Correspondence

Contents lists available at ScienceDirect

American Journal of Emergency Medicine

journal homepage:

A program of education and performance feedback reduces CT ordering in the emergency department?

Dear Editors,

We would like to share the results of a study examining the effects of a 3- part intervention on the rate of computed tomography ordering in the ED. The Institute of Medicine recently challenged medical specialties to “Choose Wisely;” and to reduce unnecessary testing [1]. The American Col- lege of Emergency Physicians (ACEP) chose 4 CT scans to target [2]. In order to address this goal, our partnership developed a 3 part educational in- tervention consisting of an initial 30 min power point presentation, subse- quent emailed supplemental material and monthly feedback. This analytic, non-randomized, multi-cohort study was conducted at three sites represen- ting three states (California, Oregon and Washington), all staffed by the same Partnership. We found a modest, 10% relative reduction (a 2% overall reduc- tion) in CT ordering, without any apparent reduction in patient safety.

educational interventions without other incentives do not often re- sult in changes in behavior. The unique portion of our intervention was attention to human factors via the non-blinded monthly report. The initial 30 min power point discussed the usual reasons for CT order- ing reduction; and the emailed supplemental material included the ex- pected information: clinical decision tools for not ordering CT scans in Blunt head trauma [3-6], a review paper on the radiation risks associat- ed with CT scanning [7], and ACEP’s choosing wisely guidelines [2].

The non-blinded monthly updated CT ordering reports were reviewed at the monthly department meetings; these reports averaged 3 pages. They compared each provider’s personal CT ordering to those of peers in the same ED. Orders were presented via unblinded bar graphs. Each provider could compare their own personal ordering practice against that of each of their peers, and vice versa.

Human factors are important yet seldom overtly addressed by simi- lar studies. We found that providers often felt that their personal rates of CT ordering were justified because their patients were sicker than the patients’ of others. We believe that comparing their personal practice styles against those of known colleagues was important although we did not design the study to test the strength of this factor separately. Re- sults were the same at three disparate sites, reducing the possibility that some non-captured confounding variable accounted for the change. Being able to compare their personal practice style against known col- leagues may have been the more important factor.

In the combined practices, 97 providers (physicians, advanced prac- tice nurses and physician’s assistants) encountered 159,493 emergency patients between January 1, 2014, and February 19, 2015. CT ordering data was obtained by abstracting the relevant CPT codes from the elec- tronic medical record. Random effects and confounders were statistically controlled. Patient characteristics did not significantly change during the pre- and post-intervention periods, nor did the providers change signifi- cantly. However, there was a modest, 2% reduction in CT ordering, from

? This paper has not been presented publically.

0735-6757/(C) 2017

17.5% of patients to 15.5% (95% C.I. 1.7%-2.4%), which is a 10% relative re- duction. After accounting for confounders, and the ED and provider level random-effects, the odds ratio for ordering a CT scan after the interven- tion was reduced to 0.936 (95% C.I. 0.887-0.990) from baseline. This is ap- proximately a 10% relative reduction in CT scanning.

Although this reduction is small, if translated nationally could bring real savings to the ED in terms of length-of-stay, monetary costs and ra- diation exposure [7,8].

Our study has limitations. The 3 sites represent 3 states, all within the same partnership. Results may not generalize to other staffing models. Quality of care was not directly measured, but the crude mea- sures of 30 day return rate and patient satisfaction were not different.

In summary, this study demonstrated that a simple, data-driven in- tervention reduced ED CT ordering by 2%, which is a 10% relative reduc- tion. We recommend that emergency department leaders interested in lowering rates of CT scans consider educating providers on CT ordering guidelines as well as on the risks and benefits of CT scans, and then share monthly dashboards of their individual utilization rates.

Grants/financial support

There was no financial support.

Conflicts of interest

No conflicts of interest for any author.

Author contributions

GM and JTS conceived the study and designed the trial. GM present- ed educational intervention. GM and JST supervised the data collection. CW, DD, and KSW searched the literature and obtained appropriate ref- erences, assisted in writing the manuscript, and helped format the ta- bles. JTS managed the data, including quality control. JTS provided statistical advice on study design and analyzed the data; VT drafted the manuscript, and all authors contributed substantially to its revision. GM takes responsibility for the paper as a whole.

Acknowledgements and appreciation

Jacqueline Cleto and L. Kriger. Study coordinators. Advised with study: Prentice Tom MD. [email protected].

Medical director at Site One: Erik Egsieker MD. [email protected]. Medical director at Site Two: Laura Cook MD. [email protected].

G. Miller

J. Tamayo-Sarver

CEP America, 2100 Powell St Suite 900, Emeryville,

CA 94608, United States E-mail addresses: [email protected] (G. Miller), [email protected] (J. Tamayo-Sarver)

V.Y. Totten*

D. Denson

C. Wishka

K.S. Whitlow Kaweah Delta District Hospital, 400W.Mineral King Dr.,

Visalia CA 93291, United States

* Corresponding author E-mail address: [email protected] (V.Y. Totten), [email protected] (D. Denson), [email protected] (C.Wishka),

[email protected] (K.S.Whitlow) http://dx.doi.org/10.1016/j.ajem.2017.02.041

References

1. http://abimfoundation.org/what-we-do/choosing-wisely. (accessed Nov 22, 2016).
2. American College of Emergency Physicians (ACEP). Choosing wisely guidelines. Avail- able at: http://www.choosingwisely.org/societies/american-college-of-emergency- physicians/. (Accessed December 10, 2015).
3. Ip IK, Raja AS, Gupta A, Andruchow J, Sodickson A, Khorasani R. Impact of clinical de- cision support on head computed tomography use in patients with mild traumatic brain injury in the ED. Am J Emerg Med 2015;33(3):320-5.
4. Haydel MJ, Preston CA, Mills TJ, Luber S, Blaudeau E, DeBlieux PM. Indications for computed tomography in patients with Minor head injury. N Engl J Med 2000; 343(2):100-5.
5. Stiell IG, Wells GA, Vandemheen K, et al. The Canadian CT Head rule for patients with minor head injury. Lancet 2001;357(9266):1391-6.
6. Kuppermann N, Holmes JF, Dayan PS, et al. Identification of children at very low risk of clinically-important brain injuries after head trauma: a prospective cohort study. Lancet 2009;374(9696):1160-70.
7. Cardis E, Vrijheid M, Blettner M, et al. Risk of cancer after low doses of ionising radi- ation: retrospective cohort study in 15 countries. BMJ 2005;331(7508):77.
8. Semelka RC, Armao DM, Elias Jr J, Huda W. Imaging strategies to reduce the risk of ra- diation in CT studies, including selective substitution with MRI. J Magn Reson Imaging 2007;25(5):900-9.

Fig. 1. (a) OOD frequency seen at JMH ED 2014-2016. Blue line indicates a clear upward trend from July-September (b) Gross number of 2 mL Naloxone vials used from July- September by the JMH ED. There is a staunch increase in utilization in 2016 (orange).

Fentanyl laced heroin and its contribution to a spike in heroin overdose in Miami-Dade County

Keywords:

Heroin overdose Fentanyl

opiate use Fentanyl laced heroin

With Heroin use increasing nationally, suppliers have been faced with an unprecedented demand. To increase supply while maintaining potency, heroin is often tainted with additives, commonly Synthetic opioids such as fentanyl or a fentanyl analog. Unfortunately, heroin users are unable to discern whether they have pure heroin or fenta- nyl/fentanyl-analog laced heroin (FLH). When a user injects what ap- pears to be pure heroin, they may in fact be injecting a substance which is hundreds of times more potent, leading to an increased risk of overdose and death [1,2,3,4].

During July-September of 2016 there was an unparalleled increase in opiate overdoses seen in Jackson Memorial Hospital’s Emergency De- partment (JMH ED) in Miami, FL. This spike was accompanied by a per- ceived increase in the number of naloxone doses that were necessary to reverse overdoses, which potentially indicates the involvement of syn- thetic opioids such as fentanyl or fentanyl-like compounds in the over- dose [5,6]. The increase in heroin overdoses coupled with higher naloxone dosing led to the postulation that FLH may have been present in the local heroin supply and driving the spike.

To test this hypothesis, a retrospective cross-sectional study was			2015	2016	% change (2015 to 2016)
performed to compare the number of opiate overdose (OOD) cases		July	92	98	7
treated at JMH ED from July-September of 2015 and 2016. Patient		August	81	158	95
charts were selected based on ICD-10 codes for OOD. The number of nal-		September	65	266	309
oxone (Narcan(R)) vials (2 mg) used from July-September 2015 and		Total	238	522	119

2016 was determined by accessing the ED pharmaceutical removal da- tabase, which documents any removal of a controlled substance. A vials/ patient ratio was calculated by dividing the total number of 2 mL nalox- one vials used by the total number of OOD cases seen in the same time frame. This ratio allowed an estimation of the average naloxone usage per case over a given period. A one-way analysis of variance (ANOVA) was used to compare the mean naloxone utilization from July-Septem- ber in 2015 and 2016 by the JMH ED.

The number of OOD cases seen at JMH ED from the July-September period increased by 119% between 2015 and 2016 (238 to 522 cases, re- spectively). The greatest increase in OOD was observed during the month of September with 65 cases in 2015 and 266 cases in 2016 (182%) (Fig. 1a/Table 1). The number of naloxone vials used from July-September rose from 212 in 2015 to 1221 in 2016, a 476% increase (Table 2a). The greatest change in naloxone utilization was also ob- served in September, jumping from 55 in 2015 to 568 in 2016, a 933% increase (Fig. 1b/Table 2a). A One-way ANOVA illustrated the difference in mean naloxone utilization by JMH ED from July-September 2015 and 2016 was statistically significant [F(1,4) = 10.67, p = 0.030] (Table 2b). The vials/patient ratio increased from 0.89 in 2015 to 2.33 in 2016.

Interestingly, while both the number of naloxone vials utilized and the frequency of OOD cases increased in 2016, they did not increase pro- portionally. The disproportionate increase in naloxone use relative to OOD suggests that the driver of overdose was not traditional heroin,

Table 1

OOD cases at JMH ED; Frequency table of OOD showing the most notable increase in Sep- tember of 2016.

overdose cases at JMH ED