Results of random drug testing in an adolescent substance abuse program.

OBJECTIVE

The objective of this study was to estimate from a random urine drug-testing program for adolescents the proportion of drug tests that are susceptible to interpretation errors.

METHODS

This was a secondary analysis of a clinical database and chart review from an adolescent outpatient substance abuse program at a large children's hospital. We analyzed from 110 adolescent patients (13-21 years of age) all 710 urine drug test results that were collected between December 2002 and July 2005 and 85 original laboratory reports for tests that were collected between December 2002 and May 2006 and were confirmed positive for opioids. We calculated the percentage of tests that were too dilute to interpret (potential false-negatives) and the percentage of confirmed positive tests for oxycodone that did not result in a positive initial screen (potential false-negatives). We also reviewed clinical information to determine whether confirmed positive tests resulted from legitimate use of prescription or over-the-counter medication (potential false-positives).

RESULTS

Of 710 drug tests, 40 negative tests were too dilute to interpret properly, and 45 of 217 positive tests resulted from prescription medication use for a total of 85 tests that were susceptible to error. Of the 85 confirmatory laboratory reports reviewed, 43 were positive for oxycodone, but only 16 of these had produced a positive opiate screen.

CONCLUSIONS

Unless proper procedures are used in collecting, analyzing, and interpreting laboratory testing for drugs, there is a substantial risk for error.