Determinants of youth suicide: the Easterlin-Holinger cohort hypothesis re-examined

American Journal of Economics and Sociology, The,  April, 1998  by Donald G. Freeman

• E-mail
• Print
• Link

I

Introduction

Youth Suicide is one of the nation's most pressing social problems. In 1993, 1,884 adolescents between the ages of 15 and 19 took their own lives, making suicide the third leading cause of death for this age group, behind only accidents and homicides. The rate at which young people kill themselves has risen unabated for three decades, from 3.3 per 100,000 population in 1959 to 10.9 in 1993, an increase of over 200 percent. From 1973 to 1993, the very young (less than 20 years) and the very old (greater than 75 years) were the only groups recording higher suicide rates; moreover, rates for the elderly have risen only slightly (less than 10%) over the past two decades, while rates for ages 10-14 and 15-19 have risen 240 and 59 percent, respectively.(1)

In 1980, the U.S. Department of Health and Human Services (DHHS) established a specific health objective:

By 1990, the rate of suicide among people 15-24 years of age

should be below 11 per 100,000 (compared with 12.4 per 100,000

in 978). [DHHS, 1980]

Despite ongoing efforts toward prevention and treatment for suicidal behaviors, including the DHHS's own 1989 Task Force Report comprising four volumes, the actual suicide rate in 1990 for the 15-24 age group was 13.2 suicides per 100,000, an increase of 6 percent over the 1978 level and 20 percent over the target. Even more tragic in that statistic is that a decline in suicide rates for the older subgroup 20-24 years from 16.9 in 1978 to 15.1 in 1990 was more than offset by the increase in the suicide rate from 8.0 to 11.1 for the younger subgroup of 15-19 year-olds. Clearly, the risk factors of this increasingly vulnerable subgroup warrant greater attention.

The phenomenon of rising youth suicide rates has inspired a considerable body of literature searching for the underlying causes of self-inflicted death at what should be a promising time in an individual's life. This literature focuses on three broad factors in youth suicide: 1) Personal characteristics, such as mental illness, drug use, school failure, sexual identity and personality; 2) family characteristics, such as family history of suicidal behavior, family violence or abuse, parental divorce or separation, and lack of family support; and 3) souci-economic factors, such as economic cycles, demographic pressures, poverty and poor social environment.(2)

Plan of the Paper

The role of personal characteristics in any individual suicide is undeniably critical, given that different individuals in similar environments behave differently. An individual attempts to satisfy "needs" (expectations or aspirations, conditioned by his social environment) with "means" (economic resources, social and family support, personal resilience). When an individual's means are not sufficient to meet needs, suicide may result.

Studies of life stress and social supports repeatedly show that family stresses such as abuse, alcoholism, history of mental illness and parental loss have significant positive effects on youth suicide (Pfeffer, 1989). In addition, individuals in families living in poverty and/or headed by single parents may be at greater risk. Bonner and Rich (1987) suggest that stressful life events and lack of family support may be associated with adolescent suicidal behavior; Morano, et.al. (1993) identify the actual absence of family members, especially parents, as producing lower levels Of family support.

This paper contributes to the body of existing research by combining demographic, economic and social determinants as proxies for external means in a time series model of youth suicide, and using that model to project suicide rates over the next decade. By focusing on a single age group that appears increasingly prone to self-destructive behavior, we may be better able to identify environmental risk factors and redirect our prevention efforts to more effective policies.

This research finds new evidence supporting the Easterlin-Holinger cohort theory of suicidal behavior. The cohort theory predicts that youth suicide rates move pari passu with the relative proportion of the youth cohort. The validity of the theory has been called into question recently, with youth suicide rates remaining high despite shrinking youth cohorts. The empirical analysis in this paper suggests that after controlling for economic and social effects, the cohort effect is still positive in sign and statistically significant, and that we may expect higher rates of youth suicide from 1994 forward as relative youth cohort size begins to rise again. Other findings suggest that higher incomes and fluctuations in income are associated with higher rates of youth suicide, and that the greater incidence of single-parent families may be contributing to the trend.

The following section presents a short review of the literature, citing earlier research on the demographic and socio-economic determinants of youth suicide. Section III of the paper describes the model and the choice of variables used to test the Easterlin-Holinger cohort hypothesis. Empirical tests of the model using multiple regression analysis are presented in Section IV, and projections of youth suicide rates for the next decade using the results of the analysis are presented in Section V. Section VI concludes with discussion and suggestions for future research.