Importance of Community Context in Understanding Substance Use Among Adolescents With Allegations of Maltreatment

Abstract

This study examined the association between community disadvantage and substance use among adolescents referred to child protective services (CPS) and the interaction role of race/ethnicity in the association between community disadvantage and adolescent substance use. Findings revealed that higher county-level alcohol/drug mortality rates were associated with a higher risk of substance use across non-Hispanic Black, non-Hispanic white, and Hispanic adolescents. An interaction test showed that the association between county-level alcohol/drug mortality rates and risk of substance use was particularly stronger for non-Hispanic Black adolescents compared to non-Hispanic white adolescents. However, county-level vacant housing rates and low median household income did not show a significant association with adolescent substance use across all racial/ethnic groups. Results suggest that practitioners should screen for community-level substance risk levels to prevent substance use among historically marginalized racial-ethnic adolescents involved in CPS.

Keywords

community disadvantage substance use child protective services

Introduction

As of 2021, 23% of adolescents reported drinking alcohol during the past 30 days, 16% used marijuana during the past 30 days, and 12% reported ever nonmedical use of prescription opioids (Hoots, 2023). Adolescent substance use is a significant public health concern as it can impact physical growth and brain development (Fuhrmann et al., 2015; Ganzer et al., 2016) and other risk behaviors, including weapon carrying (Khubchandani & Price, 2018), physical fighting (DeLisi et al., 2015), and risky sexual behaviors (Jackson et al., 2015). Adolescent substance use is also associated with mental health problems in later life, such as anxiety (Asselmann et al., 2014), depression (Scholes-Balog et al., 2015), or suicidal behaviors (Ammerman et al., 2018).

Adolescents referred to child protective services (CPS) are likely to be at increased risk of substance use. While national statistics on drug use among children referred to CPS are not frequently reported, an estimated number of 4,089 children in foster care in 2022 had child drug use listed as a removal cause (representing approximately 2% of all children in care). One prior study using national data of CPS investigations found that approximately 2.26% of children referred to CPS for alleged maltreatment had reported substance use (Children’s Bureau, 2021; LaBrenz et al., 2024; U.S. Department of Health and Human Services, 2023). Past literature has consistently shown that exposure to maltreatment increases the risk of substance use (Benedini & Fagan, 2020; Cicchetti & Handley, 2019; Tonmyr et al., 2010). According to the self-medication theory (SMT; Khantzian, 1997), substance use behaviors begin as an attempt to alleviate painful feelings or control emotional stress. Empirical evidence on maltreated children’s substance use has provided support for SMT, in which substance use of individuals with a history of child maltreatment was viewed as a coping strategy in response to emotional or psychological stress (Austin & Shanahan, 2018; Liu et al., 2018). Informed by the SMT, the literature has attempted to identify individual- and family-level disadvantages impacting substance use initiation among children who experienced maltreatment or alleged maltreatment (Aarons et al., 2008; Guibord et al., 2011; Hagborg et al., 2020; Hamburger et al., 2008; Moran et al., 2004; Shin et al., 2018; Singh et al., 2011; Traube et al., 2012; Yampolskaya et al., 2019). For example, studies have found that adolescents experiencing multiple types of maltreatment are more likely to use substances than those who experienced no or one type of maltreatment (Hagborg et al., 2020; Hamburger et al., 2008; Moran et al., 2004). In addition, family adversity, such as parental substance use and family poverty, has been associated with adolescent substance use behaviors (Singh et al., 2011), possibly through low-quality parenting and perception of substance use as a coping response (Cardoso, 2018; Eitle & Eitle, 2014; Hogarth et al., 2019; Moise et al., 2020; Smith et al., 2016; Straussner & Fewell, 2018).

During the 2000s, scholars have begun to focus on contextual disadvantages associated with maltreated adolescents. Notably, Coulton et al. (2007) emphasized the importance of community and contextual factors when assessing child maltreatment. In response to this call, an expanding body of empirical research explored the association between community disadvantage and maltreatment. For example, Thurston et al. (2021) found that the availability of alcohol establishments increased the risk of child physical abuse. Similarly, Maguire-Jack and Font (2017) found that neighborhood poverty was related to neglect, even after adjusting for family income.

While a growing body of research has investigated the relationship between contextual disadvantages and maltreatment over the last decade, there has been relatively limited attention paid to community disadvantages associated with substance use among maltreated children (Duprey et al., 2017; Handley et al., 2015). Within the limited literature, one study conducted by Handley et al. (2015) used data from 266 maltreated adolescents and 141 nonmaltreated adolescents and found that community disadvantage, measured by perceived safety, perceived drug availability, and poverty, was associated with a greater risk of drug dependence symptoms among adolescents who had experienced maltreatment, but not among non-maltreated adolescents. Another study conducted by Duprey et al. (2017) used data from 300 maltreated children and 665 non-maltreated children and reported that perceived community disorder, such as open drug activity, mediated the pathway from the severity of neglect to adolescent substance use. Their data, however, had a small sample size collected from a limited number of communities, and the studies relied on perceived rather than objective measures of community environments. Given that objectively measured community environments can be informative for designing community-focused CPS programs, an investigation of the association between objectively measured contextual factors and substance use is warranted.

Furthermore, the association between community disadvantage and substance use can vary across different racial/ethnic groups. Features of community environments have been found to be especially significant for historically marginalized racial-ethnic groups who often experience everyday discrimination, structural racism, and oppression. Considering that experiences of discrimination can diminish healthy coping mechanisms (Mannoh et al., 2021; Noonan et al., 2016), the impact of community disadvantage might be more pronounced for historically marginalized racial-ethnic adolescents who already have the burdens of discrimination. To our best knowledge, no study has yet examined racial/ethnic differences in the association between community disadvantage and substance use.

To address the gaps, the current study, guided by the SMT (Khantzian, 1997), examines community disadvantages associated with substance use among adolescents referred to CPS, after controlling for child/family-level factors. We also examine the interaction role of race/ethnicity in the relationship between community disadvantage and adolescent substance use. We included alcohol and drug mortality rates, vacant housing rates, and low median household income as community disadvantage factors. Based on SMT and prior literature, we hypothesize that adolescents exposed to more adversity at the community level (e.g., higher alcohol and drug mortality rates, higher vacant housing rates, and lower median household income) will have higher odds of substance use. We also hypothesized that the association between community disadvantage and the likelihood of substance use will be stronger for non-Hispanic Black and Hispanic adolescents compared to non-Hispanic white adolescents.

Method

Data

The present study used data from the 2019 National Child Abuse and Neglect Data System (NCANDS). The 2019 NCANDS is the most comprehensive data system of child abuse and neglect in the United States, which documents information on all children who have received a disposition of an investigation or assessment of allegations of maltreatment in 2019 from the child welfare information systems. The information was collected by state and territorial child protective agencies in 50 states, Puerto Rico, and the District of Columbia. The data files contain the demographic information on children and their perpetrators, types of maltreatment, investigation or assessment dispositions, and risk factors. Further details about the NCANDS have been described in the Children’s Bureau (2021). In addition, we characterized community characteristics related to substance use and social conditions by linking the NCANDS to the American Community Survey 2015–2019 and the Wide-Ranging Online Data for Epidemiologic Research (WONDER) 2016–2018 based on the county where the adolescent was residing at the time of the report of maltreatment.

Our study focused on adolescent substance use associated with community disadvantage, limiting the analysis to adolescents aged 15 years or older—an age range associated with the onset of substance use (Alcover & Thompson, 2020; Chen et al., 2017)—and included only those with information on both substance use status and their county of residence. For cases with multiple records in 2019 NCANDS, we included only the first referral, resulting in a sample size of 170,757. Using listwise deletion method, we excluded missing data on adolescent race/ethnicity (n = 5,651), sex (n = 329), and caregiver’s substance use (n = 32,885) because our bivariate analysis showed no significant differences in completion rates by sociodemographic factors. It resulted in 131,892 adolescents in our analytic group. This study used secondary administrative data linked and de-identified. A waiver of informed consent was obtained by the first author’s institutional review board.

Measures

Adolescent Substance Use

Adolescent substance use was measured based on the questions “if the child was reported to compulsively use or need alcohol” and “if the child was reported to compulsively use or need narcotics.” If either question was answered yes, the adolescent was deemed to have used alcohol or drugs.

Community-Level Measures

Community-level measures included alcohol and drug mortality rates at the county level (obtained from the WONDER 2016–2018), and vacant housing rates and median household income at the county level (obtained from the American Community Survey 2015–2019). We examined alcohol and drug mortality rates per 100,000 people at the county level in 2016–2018 based on the International Statistical Classification of Diseases, 10th revision codes: alcohol-induced causes, unintentional drug poisonings, suicide drug poisonings, homicide drug poisonings, undetermined drug poisonings, and all other drug-induced causes. Vacant housing rates and median household income from 2015 to 2019 were examined at the county level to characterize community socioeconomic conditions. To allow for comparability among different measures, the community-level factors were standardized to have a mean of 0 and a standard deviation of 1. In multivariate analysis, we multiplied the standardized score of median household income by -1 to reflect “low” median household income, aligning it with the directions of alcohol and drug mortality rates and vacant housing rates. Finally, urbanicity was examined at the county level by distinguishing major metropolitan areas with populations over 1 million, urbanized areas not part of major metropolitan areas with populations of 1 million or less, and nonmetro areas.

Child/Family-Level Measures

Sociodemographic characteristics included adolescent age, sex (female [reference], male), and race/ethnicity self-identified or identified by a caregiver (non-Hispanic Black, non-Hispanic white [reference], Hispanic, non-Hispanic other). In addition, we included the number of types of alleged abuse/neglect types as a continuous variable, which indicates whether the adolescent was alleged as a victim of one or multiple types of abuse/neglect among physical abuse, neglect, medical abuse, sexual abuse, emotional abuse, and other abuse types. Finally, we included a caregiver’s drug and alcohol use (yes, no [reference]). A caregiver refers to a person responsible for the care and supervision of an adolescent. A caregiver’s substance use was measured based on the questions “if alcohol abuse is a problem of the child’s caregiver(s)?” and “if drug abuse is a problem of the child’s caregiver(s)?” If either question was answered affirmatively, the caregiver was deemed to have used alcohol or drugs.

Analysis

A multilevel logistic regression analysis was conducted based on a model-building process. First, an unconditional model was estimated to calculate the intraclass correlation (ICC) coefficients between counties. With a high ICC value (ICC = 0.43), we concluded that there were sufficient between-county variations to warrant the use of multilevel modeling. Next, we included the child/family-level measures in the model. Third, we included all the community-level measures while adjusting for the individual-level measures. Fourth, we tested cross-level interactions between race/ethnicity and three community-level measures. We found a significant interaction effect between race/ethnicity and alcohol and drug mortality rates at alpha 0.10, but not between race/ethnicity and the other two community factors. Although the interaction effects by race/ethnicity were not significant for vacant housing rates and low median household income, past literature has shown different community effects by race/ethnicity (Alegría et al., 2014; Karriker-Jaffe et al., 2016; Rosenblatt et al., 2021). Finally, we performed a sensitivity analysis using multiple imputation, rather than the listwise deletion method, to handle missing data and verify the consistency of the multilevel logistic regression analysis results. Throughout the model fitting process, the Akaike information criterion (AIC) and the Bayesian information criterion (BIC) were assessed to provide evidence for the fit of the model, with lower AIC and BIC values indicating a better model fit. Analyses were conducted using StataSE 16 software program.

Results

Descriptive Statistics

Table 1 shows the sociodemographic characteristics of adolescents in our analytic sample. The sample included slightly more girls (59%) than boys (42%). Over 46% of the adolescents were identified as non-Hispanic white, and 25% and 23% identified as Hispanic and non-Hispanic Black, respectively. Caregiver drug and alcohol use rates were 8% and 4%, respectively. In addition, approximately 4% of the adolescents reported substance use.

Table 1.

Sociodemographic Characteristics (N = 131,892).

Characteristics	Sample characteristics
Age
15–16 years	72.6%
17 years or above	27.4%
Sex
Female	58.5%
Male	41.5%
Race/ethnicity
Non-Hispanic White	46.4%
Non-Hispanic Black	22.5%
Hispanic	24.8%
Non-Hispanic other	6.3%
Number of alleged abuse/neglect types (M ± SD)	1.1 ± 0.60
Caregiver drug use
Yes	7.9%
No	92.1%
Caregiver alcohol use
Yes	3.8%
No	96.2%
Adolescent substance use
Yes	4.3%
No	95.7%
County-level alcohol and drug mortality rates (M ± SD)	35.0 ± 16.98
County-level vacant housing rates (M ± SD)	10.2 ± 4.37
County-level median household income (M ± SD)	63,222.2 ± 14,701.9
Urbanicity
Metro areas with over 1 million population	58.5%
Metro areas with ≤ 1 million population	33.2%
Nonmetro areas	8.3%

Table 2 shows racial/ethnic differences in maltreatment characteristics, substance use, and community disadvantages. Notably, non-Hispanic Black adolescents exhibit a lower reported rate of substance use (3% vs. 5–7%), lower caregiver drug and alcohol use rates (drug: 6% vs. 7–10%; alcohol: 2% vs. 4–6%), and more likely to live in metro areas with over 1 million population (72% vs. 48-67%) than their counterparts in other racial/ethnic groups.

Table 2.

Racial/Ethnic Differences in Abuse/Neglect Experiences, Substance Use, and Community Characteristics (N = 131,892).

Characteristics	NH White (N = 61,209)	NH Black (N = 29,673)	Hispanic (N = 32,650)	NH Other (N = 8,360)	χ² (F-value)
Adolescent substance use					427***
Yes	5.1%	3.1%	3.3%	7.0%
No	94.9%	96.9%	96.7%	93.0%
Number of alleged abuse/neglect types					292***
0 type	7.7%	9.5%	10.2%	7.7%
1 type	73.2%	74.2%	72.4%	74.7%
2+ types	19.1%	16.3%	17.4%	17.6%
Caregiver drug use					348***
Yes	8.9%	5.7%	7.4%	9.8%
No	91.1%	94.3%	92.6%	90.2%
Caregiver alcohol use					515***
Yes	4.6%	1.9%	3.6%	6.0%
No	95.4%	98.2%	96.4%	94.0%
County-level alcohol and drug mortality rates (M ± SE)	37.5 ± 0.07	35.2 ± 0.10	29.5 ± 0.07	36.3 ± 0.17	(1,648***)
County-level vacant housing rates (M ± SE)	10.5 ± 0.02	10.2 ± 0.02	9.6 ± 0.02	10.1 ± 0.05	(265***)
County-level median household income (M ± SE)	62,446 ± 61.7	63,338 ± 80.3	64,425 ± 78.7	63,792 ± 163.2	(135***)
Urbanicity					<0.1***
Metro areas with over 1 million population	47.8%	71.7%	67.2%	56.0%
Metro areas with ≤ 1 million population	38.0%	25.1%	30.8%	36.6%
Nonmetro areas	14.1%	3.2%	2.0%	7.4%

p < .05. **p < .01. ***p < .001.

Association Between Community Disadvantage and Adolescent Substance Use

Table 3 presents the results of multilevel modeling, examining the association between community disadvantage and substance use among CPS-involved adolescents. Higher alcohol and drug mortality rates were associated with a greater likelihood of substance use (AOR = 1.31, 95% CI = 1.06–1.61) after adjusting for adolescent sociodemographic and abuse/neglect characteristics. However, county-level vacant housing rates and low median household income were not significantly associated with substance use. A marginally significant interaction effect of race/ethnicity was identified, indicating that the association between alcohol and drug mortality rates and substance use is stronger among non-Hispanic Black adolescents compared to non-Hispanic white adolescents (AOR = 1.10, 95% CI = 0.99–1.22).

Table 3.

Results of Multilevel Models Assessing the Association Between Community Disadvantage and Adolescent Substance Use (N = 131,892).

	Model 1		Model 2
Variables	OR	95% CI	OR	95% CI
Age	1.23***	[1.16, 1.32]	1.23***	[1.16, 1.32]
Male (reference: female)	1.32***	[1.24, 1.40]	1.32***	[1.24, 1.40]
Race/ethnicity (reference: non-Hispanic white)
Non-Hispanic Black	0.90*	[0.82, 0.99]	0.88*	[0.80, 0.97]
Hispanic	0.94	[0.86, 1.02]	0.93	[0.85, 1.01]
Non-Hispanic other	1.03	[0.92, 1.14]	1.02	[0.91, 1.15]
Number of alleged abuse/neglect types	1.53***	[1.45, 1.62]	1.53***	[1.45, 1.62]
Caregiver drug use (reference: no)	4.76***	[4.41, 5.15]	4.77***	[4.41, 5.15]
Caregiver alcohol use (reference: no)	1.66***	[1.47, 1.86]	1.66***	[1.47, 1.86]
County-level alcohol and drug mortality rates	1.31*	[1.06, 1.61]	1.29*	[1.04, 1.59]
County-level vacant housing rates	1.05	[0.89, 1.24]	1.05	[0.89, 1.24]
County-level low household income	0.91	[0.69, 1.20]	0.91	[0.69, 1.20]
Urbanicity (reference: metro areas with over 1 million population)
Metro areas with ≤ 1 million population	1.24	[0.71, 2.16]	1.24	[0.71, 2.15]
Nonmetro areas	1.73	[0.85, 3.55]	1.72	[0.84, 3.52]
Race/ethnicity × county-level alcohol and drug mortality rates
Non-Hispanic Black			1.10^†	[0.99, 1.22]
Hispanic			1.01	[0.90, 1.14]
Non-Hispanic other			1.01	[0.88, 1.16]

Note. Alcohol and drug mortality rates, vacant housing rates, and low household income were standardized to have a mean of 0 and standard deviation of 1. County-level low household income was calculated by multiplying “county-level household income (standardized)” by −1.

†

p < .10. *p < .05. **p < .01. ***p < .001.

Table 4 reports the association between community disadvantage and adolescent substance use by race/ethnicity. Higher alcohol and drug mortality rates were associated with a 1.42 times greater likelihood of substance use among non-Hispanic Black adolescents (AOR = 1.42, 95% CI = 1.13–1.78). This association was stronger compared to non-Hispanic white adolescents (AOR = 1.33, 95% CI = 1.07–1.65) and non-Hispanic adolescents of other races (AOR = 1.22, 95% CI = 0.93–1.62). Low median household income and vacant housing rates were not significantly associated with the likelihood of adolescent substance use across all racial/ethnic groups. A sensitivity analysis using multiple imputation, instead of the listwise deletion method, also found that higher rates of alcohol and drug-related deaths were significantly associated a greater likelihood of substance use among adolescents, particularly among Black adolescents.

Table 4.

Results of Multilevel Models Assessing the Association Between Community Disadvantage and Adolescent Substance Use (N = 131,892).

	NH White (N = 61,209)		NH Black (N = 29,673)		Hispanic (N = 32,650)		NH Other (N = 8,360)
Variables	OR	95% CI	OR	95% CI	OR	95% CI	OR	95% CI
Age	1.27***	[1.16, 1.39]	1.07	[0.90, 1.26]	1.23**	[1.07, 1.42]	1.44**	[1.15, 1.80]
Male (reference: female)	1.34***	[1.23, 1.45]	1.25**	[1.07, 1.46]	1.55***	[1.36, 1.77]	0.97	[0.80, 1.20]
Number of alleged abuse/neglect types	1.47***	[1.37, 1.59]	1.38**	[1.19, 1.58]	1.84***	[1.66, 2.05]	1.48***	[1.24, 1.77]
Caregiver drug use (reference: no)	4.55***	[4.09, 5.06]	5.00***	[4.09, 6.08]	5.01***	[4.18, 6.01]	9.43***	[7.30, 12.16]
Caregiver alcohol use (reference: no)	1.73***	[1.47, 2.03]	1.41	[0.99, 1.99]	1.15	[0.87, 1.52]	2.64***	[1.91, 3.65]
County-level alcohol and drug mortality rates	1.33*	[1.07, 1.65]	1.42**	[1.13, 1.78]	1.40*	[1.09, 1.81]	1.22	[0.93, 1.62]
County-level vacant housing rates	1.01	[0.86, 1.20]	0.97	[0.78, 1.20]	1.00	[0.82, 1.22]	1.23	[0.96, 1.56]
County-level low household income	1.03	[0.78, 1.38]	1.23	[0.91, 1.66]	1.09	[0.81, 1.45]	1.04	[0.72, 1.51]
Urbanicity (reference: metro areas with over 1 million population)
Metro areas with ≤ 1 million population	1.06	[0.60, 1.87]	0.85	[0.47, 1.53]	1.10	[0.60, 1.99]	1.57	[0.75, 3.26]
Nonmetro areas	1.31	[0.63, 2.73]	0.55	[0.20, 1.49]	2.26	[0.99, 5.18]	1.22	[0.44, 3.38]

p < .05. **p < .01. ***p < .001.

Discussion

This study extends existing literature by examining community factors associated with adolescent substance use among those referred to CPS for alleged maltreatment. Our findings, aligned with the SMT framework (Khantzian, 1997), showed that residing in communities with higher county-level alcohol and drug mortality rates is associated with an increased risk of substance use. The association remains significant even after accounting for individual-level sociodemographic and maltreatment history factors. Notably, the association is significantly stronger among non-Hispanic Black adolescents compared to non-Hispanic white adolescents. The findings align with other studies demonstrating stronger associations between community characteristics and health behaviors among historically marginalized racial-ethnic individuals compared to non-Hispanic white individuals (Alegría et al., 2014; Karriker-Jaffe et al., 2016; Rosenblatt et al., 2021).

The observed racial/ethnic differences in community effects on health behaviors may be attributed to the heightened vulnerability of historically marginalized racial-ethnic adolescents to community-level substance exposure, driven by discrimination and systematic racism-related stressors they often face in their daily lives (Brody et al., 2012; Jelsma & Varner, 2020). Living in communities with high alcohol and drug mortality rates likely results in increased substance exposure and cumulative strain from substance use within the community. This, in turn, could lead historically marginalized racial-ethnic adolescents, who may lack well-developed self-regulation or coping skills, to turn to substance use as a maladaptive coping mechanism (Cook et al., 2005). This is supported by recent findings indicating that adolescents in foster care exhibit greater emotional vulnerability and tend to use coping strategies such as emotional outbursts, social avoidance, and withdrawal in comparison to their non-maltreated peers (Delaville & Pennequin, 2020). In addition, it is possible that professionals are more inclined to have an in-depth investigation of adolescent substance use in communities with higher rates of substance use, perceiving it as a threat to child safety (Ghertner et al., 2018).

Conversely, vacant housing rates and median household income at the county level were not significantly associated with adolescent substance use in our sample. This contradicts with some recent findings, such as Shah and Watson’s (2020) study of adolescents in Pennsylvania from 2009 to 2017, which found that higher county-level median household income and education level were protective factors against adolescent substance use. It is possible that economic conditions vary within the county because the county is too large to be considered as a single community. Given the influence of community economic conditions on investments in social infrastructure and human services, as well as social norms regarding substances within communities (Brooks-Gunn et al., 1993), smaller geographic units like census tracts and block groups might better capture community-level variations.

The finding that county-level alcohol and drug mortality rates were significantly higher in communities where non-Hispanic white adolescents lived compared to adolescents of other racial/ethnic groups aligns with past research, which has shown that white patients tend to receive more prescription opioids and be more vulnerable to exposures to substances than Black patients (Harrison et al., 2018). For instance, Townsend et al. (2022) investigated the effects of prescription drug monitoring program (PDMP) laws on opioid dispensing to Black and white patients aged 18–64 years in the United States. They found no clear evidence that PDMP laws reduced opioid dispensing more among Black patients. They, however, found that opioid dispensing laws appeared to reduce high-dose opioid receipt rates more among white patients than Black patients (Townsend et al., 2022). Additional research is needed to understand racial differences in substance exposures among adolescents involved in the child welfare system in the United States.

Implications for Practice

Our findings offer crucial insights with direct implications for social work practice. Adolescents in our sample had allegedly experienced child maltreatment by a caregiver or other trusted adult, leading to their referral to CPS. Due to the importance of community disadvantage in substance use among historically marginalized racial-ethnic adolescents, who often encounter discrimination-related stressors (Brody et al., 2012; Jelsma & Varner, 2020), child and family social workers need to proactively assess adolescents for community disadvantages and possible substance use during investigations. Expanding child welfare services in communities with heightened substance exposure is essential to assist adolescents and their parents in developing more adaptive coping mechanisms.

In addition to child-level prevention and early intervention efforts, the presence of community factors (e.g., high substance mortality rates within communities), as well as family factors (e.g., parental substance use), that increase the risk of adolescent substance use highlight a need for multilevel prevention programs. In addition to alcohol and drug mortality rates, our findings reveal that adolescents with substance-using caregivers are more likely to report substance use, after accounting for individual- and community-level factors. A substantial body of literature consistently demonstrates the intergenerational transmission of substance use (Haggerty & Carlini, 2020; Neppl et al., 2020). Parents often serve as role models for coping strategies, whether adaptive or maladaptive (Liga et al., 2020). Adolescents with substance-using parents may, in turn, develop similar coping strategies as they transition into adolescence. Therefore, family-based and community-based interventions that include universal and targeted approaches with collaborative efforts spanning various sectors, including schools, families, mass media, and social work community organizations, are needed to effectively reach the adolescents most in need and address child, family, and community factors. To effectively implement these interventions, it may also be necessary to address treatment barriers, including stigma and misunderstanding of certain substance use treatments among child welfare professionals (Radel et al., 2018). Indeed, while substance use among adolescents in child welfare is understudied, several researchers have focused on stigma of child welfare-involved parents with substance use (Kenny & Barrington, 2018; Olsen, 2015; Wolfson et al., 2021). As such, strategies such as training, trauma-informed practice, and cross-sector collaboration may help meet needs of individuals with substance use who are involved in child welfare (Wolfson et al., 2021).

Limitations

Many of the limitations of the current study are inherent to administrative data. First, our study used a cross-sectional design; thus, we cannot determine the directionality of the findings. Second, we were limited by cases that had valid data on adolescent substance use and residential county. As a result, findings are not necessarily reflective of the entire country but rather states that reported this information frequently. Third, as county identification was limited to counties with at least 1,000 referrals per year, we limited our analyses to referrals in counties with a unique identifier, most of which were urban or suburban counties. Fourth, we used data that are nearly five years old, including NCANDS data from 2019, WONDER from 2016 to 2018, and the American Community Survey from 2015 to 2019. These datasets may not fully capture recent shifts in policy and substance use trends. For example, our analysis included alcohol and drug mortality rates measured from 2019, which may not reflect recent trends in adolescent substance use. Future research should examine these trends within the context of evolving policies, the substance use landscape, and the impact of COVID-19. Fifth, we included three measures of community disadvantage as exposures, alcohol and drug mortality rates, vacant housing rates, and low median household income. Adolescent substance use is influenced by a wide range of community environments, including community crime rates, physical disorder, opioid dispensing environments, and densities of alcohol outlets. Future research needs to incorporate these additional factors to provide a more comprehensive understanding of the community influences on adolescent substance use. Finally, all the variables were based on caseworker report; therefore, receipt of public assistance, parental history of substance use, and adolescent substance use may have varied based on information available to the caseworker at referral and variations in reporting by county and state.

Our findings and limitations highlight several opportunities for future social work research to understand the relationships between family-based maltreatment, community adversity, and adolescent substance use. As our analyses were limited to counties with at least 1,000 child referrals to CPS, our results are not necessarily reflective of rural counties. While the opioid crisis initially had a particularly detrimental impact on rural America, recent trends indicate that overdose fatalities disproportionately affect highly segregated Black and urban communities. Considering these shifts and our use of older data, future research could compare rural, urban, and suburban counties to example changes in overdose fatalities and substance use trends over time, focusing on different demographic and geographic contexts (Glick et al., 2020). In addition, future research could replicate the study in counties that did not report adolescent substance use data to NCANDS; this could allow us to compare findings across the United States. Finally, future research could explore three-way interactions among child, parent, and community factors to identify risk and protective factors that may impact the risk of adolescent substance use.

Conclusion

This study provides an initial glimpse at the relationship between community factors and adolescent substance use among children who have experienced alleged child maltreatment. Although child welfare systems often focus on individual or family risk and protective factors, our findings highlight the need to consider community-level disadvantage that may increase the risk of substance use. As such, it is important to take a multilevel approach in research and prevention/early intervention programs to identify and address the impact of families and communities on adolescent substance use.

Footnotes

Disposition editor: Cristina Mogro-Wilson

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research project is supported by an Interdisciplinary Research Program grant from the Office of the Vice President Research at the University of Texas at Arlington. The content is solely the responsibility of the authors and does not necessarily represent the official views of the University of Texas at Arlington.

ORCID iDs

Yeonwoo Kim

Catherine LaBrenz

References

Aarons

G. A.

Hazen

A. L.

Leslie

L. K.

Hough

R. L.

Monn

A. R.

Connelly

C. D.

Landsverk

J. A.

Brown

S. A.

(2008). Substance involvement among youths in child welfare: The role of common and unique risk factors. American Journal of Orthopsychiatry, 78(3), 340–349.

Alcover

K. C.

Thompson

C. L.

(2020). Patterns of mean age at drug use initiation among adolescents and emerging adults, 2004-2017. JAMA Pediatrics, 174(7), 725–727.

Alegría

Molina

K. M.

Chen

C. N.

(2014). Neighborhood characteristics and differential risk for depressive and anxiety disorders across racial/ethnic groups in the United States. Depression and Anxiety, 31(1), 27–37.

Ammerman

B. A.

Steinberg

McCloskey

M. S.

(2018). Risk-taking behavior and suicidality: The unique role of adolescent drug use. Journal of Clinical Child & Adolescent Psychology, 47(1), 131–141.

Asselmann

Wittchen

H.-U.

Lieb

Höfler

Beesdo-Baum

(2014). Associations of fearful spells and panic attacks with incident anxiety, depressive, and substance use disorders: A 10-year prospective-longitudinal community study of adolescents and young adults. Journal of Psychiatric Research, 55, 8–14.

Austin

A. E.

Shanahan

M. E.

(2018). Association of childhood abuse and neglect with prescription opioid misuse: Examination of mediation by adolescent depressive symptoms and pain. Children and Youth Services Review, 86, 84–93.

Benedini

K. M.

Fagan

A. A.

(2020). From child maltreatment to adolescent substance use: Different pathways for males and females? Feminist Criminology, 15(2), 147–173.

Brody

G. H.

Kogan

S. M.

Chen

Y.-f.

(2012). Perceived discrimination and longitudinal increases in adolescent substance use: Gender differences and mediational pathways. American Journal of Public Health, 102(5), 1006–1011.

Brooks-Gunn

Duncan

G. J.

Klebanov

P. K.

Sealand

(1993). Do neighborhoods influence child and adolescent development? American Journal of Sociology, 99(2), 353–395.

10.

Cardoso

J. B.

(2018). Running to stand still: Trauma symptoms, coping strategies, and substance use behaviors in unaccompanied migrant youth. Children and Youth Services Review, 92, 143–152.

11.

Chen

Lasopa

S. O.

Cottler

L. B.

(2017). Current patterns of marijuana use initiation by age among US adolescents and emerging adults: Implications for intervention. The American Journal of Drug and Alcohol Abuse, 43(3), 261–270.

12.

Children’s Bureau. (2021). Child maltreatment 2019. https://www.acf.hhs.gov/cb/data-research/child-maltreatment

13.

Cicchetti

Handley

E. D.

(2019). Child maltreatment and the development of substance use and disorder. Neurobiology of Stress, 10, 100144.

14.

Cook

Spinazzola

Ford

Lanktree

Blaustein

Cloitre

DeRosa

Hubbard

Kagan

Liautaud

(2005). Complex trauma in children and adolescents. Psychiatric Annals, 35(5), 390–398.

15.

Coulton

C. J.

Crampton

D. S.

Irwin

Spilsbury

J. C.

Korbin

J. E.

(2007). How neighborhoods influence child maltreatment: A review of the literature and alternative pathways. Child Abuse & Neglect, 31(11–12), 1117–1142.

16.

Delaville

Pennequin

(2020). Foster placement disruptions in France: Effects on children and adolescents’ emotional regulation. Child and Adolescent Social Work Journal, 37, 527–536.

17.

DeLisi

Vaughn

M. G.

Salas-Wright

C. P.

(2015). Rumble: Prevalence and correlates of group fighting among adolescents in the United States. Behavioral Sciences, 5(2), 214–229.

18.

Duprey

E. B.

Oshri

Caughy

M. O.

(2017). Childhood neglect, internalizing symptoms and adolescent substance use: Does the neighborhood context matter? Journal of Youth and Adolescence, 46(7), 1582–1597.

19.

Eitle

T. M.

Eitle

(2014). Race, coping strategies, and substance use behaviors: A preliminary analysis examining White and American Indian adolescents. Substance Use & Misuse, 49(3), 315–325.

20.

Fuhrmann

Knoll

L. J.

Blakemore

S.-J.

(2015). Adolescence as a sensitive period of brain development. Trends in Cognitive Sciences, 19(10), 558–566.

21.

Ganzer

Bröning

Kraft

Sack

P.-M.

Thomasius

(2016). Weighing the evidence: A systematic review on long-term neurocognitive effects of cannabis use in abstinent adolescents and adults. Neuropsychology Review, 26(2), 186–222.

22.

Ghertner

Waters

Radel

Crouse

. (2018). The role of substance use in child welfare caseloads. Children and Youth Services Review, 90, 83–93.

23.

Glick

J. E.

McHale

S. M.

King

(2020). Rural families and communities in the United States: Facing challenges and leveraging opportunities (Vol. 10). Springer Nature.

24.

Guibord

Bell

Romano

Rouillard

(2011). Risk and protective factors for depression and substance use in an adolescent child welfare sample. Children and Youth Services Review, 33(11), 2127–2137.

25.

Hagborg

J. M.

Thorvaldsson

Fahlke

(2020). Child maltreatment and substance-use-related negative consequences: Longitudinal trajectories from early to mid adolescence. Addictive Behaviors, 106, 106365.

26.

Haggerty

K. P.

Carlini

B. H.

(2020). Understanding the intergenerational transmission of substance use and problem behavior: Implications for future research and preventive interventions. Psychology of Addictive Behaviors, 34(8), 894–897.

27.

Hamburger

M. E.

Leeb

R. T.

Swahn

M. H.

(2008). Childhood maltreatment and early alcohol use among high-risk adolescents. Journal of Studies on Alcohol and Drugs, 69(2), 291–295.

28.

Handley

E. D.

Rogosch

F. A.

Guild

D. J.

Cicchetti

(2015). Neighborhood disadvantage and adolescent substance use disorder: The moderating role of maltreatment. Child Maltreatment, 20(3), 193–202.

29.

Harrison

J. M.

Lagisetty

Sites

B. D.

Guo

Davis

M. A.

(2018). Trends in prescription pain medication use by race/ethnicity among US adults with noncancer pain, 2000–2015. American Journal of Public Health, 108(6), 788–790.

30.

Hogarth

Martin

Seedat

(2019). Relationship between childhood abuse and substance misuse problems is mediated by substance use coping motives, in school attending South African adolescents. Drug and Alcohol Dependence, 194, 69–74.

31.

Hoots

B. E.

(2023). Alcohol and other substance use before and during the COVID-19 pandemic among high school students—Youth Risk Behavior Survey, United States, 2021. Morbidity and Mortality Weekly Report Supplements, 72(1), 84–92.

32.

Jackson

J. M.

Seth

DiClemente

R. J.

Lin

(2015). Association of depressive symptoms and substance use with risky sexual behavior and sexually transmitted infections among African American female adolescents seeking sexual health care. American Journal of Public Health, 105(10), 2137–2142.

33.

Jelsma

Varner

(2020). African American adolescent substance use: The roles of racial discrimination and peer pressure. Addictive Behaviors, 101, 106154.

34.

Karriker-Jaffe

K. J.

Liu

Johnson

R. M.

(2016). Racial/ethnic differences in associations between neighborhood socioeconomic status, distress, and smoking among US adults. Journal of Ethnicity in Substance Abuse, 15(1), 73–91.

35.

Kenny

K. S.

Barrington

(2018). “People just don’t look at you the same way”: Public stigma, private suffering and unmet social support needs among mothers who use drugs in the aftermath of child removal. Children and Youth Services Review, 86, 209–216.

36.

Khantzian

E. J.

(1997). The self-medication hypothesis of substance use disorders: A reconsideration and recent applications. Harvard Review of Psychiatry, 4(5), 231–244.

37.

Khubchandani

Price

J. H.

(2018). Violent behaviors, weapon carrying, and firearm homicide trends in African American adolescents, 2001–2015. Journal of Community Health, 43(5), 947–955.

38.

LaBrenz

C. A.

Ohri

Kim

Baiden

(2024). Adolescents with substance use referred to child protective services: Does substance use treatment prevent foster care entry? Child & Family Social Work, 29(2), 363–373.

39.

Liga

Inguglia

Gugliandolo

M. C.

Ingoglia

Costa

(2020). The socialization of coping strategies in adolescence: The modeling role of parents. Anxiety, Stress, & Coping, 33(1), 47–58.

40.

Liu

Oshri

Duprey

E. B.

(2018). Alcohol use and depressive symptoms among a nationally representative sample of youth investigated for maltreatment. Journal of Studies on Alcohol and Drugs, 79(3), 380–390.

41.

Maguire-Jack

Font

S. A.

(2017). Community and individual risk factors for physical child abuse and child neglect: Variations by poverty status. Child Maltreatment, 22(3), 215–226.

42.

Mannoh

Hussien

Commodore-Mensah

Michos

E. D.

(2021). Impact of social determinants of health on cardiovascular disease prevention. Current Opinion in Cardiology, 36(5), 572–579.

43.

Moise

Varma

Stewart

(2020). Interpersonal consequences predicting adolescent substance use: Examining the role of temptation coping, motivation, and impulsivity. Substance Use & Misuse, 55(10), 1640–1649.

44.

Moran

P. B.

Vuchinich

Hall

N. K.

(2004). Associations between types of maltreatment and substance use during adolescence. Child Abuse & Neglect, 28(5), 565–574.

45.

Neppl

T. K.

Diggs

O. N.

Cleveland

M. J.

(2020). The intergenerational transmission of harsh parenting, substance use, and emotional distress: Impact on the third-generation child. Psychology of Addictive Behaviors, 34(8), 852–863.

46.

Noonan

A. S.

Velasco-Mondragon

H. E.

Wagner

F. A.

(2016). Improving the health of African Americans in the USA: An overdue opportunity for social justice. Public Health Reviews, 37(1), 1–20.

47.

Olsen

(2015). Punishing parents: Child removal in the context of drug use. Drug and Alcohol Review, 34(1), 27–30.

48.

Radel

Baldwin

Crouse

Ghertner

Waters

(2018). Substance use, the opioid epidemic, and the child welfare system: Key findings from a mixed methods study [ASPE Research Brief] (pp. 1–9). Office of the Assistant Secretary for Planning and Evaluation.

49.

Rosenblatt

A. M.

Crews

D. C.

Powe

N. R.

Zonderman

A. B.

Evans

M. K.

Tuot

D. S.

(2021). Association between neighborhood social cohesion, awareness of chronic diseases, and participation in healthy behaviors in a community cohort. BMC Public Health, 21(1), 1–9.

50.

Scholes-Balog

K. E.

Hemphill

S. A.

Patton

G. C.

Toumbourou

J. W.

(2015). Relationships between substance use and depressive symptoms: A longitudinal study of Australian adolescents. The Journal of Early Adolescence, 35(4), 538–561.

51.

Shah

Watson

(2020). Relationship between substance use and socioeconomic variables in Pennsylvania adolescents: 2009–2017. Substance Use & Misuse, 55(11), 1856–1866.

52.

Shin

S. H.

McDonald

S. E.

Conley

(2018). Patterns of adverse childhood experiences and substance use among young adults: A latent class analysis. Addictive Behaviors, 78, 187–192.

53.

Singh

V.-A. S.

Thornton

Tonmyr

(2011). Determinants of substance abuse in a population of children and adolescents involved with the child welfare system. International Journal of Mental Health and Addiction, 9(4), 382–397.

54.

Smith

V. C.

Wilson

C. R.

Ryan

S. A.

Gonzalez

P. K.

Patrick

S. W.

Quigley

Siqueira

Walker

L. R.

(2016). Families affected by parental substance use. Pediatrics, 138(2), e20161575.

55.

Straussner

S. L.

Fewell

C. H.

(2018). A review of recent literature on the impact of parental substance use disorders on children and the provision of effective services. Current Opinion in Psychiatry, 31(4), 363–367.

56.

Thurston

Freisthler

Wolf

J. P.

(2021). Contrasting methods of measurement in spatial analyses examining the alcohol environment and child maltreatment. Child Maltreatment, 27(4), 515–526.

57.

Tonmyr

Thornton

Draca

Wekerle

(2010). A review of childhood maltreatment and adolescent substance use relationship. Current Psychiatry Reviews, 6(3), 223–234.

58.

Townsend

T. N.

Bohnert

A. S.

Lagisetty

Haffajee

R. L.

(2022). Did prescribing laws disproportionately affect opioid dispensing to Black patients? Health Services Research, 57(3), 482–496.

59.

Traube

D. E.

James

Zhang

Landsverk

(2012). A national study of risk and protective factors for substance use among youth in the child welfare system. Addictive Behaviors, 37(5), 641–650.

60.

U.S. Department of Health and Human Services. (2023). Preliminary FY¹ 2022 estimates as of May 9, 2023 [The AFCARS Report 30].

61.

Wolfson

Schmidt

R. A.

Stinson

Poole

(2021). Examining barriers to harm reduction and child welfare services for pregnant women and mothers who use substances using a stigma action framework. Health & Social Care in the Community, 29(3), 589–601.

62.

Yampolskaya

Chuang

Walker

(2019). Trajectories of substance use among child welfare-involved youth: Longitudinal associations with child maltreatment history and emotional/behavior problems. Substance Use & Misuse, 54(3), 437–448.