Adding Cannabis to State Drug-Tracking Systems Was Linked to Fewer High-Risk Prescriptions

A 2024 study found that adding medical cannabis records to some state prescription drug monitoring programs (PDMPs) was associated with fewer prescriptions for medications that may interact adversely with cannabis. The analysis also identified a broader decline in prescriptions for certain controlled substances, although the researchers cautioned that the data cannot determine whether clinicians were making safer decisions, avoiding appropriate treatment, or responding to concerns about cannabis use.

The study, published in Health Economics by Shelby R. Steuart, examined Medicaid prescription claims in states that added medical cannabis information to their PDMPs. A PDMP is a state-run electronic database that records the dispensing of controlled substances and gives authorized clinicians additional information when evaluating a patient’s medication history. The Centers for Disease Control and Prevention describes PDMPs as a tool for improving prescribing decisions and identifying potentially risky medication combinations.

During the study period, nine states incorporated medical cannabis into their PDMPs. Steuart compared medication fills before and after the policy change with patterns in states that had legal medical cannabis dispensaries but did not add cannabis records to their monitoring systems.

In the study’s main analysis, adding cannabis information to a PDMP was associated with a 14.4% reduction in the number of units dispensed per prescription for medications classified as having severe contraindications with cannabis. Prescriptions for medications with moderate contraindications fell by 7.74% in units per prescription. The study also reported reductions across other measures of medication fills, suggesting that the changes were not limited to prescription size alone.

These results are consistent with the intended purpose of expanding a PDMP: giving clinicians a more complete view of a patient’s use of controlled medications and other substances before prescribing. However, the study measured changes in prescribing and dispensing patterns rather than clinical outcomes. It therefore cannot establish that the policy reduced adverse drug events, hospitalizations, or overdoses.

The analysis also found an apparent spillover effect involving scheduled medications. Schedule IV drugs were prescribed at an 11.4% lower rate after cannabis was added to the PDMP, while the number of units per prescription fell by 16.2%. Schedule II medications showed a smaller, moderately statistically significant decline.

Those findings may reflect more cautious prescribing, but they should not automatically be interpreted as evidence of provider bias. The study used Medicaid claims data and did not directly survey clinicians, examine individual prescribing decisions, or measure patients’ experiences. A reduction in controlled-substance prescribing could represent appropriate avoidance of dangerous combinations, an overly broad response to a patient’s cannabis use, or a combination of both.

This distinction matters because PDMP information is intended to support—not replace—clinical judgment. The CDC advises clinicians to use monitoring data alongside a patient’s medical history, examination, treatment goals, and other relevant information. It also recommends applying PDMP checks consistently rather than making assumptions about patients based on the information clinicians expect to find.

Improving how such information is presented may help. The Office of the National Coordinator for Health Information Technology’s PDMP integration resources focus on connecting monitoring data with electronic health records and clinical workflows. Easier access can make it more practical for clinicians to review medication histories while reducing the risk that a database entry becomes a simplistic reason to deny treatment.

Steuart’s findings provide evidence that adding cannabis data to PDMPs can change medication-use patterns among Medicaid enrollees. They do not, however, show that the policy itself caused better health outcomes or prove that clinicians discriminated against medical cannabis patients. Further research should examine adverse events, pain control, treatment access, and patient-level prescribing decisions to determine whether the reductions represent improved safety, unintended barriers to care, or both.