Firearm homicide and other causes of death in delinquents: a 16-year prospective study

Abstract

Background: Delinquent youth are at risk for early violent death after release from detention. However, few studies have examined risk factors for mortality. Previous investigations studied only serious offenders (a fraction of the juvenile justice population) and provided little data on females.

Methods: The Northwestern Juvenile Project is a prospective longitudinal study of health needs and outcomes of a stratified random sample of 1829 youth (657 females, 1172 males; 524 Hispanic, 1005 African American, 296 non-Hispanic white, 4 other race/ethnicity) detained between 1995 and 1998. Data on risk factors were drawn from interviews; death records were obtained up to 16 years after detention. We compared all-cause mortality rates and causes of death with those of the general population. Survival analyses were used to examine risk factors for mortality after youth leave detention.

Results: Delinquent youth have higher mortality rates than the general population to age 29 years (P < .05), irrespective of gender or race/ethnicity. Females died at nearly 5 times the general population rate (P < .05); Hispanic males and females died at 5 and 9 times the general population rates, respectively (P < .05). Compared with the general population, significantly more delinquent youth died of homicide and its subcategory, homicide by firearm (P < .05). Among delinquent youth, racial/ethnic minorities were at increased risk of homicide compared with non-Hispanic whites (P < .05). Significant risk factors for external-cause mortality and homicide included drug dealing (up to 9 years later), alcohol use disorder, and gang membership (up to a decade later).

Conclusions: Delinquent youth are an identifiable target population to reduce disparities in early violent death.

Keywords: alcohol use; drug dealing; firearms; gangs; juvenile delinquents; longitudinal studies; mortality; prospective studies; public health; substance use.

FIGURE 1

Standardized all-cause mortality rates of delinquent youth after detention (N = 1829) compared with the general population. Rates for delinquent youth are given in deaths per 100 000 person-years; rates for the general population are given in deaths per 100 000 person-years in Cook County, 2005. We do not report racial/ethnic or age specific rates for groups with ≤3 deaths in our sample. Crude mortality rates across racial/ethnic categories were weighted to reflect the racial/ethnic distribution of the CCJTDC. *Indicates a statistically significant difference in mortality rates (P < .05).

FIGURE 2

Causes of death of delinquent youth at ages 15 to 29 compared with the general population (N = 1829), standardized on CCJTDC population. General population refers to all males and females ages 15 to 29 years who died in Cook County, Illinois, in 2005. Other cause of death includes internal causes and other accidents. Compared with the general population, delinquent males and females experienced a significantly (P < .05) higher percentage of deaths due to homicide and its subcategory, homicide by firearm. The total proportion of deaths due to homicide (firearm and other) were 90% for delinquent males, 39% for general population males, 42% for delinquent females, and 14% for general population females. Due to rounding error, sums of percentages in the figure may not be accurate.

FIGURE 3

Kaplan-Meier estimates of the probability of survival in delinquent youth after detention according to demographic characteristics and risk factors. Estimates for survival by gender and race/ethnicity are estimated for n =1798 participants (of the original 1829 participants, 4 identified as other race/ethnicity, 6 were missing correctional records, and 21 were incarcerated during the entire study period). Estimates for survival by gang membership and drug dealing are estimated for n =1619 participants (of the original 1829 participants, 31 died before the follow-up interview, 5 withdrew from the study, 42 were lost to follow-up, 92 were interviewed past the 4.5-year cutoff, 4 participants identified as other race/ethnicity, 6 were missing correctional records, 19 were incarcerated during the entire study period, and 11 were missing “selling drugs” and “gang membership”). Estimates for survival by alcohol use disorder are estimated for n =1615 participants (of the original 1829 participants, 31 died before the follow-up interview, 5 withdrew from the study, 42 were lost to follow-up, 92 were interviewed past the 4.5-year cutoff, 4 participants identified as other race/ethnicity, 6 were missing correctional records, 19 were incarcerated during the entire study period, and 15 were not assessed for alcohol use disorder).

FIGURE 4

Hazard ratio estimates of mortality in delinquent youth after juvenile detention: time-dependent risk factors. Hazard ratio estimates for gang membership and drug dealing are estimated for n =1619 participants (of the original 1829 participants, 31 died before the follow-up interview, 5 withdrew from the study, 42 were lost to follow-up, 92 were interviewed past the 4.5-year cutoff, 4 participants identified as other race/ethnicity, 6 were missing correctional records, 19 were incarcerated during the entire study period, and 11 were missing “selling drugs” and “gang membership”). Hazard ratio estimates for alcohol use disorder are estimated for n = 1615 participants (of the original 1829 participants, 31 died before the follow-up interview, 5 withdrew from the study, 42 were lost to follow-up, 92 were interviewed past the 4.5-year cutoff, 4 participants identified as other race/ethnicity, 6 were missing correctional records, 19 were incarcerated during the entire study period, and 15 were not assessed for alcohol use disorder).