Medical Cannabis Eased Pain, Not Cravings, in OUD on Buprenorphine
Study Facts
| Authors | Lent MR, Keen R, Ruiz M, Callahan HR, Galluzzi KE, Dugosh KL |
|---|---|
| Journal | Journal of Cannabis Research |
| Year | 2025 |
| Design | Prospective, observational intervention study, no control group, no randomization |
| Sample size | 58 approached, 47 enrolled, 40 completed 3-month assessments |
| Intervention | 1:1 THC:CBD 5 mg:5 mg daily oral capsule, discounted, alongside buprenorphine/naloxone for OUD |
| Comparator | None; single-arm pre-post design |
| Primary endpoint | Pain severity, interference, self-efficacy (BPI, PSEQ-2), and UDS-confirmed opioid use |
| Key results | Pain severity 5.18 to 4.39 (d=0.54, P<0.01); pain interference 5.21 to 4.32 (d=0.47, P<0.01); pain self-efficacy 6.55 to 8.05 (d=0.44, P<0.01); opioid-positive UDS 16% to 5% (P=0.13, not significant); sleep quality (PSQI) 12.38 to 10.95 (d=0.33, P<0.05); cravings unchanged (P=0.49) |
| Adverse events | 2 serious adverse events (inpatient alcohol detox, psychiatric hospitalization); 7 discontinuations |
| Funding | Organic Remedies, Inc. funded the study and manufactured the discounted study medication but had no role in design, analysis, or interpretation |
| Conflicts of interest | Authors declare no competing interests |
Study Snapshot
| Study type | Prospective, single-arm observational intervention study |
|---|---|
| Population | Adults on buprenorphine/naloxone for OUD with chronic pain (minimum 5/10 pain score) |
| Intervention | Daily 1:1 THC:CBD 5 mg:5 mg oral capsule, plus optional additional MC purchases |
| Comparator | None (no control group) |
| Primary outcome | Pain severity, interference, self-efficacy, and UDS-confirmed opioid use at 3 months |
| Sample size | 47 enrolled, 40 completed follow-up assessments (38 completed UDS) |
| Headline result | Significant improvements in pain and quality of life; no significant change in opioid use or cravings |
Medical Cannabis With Buprenorphine: Pain Improved, Cravings Did Not
CED Clinical Relevance
CED Clinic sees a substantial subset of patients managing chronic pain alongside opioid use disorder, often stabilized on buprenorphine/naloxone but still reporting unresolved pain. This study population, adults on buprenorphine with pain scores of 5/10 or higher, maps directly onto that clinical reality. The intervention, a low-dose 1:1 THC:CBD oral capsule, is a formulation Dr. Caplan can discuss concretely with patients already holding an MC certification for OUD or chronic pain in Pennsylvania.
The findings offer a data point for a conversation that patients raise often: whether cannabis might help them tolerate buprenorphine treatment more comfortably without undermining their recovery. The results support cautious optimism about pain and quality of life benefits while tempering expectations about cravings or illicit opioid use reduction.
Clinical Insight
When adding low-dose 1:1 THC:CBD capsules to buprenorphine/naloxone in patients with comorbid chronic pain, clinicians can reasonably expect measurable improvement in pain severity, pain interference, and self-efficacy within three months, but should not counsel patients that this intervention will reliably reduce cravings or illicit opioid use, since those endpoints did not reach statistical significance in this sample.
What Medical Cannabis With Buprenorphine Changed / How This Fits the Broader Conversation
This study adds prospective clinical data to a literature that has largely relied on retrospective surveys or theoretical models linking pain, cannabis, and opioid use disorder recovery. It confirms a pattern seen in cannabis-for-pain research generally: modest but statistically significant reductions in pain severity and interference, alongside sleep and quality-of-life gains, using a low-THC formulation.
What it does not resolve is the central controversy in the OUD field, whether cannabis meaningfully reduces cravings or illicit opioid use in this population. The non-significant findings on those outcomes, combined with an already-low baseline rate of opioid-positive UDS in a buprenorphine-stabilized sample, leave open whether a more treatment-naive or higher-craving population would show different effects. The authors explicitly call for randomized, controlled trials, a call that remains structurally difficult given cannabis’s Schedule I status.
What This Teaches Us: For Patients and Families
This study followed adults who were already taking buprenorphine, a medication that helps manage opioid use disorder, and who also lived with ongoing pain. Researchers wanted to see what happened when these patients added a low-dose cannabis capsule containing equal parts THC and CBD to their treatment.
After three months, participants reported feeling less pain, sleeping somewhat better, and feeling more confident in their ability to manage pain day to day. Their overall quality of life improved in most areas measured. However, the cannabis did not clearly reduce their cravings for opioids or their rates of using illicit opioids, though those rates were already fairly low to begin with. Two people in the study had serious health events unrelated in a straightforward way to the cannabis itself. The takeaway for families: cannabis alongside buprenorphine may ease pain and improve daily functioning for some people, but it is not a proven tool for reducing opioid cravings or use, and more rigorous research is still needed.
How Strong Is the Evidence
Low. This is a single-arm, uncontrolled, prospective observational study with 47 enrolled and 40 completers at Month 3. There is no comparator group, no randomization, and no blinding, so the pre-post changes in pain, self-efficacy, sleep, and quality of life cannot be separated from regression to the mean, natural fluctuation in chronic pain, concurrent recreational cannabis use (85% of the sample at baseline), or nonspecific effects of enrolling in a monitored study with $30 per-visit remuneration. The primary substance-use outcome, UDS-confirmed opioid use, did not reach significance (P = 0.13) despite a numerically large drop from 16% to 5%, which in a sample this size reflects a change in only four participants. The consistent, significant improvements in pain severity, interference, and self-efficacy (all P < 0.01) are the most credible signal in the dataset, but they still rest on subjective self-report instruments without a control arm. GRADE-style, this evidence sits at Low certainty: real and biologically plausible, but not causally established.
Dr. Caplan’s Take
What I find clinically useful here is not the headline finding, since opioid use did not significantly change, but the pain data underneath it. Pain severity, pain interference, and pain self-efficacy all moved in the expected direction with P values under 0.01 and moderate effect sizes (d = 0.44 to 0.54). In a population maintained on buprenorphine/naloxone with a documented pain condition, that is a meaningful signal. Undertreated pain is one of the more common reasons patients on MOUD destabilize, and a low-dose 1:1 THC:CBD capsule (5 mg:5 mg) moving pain self-efficacy from 6.55 to 8.05 over three months is the kind of change that shows up in a patient’s daily functioning, not just their questionnaire.
I want to be honest about what this study does not show. Illicit opioid use dropped from 16% to 5% positive UDS, which sounds clinically important, but it was not statistically significant and the underlying numbers are small: six positive tests at baseline versus two at Month 3, in a sample of 38 with UDS data. Cravings barely moved and the difference was not significant either. So the addiction-specific outcomes here are inconclusive, not negative and not positive. Anyone using this paper to argue that cannabis directly suppresses opioid use in this population is overstating what the data support.
The adjunctive framing in the discussion is the right one. This is not a study of cannabis replacing buprenorphine or treating OUD as a standalone intervention. It is a study of low-dose oral THC:CBD added on top of established MOUD in patients whose chronic pain was not fully controlled. That is a narrower and more defensible clinical question, and it is the one I actually get asked in practice: can we use cannabis to help a stable buprenorphine patient whose pain is still limiting their function, without destabilizing their recovery. On the safety side, two SAEs occurred (an alcohol detox admission and a psychiatric hospitalization), and neither reads to me as cannabis-attributable based on the description given, but they are a reminder that this population carries polysubstance and psychiatric risk that needs monitoring regardless of the cannabis question.
The sleep and quality-of-life findings round out a coherent picture: sleep quality improved significantly, seven of eight SF-36 domains improved, and none of this happened at the cost of increased opioid use or cravings. If I am counseling a patient in this exact situation, chronic pain, stable on buprenorphine/naloxone, pain not fully controlled, I read this paper as modest reassurance that a low-THC adjunct did not appear to destabilize their substance use status over three months while their pain scores improved. That is a reasonable, bounded claim. It is not evidence that cannabis treats opioid use disorder, and the authors do not claim that either.
What would change my mind toward more confidence is a controlled trial, even a small randomized one, with a longer follow-up than three months and enough power to look at who benefits and who does not. The 47-patient sample here cannot support subgroup analysis, and the authors say as much. Until that trial exists, I treat this as hypothesis-generating: a signal on pain, a null on substance use, and an open question on durability past the three-month mark.
What Medical Cannabis With Buprenorphine Did Not Change / What a Careful Reader Should Take Away
The pain and quality-of-life improvements are statistically significant with moderate effect sizes and are the strongest part of this dataset, but the study’s uncontrolled, single-arm design, high baseline recreational cannabis use, and small completer sample (40 of 47) mean these gains cannot be attributed to the study capsule alone; meanwhile the substance-use outcomes that motivated the study, opioid-positive UDS and craving scores, did not reach statistical significance, so this paper supports a plausible adjunctive role for low-dose oral THC:CBD in pain management for patients stable on buprenorphine/naloxone, but it does not demonstrate that medical cannabis reduces illicit opioid use or cravings in this population.
Interpreting This Evidence Through Multiple Clinical Perspectives
Overview
This study sits at the junction of two policy experiments running in parallel: expanded buprenorphine access under the MATE Act and state-level legalization of medical cannabis for OUD in Pennsylvania. The finding that matters structurally is not the pain reduction itself but the fact that adding MC to an established buprenorphine regimen did not increase illicit opioid use, cannabis use aside. That absence of harm, more than the presence of benefit, is what a regulator or health system will extract from this paper.
The deeper pattern is institutional: a cannabis manufacturer funded the study, supplied the product at a nominal price, and the resulting positive framing arrives inside a literature still starved for controlled data because of Schedule I restrictions. The paper is less a definitive clinical signal than a marker of how early-stage evidence gets built when the gold-standard tools, randomization and blinding, are legally unavailable.
- No significant increase in illicit opioid use following MC initiation, in a population already stabilized on buprenorphine
- Improvements clustered in pain and quality of life domains, not in the primary substance-use outcome
- Industry funding and discounted product access shaped both recruitment and retention
- The study exists because Schedule I status blocks the randomized trials this question actually requires
Patient Takeaway
If you are managing chronic pain on buprenorphine, this study offers a modest, honest picture: pain scores dropped by about one point on a ten-point scale, and confidence in managing pain rose noticeably. Sleep improved slightly and most quality of life domains got better. These are real but incremental changes, not a transformation.
What this study does not show is that cannabis reduced cravings for opioids or further lowered illicit opioid use beyond what buprenorphine was already achieving. Expect better function and slightly less pain interference in daily life, but do not expect cannabis to resolve craving or substitute for continued addiction treatment. The people who benefited most stayed engaged with both their MC and their buprenorphine care.
- Pain severity and interference improved modestly, roughly a one-point drop on a ten-point scale
- Confidence in managing pain increased more than pain itself decreased
- Cravings for opioids did not change in a statistically meaningful way
- Sleep and most quality of life domains improved, but general health ratings did not
Clinician’s POV
Operationally, this model requires coordination between an OUD prescriber and a separately certified MC recommender, a division of labor that most practices are not built for. Urine drug screening remained the backbone of monitoring, run at baseline and Month 3, and buprenorphine-positive UDS rates stayed essentially flat, 84% to 82%, confirming that the cohort remained adherent to MOUD throughout.
Documentation burden is nontrivial: PDMP verification of buprenorphine history, dispensary purchase tracking through the state’s MJ Freeway system, and structured interview batteries covering pain, sleep, cravings, and eight SF-36 domains. None of this is reimbursable in a typical outpatient addiction medicine workflow, and the discounted product price used here, one dollar for a 30-day capsule supply, will not reflect real-world costs once a program scales past a manufacturer-subsidized pilot.
- Requires coordinated care between MOUD prescriber and a certified MC recommender, two separate credentialing pathways
- UDS monitoring confirmed buprenorphine adherence remained stable across the study period
- Purchase data via state tracking software revealed most patients bought flower, not the study capsule
- Discounted product pricing in this study does not reflect sustainable reimbursement conditions
A Skeptical Read
Eighty-five percent of this sample was already using cannabis recreationally at baseline, for an average of 49 of the prior 90 days. Calling this an MC ‘initiation’ study is a stretch: most participants were regular cannabis users before enrollment, and the intervention largely formalized an existing behavior rather than introducing a new one. That undercuts any claim that the observed changes reflect a novel pharmacologic effect rather than continuation of prior use patterns, possibly combined with increased attention, structured monitoring, and a thirty-dollar per-visit payment.
The study was funded by a cannabis manufacturer that also supplied the discounted product, and most participants ended up purchasing flower products with a mean THC content of 23%, far from the 1:1 THC:CBD capsule the study was designed around. The primary preregistered hypothesis, reduced illicit opioid use, did not reach significance, yet the discussion section frames the paper’s message around the secondary pain and quality of life findings that did. That is a permissible move but one worth naming plainly.
- 85% of participants already used cannabis recreationally at baseline, complicating any ‘initiation’ framing
- Study funded and product-supplied by a cannabis manufacturer with no independent control condition
- Most purchases diverged from the study capsule toward higher-THC flower products
- The primary outcome, illicit opioid use, was not statistically significant despite the paper’s positive framing
Study Critic
This is a single-arm, pre-post observational design with no control group, which the authors themselves acknowledge precludes causal inference. Regression to the mean is a live concern given that participants were recruited immediately after MC consultation, likely at a point of heightened symptom salience, and any subsequent improvement could reflect natural fluctuation rather than treatment effect.
The analyzed sample of 40 completers out of 47 enrolled, drawn from 58 originally approached, introduces attrition risk, and the paper notes a significant gender difference in follow-up completion without adjusting outcome analyses for it. Multiple outcomes were tested (pain severity, interference, self-efficacy, eight SF-36 domains, sleep, cravings, UDS categories) without correction for multiple comparisons, and effect sizes for the significant findings were small to moderate, d = 0.33 to 0.54. The three-month window captured is an interim analysis from a planned one-year follow-up, so durability of any effect is unknown.
- No control group; causal attribution is explicitly precluded by design, as the authors state
- Differential attrition by gender was detected but not adjusted for in outcome analyses
- Multiple outcome comparisons run without correction, raising the risk of false positive findings
- Effect sizes for significant findings are modest, d = 0.33 to 0.54, and sample is underpowered for subgroup analysis
Compared to Past Research
The paper situates itself against a literature of cannabis-for-pain randomized trials it cites, including systematic reviews and RCTs showing analgesic effects across pain conditions, and meta-analyses suggesting an opioid-sparing effect when cannabinoids are co-administered with opioids. It also draws on a double-blind placebo-controlled trial of CBD reducing cue-induced craving in heroin use disorder, cited as rationale for testing whether MC might attenuate opioid craving in this population.
What distinguishes this study from that cited RCT evidence is design: the trials referenced were randomized and controlled, while this is an uncontrolled naturalistic cohort. The paper does not claim to replicate those controlled findings, and in fact its craving results diverge from the cited CBD-craving trial, since no significant craving reduction was observed here. This gap between RCT-based prior evidence and this study’s real-world, uncontrolled results is the paper’s own framing, not an external comparison this content invents.
- Cited RCTs and meta-analyses support cannabis’s analgesic and opioid-sparing potential in other populations
- A cited placebo-controlled trial found CBD reduced craving in heroin use disorder, a result this study did not replicate
- This study is observational and uncontrolled, unlike the RCT evidence it cites as rationale
- The paper positions itself as an early real-world test of ideas established in more rigorous prior trials, not a replication of them
Practical Considerations
Scaling this model requires three things that were present here only through study infrastructure: a certified MC recommender working near an OUD treatment site, a discounted product arrangement with a dispensary, and state-level purchase tracking software to verify what patients actually bought. Removing any one of these, particularly the one-dollar discounted capsule, changes the cost-benefit picture substantially, since the authors themselves flag MC evaluation and product costs as prohibitive for many patients with OUD.
Retention also depended on incentives, thirty dollars per visit, and even with those in place, seven of forty-seven enrolled participants did not complete follow-up. Any program considering this approach needs to budget for physician certification, dispensary coordination, urine toxicology monitoring, and structured psychosocial assessment battery administration, none of which map cleanly onto existing addiction treatment reimbursement codes.
- Requires a certified MC recommender integrated with OUD treatment access, an uncommon clinical arrangement
- Discounted pricing used in the study is not representative of real dispensary costs
- Retention relied on per-visit payment; unincentivized real-world adherence may differ
- No existing reimbursement pathway supports the assessment battery or monitoring used here
Future Directions
The authors explicitly call for randomized controlled trials with larger samples and longer duration, while acknowledging that Schedule I status is the structural barrier preventing exactly that kind of study. The one-year follow-up from which this is an interim report may eventually clarify whether the pain and quality of life gains persist or attenuate, and whether opioid use trends toward greater separation from baseline over a longer window.
An unresolved question the paper raises but cannot answer is whether outcomes differ for cannabis-naive patients versus the 85% who were already recreational users at enrollment, since the sample was underpowered for that subgroup analysis. Future work will also need to reconcile the study’s intended 1:1 THC:CBD capsule protocol with the reality that most participants purchased higher-THC flower products instead, since the formulation actually used may matter more than the one recommended.
- One-year outcome data, of which this is an interim report, may show whether gains persist
- Subgroup analysis of cannabis-naive versus prior recreational users is needed but was underpowered here
- Product formulation drift toward higher-THC flower needs to be tracked against outcomes in future analyses
- Randomized designs remain legally constrained by Schedule I status, a barrier the authors name directly
Misreadings and Bad-Faith Takes
The most likely distortion of this paper is a headline reading ‘medical cannabis reduces opioid use in addiction patients.’ The reported change in illicit opioid use, 16% to 5%, did not reach statistical significance (P = 0.13), and the authors are explicit that rates of opioid use did not differ significantly. The significant findings were in pain severity, pain interference, pain self-efficacy, sleep, and most quality of life domains, not in substance use itself.
A second distortion would be treating this as evidence that cannabis is a safe universal adjunct for OUD treatment. Two serious adverse events occurred, an inpatient alcohol detox and a psychiatric hospitalization, and the study explicitly could not attribute causality to MC given its uncontrolled design. It also cannot be generalized beyond this single-state, referred, industry-supported sample of patients already stabilized on buprenorphine.
- The reduction in illicit opioid use did not reach statistical significance; this was not a confirmed finding
- Significant results were limited to pain and quality of life measures, not substance use outcomes
- Two SAEs occurred during the study period; causality was not established either way
- Findings come from one state, one referral source, and industry-supported product access, limiting generalizability
Join the Conversation
If you manage patients on buprenorphine/naloxone with persistent chronic pain, has low-dose oral THC:CBD come up as an option in your practice, and what has shaped your comfort level with recommending it alongside MOUD? Share your experience in the comments.
Frequently Asked Questions
Did the cannabis capsule actually stop people from using illicit opioids?
Not in a way the study could confirm statistically. Positive opioid urine tests dropped from 16% at the start to 5% at three months, which looks encouraging, but with only 38 people tested this change was not statistically significant, so the researchers could not conclude that the cannabis caused it.
What kind of cannabis product did participants take?
Everyone was offered a discounted 1:1 THC to CBD oral capsule containing 5 mg of THC and 5 mg of CBD daily, though participants were also free to buy other medical cannabis products, and most did purchase additional items like flower.
Did people’s pain actually get better?
Yes. Average pain severity dropped from about 5.2 to 4.4 out of 10, and pain-related interference with daily life also dropped significantly, with patients reporting more confidence in managing their pain by Month 3.
Did cravings for opioids go down?
Cravings were mildly lower at three months (1.78 versus 2.15 at baseline) but this difference was not statistically significant. Craving levels were already fairly low at the start, likely because most participants had been on buprenorphine for some time already.
Were there any safety concerns?
Two serious adverse events occurred during the study period, an inpatient alcohol detoxification and a psychiatric hospitalization. The study does not report these as caused by the cannabis product.
Can these results apply to someone just starting buprenorphine treatment?
Not directly. Most participants in this study had already been on buprenorphine for months before enrolling, so the findings speak to adding cannabis for pain in patients already stable on MOUD, not to those newly starting opioid use disorder treatment.
Did medical cannabis actually reduce illicit opioid use in this study?
Opioid positive urine screens fell from 16% at baseline to 5% at Month 3, but this change did not reach statistical significance (P = 0.13). All participants who tested opioid positive at follow-up had also tested positive at baseline, and no one newly initiated opioid use during the study. The authors interpret this as no significant change in illicit opioid use, not a confirmed reduction.
What dose of THC and CBD did participants actually consume?
Participants were offered a discounted 1:1 THC:CBD oral capsule dosed at 5 mg THC and 5 mg CBD daily. However, tracking data showed most participants also purchased other products, including flower, and the mean THC content across all purchased products was 23.05%, with mean CBD content of only 0.64%. Actual cannabinoid exposure was therefore likely higher in THC and lower in CBD than the study capsule alone.
Were participants using cannabis recreationally before starting medical cannabis, and does that complicate the results?
Yes. At baseline, 85% of the sample reported recreational cannabis use in the prior 90 days, averaging 49 of 90 days. The authors note this as a limitation: for these participants, changes in outcomes could reflect shifts in cannabis source, strain, route of administration, or clinical guidance rather than the study medication itself.
Did everyone in the study stay on buprenorphine, and could that explain the pain and craving results?
Yes, all participants were maintained on buprenorphine/naloxone throughout the study, and buprenorphine positive urine screens were similarly high at baseline (84%) and Month 3 (82%). The authors suggest that established buprenorphine use likely explains why baseline craving scores were already low, potentially limiting the ability to detect further craving reduction from cannabis.
How many participants dropped out, and were dropouts different from completers?
Of 47 enrolled, 40 (85%) completed the Month 3 assessment battery and 38 (81%) completed urine drug screening; 7 participants did not complete follow-up. Females were significantly more likely to complete the Month 3 assessment than males. No other baseline differences were found between completers and non-completers.
What were the serious adverse events, and were they attributed to the cannabis product?
Two serious adverse events occurred: one inpatient alcohol detox admission and one psychiatric hospitalization. The paper reports these events but does not attribute them causally to the medical cannabis product, consistent with the study’s observational, non-controlled design.
