Effect Of A “No Superuser Opioid Prescription” Policy On ED Visits And Statewide Opioid Prescription

Abstract

IntroductionThe U.S. opioid epidemic has highlighted the need to identify patients at risk of opioid abuse and overdose. We initiated a novel emergency department- (ED) based interventional protocol to transition our superuser patients from the ED to an outpatient chronic pain program. The objective was to evaluate the protocol’s effect on superusers’ annual ED visits. Secondary outcomes included a quantitative evaluation of statewide opioid prescriptions for these patients, unique prescribers of controlled substances, and ancillary testing.MethodsPatients were referred to the program with the following inclusion criteria: ≥ 6 visits per year to the ED; at least one visit identified by the attending physician as primarily driven by opioid-seeking behavior; and a review by a committee comprising ED administration and case management. Patients were referred to a pain management clinic and informed that they would no longer receive opioid prescriptions from visits to the ED for chronic pain complaints. Electronic medical record (EMR) alerts notified ED providers of the patient’s referral at subsequent visits. We analyzed one year of data pre- and post-referral.ResultsA total of 243 patients had one year of data post-referral for analysis. Median annual ED visits decreased from 14 to 4 (58% decrease, 95% CI [50 to 66]). We also found statistically significant decreases for these patients’ state prescription drug monitoring program (PDMP) opioid prescriptions (21 to 13), total unique controlled-substance prescribers (11 to 7), computed tomography imaging (2 to 0), radiographs (5 to 1), electrocardiograms (12 to 4), and labs run (47 to 13).ConclusionThis program and the EMR-based alerts were successful at decreasing local ED visits, annual opioid prescriptions, and hospital resource allocation for this population of patients. There is no evidence that these patients diverted their visits to neighboring EDs after being informed that they would not receive opioids at this hospital, as opioid prescriptions obtained by these patients decreased on a statewide level. This implies that individual ED protocols can have significant impact on the behavior of patients.

The Characteristics and Effectiveness of Interventions for Frequent Emergency Department Utilizing Patients With Chronic Noncancer Pain: A Systematic Review

“…Each of the studies that measured the frequency of ED visits (n = 12) reported a decrease in the number of visits following the implementation of an intervention. [19][20][21][22][23][24][25][26][28][29][30][31] Nine of the studies reported statistically significant reductions in the frequency of ED visits (from p < 0.0001 to p < 0.05), 19,22,23,[26][27][28][29][30][31] one reported a nonsignificant reduction (p = 0.68), 21 and two studies did not report a p value. 19,24 • Patients referred to taper-to-abstinence pain management clinic run by a pain management and addiction specialist;…”

“…The types of opioid medications considered in the analyses of the studies that measured this outcome (n = 5) were hydromorphone, oxycodone, hydrocodone, tramadol, and codeine. 20,22,24,26,27 The method of measurement varied widely across studies with some reporting pill count, total number of prescriptions, or proportion of patients who received an opioid prescription. Regardless of the unit of measurement, all studies reported a statistically significant reduction in the number of opioid prescriptions following the implementation of their respective intervention (p < 0.05 to p < 0.001).…”

“…As detailed in Table 1, a total of four RCTs [19][20][21][22] and nine noncontrolled before-after studies (NCBAS) [23][24][25][26][27][28][29][30][31] were included, involving the assessment of 609 patients and 1,070 patients, respectively (total N = 1,679). The sample sizes of the RCTs ranged from 40 to 406 patients while the sample sizes for the NCBAS were slightly smaller ranging from 14 to 314.…”

“…Eight of the included studies implemented multifaceted interventions that included two or more of the identified components, 19,20,[22][23][24][25][26]30 and the remaining five studies implemented interventions that were focused around a singular component. 21,[27][28][29]31 Study Quality Assessment Four randomized studies were assessed using the Cochrane RoB 2.0 tool and nine nonrandomized studies were assessed using the ROBINS-I tool.…”

“…In nine of the 13 included studies we identified a high risk of bias due to confounding given the lack of a formal control group or a lack of procedures to control at the analysis stage. [23][24][25][26][27][28][29][30][31] We identified a high risk of bias due to participant selection in three of the 13 studies. [23][24][25] These concerns were most pronounced in the retrospective studies that participant selection for analyses occurred after the intervention had been implemented.…”