Should We Screen Patients for Opioid Abuse Potential?


New screening tools have been created, but which, if any, are right for you ED?

It is a given that pain is the most common reason why patients seek care in the ED. So emergency physicians must be experts at treating acutely painful conditions. For many years, the Joint Commission mandated treating pain as a vital sign, causing ED prescription of opioids to skyrocket. By 2012, a large study of 19 hospitals noted that 17% of discharged patients received an opioid prescription. But now there is recognition that pain control with non-opioids is just as efficacious for conditions such as non-radicular back pain and acute extremity pain. Programs that emphasize “alternatives to opioids” are popping up around the country, highlighting non-opioid pain control, regional anesthesia, and non-pharmacologic pain treatment.

Still, it would be impossible to practice EM without occasionally prescribing opioids. In the two hospitals where I work, we prescribed opioids to about 5% of discharged patients in 2017. That’s far less than the 17% of 2012, but still 1 in 20 patients, making this a common occurrence. And though there is extensive evidence that EPs are not a large contributor to the total number of opioid prescriptions, there is also evidence that a significant number of the patients we start on opioids will still be on them a year later, that some of patients who eventually misuse opioids will get them first from the ED, and that EPs who tend to prescribe more than their peers are more likely to have patients that are still on opioids at one year.

Therefore, judicious opioid prescribing from the ED is essential. Prescribing judiciously entails multiple steps. After deciding that an opioid is truly indicated, a check of your state’s prescription drug monitoring program (PDMP) is a best practice. Get into the habit of just checking it on everyone for whom you are considering writing an opioid.

Research shows that clinician gestalt is actually not very good at detecting “doctor-shopping” behavior. When an opioid is prescribed, a brief-but-focused discussion should be had with the patient, educating about safe use of opioids. Remind patients to take them only when other recommended non-opioid pain relievers are not helping, and with the goal of minimizing discomfort but not expecting to eliminate the pain entirely.  Also, it’s essential for patients to safely dispose of these medications if any are left over to prevent diversion and misuse.

As a final consideration, some guidelines recommend that patients be screened for abuse potential prior to prescribing an opioid. EDs screen for all types of conditions including depression, tobacco and alcohol use, and domestic violence. When public health crises arise, sometimes additional screening is performed, as we have seen for recent infectious disease outbreaks like Zika and MERS.  But what does screening for opioid use disorder look like, and how can it be applied in the ED setting?

Perhaps the most straightforward screening tool is the “NIDA Quick Screen,” a version of which is simply to ask the question in a straightforward and nonjudgmental manner: “How many times in the past year have you used an illegal drug or used a prescription medication for non-medical reasons?” This is also the first question of the Drug Abuse Screening Test (DAST). If the answer is more than once, it is recommended to perform the rest of the screener.

The DAST is formatted as either a 10 question or 20 question version, with the 10 question version being most appropriate for the ED setting. The DAST correlates well with people needing substance use treatment, but the main criticism is that it is very obvious. That is, the questions are very directed, such as, “Do you every feel bad or guilty about your drug use?” and, “Have you engaged in illegal activities in order to obtain drugs.” If the patient answers “yes” to questions like these, they clearly have problem use, but it would also be very easy to avoid telling the truth, especially if the goal of a patient may be to obtain an opioid prescription from the ED for non-medical use.

To that end, two other screening tools were developed that ask questions in a more subtle fashion: the Opioid Risk Tool (ORT) and the Screener and Opioid Assessment for Patients with Pain-Revised (SOAPP-R). The ORT gives points if the patient has a family or personal history of alcohol or drug use.

Additionally, there are points for having a high-risk age group (16-45 years), a history of preadolescent sexual abuse, and psychological disease (e.g. bipolar, schizophrenia and depression). The questions are not as pointed as the DAST, but rather capture risk factors associated with higher likelihood of opioid misuse. Whereas the ORT was developed by consensus, the SOAPP-R was rigorously developed with a derivation phase and validation phase followed by further refinement.

The result was a 24-question screening tool with more subtle questions, such as “How often do you have mood swings?” and “How often have you had to borrow pain medications from your family or friends?” in addition to the more obvious questions like “How often have you taken more pain medication than you were supposed to?”

It’s also necessary to recognize what screening tools like the SOAPP-R are actually predicting. The SOAPP-R, for example, was initially used in chronic pain patients seen at pain clinics. The study followed the patients out several months to determine presence of aberrant behaviors such as seeking medications from multiple providers or having positive toxicology screens for substances of abuse.

There is one glaring problem with all of these tools, which is that they weren’t developed or validated in the ED setting. We attempted to address this with a study comparing SOAPP-R with PDMP results in ED patients that were being prescribed opioids. Our criteria were that if you had 4 or more opioid prescriptions and 4 or more prescribers of controlled substances in one year, we determined that you had “aberrant drug-related behavior.”

Of course, this is by no means a perfect definition; there are legitimate reasons why some patients will have multiple doctors prescribing pain meds. Interestingly, the SOAPP-R had a high negative predictive value (89%) but was not very sensitive for detecting the outcome (54%). That is, if a patient had a lower score on the SOAPP-R, then it was very unlikely if they had a suspicious PDMP profile.

An additional problem with the SOAPP-R is its length. At first blush, the idea of performing a 24 question screener in the ED doesn’t sound very feasible. Surprisingly, though, when we programmed the SOAPP-R into a tablet computer and had patients complete it by themselves, 75% of participants completed the study in less than 3 minutes and 93% rated the ease of completion to be “very easy.”
One third of the patients screened positive, indicating that they were taking the survey seriously and probably answering truthfully.

For the efficiency nerds, two techniques – “curtailment” and “stochastic curtailment” – can further shorten screening tools. The idea of curtailment is simple: the screener stops once the test is “positive” or if there is no longer a chance that it would become positive with the number of remaining questions. This requires that the questions be stratified and works well for those with just one cutoff score that makes someone screen positive.

Stochastic curtailment uses probability to shorten the screener even further, like when you have 95% confidence that the patient will screen positive based on the previous questions answered. In one study of the SOAPP-R, curtailment shortened the length by 26% and stochastic curtailment reduced it by an average of 65%, all with very similar sensitivity and specificity to the original version. Additionally, a new study determined that by using only the most predictive questions, an 8-question version performed better than even the original 24-question screener.

Even so, there is great skepticism with prediction systems for opioid abuse. The recent CDC opioid prescribing guidelines state: “Clinicians should always exercise caution when considering or prescribing opioids for any patient with chronic pain outside of active cancer, palliative, and end-of-life care and should not overestimate the ability of these tools to rule out risks from long-term opioid therapy.”

Putting it all together, screening patients for opioid risk is problematic in the ED setting. We are already overburdened with tasks, and adding yet another step for evaluation must be done cautiously. What I recommend is this: no opioid prescription should be written without a careful consideration of the risks for all patients. As above, patients must be counseled on the safe and appropriate use of these medications, and it is our role as the prescribing physician to do this. We should always check the PDMP as well, but like with screening tools, recognize that the PDMP is only helpful if positive. It should not falsely reassure you if negative – patients may be obtaining their opioids from other sources. However, if you are pressed for time, at least ask this one-question screener: “How many times in the past year have you used an illegal drug or used a prescription medication for nonmedical reasons?”

We must be careful with all patients to whom we prescribe opioids. Taking a little extra time will help you recognize higher risk patients and counsel them appropriately about their opioid prescriptions.


Scott Weiner, MD, MPH is an emergency physician and director of the opioid stewardship program at Brigham and Women's Hospital, Boston, MA.


  1. Scott Weiner on

    Thanks for the embedded video – not only was it worth a good chuckle but also makes a fantastic point that resources are needed when a patient screens positive, and not everything should always fall on the backs of EPs. Nicely done.

  2. To be brutally candid, I did not read the article by Dr Weiner. I am, unfortunately, a painfully slow reader and just could not afford the time.

    But, I think all of this screening is BS. Patients come for a headache, a sprained ankle, a laceration, a bloody nose. Asking questions that are not germane to their problem is an inappropriate invasion of their privacy. It is none of our business if they have used illegal substances, or if they feel safe at home or whether their immunizations are up to date or if they are feeling depressed — or did they get their flu shot, want an HIV test or do they want a hepatitis C test if they are baby boomers? What is the NNS — number needed to screen. How many people will need to be asked these questions to get a positive (and will it be a true-positive) — and, then, what can we do about it? Do EDs have ready access to social workers and all of the other resources needed — most won’t. I can tell you that I have been an ED patient a time or two and I would find it extraordinarily annoying to be asked a battery of these questions that have nothing to do with why I was there and which I might find unnecessarily intrusive.

Leave A Reply