Incidence and outcomes of mental disorders in a regional population: the Northern Rivers Mental Health Study

OBJECTIVES: To estimate the incidence of mental disorders in a cohort of previously symptom-free individuals who are representatives of a regional Australian population. To map changing patterns of diagnosis and comorbidity within the cohort over a 2 year period. METHOD: Two year follow-up of a community-based cohort drawn from a telephone screening of 9191 randomly selected adults. Subjects were administered a comprehensive face-to-face interview which included the Composite International Diagnostic Interview. A total of 1407 subjects were interviewed at baseline, and 968 subjects were reinterviewed (a 68.8% follow-up rate). RESULTS: There was considerable change in disorder status over the study period, and analysis of the Composite International Diagnostic Interview scoring suggests that these changes reflected real changes in symptomatology. Of subjects interviewed at both baseline and follow-up, 638 were classified as disorder-free at their entry to the study. After 2 years, 98 of these met criteria for a mental disorder during the preceding 12 months. After adjusting for sampling and gender, the 12 month incidence of any mental disorder among subjects who had been disorder-free 2 years previously was 9.95 per hundred person-years at risk. At baseline, a further 330 subjects met ICD-10 criteria for a mental disorder during the previous 12 months. Two years later, 167 of these subjects (50.6%) were disorder-free, and 163 still met the criteria for a mental disorder, although there had often been considerable change in their diagnosis. Subjects with a mental disorder at the commencement of the study were significantly more likely than those without a disorder to have a positive diagnosis 2 years later (p < 0.001). The number of diagnoses at baseline was a strong predictor of the number of diagnoses at follow-up (p < 0.001), and each additional comorbid diagnosis at baseline also increased the probability of a persisting disorder at follow-up (p < 0.001). CONCLUSIONS: Over a 2 year period, the majority of subjects with a mental disorder will become disorder-free, while a significant number of previously disorder-free individuals will develop a positive diagnosis. Health services need to be designed to meet this labile demand.