Prevalence and frequency of mental health care provided by general practitioners: differences between 2 national data sources for the same population.

OBJECTIVE: Estimates from Canada's first national mental health surveillance initiative-which is based on diagnostic codes in administrative health care utilization databases-indicate that the proportion of Canadians who receive mental health care is more than twice as high as reported in Canada's national mental health survey. Our study examines and clarifies the nature and extent of differences between 2 predominant types of data that are used for mental health services research and planning. METHOD: A person-by-person data linkage was conducted between the Canadian Community Health Survey: Mental Health and Well-Being and administrative health care utilization records (British Columbia Ministry of Health Services-Medical Services Plan, and Hospital Discharge Abstract Database) within a universal-access, publically funded health care system, to examine the level of agreement between the data sources and respondent characteristics associated with agreement (N = 2378). RESULTS: The prevalence of mental health care from general practitioners (GPs) was higher in administrative data (19.3%; 95% CI 17.7% to 20.9%) than survey data (8.5%; 95% CI 7.5% to 9.8%). Agreement between prevalence estimates from the 2 data sources was associated with age, mental health characteristics, and the number of GP visits. The median number of visits per person was significantly higher in the survey data. CONCLUSIONS: GPs saw more than twice as many patients for mental health issues according to administrative data, compared with survey data; however, the number of visits per patient was higher in survey data.

Primary mental health care visits in self-reported data versus provincial administrative records.

BACKGROUND: Survey data and provincial administrative health data are the major sources of population estimates of mental health care visits to General Practitioners (GPs). Previous research has suggested that self-reported estimates of the number of mental health-related visits per person to health professionals may exceed estimates obtained from physician reimbursement records. DATA AND METHODS: Self-reported data from the 2002 Canadian Community Health Survey (CCHS): Mental Health and Well-being and administrative records from the Medical Services Plan of British Columbia were linked. The analytic sample consisted of 145 CCHS respondents who had at least one mental health visit to a GP in the past 12 months according to both data sources. High Reporters (self-reported visits exceeded number in administrative data), Low Reporters (self-reported visits were less than number in administrative data), and Exact Matches were analyzed in two ways. The first analysis used diagnostic codes to identify mental health-related visits in the administrative data. For the second analysis, all GP visits in the administrative data were counted as "possibly" mental health-related. Differences were described based on the median number of visits. RESULTS: When diagnostic codes were used to identify mental-health-related visitis in the administrative data, High Reporters (49%) substantially exceeded Low Reporters (24%). The remaining 27% were Exact Matches. Based on a broader definition of a mental health visit, 51% were Exact Matches. High reporting was common among people with mental disorders. INTERPRETATION: Self-reported data and administrative data provide different estimates of the number of mental health visits per person to GPs. The discrepancy can be large.