Background. Patients whose symptoms are ' unexplained by disease ' often have a poor symptomatic outcome after specialist consultation, but we know little about which patient factors predict this. We therefore aimed to determine predictors of poor subjective outcome for new neurology out-patients with symptoms unexplained by disease 1 year after the initial consultation.Method. The Scottish Neurological Symptom Study was a 1-year prospective cohort study of patients referred to secondary care National Health Service neurology clinics in Scotland (UK). Patients were included if the neurologist rated their symptoms as ' not at all ' or only ' somewhat explained ' by organic disease. Patient-rated change in health was rated on a five-point Clinical Global Improvement (CGI) scale (' much better ' to ' much worse ') 1 year later.Results. The 12-month outcome data were available on 716 of 1144 patients (63 %). Poor outcome on the CGI (' unchanged ', ' worse ' or ' much worse ') was reported by 482 (67 %) out of 716 patients. The only strong independent baseline predictors were patients' beliefs [expectation of non-recovery (odds ratio [OR] 2.04, 95 % confidence interval [CI] 1.40-2.96), non-attribution of symptoms to psychological factors (OR 2.22, 95 % CI 1.51-3.26)] and the receipt of illness-related financial benefits (OR 2.30,. Together, these factors predicted 13 % of the variance in outcome.Conclusions. Of the patients, two-thirds had a poor outcome at 1 year. Illness beliefs and financial benefits are more useful in predicting poor outcome than the number of symptoms, disability and distress.