Video Abstract

Video Abstract

Close modal
BACKGROUND:

Opioid overdose and abuse have reached epidemic rates in the United States. Medical prescriptions are a large source of opioid misuse. Our quality improvement initiative aimed to reduce opioid exposure from the pediatric emergency department (ED). Objective was to reduce opioid doses prescribed weekly from our ED by 50% within 4 months.

METHODS:

Three categories of interventions were implemented in Plan-Do-Study-Act cycles: guidelines and education, electronic medical record optimization, and provider-specific feedback. Primary measures were opioid doses prescribed weekly from the ED and opioid doses per 100 ED visits. Process measures were opioid prescriptions, opioid doses per prescription, and opioid prescriptions for unspecified abdominal pain, headache, and viral upper respiratory infection. Balancing measures were phone calls and return visits for poor pain control in patients prescribed opioids and reports of poor pain control in call backs to orthopedic reduction patients. We used statistical process control to examine changes in measures over time.

RESULTS:

Opioid doses decreased from 153 to 14 per week and from 8 to 0.7 doses per 100 ED visits in 10 months, sustained for 9 months. Opioid prescriptions, opioid doses per prescription, and prescriptions for unspecified abdominal pain, headache, and viral upper respiratory infection decreased. Phone calls and return visits in patients prescribed opioids did not increase. There were 2 reports of poor pain control among 152 orthopedic reduction patients called back.

CONCLUSIONS:

We decreased opioid doses prescribed weekly from the pediatric ED by 91% while minimizing return visits and reports of poor pain control.

The United States is in the midst of an opioid crisis. There is an epidemic of nonmedical use of prescription opioids, accidental and intentional drug poisonings, and a need for opioid associated treatment. In 2017, the number of overdose deaths attributed to opioids, prescription and illegal, was 6 times higher than in 1999.1 

The opioid epidemic affects children as well as adults. Reports to poison control centers, emergency department (ED) visits, hospitalizations, and intensive care admissions in children related to opioid exposure have increased in recent years.24  Opioid misuse and medical prescription use in adolescents has been associated with future adult opioid misuse.5  In a recent study, 1 in 12 high school students surveyed reported nonmedical prescription opioid use in the past year.6  The availability of opioids at home is often cited as a risk factor for children and teenagers ingesting opioids, whether intentionally or unintentionally.4,7,8  Opioids are often prescribed in excess quantities, with much of the medication left over, unconsumed, and improperly disposed, contributing to this problem.8,9  In addition, there is growing evidence that nonopioid pain medications are as effective as opioids in treating acute pain in children, particularly nonsteroidal anti-inflammatory agents for orthopedic conditions.1012 

In our San Diego region, medical examiner data from 2000 to 2016 identified 174 deaths involving opioids in patients 0 to 21 years old. We identified that 46% of those patients had been previously evaluated at Rady Children’s Hospital San Diego (RCHSD) at least once. We identified a need to reduce opioid prescribing from the ED and better educate patients and families regarding opioids. Our global aim was to decrease our opioid exposure on the San Diego community with the specific objective to reduce the total number of opioid doses prescribed weekly from the ED by 50% within 4 months.

The RCHSD ED is an urban, pediatric ED serving ∼100 000 children per year and is the only freestanding pediatric ED serving San Diego, Imperial, and South Riverside counties. It is part of a health network composed of 220 primary care and 370 specialty physicians, serving 90% of pediatric patients in these counties. In 2019, the race and ethnicity distribution of patients discharged from our ED was 53% Hispanic, 31% white non-Hispanic (NH), 7% Black NH, 5% Asian NH, and 4% other (including Pacific Islander, American Indian or Alaskan native, and other). In 2019, the insurance payer distribution for patients seen in our ED was 63% government (including Medi-Cal, Medicaid, and Medicare), 27% private, 7% military, and 3% self-pay. The majority (>95%) of opioid prescriptions are written in our ED by ∼35 pediatric emergency medicine (PEM) faculty.

The project was done in collaboration with a hospital-wide opioid task force, composed of physicians, nurses, electronic medical record (EMR) analysts, and pharmacists from outpatient and inpatient specialties. In addition, we assembled a local multidisciplinary ED quality improvement (QI) team that included PEM physicians, ED nurses, EMR analysts, and an epidemiologist-biostatistician. The ED QI team reviewed baseline data from the preceding 6 months, determined a baseline ED discharge opioid prescribing rate of 153 doses per week, and set the goal of reducing this number by 50% to 76 doses per week within a 4-month period. Using process mapping and failure mode and effects analysis, we determined our key drivers and interventions and constructed a key driver diagram (Fig 1). The key driver diagram was periodically updated over the course of the QI initiative. We constructed a Pareto chart (Fig 2) of the most-frequent diagnoses for which the ED had prescribed opioids at discharge in the preceding 6 months. We determined that orthopedic diagnoses constituted 61%, with 62% of those undergoing reduction in the ED, highlighting the importance of addressing those diagnoses in our initiative.

There were 3 categories of interventions implemented over the course of 14 months using Plan-Do-Study-Act cycles: (1) guidelines and education, (2) EMR optimization, and (3) provider-specific feedback.

Guidelines and Education

As part of our initial interventions in February 2018, RCHSD disseminated guidelines recommending a maximum of a 3-day supply per opioid prescription, in agreement with the Centers for Disease Control and Prevention and American Academy of Emergency Medicine guidelines regarding opioid prescribing.13,14  In addition, every patient and family prescribed an opioid automatically received an educational handout as part of their after-visit summary, which included information on nonopioid pain control, signs of opioid overdose, and how to safely dispose of excess opioids. A revised version of the handout, with additional options for safe medication disposal and information about naloxone administration, was implemented in January 2019. Specific orthopedic reduction pain control discharge instructions were implemented in the ED in April 2019 in response to the significant decrease in opioid prescribing for this patient population.

In February 2018, the ED concurrently published guidelines recommending a maximum of 5 doses per prescription. The ED decided to be stricter in its prescribing recommendations as the access point of the hospital and because, based on our Pareto chart, the diagnoses for which we prescribed had good primary care and specialist follow-up. These guidelines were disseminated to all ED providers through e-mail and presentations at division meetings. In June 2018, these guidelines were included in an educational poster that was displayed throughout the ED. The poster included the recommendation to not prescribe opioids for the diagnoses of unspecified abdominal pain, headache, and viral upper respiratory infection (URI). In July 2018, an additional 6 new PEM providers joined the group and were oriented to the initiative and the opioid prescribing guidelines.

EMR Optimization

As part of our initial interventions in February 2018, all ED discharge opioid prescriptions (tablet and liquid) were automated to 5 doses in the EMR. ED providers had the ability to override this default when they felt it was clinically appropriate. Examples shared with providers as to when they might override this default included patients with sickle cell crisis or nephrolithiasis discharged from the ED before the weekend and/or with anticipated delays in follow-up. In January 2019, we further optimized the liquid prescription defaults. We made modifications so that providers did not have to calculate the total volume to dispense 5 doses because the QI team had noted numerous errors in these calculations. Our EMR does not auto-calculate the total volume for as-needed (PRN) prescriptions, so we circumvented that restriction and developed weight range options for the total volume that were equivalent to 5 to 7 doses.

Provider-Specific Feedback

In June 2018, we initiated provider-specific e-mail feedback. Each of the PEM faculty members received an individualized e-mail with their prescribing data compared with the PEM group mean. Reported metrics included number of opioid prescriptions per month and opioid doses per prescription. Providers were alerted if they were in the top 2 or top 5 for those metrics and if they prescribed 0 opioids. E-mails to high prescribers typically led to further e-mail and in-person discussion regarding individual prescribing practices and ED guidelines. E-mail feedback was continued in November 2018 and April 2019. Based on feedback from providers, the data in the individualized e-mails was better contextualized, accounting for the variation in the number of hours that providers work and the number of patients they see. In April 2019, an anonymized, comparative display of individual provider performance during the initial and later stages of the initiative was shared with the PEM group.

The primary measures were total opioid doses prescribed weekly from the pediatric ED and opioid doses per 100 ED visits, to account for variations in ED volume. Additional process measures tracked weekly included total opioid prescriptions, mean opioid doses per prescription, and opioid prescriptions for unspecified abdominal pain, headache, and viral URI diagnoses. As a balancing measure, we tracked telephone encounters and return visits for pain documented in our EMR to the ED, primary care, or specialist within a week for patients prescribed opioids. Given that our health network serves 90% of pediatric patients in our catchment area and that our integrated EMR also captures encounters at other large health networks in the area, we are confident that we captured the vast majority of return visits. Because our largest ED population prescribed opioids at discharge preintervention were those who underwent orthopedic reduction in the ED (38%), we also made phone calls within a week of discharge to all of these patients regardless of whether they were prescribed an opioid. We asked whether they had picked up and taken opioid prescriptions and assessed their pain control status.

Our institution’s institutional review board granted the study a QI exemption. Statistical analysis of patient demographic characteristics was performed using SPSS, version 24 (IBM SPSS Statistics, IBM Corporation). Student t test was used for continuous variables and χ2 test was used for categorical variables. We used Microsoft Excel and QI Macros to develop statistical process control charts to examine changes in measures over time. We used established rules for interpretation of control charts.15  The centerline and control limits were revised when an intervention was associated with special cause variation, as defined by 8 consecutive points above or below the mean. Special cause variation in the form of a single data point outside of the control limits were investigated for possible causes to guide improvement.

The demographic characteristics of the ED patients prescribed opioids at discharge at baseline and during the QI initiative are displayed in Table 1. There were no statistically significant differences in the average age, sex distribution, or race and ethnicity distribution of the patients between those time periods.

Over the 10 months from the start of the initiative, we achieved 3 statistical process control chart shifts associated with our interventions and reduced opioid doses prescribed from the ED from 153 to 14 doses per week (Fig 3) and from 8 to 0.7 doses per 100 ED visits (Fig 4). This change was sustained for an additional 9 months from the time of the last shift and 5 months from the time of the last intervention.

As shown in Fig 4, 9 consecutive data points below the center line were noted between November 2017 and January 2018. However, a control chart shift was not made at that time because this special cause variation was presumed to be due to baseline seasonal variation in opioid prescribing, because orthopedic complaints present more commonly in the summer and fall months. In addition, shifts were only made after initiation of and in association with interventions.

The first shift from 153 to 59 doses per week and 8 to 3 doses per 100 ED visits was achieved within 4 months, exceeding our initial aim of a reduction to 76 doses per week within this time frame. However, substantial amounts of special cause variation in the form of single data points outside of the control limits were still noted in the data. On investigation, they were found to be related to differences in provider prescribing practices. Prescribing numbers varied on the basis of which providers were working shifts each week. Provider-specific feedback was then initiated, with the goal to further reduce variation. A new aim to reduce doses per week to 15 was chosen on the basis of the potential for further improvement with provider-specific feedback and the potential for a shift in opioid doses per prescription, which had not yet occurred.

Prescriptions per week decreased from 12.8 to 2.3 with 3 control chart shifts in 10 months, sustained for 9 additional months (Fig 5). Doses per prescription decreased from 10.9 to 6.1 with a sustained shift occurring 9 months after the start of the initiative (Fig 6). Prescriptions for unspecified abdominal pain, headache, and viral URI diagnoses decreased from 0.5 (upper control limit [UCL]: 2.6, lower control limit [LCL]: 0) to 0.03 (UCL 0.5, LCL 0). Weeks between a prescription for any of these diagnoses increased, and special cause variation was achieved (Fig 7).

The funnel plots in Figs 8 and 9 depict the changes in individual provider opioid prescribing over the course of the initiative. Between the February to June 2018 reporting period and the October 2018 to January 2019 reporting period, mean provider opioid doses prescribed per 1000 ED patient visits decreased from 37 to 12, a 68% reduction. In addition, 3 notable outlying prescribers moved closer to the group average.

Return visits and telephone encounters per opioid prescription did not increase, remaining at 0.05 (UCL: 0.4, LCL: 0) over the course of the QI initiative. Opioid prescriptions between a return visit or telephone encounter did not decrease (Fig 10). From June 2018 to June 2019, there were 881 patients who underwent orthopedic reduction in the ED. A total of 598 (68%) were called, 397 (45%) were left a voice message encouraging them to call back, and 152 (17%) were reached. Of the 146 of 152 (96%) patients reached not prescribed opioids at discharge, only 2 patients reported poor pain control. Of the 6 (4%) patients prescribed opioids, all of them picked up the prescriptions, 5 took the medication, and none took all 5 doses (mean doses taken = 2.8). Overall, there were no complaints regarding the initiative or lack of opioid prescribing received through our patient complaint system.

Using QI methodology, we reduced total opioid doses prescribed on discharge from the pediatric ED by 91% within 10 months and sustained this change for an additional 9 months. The largest reduction (61%) in opioid prescribing occurred after the initial release of the ED prescribing guidelines and the EMR default of 5 doses for all ED discharge opioid prescriptions. Given the higher level of reliability of default dosing, we suspect that this intervention had the most impact. Default dosing is an example of a clinical nudge. Clinical nudges change the choice architecture of a decision and alter provider behavior in a predictable way without taking away options.16  Previous literature suggested that default dosing might be the best way to decrease provider variation in opioid prescribing for pediatric upper extremity fractures.17  The greater reductions made by our initiative in comparison with previous work that used only educational interventions to reduce ED opioid prescribing support this theory.18 

Another key intervention to this initiative’s success was provider-specific feedback to PEM faculty. Provider-specific feedback is an effective QI intervention, particularly when administered by a supervisor or colleague, provided more than once, and delivered verbally and in writing.19  Although initially implemented with apprehension, individual e-mails to high prescribers led to productive conversations initiated by those individuals that ultimately led to reduced prescribing. Previous study of provider feedback administered in an adult ED setting also demonstrated reductions in opioid prescribing.20 

The decrease in total prescribed opioid doses from the ED did not result in an increase in return visits or telephone encounters in patients prescribed opioids nor in significant reports of poor pain control in patients who underwent orthopedic reduction in the ED. This work aligns with the results of researchers in previous studies who have demonstrated the equivalent efficacy of nonsteroidal anti-inflammatory drugs to opioids in treating pediatric pain in the setting of acute extremity injuries and postfracture care.1012  In the call-back component of this initiative, none of the patients who underwent orthopedic reduction in the ED who were prescribed opioids at discharge took all 5 doses, in agreement with results of a previous study indicating that discharge opioids in children are often prescribed in excess quantities.8,9  Finally, we did not receive negative feedback regarding this initiative through our patient complaint system, which may help to alleviate concerns by other areas of our hospital or other EDs looking to implement similar work.

There were several limitations to our study. To our knowledge, there are currently no consensus guidelines available regarding opioid prescribing in the setting of acute pain in children. We developed our hospital and ED guidelines on the basis of the Centers for Disease Control and Prevention and American Academy of Emergency Medicine guidelines13,14  and chose to be more strict in our ED guidelines as the access point of the hospital and because, in general, patients seen at our institution have reliable specialty and primary care follow-up. The accessibility of follow-up at our hospital and in our region potentially limits the generalizability of our work to other hospitals and regions. In addition, recent literature has described significant variations in opioid prescribing practices by race and ethnicity and insurance status for pediatric fractures.21,22  Although we did not note a change in opioid prescribing patterns by race and ethnicity as a result of our QI initiative, this work may not be generalizable to other settings with different race and ethnicity and insurance status distributions.

In recent years, there has been greater national awareness of the risks of opioid prescribing. Opioid prescribing rates in children are decreasing, although vary by diagnosis, demographics, and region.23,24  We cannot assess the impact that this trend may have had on our local prescribing practices. Finally, our phone calls to patients who underwent orthopedic reduction in the ED were performed only as a balancing measure for this QI initiative. We did not formally study patient pain scores or administer parent satisfaction surveys in these patients as part of that evaluation.

Using QI methodology, we successfully decreased total opioid doses prescribed weekly from a pediatric ED by 91% within 10 months and sustained this change for an additional 9 months while minimizing return visits and reports of poor pain control. In the future, we plan to spread our key interventions to other areas of the hospital and other EDs in the region.

Thank you to Dr Cynthia Kuelbs, Dr Mario Bialostozky, Nicholas Fusco RPh, and Paul Boynay for their informatics support. Thank you to Deborah Hershberger MSN, RN for her mentorship in QI methodology. Thank you to Dr John Kanegaye and Dr Erin Fisher for reviewing this article and providing feedback. Thank you to Dr Kristy Schwartz and the RCHSD ED RAs for the use of the Rady ED call back program. Thank you to Dr Vidyadhar Upasani, Dr Paula Aristizabal, and the members of the RCHSD ED QAPI committee and RCHSD opioid task force for their valuable input with this initiative.

Dr Bryl conceptualized the study, designed the quality improvement (QI) interventions, supervised data collection, analyzed the data, and drafted and critically revised the manuscript; Dr Demartinis designed the QI interventions, led the data collection, and drafted the manuscript; Dr Etkin designed the QI interventions and critically revised the manuscript; Dr Hollenbach designed and supervised data collection and critically revised the manuscript; Drs Huang and Shah conceptualized the study, designed the QI interventions, and critically revised the manuscript; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

FUNDING: No external funding.

     
  • ED

    emergency department

  •  
  • EMR

    electronic medical record

  •  
  • LCL

    lower control limit

  •  
  • NH

    non-Hispanic

  •  
  • PEM

    pediatric emergency medicine

  •  
  • QI

    quality improvement

  •  
  • RCHSD

    Rady Children’s Hospital San Diego

  •  
  • UCL

    upper control limit

  •  
  • URI

    upper respiratory infection

1
Scholl
L
,
Seth
P
,
Kariisa
M
,
Wilson
N
,
Baldwin
G
.
Drug and opioid-involved overdose deaths - United States, 2013–2017
.
MMWR Morb Mortal Wkly Rep
.
2018
;
67
(
5152
):
1419
1427
2
Allen
JD
,
Casavant
MJ
,
Spiller
HA
,
Chounthirath
T
,
Hodges
NL
,
Smith
GA
.
Prescription opioid exposures among children and adolescents in the United States: 2000–2015
.
Pediatrics
.
2017
;
139
(
4
):
e20163382
3
Kane
JM
,
Colvin
JD
,
Bartlett
AH
,
Hall
M
.
Opioid-related critical care resource use in US children’s hospitals
.
Pediatrics
.
2018
;
141
(
4
):
e20173335
4
DePhillips
M
,
Watts
J
,
Lowry
J
,
Dowd
MD
.
Opioid prescribing practices in pediatric acute care settings
.
Pediatr Emerg Care
.
2019
;
35
(
1
):
16
21
5
Miech
R
,
Johnston
L
,
O’Malley
PM
,
Keyes
KM
,
Heard
K
.
Prescription opioids in adolescence and future opioid misuse
.
Pediatrics
.
2015
;
136
(
5
). Available at: www.pediatrics.org/cgi/content/full/136/5/e1169
6
McCabe
SE
,
West
BT
,
Teter
CJ
,
Boyd
CJ
.
Medical and nonmedical use of prescription opioids among high school seniors in the United States
.
Arch Pediatr Adolesc Med
.
2012
;
166
(
9
):
797
802
7
Sung
HE
,
Richter
L
,
Vaughan
R
,
Johnson
PB
,
Thom
B
.
Nonmedical use of prescription opioids among teenagers in the United States: trends and correlates
.
J Adolesc Health
.
2005
;
37
(
1
):
44
51
8
Monitto
CL
,
Hsu
A
,
Gao
S
, et al
.
Opioid prescribing for the treatment of acute pain in children on hospital discharge
.
Anesth Analg
.
2017
;
125
(
6
):
2113
2122
9
Huang
JS
,
Kuelbs
CL
.
Clinician opioid prescribing practices and patient utilization of prescribed opioids in pediatrics
.
J Opioid Manag
.
2018
;
14
(
5
):
309
316
10
Drendel
AL
,
Gorelick
MH
,
Weisman
SJ
,
Lyon
R
,
Brousseau
DC
,
Kim
MK
.
A randomized clinical trial of ibuprofen versus acetaminophen with codeine for acute pediatric arm fracture pain
.
Ann Emerg Med
.
2009
;
54
(
4
):
553
560
11
Friday
JH
,
Kanegaye
JT
,
McCaslin
I
,
Zheng
A
,
Harley
JR
.
Ibuprofen provides analgesia equivalent to acetaminophen-codeine in the treatment of acute pain in children with extremity injuries: a randomized clinical trial
.
Acad Emerg Med
.
2009
;
16
(
8
):
711
716
12
Poonai
N
,
Bhullar
G
,
Lin
K
, et al
.
Oral administration of morphine versus ibuprofen to manage postfracture pain in children: a randomized trial
.
CMAJ
.
2014
;
186
(
18
):
1358
1363
13
Motov
S
,
Strayer
R
,
Hayes
BD
, et al
.
The treatment of acute pain in the emergency department: a white paper position statement prepared for the American Academy of Emergency Medicine
.
J Emerg Med
.
2018
;
54
(
5
):
731
736
14
Dowell
D
,
Haegerich
TM
,
Chou
R
.
CDC guideline for prescribing opioids for chronic pain - United States, 2016
.
MMWR Recomm Rep
.
2016
;
65
(
1
):
1
49
15
Provost
LP
,
Murray
SK
.
The Health Care Data Guide: Learning From Data for Improvement
, 1st ed.
San Francisco, CA
:
Jossey-Bass
;
2011
16
Nagtegaal
R
,
Tummers
L
,
Noordegraaf
M
,
Bekkers
V
.
Nudging healthcare professionals toward evidence based medicine: a systematic scoping review
.
J Behav Public Adm
.
2019
;
2
(
2
):
1
20
17
Chua
WJ
,
Klein
EJ
,
Al-Haddad
BJS
,
Quan
L
.
Factors associated with opioid prescribing for distal upper extremity fractures at a pediatric emergency department [published online ahead of print August 20, 2019]
.
Pediatr Emerg Care
.
doi:10.1097/PEC.0000000000001908
18
Kline
TV
,
Savage
RL
,
Greenslade
JH
,
Lock
CL
,
Pattullo
C
,
Bell
AJ
.
Affecting emergency department oxycodone discharge prescribing: an educational intervention
.
Emerg Med Australas
.
2019
;
31
(
4
):
580
586
19
Ivers
N
,
Jamtvedt
G
,
Flottorp
S
, et al
.
Audit and feedback: effects on professional practice and healthcare outcomes
.
Cochrane Database Syst Rev
.
2012
;(
6
):
CD000259
20
Boyle
KL
,
Cary
C
,
Dizitzer
Y
,
Novack
V
,
Jagminas
L
,
Smulowitz
PB
.
Reduction of opioid prescribing through the sharing of individual physician opioid prescribing practices
.
Am J Emerg Med
.
2019
;
37
(
1
):
118
122
21
Drendel
AL
,
Brousseau
DC
,
Casper
TC
, et al
.
Opioid prescription patterns at emergency department discharge for children with fractures
.
Pain Med
.
2020
;
21
(
9
):
1947
1954
22
Goyal
MK
,
Johnson
TJ
,
Chamberlain
JM
, et al.;
Pediatric Emergency Care Applied Research Network (PECARN)
.
Racial and ethnic differences in emergency department pain management of children with fractures
.
Pediatrics
.
2020
;
145
(
5
):
e20193370
23
Tomaszewski
DM
,
Arbuckle
C
,
Yang
S
,
Linstead
E
.
Trends in opioid use in pediatric patients in US emergency departments from 2006 to 2015
.
JAMA Netw Open
.
2018
;
1
(
8
):
e186161
24
Pezalla
EJ
,
Rosen
D
,
Erensen
JG
,
Haddox
JD
,
Mayne
TJ
.
Secular trends in opioid prescribing in the USA
.
J Pain Res
.
2017
;
10
:
383
387

Competing Interests

POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.

FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.