Employment Status and Substance Use: The Unique Risks of Underemployment

Background/Purpose: Employment insecurity is an important social determinant of health and is closely associated with substance use. Substance use is often shaped by stress and financial strain, making it relevant in the context of labor market disadvantage. While unemployment has been widely studied, underemployment remains less well understood. Defined as working below one’s full capacity, such as working fewer hours than desired or earning insufficient income, underemployment is often grouped with secure employment despite representing a distinct form of employment insecurity. Moreover, underemployment is often treated as a uniform category, with limited attention to distinctions between hour-based and income-based forms, which may reflect different dimensions of vulnerability. This study investigates how employment insecurity, including distinct types of underemployment, is associated with different substance use behaviors.

Methods: Using a nationally representative panel dataset with 23 biweekly surveys (N = 4,803), this study examined associations between employment status and multiple substance use outcomes among U.S. working-age adults (ages 18-64), including heavy drinking, cannabis use, vaping, and cigarette use. Substance use outcomes were dichotomized: cannabis, vaping, and cigarette use (any use vs. none), and heavy drinking (yes vs. no). Employment status was categorized as securely employed (reference group), exclusively hour-based underemployment (i.e., involuntary part-time work), exclusively income-based underemployment (i.e., earnings below 125% of the poverty threshold), multidimensional underemployment (i.e., both hour- and income-based), unemployed, and other (i.e., retired, not in labor force but not retired, unknown labor force status). Mixed-effects logistic regression models were used to estimate time-varying, concurrent, within-wave (cross-sectional) associations over time between employment status and each substance use outcomes. Models adjusted for age, gender, education, race/ethnicity, marital status, and survey wave.

Results: Participants were securely employed (33.6%), underemployed (hour-based 19.6%, income-based 7.6%, multidimensional 4.8%), unemployed (20.0%), and other (14.4%). Hour-based underemployment (vs. secure employment) was associated with higher odds of heavy drinking (OR = 1.40, 95% CI [1.25, 1.56]). For nicotine-related outcomes, income-based (vaping: OR = 1.36 [1.06, 1.74]) and multidimensional underemployment (vaping: OR = 1.37 [1.02, 1.83]) were associated with higher odds of use, as were unemployment (vaping: OR = 1.40 [1.12, 1.76]) and other statuses (vaping: OR = 1.51 [1.15, 2.00]; cigarette: OR = 1.37 [1.03, 1.81]). For cannabis use, unemployment (OR = 1.48 [1.24, 1.77]) and other statuses (OR = 1.69 [1.35, 2.12]) were associated with higher odds of use.

Conclusions/Implications: Findings suggest that different forms of employment insecurity are associated with distinct patterns of substance use behaviors. Income-based underemployment may increase reliance on nicotine use as a coping strategy, while hour-based underemployment may contribute to higher likelihood of heavy drinking. Focusing solely on unemployment may overlook individuals experiencing other forms of employment insecurity, including underemployment. Distinguishing between types of underemployment is important, as treating underemployment as a single category may obscure differences in substance use risk. Social work practitioners and policymakers should move beyond traditional dichotomous assessments of employment insecurity, when evaluating substance use risk. Policies addressing income insecurity and inadequate work hours among workers may help mitigate substance-related harm and reduce behavioral health disparities.