Adolescents were at-risk for poor mental health and substance use prior to the COVID-19 pandemic. Data from the 2019 Youth Risk Behavior Survey (YRBS), a cross-sectional, national survey of U.S. high-school students, revealed that approximately 37% experienced persistent feelings of sadness or hopelessness; 19% seriously considered attempting suicide; and 20% were bullied at school (Centers for Disease Control and Prevention, 2023). In addition, approximately 21.7% of students reported current marijuana use, 29.2% reported current alcohol use, and 7.2% reported prescription opioid misuse (Jones et al., 2020).

Pandemic-related disruptions to adolescents’ daily lives, including schooling, family dynamics, work, and social activities heightened this vulnerability. The disruptions adversely affected their health and well-being, with studies finding increases in the number of mentally unhealthy days and depression and anxiety symptoms (Hertz et al., 2022; Panchal et al., 2023; Temple et al., 2022; Wang et al., 2021). COVID-19 led to business closures and reduced hours, resulting in layoffs and increased unemployment (Smith et al., 2021). Adolescents were not immune to these job losses (Inanc, 2020). However, research describing the possible effects of job loss during the COVID-19 pandemic on adolescent health behaviors, including substance use, is limited. A systematic review examining substance use among youth during the COVID-19 pandemic revealed that most studies reported decreases in prevalence of substance use among adolescents. These decreases were attributed to reduced accessibility to substances and limited peer group interactions where substance use is likely to occur as a result of business closures, remote schooling, and restrictions to social interactions (Layman et al., 2022). Only a few of these studies considered the potential impact of parental job loss during the COVID-19 pandemic on adolescent substance use, and none of the studies examined adolescent job loss during the COVID-19 pandemic as a risk factor for substance use.

Pre-pandemic research has indicated a potential link between adolescent employment status and substance use (Lee et al., 2015; Osilla et al., 2013; Unger et al., 2004; Valois et al., 1999). Among a cohort of predominantly African-American students, researchers found that increasing time spent working for pay was associated with current tobacco use, and those students who worked more than 10 h per week were more likely to initiate tobacco use earlier than their peers (Ramchand et al., 2007). Other studies found that working more than 10 hours per week was predictive of heavy drinking among U.S. adolescents (Paschall et al., 2004), and working either full time or part time was associated with increased the odds of binge alcohol and heavy alcohol use among students aged 12–17 years (Wu et al., 2003). No study to date has examined the possible effects of adolescent job loss on substance use.

Several studies of adolescents have found associations between stress, including finance-related stress, and changes in substance use early in the COVID-19 pandemic including both increases and decreases in alcohol, cigarette and electronic vapor product (EVP), and drug use (Chaffee et al., 2024; Clendennen et al., 2021; Hoots et al., 2023; Kreslake et al., 2021; Layman et al., 2022; Temple et al., 2022). U.S. adolescents whose families experienced material hardship during May—August, 2020, including job loss of any household member, were more likely to report past 30-day substance use compared to pre-pandemic (during 2018, 2019, or January 2020) (Pelham et al., 2021). Sustained or increased marijuana, e-cigarette, and cigarette use attributed to COVID-19-related stressors was reported by more than 75% of participants of a 2020 study and higher levels of perceived stress were associated with a statistically significant increase in marijuana use (Clendennen et al., 2021). Negative changes in self-reported family financial situation from before to during the COVID-19 pandemic were associated with increased likelihood of alcohol consumption among Texas adolescents (Temple et al., 2022).

The Adolescent Behaviors and Experiences Survey (ABES) presents an opportunity to look at associations between student and parental pre-pandemic employment status and job loss during the COVID-19 pandemic and adolescent substance use behaviors, including cigarette smoking, EVP use, alcohol use, and drug use. ABES was conducted by the Centers for Disease Control and Prevention (CDC) from January through June 2021 to collect nationally representative data on the potential effects of the COVID-19 pandemic on U.S. high-school students’ health-related experiences and behaviors. Prior research using ABES data has examined student mental health, perceived racism, and other adverse experiences (Jones et al., 2023; Krause et al., 2022; Mpofu et al., 2022). Additionally, Brener et al. (2022) previously examined substance use through analyses of the ABES data but did not include the student or parental employment status variables in their study. The current study builds on previous work with the ABES data, and the results provide insight into the effects that job loss (both parental and adolescent) may have on adolescent health and well-being.

Among U.S. high-school students in 2021, 9.0% reported job loss and 59.8% reported not working before the COVID-19 pandemic (Table 2). Reported parental job loss during the COVID-19 pandemic was 27.4% and the reported parental unemployment prior to the COVID-19 pandemic was 3.7%. White, non-Hispanic students were most likely to report job loss during the COVID-19 pandemic (10.3%), while Black, non-Hispanic students were most likely to report parental job loss (5.0%). Student job loss increased with increasing grade level. Heterosexual (9.2%) and LGB (9.1%) students had similar job loss prevalences, although LGB students were less likely to be employed during the pandemic.

The most common substance use behavior among students was current alcohol use (19.5%) and the least common was current cigarette smoking (3.3%) (Table 3). Female students were significantly more likely than male students to report current alcohol use (22.4% vs. 16.4%) and current opioid prescription misuse (5.4% vs. 3.2%). White, non-Hispanic students were significantly more likely to report current EVP use (20.3%) compared to Black, non-Hispanic (10.6%); other or multi-race, non-Hispanic (12.3%); or Hispanic (9.7%) race and ethnicity groups. The prevalence for most substance use behaviors increased with increasing grade level. Current cigarette smoking and vaping were similar for 9th and 10th graders but were higher among 11th graders and the highest among 12th graders. The major exception to these trends was exception of current prescription opioid misuse, which was most common among 9th-grade students (5.3%) and lowest among 12th graders (3.6%), although confidence intervals for all grades overlapped. LGB students reported higher prevalences for each substance use behavior than heterosexual and other or questioning students. For the drug use measures, the prevalences for LGB students were significantly higher than those for heterosexual students.

Student pre-pandemic employment status and job loss during the pandemic were associated with differences in use of most substances except for current cigarette smoking and current opioid misuse (Table 4). Among students who reported job loss during the pandemic, the unadjusted prevalences for all substance use behaviors were higher than among students who did not lose their job during the pandemic and students who had no job prior to the pandemic (Table 4). The prevalences for current EVP use (29.9%), drinking more alcohol during the pandemic (25.3%), current marijuana use (25.8%), and using more drugs during the pandemic were significantly elevated among students who reported job loss compared to students who did not lose their job during the pandemic and students who had no job prior to the pandemic. Similar to the prevalences, the aPRs for students reporting job loss during the pandemic were significantly elevated compared to students who did not lose their job during the pandemic for current EVP use (1.43; 95% CI: 1.15–1.76), current alcohol use (1.27; 95% CI: 1.05–1.55), drinking more alcohol during the pandemic (1.43; 95% CI: 1.14–1.80), current marijuana use (1.51; 95% CI: 1.20–1.90), and using more drugs during the pandemic (1.65; 95% CI: 1.30–2.09). Students who had no paying job reported significantly less substance use than students who did not lose a job during the pandemic except current opioid misuse, with aPRs below 1.0 for almost every substance use behavior.

Parental employment experience was also associated with differences in prevalence of student substance use measures. Students who reported their parent had no job prior to the pandemic had the highest unadjusted prevalences for current cigarette smoking (6.3%), current EVP use (19.1%), drinking more alcohol during the pandemic (21.4%), current prescription opioid misuse (9.5%), and using more drugs during the pandemic (22.3%) (Table 5). In addition, the aPRs for students whose parents had no job prior to the pandemic were significantly elevated for each of the substance use behaviors except for current alcohol and marijuana use; the highest aPRs were current prescription opioid misuse (2.79; 95% CI: 1.61–4.84) and current cigarette smoking (2.44; 95% CI: 1.22–4.89) in comparison with those students who parents did not lose a job during the pandemic.

Students who reported their parent lost their job during the pandemic had the highest prevalence for current alcohol use (24%) and current marijuana use (15.5%) (Table 5). The aPRs were significantly elevated for each substance use behavior among students whose parent lost a job compared to those who retained their jobs, with current cigarette smoking (1.63; 95% CI: 1.08–2.46); current EVP use (1.46; 95% CI: 1.23–1.74); current prescription opioid misuse (1.45; 95% CI: 1.06–1.99); and current marijuana use (1.44; 95% CI: 1.24–1.67) having the highest aPRs.

For the parent/student employment dyads, students reporting both parent and student job loss during the pandemic had the highest prevalences for current EVP use (34.2%), current alcohol use (36.2%), and current marijuana use (27.7%) (Table 6). Students who reported they lost their job but their parent did not lose a job had the highest prevalences of drinking more alcohol (26.0%) and using more drugs (26.8%) during the pandemic. Students who had no job prior to the pandemic and whose parent retained their job had the lowest prevalences for every substance use variable.

The aPRs for every substance use variable except current prescription opioid use were significantly elevated for students reporting both they and their parent lost jobs in comparison with the reference group of students who reported both they and their parent retained their jobs (Table 6). For dyads in which both parent and student both lost jobs, the highest aPR was for current EVP use (1.95; 95% CI: 1.49–2.56). For dyads in which the parent lost their job, but the student retained their job, aPRs for each substance use measure except increased alcohol use during the pandemic were significantly elevated. The highest aPR in these dyads was for current cigarette smoking (2.10; 95% CI: 1.15–3.81). The lowest levels of substance use were reported by students who had no job prior to the pandemic and whose parent retained their job; aPRs for each substance use measure were below 1.0 for this group.

Both student and parental job loss during the pandemic, as well as parental unemployment prior to the pandemic, were associated with elevated levels of reported substance use. When student and parental employment status were considered together, dyads in which both lost jobs and those in which the parent lost the job and the student retained the job had the highest prevalences for multiple substance use metrics: current cigarette smoking, EVP use, alcohol use, and marijuana use. Substance use was also elevated, although to a lesser extent, among dyads in which the parent was unemployed before the pandemic and those in which the parent retained their job but the student lost their job. While, in general, parental job loss had the larger effect on most substance use measures, student job loss was also associated with increased substance use. Parental and student job loss during the first 18 months of the COVID-19 pandemic both appear to have been risk factors for adolescent substance use.

Current cigarette smoking and EVP use were elevated among those students reporting parental unemployment prior to the pandemic or parental job loss (with or without student job loss) during the pandemic. Stress, anxiety, and other mental health outcomes have been previously associated with increased use or risk of cigarette smoking and EVP use among adolescents (Becker et al., 2021; Kwon et al., 2018; Low et al., 2012). A study of Southern California middle-school students found that the specific stressor of job loss for a household member was a statistically significant risk factor for cigarette smoking (Unger et al., 2004).

Lower socioeconomic status (SES) has also been associated with higher usage rates of e-cigarettes, blunts (cigars filled with marijuana), and cigarettes among high-school students in a previous longitudinal study (Simon et al., 2017). In addition to stress and SES, having at least one parent who smokes has been found to substantially increase the risk that their children will become smokers or EVP users (Han & Son, 2022; Leonardi-Bee et al., 2011; Peterson et al., 2006). Vulnerable adults, including those who were economically disadvantaged or unemployed during the COVID-19 pandemic, were found to increase their cigarettes per day compared to pre-pandemic (Wiley et al., 2023). The elevated cigarette and EVP use observed among students in the current study whose parents were unemployed before or lost jobs during the pandemic may reflect the context of increased parental smoking.

The prevalence and changes in prevalence of drug use varied by substance but was generally lowest among dyads in which the parent and student did not lose jobs during pandemic (including dyads in which the student had no job before the pandemic). The highest prevalence of increased drug use during the COVID-19 pandemic was reported by students who lost a job but whose parent retained their job but was also high among dyads where the parent was unemployed or both student and parent lost jobs. Prevalence of current marijuana was the highest among the last group. The prevalence of current prescription opioid misuse was the highest among students whose parents were unemployed before the pandemic, followed by dyads in which the parent but not the student lost a job. As with cigarette and EVP use, family SES measures have been examined in conjunction with adolescent drug use. Lower parental education has been linked to increased likelihood of adolescent drug use compared to higher parental education (Bello et al., 2019; Peters et al., 2018). Other U.S. studies have also found that SES based on multiple indicators (parental or family income, parental occupation, parental education) is associated with adolescent substance use, with lower SES associated with increased cannabis consumption (Daniel et al., 2009; Duncan et al., 1997; Jeynes, 2002). Few, if any of these studies, considered employment status or job loss of the parent or the adolescent. Familial conflict as defined by the Family Environment Scale was significantly associated with substance use disorders in a longitudinal study (Skeer et al., 2009). Parental job loss during the COVID-19 pandemic has previously been linked to increased parent–child conflict, particularly in families that are more economically vulnerable (Kalil et al., 2020; Wang et al., 2021; 2024;).

Our study contributes the new finding that student drug use was elevated among students who lost a job while their parent retained their job. This finding may reflect conditions specific to the COVID-19 pandemic and these dyads, such as more unsupervised time alone, while the parent was working due to school closures and student job loss; parental monitoring has been noted as a preventative factor in adolescent substance use in other research (Kuntsche & Kuntsche, 2016; Pelham et al., 2024). Ultimately, there is likely a complex, multi-factorial relationship between adolescent substance use and student job loss, parental unemployment or job loss, SES, time available to parents to supervise children, family economic security, parent–child conflict, and other variables not measured in this dataset. Exploration of emergent and increasing drug use in the context of a longitudinal dataset with more information about these topics, during a less economically chaotic time period, could help elucidate the mechanisms behind our findings.

Current alcohol use and drinking more during the pandemic were associated with both student and parental job loss in our study. These findings are consistent with a recent COVID-19 related longitudinal study of Texas adolescents, which found that family financial instability, including job loss of an immediate family member, was linked to increases in poor mental health symptoms and alcohol use among students (Temple et al., 2022). Among a longitudinal sample of U.S. adults, job loss was associated with increased alcohol consumption, with men and women both reporting significant increases in daily alcohol consumption (Popovici & French, 2013). The same study found that both unemployed men and women reported more binge drinking days and increased odds of being diagnosed with an alcohol abuse or dependence condition (Popovici & French, 2013); similarly, parental unemployment was associated with increased alcohol use during the COVID-19 pandemic (though not current alcohol use) by students in our study.

In this study, parental job loss and unemployment, as well as student job loss, were associated with elevated prevalences of substance use, as well as reported increased use during the COVID-19 pandemic. This finding may indicate that stress among the adolescents who experience job loss leads to an increase in substance use, similar to studies that linked unemployment and job loss in young adults aged 22–33 years to increased substance use (Lee et al., 2015). The findings from the current study are also reflective of the disruptions of the early COVID-19 pandemic, a unique and uncertain time. Adolescents dealt with disruptions to school, minimized peer group social interactions, and changes to family dynamics. Examining parental and student job loss in conjunction with substance use outside the context of a public health emergency could help elucidate the effects of both parental and student employment status and job loss. If findings are similar to those during the COVID-19 pandemic, increasing awareness of school and community health professionals about the effects of job loss could improve the direction of mental health and economic resources to families experiencing these job losses. School-based prevention and intervention programs have been effective at reducing substance use among adolescents (Das et al., 2016) and inclusion of job loss support may improve this effectiveness. Additional resources may help reduce adolescent substance use.