The assessment of need for mental health services

BACKGROUND: The field of psychiatric epidemiology has yielded several large and important studies of the prevalence of psychiatric disorders. These surveys have been enhanced by the inclusion of methodologies that reflect the needs for care of the population in question. Clinical studies of psychiatric disorders and unmet needs have focussed on identifying needs and correlating them with service evaluation and satisfaction measures. The association between prevalence, service use and unmet need requires review in order to establish whether there are trends and consistent findings. METHOD: Peer-reviewed studies involving the assessment of need up to the present were included. These were of two broad groups: population-based studies and clinical studies. Studies based on outcome measures, review articles, and child psychiatry, old age, and intellectual disability samples were excluded. We conducted a search of MEDLINE and PSYCHINFO using the key words, "mental health needs","assessment of need", and "needs assessment". In addition, we hand-searched key journals and sought personal communication with researchers in the field. RESULTS: A total of 14 population studies and 19 clinical studies were retrieved. The percentage of the general population who reported at least one need for care ranged from 9.5% to 13.8%. The numbers of needs in psychiatric patients ranged from 3.3 to 8.6. Correlates of unmet need include: unemployment, single status, low quality of life and high disability scores, and the presence of certain psychiatric diagnoses, such as affective psychosis and personality disorders. Higher rates of met need and service satisfaction are correlated with a longer duration of service contact. CONCLUSION: The assessment of need to date has been established either by service use data to make inferences about unmet need, by asking limited guided questions about needs, or by using established needs assessment instruments. The high prevalence of psychiatric disorders and the weak correlation between disorder and unmet need require a combined approach towards service development.