Persons living with HIV (PLWH) are significantly more likely to die by suicide compared to the general population. This is the first study to examine the impact of posttraumatic stress disorder (PTSD), major depressive disorder (MDD), insomnia, and substance use disorders on suicidal ideation/behavior among PLWH using electronic medical record (EMR) data. We also evaluated the mutual influence of interactions between PTSD and substance use disorders on suicide risk, consistent with the substance abuse, violence exposure, and HIV/AIDS “SAVA” syndemic model. Participants (n = 2336) were HIV+ individuals recruited through the Center for AIDS Research (CFAR). Participants provided informed consent for extraction of their EMR. As hypothesized, univariate analyses revealed that PTSD, MDD, insomnia, alcohol and other substance use disorders (cocaine abuse and cocaine dependence, opioid abuse and dependence, cannabis abuse, other psychoactive substance abuse and dependence, and polysubstance use disorder) were each positively associated with suicidal ideation/behavior. Also as hypothesized, a multivariable analysis found that alcohol and cocaine dependence, MDD, and PTSD were significant predictors of suicidal ideation/behavior. Contrary to hypotheses, none of the interactions between PTSD and substance use disorders were significantly associated with suicidal ideation/behavior.

Prior studies of prolonged exposure therapy (PE) suggested that reduction of posttraumatic stress disorder (PTSD) precedes reduction in depression, yet no research has collapsed data across multiple studies to examine whether the directionality of reduction remains consistent in larger and diagnostically diverse samples. Thus, the objective of this study is to conduct an evaluation of bidirectional associations between PTSD and depression in PE. Participants (n = 216) from three randomized controlled trials of PE alone, PE + alcohol use disorder treatment, and PE + nicotine use disorder treatment completed weekly PTSD and depression severity measures. First, we analyzed the directional relationship between PTSD and depression over time in 2 single models to separately examine the effects of PTSD on depression and vice versa. Second, we analyzed a combined model to examine the simultaneous effects of reduction in PTSD on reduction in depression over and above the effects of reduction in depression on reduction in PTSD, and vice versa. Two single models suggested that reductions in PTSD lead to reductions in depression and vice versa. The combined models suggested that both directions of change are important and reciprocal. The strength of predictive power from PTSD to depression, and vice versa, is approximately equal. Most significant prediction of PTSD from depression and vice versa occurred early in treatment. The relationship between reductions in PTSD and depression during PE is transactional. Regardless of whether PTSD or depression decreases first, reduction in the other symptom cluster is likely to follow. (PsycINFO Database Record