Results & Conclusions

Results & Conclusions


This descriptive study of over 900,000 UDTs ordered by physicians to monitor their patients’ appropriate use of opioids found alarmingly high rates of potential problems:

  • Approximately 11% of screens detected an illicit drug.
  • 29% found nonprescribed medications.
  • 38% found that a prescribed medication was not present.
  • 15% found opioid levels lower than expected ranges.
  • 27% found opioid levels higher than expected.

It is important to note that urine samples can fall into multiple categories, and thus patient sample result categories add to greater than 100%. Of the total samples screened, 75% had one or more of these problems. While this rate of problematic tests is considerably higher than that found by Katz and colleagues (46%), Katz did not consider the absence of prescribed opioid in the urine a “problem” test result, whereas a considerable number of patients (38%) fit into this category in the current study.1 These results clearly confirm that physicians and health plans should be vigilant in periodically screening patients who are prescribed long-term opioid therapy.

High rates of likely nonadherence and drug misuse are not surprising given the propensity for misuse, abuse, and diversion associated with these medications.

Vermeire et al reported that up to 50% of patients are nonadherent to their chronic medication regimens, regardless of disease, prognosis, or setting.2

Although the amount of demographic and clinical information available for this analysis was limited, the study confirms previously reported findings that “inappropriate” drug use spans all demographic groups, although to different degrees.

  • While males were approximately 75% more likely to have an illicit drug detected, in comparison with females, gender differences for other categories of findings were not observed.
  • Age differences with respect to illicit substances detected were quite pronounced, with illicit drug use nearly twice as likely in patients between the ages of 12–21 as patients between the ages of 35–50.
  • A reverse pattern was observed in individuals who had no prescribed medication present, with patients ages 35–50 years being over 30% more likely than patients ages 12–21 years to have no prescribed medication detectable in their urine.

Future multivariate analysis is needed to identify which, if any, of these characteristics appears to drive potentially inappropriate use. This in turn will help guide development of risk stratification models and characteristic-specific screening schedules.