The self-medication hypothesis proposes that individuals experiencing posttraumatic stress disorder (PTSD) symptoms may use alcohol (or other substances) to self-medicate distress, while the mutual maintenance hypothesis proposes that alcohol use also exacerbates or maintains distress. Existing research largely supports the self-medication hypothesis. However, findings are often based upon retrospective reporting of problems and assessment of symptoms anchored to a trauma occurring in the remote past. To improve on these retrospective designs, the current study examined the relationship between daily PTSD symptom severity and alcohol-related outcomes during the early phase of recovery following a traumatic physical injury. Specifically, 36 injury victims reported on PTSD symptom severity, alcohol craving, alcohol consumption, and negative drinking consequences thrice daily (morning, afternoon, night) over 7 days beginning 6 weeks postinjury. Results indicated relationships between PTSD symptom severity and alcohol craving/negative consequences when these experiences were assessed concurrently within the same signal (particularly during the nighttime assessments). Prospective models found that nighttime PTSD symptom severity was related to negative drinking consequences occurring the following morning, even after controlling for consumption level, suggesting a more general behavioral regulation problem. Results were less supportive of the mutual maintenance hypothesis. Collectively, these results suggest that health interventions targeting PTSD symptom severity in recent injury victims—particularly when delivered at night—may be able to prevent problematic alcohol use.