The AHRQ finding

The Agency for Healthcare Research and Quality’s 2020 comparative effectiveness review, Opioid Treatments for Chronic Pain, asked a specific question — its Key Question 5 — about the effectiveness of risk mitigation strategies in patients prescribed opioid therapy, listing eleven of them including urine drug screening, prescription monitoring data review, pill counts, risk assessment instruments and abuse-deterrent formulations.

No study evaluated the effectiveness of risk mitigation strategies, such as use of risk assessment instruments, opioid management plans, patient education, urine drug screening, prescription drug monitoring program data review, monitoring instruments, more frequent monitoring intervals, pill counts, abuse-deterrent formulations, or avoidance of co-prescribing of benzodiazepines on risk of overdose, addiction, abuse or misuse.

Chou et al., AHRQ Comparative Effectiveness Review No. 229, 2020

Read that carefully, because it is stronger and stranger than “the evidence is weak.” It is not that studies found no effect. It is that no study evaluated the question at all — for testing or for most of the rest of the standard risk-mitigation toolkit. An entire apparatus of clinical practice, mandated in many places, rests on an evidence base that has not been assembled.

The predecessor AHRQ report in 2014 reached the same conclusion in more general terms. An earlier systematic review of treatment agreements and urine drug testing found 11 eligible studies from 102 screened, all observational and of poor to fair quality, with modest reductions in misuse (7% to 23%) in the four that had comparison groups, and concluded the evidence was “relatively weak.”

And the results are frequently misread

A separate problem, and in practice a more dangerous one. A 2024 survey of 911 clinical and laboratory professionals across the US and Canada tested interpretation of urine drug test results with six questions. Toxicologists scored 4.82 out of 6 and laboratory PhDs 4.63. Physicians in pathology, emergency medicine, primary care and internal medicine “displayed concerning knowledge gaps.” The largest gaps were in identifying simulated compliance, understanding opioid metabolism, and immunoassay cross-reactivity. More than 30% said they would seek guidance on interpretation from the internet or from peers rather than from a laboratory expert.

The people acting on these results are frequently not the people who can read them.

False positives are routine and consequential

A review of 173 articles, condensed to 62 containing false-positive data, tabulated documented immunoassay false positives across amphetamines, opiates, benzodiazepines, cannabinoids, tricyclic antidepressants, phencyclidine, LSD and barbiturates. Its conclusion is the operational rule that should govern every use of these tests: immunoassay positives are presumptive until confirmed by a second, independent chemical technique.

A published case series makes the stakes concrete. Three pregnant or postpartum patients taking labetalol — a common antihypertensive in pregnancy — produced presumptive-positive fentanyl immunoassay results. In that setting an unconfirmed fentanyl positive can trigger child-protective reporting.

What the guideline says about consequences

CDC’s 2022 guideline addresses the downstream use of these results directly:

Clinicians should not dismiss patients from care on the basis of a toxicology test result. Dismissal could have adverse consequences for patient safety, potentially including the patient obtaining opioids from alternative sources and the clinician missing opportunities to facilitate treatment for a substance use disorder.

Dowell et al., 2022 — recommendation 10

It also states that clinicians should explain to patients that toxicology testing will not be used punitively and is intended to improve their safety.

One thing we could not find, and will not imply exists: no peer-reviewed study measures the rate of dismissal from care following a drug test result, and there is no validated measure of testing-associated stigma. The CDC statement above is normative — it says what should not happen, citing the misapplication literature, not a measured dismissal rate. The nearest empirical proxy is the finding that tapering was associated with subsequent termination of care at an adjusted odds ratio of 4.3.

So what should an agency conclude

Not that testing should stop. Testing produces real information that clinicians use, and the realistic choice in any jurisdiction is not between testing and no testing.

What follows is narrower:

• Do not fund testing on the basis that it is proven to mitigate risk. It is not, and the review that establishes this is a federal one.
• Require confirmatory testing before any result is acted on. This single rule prevents most of the documented harm.
• Require chain of custody, so a result belongs to the person it is attached to — see custodial toxicology and DNALock.
• Require that results are returned with interpretation, by or with access to someone competent to give it.
• Prohibit dismissal from care on the basis of a result, in the contract, in line with CDC.
• Measure whether dismissals are happening anyway — see measurement and evaluation.

Questions this raises

Is there evidence that drug testing reduces overdose or misuse?

No. The AHRQ 2020 comparative effectiveness review found that no study evaluated the effectiveness of urine drug screening — or of most other standard risk-mitigation strategies — on risk of overdose, addiction, abuse or misuse. That is an absence of evaluation, not a negative result, and the distinction matters. See absence of evidence is not evidence of absence.

What is the single most important rule for using drug tests?

Confirm before acting. Immunoassay results are presumptive, documented false positives exist across every major drug class, and labetalol prescribed in pregnancy has produced presumptive fentanyl positives with child-protective consequences. See presumptive is not positive.