use and PA are independently associated with higher odds of self-reported clinician-diagnosed sleep disorders. Encouraging achievable amounts of transportation- and leisure-time PA may benefit sleep health, including among adults using prescription opioids. Results were directionally consistent when short sleep (<7 h/night) was used as a current sleep indicator. Longitudinal studies are needed to clarify temporality and dose-duration effects.