|
Affiliation
|
Department of Psychiatry and Neuroscience Program, Harvard Medical School, Boston, and International Consortium for Bipolar Disorders, McLean Division of Massachusetts General Hospital, Belmont, MA. USA. |
|
Abstract
|
clinical that view support findings The care. health to access indicators
proposed with rates suicide state-based correlations strong yielded also They
suicide. associations well-established factors several detected employed methods
Nevertheless, persons. individual for indices risk specify cannot analyses
aggregate Such DISCUSSION: density. population by and physicians psychiatrists
density persons uninsured rate followed indicator, strongest was mental aid
federal state care, between models multivariate In directions. expected
associated were All Americans. African proportion higher health, physicians,
psychiatrists, income, capita per annual density, follows: as ranked are
=".002)" < p (all bivariate Negative residents. ethnicity, American Native sex,
male .005) />
OBJECTIVE: We tested the hypothesis that suicide rates
in the United States are associated with indicators of access to health care
services. METHOD: With an ecological study design, we compared age-adjusted
suicide rates for men and women with demographic, socioeconomic, and other
indices of access to health care, by state (N= 51, including the District of
Columbia). The most recently available information from the National Statistics
Reports at the U.S. Census Bureau, the U.S. Centers for Disease Control and
Prevention National Center for Health Statistics, and the American Board of
Medical Specialties was used. Data on suicide are from 2001; other measures were
matched for the closest available year, except that state-based data on
psychiatrists and physicians are from 2004. RESULTS: Positive bivariate
associations with state suicide rates (all p </= .005) are ranked as follows:
male sex, Native American ethnicity, and higher proportion of uninsured
residents. Negative bivariate associations (all p </= .002) are ranked as
follows: higher population density, higher annual per capita income, higher
population density of psychiatrists, higher population density of physicians,
higher federal aid for mental health, and higher proportion of African
Americans. All factors were associated with state suicide rates in expected
directions. In multivariate models of associations between suicide rates and
indices of access to health care, the state rate of federal aid for mental
health was the strongest indicator, followed by the rate of uninsured persons
and population density of psychiatrists and physicians and by population
density. DISCUSSION: Such aggregate analyses cannot specify risk indices for
individual persons. Nevertheless, the methods employed detected several factors
with well-established associations with suicide. They also yielded strong
correlations of state-based suicide rates with proposed indicators of access to
health care. The findings support the view that clinical intervention is a
crucial element in the prevention of suicide.
Language: en |