Access to the PPM Journal and newsletters is FREE for clinicians.
10 Articles in Volume 14, Issue #3
Practical Guide To Safe Use of Nonprescription Pain Medications
Common Causes of Acute Abdominal Pain
Early Treatment of TMD May Prevent Chronic Pain and Disability
Insomnia: Focus on New Dosing Concerns In Women
Is Marijuana Use Associated With Non-adherence To Opioid Therapy—Insights Gained From Urine Drug Monitoring
New Evidence-Based Diagnosis Criteria for TMD
New Rating Scale Helps Evaluate Refractory Chronic Migraine Patients
The Effect of Prolonged Knee Extension Immobilization on Knee Active Range of Motion: A Case Study on Arthrofibrosis
Opioid Bias Hurts Pain Patients
Can misoprostol be used for refractory chronic constipation?

Is Marijuana Use Associated With Non-adherence To Opioid Therapy—Insights Gained From Urine Drug Monitoring

A growing issue of concern among pain practitioners is whether people who use marijuana may also abuse or misuse prescribed pain relievers. This study examines the association between marijuana use and potential non-adherence in patients prescribed hydrocodone.

Marijuana and its active ingredient tetrahydrocannabinol (THC) have been used for pain relief for centuries. While the legalization of marijuana for recreational and/or medical use was being hotly debated across the country, investigators from the Substance Abuse and Mental Health Services Administration (SAMHSA) were compiling the latest statistics on THC use and abuse. In June 2012, SAMHSA published findings showing that 18% of patients admitted to substance abuse treatment centers reported marijuana as their primary substance of abuse, with 58% of those reporting abuse of additional substances, including opiates, sedatives, and alcohol.1

Of all the illicit drugs, marijuana has the largest number of users followed closely by pain relievers.2 Nearly 5 million people use marijuana on a daily or almost daily basis.2 Twenty states and the District of Columbia have approved the use of medical marijuana, citing benefits across a number of disease states, including nausea/vomiting and pain. Two states, Colorado and Washington, recently legalized the use of marijuana for recreational purposes.3-6

A number of studies have found that marijuana use, particularly in teens and young adults, is associated with abuse of prescription opioids and other drugs, as well as impaired driving ability.7-9 In one study, both male and female young adults (ages 18-25) with previous marijuana usage were 2.5 times more likely than those with no previous marijuana usage to have subsequently abused prescription opioids.7

Studies in chronic pain patients have demonstrated an association between prescription medication misuse and illicit drug use.10 In a recent report analyzing urine drug samples from 527,000 patients, 11.1% of the samples were found to contain one or more illicit drugs, 35.9% of samples did not contain the drug prescribed by the doctor, and 32.3% contained a drug not prescribed by the doctor.11 In these chronic pain patients, it is unknown if the association extends to THC or is limited to agents such as cocaine, heroin, etc. Therefore, we sought to investigate the relationship of marijuana use to potential medication non-adherence in patients prescribed hydrocodone, the second most frequently prescribed pain medication in the United States.12 The findings of our study may help clinicians clarify their practice policy on THC to achieve optimum outcomes for patients with chronic pain while minimizing risks.13 

Current Study Design

A retrospective review was conducted on a database of urine drug monitoring (UDM) results from samples submitted to Ameritox for patients prescribed hydrocodone from May 16, 2011 to May 15, 2012. Samples were included in the analysis if physicians ordered testing for both the marijuana metabolite, 11-nor-A9- tetrahydrocannabinol-9-carboxylic acid (THCA), and cocaine metabolite, benzoylecgonine. Information collected on the Ameritox lab requisition form included the dose, frequency, time of last dose, and whether hydrocodone was prescribed prn (as needed). Samples were excluded if hydrocodone was prescribed as needed, or if medical marijuana was listed among the prescribed medications.

The results of the UDM were grouped as having the prescribed hydrocodone not found by liquid chromatography tandem mass spectrometry (LC/MS/MS; below a cutoff of 100 ng/mL) or as having a non-prescribed medication found. This later finding included prescription medications that were not listed as being prescribed on the laboratory requisition form and illicit agents.

The results of UDM were further sorted into 3 categories for analysis based on the presence or absence of illicit agents: THC only, cocaine only (as a comparator because cocaine is typically considered a serious finding by clinicians), and no illicit drugs found—meaning tests were negative for all tested illicit agents, including THC, cocaine, phencyclidine (PCP), and MDMA (3,4 methylenedioxy-N-methylamphetamine).

Additional analysis was done to look for specific classes of non-prescribed drugs found. This analysis included opioids (codeine, morphine, hydrocodone, oxycodone, oxymorphone, buprenorphine, fentanyl, methadone, meperidine, propoxyphene, tramadol, and tapentadol), sedative hypnotics (benzodiazepines, barbiturates, and carisoprodol), and stimulants (amphetamine, methamphetamine, methylphenidate, and nicotine metabolite). The detection time and mass spectrometry lab cutoff for hydrocodone, THCA, and benzoylecgonine were as follows: hydrocodone (cutoff 100 ng/mL, detection time up to 3 days), THCA (cutoff 10 ng/mL, detection time up to 30 to 45 days in a chronic user), and benzoylecgonine (cutoff 100 ng/mL, detection time up to 3 days).14 A sample may be included in 2 categories if multiple non-prescribed medications were found. Results are presented as the aggregate of all eligible samples. Data was examined from an existing database and analyzed by researchers as de-identified, so Institutional Review Board approval was not obtained.


From May 16, 2011 to May 15, 2012, 250,397 eligible urine samples were received by Ameritox from individuals prescribed hydrocodone-containing medications. Of those, 116,001 samples (46%) had a physician order to test for marijuana (THC) and cocaine (COC) and were included in the study. Further breakdown of the study samples found that the vast majority of samples (99,115 or 85%) had no illicit drugs present. Of the remainder of samples, 15,153 (13%) were positive for THC only and 1,731 (2%) were positive for cocaine only. A further analysis of the samples positive for marijuana and cocaine found that they were more likely than the samples that were negative for marijuana and cocaine to be negative for hydrocodone (THC: 36.5% vs. 29.7%; COC: 59.6% vs. 29.7%) (Figure 1, Table 1). The samples that were positive for either marijuana or cocaine were more than likely to be positive for an additional medication compared with samples with no illicit drugs at all (THC: 29.1% vs. 22.0%; COC: 29.9% vs. 22.0%) (Figure 2, Table 1). Additional analysis showed that both marijuana- and cocaine-positive samples were more likely to have a non-prescribed opioid, sedative/hypnotic, and/or stimulant present compared with samples with no illicit drug present (Figure 3).


For the clinician caring for patients suffering from chronic pain, the challenges are many. Given the millions of marijuana users, deciding what to do with a positive marijuana urine drug test in a patient prescribed opioids is a common dilemma. Some clinicians have opted to simply not test for THC, given the fact that in many circles, including some state legislatures, marijuana is considered to possess medicinal value and perhaps be suitable for recreational or medicinal use.

The results presented here suggest that for a small percentage of patients, marijuana use is associated with misuse of prescription medications. The study found that patients who are prescribed hydrocodone and who tested positive for marijuana also tested positive for missing their opioid and/or were found to be taking some other non-prescribed medicine more often than patients who were not using an illicit substance (trended higher but did not reach statistical significance). For chronic pain patients who are prescribed an opioid, the odds of finding some other drug in a UDS were similar for those who were positive for THC and those who were positive for cocaine. The odds ratio for finding non-prescribed stimulants and sedative/hypnotics was higher among those positive for THC than among those positive for cocaine. The odds ratio of finding a non-prescribed opioid was higher in the cocaine users, however.

Limitation of the Study

There are several limitations of this study. The population we focused on was patients prescribed hydrocodone, and other prescribed opioids have not yet been analyzed.

The results of UDM may not be reflective of the overall population of patients because physicians may test patients suspected of medication misuse more frequently, thereby resulting in a possible selection bias.

As categories of urine drug testing results are not mutually exclusive, the samples with multiple abnormal results may be contributing to an overstatement of the problem. The analysis is dependent on the accuracy of the information provided on the laboratory requisition form.

The results are strictly looking at objective measures of drug in the urine and the authors have no way of knowing the reasons for use of illicit medication and or non-adherence to prescribed therapy.


This data suggest that marijuana use is associated with significant medication non-adherence. With respect to the presence of a non-prescribed medication, the association is equal to that reported in this study for cocaine. These results indicate that marijuana use should be explored via UDM in chronic pain patients prescribed opioids. For clinicians who continue to prescribe opioids even if UDM confirms the use of marijuana, this data suggests that these patients should be considered at higher risk of medication misuse than patients not using illicit drugs. Consistent with recently published expert recommendations,15-17 a high-risk patient warrants more frequent follow-up, including the use of clinical tools such as prescription drug monitoring programs and more frequent urine drug testing.


Last updated on: May 25, 2017
close X