0390 Associations of Mental Health and Pain with Actigraphically-Assessed Sleep: A Prospective Study of Operational Firefighters

Abstract Introduction Prior firefighter studies have rarely assessed actigraphic sleep beyond 14 days or evaluated mental health’s relationship to objective sleep, the aims of this prospective study. Methods Thirty-two participating firefighters were assessed via actigraphy (Philips Spectrum Plus) for 21 days and follow-up questionnaires including: Generalized Anxiety-7 questionnaire-assessed anxiety(categorized as mild[5-9], moderate[10-14], severe[≥15]), Patient Health Questionnaire-9-assessed depression(categorized as mild[5-9], moderate[10-14], severe[≥15]), PTSD-5-assessed PTSD(≥4-presence versus 0-3-none), pain(presence versus absence), and work exposures for their 21-day cycle. Actigraphic measures were assessed continuously and categorically--sleep latency(SL)< 8min, total sleep time(TST)≤180.5min(≤75thpercentile), wake after sleep onset(WASO>31.5min[>75thpercentile]), and sleep efficiency(SE)≤80%. Unadjusted and adjusted generalized estimating equations linear regression models and robust Poisson regressions estimated associations of mental health and pain with mean and categorized sleep, respectively, summarized for 657 person-days adjusting for age, poor sleep quality, comorbidity, sleep apnea, chronotype, occupational measures. Results Mean actigraphic sleep measures over 657 person-days were: 11min(SD=29.9), 284min(SD=128.4) TST, 19.5(SD=18.7) WASO, 84.7%(SD=13.2) SE. After adjustments, severe anxiety (versus mild) was significantly associated with increased shorter mean SL(β=-11.6,p=0.0245); moderate and severe depression (versus mild), with longer mean SL(β=24.0,p=0.0245 and β=32.4,p=0.0011, respectively) and TST(β=77.6,p< 0.0001 and β=66.0,p=0.0144, respectively); pain(versus absence), with shorter mean SL (β= -15.5,p< 0.0001) and TST(β=-37.3,p=0.0080). In adjusted Poisson models, moderate (versus mild) depression lowered SL< 8min risk(versus ≥8)(RR=0.61,p=0.02); moderate (RR=1.71,p< 0.0001) and severe (RR=1.89,p=0.0002) anxiety(versus mild) significantly increased TST≤180.5min risks(versus>180.5), while moderate (RR=0.68,p=0.0027) and severe (RR=0.50,p=0.0225) depression(versus mild) significantly lowered short TST≤180.5min risks (versus>180.5); severe(versus mild) anxiety (RR=0.41,p=0.0354) significantly lowered SE≤80% risks(versus >80%); moderate (RR=8.36,p=0.0005) and severe (RR=24.89,p=0.0005) depression(versus mild) elevated SE≤80% risks(versus>80%). Conclusion After adjustments, pain increased poorer sleep health risks (exempting SE). Moderate and severe depression and PTSD decreased shorter TST risks. In contrast, moderate and severe anxiety increased shorter TST risks, suggesting mental health conditions may pose differential risks for poor sleep in operational firefighters. Support (if any) This research was supported by grants from the U.S. Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health to the Johns Hopkins Education and Research Center for Occupational Safety and Health (award No. T42 OH0008428) and the Johns Hopkins P.O.E. Total Worker Health Center in Mental Health (POE Center, award No. U19OH012297).