Purpose: Health inequities among street-involved youth are theorized to be due to transience, which disconnects them from social capital, however, transience has been variously defined and seldom studied in health research. We derived two measures in the same survey, geographical transience (the movement between cities or regions) and residential transience (numbers of types of housing), to test whether transience is linked to social disconnection, health risk behaviours and service access among street-involved youth in Canada.
Methods: Secondary analysis of the 2006 British Columbia Street Youth Survey (N762 adolescents, age 12 to 18 years, 48% girls, 54% Aboriginal) from 9 cities. Residential transience (RES) was calculated from the number of types of shelter/housing over the past year (M 1.5), and geographical transience (GEO) by number of cities where youth had been street-involved (M1.7). A 3-category gradient, Low (0 -1 move), Medium (2-3 moves) and High (4 moves) was created for both. Youth were more likely to report residential (41.5% Low, 44.9% Medium, 13.6% HighRES) than geographical transience (64.6% Low, 19.7% Medium, 16.7 % HighGEO) with no gender differences, and very little overlap between groups: only 3.5% reported both high RES and GEO transience (kappa .07, s.e. .028). Comparisons across groups included family and school connectedness, past month somatic complaints, mental health conditions, suicidality, sexual exploitation, and sharing needles, using ANOVAs with post-hoc LSD, and crosstabs with chi-square, depending on the variable.
Results: As transience increased, social connection decreased and health risk behaviours increased. HighRES transient youth had lower family connectedness scores (LowRES M(sd): 2.22(.48), Medium: 2.17(.48), High: 1.95(.50), F(2,680) 17.37, p.001) as did HighGEO (Low: 2.18, Medium: 2.03, High: 2.05, F(2,730) 6.99, p.001). School connectedness was also lower on average for HighRES transience (F(2,687) 13.84, p.001) and for HighGEO transience (Low M(sd): 3.34(.99), Medium: 3.03(.98); High: 2.79 (1.1), F(2,685) 16.34, p.001). HighRES only were less likely to be currently attending school (44.6% vs. 77.0% LowRES, p.001). They also reported experiencing more types of violence, whether HighRES (F(2,687) 16.21, p.001) or HighGEO (F(2,740) 22.41, p.001). High transience youth were more likely to report current precarious housing (HighGEO-37.4% v. 15.7% Low; HighRES-46.7% v. 5.8% LowRES, both p.001). They reported higher levels of emotional distress, and were more likely to report suicidal ideation (HighRES-41.0% v. Low 23.5%; HighGEO-31.6%, Med-42.0%, Low-25.7%; both p.001) and suicide attempts. High transience youth reported using more kinds of substances in past 20 days (p.01 RES and GEO), sharing needles or crack pipes (HighGEO 39.7%-Low 19.7%, X2 26.17, p.001; HighRES 46.2%-Low14.5%, X2 60.35, p.001), and sharing razors/piercing equipment (HighRES-25.4% vs. LowRES-10.9%, p.01). They also were more likely to have been pregnant or caused a pregnancy (both p.01).
Conclusions: Frequent relocation by street youth, whether from city to city, or between living situations within a city, clearly disconnects them from social supports–family and school–and is associated with health inequities and risk behaviours. Services for homeless and street-involved youth should assess transience levels and tailor interventions, and research should incorporate measures of transience in studying determinants of health among street youth.