OBJECTIVE To determine the effect of a novel brief general practitioner (GP)-led narrative exposure intervention on post-traumatic stress disorder (PTSD) symptoms after intensive care. DESIGN Multicentre, observer blind, randomised controlled trial (PICTURE). SETTING Primary care in 319 general practices across Germany. PARTICIPANTS 319 adults (18-85 years) who have survived critical illness with symptoms of PTSD, discharged from intensive care and randomised to receive the intervention (n=160) or improved usual care (n=159) from a general practitioner. INTERVENTIONS Intervention group participants had three narrative exposure consultations with a general practitioner and eight scheduled contacts with a nurse. Control group participants received improved treatment as usual based on the German PTSD guideline. MAIN OUTCOME MEASURES The primary clinical outcome was self-reported PTSD symptoms using the Post-Traumatic Diagnostic Scale for DSM-5 (PDS-5, range 0-80, higher scores indicating more severe symptoms) at six months. The minimal clinically important difference was six points. Secondary outcomes included changes in depression, anxiety, patient activation, health related quality of life and disability at six and 12 months. RESULTS Between 21 October 2018 and 18 January 2023, 1283 patients discharged from an intensive care unit were screened for PTSD symptoms. 319 study participants were randomly assigned either to the control group (n=159) or the intervention group (n=160). The mean patient age was 57.7 years (standard deviation (SD) 12.7), and 61% of participants were male. The mean baseline PDS-5 score was 30.6 (SD 13.3) in both groups. 271 (85%) study participants completed follow-up assessment after six months and 247 (77%) after 12 months. The intervention effect showed a mean between-group difference in the PDS-5 score of 4.7 points ((95% confidence interval 1.6 to 7.8); P=0.003, Cohen’s d=0.37)) at six months and 5.4 points ((1.8 to 9.0); P=0.003, Cohen’s d=0.41)) at 12 months. Among secondary outcomes, patients in the intervention group had greater improvements in depression, health related quality of life, and disability. CONCLUSIONS In adults with symptoms of PTSD after critical illness, a brief narrative exposure intervention was feasible and showed a reduction of symptoms, which was less than the predefined minimal clinically important difference. The effect was found to be sustained at 12 months’ follow-up. These findings support the further evaluation of this intervention in primary care.