Background: Research on long-term pharmacotherapy for trauma-affected refugees is scarce. The purpose of this follow-up study of a randomised trial was to investigate the effects of sertraline compared to venlafaxine in combination with psychotherapy, 6 and 18 months after end of trial. Method: The primary outcome was PTSD symptoms, measured by the Harvard Trauma Questionnaire (HTQ). The secondary outcomes included: Hopkins Symptom Checklist-25 (HSCL-25), somatisation items of the Symptoms Checklist-90 (SCL), pain on a visual analogue scale, well-being on the WHO-5, Sheehan Disability Scale, Hamilton Depression and Anxiety scales and Global Assessment of Functioning. Moreover, the shorter version of the Recent Life Events (IRLE) was adopted to obtain information regarding the patients’ treatment and life events between the follow-up periods. Results: Out of 195 patients eligible for intention-to-treat analyses during trial, 116 participated in the 6-month follow-up and 97 participated in the 18-month follow-up. The results of our intention-to-treat analyses revealed no significant long-term differences between the groups on the primary outcome assessing PTSD symptoms (HTQ). For the secondary outcomes significant differences were found at the 18-month follow-up in favour of venlafaxine assessing symptoms of anxiety, depression and somatisation (HSCL-25 and SCL), although only in intention-to-treat and not per-protocol analyses. Conclusions: No conclusions could be drawn due to conflicting results between our intention-to-treat and per-protocol analyses.