This study aimed to identify the predictors of outcome in 102 patients following their first admission with acute chest pain. Outcome was measured at three months by interview and at five years by questionnaire. Chest pain, change in physical activity, return to work, smoking, psychiatric disorder, and mortality were assessed. The principal predictors of chest pain and smoking were previous psychiatric disorder and a diagnosis of non-specific chest pain; a previous history of psychiatric disorder was associated with a five-fold increase in the risk of continued chest pain at five-year follow-up (95% CI = 1.1-25.0). Psychiatric disorder at five years was predicted by psychiatric disorder at admission (adjusted odds ratio (adj OR) = 3.2; 95% CI 1.0-11.0) and non-specific chest pain (adj OR = 7.5; 95% CI = 1.7-32.1). Mortality at five-year follow-up was independently associated with older age (adj OR = 1.1; 95% CI = 1.01-1.2), an elevated Norris score (adj OR = 1.41; 95% CI = 1.01-1.96) and a previous history of psychiatric disorder (adj OR = 5.06; 95% CI = 1.13-22.0). These findings suggest that prediction of outcome, irrespective of underlying diagnosis, requires careful assessment of previous or current psychiatric symptoms in patients admitted with chest pain. Early intervention with psychological treatment for patients with non-specific chest pain should be considered; this may also involve help to reduce smoking. The study provides further evidence that mortality following myocardial infarction is closely linked to psychiatric disorder, but suggests that prior psychiatric disorder may be more important than 'post-infarction' depression. A larger study is needed to confirm these results.