Lin E, Dewa C, Kessler RC; Association for Health Services Research. Meeting.

Abstr Book Assoc Health Serv Res Meet. 1999; 16: 219-20.

Health Systems Research Unit, Centre for Addiction and Mental Health, Toronto, ON M5T 1R8, Canada.

RESEARCH OBJECTIVE: Mental health care has been characterized as being governed by demand-side constraints in the U.S. and by supply-side constraints in Canada. However, previous analyses of epidemiologic data from the U.S. and Ontario, the most populated province in Canada, indicate that demand-side controls as represented by insurance coverage only partially explain American patterns of mental health care use. This presentation expands these analyses to examine two further questions: 1) what are the relative effects of factors which influence help-seeking and those reflecting service supply on the frequency of mental health care visits? 2) does the pattern of these effects differ when mental health coverage is universal versus when it is not? STUDY DESIGN: Data are drawn from four sources. Two are psychiatric epidemiologic surveys -- the U.S. National Comorbidity Survey (NCS) and the Mental Health Supplement to the Ontario Health Survey (the Supplement), and two are county-level resource files specifically designed to augment the two population surveys. The NCS and the Supplement were both conducted in 1990/1991 and used nearly identical instruments to assess psychiatric disorder (the University of Michigan's modification of the World Health Organization's Composite International Diagnostic Interview -- UM-CIDI) and the use of mental health services. Other variables common to the two surveys included questions regarding self-perceived mental health and attitudes toward the use of mental health care. The county-level data for the NCS was derived from the 1990 U.S. Census Bureau's Area Resource File (ARF), and parallel data for Ontario was compiled from various government sources. Analyses focus on those respondents reporting past-year use of mental health care and are based on two sets of independent variables: sociodemographic and attitudinal variables which might influence help-seeking behavior and measures of physician and hospital bed supply within each U.S. and Ontario county. In addition, multivariate analyses control for mental health insurance coverage for the U.S. respondents. The descriptive (parametric and non-parametric) and multiple regression software used for analysis incorporate Taylor-series linearization to adjust for the complex sampling designs of the two population surveys. PRINCIPAL FINDINGS: Examination of the county-level supply variables reveals a much higher per capita rate of psychiatrists in the U.S. (nearly double Ontario's average rate) but fewer psychiatric hospital beds per capita. Multivariate analyses of the Ontario sample show that, once sociodemographic characteristics are controlled, factors related to help-seeking are still important predictors of intensity of service use. Particularly strong associations are found for measures of need and respondent perceptions of mental health services and of their own health. The presence of multiple mental disorders versus no disorder is associated with increased use of services. In addition, fear of potential embarrassment from discovery negatively impacts numbers of visits while comfortableness with discussing problems with a health professional is positively associated. Furthermore, those who rate their mental health as excellent to good used fewer services. When mental health visits to all sectors are considered, the supply of services has no significant effect. However, the picture changes when the focus turns specifically to medical visits. While need and perception remain important factors associated with intensity of use, psychiatrist supply also comes into play and is linked to increased service use. Additional multivariate analyses will examine the comparative impact of these factors in the U.S. market where type of insurance coverage also influences intensity of mental health care service use. CONCLUSIONS; While the frequency of mental health visits to the medical sector is related to physician supply under the universal coverage system practised in Ontario, help-seeking factors -- particularly psychiatric status, self-perceived mental health, and expectations of the mental health care system -- still assume a strong role. Because of different systems of care and cultural values, these factors may not have the same weight in the U.S. sample. IMPLICATIONS FOR POLICY, DELIVERY OR PRACTICE: Whether or not a universal coverage system would be successful in the United States has long been debated. Our findings show that both the supply of providers as well as need, individual perceptions, and attitudes are associated with the use of mental health care in Ontario. These results are consistent with the fact that universal health care in Canada is viewed as more than a mere payment mechanism but rather reflects the larger societal values -- where health care is viewed as every individual's right and prudent use of services as every individual's responsibility. (ABSTRACT TRUNCATED)

Publication Types:

• Meeting Abstracts

Keywords:

• Canada
• Comorbidity
• Ethnic Groups
• Health Services Needs and Demand
• Insurance Coverage
• Insurance, Psychiatric
• Mental Disorders
• Mental Health
• Mental Health Services
• Michigan
• Ontario
• United States
• Universal Coverage
• economics
• methods
• supply & distribution
• utilization
• hsrmtgs

Other ID:

• HTX/20602935

UI: 102194624

From Meeting Abstracts