A 2025 national survey of nearly 8,000 cancer survivors found that those carrying the highest burden of childhood adversity were four times as likely to report cannabis use. The association is graded and statistically robust, but the study cannot determine whether cannabis serves as self-directed therapy, a coping mechanism rooted in trauma, or both.

Cancer survivors already navigate a complex landscape of physical symptoms, psychological distress, and treatment side effects that frequently prompt interest in cannabis. If childhood trauma meaningfully shapes who turns to cannabis and why, oncology teams may be missing a critical piece of the clinical puzzle by failing to ask about early-life adversity. This study is the first to directly examine that intersection in a large, population-based sample, raising a question that cancer care has largely ignored: does what happened to patients decades ago influence the health decisions they are making right now?

Cannabis use among cancer survivors has attracted growing clinical and research interest, with prior studies estimating prevalence rates between 8% and 31%. Simultaneously, the adverse childhood experiences (ACE) literature has established strong associations between cumulative childhood trauma and a range of adult health behaviors, including substance use. However, no previous study had examined whether the ACE-substance use relationship extends specifically to cannabis in cancer survivors, a population that faces distinct physical and psychosocial pressures that could motivate both therapeutic and coping-driven use. This study drew on the 2020 Behavioral Risk Factor Surveillance System to investigate that question using data from 28 states that included the optional ACE module.

Among 7,896 cancer survivors, the weighted prevalence of cannabis use was 6.0%, with raw prevalence rising sharply from 2.6% in survivors with zero ACEs to 15.4% in those with four or more. After multivariable adjustment for age, sex, race, marital status, education, income, smoking, general health status, and depression, a clear dose-response gradient persisted: survivors with two to three ACEs had 2.56 times the odds of cannabis use (95% CI: 1.57 to 4.27), while those with four or more ACEs had 4.10 times the odds (95% CI: 2.54 to 6.64), compared to survivors with no ACEs. The authors appropriately acknowledged that the cross-sectional design precludes causal inference, that the binary cannabis measure cannot distinguish frequency, formulation, or intent, and that state-level cannabis legal status was not included as a covariate. They call for prospective studies and recommend that oncology clinicians consider incorporating ACE screening into survivorship care.

Childhood Adversity and Cannabis Use in Cancer Survivors: A Signal Worth Taking Seriously

Of nearly 8,000 cancer survivors surveyed across the United States, those who had experienced four or more adverse childhood events were four times as likely to be using cannabis. That is a finding that raises a question every oncologist should be asking: when a patient reaches for cannabis, what are they really reaching for? This study by Gao and colleagues does not answer that question, but it frames it with a clarity that has been missing from the oncology literature. What the authors actually tested is whether a cumulative ACE score, recalled retrospectively and measured at the same moment as cannabis use status, predicts the odds of reporting any cannabis use. What they found is genuinely valuable: a graded, dose-dependent association that survives reasonable statistical adjustment and that represents the first time this specific intersection has been examined in a cancer survivor population. Asking someone today whether they had a difficult childhood and whether they currently use cannabis is like taking a single photograph to determine whether one event caused another. You can see they appear together in the frame, but you have no way of knowing what happened before the shutter clicked. The study design structurally cannot tell us whether trauma drives cannabis initiation, whether cannabis use reflects a broader pattern of health behaviors that cluster with adversity, or whether both are shaped by a third factor the model does not capture.

The most consequential blind spot, in my reading, is the absence of any data on why survivors are using cannabis. Measuring cannabis use as yes or no is like asking whether a patient “takes medication.” It tells you almost nothing clinically useful without knowing what, how much, how often, and why. A cancer survivor using a low-dose CBD tincture for chemotherapy-induced nausea and a survivor using high-THC flower to numb trauma-related anxiety are occupying profoundly different clinical spaces, yet this study counts them identically. The omission of state-level cannabis legal status as a covariate compounds this problem. Comparing cannabis use rates across states with different legal frameworks without adjusting for those laws is like comparing umbrella sales across cities with different climates and concluding that something about the people, not the weather, explains the difference. These are not minor quibbles. They represent the exact variables that would allow us to move from association to understanding, and their absence keeps the study firmly in hypothesis-generating territory.

What would I tell a patient sitting across from me? If you have experienced significant childhood adversity and you are navigating cancer, I want to understand more about why you are using cannabis and whether it is serving you well, or whether we should be offering other forms of support alongside it or instead of it. To a colleague, I would say the dose-response pattern here is compelling enough to start asking about early-life adversity in survivors who are using cannabis, particularly those presenting with depression, poor self-rated health, or multiple risk factors. To a policymaker, I would say this is early epidemiologic evidence that childhood trauma is clinically relevant to cannabis use in cancer survivors, and it warrants investment in prospective research, but it is not yet a foundation for mandating screening or changing access policies. A statistically compelling cross-sectional association, even a large, dose-graded, well-adjusted one, is an invitation to design the next study, not a conclusion. The clinical questions that matter most, why high-ACE survivors use cannabis and whether it is helping or harming them, require study designs that this paper structurally cannot answer.

This study sits early in the research arc connecting childhood adversity to substance use patterns specifically in oncology populations. The general ACE-substance use relationship is well established in population health literature, with meta-analytic evidence showing graded associations between cumulative ACEs and tobacco, alcohol, and illicit drug use. What this paper adds is specificity: it narrows the lens to cannabis use and cancer survivorship, providing preliminary evidence that the broader population-level finding holds within a clinically distinct group. However, without prospective data, motivational assessment, or intervention testing, this remains a first-stage epidemiologic observation that identifies a pattern rather than explaining or addressing it.

From a pharmacological and safety standpoint, clinicians should note that the co-occurrence of depression, smoking, and poor self-rated health with cannabis use in this sample points to a multi-morbidity context in which cannabis interactions with psychiatric medications, chemotherapeutic agents, and other substances may be particularly consequential. High-ACE patients may also be less likely to disclose cannabis use proactively, especially in states with restrictive legal environments. The single most actionable takeaway for practicing clinicians is to normalize open conversations about cannabis use with cancer survivors, and to consider asking about childhood adversity as a way of understanding the context behind that use, rather than treating cannabis disclosure as an isolated behavioral data point.

This is a cross-sectional analysis of publicly available 2020 BRFSS telephone survey data, employing weighted multivariable logistic regression. In the evidence hierarchy, cross-sectional studies occupy a lower tier than cohort studies or randomized trials because exposure and outcome are measured simultaneously. The single most important inference constraint is that no temporal sequence can be established: we cannot know whether ACE exposure preceded cannabis initiation, whether cannabis use affects ACE recall, or whether both are driven by unmeasured factors.

The finding of a graded ACE-cannabis association is consistent with the broader ACE-substance use literature, including meta-analytic work showing that individuals with four or more ACEs have significantly elevated risk for multiple health-risk behaviors. Prior studies estimating cannabis use prevalence among cancer survivors at 8% to 31% are higher than the 6% found here, a discrepancy likely explained by geographic restriction to 28 states and the older, predominantly White composition of this sample. ASCO’s existing guidelines already call for evidence-based consideration of cannabis in oncology, though they do not address the role of childhood adversity in shaping use patterns.

This study extends the existing literature rather than challenging it. It takes a well-established general-population finding and demonstrates that it applies, potentially with even larger effect sizes, in a clinically vulnerable subgroup. What remains absent from the literature is any prospective or interventional evidence testing whether addressing ACE-related trauma changes cannabis use patterns or survivorship outcomes.

The most consequential analytic choice was the omission of state-level cannabis legal status as a covariate or effect modifier. In 2020, the 28 participating states spanned a wide range of legal frameworks, from full prohibition to recreational legalization. If high-ACE survivors were disproportionately concentrated in states with more permissive cannabis laws, the observed association could be partly or substantially explained by legal access rather than trauma-driven behavior. Including state cannabis policy as a fixed effect or stratifying analyses by legal status would have substantially strengthened the study’s ability to isolate the ACE-specific signal.

Additionally, the use of listwise deletion for missing data, particularly income data (missing in approximately 18.7% of respondents), without sensitivity analysis or multiple imputation raises concern that the analytic sample may systematically differ from the full sample. If missingness is related to both ACE burden and cannabis use, the resulting estimates could be biased in either direction.

The most likely overinterpretation is reading this study as evidence that childhood trauma causes cannabis use in cancer survivors. The cross-sectional design establishes co-occurrence, not causation; ACEs and cannabis use were measured at the same moment, and no temporal sequence can be determined. The dose-response gradient is compelling and consistent with a causal relationship, but it is equally consistent with unmeasured confounders that themselves increase in parallel with ACE burden, such as cumulative psychiatric illness or poverty severity.

A second common misreading would be to conclude that these cancer survivors are using cannabis problematically or for non-medical reasons. The study collected no data on motivation, frequency, dosage, or formulation. Some proportion of this cannabis use may represent well-considered symptom management, and conflating all cannabis use among high-ACE survivors with harmful coping is both empirically unsupported and clinically unhelpful.

This study establishes a graded, statistically robust association between adverse childhood experiences and cannabis use among cancer survivors, with those reporting four or more ACEs showing roughly four times the odds of use. It does not establish causation, determine cannabis use motivation, or test any clinical intervention. For practice today, it supports asking cancer survivors about childhood adversity as a way of understanding the context behind cannabis use, while we await the prospective and interventional evidence needed to guide formal protocols.

Does this study prove that childhood trauma causes cancer survivors to use cannabis?

No. The study measured both childhood adversity and cannabis use at the same point in time, which means it can show that the two tend to occur together but cannot determine which came first or whether one caused the other. Many other factors that were not measured in this study could explain the connection.

Should I be concerned if I am a cancer survivor who uses cannabis and had a difficult childhood?

This study does not suggest that cannabis use is necessarily harmful for people with childhood trauma. It simply shows the two are linked statistically. What matters clinically is whether your cannabis use is serving you well, whether your healthcare team knows about it, and whether you have access to other forms of support for any trauma-related distress you may be experiencing.

Why couldn’t the researchers tell whether survivors were using cannabis for medical or recreational reasons?

The national survey used in this study asked only whether respondents had used cannabis, with a simple yes or no answer. It did not collect information on the reason for use, how often it was used, what form it took, or what dose was consumed. This is a significant limitation because the clinical implications differ greatly depending on motivation and pattern of use.

Should oncologists start screening for childhood trauma based on this study?

The study’s authors recommend considering ACE screening in oncology, and the idea has clinical logic behind it. However, this study alone does not prove that screening would improve outcomes. It identifies a statistical pattern that justifies further research and that may encourage clinicians to ask about early-life experiences as part of a broader conversation about a patient’s health and coping strategies.

Have thoughts on this? Share it: