The current study examines the unique and interactive effects of posttraumatic stress disorder (PTSD) symptoms and days using alcohol, opioids, and marijuana on PTSD symptoms, suicidal ideation, and suicidal behavior up to 1 year, later in a high-risk sample of military personnel not active in mental health treatment. Current and former military personnel at risk for suicide (N = 545; M age = 31.91 years, standard deviation = 7.27; 88.2% male) completed self-report measures of PTSD symptoms, past 30 days heavy alcohol use, opioid use, marijuana use, and current suicidal ideation via telephone at baseline and 1, 3, 6, and 12 months later. PTSD symptoms and the substance use variables (and relevant covariates) were entered as predictors of changes in PTSD symptoms, the likelihood of suicidal ideation, suicidal ideation severity, and the likelihood of suicidal behavior during the 11-month follow-up period. PTSD symptoms predicted PTSD symptoms 1 month later. PTSD symptoms and marijuana use predicted the likelihood of suicidal ideation 1 month later and suicidal behavior during the 11-month follow-up period. The interaction between PTSD symptoms and marijuana use significantly predicted increased PTSD symptoms over time and suicidal behavior. At high, but not low levels of PTSD symptoms, more days using marijuana predicted increased PTSD symptoms over time and the likelihood of suicidal behavior. Results suggest marijuana, especially for military personnel experiencing elevated PTSD symptoms may negatively impact suicidal thoughts and behavior. These results are relevant to suggestions that medical marijuana could be used in treating or augmenting treatment for PTSD. © 2019 Wiley Periodicals, Inc.