Associations Between Lifetime Panic Attacks, Posttraumatic Stress Disorder, and Substance Use Disorders in a Nationally Representative Sample.

Objective: Rates of lifetime substance use disorder (SUD) are high among people with lifetime posttraumatic stress disorder (PTSD). Panic attacks are also prevalent among trauma survivors and people with SUD, yet studies on PTSD/SUD have rarely examined comorbid panic. This potentially creates additional barriers to effective treatment for people with PTSD/SUD, in that panic may be under-diagnosed among people with PTSD/SUD and consequently attenuate treatment outcome. Additionally, research on PTSD/SUD often combines people with alcohol use disorder (AUD) and people with drug use disorders (DUDs) into a single group despite evidence that these two PTSD/SUD subgroups differ along important sociodemographic and clinical variables. This study tested the hypothesis that among adults with lifetime PTSD, panic attacks would be associated with greater lifetime risk for both AUD and DUD. We also explored whether panic attacks were associated with specific DUDs that frequently co-occur with PTSD (cannabis, sedatives/tranquilizers, heroin/opioids, and cocaine). Methods: Data were drawn from the National Epidemiologic Survey on Alcohol and Related Conditions-III (NESARC-III), a cross-sectional national study. Adults with lifetime PTSD (N = 2,230) were classified into one of three groups based on diagnostic interview data: adults with PTSD/AUD (i.e., met criteria for PTSD and AUD but not DUD; n = 656), adults with PTSD/DUD (i.e., met criteria for PTSD and DUD, regardless of AUD diagnostic status; n = 643), or adults with PTSD-only (i.e., met criteria for PTSD but not AUD or DUD; n = 1,031). Results: Weighted logistic regression analyses showed that lifetime risk of PTSD/AUD and PTSD/DUD, each relative to PTSD-only, was greater for adults who were younger at the time of data collection, were male, and had a history of panic attacks. Panic attacks did not predict specific DUD diagnoses comorbid with PTSD in exploratory analyses adjusting for sociodemographic and clinical covariates. Conclusions: Findings highlight the importance of assessing and targeting panic in PTSD/SUD clinics, but suggest panic may not discriminate between specific DUDs that commonly co-occur with PTSD. Study limitations and future directions are discussed.