Street youth mortality: leaning with intent to fall.

Abstract

THE WORDS “STREET YOUTH” CONJURE DIFFERENT IMages for each reader based on experience or suppositions. Clinicians who are newly introduced to a group of homeless or marginally housed adolescents and young adults are invariably surprised by what they had never noticed before. They begin to recognize young people on the street they had previously ignored and identify street youth among their own patients. Street youth are often invisible, blending into the urban, suburban, or rural landscape. To complicate matters, street youth do not identify themselves as homeless or marginally housed unless such identifiers provide a specific advantage. Even then, street youth often resist the label because they stereotype homeless people as substance-abusing adults who should be avoided as dangerous, not individuals with whom to identify. Alternatively, street youth often depict themselves as adventurers and self-determined individualists who are living life on their own terms. Even those with a long history of neglect and abuse rarely tolerate being treated like dependents or objects of pity. The study of mortality among Montreal street youth by Roy et al in this issue of JAMA is a reminder of the vulnerability of street youth that belies their bravado. In this prospective study, 1013 street youth aged 14 to 25 years were followed every 6 months, with an average follow-up of 33.4 months between January 24, 1995, and September 30, 2000. Twenty-six of these young individuals died. The high mortality rate, more than 11 times that of the general youth population in Montreal, is consistent with other investigators’ reports of elevated mortality ratios in other cities in Canada and the United States. Unlike earlier investigations of the mortality of homeless populations, this study treated homelessness as a time-dependent variable, permitting the authors to further conclude that mortality risk is increased during homeless periods; thus, homelessness in itself, not just being a street youth with tenuous housing and risky behavior, is an independent predictor of death. Factors mediating the effect of homelessness on the death rate could not be defined in this study but are certainly suggested by the leading causes of death: suicide and drug overdose. Infection with human immunodeficiency virus (HIV) was the strongest predictor of mortality in this study, with an adjusted hazard ratio of 5.6. This is not surprising given the rates of HIV and HIV behavior risk reported in the Montreal cohort. In a recent study of HIV-positive youth aged 12 to 18 years at baseline who were infected through sex or injection drug use and were receiving routine health care, fully 27% of both females and males reported having been homeless. Mortality was due to the complications of HIV and its treatment. For 90 male youth followed for a total of 203.3 person-years (mean follow-up time, 2.26 years), the observed death rate was 490 per 100000 person-years over the observation period; for 262 females followed for 634.4 person-years (mean follow-up time, 2.42 years), the observed death rate was 1100 per 100000 person-years (a rate much greater than the 2001 death rate of 80.7 per 100000 youth aged 15-24 years). Depression occurred in 43% of these individuals. Depression and health anxiety were related to substance abuse in this HIV-infected cohort. In the study by Roy et al, the death of 1 HIV-infected street youth was attributed to complications of HIV; 2 of the other 3 deaths among HIV-infected youth were drug overdoses. Although Roy et al did not gather specific information on depression, it is likely that depressive disorder and substance abuse coupled with HIV could increase the effect of an already high HIV death rate on mortality. The ability of Roy et al to follow 1013 street youth for an average of 33.4 months per participant is noteworthy. Tracking street youth for this length of time is difficult. The mobility of unattached youth and their friends, some of whom may be nomadic and travel to other cities over the observation period, as well as the mobility of family and other “stationary” contacts makes tracking highly dependent on relationship-building with youth, their contacts, and with organizations providing youth services. While the study participants were given $20 per visit, this amount, although respectable, will not induce many youth to make an unusual effort to participate in a study, as evidenced by the 12% refusal rate of youth approached for enrollment. The characteristics of street youth who do not participate, particularly