Youth and young adults who live or work on the street contend with an array of health and social challenges, including engagement in injection drug use( Reference Barker, Kerr and Dong 1 ), sex work( Reference Hayashi, Daly-Grafstein and Dong 2 ) and substandard housing( Reference Krusi, Fast and Small 3 ). In turn, these challenges have resulted in elevated morbidity and mortality among this population relative to other youth( Reference Boivin, Roy and Haley 4 , Reference Roy, Haley and Leclerc 5 ). Street-involved youth also frequently struggle to obtain nutritious and reliable food supplies( Reference Tarasuk, Dachner and Poland 6 , Reference Whitbeck, Chen and Johnson 7 ), which has been associated with nutritional deficiencies( Reference Kirkpatrick and Tarasuk 8 – Reference Tarasuk, Dachner and Li 10 ), kidney disease( Reference Crews, Kuczmarski and Grubbs 11 ), poor mental health( Reference Davison, Gondara and Kaplan 12 , Reference Jones 13 ), obesity( Reference Sirotin, Hoover and Shi 14 ) and diabetes( Reference Seligman, Bindman and Vittinghoff 15 ) in high-risk populations and in the general adult population. In addition, mental illness, such as major depressive disorder, disproportionately affects street-involved youth( Reference Ensign and Santelli 16 , Reference Gwadz, Gostnell and Smolenski 17 ).
Food insecurity is a broad conceptual term referring to the inability to acquire, through socially acceptable means, adequate food supplies for a healthy life( Reference Campbell 18 ). Food insufficiency is an individual-level stage of food insecurity that is defined by the experience of hunger( Reference Scott and Wehler 19 ). Previous studies have found associations between experiences of food insecurity and depressive symptoms among selected populations, such as people who are HIV-positive( Reference Heylen, Panicker and Chandy 20 , Reference Kapulsky, Tang and Forrester 21 ), adults who use injection drugs( Reference Anema, Wood and Weiser 22 ), recent mothers( Reference Dewing, Tomlinson and le Roux 23 ), individuals with diabetes( Reference Montgomery, Lu and Ratliff 24 ) and youth( Reference McIntyre, Williams and Lavorato 25 , Reference Romo, Abril-Ulloa and Kelvin 26 ). However, to our knowledge, there has been no investigation of the relationship between food insecurity and symptoms of depression among populations of street-involved youth who use illicit drugs.
Drawing on a prospective cohort study of street-involved youth who use illicit drugs in a Canadian setting, we undertook the present study to examine the relationship between food insufficiency and symptoms of depression, including changes in depression with rising levels of food insufficiency.
Methods
The At-Risk Youth Study (ARYS) is an open prospective cohort study of street youth who use illicit drugs based in Vancouver, Canada. Youth aged 14–26 years who have used any illicit drug (other than or in addition to cannabis) in the preceding 30d are eligible for study enrolment. Recruited youth are street-involved at baseline, defined as having been without stable housing or having accessed street-based services in the preceding 6 months( Reference Boivin, Roy and Haley 4 , Reference DeMatteo, Major and Block 27 , Reference Kerr, Marshall and Miller 28 ). Street-based outreach is used to enhance study recruitment both during daytime and night-time hours in a range of neighbourhoods throughout Vancouver where street youth are known to congregate. Snowball sampling is also used to maximize study enrolment. After providing informed consent, participants complete an interviewer-administered questionnaire regarding sociodemographic and socio-economic details, engagement with health and social services, substance use patterns and other behavioural data. All participants are provided with monetary compensation for their time ($CAN 30).
The present analysis draws on baseline data from youth recruited between April 2006 and November 2013. The primary outcome of interest was depression as measured by the Center for Epidemiological Studies Depression (CES-D) scale. CES-D is a scored twenty-item survey measuring depressive symptoms( Reference Radloff 29 ) that has been shown to be both valid and reliable when administered to young people( Reference Radloff 30 ) and homeless populations( Reference Fitzpatrick, Irwin and Lagory 31 ). A cut-off of ≥22 has been used successfully in previous studies as a measure of depression among high-risk populations( Reference Hallfors, Waller and Bauer 32 – Reference Weissman, Sholomskas and Pottenger 34 ). The primary exposure variable of interest, food insufficiency, was based on responses to the following ARYS survey question: ‘I am often hungry but I don’t eat because I can’t afford enough food’. Participants who responded ‘never true’ were compared on CES-D scores with those who responded ‘sometimes true’ and with those who responded ‘often true’. This individual statement was extracted from a validated food insecurity scale published by Radimer/Cornell( Reference Radimer, Olson and Campbell 35 ). This item prompts respondents for their current hunger status and has been shown to have good specificity and sensitivity compared with dietary proxies of food insufficiency( Reference Kendall, Olson and Frongillo 36 ). Because this population of youth experiences high rates of housing instability and homelessness, our analysis focuses on the relationship between depression and individual-level food insecurity, as opposed to household-level food insecurity.
To adjust for variables that might confound the relationship between food insufficiency and depressive symptoms, we examined an array of covariates, including: age (per year older, continuous); gender (male v. female); Caucasian/white ethnicity (yes v. no); incarceration in last 6 months (detention, prison or jail; yes v. no); currently in a stable relationship (legally married/common law, separated, divorced/widowed or regular partner; yes v. no); homelessness in the last 6 months (yes v. no); employment in the last 6 months (yes v. no); high school completion (yes v. no); sex work in the last 6 months (received money, gifts, food, shelter, clothes or drugs in exchange for sex; yes v. no); daily cocaine use in the last 6 months (yes v. no); daily heroin use in the last 6 months (yes v. no); daily meth use in the last 6 months (yes v. no); daily crack use in the last 6 months (yes v. no); any injection drug use in the last 6 months (yes v. no); drug or alcohol treatment in the last 6 months (yes v. no); and heavy alcohol use (yes v. no). Heavy alcohol use was defined as: >14 drinks/week or >4 drinks on one occasion for men; and >7 drinks/week or >3 drinks on one occasion for women. This measure of heavy alcohol use has been described by the National Institute on Alcohol Abuse and Alcoholism( Reference Willenbring, Massey and Gardner 37 ).
Initially, we examined the descriptive characteristics, stratified by depression (CES-D score ≥22) at the first study visit. Comparisons were made using the Pearson’s χ 2 test for binary variables (Fisher’s exact test when cell counts were less than or equal to 5) and the Wilcoxon rank-sum test for continuous variables. Next, we examined the bivariate associations between each explanatory variable and depression using logistic regression. As the last step, we fitted a multivariable model, considering all variables in bivariate analyses as the full model. All statistical analyses were performed using RStudio version 0.99.892 (R Foundation for Statistical Computing, Vienna, Austria). All P values were two-sided and tests were considered significant at P<0·05.
Results
Of 1066 street-involved youth who were eligible for inclusion in the analysis, 340 (31·9 %) identified as female, 726 (68·1 %) identified as Caucasian/white, and the median age was 21·8 years (interquartile range=19·9–23·6 years). In total, 724 (67·9 %) reported some food insufficiency and 565 (53·0 %) reported CES-D score ≥22. Among those who reported food insufficiency, 241 (33·3 %) reported experiencing it ‘often’ and 483 (66·7 %) reported experiencing it ‘sometimes’.
Table 1 lists sociodemographic, socio-economic and substance use characteristics stratified by CES-D score ≥22. Factors positively associated with depression in bivariable analyses included: food insufficiency ‘sometimes’; food insufficiency ‘often’; homelessness; daily heroin use; daily meth use; daily crack use; any injection drug use; sex work; and drug or alcohol treatment. Male gender, high school completion and employment were negatively associated with depression.
Bivariate analyses of factors associated with depression among a cohort of street-involved youth (n 1066); At-Risk Youth Study (ARYS), Vancouver, Canada, April 2006–November 2013
CES-D, Center for Epidemiological Studies Depression scale; IQR, interquartile range; Ref., reference category.
* In the last 6 months.
Table 2 shows the adjusted odds ratios (AOR) for CES-D score ≥22 and variables of interest. Variables independently positively associated with depression in the multivariable analysis included: food insufficiency ‘sometimes’ (AOR=1·99; 95 % CI 1·47, 2·70); food insufficiency ‘often’ (AOR=2·52; 95 % CI 1·74, 3·67); daily crack use (AOR=1·76; 95 % CI 1·20, 2·61); any injection drug use (AOR=1·81; 95 % CI 1·31, 2·49); and sex work (AOR=2·43; 95 % CI 1·49, 4·09). Male gender remained negatively associated with depression (AOR=0·61; 95 % CI 0·45, 0·84).
Multivariate analysis of factors associated with depression among a cohort of street-involved youth (n 1066); At-Risk Youth Study (ARYS), Vancouver, Canada, April 2006–November 2013
AOR, adjusted odds ratio.
* In the last 6 months.
Discussion
In the present study of street-involved youth who use illicit drugs, we observed a high prevalence of both depression and food insufficiency, with more than five in ten youth reporting CES-D score ≥22, and nearly seven in ten youth reporting some level of food insufficiency. Compared with youth who reported having sufficient food, the odds of depression (CES-D score ≥22) were higher among youth who reported any level food insufficiency. In addition, the odds of depression increased as the level of food insufficiency increased and we observed little change in these estimates after adjustment for confounders. Depression was also independently positively associated with daily crack use, injection drug use and sex work, and was independently negatively associated with male gender.
Our findings build on previous studies that have found associations between food insecurity and depression among youth( Reference Romo, Abril-Ulloa and Kelvin 26 , Reference McIntyre, Wu and Kwok 38 ) and among adults who use injection drugs( Reference Anema, Wood and Weiser 22 ). One study using a nationally representative sample of Canadian youth found that hunger was significantly associated with future risk of depression, even after accounting for previous depressive symptoms( Reference McIntyre, Wu and Kwok 38 ). Street-involved youth who are food insufficient often experience poverty, which has been repeatedly associated with depression( Reference Catz, Gore-Felton and McClure 39 , Reference Wu, Munoz and Espiritu 40 ). However, one study found that while family food insufficiency was independently associated with depressive symptoms (dysthymia and suicide symptoms) among adolescents, low family income was not( Reference Alaimo, Olson and Frongillo 41 ). These findings suggest that pathways other than poverty may mediate the relationship between food insufficiency and depression. For instance, stress and anxiety have been associated with both food insecurity and depression( Reference Williamson, Birmaher and Anderson 42 , Reference Maynard, Perlman and Kirkpatrick 43 ). Further, nutritional deficiencies( Reference Cabrera, Mesas and Garcia 44 ), shame( Reference Andrews, Qian and Valentine 45 ), childhood trauma( Reference Hadland, Marshall and Kerr 46 ) and hopelessness( Reference Kuo, Gallo and Eaton 47 ) have all been associated with symptoms of depression and may mediate this relationship. There is also evidence of a bidirectional relationship between food insecurity and depression( Reference Huddleston-Casas, Charnigo and Simmons 48 ), which may suggest that depression contributes to food insecurity in street-involved youth by acting as a barrier to services( Reference Aviles and Helfrich 49 ) or by impeding the purchase and preparation of meals. The associations found in the current analysis between depression and daily crack use, injection drug use and female gender are consistent with what has previously been found in the literature( Reference Hadland, Marshall and Kerr 50 , Reference Pettes, Kerr and Voon 51 ).
The findings from the current study suggest that there is a need to increase access to quality mental health and food security interventions among this population of youth. Previous studies have found that street-involved youth experience high rates of mental illness( Reference Ensign and Santelli 16 , Reference Gwadz, Gostnell and Smolenski 17 ). The high-risk environment and trauma associated with living or working on the streets often intensify pre-existing mental health issues and make it challenging for clinicians to diagnose youth appropriately( Reference Edidin, Ganim and Hunter 52 ). In addition to these high rates of mental health issues, unstably housed youth face many barriers to care including fear of discrimination, long waiting lists and lack of specialized youth services( Reference Christiani, Hudson and Nyamathi 53 – Reference Hadland, Kerr and Li 56 ). Further, due to the large nutritional gap that still exists for Canada’s most vulnerable populations, food security interventions (e.g. food banks) are often unable to meet demands( Reference Tarasuk, Dachner and Loopstra 57 ) and some homeless youth have described food assistance programmes to be of poor quality and associated with food sickness( Reference Dachner and Tarasuk 58 ).
Because of the tremendous health and economic burden associated with both food insecurity and mental illness, it is critical that policies and programmes focus on the social and environmental determinants of food access, such as formal employment (hindered by factors such as educational limitations, discrimination and incarceration)( Reference Gwadz, Gostnell and Smolenski 17 ) and stable housing (hindered by factors such as family breakdown, abuse and ‘ageing out’ of the foster care system)( Reference Edidin, Ganim and Hunter 52 ). It is also important that mental health programmes consider how food insufficiency may act as an underlying cause or contributing factor to depression, and that interventions addressing food insecurity gain capacity to support youth who are experiencing mental illness and refer them to appropriate services. There is evidence that social support may moderate the relationship between food insufficiency and depression( Reference Kapulsky, Tang and Forrester 21 , Reference Tsai, Tomlinson and Comulada 59 ), which highlights the potential benefits of incorporating social support interventions into food security programmes.
There are limitations to the present study. First, because it is a cross-sectional study, we are unable to determine the direction of the relationship between food insufficiency and depression. Second, our sample was not randomly selected (as there are no registries of street youth to draw upon) and included only youth who had used illicit drugs in the 30d prior to baseline. Therefore, our sample may not be representative of all street youth in Vancouver. However, we note that the characteristics of the ARYS sample are similar to those from other studies of high-risk youth( Reference Evans, Hahn and Lum 60 – Reference Steensma, Boivin and Blais 62 ). Third, we relied on self-report, which may have been subject to response biases, including recall bias and socially desirable responding, although we know of no reason why this would explain the associations we identified in our study. Fourth, the Radimer/Cornell scale is no longer frequently used in food security research, as newer scales such as the Household Food Security Survey Module (HFSSM)( 63 ) have been developed. However, we note that the individual item on the Radimer/Cornell scale used in the current study has been incorporated into several recently developed food insecurity scales (including the HFSSM) and has remained relatively consistent across them( 64 ). Lastly, the current study did not include a detailed analysis of race or ethnicity. To control for confounding, we included an ‘ethnicity’ variable that was limited to ‘Caucasian/white’ v. ‘other’. However, given that youth who identify as ‘non-white’ are more likely to experience unstable housing relative to youth who identify as ‘white’( Reference Fowler, Toro and Miles 65 ), and that race has previously been associated with food insecurity( Reference Myers and Painter 66 ), it would have been useful to further examine whether race/ethnicity has a modifying effect on the relationship between food insufficiency and depression.
To our knowledge, the present study is the first to document a relationship between food insufficiency and depression among street-involved youth who use illicit drugs. These findings call attention to the limitations of current food security interventions in meeting their nutritional needs, as well as the importance of addressing structural factors that contribute to the high rates of food insufficiency within this population. The current study also highlights the importance of addressing the mental health concerns of youth who are already food insufficient. Further research is needed to identify potential pathways by which food insufficiency may lead to depression and how depression may lead to further food insufficiency by hindering access to food supplies and/or nutritional support programmes.
Acknowledgements
Acknowledgements: The authors thank the study participants for their contribution to the research, as well as current and past researchers and staff. Financial support: The study was supported by the US National Institutes of Health (grant number U01DA038886). K.D. is supported by a Michael Smith Foundation for Health Research/St. Paul’s Hospital Foundation–Providence Health Care Career Scholar Award and a Canadian Institutes of Health Research New Investigator Award. This research was undertaken, in part, thanks to funding from the Canada Research Chairs programme through a Tier 1 Canada Research Chair in Inner City Medicine which supports Dr Evan Wood. The funding agencies had no role in the design, analysis or writing of this article. Conflict of interest: The authors declare no conflicts of interest. Authorship: J.G.-H., K.D. and T.K. conceptualized the study design. E.N. performed the statistical analyses, and J.G.-H. and T.K. interpreted the results. J.G.-H. drafted the initial manuscript. T.K., K.D. and L.D. provided substantial revisions of the manuscript and guidance throughout the writing. All authors have read and approved the final manuscript. Ethics of human subject participation: This study was conducted according to the guidelines laid down in the Declaration of Helsinki and all procedures involving human subjects were approved by the University of British Columbia and Providence Health Care Research Ethics Board. Written informed consent was obtained from all subjects.
