Tobacco and other substance co-use among adolescents and young adults with cancer who use tobacco: prevalence and associations with nicotine dependence and depression
Northwell Health Cancer Institute, Lake Success, New York, USA
Institute of Health System Science, Northwell Health Feinstein Institutes for Medical Research, Manhasset, New York, USA
Department of Psychology, State University of New York at Buffalo, Buffalo, New York, USA
Department of Public Health Sciences, University of Virginia School of Medicine, Charlottesville, Virginia, USA
Institute of Behavioral Science, Northwell Health Feinstein Institutes for Medical Research, Manhasset, New York, USA
Departments of Psychiatry, Molecular Medicine, and Psychology, Hofstra University, Hempstead, New York, USA
Abstract
Introduction
Tobacco and other substance co-use has not been examined in adolescent and young adult (AYA) cancer survivors. We compared the prevalence of past-month co-use of tobacco+cannabis, alcohol and illicit drugs between AYAs with and without a cancer history and considered associations between co-use and nicotine dependence in AYA cancer survivors who use tobacco, exploring if past-year major depression moderates this relationship.
Methods
2015–2019 National Survey on Drug Use and Health data were used to analyse past-month co-use in 7793 AYAs (228 with cancer; 7565 without cancer). Weighted univariable and multivariable logistic regression models estimated associations between cancer history and co-use and co-use with nicotine dependence (among AYA cancer survivors) incorporating moderation by major depression.
Results
AYA cancer survivors had lower reported past-month cannabis co-use than those without cancer (29% vs 39%), but cancer history was not associated with cannabis co-use in multivariable models (adjusted OR (aOR): 0.83, 95% CI=0.54, 1.28). When AYA cancer survivors who use tobacco had major depression, alcohol co-use was associated with lower rates of nicotine dependence (aOR=0.08, 95% CI=0.01, 0.53).
Conclusions
There are high rates of substance co-use among AYAs who use tobacco, consistent across cancer history. Unlike previous research, alcohol co-use was associated with lower rates of nicotine dependence, but only for those with major depression. This finding could be related to neurochemical dysregulation due to co-use and warrants further exploration. Future research should also examine more nuanced definitions of substance use including modes, patterns and initiation of use, and explore motivation to change tobacco behaviour in AYA cancer survivor populations.
Untitled section
Keywords: Tobacco Use Cessation, Public Health Surveillance, Epidemiology, Health Behavior, Medical Oncology
Article notes
Untitled section
Received 2025 Mar 10; Accepted 2025 Aug 1; Collection date 2025.
Boxed Text
WHAT IS ALREADY KNOWN ON THIS TOPIC
- Adolescents and young adults (AYA) diagnosed with cancer represent a population uniquely at risk for the concerning long-term effects of tobacco use, independently and in conjunction with additional substances. Prior research has compared rates of single substance use (eg, tobacco, alcohol, cannabis and other drugs) between AYAs with and without a cancer history.
WHAT THIS STUDY ADDS
- Prior research has not explored tobacco + other substance co-use in AYA cancer survivors or determined the relationship between co-use and nicotine dependence in this population, even though co-use is associated with lower likelihood to quit tobacco in other populations. We found similar rates of co-use between AYAs with and without a cancer history. For AYA cancer survivors who use tobacco and have major depression, alcohol co-use was associated with lower rates of nicotine dependence.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
- Given the high and comparable rates of tobacco + other substance co-use in AYA cancer survivors, our results highlight the need to consider co-use behaviours in AYA cancer survivors. Our study findings challenge results from previous studies that found tobacco+alcohol co-use was associated with increased nicotine dependence, opening new research opportunities for the development of hypotheses and interventions to address co-use behaviours in AYA cancer survivors.
Introduction
Adolescents and young adults (AYA) diagnosed with cancer between ages 15 and 39 have high 5-year relative survival rates (85.8%),1 necessitating an increased focus on their long-term health. AYA cancer survivors have an increased risk for infertility, psychological disorders, cardiovascular illness and subsequent malignancies.2,4 In addition, AYA cancer survivors report alarmingly high tobacco use rates (38% are regular users of tobacco),5,9 a concerning trend because tobacco use independently increases the risk for recurrence and secondary cancers.10 11 Examining AYA cancer survivors who use tobacco and how they compare to their peers without cancer is necessary to inform strategies to reduce tobacco use and improve their long-term health.
Nationally representative data found that AYA cancer survivors have a higher prevalence of regular tobacco use (38%) and daily use (28%) compared with adult survivors overall (17% regular users, 13% daily users),5 6 and these rates are comparable to AYAs without cancer (33% regular users).6 AYA cancer survivors who use tobacco are more likely than AYA survivors who do not use tobacco to report psychological distress, depression, anxiety and poor health status.812,14 Unfortunately, few tobacco cessation interventions developed for AYA cancer survivors have effectively reduced smoking rates,15 16 underscoring the urgent need to search for alternative intervention targets.
One approach to improve tobacco cessation interventions for AYA cancer survivors could be exploring the feasibility of multiple behaviour change interventions, particularly tobacco plus other substance co-use. Substance co-use is the concurrent or sequential use of two or more substances by the same person within a designated period and is associated with a higher risk for nicotine dependence, substance use disorder, psychological and medical sequelae, and reduced tobacco cessation.17 18 Additionally, other substance use, specifically alcohol, is an established risk factor for cancer,19 making it a relevant target in prevention of secondary cancers for AYA cancer survivors. However, no studies have explored co-use in AYA cancer survivors despite their high rates of individual substance use (eg, alcohol, cannabis).7 20 21 The lack of exploration of co-use is a concerning gap because research in non-cancer populations has demonstrated that tobacco cessation programmes are not as effective in populations with co-use.10 22
Adolescence to young adulthood is also a key period for neurodevelopment, which can be impacted by substance use.23 24 Co-use might also exacerbate withdrawal-related anxiety and depressive symptoms through shared pathophysiological mechanisms, such as oxidative stress and neuroinflammation.24,27 For example, brain-derived neurotrophic factor (BDNF) alterations are implicated in both depression and substance use withdrawal25,27 and have been shown to be elevated among adults who smoke compared with adults who formerly smoked or who never smoked.28 Cancer diagnosis and treatment during adolescence and young adulthood can also influence neurodevelopment, including the development of anxiety and depression.29 30 Yet, the potential intersection between cancer diagnosis, treatment, substance use or co-use, and mental health has not been thoroughly explored.
In this study, we aimed to improve our understanding of tobacco+other substance co-use (hereafter described as co-use) among AYA cancer survivors through two primary objectives: (1) estimate the prevalence of past-month co-use among a nationally representative sample of AYAs and compare the co-use prevalence rates between cancer survivors and a non-cancer comparison group; and (2) test the association between past-month co-use and nicotine dependence in AYA cancer survivors who use tobacco and determine if this association is moderated by mental health status.
Materials & methods
Data source
We used data from the 2015–2019 National Survey on Drug Use and Health (NSDUH) surveys. The NSDUH is an annual, nationally representative, cross-sectional survey on drug use and mental health among adolescents and adults (individuals 12 years and up) residing in the USA. Households in all 50 states and the District of Columbia are eligible, and response rates for the surveys ranged from 71% to 80%.31 NSDUH data are provided at no cost for secondary data analyses as de-identified datasets. Institutional Review Board approval is not required to access and use these de-identified datasets. We elected to use the data collected from 2015 to 2019 because the NSDUH survey format was overhauled substantially in 2015, and the COVID-19 pandemic required different data collection methodologies in 2020. As a result, the survey developers do not recommend comparisons to other years. Additionally, AYAs diagnosed with cancer during the COVID-19 pandemic may have had a different experience than AYAs diagnosed before the COVID-19 pandemic in ways that could influence their substance use.
Patient and public involvement
No patients or members of the public were involved in this secondary analysis of NSDUH data.
Sample
The flow diagram of the cohort selection process is shown in online supplemental figure 1. The final sample included AYAs who currently used tobacco with or without a history of cancer (excluding non-melanoma skin cancer). The sample was first restricted by age. The National Cancer Institute (NCI) defines AYA cancer survivors as any individual diagnosed with cancer between the ages of 15 and 39.32 After age 21, NSDUH data restricts age information by only providing respondent age in categories. The category that included the last ages in the AYA definition included a range of 35–49 years. Therefore, to be conservative, we elected to use a smaller age range for eligibility (ages 16–34) to ensure only AYAs were included. AYAs with missing information about cancer history were then excluded, and those remaining were then categorised as being a cancer survivor if they reported a history of cancer. We then excluded any AYA cancer survivors who reported that their cancer diagnosis occurred in the past year, since our interest was in behaviours during post-treatment survivorship. Then, AYA cancer survivors diagnosed with non-melanoma skin cancer were excluded. Lastly, we excluded any AYA who reported not using tobacco, resulting in a final sample of 7793 AYAs who currently used tobacco.
Measures
Substance use
We examined past-month use of tobacco, illicit drugs, alcohol and cannabis because prior research has already examined past-year use,6 and past-month use better identifies regular use and those who are likely to require cessation services. Tobacco use was recoded from individual variables reporting past-month use of cigarettes, cigars, pipes and smokeless tobacco. AYAs were coded as having tobacco use in the past month if they reported past-month use of any of these products. Use of non-cannabis illicit drugs was recoded from individual variables reporting past-month use of cocaine, crack, heroin, hallucinogens, LSD, PCP, ecstasy, DMT/AMT/Foxy, ketamine, salvia, inhalants, methamphetamine and prescription medications without a doctor’s prescription (pain relievers, oxycontin, tranquillisers, stimulants, sedatives, psychotherapeutics). AYAs were coded as having used illicit drugs in the past month if they reported past-month use of any of these illicit drugs. For both alcohol and cannabis, AYAs were coded as having used alcohol or cannabis in the past month if they reported that their most recent time consuming either of these substances was within the last 30 days. Individuals were asked to report the age at which they first used three of the measured tobacco products (cigarettes, cigars, smokeless tobacco), and we dichotomised these variables at the median value of 16, 18 and 17 years, respectively.
Co-use
We created three composite variables measuring past-month co-use of tobacco+cannabis, tobacco+alcohol and tobacco+illicit drugs. For each tobacco co-use combination, individuals were categorised as 0=past-month use of tobacco only and 1=past-month co-use. Therefore, the specific patterns of co-use behaviour (eg, using on the same day or using at the same time) were not considered. Categorisation in each co-use variable was not conditioned on use of other products (ie, someone with tobacco+cannabis co-use may have also used alcohol that month).
Mental health
Mental health was operationalised as a major depressive episode in the past year and was measured using items adapted from the National Comorbidity Survey-Replication (adults) and National Comorbidity Survey-Adolescent (youth) to assess symptom criteria for major depression based on the Diagnostic and Statistical Manual of Mental Disorders (DSM-5), 5th edition.33 AYAs were categorised as reporting a prior year major depressive episode if they had at least five of the nine symptoms in the same 2-week period, and reported feeling depressed or loss of interest in daily activities for a 2-week period in the past year.
Nicotine dependence
Nicotine dependence was measured using the Nicotine Dependence Syndrome Scale (NDSS).34 NDSS scores were calculated based on 17 of the 19 items per recommendations by scale developers because two items (both pertaining to non-smoking friends) have high non-response.34 Four items were reverse scored, and a sum across all items was then calculated and divided by 17. Higher scores indicate more nicotine dependence. AYAs who use tobacco were defined as nicotine dependent if their NDSS score was >2.75,34 and scores for AYAs that answered <15/17 items were not calculated and coded as not nicotine dependent.
Sociodemographic and clinical variables
Sociodemographic and clinical variables previously considered as correlates of tobacco use were included as potential covariates. These included age range at survey, sex, insurance status, race and ethnicity, sexual orientation, marital status, cancer type (blood cancer (yes vs no or unspecified) and multiple cancer diagnoses (2+ vs 1 or unspecified)), comorbidities and overall health status. Due to the age of individuals included in our sample, it is expected that some were full-time students (eg, 16–17-year-olds) whereas others were full-time employees (eg, 26–34-year-olds). Therefore, we created a mutually exclusive, composite variable of both school and work status items.
Statistical analysis
First, we conducted descriptive analyses to examine the weighted percentages and unweighted frequencies of sociodemographic and clinical characteristics and the prevalence rates of substance co-use between AYAs who use tobacco with and without a history of cancer. Next, we performed bivariate analyses using a χ² test with Rao & Scott’s second-order correction to compare the prevalence rates of co-use, sociodemographic and clinical characteristics between AYAs who use tobacco with and without a history of cancer. To answer Aim 1, we first conducted weighted univariable logistic regression to assess the association between cancer history and potential control variables with past-month co-use to produce unadjusted OR and 95% CI. Next, we performed weighted multivariable logistic regression to compare co-use between AYAs who use tobacco with and without a cancer history, controlling for all included sociodemographic and clinical factors. Each co-use variable was tested in a separate model. Multivariable logistic regression models provided adjusted OR (aOR) and 95% CIs.
For Aim 2, we restricted the sample to AYA cancer survivors who use tobacco (n=228). We first evaluated the association of co-use of each substance with nicotine dependence using a χ² test with Rao & Scott’s second-order correction. Next, we used weighted univariable logistic regression to analyse unadjusted relationships between co-use, major depressive episode and potential control variables, including cancer history information with nicotine dependence. Finally, weighted multivariable logistic regression with an interaction was performed to examine if the relationship between co-use and nicotine dependence was moderated by mental health status (assessed via past-year major depressive episode) while controlling for other sociodemographic and clinical factors. We included an interaction term between co-use and mental health status, adjusting for the control variables included in the multivariable model for each co-use outcome. The interaction term was dropped from the final model if it was not significant at p<0.05.
All the analyses were weighted using the NSDUH survey weight to achieve nationally representative estimates. The survey weight accounted for weighting and clustering effects such as unequal probabilities of non-response and post-stratification adjustments.35 The nesting variables were also used to capture explicit stratification and identify clustering within the data, which are needed to accurately compute the variance estimates.36 All statistical analyses were performed using R version 4.2.2.
Data availability
The data are publicly available through the Substance Abuse and Mental Health Services Administration website: samhsa.gov.
Results
Sample characteristics
The final sample included 228 AYA cancer survivors who use tobacco and 7565 AYAs without a history of cancer who use tobacco. Among AYA cancer survivors who use tobacco, the majority were >21 years (98%); 70% were female; 77% identified as non-Hispanic White, 6% identified as non-Hispanic Black and 10% identified as Hispanic; 21% identified as lesbian/gay/bisexual and 79% identified as heterosexual/straight. A higher prevalence of poor or fair overall health was observed among AYA cancer survivors who use tobacco compared with AYA non-cancer survivors who use tobacco (29% vs 18%, respectively). A full description of the sample is in table 1.
| Characteristic | Overall | AYAs with cancer | AYAs without cancer | P value |
|---|---|---|---|---|
| (N=7793) | (N=228) | (N=7565) | ||
| Unweighted N (weighted %) | Unweighted N (weighted %) | Unweighted N (weighted %) | ||
| Age | <0.001 | |||
| 16–20 | 1685 (15.4%) | 11 (2.5%) | 1674 (15.8%) | |
| 21–25 | 2824 (26.8%) | 67 (22.6%) | 2757 (26.9%) | |
| 26–34 | 3284 (57.8%) | 150 (74.9%) | 3134 (57.3%) | |
| Sex | <0.001 | |||
| Male | 3985 (54.1%) | 69 (30.1%) | 3916 (54.9%) | |
| Female | 3808 (45.9%) | 159 (69.9%) | 3649 (45.1%) | |
| Race/ethnicity | 0.039 | |||
| Non-Hispanic Black | 1006 (14.1%) | 11 (6.0%) | 995 (14.4%) | |
| Non-Hispanic White | 4950 (65.5%) | 178 (77.2%) | 4772 (65.1%) | |
| Hispanic | 994 (13.2%) | 15 (10.3%) | 979 (13.3%) | |
| Other | 843 (7.2%) | 24 (6.5%) | 819 (7.2%) | |
| Insurance | 0.020 | |||
| Uninsured | 1239 (17.5%) | 39 (18.5%) | 1200 (17.5%) | |
| Private insurance | 3740 (48.4%) | 97 (41.5%) | 3643 (48.6%) | |
| Public insurance | 2502 (30.3%) | 88 (39.1%) | 2414 (30.0%) | |
| Other insurance | 312 (3.8%) | 4 (0.9%) | 308 (3.9%) | |
| Education/employment status | 0.104 | |||
| No school & unemployed | 1930 (25.1%) | 77 (35.2%) | 1853 (24.7%) | |
| No school (or part-time student) & employed full time | 3417 (48.1%) | 101 (42.4%) | 3316 (48.3%) | |
| Full-time student & unemployed (or employed part-time) | 1032 (10.4%) | 20 (8.3%) | 1012 (10.5%) | |
| Working and studying full time | 332 (3.7%) | 5 (3.4%) | 327 (3.8%) | |
| Neither above | 1082 (12.7%) | 25 (10.8%) | 1057 (12.8%) | |
| Marital status | <0.001 | |||
| Never been married | 5682 (69.5%) | 125 (51.1%) | 5557 (70.1%) | |
| Married | 1491 (21.6%) | 69 (33.1%) | 1422 (21.2%) | |
| Widowed/divorced/separated | 620 (8.9%) | 34 (15.9%) | 586 (8.7%) | |
| Sexual identity | 0.282 | |||
| Heterosexual/straight | 5971 (83.1%) | 178 (79.2%) | 5793 (83.3%) | |
| Lesbian/gay/bisexual | 1284 (16.9%) | 47 (20.8%) | 1237 (16.7%) | |
| Overall health status | 0.003 | |||
| Excellent/very good | 3535 (44.3%) | 82 (34.2%) | 3453 (44.7%) | |
| Good | 2885 (37.3%) | 77 (37.4%) | 2808 (37.3%) | |
| Poor/fair | 1373 (18.4%) | 69 (28.5%) | 1304 (18.0%) | |
| Comorbidities | 0.098 | |||
| 0 or 1 | 6580 (84.3%) | 184 (78.4%) | 6396 (84.5%) | |
| 2+ | 1213 (15.7%) | 44 (21.6%) | 1169 (15.5%) |
Prevalence of co-use among AYAs who use tobacco with and without cancer
Among AYAs who use tobacco, 29% of AYA cancer survivors reported past-month tobacco+cannabis co-use, 66% reported tobacco+alcohol co-use and 14% reported tobacco+illicit drug co-use (online supplemental table 1). In univariable logistic regression comparing AYAs with and without a history of cancer (online supplemental table 2), AYA cancer survivors were significantly less likely to report past-month tobacco+cannabis co-use (OR=0.64, 95% CI=0.44, 0.94). In multivariable analyses (table 2), cancer history was not significantly associated with past-month co-use of any substance.
| Characteristic | Model 1 | Model 2 | Model 3 | |||
|---|---|---|---|---|---|---|
| Cannabis co-use | Alcohol co-use | Illicit drug co-use | ||||
| aOR (95% CI) | P value | aOR (95% CI) | P value | aOR (95% CI) | P value | |
| AYA group | ||||||
| AYAs comparison group | Ref | Ref | Ref | |||
| AYAs with cancer | 0.83 (0.54, 1.28) | 0.395 | 0.95 (0.65, 1.37) | 0.759 | 0.84 (0.52, 1.38) | 0.481 |
| Age | ||||||
| 16–20 | Ref | Ref | Ref | |||
| 21–25 | 0.84 (0.70, 1.01) | 0.058 | 2.48 (1.99, 3.08) | <0.001 | 1.09 (0.85, 1.41) | 0.481 |
| 26–34 | 0.65 (0.52, 0.82) | <0.001 | 2.01 (1.62, 2.48) | <0.001 | 0.85 (0.64, 1.13) | 0.252 |
| Sex | ||||||
| Male | Ref | Ref | Ref | |||
| Female | 0.69 (0.59, 0.81) | <0.001 | 0.93 (0.80, 1.10) | 0.397 | 0.84 (0.68, 1.04) | 0.104 |
| Race/ethnicity | ||||||
| Non-Hispanic Black | Ref | Ref | Ref | |||
| Non-Hispanic White | 0.66 (0.53, 0.84) | 0.001 | 0.78 (0.64, 0.95) | 0.015 | 2.18 (1.60, 2.98) | <0.001 |
| Hispanic | 0.72 (0.53, 0.99) | 0.043 | 1.00 (0.77, 1.32) | 0.976 | 2.10 (1.37, 3.22) | 0.001 |
| Other | 0.74 (0.54, 1.00) | 0.051 | 0.91 (0.69, 1.20) | 0.493 | 1.95 (1.24, 3.08) | 0.005 |
| Insurance | ||||||
| Uninsured | Ref | Ref | Ref | |||
| Private insurance | 0.68 (0.57, 0.81) | <0.001 | 1.59 (1.27, 1.99) | <0.001 | 0.78 (0.62, 1.00) | 0.051 |
| Public insurance | 0.72 (0.57, 0.91) | 0.007 | 0.62 (0.48, 0.80) | <0.001 | 0.77 (0.57, 1.05) | 0.092 |
| Other insurance | 0.74 (0.53, 1.04) | 0.085 | 0.79 (0.54, 1.14) | 0.194 | 0.48 (0.28, 0.83) | 0.010 |
| Education/employment status | ||||||
| No school & unemployed | Ref | Ref | Ref | |||
| No school (or part-time student) & employed full time | 0.92 (0.76, 1.12) | 0.407 | 1.80 (1.45, 2.22) | <0.001 | 0.82 (0.66, 1.02) | 0.073 |
| Full-time student & unemployed (or employed part-time) | 1.50 (1.21, 1.87) | <0.001 | 1.92 (1.42, 2.60) | <0.001 | 1.21 (0.84, 1.72) | 0.291 |
| Working and studying full time | 1.13 (0.83, 1.54) | 0.423 | 2.23 (1.50, 3.30) | <0.001 | 1.19 (0.72, 1.96) | 0.486 |
| Neither above | 1.17 (0.95, 1.43) | 0.134 | 1.48 (1.20, 1.81) | <0.001 | 1.08 (0.81, 1.45) | 0.577 |
Association between co-use and nicotine dependence among AYA cancer survivors who use tobacco
Overall, 33% of AYA cancer survivors who use tobacco were nicotine dependent, and 25% had a major depressive episode in the past year (online supplemental table 3). In bivariate analyses, those with alcohol co-use had lower prevalence of nicotine dependence compared with those who only use tobacco (24% vs 50%, p=0.001), and similar prevalence of major depression (24% vs 25%). Those with illicit drugs co-use had identical prevalence rate of nicotine dependence compared with those who only use tobacco (33% vs 33%) and were more likely to report a major depressive episode in the past year (44% vs 22%, p=0.035). Nicotine dependence (36% vs 32%) and major depression (30% vs 23%) were similar between those with cannabis co-use and those who only use tobacco.
In univariable logistic regression (online supplemental table 4), alcohol co-use was associated with lower nicotine dependence compared with tobacco-only use (OR=0.32, 95% CI=0.16, 0.64). Illicit drug co-use and cannabis co-use were not associated with nicotine dependence, and past-year major depressive episode was not associated with nicotine dependence in any model.
Neither cannabis co-use (aOR=1.29, 95% CI=0.56, 2.98) nor illicit drug co-use (aOR=0.92, 95% CI=0.38, 2.20) was associated with nicotine dependence in adjusted models (table 3). There was a significant interaction between alcohol co-use and a past-year major depressive episode (p=0.013) but not for cannabis nor illicit drug co-use. Among AYA cancer survivors who use tobacco and had a major depressive episode in the past year, alcohol co-use was associated with lower odds of nicotine dependence compared with tobacco-only use (aOR=0.08, 95% CI=0.01, 0.53).
| Characteristic | Model 1 | Model 2 | Model 3 | |||
|---|---|---|---|---|---|---|
| Nicotine dependence | Nicotine dependence | Nicotine dependence | ||||
| aOR (95% CI) | P value | aOR (95% CI) | P value | aOR (95% CI) | P value | |
| Cannabis co-use | ||||||
| Used tobacco only | Ref | |||||
| Used both tobacco and cannabis | 1.29 (0.56, 2.98) | 0.531 | – | – | ||
| Alcohol co-use | ||||||
| Used tobacco only | Ref | |||||
| Used both tobacco and alcohol | – | 0.67 (0.24, 1.83) | 0.409 | – | ||
| Alcohol co-use × MDEYR | ||||||
| Used both tobacco and alcohol×yes | – | 0.08 (0.01, 0.53) | 0.013 | – | ||
| Illicit drug co-use | ||||||
| Used tobacco only | Ref | |||||
| Used both tobacco and illicit drugs | – | – | 0.92 (0.38, 2.20) | 0.834 | ||
| MDEYR | ||||||
| No/unknown | Ref | Ref | Ref | |||
| Yes | 1.83 (0.79, 4.26) | 0.148 | 11.1 (2.06, 60.0) | 0.008 | 1.90 (0.81, 4.45) | 0.131 |
| Age | ||||||
| 16–20 | Ref | Ref | Ref | |||
| 21–25 | 4.41 (0.68, 28.6) | 0.112 | 5.10 (0.58, 44.9) | 0.132 | 4.47 (0.65, 31.0) | 0.121 |
| 26–34 | 2.07 (0.27, 15.6) | 0.457 | 2.19 (0.19, 24.7) | 0.502 | 2.02 (0.26, 16.0) | 0.483 |
| Gender | ||||||
| Male | Ref | Ref | Ref | |||
| Female | 1.31 (0.50, 3.47) | 0.559 | 1.07 (0.40, 2.82) | 0.890 | 1.30 (0.49, 3.44) | 0.574 |
| Race/ethnicity | ||||||
| Non-Hispanic Black | Ref | Ref | Ref | |||
| Non-Hispanic White | 0.83 (0.13, 5.19) | 0.831 | 0.87 (0.15, 5.11) | 0.865 | 0.87 (0.14, 5.53) | 0.873 |
| Hispanic | 0.12 (0.01, 2.33) | 0.151 | 0.15 (0.01, 3.69) | 0.227 | 0.14 (0.01, 2.29) | 0.158 |
| Other | 1.45 (0.15, 14.1) | 0.732 | 1.65 (0.20, 13.8) | 0.625 | 1.65 (0.18, 15.5) | 0.642 |
| Insurance | ||||||
| Uninsured | Ref | Ref | Ref | |||
| Private insurance | 0.64 (0.26, 1.56) | 0.303 | 0.64 (0.24, 1.69) | 0.344 | 0.59 (0.24, 1.49) | 0.249 |
| Public insurance | 1.02 (0.37, 2.76) | 0.974 | 0.71 (0.24, 2.11) | 0.514 | 0.99 (0.37, 3.70) | 0.989 |
| Other insurance | 0.36 (0.02, 6.35) | 0.466 | 0.35 (0.02, 5.59) | 0.431 | 0.34 (0.02, 5.73) | 0.433 |
| Education/employment status | ||||||
| No school & unemployed | Ref | Ref | Ref | |||
| No school (or part-time student) & employed full time | 0.57 (0.20, 1.68) | 0.289 | 0.51 (0.18, 1.49) | 0.203 | 0.58 (0.20, 1.70) | 0.300 |
| Full-time student & unemployed (or employed part-time) | 0.41 (0.06, 2.96) | 0.353 | 0.44 (0.07, 2.78) | 0.357 | 0.44 (0.06, 3.08) | 0.383 |
| Working and studying full time | 0.23 (0.01, 3.77) | 0.280 | 0.39 (0.02, 7.78) | 0.514 | 0.25 (0.02, 3.97) | 0.304 |
| Neither above | 0.26 (0.07, 0.88) | 0.033 | 0.21 (0.06, 0.75) | 0.019 | 0.26 (0.07, 0.93) | 0.040 |
| Blood cancer | ||||||
| No (or unspecified) | Ref | Ref | Ref | |||
| Yes | 0.63 (0.23, 1.70) | 0.335 | 0.63 (0.22, 1.76) | 0.352 | 0.63 (0.23, 1.74) | 0.353 |
| Multiple cancer diagnoses | ||||||
| No (1 or unspecified) | Ref | Ref | Ref | |||
| Yes (2+) | 0.65 (0.20, 2.16) | 0.461 | 0.84 (0.21, 3.32) | 0.790 | 0.67 (0.20, 2.21) | 0.484 |
Discussion
The recent NCI monograph on tobacco use and cessation programmes in cancer survivors highlights AYAs as a high-risk group that needs additional focus.37 This study is the first to explore the prevalence of past-month co-use of other substances as a potential risk factor for higher nicotine dependence and compare these to the rates of AYAs without a history of cancer. We found that AYA cancer survivors who use tobacco have high rates of co-use; their rates of co-use were similar to their non-cancer peers. Among AYA cancer survivors who use tobacco with a past-year major depressive episode, alcohol co-use was associated with a far lower likelihood of nicotine dependence.
Our study explored a potential target for cessation interventions—co-use. In non-cancer populations, co-use has been associated with lower rates of tobacco cessation17 18 and tobacco cessation interventions are less effective for those who co-use other substances,17 22 across varying definitions of co-use.17 We found that AYA cancer survivors who use tobacco have high rates of past-month co-use. However, we did not find a significant relationship between co-use and nicotine dependence. For tobacco+cannabis co-use, the non-significant relationship may be driven by the specific mode of cannabis consumption (eg, edibles vs inhaled mode) or how the co-use occurred (ie, concurrently vs sequentially); however, the details of cannabis use were not collected as part of NSDUH data and could not be explored in the current study. Cannabis is increasingly being prescribed during cancer treatment as a symptom management tool,38,40 and providers often encourage individuals to consume it through non-inhaled modes.41 42 However, among adolescents in particular, there is limited evidence examining this co-use behaviour.24 The relationship between tobacco+cannabis co-use and reduced tobacco cessation may be driven by using the substances together (side-by-side or mixed formulations) or because they involve similar behavioural actions. Future research should focus on a more detailed examination of modes of cannabis consumption, history of use for cancer symptom management and the order of substance initiation to identify potential intervention targets.
Among AYA cancer survivors who use tobacco and who had a major depressive episode in the past year, co-use of alcohol and tobacco was associated with significantly lower rates of nicotine dependence compared with the use of tobacco alone. This finding is unexpected because major depression is often associated with increased nicotine dependence.43 44 The mechanistic understanding for the association between depression and dependence is based on the neurobiology of addiction wherein increased substance use is used to mitigate the negative affect state caused by withdrawal.45 However, previous evaluations of substance use and depression have primarily focused on single use patterns, not co-use behaviours.46 The shared pathophysiological mechanisms between depression and anxiety and substance use withdrawal through oxidative stress and neuroinflammation24,27 may mean that this finding is neurobiologically mediated. For example, animal studies have demonstrated that substances such as biotin and vitamin C can mitigate alcohol withdrawal-induced depression via serotonergic and antioxidant mechanisms and suggest shared potential neurobiological mechanistic overlaps.27 47 Another previous study found that shared exposure to alcohol and nicotine in adolescence only increased alcohol consumption (and not alcohol and nicotine consumption) compared with exposure to alcohol only,48 which could mean that when co-using alcohol and tobacco, reinforcing behaviours are following an alcohol-mediated pathway over a nicotine-mediated pathway to address symptoms of depression. However, these findings have not been replicated in human studies. It is also possible that among those with co-use of alcohol and tobacco, the decreased association with nicotine dependence among those who had a major depressive episode in the last year could be due to differences in smoking behaviours in social settings. Specifically, the patterns of tobacco use may be different for those who co-use with alcohol, and the social drivers of use may be impacted by major depressive episodes, reducing overall exposure to nicotine and resulting in lower dependence. Future research comparing tobacco use patterns between those who only use tobacco and those with alcohol co-use is needed to explore this hypothesis.
The strengths of the current study are the use of a nationally representative dataset and the focus on past-month substance co-use. However, the study has limitations. First, limitations with age categorisations in the NSDUH data resulted in limiting the available age range for study eligibility compared with the NCI definition of an AYA (ie, ages 16–34 years in the current study vs 15–39 years as defined by the NCI). Second, NSDUH also did not ask about e-cigarette use. E-cigarette use is one of the main modes of tobacco use among our target population,5 49 50 and therefore, tobacco use is likely even higher than what is estimated here. However, while our evaluation of the association between co-use and nicotine dependence is limited to those who use products other than electronic cigarettes, the nicotine dependence outcome remains relevant across products and the shared pathophysiological mechanisms would be similar across products as this mechanism is based on nicotine use. Third, past-month use included any amount of non-tobacco substance use. It is possible that more restrictive definitions (eg, daily use or past-month binge drinking) may modify the study’s conclusions. Fourth, cannabis use did not include assessment of the mode of consumption. Finally, we did not condition co-use on other substance use.
The findings of this study provide important implications for cessation intervention research. Although we did not find a significant relationship between co-use and nicotine dependence, more focus on co-use is warranted because co-use rates are high in AYA cancer survivors. Additionally, future research should examine neurobiological correlates of co-use and mental health as well as examining the social context in which substance use occurs, other psychosocial characteristics (eg, impulsivity), or the use of cannabis during cancer treatment as potential intervention targets for AYA cancer survivors.
Additional information
NSDUH data files are publicly available. Data and user guides can be found online: https://www.samhsa.gov/data/data-we-collect/nsduh-national-survey-drug-use-and-health/datafiles. Methodological reports with scripts for analyses can also be found online: https://www.samhsa.gov/data/data-we-collect/nsduh-national-survey-drug-use-and-health/methodology/2019.
Supplementary material
Footnotes
Footnote Group
Data availability statement
Data are available in a public, open access repository.
References
Untitled section
References
- 1.National Cancer Institute Cancer stat facts: cancer among adolescents and young adults (AYAS) (ages 15–39) 2023. https://seer.cancer.gov/statfacts/html/aya.html#:~:text=Statistics%20at%20a%20Glance&text=85.8%25%20of%20AYAs%20diagnosed%20with,for%205%20years%20after%20diagnosis.&text=Cancers%20diagnosed%20among%20AYAs%2C%20ages%2015%E2%80%9339%2C%204.4%25 Available.
- 2.Perez GK, Salsman JM, Fladeboe K, et al. Taboo Topics in Adolescent and Young Adult Oncology: Strategies for Managing Challenging but Important Conversations Central to Adolescent and Young Adult Cancer Survivorship. Am Soc Clin Oncol Educ Book. 2020;40:1–15. doi: 10.1200/EDBK_279787.
- 3.Muffly LS, Hlubocky FJ, Khan N, et al. Psychological morbidities in adolescent and young adult blood cancer patients during curative-intent therapy and early survivorship. Cancer. 2016;122:954–61. doi: 10.1002/cncr.29868.
- 4.Suh E, Stratton KL, Leisenring WM, et al. Late mortality and chronic health conditions in long-term survivors of early-adolescent and young adult cancers: a retrospective cohort analysis from the Childhood Cancer Survivor Study. Lancet Oncol. 2020;21:421–35. doi: 10.1016/S1470-2045(19)30800-9.
- 5.Salloum RG, Huo J, Lee J-H, et al. Tobacco and E-cigarette use among cancer survivors in the United States. PLoS One. 2019;14:e0226110. doi: 10.1371/journal.pone.0226110.
- 6.Ji X, Cummings JR, Mertens AC, et al. Substance use, substance use disorders, and treatment in adolescent and young adult cancer survivors—Results from a national survey. Cancer. 2021;127:3223–31. doi: 10.1002/cncr.33634.
- 7.Deleemans JM, Zwicker HM, Reynolds KA, et al. Associations Among Health Behaviors and Psychosocial Outcomes in Adolescent and Young Adult Cancer Survivors. J Adolesc Young Adult Oncol. 2021;10:675–81. doi: 10.1089/jayao.2020.0224.
- 8.Warner EL, Nam GE, Zhang Y, et al. Health behaviors, quality of life, and psychosocial health among survivors of adolescent and young adult cancers. J Cancer Surviv. 2016;10:280–90. doi: 10.1007/s11764-015-0474-7.
- 9.Stroske I, Geue K, Friedrich M, et al. Health Behavior and Associated Factors in Young Adult Cancer Patients. Front Psychol. 2021;12:697096. doi: 10.3389/fpsyg.2021.697096.
- 10.Lee JS, DuBois SG, Coccia PF, et al. Increased risk of second malignant neoplasms in adolescents and young adults with cancer. Cancer. 2016;122:116–23. doi: 10.1002/cncr.29685.
- 11.Antwi GO, Lohrmann DK, Jayawardene W, et al. Associations between Cigarette Smoking and Health-Related Quality of Life in Adult Survivors of Adolescent and Young Adult Cancer. J Cancer Educ. 2022;37:508–16. doi: 10.1007/s13187-020-01837-8.
- 12.Abdelhadi OA, Pollock BH, Joseph JG, et al. Psychological distress and associated additional medical expenditures in adolescent and young adult cancer survivors. Cancer. 2022;128:1523–31. doi: 10.1002/cncr.34064.
- 13.Asvat Y, King AC, Smith LJ, et al. Substance use behaviors in adolescent and young adult cancer patients: Associations with mental and physical health. Psychooncology. 2020;29:1068–76. doi: 10.1002/pon.5378.
- 14.Grégoire S, Lamore K, Laurence V, et al. Coping Strategies and Factors Related to Problematic Substance Use and Behavioral Addictions Among Adolescents and Young Adults with Cancer. J Adolesc Young Adult Oncol. 2020;9:639–50. doi: 10.1089/jayao.2019.0171.
- 15.Pugh G, Gravestock HL, Hough RE, et al. Health Behavior Change Interventions for Teenage and Young Adult Cancer Survivors: A Systematic Review. J Adolesc Young Adult Oncol. 2016;5:91–105. doi: 10.1089/jayao.2015.0042.
- 16.Klesges RC, Krukowski RA, Klosky JL, et al. Efficacy of a tobacco quitline among adult survivors of childhood cancer. Nicotine Tob Res. 2015;17:710–8. doi: 10.1093/ntr/ntu216.
- 17.McClure EA, Rabin RA, Lee DC, et al. Treatment Implications Associated with Cannabis and Tobacco Co-Use. Curr Addict Rep. 2020;7:533–44. doi: 10.1007/s40429-020-00334-8.
- 18.Berg CJ, Duan X, Romm K, et al. Young Adults’ Vaping, Readiness to Quit, and Recent Quit Attempts: The Role of Co-use With Cigarettes and Marijuana. Nicotine Tob Res. 2021;23:1019–29. doi: 10.1093/ntr/ntaa265.
- 19.US Department of Health and Human Services Alcohol and cancer risk: the U.S. surgeon general’s advisory. 2025
- 20.Sanford NN, Sher DJ, Xu X, et al. Alcohol Use Among Patients With Cancer and Survivors in the United States, 2000-2017. J Natl Compr Canc Netw. 2020;18:69–79. doi: 10.6004/jnccn.2019.7341.
- 21.Podda M, Pagani Bagliacca E, Sironi G, et al. Cannabinoids use in adolescents and young adults with cancer: a single-center survey. Tumori. 2020;106:281–5. doi: 10.1177/0300891620912022.
- 22.Walsh H, McNeill A, Purssell E, et al. A systematic review and Bayesian meta-analysis of interventions which target or assess co-use of tobacco and cannabis in single- or multi-substance interventions. Addiction. 2020;115:1800–14. doi: 10.1111/add.14993.
- 23.Squeglia LM, Jacobus J, Tapert SF. The influence of substance use on adolescent brain development. Clin EEG Neurosci. 2009;40:31–8. doi: 10.1177/155005940904000110.
- 24.Steinfeld MR, Torregrossa MM. Consequences of adolescent drug use. Transl Psychiatry. 2023;13:313. doi: 10.1038/s41398-023-02590-4.
- 25.Correia AS, Cardoso A, Vale N. BDNF Unveiled: Exploring Its Role in Major Depression Disorder Serotonergic Imbalance and Associated Stress Conditions. Pharmaceutics. 2023;15:2081. doi: 10.3390/pharmaceutics15082081.
- 26.Gellé T, Vinais T, Lacroix A, et al. Serum BDNF and pro-BDNF levels in alcohol use disorders according to depression status: An exploratory study of their evolution two months after withdrawal. Heliyon. 2024;10:e38940. doi: 10.1016/j.heliyon.2024.e38940.
- 27.Haidary M, Alipour A, Rezayee H, et al. Ameliorative Effects of Vitamin C on Ethanol Withdrawal-Induced Anxiety, Depression, and Cortical Oxidative Stress in Rats. Alcohol Treat Q. 2025;43:468–84. doi: 10.1080/07347324.2025.2460827.
- 28.Shafiee A, Beiky M, Mohammadi I, et al. Effect of smoking on Brain-Derived Neurotrophic Factor (BDNF) blood levels: A systematic review and meta-analysis. J Affect Disord. 2024;349:525–33. doi: 10.1016/j.jad.2024.01.082.
- 29.Jim HSL, Jennewein SL, Quinn GP, et al. Cognition in Adolescent and Young Adults Diagnosed With Cancer: An Understudied Problem. J Clin Oncol. 2018;36:2752–4. doi: 10.1200/JCO.2018.78.0627.
- 30.Marusak HA, Iadipaolo AS, Harper FW, et al. Neurodevelopmental consequences of pediatric cancer and its treatment: applying an early adversity framework to understanding cognitive, behavioral, and emotional outcomes. Neuropsychol Rev. 2018;28:123–75. doi: 10.1007/s11065-017-9365-1.
- 31.Han B. Key substance use and mental health Indicators in the United States: results from the 2019 national survey on drug use and health. Rockville, MD: Substance Abuse and Mental Health Services Administration; 2019.
- 32.National Cancer Institute . Closing the gap: research and cancer care imperatives for adolescents and young adults with cancer. Bethesda, MD: National Cancer Institute; 2006.
- 33.National Comorbidity Survey (NCS) Harvard medical school. 2005. [08-Aug-2023]. https://www.hcp.med.harvard.edu/ncs/ Available. Accessed.
- 34.Shiffman S, Waters A, Hickcox M. The nicotine dependence syndrome scale: a multidimensional measure of nicotine dependence. Nicotine Tob Res. 2004;6:327–48. doi: 10.1080/1462220042000202481.
- 35.Substance Abuse and Mental Health Services Administration 2019 methodological summary and definitions. 2020. [21-Feb-2023]. https://www.samhsa.gov/data/report/2019-methodological-summary-and-definitions Available. Accessed.
- 36.Center for Behavioral Health Statistics and Quality . Rockville, MD: 2020. 2019 national survey on drug use and health: public use file codebook.
- 37.U.S. National Cancer Institute . Vol national cancer institute tobacco control monograph 23. Bethesda, MD: U.S. Department of Health and Human Services, National Institutes of Health, National Cancer Institute; 2022. Treating smoking in cancer patients: an essential component of cancer care.
- 38.Martell K, Fairchild A, LeGerrier B, et al. Rates of Cannabis Use in Patients with Cancer. Curr Oncol. 2018;25:219–25. doi: 10.3747/co.25.3983.
- 39.Macari DM, Gbadamosi B, Jaiyesimi I, et al. Medical Cannabis in Cancer Patients: A Survey of a Community Hematology Oncology Population. Am J Clin Oncol. 2020;43:636–9. doi: 10.1097/COC.0000000000000718.
- 40.O’Brien K. Cannabidiol (CBD) in Cancer Management. Cancers (Basel) 2022;14:885. doi: 10.3390/cancers14040885.
- 41.Busse JW, Vankrunkelsven P, Zeng L, et al. Medical cannabis or cannabinoids for chronic pain: a clinical practice guideline. BMJ. 2021;374:n2040. doi: 10.1136/bmj.n2040.
- 42.Hughes PR, Nwokocha J. Medical Cannabis or Cannabinoids for Chronic Pain: BMJ Rapid Recommendation. Am Fam Physician. 2022;106:208–9.
- 43.Breslau N, Kilbey M, Andreski P. Nicotine dependence, major depression, and anxiety in young adults. Arch Gen Psychiatry. 1991;48:1069–74. doi: 10.1001/archpsyc.1991.01810360033005.
- 44.Lyons M, Hitsman B, Xian H, et al. A twin study of smoking, nicotine dependence, and major depression in men. Nicotine Tob Res. 2008;10:97–108. doi: 10.1080/14622200701705332.
- 45.Semaan A, Khan MK. StatPearls. Treasure Island (FL): StatPearls Publishing; 2025. Neurobiology of addiction.https://www.ncbi.nlm.nih.gov/books/NBK597351/ Available.
- 46.Quittschalle J, Pabst A, Löbner M, et al. Association of Alcohol and Tobacco Consumption with Depression Severity in the Oldest Old. Results from the Age Different Old Age Cohort Platform. Int J Environ Res Public Health. 2021;18:7959. doi: 10.3390/ijerph18157959.
- 47.Hossaini D, Alipour AK, Sajjadi M, et al. Biotin Mitigates Alcohol Withdrawal-Induced Anxiety and Depression by Regulating Serotonin Metabolism, BDNF, Inflammation, and Oxidative Stress in Rats. Neuropsychopharmacol Rep. 2025;45:e12523. doi: 10.1002/npr2.12523.
- 48.Lárraga A, Belluzzi JD, Leslie FM. Nicotine Increases Alcohol Intake in Adolescent Male Rats. Front Behav Neurosci. 2017;11:25. doi: 10.3389/fnbeh.2017.00025.
- 49.Lanza HI, Bello MS, Cho J, et al. Tobacco and cannabis poly-substance and poly-product use trajectories across adolescence and young adulthood. Prev Med. 2021;148:106545. doi: 10.1016/j.ypmed.2021.106545.
- 50.Evans-Polce RJ, Patrick ME, McCabe SE, et al. Prospective associations of e-cigarette use with cigarette, alcohol, marijuana, and nonmedical prescription drug use among US adolescents. Drug Alcohol Depend. 2020;216:108303. doi: 10.1016/j.drugalcdep.2020.108303.
Associated Data
Supplementary Materials
Data Availability Statement
The data are publicly available through the Substance Abuse and Mental Health Services Administration website: samhsa.gov.
Data are available in a public, open access repository.