With no perfect way to assess risk, an objective tool combined with urine drug testing can help emergency physicians assess risk of dependence on long-term opioid use in patients with no history of addiction
Explore This IssueACEP Now: Vol 33 – No 04 – April 2014
“In chronic-pain patients who have been properly screened for addiction risk using the Opioid Risk Tool (see Table 1), the risk of addiction for those who are scored at low risk is less than 0.2 percent”
That 0.2 percent is a great stat! Does anyone know where exactly it comes from?
In response to the great question from Alyssa Berns, here are my comments.
The data on risk have been accumulated in our nine community-based chronic non-cancer pain clinics over a period of three years. In our clinics, we average 60,000 visits per year, representing approximately 8,000 patients. All patients fill out the Opioid Risk Tool (ORT), along with many other validated assessment forms, at the time of entry into the clinics. Patients also sign an opioid agreement and are required to undergo random urine drug screening at a frequency determined by that initial risk stratification (anywhere from two per year to monthly).
Our patient retention is high, so we have fairly consistent evidence with respect to compliance and substance misuse. Others, however, have not reported such consistency in the value of any screening tool they have used. Witkin et al (J Opioid Manag. 2013;9:177-187) found that 38 percent of their low-risk patients had aberrant drug-related behavior (ADRB). It is important to note that ADRB is not the equivalent of addiction and represents acting-out behavior seen both with oligoanalgesia and addiction.
The Canadian Opioid Guideline (cited in the article) recommends the use of the ORT to establish baseline risk; it is the preferred screening tool at present. Another tool, the Drug Abuse Screening Test (DAST-20), specifically looks at presence of addiction risk at the point in time of seeing the patient rather than predicting risk in the future, as the ORT is meant to do. In Gorchynski et al (Cal J Emerg Med. 2005;6:3-8), the DAST-20 identified 50 percent of the patients with addiction who would not have been identified by the clinician. This tool may be another one to consider.
We, as clinicians, cannot judge in a few minutes why the patient is manifesting ADRB…To help us better judge, validated tools and objective testing should be to what we turn.
A caveat: as per Brown et al (J Opioid Manag. 2011;7:467-483), the majority of patients seen in a family practice do not fall into the low-risk category; I would assume that would be equally true in the ED. Assessment of risk is to establish a level of comfort, along with the subjective (and biased) assessment of the caregiver; the latter has been shown to be fraught with error. Urine drug testing can also be of value: no one already testing positive for cocaine, for example, should be leaving any ED with a prescription for any controlled substance. The other important objective resource is the state-based drug-monitoring programs that allow us to see who has received what prescriptions.
There is, thus, no perfect way to assess risk. Use of an objective tool plus or minus urine drug testing does shift the weight toward a more objective assessment. The best example I can provide is that roughly 40 percent of patients presenting with a vaso-occlusive crisis are found to have ADRB—not surprising given the poor pain management most receive. In the Annals of Internal Medicine (2008;148:94), it was reported that 30 percent of patients with sickle cell disease experienced pain on at least 95 percent of days. When properly treated, however, only about 2 percent are found to have an addiction disorder, demonstrating that almost all ADRB seen is due to poor pain control in that population. One BMJ study found that only three of 1,900 patients with sickle cell disease were also suffering from addiction. Fishbain et al (Pain Med. 2008;9:444-459) show that, in patients with no past or current addiction problems, the risk of addiction from long-term opioid use is 0.19 percent, whereas the rate of ARDB is 11 percent. This, again, demonstrates the need to recognize that for the large majority of patients ADRB is not associated with addiction but rather with poor pain control.
We, as clinicians, cannot judge in a few minutes why the patient is manifesting ADRB; unfortunately, most of us see ADRB and assume it is due to diversion or addiction when in the large majority of cases it is due instead to oligoanalgesia. To help us better judge, validated tools and objective testing should be to what we turn.
—Jim Ducharme, MD, CM, FRCP