A 2026 commentary in Clinical Therapeutics proposes “Smart Cannabis,” a conceptual model pairing medical cannabis with FDA-regulated smartphone-based digital therapeutics for real-time monitoring and personalized dosing. The idea is intellectually compelling, but no clinical trial, pilot study, or feasibility data exist to support it. The proposal’s value lies in its diagnosis of the problem, not in a proven solution.

Millions of patients in the United States use medical cannabis with almost no clinical infrastructure guiding their care. There is no standardized dosing, no integrated electronic health record documentation, and virtually no closed-loop feedback between patient response and treatment adjustment. At a time when cannabis legalization is accelerating and digital health tools are gaining FDA clearance for a growing list of conditions, the question of whether these two trajectories can be linked is clinically urgent. This commentary forces that question into focus, even if it cannot yet provide the answer.

Medical cannabis care in the United States currently operates without the clinical guardrails that define standard pharmacotherapy. Patients receive product recommendations from dispensary staff rather than clinicians, dosing is largely self-directed, and cannabis use is rarely documented in electronic health records due to scheduling barriers and stigma. Against this backdrop, Lakhan and Driscoll propose “Smart Cannabis,” a conceptual framework in which FDA-regulated prescription digital therapeutics, delivered via smartphone, would be paired with medical cannabis to create a closed-loop system of real-time monitoring, personalized dosing guidance, and clinical data capture. The authors draw on the growing evidence base for standalone PDTs in conditions such as chronic insomnia, PTSD, depression, and migraine, and on separately documented cannabis use in oncology and pediatric epilepsy, to argue that combining these modalities could address the structural failures of current cannabis care.

No trial data, pilot results, or feasibility outcomes are presented. The commentary proposes Massachusetts as a near-term pilot site, citing its existing Cannabis Control Commission infrastructure and academic medical ecosystem. Policy recommendations include Smart Cannabis pilot programs, improved product labeling, bundled reimbursement models, and clinician education. The authors briefly acknowledge that smartphone-only tools may lack the physiological precision necessary for true feedback-driven dosing and that wearable technology requirements could introduce equity barriers. They are transparent that the model is aspirational and requires empirical validation, though the extent of that gap is arguably understated given the confident framing of their proposal.

Could a Smartphone Fix Medical Cannabis? What the “Smart Cannabis” Commentary Actually Claims

Most medical treatments come with a dose, a diagnosis code, and a follow-up appointment. Medical cannabis comes with a budtender and a best guess. A new commentary in Clinical Therapeutics asks whether smartphone-based digital tools could finally give cannabis the clinical scaffolding it has always lacked, and the honest answer is: we do not know yet. What Lakhan and Driscoll have done well is articulate a problem that those of us who work in cannabis medicine recognize daily. The current system is, as they describe it, an open-loop arrangement in which patients self-titrate with minimal clinical feedback, products are labeled inconsistently, and physicians are largely cut out of the therapeutic relationship after the initial recommendation. That problem statement is not speculative. It is real, it is documented, and it harms patients. Where the commentary genuinely contributes is in naming this structural failure clearly and mapping it against a precision-medicine standard that virtually every other therapeutic domain has at least approached. I give the authors credit for that clarity. It is a diagnosis worth taking seriously.

The proposed solution, however, requires several leaps of inference that the commentary cannot support. The authors draw an analogy to continuous glucose monitoring and insulin pump systems, where a sensor reads a validated biomarker in real time and an algorithm adjusts drug delivery accordingly. It is an elegant analogy and an aspirational one. But imagine a “smart thermostat” for your health that adjusts your treatment in real time, except the thermostat has no temperature sensor, only a survey asking how warm you feel. That is the current state of cannabinoid feedback measurement. No validated real-time biomarker for cannabinoid effect exists, and subjective symptom scores used as proxies are vulnerable to expectancy bias, reporting inconsistency, and placebo effects. Without a reliable feedback signal, the closed-loop model cannot function as described, regardless of how sophisticated the app may be. Similarly, knowing that both PDTs and cannabis show promise for certain conditions independently does not tell you what happens when you package them together. Running shoes and protein shakes may each improve athletic performance, but selling them as a bundled program requires its own trial. The commentary skips that essential step entirely.

There are also blind spots that trouble me as a clinician who works with cannabis patients. A system that continuously monitors federally illegal substance use behavior raises profound privacy and legal risks that are never meaningfully addressed. The equity question is acknowledged but not resolved: the patients most failed by current cannabis care are often those least likely to have reliable smartphone access, digital health literacy, or the financial resources for wearable devices. And the commentary does not engage with existing structured cannabis care models that already provide some of the proposed clinical oversight without requiring novel digital infrastructure. The “Smart Cannabis” commentary is best read as a well-reasoned call to action, not a clinical recommendation. It correctly identifies that the current open-loop model for medical cannabis care fails patients, particularly the most vulnerable ones, and it correctly points toward digital therapeutics as a technology class with genuine, if separately demonstrated, potential. What it does not yet provide, and cannot provide as a commentary, is evidence that the combined system works, that it is equitable, that it is safe for patients whose cannabis use could have legal implications, or that it is feasible to implement within existing regulatory frameworks. The structural problems with medical cannabis care are real and well-characterized; the solutions remain speculative. Any proposed technology fix for a governance problem must first prove it does not introduce new harms, including surveillance risk, equity gaps, and expectancy inflation, faster than it solves the old ones.

For clinicians who already work in cannabis medicine, the problems this commentary identifies will feel familiar. The absence of standardized dosing protocols, the disconnection between dispensary recommendations and clinical records, and the lack of any structured follow-up mechanism represent the daily reality of cannabis-based care. This paper sits at the very beginning of a research arc: it is a concept paper that formulates a testable hypothesis. It does not provide any of the pilot data, feasibility results, or controlled outcomes that would be necessary before any practice change could be considered.

From a pharmacological and safety standpoint, the proposal raises questions it does not answer. Cannabis interacts with numerous medications through cytochrome P450 pathways, and any real-time dosing guidance system would need to account for these interactions, a technical challenge the commentary does not address. The privacy implications of digitally tracking a federally scheduled substance are substantial and remain unexplored. Clinicians should continue to use existing structured approaches to cannabis care, including careful documentation, patient-reported outcome tracking, and regular follow-up visits. The most actionable takeaway from this commentary is not to adopt a new tool, but to recognize that the clinical infrastructure gap it describes is real and worth addressing through whatever means are currently available.

This is a commentary published in a peer-reviewed clinical journal, representing expert opinion at the lowest tier of the evidence hierarchy. It presents no original data, no systematic literature search, and no empirical analysis of any kind. Its authority rests entirely on the quality of its reasoning and the representativeness of the literature it selectively cites. The single most important inference constraint is that no claim about the efficacy, safety, or feasibility of the proposed Smart Cannabis model can be evaluated beyond face plausibility.

The problem characterization in this commentary aligns with a substantial body of literature documenting the structural deficiencies of medical cannabis care, including poor EHR integration, inconsistent product labeling, and minimal clinician involvement after initial certification. The PDT evidence the authors cite, drawn from trials in chronic insomnia and PTSD, is real but has been demonstrated only for those standalone digital interventions, not for a combined cannabis-PDT system. No published study has tested the integration of digital therapeutics with cannabis therapy for any indication. The commentary thus extends the conversation by connecting two separately developing literatures, but the bridge between them remains entirely hypothetical. This positions the paper as a plausible hypothesis-generating contribution, similar in function to earlier conceptual proposals for applying telehealth models to cannabis care, none of which have yet produced definitive trial evidence.

Because this is a commentary rather than an empirical study, the relevant question is not about statistical analysis but about argumentative choices. The most consequential choice was the authors’ reliance on the continuous glucose monitoring analogy to justify the closed-loop model. Had they instead grounded their proposal in a more analogous clinical scenario, such as digital symptom tracking in mental health care where validated biomarkers are similarly absent, the argument would have been more technically honest but less compelling. A systematic rather than selective review of PDT evidence would also have produced a more nuanced picture, likely revealing both the genuine promise and the significant failure rates and engagement drop-offs that characterize real-world PDT implementation. These choices do not invalidate the commentary’s core insight, but they do inflate the perceived readiness of the proposed solution.

The most likely overinterpretation is reading this commentary as evidence that the Smart Cannabis model improves clinical outcomes. It presents no outcome data of any kind. The confident clinical language, the specific disease-area examples, and the named pilot site (Massachusetts) create an impression of a program that is further along than it actually is. In reality, no prototype exists, no pilot has launched, and no patient has been enrolled in a Smart Cannabis study. Readers should also avoid inferring that separately demonstrated efficacy of PDTs for insomnia or PTSD means a combined cannabis-PDT system will be effective. Each component may work in isolation; the combination is an untested hypothesis. Treating Massachusetts as an active implementation site rather than a hypothetical proposal is another common misreading to guard against.

This commentary contributes a clearly articulated conceptual framework for integrating medical cannabis with prescription digital therapeutics and accurately identifies the structural failures of current cannabis care. It does not establish that the proposed Smart Cannabis model is effective, safe, feasible, equitable, or implementable within current regulatory structures. No practice change is warranted at this time. The research agenda the paper implies, particularly feasibility pilots and controlled comparisons, is genuinely worth pursuing.

Is there a “Smart Cannabis” app I can use right now?

No. Smart Cannabis is a conceptual proposal described in a published commentary. No app, product, or clinical program based on this framework currently exists. The authors suggest it should be built and tested, but that work has not yet begun.

Does this paper prove that combining digital apps with cannabis is safe or effective?

No. This commentary presents no clinical trial data, pilot results, or patient outcomes. It argues that the combination could theoretically work based on separately demonstrated evidence for digital therapeutics and cannabis, but the combined approach has never been tested in any patient population.

What should I do differently with my medical cannabis based on this paper?

Nothing needs to change based on this commentary alone. However, the problems it describes are real. You can take practical steps now by keeping a symptom journal, sharing your cannabis use openly with your physician, and scheduling regular follow-up visits to discuss your response. These simple actions address some of the same gaps the Smart Cannabis model aims to fill.

Why does this matter if it is just an opinion piece?

Commentaries in major clinical journals can shape research priorities, attract grant funding, and influence policy discussions. This piece names a real and well-documented problem in medical cannabis care and proposes a specific, testable framework for addressing it. If researchers and policymakers take it seriously, it could lead to clinical trials and pilot programs that generate the evidence patients and clinicians actually need.

1. Lakhan SE, Driscoll B. Smart Cannabis: A prescription digital therapeutic framework for enhancing medical cannabis care. Clinical Therapeutics. 2026;48:46-50. doi:10.1016/j.clinthera.2025.12.002

2. Cannabis and EHR integration barriers under Schedule I status (cited as ref 1 in original; full citation not available in extracted text).

3. Definition and regulatory framework of prescription digital therapeutics (cited as ref 2 in original; full citation not available in extracted text).

4. Underreporting of cannabis use in clinical encounters and EHRs (cited as ref 3 in original; full citation not available in extracted text).

5. FDA regulatory definition of PDTs (cited as ref 4 in original; full citation not available in extracted text).

6. PDT for chronic insomnia via CBT-based digital intervention (cited as ref 5 in original; full citation not available in extracted text).

7. PDT for major depressive disorder via cognitive-emotional training (cited as ref 6 in original; full citation not available in extracted text).

8. Digital exposure therapy and autonomic training for PTSD (cited as ref 7 in original; full citation not available in extracted text).

9. Cannabis use in oncology for nausea, pain, sleep, fatigue, and mood (cited as ref 8 in original; full citation not available in extracted text).

10. PDT for migraine symptom-free days (cited as ref 9 in original; full citation not available in extracted text).

11. Cannabis for pediatric epilepsy management (cited as ref 10 in original; full citation not available in extracted text).

Have thoughts on this? Share it: