Mental health need: use of administrative data and record linkage to inform mental health policy and practice

Abstract

BackgroundWith mental ill-health on the rise globally, it is crucial to investigate whether the needs of individuals with mental ill-health are fully addressed. Attempts to measure negative consequences of unmet needs have been limited by the use of cross-sectional study designs or self-report measures. We aimed to investigate the interplay between perceived mental ill-health and unmet need in relation to mental health on a population level. MethodsA record linkage methodology was implemented drawing information from the 2011 Northern Ireland Census returns and a population-wide prescribing database (n=286 717). Chronic mental ill-health was assessed through the Census self-reported mental health question (presence of an emotional, psychological, or mental health condition that has lasted or is expected to last at least 12 months) and compared with regular psychotropic medication use (monthly dosage of antidepressant, anxiolytic, antipsychotic, or antimania medication) in the 6 and 12 months after the Census. Logistic regression models adjusted for demography (age, sex, ethnicity, marital status, educational attainment, occupational social class), household (tenure, car availability), and area variables (urbanicity, deprivation). FindingsOverall, 23 803 individuals (8%) aged 25–74 years reported a chronic mental health condition, with low rates among ethnic minorities (129 [3%] of 3897 non-White individuals in receipt of medication). Of the individuals with self-reported mental ill-health, 5246 (22%) did not use psychotropic medication over the following 6 months, and 4412 (19%) did not use them by 12 months. Lower uptake was noted among men (odds ratio 0·56, 95% CI 0·52–0·60), non-white ethnic minorities (0·38, 0·26–0·54), and individuals separated, divorced, or widowed (0·75, 0·68–0·82) or unemployed (0·65, 0·53–0·81). InterpretationDiscrepancies between population mental ill-health and uptake of pharmacological treatment were more evident among men, ethnic minorities, and the economically disadvantaged. This study indicates that administrative data linkages can provide a valuable resource to define population characteristics, and inform policy and practice. However, the findings might be limited by availability of data on psychosocial and non-pharmacological interventions, use of proxy measures of mental health treatment, and the self-reported nature of the Census. Further research should explore whether this variation is due to stigma or lack of understanding or knowledge of available health-care services. FundingNone.