Racial Differences in Cigarette Smoking Among Homeless Youth

Introduction

Several studies have reported pronounced racial/ethnic differences in smoking behavior among homeless youth. Better understanding the factors underlying racial/ethnic differences in daily smoking among homeless youth may help inform programs to reduce smoking in this population.

Methods

Data come from a probability sample of homeless youth in Los Angeles County collected between 2008 and 2009. The sample includes 116 African American, 99 Hispanic, and 119 White youth with ages ranging from 13 to 24. Chi-square tests were used to test for differences in daily smoking among African American, Hispanic, and White youth. Propensity score and doubly robust methods were used to produce a less biased estimate of the association between daily smoking and race/ethnicity after having removed the effect of potential confounders.

Results

The daily smoking rate for White youth was 70.1%, more than 31 percentage points than the rates for either African American or Hispanic youth. Propensity score analysis revealed that the majority of the racial/ethnic differences in smoking rates could be explained by differences in homelessness severity, although background characteristics and comorbidity were relevant as well.

Conclusions

As programs are developed to reduce smoking among homeless youth, results suggest that additional outreach may be needed to engage White youth in services. Also, smoking prevention programs may benefit from incorporating a social network-based approach that assists youth in fostering relationships with lower-risk peers, as well as addressing other forms of substance use. Incorporating these elements may help reduce the large racial/ethnic disparities in daily smoking among homeless youth.

Implications

This report extends the small existing literature on racial/ethnic differences in smoking among homeless youth in two important respects. First, it confirms differences in daily smoking, an important indicator of dependence, across racial/ethnic groups. Second, it seeks to understand the extent to which differences in smoking can be explained by demographic characteristics (other than race/ethnicity), background factors, homelessness severity, and comorbidity characteristics known to be associated with substance use among homeless youth.