Associations between neighborhood disadvantage, cannabis use, and patient-reported outcomes among patients with cancer
Department of Psychology at the State University of New York, University at Buffalo, United States
University of Pennsylvania School of Nursing, United States
Sidney Kimmel Cancer Center, Thomas Jefferson University, United States
Abstract
Introduction
Cannabis use is proliferating among oncological populations. However, research is limited on whether cannabis routes and sources are associated with neighborhood disadvantage and whether it moderates the relationship between neighborhood disadvantage and patient-reported outcomes (PROs). This analysis evaluates the relationship between neighborhood disadvantage and cannabis source and route, and their association with PROs.
Methods
Analyses included 106 patients with cancer from varying levels of neighborhood disadvantage. Logistic regressions tested the association of neighborhood disadvantage with cannabis source (i.e., regulated vs. unregulated) and route (i.e., inhaled vs. consumed/other formulations). Multilevel regressions tested the association of neighborhood disadvantage with PROs (e.g., anxiety, depression, sleep quality, pain, quality of life) and moderation by cannabis variables.
Results
Neighborhood disadvantage did not uniquely predict cannabis source or route (p’s > .05) or PROs (p’s > .05). Conversely, household income predicted pain interference and severity, quality of life, and depression, indicating that lower household income was associated with worse outcomes (p’s < .05). There was a significant interaction between neighborhood disadvantage and visit for both pain interference (b=-1.37, 95%CI [-2.4,-0.3], p = .01) and sleep quality (b=1.74, 95%CI [-0.001,3.5], p = .05). Cannabis route significantly moderated the relationship between neighborhood disadvantage and anxiety (b=-5.32, 95%CI [-10.2,-0.4], p = .03). The interaction was driven by reversed scores between patients using inhaled and consumed/other formulations at different levels of neighborhood disadvantage.
Conclusion
Although the composite index of neighborhood disadvantage was not associated with cannabis use or PROs, our data suggest that income and cannabis route may be key factors associated with health disparities among patients with cancer.
Untitled section
Keywords: Cannabis, Cancer, Neighborhood disadvantage, Health outcomes, Cannabis source, Cannabis route
Highlights
- •A third of patients with cancer use cannabis as a symptom management option.
- •Pain interference and sleep quality scores vary across ADI groups over time.
- •Cannabis route affects anxiety differently across ADI groups.
- •Higher income was associated with better patient reported outcomes.
- •Aspects of neighborhood disadvantage may elucidate cannabis use and health differences.
Article notes
Untitled section
Received 2026 Feb 5; Revised 2026 May 5; Accepted 2026 May 20; Collection date 2026 Jun.
1.Introduction
Medical advancements in cancer treatments have reduced cancer’s health risks; however, access to effective symptom management remains sparse for patients living in disadvantaged areas. Neighborhood disadvantage refers to residential areas that face systemic and structural inequities, including limited access to socioeconomic resources (e.g., adequate healthcare, healthy food) and high concentrations of environmental stressors (e.g., poverty, unemployment) stemming from historic discriminatory practices (Riley, 2018). In the oncology context, neighborhood disadvantage has been negatively associated with adverse health outcomes and suboptimal cancer care (Chiang et al., 2023, Hassan et al., 2023), characterized by lower quality of life and greater time to care (Li et al., 2019, Unger et al., 2021). Accordingly, neighborhood disadvantage may have important implications for patients residing in disadvantaged areas, making it crucial to further delineate its association with patient-reported outcomes (PROs).
Patients living in disadvantaged neighborhoods may be exposed to profound challenges, including lower income and safety concerns. Collectively, these contextual factors are associated with chronic stress (Santiago et al., 2011), compound cancer-related stressors (e.g., screenings) (Kurani et al., 2020), and may in turn exacerbate common cancer-related PROs such as pain (Choi et al., 2023), anxiety and depression (Balogun et al., 2024, Goel et al., 2023, Segrin et al., 2025), and sleep disturbance (Segrin et al., 2025). For instance, mood disturbance may arise due to financial insecurity and having to manage daily stressors (Guan et al., 2022, Ross, 2000). Moreover, environmental factors in disadvantaged neighborhoods, such as crime and noise, may disrupt sleep (Fuller-Rowell et al., 2016, Hill et al., 2009). Additionally, stress associated with living in disadvantaged neighborhoods may contribute to adverse physiological outcomes like increased inflammation, thereby increasing the risk of chronic pain (Hannibal and Bishop, 2014). Furthermore, the influence of socioeconomic factors on PROs may be sustained by limited access to health resources, including inequitable access to pain prescriptions and mental health services for low-income patients (Meghani et al., 2020, Wen et al., 2025). Given the significant impact of neighborhood disadvantage on PROs, it is also essential to investigate its role in contemporary symptom management options, such as cannabis.
One-third of patients use cannabis to manage cancer and cancer-treatment-related symptoms (Ellison et al., 2024). Cannabis has purported medicinal properties; however, findings on its effects have been mixed. Evidence suggests cannabis reduces pain and sleep issues (Blake et al., 2018, Wang et al., 2021), but may have negligible (Crichton et al., 2024, Zaki et al., 2017) or adverse effects on anxiety and mood (Steigerwald et al., 2018, Worster et al., 2022). Indeed, one study (N = 267) found that compared to those who did not use cannabis, patients who reported using cannabis to treat physiological and psychological symptoms exhibited greater symptom severity on these outcomes (Azizoddin et al., 2023), which is in contrast to reported symptom improvement among patients using cannabis in other studies (McClure et al., 2023, Salz et al., 2023). In light of these mixed findings, the emerging literature has increasingly focused on examining sociodemographic and neighborhood-level factors that may help elucidate the complex relationship between cannabis use and PROs.
Prior work indicates that underserved communities may be more likely to engage in cannabis use patterns associated with higher risk, which may be driven by neighborhood-level factors (Ashare et al., 2023, Azizoddin et al., 2023, Jeffers et al., 2021). For instance, Black patients may be more likely to purchase cannabis from non-licensed retailers (Ashare et al., 2023), and these retailers are more prevalent in disadvantaged neighborhoods (Rhew et al., 2022). Moreover, low-income patients are affected by cannabis pricing and accessibility (Ashare et al., 2023). This has important implications, as obtaining cannabis from non-licensed retailers, likely because of accessibility, may confer a greater risk of exposure to harm due to the lack of regulatory oversight. Additionally, combustible cannabis formulations, which are more prevalent in socioeconomically disadvantaged neighborhoods (Brasky et al., 2025), may increase the risk of carcinogen exposure (Jett et al., 2018). Taken together, neighborhood disadvantage may be associated with cannabis source and route, and these cannabis-related variables may disproportionately affect patients living in disadvantaged neighborhoods due to a theorized elevated risk of cannabis-related adverse outcomes stemming from limited product safety and established health risks.
These adverse outcomes may also manifest in PROs; however, this topic remains understudied in oncology populations. Further research would be beneficial for clarifying the role of cannabis in PROs and for further delineating how the influence of neighborhood disadvantage on PROs differentially affects patients residing in disadvantaged neighborhoods compared with those who do not. To address this gap, the present study evaluates whether neighborhood disadvantage is associated with cannabis source or route and whether these cannabis variables modify the association between neighborhood disadvantage and PROs. We utilized the ADI (2023) to investigate the association of neighborhood disadvantage with cannabis route (i.e., inhaled [smoked, vaped], consumed/other [e.g., topicals, pills]) and source (i.e., regulated [licensed dispensaries] vs. unregulated [family and friends, unlicensed sellers, Native American reservations, smoke shops]). Separately, we assessed the relationship between ADI and key PROs (i.e., pain, sleep quality, anxiety, depression, and quality of life) across time (i.e., visit; coded Month 0, 1, 2, and 3). We hypothesized that greater neighborhood disadvantage would be associated with: (1) obtaining cannabis from unregulated versus regulated sources; (2) using inhaled formulations compared to consumed/other formulations; and (3) greater symptom severity in PROs and greater increases in symptom severity in PROs across time. An exploratory aim tested whether cannabis source and route moderated the association between neighborhood disadvantage and PROs. Additionally, we assessed the relationship between individual socioeconomic characteristics (that were retained as covariates) and PROs.
2.Methods
2.1.Study design, population, & setting
This is a secondary analysis of an ongoing multi-site study (University at Buffalo [UB], University of Pennsylvania [UP], Thomas Jefferson University [TJU], NCT06037681) investigating the association among PROs and opioid and cannabis use over 12 months. Electronic medical records (EMR) were used to identify potentially eligible participants. Additional recruitment methods included flyers and the Buffalo Research Registry. Data collection began in March 2023. Participants in the current sample were scheduled to complete Month 3 of the parent study by March 28, 2025 and were recruited from the overall study sample at this time (N = 361). This study was approved by the UB Institutional Review Board.
Eligible participants were at least 21 years old (the legal age to purchase cannabis), receiving outpatient treatment for solid tumors, and self-reported cancer or cancer treatment-related pain for greater than 30 days in the past three months. Skin cancer was the only cancer diagnosis excluded. At baseline (Month 0), patients were categorized into one of two groups: people who used cannabis (self-reported at least weekly cannabis use over the past month) and people who did not use cannabis (self-reported no cannabis use in the past three months). The parent study aimed to recruit a sample representing 50% of each group.
The present analyses were restricted to the first four months of data collection (Month 0 – Month 3). Participants were included if they used cannabis at baseline, had at least one time point with cannabis route and source information to assess between-subject differences, and at least two time points for one of the PRO variables to identify change over time.
2.2.Study procedures
Participants completed an initial screening to determine eligibility. Eligible participants completed a baseline visit, including informed consent and coordinator-administered surveys. Participants then completed surveys at monthly study visits and seven days of ecological momentary assessment (EMA) at each time point (4 surveys/per day: 1 morning and 3 random surveys) to assess PROs, cannabis, and opioid use. Following baseline, EMA data were collected for one week per month for 3 months (total of 21 days). EMAs were administered either through a smartphone-based application (mEMA; ilumivu, Inc) or via REDCap links sent through a third-party application (e.g., Mosio, Inc).
2.3.Measures
2.3.1.Sociodemographic and clinical data
Sociodemographic variables (e.g., race, income) were ascertained at baseline. Medical data (e.g., cancer stage, treatment status, cancer type) were obtained from EMRs.
2.3.2.Cannabis use
Cannabis Frequency. Past month average cannabis frequency was collected during monthly study visits.
Cannabis Source and Route. Categorization of cannabis route (i.e., inhaled vs. consumed/other route) was based on both prior literature (Ellison et al., 2024) and observed patterns in the current data. Categorization of cannabis source (i.e., regulated vs unregulated) was based on two primary reasons: (1) dispensaries are the most common source of obtaining cannabis (42.5%) (Myers et al., 2024); (2) and are the only regulated cannabis source.
Data on cannabis source and route were collected at monthly study visits and EMAs. Source and route were first determined via EMA data to reduce potential recall bias. If EMA data were missing, visit data were used. For participants who reported obtaining cannabis from more than one source, the source that was reported most frequently was designated as the primary cannabis source. In cases where there was an equal number of regulated and unregulated sources, the regulated source was selected as the primary source based on the premise that access to regulated sources may reflect higher levels of accessibility. A similar process was conducted to identify the primary route. If routes were reported with equal frequency, inhaled formulations took precedence over other methods based on literature indicating that inhaled formulations may be associated with worse health outcomes (Jett et al., 2018). Few participants showed within-person variability in cannabis source (16.0%) and route (29.3%), reflecting that the majority of participants didn’t change their route or source over time.
Cannabis Use Disorder Identification Test- Revised (CUDIT-R). The 8-item CUDIT-R assessed risk for hazardous cannabis use (Adamson et al., 2010, Adamson and Sellman, 2003). Scores range from 0 to 32, with a score of 8 or more indicating “hazardous cannabis use” and 13 suggesting “possible cannabis use disorder.” Internal consistency in our study was poor (Cronbach’s α=0.55) but excellent in prior studies (Cronbach’s α=0.91) (Adamson et al., 2010).
2.3.3.Neighborhood disadvantage
The Area Deprivation Index (ADI) is a well-validated measure of neighborhood disadvantage and uses 17 metrics from census-tract data to produce a composite score (Kind and Buckingham, 2018, Kind et al., 2014). In prior oncology studies, the ADI has been linked to PROs such as anxiety and has informed recommendations for its use in oncology populations, including recommended screening for neighborhood-level socioeconomic issues and assisting with patient navigation to improve targeted patient care (Rosenzweig et al., 2021). The ADI defines a neighborhood as a census tract, which was identified here by entering an individual’s full address (including zip code) into the ADI’s website. Neighborhood disadvantage is quantified by the ADI’s national decile, which is a composite score that combines key socioeconomic indicators (e.g., income, education) and is weighted and standardized to produce an overall score for each neighborhood (0–100 %ile; higher scores, greater disadvantage). As in prior work (Goel et al., 2023), we used a categorical ADI in all analyses, which partitioned participants’ national deciles into three tertiles (0–33.3% Least Disadvantaged; 33.3–66.6% Moderately Disadvantaged; 66.6–100% Most Disadvantaged).
2.3.4.Patient-reported outcomes
All PROs data were collected from study visits.
Pain. The Brief Pain Inventory Short Form (BPI-SF) assesses clinical pain (Caraceni et al., 1996, Cleeland, 1991). It includes two domains: pain severity and interference with sub-items scored on a scale from 0 (no pain/no interference) to 10 (worse pain imaginable/complete interference). In our study, internal consistency was excellent (Cronbach α=.95).
Sleep Quality. The Pittsburgh Sleep Quality Index (PSQI) is a 9-item measure that assesses sleep quality and patterns over the past month. The outcome is the global PSQI score, where a sum of 5 or greater indicates poor sleep (Buysse et al., 1989). The PSQI had poor internal consistency in our study (Cronbach α=.63), likely due to the variable sleep quality and duration patients with cancer often experience.
Anxiety. The Generalized Anxiety Disorder-7 (GAD-7) is a 7-item measure that assesses clinical severity of anxious symptoms (Spitzer et al., 2006). Total scores ranged from 0 to 21, with the clinical threshold being a score of 10 or higher. Internal consistency in our study was excellent (Cronbach α=.93).
Depression. The Patient Health Questionnaire-8 (PHQ-8) is an 8-item measure of clinical depression (Kroenke et al., 2009). The scaling is the same as the GAD-7. Total scores ranged from 0 to 24, with a clinical threshold of 10 or higher. Internal consistency in our study was good (Cronbach α=.88).
Quality of Life. The Functional Assessment of Cancer Therapy-General 7 (FACT-G7) scale is a 7-item cancer-specific quality of life index adapted from the 27-item FACT-G (Cella et al., 1993, Mah and Swami, 2020). In our study, internal consistency was excellent (Cronbach α=.91). It has also been shown to decrease with worsening disease in patients with cancer (Pearman et al., 2014).
2.3.5.Other substance use
Opioid Use. Opioids are frequently prescribed to patients with cancer for pain management (Mudumbai et al., 2024). We assessed opioid utilization by querying participants about their opioid consumption during the past month before each visit. Responses were coded as 1 = ‘yes’ or 0 = ‘no’.
Alcohol and Tobacco Use. Because alcohol and tobacco use correlate with cannabis use (Lemyre et al., 2019, Yurasek et al., 2017), these were measured at baseline as statistical control variables. The Alcohol Use CAGE is a 4-item, self-report measure of alcohol dependence (Ewing, 1984) with high-test retest reliability (.80–.95) and adequate correlations (.48–.70) with other screening instruments (Dhalla and Kopec, 2007). Participants reported past-month tobacco use during visits. No patients reported current alcohol or tobacco use; consequently, these variables were not included in analyses.
2.4.Data analysis
2.4.1.Statistical power
Because the goal of this study is to generate data to inform future studies, a power analysis was not conducted.
2.4.2.Statistical analysis
Descriptive statistics were calculated to compare patients across ADI tertiles using bivariate analyses (Chi-square for categorical data; one-way ANOVAs for continuous data). Analyses that violated Bartlett’s assumption of equal variances were then analyzed using non-parametric tests. Covariates associated with study outcomes with a p < .1 were retained in subsequent models. For all analyses, including neighborhood disadvantage, the reference group was the most disadvantaged group. Logistic regressions were used to determine whether neighborhood disadvantage was associated with primary cannabis source (regulated vs. unregulated) and route (inhaled vs. consumed/other), respectively. Because few patients showed within-person variability for cannabis variables, the logistic regression analyses modeled between-person variability only.
The PROs data conformed to a multilevel structure with observations at each visit nested within patients. We estimated two-level mixed-effect models with neighborhood disadvantage as a categorical predictor to describe its association with each PRO, including random effects of visit. A series of models were fitted to identify the most appropriate model for each PRO. First, we estimated the intercept-only model for each PRO and then added patient as a random effect to evaluate differences in baseline-level outcomes. Next, we added visit (i.e., time, coded Month 0, 1, 2, 3) as a fixed effect to evaluate the average change in a specified PRO and then added a random effect for visit to examine individual differences in change. Likelihood-ratio tests were used to determine the best-fitting model for each PRO. Results yielded a random-intercept only model for anxiety and depression, a random intercept and fixed effect of visit for pain interference, sleep quality, and quality of life, and a random intercept and random effect of visit for pain severity. All models used maximum likelihood estimation. Models with pain severity as the outcome had an unstructured covariance structure to allow for correlations between random effects. Final models included all study covariates (race, income, and site).
We tested the interaction between neighborhood disadvantage and visit predicting PROs when nested model tests supported the inclusion of visit. Exploratory analyses included testing the interaction between neighborhood disadvantage and cannabis source and route, respectively, on each PRO. Post hoc analyses for covariates were also conducted.
3.Results
3.1.Sociodemographic, cannabis, and clinical characteristics
Table 1 contains sociodemographic and cannabis use characteristics by neighborhood disadvantage. Fig. 1 outlines the participant flowchart. The sample comprised 106 patients with cancer and was majority female (56.6%), white (65.1%), retired or disabled (58.5%), low (<$35,000) to lower-middle-income ($35,000-$99,000; 79.1%), and used cannabis daily or more (61.9%). Relative to those in the least and moderately disadvantaged groups, more patients from the most disadvantaged group were African-American/Black (p < .001), low-income (p < .01), and found it difficult to get by on their current income (p = .01). UB enrolled more patients from the most disadvantaged group than other sites (p < .01). Additional characteristics were similar across groups (see Table 1 and Supplemental Table 1).
| Characteristics | Most Disadvantaged | Moderately Disadvantaged | Least Disadvantaged | Overall | p-value | ||
|---|---|---|---|---|---|---|---|
| (n = 44) | (n = 34) | (n = 28) | (n = 106) | ||||
| Site n (%) | < .01 | ||||||
| University at Buffalo (UB) | 18 (40.9%) | 16 (47.1%) | 3 (10.7%) | 37 (34.9%) | |||
| University of Pennsylvania (UP) | 10 (22.7%) | 6 (17.7%) | 16 (57.1%) | 32 (30.2%) | |||
| Thomas Jefferson University (TJU) | 16 (36.4%) | 12 (35.3%) | 9 (32.1%) | 37 (34.9%) | |||
| Age mean (SD) | 54 (13) | 55 (12) | 55 (11) | 55 (12) | .88 | ||
| Sex n (female %) | 24 (54.6%) | 22 (64.7%) | 14 (50.0%) | 60 (56.6%) | .48 | ||
| Race n (black %) | 27 (61.4%) | 6 (17.7%) | 4 (14.3%) | 37 (34.9%) | < .001 | ||
| State n (%) | .04 | ||||||
| Pennsylvania | 23 (52.3%) | 13 (38.2%) | 22 (78.6%) | 58 (54.7%) | |||
| New York | 18 (40.9%) | 16 (47.1%) | 3 (10.7%) | 37 (34.9%) | |||
| New Jersey | 2 (4.6%) | 4 (11.8%) | 3 (10.7%) | 9 (8.5%) | |||
| Other | 1 (2.3%) | 1 (2.9%) | 0 (0.0%) | 2 (1.9%) | |||
| Marital Status n (%) | .24 | ||||||
| Married or Cohabitating | 17 (38.6%) | 20 (58.8%) | 17 (60.7%) | 54 (50.9%) | |||
| Divorced, Separated, or Widowed | 13 (29.6%) | 9 (26.5%) | 6 (21.4%) | 28 (26.4%) | |||
| Never Married | 14 (31.8%) | 5 (14.7%) | 5 (17.9%) | 24 (22.6%) | |||
| Education n (%) | .17 | ||||||
| High School Degree or Lower | 13 (29.6%) | 4 (12.1%) | 4 (14.3%) | 21 (20.0%) | |||
| Some College/Technical School | 20 (45.5%) | 15 (45.5%) | 11 (39.3%) | 46 (43.8%) | |||
| College Graduate or Greater | 11 (25.0%) | 14 (42.4%) | 13 (46.4%) | 38 (36.2%) | |||
| Employment n (%) | .23 | ||||||
| Employed | 8 (18.2%) | 9 (26.5%) | 9 (32.1%) | 26 (24.5%) | |||
| Retired or Disabled | 32 (72.7%) | 17 (50.0%) | 13 (46.4%) | 62 (58.5%) | |||
| Unemployed | 4 (9.1%) | 6 (17.7%) | 4 (14.3%) | 14 (13.2%) | |||
| Other | 0 (0.0%) | 2 (5.9%) | 2 (7.1%) | 4 (3.8%) | |||
| Health Insurance n (%) | .22 | ||||||
| Employer Plan | 10 (22.7%) | 14 (41.2%) | 13 (46.4%) | 37 (34.9%) | |||
| Private | 2 (4.6%) | 2 (5.9%) | 3 (10.7%) | 7 (6.6%) | |||
| Medicaid/Medicare | 29 (65.9%) | 16 (47.1%) | 12 (42.9%) | 57 (53.8%) | |||
| Other | 3 (6.8%) | 2 (5.9%) | 0 (0.0%) | 5 (4.7%) | |||
| Income n (%) | .01 | ||||||
| Less than $35,000 | 25 (56.8%) | 10 (30.3%) | 5 (17.9%) | 40 (38.1%) | |||
| $35,000 - $99,000 | 15 (34.1%) | 15 (45.5%) | 13 (46.4%) | 43 (41.0%) | |||
| $100,000 or Greater | 4 (9.1%) | 8 (24.2%) | 10 (35.7%) | 22 (21.0%) | |||
| Income Satisfaction n (%) | .01 | ||||||
| Living Comfortably on Current Income | 4 (9.1%) | 12 (36.4%) | 13 (46.4%) | 29 (27.6%) | |||
| Getting by on Current Income | 22 (50.0%) | 8 (24.2%) | 9 (32.1%) | 39 (37.1%) | |||
| Finding it Difficult on Current Income | 13 (29.6%) | 7 (21.2%) | 4 (14.3%) | 24 (22.9%) | |||
| Finding it Very Difficult on Current Income | 5 (11.4%) | 6 (18.2%) | 2 (7.1%) | 13 (12.4%) | |||
| Cannabis Use Frequency n (%) | .96 | ||||||
| A Few Times a Month | 3 (6.8%) | 2 (6.1%) | 3 (10.7%) | 8 (7.6%) | |||
| Once to Several Days a Week | 13 (29.6%) | 10 (30.3%) | 9 (32.1%) | 32 (30.5%) | |||
| Daily or More | 28 (63.6%) | 21 (63.6%) | 16 (57.1%) | 65 (61.9%) | |||
| Used Multiple Cannabis Routes n (yes %) | 16 (36.4%) | 9 (26.5%) | 6 (21.4%) | 31 (29.3%) | .36 | ||
| Went to Multiple Cannabis Sources n (yes %) | 8 (18.2%) | 5 (14.7%) | 4 (14.3%) | 17 (16.0%) | .88 | ||
| Cannabis Dependence mean (SD) | 7 (4.0) | 7 (3.0) | 5 (2.0) | 6 (3.0) | .10 | ||
| Opioid Prescription Receipt n (yes %) | 27 (61.4%) | 23 (67.7%) | 23 (82.1%) | 73 (68.9%) | .18 | ||
3.2.Cannabis source and route
3.2.1.Cannabis source
A majority of participants reported obtaining cannabis from regulated versus unregulated sources (53.8% vs. 46.2%). Cannabis source by ADI groups is reflected in Fig. 2. The odds of obtaining cannabis from regulated versus unregulated sources were comparable between patients from the least and most disadvantaged groups (OR=0.92, 95%CI [0.2,3.5], p = .90) and patients from moderately and most disadvantaged groups (OR=0.84, 95%CI [0.3,2.6], p = .75). Patients enrolled at UP (OR=8.27, 95%CI [2.2,30.5], p < .01) or TJU (OR=6.46, 95%CI [1.9,21.4], p < .01) were significantly more likely to obtain cannabis from regulated sources compared to patients from UB. Patients who were middle-upper-income (>$100,000) and lower-middle-income were more likely to obtain cannabis from regulated sources compared to low-income patients (OR=5.3, 95%CI [1.3,22.1], p = .02; OR=2.9, 95%CI [1.0,8.3], p = .05, respectively).
3.2.2.Cannabis route
Most participants reported using inhaled cannabis formulations compared to consumed/other formulations (53.8% vs. 46.2%). Shown in Fig. 2, patients from the least and most disadvantaged groups (OR=2.51, 95%CI [0.7,8.7], p = .15) and moderately and most disadvantaged groups (OR=2.10, 95%CI [0.7,6.2], p = .18) had comparable likelihood of using consumed/other versus inhaled formulations. Compared to low-income patients, lower-middle-income patients were 4.1 times more likely to use consumed/other formulations than inhaled (95%CI [1.4,11.4], p = .01); however, this effect was not observed for middle-higher-income patients (OR=2.89, 95%CI [0.8,10.5], p = .11).
3.3.Multilevel regressions predicting PROs
Main Effects. The main effect of neighborhood disadvantage was non-significant in all models, indicating that patients in the moderately or least disadvantaged groups did not differ in mean PRO scores from patients in the most disadvantaged group (p’s = .13–.79 and.27–.96, respectively). Visit was significant for pain interference (b=-0.18, 95%CI [-0.3,-0.04], p = .01), sleep quality (b=-0.34, 95%CI [-0.6,-0.1], p < .01), and marginal for quality of life (b=0.69, 95%CI [-0.05,1.4], p = .07). On average, outcomes improved over time (i.e., lower pain interference and higher sleep quality and quality of life). Patient income was a significant predictor in all models, and marginal for sleep quality and anxiety. Compared to the low-income group, being lower-middle-income or middle-upper-income was associated with better outcomes for pain interference and severity, quality of life, and depression (p’s = .01–.03 and <.001–.04, respectively). Associations for neighborhood disadvantage and all covariates with PROs are depicted in Table 2.
| β | 95% CI | p-value | |||
|---|---|---|---|---|---|
| Pain Interference | |||||
| Neighborhood Disadvantage | |||||
| Moderately Disadvantaged | -0.81 | (-1.8,0.2) | .13 | ||
| Least Disadvantaged | -0.29 | (-1.5,0.9) | .64 | ||
| Income | |||||
| $35,000 - $99,000 | -1.14 | (-2.1,-0.2) | .02 | ||
| $100,000 or Greater | -2.35 | (-3.6,-1.1) | < .001 | ||
| Race | |||||
| African American/Black | -0.2 | (-1.3,0.9) | .71 | ||
| Site | |||||
| University of Pennsylvania (UPenn) | 0.77 | (-0.4,1.9) | .19 | ||
| Thomas Jefferson University (TJU) | 0.69 | (-0.4,1.7) | .20 | ||
| Pain Severity | |||||
| Neighborhood Disadvantage | |||||
| Moderately Disadvantaged | -0.23 | (-1.6,0.7) | .63 | ||
| Least Disadvantaged | 0.09 | (-1.0,1.2) | .86 | ||
| Income | |||||
| $35,000 - $99,000 | -1.11 | (-2.0,-0.3) | .01 | ||
| $100,000 or Greater | -2.19 | (-3.3,-1.1) | < .001 | ||
| Race | |||||
| African American/Black | 0.91 | (-0.05,1.9) | .06 | ||
| Site | |||||
| University of Pennsylvania (UPenn) | 0.80 | (-0.2,1.8) | .13 | ||
| Thomas Jefferson University (TJU) | 0.38 | (-0.6,1.3) | .43 | ||
| Sleep Quality | |||||
| Neighborhood Disadvantage | |||||
| Moderately Disadvantaged | 0.59 | (-1.1,2.3) | .49 | ||
| Least Disadvantaged | -0.05 | (-2.0,1.9) | .96 | ||
| Income | |||||
| $35,000 - $99,000 | -1.49 | (-3.0,0.05) | .06 | ||
| $100,000 or Greater | -1.72 | (-3.7,0.3) | .10 | ||
| Race | |||||
| African American/Black | 0.24 | (-1.5,2.0) | .79 | ||
| Site | |||||
| University of Pennsylvania (UPenn) | -0.04 | (-1.9,1.8) | .97 | ||
| Thomas Jefferson University (TJU) | 0.07 | (-1.6,1.8) | .94 | ||
| Quality of Life | |||||
| Neighborhood Disadvantage | |||||
| Moderately Disadvantaged | 4.96 | (-3.6,13.5) | .26 | ||
| Least Disadvantaged | 5.51 | (-4.2,15.3) | .27 | ||
| Income | |||||
| $35,000 - $99,000 | 8.46 | (0.7,16.3) | .03 | ||
| $100,000 or Greater | 10.67 | (0.5,20.8) | .04 | ||
| Race | |||||
| African American/Black | 3.1 | (-5.8,12.0) | .49 | ||
| Site | |||||
| University of Pennsylvania (UPenn) | -0.62 | (-10.1,8.8) | .90 | ||
| Thomas Jefferson University (TJU) | -4.77 | (-13.3,3.7) | .27 | ||
| Anxiety | |||||
| Neighborhood Disadvantage | |||||
| Moderately Disadvantaged | -0.34 | (-2.9,2.2) | .79 | ||
| Least Disadvantaged | -1.24 | (-4.1,1.7) | .41 | ||
| Income | |||||
| $35,000 - $99,000 | -2.09 | (-4.5,0.3) | .09 | ||
| $100,000 or Greater | -1.20 | (-4.3,1.9) | .44 | ||
| Race | |||||
| African American/Black | -1.4 | (-4.1,1.2) | .28 | ||
| Site | |||||
| University of Pennsylvania (UPenn) | -0.50 | (-3.3,2.3) | .73 | ||
| Thomas Jefferson University (TJU) | 2.41 | (-0.1,4.9) | .06 | ||
| Depression | |||||
| Neighborhood Disadvantage | |||||
| Moderately Disadvantaged | -0.94 | (-3.6,1.8) | .50 | ||
| Least Disadvantaged | -0.91 | (-4.1,2.3) | .58 | ||
| Income | |||||
| $35,000 - $99,000 | -3.16 | (-5.6,-0.7) | .01 | ||
| $100,000 or Greater | -3.54 | (-6.7,-0.3) | .03 | ||
| Race | |||||
| African American/Black | -1.12 | (-3.9,1.7) | .43 | ||
| Site | |||||
| University of Pennsylvania (UPenn) | -0.29 | (-3.3,2.8) | .85 | ||
| Thomas Jefferson University (TJU) | 1.91 | (-0.7,4.6) | .16 | ||
Interactions. There was a significant interaction between neighborhood disadvantage and visit for both pain interference (b=-1.37, 95%CI [-2.4,-0.3], p = .01) and sleep quality (b=1.74, 95%CI [-0.001,3.5], p = .05). For pain interference, patients in the moderately disadvantaged group exhibited greater declines in pain interference scores over time, with significantly lower levels at Month 3 compared to those in the most disadvantaged group (Fig. 3a). In contrast, the most disadvantaged group demonstrated minimal change across visits. Simple effects analyses indicated that patients in the moderately disadvantaged group experienced reductions in pain interference scores at all time points (p = .001–.03). For sleep quality, there was overall improvement over time among patients in the most disadvantaged group; however, this improvement was attenuated among those in the least disadvantaged group (Fig. 3b). Indicated by simple effects, there were improvements in sleep disturbance at Month 3 for patients in the most disadvantaged group (p < .05), with smaller changes observed in the least disadvantaged group. The interaction was non-significant for pain severity (p = .26–.97) and quality of life (p = .13–.69). Because the best-fitting model for anxiety, depression, and pain severity did not include visit as a predictor, the interaction was not tested.
3.4.Cannabis source added to main effect and interaction models
Main Effects. Cannabis source was not a significant predictor for any PRO (p’s = .24–.93). Adding cannabis source did not change the main effects of visit and income and marginal effects of race described above and in Table 2. However, when controlling for cannabis source, race significantly predicted pain severity, where Black patients had significantly higher pain severity scores compared to White patients (b=0.94, 95%CI [-0.01,1.9], p = .05). Additionally, site significantly predicted anxiety such that patients from TJU had significantly higher anxiety scores than patients recruited from UB (b=2.63, 95%CI [-0.03,5.3], p = .05).
Interactions. The neighborhood disadvantage x cannabis source interaction was non-significant across all models (p’s = .14–.81).
3.5.Cannabis route added to main effect and interaction models
Main Effects. Cannabis route was not a significant predictor for any PRO (p’s = .40–.76) and adding it to the model did not significantly change prior main effects of visit and covariates.
Interactions. There was a significant neighborhood disadvantage x cannabis route interaction on anxiety scores, across levels of neighborhood disadvantage (b=-5.32, 95%CI [-10.2,-0.4], p = .03). For patients using inhaled formulations, moderate levels of neighborhood disadvantage were associated with higher anxiety compared to the highest level of neighborhood disadvantage (Fig. 4). In contrast, among patients using consumed/other formulations, this association was reversed, as moderate levels of neighborhood disadvantage were associated with lower anxiety scores. Simple effects were non-significant (p = .12–.51). The interaction effect was non-significant in all other models (p’s = .17–.97).
3.6.Supplemental analyses
Income is a significant patient-level factor that unveils health disparities, is an indicator of neighborhood disadvantage, and was associated with study outcomes. Therefore, we assessed its relationship with sociodemographic, cannabis, and clinical variables in supplemental analyses.
4.Discussion
Our study explored differences in cannabis route and source and PROs among patients with cancer across varying levels of neighborhood disadvantage. Changes in pain interference and sleep quality scores over time differed between levels of neighborhood disadvantage. Cannabis route moderated the relationship between neighborhood disadvantage and anxiety. Inconsistent with the extant literature, neighborhood disadvantage was not associated with our outcomes when accounting for individual characteristics. Given the novelty of exploring cannabis use among oncology populations and its relationship with neighborhood disadvantage, these findings may guide future oncology studies on PROs and cannabis access.
We assessed the relationship between neighborhood disadvantage and cannabis route, given that SDoH are associated with cannabis access (Ashare et al., 2023) and that inhaled cannabis formulations may pose potential health risks (Jett et al., 2018). Although non-significant, our findings were in the predicted direction as patients from more advantaged neighborhoods were more likely to use consumed/other formulations than inhaled, compared to those from the most disadvantaged neighborhoods. This aligns with literature suggesting that inhaled formulations are more prevalent in higher deprivation areas (Brasky et al., 2025) and that engaging in risky substance use behaviors is more common among patients facing greater social disparities (Simmons et al., 2016). Additionally, inhaled formulations may cost less, enhancing accessibility to more disadvantaged patients.
Our findings indicate that patients in the moderately disadvantaged group showed greater improvements in pain interference scores over time than those in the most disadvantaged group. This pattern is consistent with prior literature suggesting that residence in impoverished neighborhoods is associated with more pronounced pain disparities (Maly and Vallerand, 2018). Additionally, salient sociodemographic characteristics may have differentiated these groups. Black and low-income patients disproportionately experience more severe pain-related disparities, which are linked to limited access to care (Meghani et al., 2020, Youn et al., 2024). In our sample, patients in the moderately disadvantaged group were predominantly White and had higher incomes than the most disadvantaged group, who were largely Black and low-income. Consequently, when experiencing pain interference, they may have had greater socioeconomic resources to facilitate symptom management changes compared with the most disadvantaged group. The present findings further substantiate this hypothesis. In our sample, higher income was uniquely associated with lower average pain interference scores, and prior research has similarly identified this association (Peoples et al., 2024). In addition to pain interference, there were also group differences in sleep quality scores over time. Patients in the least disadvantaged group demonstrated worse sleep quality scores over time relative to those in the most disadvantaged group, a pattern that is inconsistent with existing literature (Fuller-Rowell et al., 2016, Troxel et al., 2018). Further investigation of clinical, sociodemographic, and cannabis-related factors may help to elucidate these unexpected findings.
Patients using inhaled formulations and residing in moderately disadvantaged neighborhoods had higher anxiety scores than those residing in the most disadvantaged neighborhoods; this relationship was reversed in patients using consumed/other formulations. While existing literature links neighborhood disadvantage to anxiety (Balogun et al., 2024), research on cannabis’ impact on anxiety remains mixed (Van Ameringen et al., 2020). Several factors may explain these findings, such as patient-level factors, including income and cannabis use patterns, like quantities consumed (Ko et al., 2016). Patients who use consumed/other formulations and reside in disadvantaged neighborhoods may be under-consuming these products or taking an ineffective dose, given their higher price point and limited access. Inhaled formulations may be used more consistently among patients residing in the most disadvantaged neighborhoods due to their affordability and accessibility, likely reducing their anxiety. Conversely, patients using consumed/other formulations residing in moderately disadvantaged neighborhoods had lower anxiety scores than those in the most disadvantaged group. Understanding cannabis characteristics like dosage and effectiveness, and their relation to socioeconomic factors, may help delineate these trends in anxiety. These postulations are speculative, and further research is required.
The interaction between neighborhood disadvantage and cannabis route was non-significant for all other PROs. Given that all participants used cannabis, this may have attenuated our ability to detect effects due to self-medication of PROs with cannabis use, but this was not tested, given the absence of a control group who did not use cannabis. Meghani and colleagues (2021) found that racial disparities in self-reported pain among patients with cancer were no longer significant after accounting for cannabis use. Thus, in our study, cannabis may have minimized potential disparities in PROs.
Descriptively, more patients from the least disadvantaged group obtained cannabis from regulated sources than those from moderately and most disadvantaged groups. Controlling for covariates, this effect dissipated. Since neighborhood disadvantage did not independently predict this or other outcomes in our study, it is important to explore how patient-level factors may contribute to cannabis access disparities. For instance, dispensaries are typically more expensive than unlicensed markets because of their higher-priced products and greater cannabis tax on people who purchase cannabis (Han and Shi, 2025). Thus, income may explain our descriptive differences in cannabis source across levels of neighborhood disadvantage. This is supported by our finding that income independently predicted cannabis source and route and PROs, consistent with prior studies (Walker et al., 2021, Youn et al., 2024). Thus, income should be explored as a predictor in future work to better contextualize differences in cannabis use and PROs among patients with cancer. Additionally, given that household income is an indicator of neighborhood disadvantage, exploring income as a community-level factor may help explain differences across levels of neighborhood disadvantage.
4.1.Limitations
Several limitations should be considered. Since ADI uses census-tract data, it may provide similar scores to neighborhoods that share a zip code, not accounting for potential variability in SDoH that differentiates them. Our sample size may not have been adequate to identify many possible associations from patient-level ADI indicators, especially in multilevel analyses. Thus, future studies should recruit larger samples. Similarly, there was less representation from the least disadvantaged group in the current sample, which may have skewed analyses. All data were self-reported and subject to recall bias. Similarly, some patients may have been unable to discern between regulated (e.g., licensed dispensaries) and unregulated sources (e.g., smoke shops selling cannabis). Indeed, a previous study found that participants misclassified nicotine smoke and vape shops (Giovenco, 2018). Although descriptive differences between the least and most disadvantaged groups on cannabis source in our study suggest this may not have posed as a substantial limitation, it should be investigated in future work. Lastly, our findings may not be generalizable to oncology populations residing outside of the East Coast, who may have distinct structural differences that contribute to neighborhood disadvantage, or general populations who may differ in symptom severity and cannabis-use patterns from oncology populations.
5.Conclusion
Our study contributes to the emerging literature on cannabis use and inequities in oncology populations. Preliminary findings suggest that sociodemographic factors, like income, may help elucidate observed differences in PROs. Cannabis route emerges as a critical variable warranting further investigation, as it may explain variations in PROs and be a target for future intervention research. Future studies should examine behavioral patterns associated with cannabis use to enhance our understanding of the variability in PROs and cannabis consumption among patients with cancer.
Contributors
BG, RA, and SM developed the manuscript concept. BG and SG conducted data analysis. All authors drafted and edited the initial manuscript. All authors approved the final version of the manuscript for submission.
Declaration of Generative AI and AI-assisted technologies in the writing process
During the preparation of this work the authors used Grammarly (Grammarly.com) to assist with grammar, spell-check, and clarity. After using this tool, the authors reviewed and edited the content as needed and takes full responsibility for the content of the published article.
Role of funding source
This research was supported by the National Cancer Institute under award number R01CA270483–02S1.
Declaration of Competing Interest
The authors have no financial conflicts of interest in regard to this research.
Footnotes
Footnote Group
Appendix A.Supplementary material
References
Untitled section
References
- Adamson S.J., Kay-Lambkin F.J., Baker A.L., Lewin T.J., Thornton L., Kelly B.J., Sellman J.D. An improved brief measure of cannabis misuse: the Cannabis Use Disorders Identification Test-Revised (CUDIT-R) Drug Alcohol Depend. 2010;110(1-2):137–143. doi: 10.1016/j.drugalcdep.2010.02.017.
- Adamson S.J., Sellman J.D. A prototype screening instrument for cannabis use disorder: the Cannabis Use Disorders Identification Test (CUDIT) in an alcohol-dependent clinical sample. Drug Alcohol Rev. 2003;22(3):309–315. doi: 10.1080/0959523031000154454.
- Ashare R.L., Turay E., Worster B., Wetherill R.R., Bidwell L.C., Doucette A., Meghani S.H. Social determinants of health associated with how cannabis is obtained and used in patients with cancer receiving care at a cancer treatment center in Pennsylvania. Cannabis. 2023;6(2):1. doi: 10.26828/cannabis/2023/000151.
- Azizoddin D.R., Cohn A.M., Ulahannan S.V., Henson C.E., Alexander A.C., Moore K.N., Holman L.L., Boozary L.K., Sifat M.S., Kendzor D.E. Cannabis use among adults undergoing cancer treatment. Cancer. 2023;129(21):3498–3508. doi: 10.1002/cncr.34922.
- Balogun Z., Gardiner L.A., Li J., Moroni E.A., Rosenzweig M., Nilsen M.L. Neighborhood deprivation and symptoms, psychological distress, and quality of life among head and neck cancer survivors. JAMA Otolaryngol. Head. Neck Surg. 2024;150(4):295–302. doi: 10.1001/jamaoto.2023.4672.
- Blake A., Wan B.A., Malek L., DeAngelis C., Diaz P., Lao N., Chow E., O’Hearn S. A selective review of medical cannabis in cancer pain management. Ann. Palliat. Med. 2018;6(2):S215–S222. doi: 10.21037/apm.2017.08.05. S221S222.
- Brasky T.M., Lee S., McBride B., Newton A.M., Baltic R.D., Wagener T.L., Conroy S., Hays J.L., Stevens E.E., Adib A. Prevalence of cannabis and medication use by indices of residential urbanicity and deprivation among Ohio cancer patients. Cancer Causes & Control. 2025;36(7):719–724. doi: 10.1007/s10552-025-01972-x.
- Buysse D.J., Reynolds III C.F., Monk T.H., Berman S.R., Kupfer D.J. The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res. 1989;28(2):193–213. doi: 10.1016/0165-1781(89)90047-4.
- Caraceni A., Mendoza T.R., Mencaglia E., Baratella C., Edwards K., Forjaz M.J., Martini C., Serlin R.C., De Conno F., Cleeland C.S. A validation study of an Italian version of the Brief Pain Inventory (Breve Questionario per la Valutazione del Dolore) Pain. 1996;65(1):87–92. doi: 10.1016/0304-3959(95)00156-5.
- Cella D.F., Tulsky D.S., Gray G., Sarafian B., Linn E., Bonomi A., Silberman M., Yellen S.B., Winicour P., Brannon J. The functional assessment of cancer therapy scale: development and validation of the general measure. J. Clin. Oncol. 1993;11(3):570–579. doi: 10.1200/JCO.1993.11.3.570.
- Chiang S.N., Finnan M.J., Varagur K., Skolnick G.B., Sacks J.M., Christensen J.M. The association between breast cancer related lymphedema and area deprivation index. J. Reconstr. Microsurg. 2023;39(06):444–452. doi: 10.1055/a-1947-8212.
- Choi Y.Y., Rha S.Y., Park J.S., Song S.K., Lee J. Cancer coping self-efficacy, symptoms and their relationship with quality of life among cancer survivors. Eur. J. Oncol. Nurs. 2023;66 doi: 10.1016/j.ejon.2023.102373.
- Cleeland C. Research in cancer pain. What we know and what we need to know. Cancer. 1991;67(S3):823–827. doi: 10.1002/1097-0142(19910201)67:3+<823::aid-cncr2820671412>3.0.co;2-s.
- Crichton M., Dissanayaka T., Marx W., Gamage E., Travica N., Bowers A., Isenring E., Yates P., Marshall S. Does medicinal cannabis affect depression, anxiety, and stress in people with cancer? A systematic review and meta-analysis of intervention studies. Maturitas. 2024;184 doi: 10.1016/j.maturitas.2024.107941.
- Dhalla S., Kopec J.A. The CAGE questionnaire for alcohol misuse: a review of reliability and validity studies. Clin. Invest. Med. 2007;30(1):33–41. doi: 10.25011/cim.v30i1.447.
- Ellison G.L., Helzlsouer K.J., Rosenfield S.M., Kim Y., Ashare R.L., Blaes A.H., Cullen J., Doran N., Ebbert J.O., Egan K.M. Perceptions, prevalence, and patterns of cannabis use among cancer patients treated at 12 NCI-Designated Cancer Centers. JNCI Monogr. 2024;2024(66):202–217. doi: 10.1093/jncimonographs/lgae029.
- Ewing J.A. Detecting alcoholism: the CAGE questionnaire. JAMA. 1984;252(14):1905–1907. doi: 10.1001/jama.252.14.1905.
- Fuller-Rowell T.E., Curtis D.S., El-Sheikh M., Chae D.H., Boylan J.M., Ryff C.D. Racial disparities in sleep: the role of neighborhood disadvantage. Sleep. Med. 2016;27:1–8. doi: 10.1016/j.sleep.2016.10.008.
- Giovenco D.P. Smoke shop misclassification may cloud studies on vape shop density. Nicotine Tob. Res. 2018;20(8):1025–1026. doi: 10.1093/ntr/ntx136.
- Goel N., Hernandez A.E., Ream M., Clarke E.S., Blomberg B.B., Cole S., Antoni M.H. Effects of neighborhood disadvantage on cortisol and interviewer-rated anxiety symptoms in breast cancer patients initiating treatment. Breast Cancer Res. Treat. 2023;202(1):203–211. doi: 10.1007/s10549-023-07050-7.
- Guan N., Guariglia A., Moore P., Xu F., Al-Janabi H. Financial stress and depression in adults: a systematic review. PLoS One. 2022;17(2) doi: 10.1371/journal.pone.0264041.
- Han B., Shi Y. Multilevel determinants of cannabis prices in legal markets: evidence from products sold in nearly 4000 cannabis dispensaries in the United States. Int. J. Drug Policy. 2025;137 doi: 10.1016/j.drugpo.2025.104722.
- Hannibal K.E., Bishop M.D. Chronic stress, cortisol dysfunction, and pain: a psychoneuroendocrine rationale for stress management in pain rehabilitation. Phys. Ther. 2014;94(12):1816–1825. doi: 10.2522/ptj.20130597.
- Hassan A.M., Nguyen H.T., Corkum J.P., Liu J., Kapur S.K., Chu C.K., Tamirisa N., Offodile 2nd A.C. Area deprivation index is associated with variation in quality of life and psychosocial well-being following breast cancer surgery. Ann. Surg. Oncol. 2023;30(1):80–87. doi: 10.1245/s10434-022-12506-z.
- Hill T.D., Burdette A.M., Hale L. Neighborhood disorder, sleep quality, and psychological distress: testing a model of structural amplification. Health Place. 2009;15(4):1006–1013. doi: 10.1016/j.healthplace.2009.04.001.
- Jeffers A.M., Glantz S., Byers A., Keyhani S. Sociodemographic characteristics associated with and prevalence and frequency of cannabis use among adults in the US. JAMA Netw. Open. 2021;4(11) doi: 10.1001/jamanetworkopen.2021.36571.
- Jett J., Stone E., Warren G., Cummings K.M. Cannabis use, lung cancer, and related issues. J. Thorac. Oncol. 2018;13(4):480–487. doi: 10.1016/j.jtho.2017.12.013.
- Kind A.J., Buckingham W.R. Making neighborhood-disadvantage metrics accessible—the neighborhood atlas. N. Engl. J. Med. 2018;378(26):2456. doi: 10.1056/NEJMp1802313.
- Kind A.J., Jencks S., Brock J., Yu M., Bartels C., Ehlenbach W., Greenberg C., Smith M. Neighborhood socioeconomic disadvantage and 30-day rehospitalization: a retrospective cohort study. Ann. Intern. Med. 2014;161(11):765–774. doi: 10.7326/M13-2946.
- Ko G.D., Bober S.L., Mindra S., Moreau J.M. Medical cannabis–the Canadian perspective. J. Pain. Res. 2016:735–744. doi: 10.2147/JPR.S98182.
- Kroenke K., Strine T.W., Spitzer R.L., Williams J.B., Berry J.T., Mokdad A.H. The PHQ-8 as a measure of current depression in the general population. J. Affect. Disord. 2009;114(1-3):163–173. doi: 10.1016/j.jad.2008.06.026.
- Kurani S.S., McCoy R.G., Lampman M.A., Doubeni C.A., Finney Rutten L.J., Inselman J.W., Giblon R.E., Bunkers K.S., Stroebel R.J., Rushlow D. Association of neighborhood measures of social determinants of health with breast, cervical, and colorectal cancer screening rates in the US Midwest. JAMA Netw. Open. 2020;3(3) doi: 10.1001/jamanetworkopen.2020.0618.
- Lemyre A., Poliakova N., Bélanger R.E. The relationship between tobacco and cannabis use: a review. Subst. Use & Misuse. 2019;54(1):130–145. doi: 10.1080/10826084.2018.1512623.
- Li C.-C., Matthews A.K., Asthana A., Shah R.C. The impact of neighborhood disadvantage on health-related quality of life among African American and White cancer survivors. Transl. Cancer Res. 2019;8(4):S313. doi: 10.21037/tcr.2019.05.30.
- Mah K., Swami N., Le L.W., Chow R., Hannon B.L., Rodin G., Zimmermann C. Validation of the 7-item Functional Assessment of Cancer Therapy-General (FACT-G7) as a short measure of quality of life in patients with advanced cancer. Cancer. 2020;126(16):3750–3757. doi: 10.1002/cncr.32981.
- Maly A., Vallerand A.H. Neighborhood, socioeconomic, and racial influence on chronic pain. Pain. Manag. Nurs. 2018;19(1):14–22. doi: 10.1016/j.pmn.2017.11.004.
- McClure E.A., Walters K.J., Tomko R.L., Dahne J., Hill E.G., McRae-Clark A.L. Cannabis use prevalence, patterns, and reasons for use among patients with cancer and survivors in a state without legal cannabis access. Support. care Cancer. 2023;31(7):429. doi: 10.1007/s00520-023-07881-6.
- Meghani S.H., Quinn R., Ashare R., Levoy K., Worster B., Naylor M., Chittams J., Cheatle M. Impact of cannabis use on least pain scores among African American and white patients with cancer pain: a moderation analysis. J. Pain. Res. 2021:3493–3502. doi: 10.2147/JPR.S332447.
- Meghani S.H., Rosa W.E., Chittams J., Vallerand A.H., Bao T., Mao J.J. Both race and insurance type independently predict the selection of oral opioids prescribed to cancer outpatients. Pain. Manag. Nurs. 2020;21(1):65–71. doi: 10.1016/j.pmn.2019.07.004.
- Mudumbai S.C., He H., Chen J.-Q., Kapoor A., Regala S., Mariano E.R., Stafford R.S., Abnet C.C., Pfeiffer R.M., Freedman N.D. Opioid use in cancer patients compared with noncancer pain patients in a veteran population. JNCI Cancer Spectr. 2024;8(2):pkae012. doi: 10.1093/jncics/pkae012.
- Myers M.G., Dowling M.A., Bohnert K.M. Where do adults in the United States obtain cannabis? A nationally representative study examining the relationships between sociodemographic factors, cannabis use characteristics and sources of cannabis. J. Stud. Alcohol Drugs. 2024;85(3):296–305. doi: 10.15288/jsad.23-00254.
- Pearman T., Yanez B., Peipert J., Wortman K., Beaumont J., Cella D. Ambulatory cancer and US general population reference values and cutoff scores for the functional assessment of cancer therapy. Cancer. 2014;120(18):2902–2909. doi: 10.1002/cncr.28758.
- Peoples J., Tanner J.J., Bartley E.J., Domenico L.H., Gonzalez C.E., Cardoso J.S., Lopez-Quintero C., Losin E.A.R., Staud R., Goodin B.R. Association of neighborhood-level disadvantage beyond individual sociodemographic factors in patients with or at risk of knee osteoarthritis. BMC Musculoskelet. Disord. 2024;25(1):887. doi: 10.1186/s12891-024-08007-7.
- Rhew I.C., Guttmannova K., Kilmer J.R., Fleming C.B., Hultgren B.A., Hurvitz P.M., Dilley J.A., Larimer M.E. Associations of cannabis retail outlet availability and neighborhood disadvantage with cannabis use and related risk factors among young adults in Washington State. Drug Alcohol Depend. 2022;232 doi: 10.1016/j.drugalcdep.2022.109332.
- Riley A.R. Neighborhood disadvantage, residential segregation, and beyond—lessons for studying structural racism and health. J. Racial Ethn. Health disparities. 2018;5:357–365. doi: 10.1007/s40615-017-0378-5.
- Rosenzweig M.Q., Althouse A.D., Sabik L., Arnold R., Chu E., Smith T.J., Smith K., White D., Schenker Y. The association between area deprivation index and patient-reported outcomes in patients with advanced cancer. Health Equity. 2021;5(1):8–16. doi: 10.1089/heq.2020.0037.
- Ross C.E. Neighborhood disadvantage and adult depression. J. Health Soc. Behav. 2000:177–187.
- Salz T., Meza A.M., Chino F., Mao J.J., Raghunathan N.J., Jinna S., Brens J., Furberg H., Korenstein D. Cannabis use among recently treated cancer patients: perceptions and experiences. Support. care Cancer. 2023;31(9):545. doi: 10.1007/s00520-023-07994-y.
- Santiago C.D., Wadsworth M.E., Stump J. Socioeconomic status, neighborhood disadvantage, and poverty-related stress: prospective effects on psychological syndromes among diverse low-income families. J. Econ. Psychol. 2011;32(2):218–230.
- Segrin C., Badger T.A., Morrill K.E., Sikorskii A. Neighborhood disadvantage and social isolation are associated with psychological and physical symptom burden in cancer survivors. Psychol. Health Med. 2025:1–15. doi: 10.1080/13548506.2025.2564313.
- Simmons V.N., Piñeiro B., Hooper M.W., Gray J.E., Brandon T.H. Tobacco-related health disparities across the cancer care continuum. Cancer Control. 2016;23(4):434–441. doi: 10.1177/107327481602300415.
- Spitzer R.L., Kroenke K., Williams J.B., Löwe B. A brief measure for assessing generalized anxiety disorder: the GAD-7. Arch. Intern. Med. 2006;166(10):1092–1097. doi: 10.1001/archinte.166.10.1092.
- Steigerwald S., Wong P.O., Cohen B.E., Ishida J.H., Vali M., Madden E., Keyhani S. Smoking, vaping, and use of edibles and other forms of marijuana among US adults. Ann. Intern. Med. 2018;169(12):890–892. doi: 10.7326/M18-1681.
- Troxel W.M., DeSantis A., Richardson A.S., Beckman R., Ghosh-Dastidar B., Nugroho A., Hale L., Buysse D.J., Buman M.P., Dubowitz T. Neighborhood disadvantage is associated with actigraphy-assessed sleep continuity and short sleep duration. Sleep. 2018;41(10):zsy140. doi: 10.1093/sleep/zsy140.
- Unger J.M., Moseley A.B., Cheung C.K., Osarogiagbon R.U., Symington B., Ramsey S.D., Hershman D.L. Persistent disparity: socioeconomic deprivation and cancer outcomes in patients treated in clinical trials. J. Clin. Oncol. 2021;39(12):1339–1348. doi: 10.1200/JCO.20.02602.
- Van Ameringen M., Zhang J., Patterson B., Turna J. The role of cannabis in treating anxiety: an update. Curr. Opin. Psychiatry. 2020;33(1):1–7. doi: 10.1097/YCO.0000000000000566.
- Walker Z.J., Xue S., Jones M.P., Ravindran A.V. Depression, anxiety, and other mental disorders in patients with cancer in low-and lower-middle–income countries: a systematic review and meta-analysis. JCO Glob. Oncol. 2021;7:1233–1250. doi: 10.1200/GO.21.00056.
- Wang L., Hong P.J., May C., Rehman Y., Oparin Y., Hong C.J., Hong B.Y., AminiLari M., Gallo L., Kaushal A. Medical cannabis or cannabinoids for chronic non-cancer and cancer related pain: a systematic review and meta-analysis of randomised clinical trials. bmj. 2021;374 doi: 10.1136/bmj.n1034.
- Wen Y.-P., Brauer E.R., Choi K. A systematic review of the influence of social determinants of health on mental health service utilization and outcomes among Asian american cancer survivors. J. Racial Ethn. Health disparities. 2025:1–14. doi: 10.1007/s40615-024-02275-8.
- Worster B., Hajjar E.R., Handley N. Cannabis use in patients with cancer: a clinical review. JCO Oncol. Pract. 2022;18(11):743–749. doi: 10.1200/OP.22.00080.
- Youn N., Sorensen J., Howland C., Gilbertson-White S. Social determinants of health and cancer pain in the US: scoping review. Clin. Nurs. Res. 2024;33(5):416–428. doi: 10.1177/10547738241232018.
- Yurasek A.M., Aston E.R., Metrik J. Co-use of alcohol and cannabis: a review. Curr. Addict. Rep. 2017;4(2):184–193. doi: 10.1007/s40429-017-0149-8.
- Zaki P., Blake A., Wolt A., Chan S., Wan A., Lam H., DeAngelis C., Slaven M., Shaw E., Ganesh V. The use of medical cannabis in cancer patients. J. Pain. Manag. 2017;10(4)