Psychological distress and unmet mental health needs of low-income food assistance recipients who smoke Authors: Erin S. Rogers, Ashley Crawford, Peter Gado, Wendy Lliguichuzhca, George Novillo, Reynalry Rozon, Charlie Thao Published: 2026 ABSTRACT Food assistance recipients have high smoking rates. Mental health factors that may affect their smoking are understudied. This study examined psychological distress and mental health service use among food assistance recipients enrolled in a smoking cessation trial. The trial sample comprised N = 352 adults aged ≥18, earning below 200% of the federal poverty level, smoking ≥5 cigarettes daily, and receiving public food assistance (e.g. SNAP). Psychological distress was assessed via the Kessler-6. Recent mental health service use (yes/no), therapeutic satisfaction (scale of 1–10), and potential usefulness of mental health services (scale of 1–10) were also assessed. Overall, 23% of participants had severe distress, 53% moderate distress, and 24% no/mild distress. Severe psychological distress was associated with younger age, high levels of financial stress, and food insecurity, but not with tobacco behaviors or motivations. Recent mental health service use was more common among individuals with severe distress, while therapeutic satisfaction was high across all distress levels. Participants who had not received mental health services reported a high level of potential usefulness if they were to receive services, especially among participants with severe distress. Results suggest that food assistance recipients who smoke may have unmet mental health care needs. Introduction Tobacco use is a leading preventable cause of morbidity and mortality in the United States (U.S., 2014).1 Although the rates of tobacco use have declined significantly in the general population, tobacco use remains high among adults experiencing food insecurity, including the 42 million Americans enrolled in the Supplemental Nutrition Assistance Program (SNAP) program (Adams, 2018; Kim-Mozeleski et al., 2020, 2022, 2024). In recognizing the high rates of tobacco use among SNAP participants, public health researchers and practitioners have recommended integrating smoking cessation services and food assistance programming to provide an accessible and synergistic point of entry into smoking cessation treatment for food assistance recipients (Adams, 2018; Kim-Mozeleski et al., 2020). However, limited research has focused on understanding the needs and barriers to smoking cessation among people enrolled in food assistance programs. Food insecurity is strongly and positively associated with psychological distress, and this relationship appears to be bidirectional in nature (Myers, 2020). Food insecurity can increase psychological distress through mechanisms of hunger, nutritional deficiencies, sleep disturbances, and shame or frustration at being unable to obtain sufficient food (Brady et al., 2023; Kim-Mozeleski et al., 2020). Psychological distress may in-turn contribute to food insecurity by increasing social isolation and withdrawal, increasing anhedonia or lethargy that reduces motivation and energy to purchase and prepare food, and limiting workforce participation and income to afford sufficient food (Myers, 2020). There are also broader structural factors that place individuals at increased risk for both food insecurity and psychological distress, including structural racism, sexism, ableism, and immigration policies that contribute to concentrated poverty and reduced access to food, health insurance, and mental health care (Myers, 2020). Psychological distress is also a well-documented risk factor for cigarette smoking and relapse (Alpers et al., 2024; Schroeder & Morris, 2010; Sinha, 2008; Zvolensky et al., 2018). Previous research has demonstrated that there is a greater and increasing prevalence of serious psychological distress among people who smoke (Zvolensky et al., 2018). Of relevance for food assistance programs, research demonstrates bidirectional causal relationships between psychological distress, food insecurity, and smoking (Brady et al., 2023; Kim-Mozeleski et al., 2020). In 2020, Kim-Mozeleski and Pandey conducted a scoping review of the literature on the relationships between food insecurity and tobacco use, the results of which informed an intersecting theoretical model through which food insecurity, distress, and tobacco use are sequentially associated and reinforce one another (Kim-Mozeleski & Pandey, 2020). Their model proposes that individuals experiencing food insecurity may turn to smoking cigarettes to cope with distress, hunger, and broader social conditions (e.g. discrimination) (Brady et al., 2023; Kim-Mozeleski & Pandey, 2020; Kim-Mozeleski et al., 2020), while smoking can exacerbate food insecurity and distress through multiple pathways including financial strain (e.g. spending on cigarettes limits money available for food and other basic needs), hunger suppression, modified food acquisition practices (e.g. purchasing unhealthy food at retailers that sell cigarettes), and declining health that limits workforce participation and income (Brady et al., 2023; Schroeder & Morris, 2010). The above evidence and theory suggest that psychological distress may be a central and modifiable therapeutic target for helping people enrolled in food assistance programs quit smoking. However, the psychological health and mental health treatment experiences of food assistance recipients who smoke have been understudied. Specifically, to our knowledge, no published studies have simultaneously examined relationships between psychological distress, tobacco use behaviors, and mental health care utilization and experiences of food assistance recipients who smoke. To address this empirical gap and inform intervention approaches for food assistance recipients who smoke, this study explored psychological distress and mental health service utilization among public food assistance recipients enrolled in an active smoking cessation trial for adults with income below 200% of the federal poverty level. Materials and methods Data source and participants Data were collected from 2022–2025 as part of the baseline survey for a randomized controlled trial (RCT) testing a smoking cessation intervention to address financial stress as a barrier to smoking cessation among low-income adults living in New York City (NYC). The parent RCT was approved by NYU Langone Health Institutional Review Board (#i21–00986). Participants were recruited for the RCT using three methods: (1) Digital study ads were developed and deployed by BuildClinical, a technology company that uses data mining and machine-learning to run research study ads on major digital platforms including Facebook, Instagram, TikTok, Twitter, Snapchat, YouTube, and Google Search. Once a potential participant clicked on the ad, they were taken to a secure study landing page that provided additional information about the study, an eligibility screener, and a page to enter their contact information. A research assistant reached out via phone, text, and email to all potential participants who were eligible based on the online screener. Once contact was made, the research assistant provided more study information, re-screened for eligibility, and scheduled individuals for an in-person informed consent process. (2) We used the electronic health record (EHR) systems at two safety net health care systems – NYU Langone Health Brooklyn and NYC Health and Hospitals (H+H) Bellevue – to identify patients who had been screened to be currently smoking cigarettes in the prior 12 months. For patients with documented past-year cigarette use and who did not opt-out of research contact, the study team mailed a letter about the study and study contact information to either learn more about the study or opt-out of further study contact. For patients who did not opt-out of further contact, a research assistant called the patient to discuss the study and screen for eligibility. (3) NYU Langone Health patients could self-refer to the study using the iConnect, an online portal that patients can use to learn about research studies being conducted throughout the health care system and submit their contact information to study investigators via a HIPAA-compliant portal. A research assistant called self-referred patients to discuss the study and screen for eligibility. People were eligible for the RCT if they met the following eligibility criteria: 1) live in NYC, 2) aged ≥18 years, 3) smoke at least five cigarettes per day, 4) have an annual household income below 200% of the federal poverty level (FPL), 5) speak English or Spanish, and 6) manage their own funds. All participants completed an in-person informed consent process and signed an IRB-approved informed consent form prior to completing any study procedures. The current analysis includes N = 352 participants who reported on the parent RCT’s self-administered baseline survey that they were receiving any form of government food assistance benefits at study enrollment (e.g. Supplemental Nutrition Assistance Program). Measures Psychological distress Psychological distress was measured with the Kessler-6 scale (Chronbach’s α = 0.89) (Kessler et al., 2002, 2003; Prochaska et al., 2012; Umucu et al., 2022). The Kessler-6 uses 5-item Likert scales (1=none of the time to 5=all of the time) to measure the past 30-day frequency with which participants experienced six symptoms of psychological distress (sadness, nervousness, restlessness, hopelessness, effort, worthlessness). The Kessler-6 has undergone clinical validation studies showing very good concordance with clinical diagnoses of serious mental illness (e.g. past 12-month DSM-IV disorder with a GAF score <60) in the general population and demonstrating good sensitivity at discriminating clinical cases of serious mental illness from non-cases in the general population (0.86 area under the receiving operating characteristic curve) (Kessler et al., 2002, 2003; Umucu et al., 2022). Following the measure’s standardized scoring approach, a sum of the six items was calculated to create a total Kessler-6 score for each participant. The total scores were then used to categorize participants as having no/mild psychological distress (score 0–4), moderate distress (score 5–12), or severe distress (score ≥13). Mental health service use, satisfaction, and potential usefulness To assess recent mental health service utilization, the survey asked three questions developed by the study team: ‘Are you currently using or have recently (past 3 months) used mental health services?’ Participants who responded ‘yes’ were asked a follow-up question: ‘On a scale of 1–10, how satisfied are you with your therapist/clinician (1 being not satisfied and 10 being extremely satisfied)?’ Participants who responded ‘no’ to the past 3-month mental health service utilization question were asked a follow-up question: ‘If you are not currently receiving counseling or therapy, on a scale of 1–10, how helpful or useful do you think it would be for you (1 being not at all useful and 10 being extremely useful)?’ Tobacco use characteristics, motivations, and spending The survey used questions from the PhenXToolkit to assess participants’ tobacco use characteristics, including the number of cigarettes smoked per day (open text box) and their current levels of motivation and confidence to quit smoking using 0–10 scales (0=no motivation/confidence to 10=highest level of motivation/confidence) (Hamilton et al., 2011). An item from the Fagerstrom Test for Nicotine Dependence asked participants ‘How soon after waiting do you smoke your first cigarette?’ (response options: within 5 minutes, 5–30 minutes, 31–60 minutes, >60 minutes). Lastly, the survey assessed tobacco expenditures with three questions developed by the study team. Participants were first asked ‘When you buy cigarettes, do you typically buy them by the pack or individual cigarettes (loosies)?’ Follow-up questions were: ‘When you buy by the pack, how much money do you typically spend on one pack of cigarettes?’ and ‘When you buy by the cigarette, how much money do you typically spend on one cigarette?’ We estimated each participant’s average daily tobacco spending by multiplying the number of cigarettes they smoke per day by the typical amount they spend on one cigarette (either price-per-loosie or price-per-pack divided by 20 cigarettes, depending on the participant’s most common purchasing approach). Financial stress The survey assessed financial stress with the 8-item InCharge Financial Distress/Financial Well-being Scale (IFDWS; Cronbach’s α = 0.96) (Prawitz et al., 2006). The IFDWS measures financial distress/well-being in eight domains that examine the level or frequency with which participants experience different components of financial stress. Example questions include: (1) What do you feel is the level of your financial stress today? and (2) How often do you worry about being able to meet normal monthly living expenses? For each question, participants respond on a scale from 1 to 10 (e.g. 1=no stress at all/never worry to 10=overwhelming stress/worry all the time). In accordance with IFDWS guidance (Prawitz et al., 2006), we created eight binary indicators (yes/no) categorizing participants with domain scores ≥7 as having high a level of stress in the given domain. One domain (confidence in affording a $1,000 emergency) was reverse-coded (1=low confidence to 10=high confidence), and a score ≤4 categorized as little to no confidence. Smoking-induced deprivation Smoking-induced deprivation was assessed with a binary (Yes/No) question adapted from the literature (Siahpush et al., 2012): ‘In the last 30 days, has there been a time when the money you spent on cigarettes resulted in not having enough money for any of these items: housing, food, household utilities, health care, transportation, and necessary clothing?’ Food insecurity Past-year food insecurity was assessed with the United States Department of Agriculture 6-item food security module (Chronbach’s α >0.80) (Blumberg et al., 1999; Hamilton et al., 1997). A sum of affirmative responses to the six items was calculated to create a total food insecurity score for each participant. The total scores were then used to categorize participants as having no/low food insecurity (score 0–1), moderate food insecurity (score 2–4), or severe food insecurity (score 5–6). Sociodemographics The survey adapted questions from the Behavior Risk Factor Surveillance System (BRFSS) to assess participants’ age (open text box), gender (cisgender female, cisgender male, transgender female, transgender male, non-binary/genderqueer, none of the above or prefers a different description), race/ethnicity/heritage (Hispanic or Latino/a/x, Middle Eastern or North African, Black or African American, American Indian or Alaskan Native, East Asian, South Asian, Southeast Asian, Central Asian, Native Hawaiian/Pacific Islander, White or European, Other), health insurance (check all that apply: Medicaid, Medicare, Private, other form of insurance, I do not have health insurance), highest level of education (8th grade or less, some high school, high school graduate/GED, Associate’s degree, some college, 4-year college graduate or higher, other education program such as trade school), marital status (married or living with partner, separated, divorced or widowed, never married and not living with a partner, other status), and current employment status (yes, no). Participants also reported their total annual household income by responding to the question ‘In the past 12 months, what is your best estimate of the total income of all members in your household before taxes?’ (Hamilton et al., 2011) Analysis All analyses were conducted using IBM SPSS software Version 29. Descriptive statistics (frequencies, means, standard deviations) were used to summarize variables. Bivariate analyses (χ2 tests, ANOVA tests) were first computed to compare participants with no/mild, moderate, and severe psychological distress on their sociodemographic characteristics (e.g. average age, race/ethnicity, health insurance), tobacco use characteristics (e.g. average cigarettes smoked per day, average motivation to quit), prevalence of experiencing high financial stress in each of the eight IFDWS domains, prevalence of past 30-day smoking-induced deprivation, prevalence of past-year food insecurity (no/low, moderate, severe), prevalence of past 3-monthmental health service use, average mental health service satisfaction among participants with past 30-day use, and average perceived potential usefulness of mental health services among participants without past 3-month use. For the subsample of participants with moderate or severe psychological distress, multivariable logistic regression was used to estimate socioeconomic factors that were independently associated with our binary (yes/no) outcome variable of past 3-month mental health service use, and multivariable linear regression was used to estimate socioeconomic factors that were independently associated with mental health service satisfaction ratingsand potential mental health service usefulness ratings. Sociodemographic factors included in the logistic and linear regression models were: age (ref: 18–50 vs. >50), gender (ref: cisgender female vs. cisgender male), race/ethnicity (ref: Hispanic/Latino/a/x vs. Non-Hispanic Black/African American, Non-Hispanic White), current employment status (ref: unemployed vs. employed), highest level of education (ref: high school/GED or less vs. >high school/GED), insurance (ref: Medicaid only vs. all other insurance situations), and whether the participant reported a level of stress about personal finances (ref: No vs Yes). Multicollinearity diagnostics showed that variance inflation factor (VIF) values were <2.0 for all variables in the model. Results Table 1 shows sample characteristics by psychological distress severity. Overall, 47% (n = 164) of participants had no or mild psychological distress, 30% (n = 107) moderate distress, and 23% (n = 81) severe distress. Participant gender, employment, education, marital status, health insurance, annual income, and tobacco use characteristics (cigarettes per day, quit motivation, quit confidence, addiction, expenditures) did not vary significantly by psychological distress severity (all p >0.05). However, participants with no or mild psychological distress were significantly older, more likely to identify as Black/African American, and less likely to report high levels of financial stress, smoking-induced deprivation, and food insecurity (all p <0.05). Overall, 36% (n = 111) of participants had received mental health services in the prior 3 months (Table 1), and mental health service use increased with distress severity (p = 0.004). Participants who had recently received mental health services reported a high level of satisfaction with their therapist (M = 8.02, SD = 2.4), and therapeutic satisfaction did not vary significantly by psychological distress severity (p = 0.51). Participants who had not received mental health services in the prior 3 months reported a high level of potential usefulness if they were to receive such services (M = 7.40, SD = 2.8), and ratings of potential usefulness increased with distress severity (p = 0.002). Table 2 shows factors independently associated with mental health service use, satisfaction, and potential usefulness among the subgroup of participants with moderate or severe psychological distress. Male participants were significantly less likely than female participants to have received mental health services in the prior 3 months (28% vs. 53%, p <0.01) and Non-Hispanic Black participants were significantly less likely than Hispanic and non-Hispanic White participants to have received mental health services in the prior 3 months (33% vs. 48% and 51%, respectively, p <0.05). Among participants who had received mental health services in the prior 3 months, male participants were significantly more satisfied with their therapist (M = 8.9, SD = 1.3) than female participants (M = 7.8, SD = 2.5, p <0.01). When examining potential usefulness of mental health services among participants who had not received mental health services in the prior 3 months, participants aged 18–50 had a significantly higher usefulness rating (M = 8.2, SD = 2.2) than participants aged 50 years and older (M = 7.2, SD = 2.8, p <0.01), employed participants had a significantly higher potential usefulness rating (M = 9.0, SD = 1.7) than unemployed participants (M = 7.6, SD = 2.6, p <0.05), and participants with Medicaid-only insurance had a significantly lower potential usefulness rating (M = 7.6, SD = 2.6) than participants with other insurance statuses such as Medicare-only or private insurance (M = 8.1, SD = 2.3, p <0.05). Discussion To our knowledge, this is one of the few studies to examine psychological distress among food assistance recipients who smoke. Psychological distress was common in the sample, with over half of participants experiencing at least moderate distress and almost one-in-four experiencing severe psychological distress. The prevalence of severe distress observed in the current study is higher than prior estimates of psychological distress among people who receive food assistance and significantly higher than the 5% general prevalence of severe psychological distress in New York City (the study location) (Hamwey et al., 2024; Oddo & Mabli, 2015; Zvolensky et al., 2018). The current study further found that while tobacco use characteristics did not vary by psychological distress severity, participants with severe psychological distress had significantly higher levels of financial stress and increased odds of smoking-induced deprivation and food insecurity than individuals with lower levels of distress. Additionally, participants self-reported spending on average $8.00 per day on tobacco (extrapolated to $2,900 per year), an amount exceeding the average annual household SNAP benefit that may be crowding-out purchasing of food and other needs that promote psychological health (USDA, 2024). The current study’s cross-sectional design precludes drawing conclusions about the directions of these relationships. However, these findings are consistent with Kim-Mozeleski and Pandey’s intersecting theoretical framework (Kim-Mozeleski & Pandey, 2020) and prior research showing robust relationships between financial hardship, food insecurity, psychological distress, and smoking (Ejiohuo et al., 2024; Jandaghian-Bidgoli et al., 2024; Ryu & Fan, 2023). Strengthening policies and infrastructures to increase access to effective smoking cessation treatment for food assistance recipients who smoke may reduce distress and increase money available for food and other household needs. The potential impact of expanded access to smoking cessation treatment on household spending, psychological health, and food insecurity among food assistance recipients could be explored in future longitudinal, population-based research. The current study further examined mental health care utilization, experiences, and beliefs among food assistance recipients who smoke. Overall, about one-third of participants in the current analysis had received mental health services in the prior three months and therapeutic satisfaction was high across all distress levels. However, roughly half of people with severe distress and nearly two-thirds of people with moderate distress were not receiving mental health care, yet they reported a high level of potential usefulness if they were to receive services. These results imply that there may be unmet mental health care needs among food assistance recipients who smoke, particularly among male and Black adults with moderate-to-severe distress who were least likely to receive mental health care. It is notable that male participants who had received mental health care were more satisfied with their therapists than women, while perceptions of potential service usefulness were low among older participants, unemployed participants, and participants with Medicaid insurance. Further research is needed to understand these patterns. It is plausible that multiple and intersecting social factors are creating compounding stressors and compounding barriers to mental health care utilization and mental health care perceptions among socially marginalized food assistance recipients who smoke. Limitations Readers should interpret results in the context of study limitations. The data are from a smoking cessation clinical trial in New York City; therefore results may not generalize to other settings or the general food assistance population. The study also used a cross-sectional design, so we cannot determine causality between variables, and the small sample size in this secondary analysis may limit statistical power to detect some clinically meaningful group differences. Lastly, the study did not assess self-reported or clinician-diagnosed mental health conditions, so we are unable to account for mental health diagnostic type or severity in our analysis examining the relationships between psychological distress and tobacco use behaviors. Conclusions This cross-sectional study revealed high rates of psychological distress and potential unmet mental health care needs in a sample of food assistance recipients who smoke in New York City. Results suggest a need for more research to develop concurrent interventions that simultaneously address food needs, smoking, and psychological health. Longitudinal, population-based studies are needed to confirm findings in the general food assistance population and to inform synergistic policies and intervention approaches for food assistance recipients who smoke.