Graduation Year


Document Type




Degree Name

Doctor of Philosophy (Ph.D.)

Degree Granting Department

Public Health

Major Professor

Skai W. Schwartz, Ph.D.

Committee Member

Amy C. Alman, Ph.D.

Committee Member

Getachew A. Dagne, Ph.D.

Committee Member

Adam P. Goode, D.P.T., Ph.D.


Nightmares, Obstructive Sleep Apnea, Positive Airway Pressure, Post-Traumatic Stress Disorder


Low back pain (LBP) is a common problem that is the primary contributor to years lived with disability worldwide. In the U.S. LBP is disproportionately experienced by military veterans. In recent years interest has grown in the impact of sleep and sleep-related diagnoses on musculoskeletal pain complaints (including LBP) and despite growth in the scientific literature investigating this, gaps in our understanding of how sleep diagnoses impact LBP remain. To that end, this dissertation uses administrative electronic health record data from veterans seeking care through the Veterans Health Administration in three manuscripts to address the following specific aims: (1) investigate obstructive sleep apnea (OSA) as a mediator of the effect of post-traumatic stress disorder and incident LBP, (2) investigate the effect of positive airway pressure treatment adherence on LBP among those with OSA, and (3) investigate the association between nightmare disorder diagnosis and LBP. All three manuscripts utilize survival analysis in some form to achieve these specific aims. Results from the first manuscript (addressing specific aim 1) indicate that OSA is both a partial mediator and an effect modifier of the effect of PTSD on incident LBP. However, most of the effect of PTSD on LBP is due to factors other than mediation or interaction with OSA. The second manuscript (addressing specific aim 2) showed that individuals who were adherent to PAP had a longer LBP-free period than those who were non-adherent. This effect appears to be stronger among those with prior LBP within the 12 months before PAP initiation. The third manuscript (addressing specific aim 3) found hazard ratios indicated a statistically significant effect of nightmare disorder on time-to-LBP that was time-dependent, becoming smaller and non-significant at ≥12 months. Each of these manuscripts adds to the limited literature to-date on how sleep diagnoses impact LBP.

Included in

Epidemiology Commons