The association between experiencing homelessness in childhood or youth and adult housing stability in Housing First

Table 4 Sensitivity analyses - least absolute shrinkage and selection operator analyses to estimate the association between first experiencing homelessness in childhood or youth (< 25 years) and housing stability (≥90%) among Vancouver At Home participants randomized to Housing First

Control Covariates Added	Primary Independent Variable	Unadjusted Odds Ratio (95% CI)	P value	Adjusted Odds Ratio (95% CI)	P value
Family separation (n = 274)¹	Age of first homeless
	< 25 years	0.51 (0.31, 0.85)	0.009	0.54 (0.31, 0.95)	0.032
	≥ 25 years	Reference		Reference	0.032
Adverse childhood experiences score (n = 229)²	Age of first homeless
	< 25 years	0.55 (0.32, 0.96)	0.035	0.60 (0.33, 1.11)	0.103
	≥ 25 years	Reference		Reference	0.103

¹In addition to control covariates included in the primary LASSO multivariable model, two family separation variables were added: 1) “Lived away from parents for any reason under the age of 18” (yes/no) and 2) “Ever placed in foster care” (yes/no)
²In addition to control covariates included in the primary LASSO multivariable model, the following variable was added: “Adverse childhood experiences” (score)

ISSN: 1471-244X