Characteristics Predictive of Opioid Nonadherence

Understanding factors that influence medication nonadherence may guide clinicians in implementing adherence strategies.

Research presented at the 32nd annual meeting of the American Academy of Pain Medicine (AAPM) revealed that nonadherence to prescribed opioid treatment is common and varies with different patient characteristics.

Characteristics that predicted opioid nonadherence included gender, age, and type of insurance.

“Urine drug monitoring allows the provider to identify potential nonadherence early in treatment. Nonadherence does not always mean abuse, misuse, or diversion,” Patricia Woster, PharmD, from Ameritox told Clinical Pain Advisor. “The patient may be experiencing side effects of the drug and therefore not taking the medication as prescribed, or the medication may not be alleviating the pain so the patient is not taking the medication.”

TRENDING ON CPA: Obama Pledges Additional $1.1B to Combat Opioid Abuse 

In order to gain a better understanding of which factors might influence nonadherence of opioid prescriptions, Dr Woster and colleagues used liquid chromatography/tandem mass spectrometry to analyze 447 435 urine samples from patients who were prescribed opioid medications between January 2013 and July 2015. If parent drug and/or metabolite(s) were detected, samples were confirmed as positive for opioid medication. If neither was detected, samples were confirmed as negative.

The researchers found that of the samples analyzed, 43.0% tested negative for the opioid medication prescribed. Potential nonadherence was lower in men compared with women (41.7% vs 44.0%; adjusted odds ratio [aOR], 0.94; 95% confidence interval [CI], 0.93-0.96).

Potential nonadherence also decreased as the decade of participants’ age increased (62.2% [10 to 19 years] to 34.5% [80 to 89 years]; aOR, 1.50; 95% CI, 1.29-1.73 [10 to 19 years] and 0.48; 95% CI, 0.45-0.51 [80 to 89 years] compared with 20 to 29 years).

“The most surprising findings were the potential nonadherence rates in the later decades of life,” Dr Woster said. “We saw potential nonadherence rates of 39.3% in the 60- to 69-year-old age group, 36.9% potential nonadherence in the 70- to 79-year-old age group, and 34.5% potential nonadherence rate in the 80- to 99-year-old age group.”

Another surprising aspect of their findings was a potential nonadherence rate of 50.8% in those prescribed more than one opioid medication. “We don’t know how the two opioids were prescribed in these patients (eg, one prescription to be taken around the clock and one for breakthrough pain PRN, or if there were 2 prescriptions written to be taken concurrently), but it was an unexpected finding,” Dr Woster noted.

The researchers also found that compared with self-pay patients, potential nonadherence was higher for Medicaid patients (52.0% vs 43.6%; aOR, 1.36; 95% CI, 1.32-1.39) and lower for patients with commercial insurance, Medicare, or Workers’ Compensation as the primary payer (36.1% to 40.8%; aOR ranging from 0.79 to 0.90).

“Understanding factors that influence potential medication nonadherence may guide clinicians in patient identification and implementing adherence strategies,” the researchers noted.

“We strive to present results that highlight the use and potential misuse of prescription pain medications,” Dr Woster said. “One thought [for future research] would be to look at medication adherence among the elderly (>60 years of age) by including the breakdown of illicit substances detected.”


Woster P, Ko M, Smith T, Fileger M. Predictors of medication adherence assessed by urine drug monitoring in patients prescribed opioid medications. Presented at: 32nd Annual Meeting of the American Academy of Pain Medicine.  February 18-21, 2016; Palm Springs, California. Abstract 220.