Medical Resources and Suicide Prevention

Abstract

Doerner and Speir (1986) argued that the regional variation in homicide rates might be a result in part of the differential distribution of medical resources, and they demonstrated that the availability of hospital beds, physicians, and nurses in Florida counties was related to the homicide rate. 'Ilus suggests the hypothesis that the regional variation in suicide rates might also be related to the differential distribution of medical resources. To test this hypothesis, data from 1980 for the number of beds per capita and the number of physicians per capita in each of the 48 continental states were obtained from the Census Bureau (1982-1983), and rates of suicide and homicide were obtained from the National Center for Health Statistics (1983). The suicide rate was negatively related to the number of beds per capita (Pearson r = -0.27, p= ,031 but not signiEicantly to the number of physicians per capita (r=-0.21, p = .07). The homicide rate was positively related to the number of beds per capita (r = 0.40, p = .003) but not to the number of physicians per capita (r = -0.06). To explore this further, suicide and homicide rates from 15 Western European nations' for 1980 obtained from the World Health Organization (WHO, annual) were correlated with each of the three measures used by Doerner and Speir-population per physician, per nurse, and per hos~itd bed obtained from the World Bank (1983). . . Suicide rates were negatively but not significantly correlated with the population per physician, nurse, and bed (Pearson rs = -0.42, -0.44, and -0.40, respectively). The higher the suicide rate, the lower the population per medical resource, that is, the better the medical resources. This is contrary to the expectation. The homicide rates were also not significantly correlated with anv of the three medical resource measures Irs = -0.19. -0.22 and -0.28). The present analyses then give Little support for the Doerner and Speir hypothesis using their measures of medical resources except for the measure of beds per capita in the states of America. However, they suggested a better measure of lethality from homicide-homicides divided by homicides and assaults. A better measure of mortality from suicide would be completed suicides divided by completed plus attempted suicides. Unfortunately, counts of attempted suicides are not available for American states or nations at present.