From an original sample of 2454 participants free of self-reported psychological distress, 1463 workers completed a 15-month follow-up. Baseline measures included exposure to job demands, decision latitude, social support and need for recovery. Psychological distress was assessed using the General Health Questionnaire at baseline and at follow-up. The findings showed that medium and high exposure to job demands and social support increased the risk of reporting psychological distress at 15-months (relative risk (RR) = 1.65, 1.45). The highest adjusted RR was observed for workers reporting a high need for recovery after work (RR 2.12, 1.90) and this finding was independent of the effects of job demands, decision latitude and social support. Neither decision latitude, nor low back problems increased the risk of reporting future psychological distress, although neck problems (RR = 1.66) and hand/wrist problems (RR = 1.45) did. It was concluded that need for recovery appears to be an important indicator of individual workers who are at risk of developing psychological distress long term. STATEMENT OF RELEVANCE: This paper reports the findings of a longitudinal study showing that need for recovery from work was the strongest predictor, relative to psychosocial work characteristics (job demands, decision latitude and social support), and musculoskeletal problems, of psychological distress 15 months later in individuals initially free from distress.