Evidence for the Use of Opioid Medication for Pediatric Acute Pain in the Outpatient Setting: Technical Report

Opioids are often prescribed to children with acute pain; however, there are no general pediatric opioid prescribing guidelines specific to the outpatient setting. It is critical for clinicians to understand when opioids may be indicated to treat pain and how to safely prescribe them in a manner that minimizes the risk for opioid use disorder (OUD), poisoning, overdose, and other harms.

The American Academy of Pediatrics (AAP) has developed a clinical practice guideline (CPG) to guide the prescription of opioids to children in the outpatient setting.¹ This technical report summarizes the results of the literature review designed to support the accompanying CPG. The CPG was written by an AAP opioids guideline panel that included a chair, vice-chair, 2 methodologists, a range of pediatric primary and tertiary care providers, and pediatric experts in pain medicine, emergency medicine, toxicology, surgery, adolescent and addiction medicine, and implementation science. The opioids guideline panel included a parent representative and an informatics and implementation scientist.

The literature review was designed to answer 2 overarching population, intervention, comparison group, outcome, time frame (PICOT) questions. PICOT 1 focused on opioid prescribing to children who are experiencing acute pain in the outpatient setting, as compared with other pharmacologic and nonpharmacological treatments, or between opioids. The search assessed multiple outcomes, including: Short-term safety, short-term efficacy and effectiveness, persistent postoperative or postprocedural pain (>2 weeks later), unintentional ingestion, diversion, misuse, excess opioid medication, opioid-related overdose, and incident or recurrent OUD (within 1 year of prescription).

PICOT 2 focused on individual-, family-, or health systems-level interventions that seek to increase safe outpatient/ambulatory opioid prescribing for children, as compared with standard practice. The outcomes included were: Excess opioids after a prescription and risk of unintentional ingestion, diversion, misuse, opioid-related overdose, or addiction.

The opioids guideline panel engaged 2 health librarians to assist in searches of PubMed and Excerpta Medica Database (Embase). The search protocol was developed in the fall of 2021; the initial PubMed query was conducted in March 2022. An additional search was conducted in June 2023 to update the review with more recent publications. A subsequent Embase search was completed in July 2022 to identify additional studies published outside of the United States.

For the PubMed search, the following search strategy was used: (“Analgesics, opioid” [medical subject heading (MeSH) terms] or “analgesics opioid” [pharmacological action]) and “pain” (MeSH terms) and (“child*” [all fields] or “pediatr*” [all fields] or “paediatr*” [all fields]). The strategy used PubMed limitations of “clinical trial” or “randomized control trial (RCT)” or “systematic review (SR)” published since January 1, 2010.

For the Embase search, the following search strategy was used: (“Opiate”/exp or opiate) and (“analgesia”/exp or analgesia) and ([adolescent]/lim or [child]/lim or “pediatrics”/exp or pediatrics) and (RCT/de or SR/de or [Cochrane review]/lim) published since January 1, 2010.

The technical report only included RCTs and SRs on the basis of the AAP evidence grading system, which prioritizes these types of studies to make strong recommendations. In this system, there is little incremental value to the review of nonrandomized observational studies in assigning a strength to the resulting recommendations. Observational studies that were included in decision-making around the strength of evidence are summarized in the main CPG document.

All studies were required to include children up to 18 years of age. Studies could also include young adults up to 21 years of age, if this population was stratified from older adult participants and if children younger than 18 years were also included in the study. Studies had to focus on children with acute pain who were prescribed an opioid for use in the home setting.

Studies were excluded that:

• Assessed opioids that were only administered within the emergency department, urgent care, or inpatient settings (eg, intranasal fentanyl, nerve blocks);

• Did not measure an outcome in the home setting;

• Assessed opioids administered for management of neonatal opioid withdrawal syndrome; and

• Assessed opioids administered for acute episodes of pain related to chronic medical conditions (eg, cancer, sickle cell disease, palliative care).

The opioids guideline panel used Covidence, a program for online collaboration and management of systematic reviews, to manage the review process (https://www.covidence.org/). All abstracts were screened by the opioids guideline panel’s 2 methodologists. The methodologists assessed the study’s inclusion in the full-text review process. All conflicts were discussed and resolved. Articles excluded at this stage were assigned a reason for the exclusion.

The methodologists extracted the study data, and conflicts were discussed and resolved. The included RCTs were assigned a quality assessment using the Jadad scoring system² (see Table 1). The scope of the studies that were included covered a very small proportion of the PICOT questions.

The evidence tables based on the above PICOT questions and search results were presented at an in-person opioids guideline panel meeting held in November 2022. The resulting discussion identified additional clinically relevant questions that were not specifically addressed in the original literature review.

In response, the methodologists used the original literature PubMed search with an additional specification term (“narcotics/administration and dosage” [MeSH]). The resulting studies were rescreened and rereviewed to identify any additional, relevant studies.

The PubMed search identified 452 unique studies that met the inclusion criteria. The Embase search identified 236 additional studies, 34 of which were duplicate references with the initial PubMed search. After screening titles, 468 studies were excluded. Of the remaining 186 studies, 129 were excluded after abstract review. The most common reason for exclusion was a focus on the inpatient setting related to postsurgical inpatient use of opioids. An additional 19 publications were retained as background references but excluded from the qualitative analysis.

Between the 2 databases, the methodologists appraised 38 articles. After full text review, all 4 of the SRs were excluded because of the inclusion of children without separate results or because they were out of the scope of the PICOT questions. After full-text review, 11 RCTs (9 from PubMed and 2 from Embase) and 0 SRs were included in the final evidence table (see Fig 1).

The review identified 5 studies related to PICOT 1 (Supplemental Table 2). These included RCTs that compared either (1) opioids versus nonopioids for acute pain, (2) one opioid versus another for acute pain, or (3) scheduled versus as-needed dosing for acute pain.

Three studies compared the efficacy of morphine versus ibuprofen for acute pain.^3–5 Two of these studies focused on postsurgical pain (orthopedic or tonsillectomy); the third focused on pain from fracture.

Poonai used the Faces pain scale⁶ to assess the efficacy of morphine (0.5 mg/kg) to ibuprofen (10 mg/kg every 6 hours) after orthopedic surgery (2014) and orthopedic trauma (2017). In both studies, there was no difference in pain scores between the 2 groups. However, there were significant differences in adverse events (AEs) between the 2 treatment arms. After surgery, 69% of children in the morphine group experienced an AE as compared with 39% of children in the ibuprofen group. Similarly, more participants in the morphine group (56%) had an AE as compared with children in the ibuprofen group (31%) in the postfracture setting. Nausea (46% vs 19%), vomiting (18% vs 4%), drowsiness (48% vs 22%), and dizziness (31% vs 6%) were significantly more common in the morphine group. In the postfracture study, nausea (27% vs 6%) and vomiting (12% vs 3%) occurred more commonly in the morphine group.

Kelly et al compared the efficacy of morphine versus ibuprofen after tonsillectomy plus or minus adenoidectomy for sleep-disordered breathing.³ All children received acetaminophen every 4 hours and were randomized to receive additional analgesia with either morphine (0.2–0.5 mg/kg per dose every 4 hours) or ibuprofen (10 mg/kg per dose every 6 hours). The primary outcome in this study was the number of desaturations on the first postoperative day as compared with baseline, measured by a standardized home pulse oximeter. Pain and AEs were secondary outcomes. Pain was assessed on days 1 and 5 by parents using the validated Objective Pain Scale⁷ and by children using the Faces Pain Scale. The number of desaturation events per hour decreased in the ibuprofen arm (1.79 ± 7.57) and increased in the morphine arm (11.17 ± 15.02). There were no statistically significant differences between study groups in terms of pain or adverse drug reactions. Taken together, these 3 RCTs do not demonstrate an analgesic benefit of morphine versus ibuprofen in these postoperative or postfracture settings. However, in each study, there were significantly more side effects in the morphine arms, including nausea and vomiting (2 studies), drowsiness and dizziness (1 study), and hypoxia (1 study). Therefore, the balance of risks and benefits favors nonopioid therapy.

One RCT compared use of tramadol to codeine/acetaminophen for pain control after tonsillectomy.⁸ Children 4 to 15 years of age were randomly assigned to receive either liquid tramadol (1.05 mg/kg; maximum 52.5 mg) by mouth every 6 hours, plus the same dose every 3 hours, as needed, or codeine/acetaminophen (0.72 mg/kg codeine component) by mouth every 6 hours, plus the same dose every 3 hours as needed. The as-needed doses were limited to 3 per day. Children in each group received scheduled medication (as above) for the first 5 postoperative days. On days 6 to 10, dosing was every 4 hours as needed. Pain was measured by the Faces Pain Scale Revised⁷ (ages 4–7) or the Numerical Rating Scale (ages 11–15).⁹ Children 8 to 10 years of age were allowed to pick either scale. Patient- and parent-reported pain and side effects were recorded at home by parents using a standard diary for 10 days. Specific side effects that were queried included: Nausea, vomiting, fever, itching, rash, sweating, dizziness, headache, constipation, and oversedation.

Overall, 48 children were randomly assigned to each group, of which 38 children in the codeine group and 36 children in the tramadol group (total n = 74) completed the study. There were no differences in pain between the 2 treatment groups. AEs were common, with more than half of the participants reporting nausea, vomiting, and constipation. There were no statistical differences in the incidence of side effects between the 2 groups except for itching, which occurred in 33% of tramadol recipients but only 13% of codeine recipients.

Although this was a well-designed study, it is of little clinical relevance, because the US Food and Drug Administration has restricted use of both of these drugs in children.¹⁰ 

One other RCT assessed the efficacy of scheduled versus as-needed dosing for children receiving acetaminophen and hydrocodone for pain during the first 3 days after tonsillectomy.¹¹ All children received acetaminophen and hydrocodone solution (167 mg/2.5 mg/5 mL) and were randomly assigned to 3 dosing strategies: Group A: Dose every 4 hours as needed, with standard postoperative instructions; group B: Dose every 4 hours around-the-clock, with standard postoperative instructions; group C: Dose every 4 hours around-the-clock, with standard postoperative instructions and nurse coaching. Standard instructions were provided using a printed teaching booklet. The nurse coaching intervention (Group C) consisted of a review of the same materials, both at the initial study visit and on the day 1 and 2 follow-ups. Side effects assessed in this study included daytime sedation, lightheadedness, nightmares, nausea, vomiting, and constipation. Outcomes were assessed on the day of surgery and the subsequent 3 days.

The primary outcome was pain measured by the child at bedtime and upon awakening. Of note, there were no differences, specifically in the amount of analgesic received, between the 2 around-the-clock groups. For this reason, groups B and C were combined in the primary analysis. The authors concluded that pain relief was statistically superior in the around-the-clock group, although of unclear clinical significance.

The study also found no differences in the incidence of side effects between the around-the-clock and as-needed groups, with the exception of constipation, which occurred more frequently on each day after surgery in the around-the-clock group but not in the as-needed group.

Overall, 5 RCTS were identified that involved the use of opioids in children in the outpatient setting. The 3 studies that compared opioids versus no opioids (eg, use of nonsteroidal antiinflammatory drugs [NSAIDs]) support the use of NSAIDs as first-line therapy for postoperative pain on the basis of similar pain control and fewer side effects. The study comparing tramadol versus codeine/acetaminophen after tonsillectomy is no longer relevant to current clinical practice because of Food and Drug Administration recommendations recommending against the use of codeine and tramadol in this clinical context. The study comparing as-needed versus around-the-clock opioid administration suggests that scheduled analgesia may provide more consistent pain relief but with greater risk for constipation; notably, this study may also be less clinically relevant in light of professional society recommendations that opioids are not routinely needed after tonsillectomy.^8,12 

Within the same pediatric age groups, care settings, and time frame defined in PICOT 1, the second PICOT question focused on individual-, family-, or health systems-level interventions to increase safe outpatient and ambulatory opioid prescribing versus usual practice. Outcomes of interest were excess opioids after a prescription and risk of unintentional ingestion, diversion, misuse, opioid-related overdose, or addiction.

Six RCTs were identified related to PICOT 2 (Supplemental Table 3). All 6 included education as all or part of either the intervention or comparison group. These RCTs evaluated interventions in the ambulatory and short stay setting and included either:

1. parent, child, or adolescent educational interventions ranging from information sheets to more complex programs^13–16; or

2. provision of a drug disposal bag with education.^17,18 

A quantitative meta-analysis was not possible, however, because of the differences in study intervention, outcome, and target population.

Educational interventions for opioid prescription focus on outlining the benefits and risks of opioid medication. They aim to improve patient and family knowledge, perceptions, intentions, and behavior about opioid risks (eg, how to administer medication for maximum safety and effectiveness, how to recognize adverse effects, how to dispose of excess medication). The educational interventions tested in these 6 RCTs ranged in content and complexity, from a patient information sheet for oxycodone,¹³ to Scenario-Tailored Opioid Messaging Program (STOMP), an interactive Web-based program that “presents descriptive, clinically relevant pain and risk situations in which parents consider common scenarios and make intentional analgesic use and handling decisions.”^14–17 The sixth study used an education sheet and video about opioid use/storage and disposal as a comparator group (this study is further discussed below in the disposal method section).¹⁸ 

A range of primary outcomes was assessed, including parent satisfaction and knowledge¹³; proper¹⁸ and/or prompt disposal and intention to retain opioid medication¹⁷; more detailed parental knowledge and perceptions about AEs, analgesic self-efficacy (ie, confidence in administering analgesics)¹⁴; parents’ perceived risk of keeping or sharing opioids, intention to dispose, and final retention decisions¹⁵; and adolescents’ analgesic misuse perception.¹⁶ Three studies examined opioid use (eg, total doses, opioid days) as a secondary outcome.^14,16,18 Other secondary outcomes included pain,¹³ opioid storage location,¹⁸ scenario-based decision responses,¹⁴ and adolescent opioid misuse behavior and intentions.¹⁶ Outcomes were most commonly assessed up to 14 days after discharge, although several looked at outcomes at 1 to 3 months after initial prescription. Comparison groups were usual prescription education/controls,^13,14,16,17 opioid education with a link to website about disposal,¹⁵ disposal bag alone,¹⁸ and education with disposal bag.¹⁷ 

Across these studies with heterogeneous outcomes, 2 studies showed a benefit for opioid-specific education intervention versus a control group. A patient education sheet about oxycodone improved parent satisfaction and knowledge about opioid use at 7 days.¹³ The STOMP intervention decreased parents’ intentions to save unused opioid medications (5.6% vs 12% in parents who did not receive STOMP) and showed that parents’ higher perception of the risks of opioid decreased intentions to save unused opioids.¹⁷ 

Three studies by the same research group examined the interactive, Web-based STOMP intervention versus routine opioid education.^14–16 The STOMP intervention reduced the likelihood of parent retention (adjusted odds ratio 0.48; 95% confidence interval [CI] 0.25–0.93).¹⁵ There were small differences in parents’ awareness of AEs, parents’ perceived risk of addiction and child misuse, parental analgesic self-efficacy, adolescent risk perception, and parental decision-making in hypothetical scenarios.^14–16 There were no differences associated with the STOMP intervention on its own for the number of opioid doses, disposal intentions, or adolescent opioid misuse behavior or intention.

In summary, results of these 5 trials about educational interventions favored the intervention on a range of outcomes in the postambulatory surgical setting. Only 1 study examined an outcome specified in the PICOT question (eg, adolescent opioid misuse behavior). Only 2 outcomes were measured in more than 1 trial: Intention to dispose and/or retain. Other than for parent retention, the magnitude of any observed associations were small. It is important to note that, although the STOMP program is relatively complex and highly structured, results about its effect on actual behavior are mixed.

A disposal method and instructions for use were evaluated in 2 studies. Lawrence¹⁸ provided a bag with activated charcoal capsule, and Voepel-Lewis¹⁷ provided a bag with coffee grounds. Both studies compared the provision of the disposal method accompanied with routine postoperative discharge instructions on opioid use, storage, and disposal to routine education alone (written instructions with or without a video). Voepel-Lewis¹⁷ included 2 additional comparator groups (disposal method with the STOMP education intervention, and a STOMP intervention only group).

The main outcomes were proper¹⁸ or prompt¹⁷ disposal (ie, immediate disposal of leftovers after use), planned retention (intention to keep leftovers),¹⁷ and whether opioid prescriptions were filled or any opioids remained unused after the resolution of postoperative pain (secondary outcome measured at 7 and 14 days¹⁷ or 2 and 4 weeks¹⁸ after discharge). Providing a disposal method with routine education increased the proportion of patients who disposed of excess opioid postsurgery versus education alone: 71.7% versus 56.2%, a difference of 15.5% decrease (95% CI, 1.7–29.3) observed in Lawrence¹⁸ and a difference of 13.8% (33.0% vs 19.2%) observed by Voepel-Lewis.¹⁷ Prompt disposal behavior was highest (38.5%) for the group who received both the STOMP education and disposal method.

Compared with routine education only, providing a disposal method alone increased the odds of proper disposal (adjusted odds ratio 1.78; 95% CI 1.16–2.73) but did not independently affect planned retention.¹⁷ A lower percentage of families who received a disposal bag filled their child’s opioid prescription as compared with those who received only routine education, but the proportion of families with leftover opioids and appropriate opioid storage location did not differ by group.¹⁸ 

These 2 RCTs provide evidence that providing an opioid disposal method with education has a positive impact on proper or prompt disposal in the first weeks after surgery as compared with education alone in the postambulatory surgery setting.

These included studies are heterogeneous in their target group (family, child/adolescent) and measured outcomes. The variation in outcomes reflects the multiple time points in the postoperative decision-making period (ie, awareness, intention, behavior).

In terms of RCT quality, many of these studies were rated as lower, because the nature of the intervention was difficult to implement in a blinded fashion. The strength of the associations observed was not large, raising questions about the definition of a minimal clinical benefit. There are no results that examine any AEs. Similarly, many of the studies examined postsurgical settings, which may or may not translate to opioid prescribing in other outpatient settings. In summary, qualitative analyses of this body of literature suggest that family and patient education can be an avenue to decrease risk associated with opioid prescription. Additionally, providing disposal methods can increase opioid disposal and ultimately decrease excess opioid in the home.

More research needs to be conducted to define “education” and its effect on the outcomes of interest and to quantify the magnitude of the effect, as well as measure the effectiveness in a broad range of outpatient settings. Importantly, educational interventions are widely known to have a low level of reliability, and instrumental support (eg, providing a disposal method such as a mail-back bag for prescriptions rather than merely educating families to dispose of medications) is likely to result in more consistent effectiveness across study settings.^19,20 Last, a focus on the implementation of the use of disposal methods across settings will also yield additional evidence for approaches that can maximize the impact of effective interventions.

As noted, after the in-person panel meeting in November 2022, the methodologists performed a second review of the original literature search to focus on opioid dosing. This search identified 209 potential studies related to dosing, of which 189 were excluded based on screening of title and abstract.

Of the 20 remaining studies, 1 was already included in the evidence table, 5 were SRs that had been excluded, and 10 were excluded because of the setting, study design, or population. After discussion with the CPG opioids guideline panel leadership team, 4 were excluded because they were pharmacokinetic/pharmacodynamics research studies. After these exclusions, none of the 20 studies were included in the final literature review.

Despite an extensive literature search including 2 international databases, few RCTs were found related to the safety and efficacy of opioids in children with acute pain. In summary, studies that compared opioids versus no opioids (eg, NSAIDs) support the use of NSAIDs as first-line therapy for postoperative pain on the basis of similar pain control and fewer side effects.

Similarly, few studies were found that assessed individual-, family-, or health systems-level interventions to increase safe outpatient opioid prescribing. The review suggests that family and patient education can be an avenue to decrease risk associated with opioid prescription. Additionally, providing disposal methods can increase opioid disposal, and ultimately, may decrease excess opioid medication in the home. However, in addition to a heterogenous range of comparisons and outcomes, most of the included studies were relatively small, single-site studies that did not have masked outcome assessments, which limits the overall strength and scope of any resulting CPG recommendations.

Because the AAP evidence grading scheme places the greatest priority on RCTs and SRs, the methodologists focused on these studies with a higher level of evidence. However, the RCTs and SRs only covered a small portion of the PICOT questions. As a result, many of the CPG’s key action statements are based on expert opinion and observational studies. These observational studies are not summarized in this technical report.

The CPG was intended to be relevant to prescribers in the outpatient setting. However, most of the RCTs included in this review addressed pain related to surgery or trauma and did not examine opioid prescriptions that originated in an office-based outpatient setting. The primary conditions examined were tonsillectomy and fractures, so these results may not be generalizable to other acute painful conditions.

The PICOT questions aimed to examine the outcomes of interest up to 1 year after prescription. Most of the included studies were designed to assess short-term efficacy or effectiveness, and they did not all measure acute AEs. Therefore, these controlled studies could not assess the risk of overdose, OUD, and other long-term harms.

The scope of this review was intended to focus on opioid prescription, including nonopioid pharmacologic and nonpharmacologic interventions versus opioid medication. We are not able to address broader questions of acute pain management in children, including comparative effectiveness of nonpharmacologic treatments.

This review identified 6 research gaps that could be considered priorities toward building evidence to inform opioid prescribing:

1. High-quality clinical trials to quantify the ideal opioid dose and duration of opioid treatment;

2. High-quality clinical trials evaluating multimodal analgesia, including opioids, nonopioid medications, and nonpharmacologic approaches;

3. Evaluating opioid prescription in outpatient settings in pediatric subgroups, incorporating considerations of health equity, by race, ethnicity, preferred language, and insurance status;

4. Characterizing longitudinal research about age-specific and cumulative exposure over childhood and adolescence;

5. Assessing the impact of patient and family understanding, expectations, and preferences, as well as interventions to modify these factors; and

6. Leveraging system-level interventions for practice change and deimplementation.

To fill the dearth of RCT evidence relevant to this PICOT question, the first research priority for pediatric opioid prescribing in the outpatient setting is high-quality clinical trials that can guide decisions about appropriate types, dose, dosing schedules, and duration of opioid treatment in the outpatient setting to minimize both acute and long-term risks while still effectively treating acute pain.

Second, although this review synthesized the existing research comparing opioids and nonopioid alternatives, additional high-quality, masked RCTs that compare opioids alone and in combination to both pharmacologic and nonpharmacologic pain treatments are needed. These studies should, on clinical comparisons and outcomes that are relevant to real-world settings, examine the best combinations and timing of pharmacologic and nonpharmacologic treatments to minimize opioid use and use outcomes that are relevant to patients and families. As current literature is focused on a few conditions, particularly postprocedural pain, further detailed studies are needed to build evidence for setting- and condition-specific multimodal approaches. Third, the primary PICOT questions were intended to inform pain management in the outpatient setting. In this situation, patients and family members monitor pain symptoms at home, rather than clinicians doing so at the bedside. The RCTs reviewed were mostly in postprocedural settings or emergency departments; few studies examined specific age groups. Additionally, many studies used restrictive selection criteria, often excluding children with comorbidities such as disabilities or mental health conditions. Because some of these children and adolescents may experience more severe distress from pain and have the greatest need for support, there is a great need for studies that include these populations in outpatient settings. Finally, as it is known that there are differences in opiate prescribing on the basis of patients’ sociodemographic characteristics, it is critical that future studies consider health equity and examine prescribing by race, ethnicity, preferred language, and insurance status.

Fourth, since a large share of pediatric exposure to opioids is related to perioperative pain for common childhood procedures related to a single episode of pain, there is a need for research to understand the impact of repeated or cumulative exposure over the full span of childhood. To generate this type of evidence, studies need to include follow-up that extends beyond the short-term period (days and weeks) to years. Adolescence is a developmental period with both an elevated likelihood of being prescribed opioids compared with other age groups and an increased potential for emerging risk behaviors. Future research should examine the biological and psychological consequences of short-term, long-term, and repeated opioid exposure over critical child and adolescent developmental time periods. The results will both inform pediatricians and other pediatric health care providers and increase understanding of interventions that may minimize long-term risks such as opioid misuse and substance use.

Fifth, specific to PICOT 2, once an opioid has been prescribed, additional research is needed to understand how to increase safety outcomes by focusing on family, institutional, and community contexts. A critical research gap is about patient and parent understanding, expectations, and preferences about pain treatment with opioid medications and their effect on opioid safety and effectiveness. Ultimately, decisions about how to use and store opioid medication occur in the home setting and are made by the patients and families themselves. Therefore, research about the content, mode, mechanisms, and effectiveness of education on opioid-related knowledge and behavior is important.

Lastly, the sixth research gap centers on the impact of institutional and community policies, laws, and regulations on appropriate prescribing of opioid medication for children.

Current RCT evidence focused on education about medication use and safe disposal but should expand into both individual- and system-level interventions that promote safe storage, use, and disposal.

Evaluation of policies should also include unintended consequences, because prescribing policies have the potential to prevent the prescription of opioids for those with prolonged or chronic pain management needs. Additionally, implementation science research that focuses on institutional and community approaches to deimplement aspects of current practices is needed to accelerate the appropriate use of opioids.

Although RCT data are limited, evidence was not found that pain control by opioids is superior to nonopioid alternatives; further, opioids are often associated with AEs. Future studies can build on this foundation of evidence to support the appropriate use of opioids for acute pain in children.

Sudha R. Raman, PhD, Methodologist

Michael J. Smith, MD, MSCE, Methodologist

Kymika Okechukwu, MPA, Senior Manager, Evidence-Based Medicine Initiatives

Cordarrell Nunnery, CAPM, Project Manager, Evidence-Based Medicine Initiatives

Susan K. Flinn, MA, Medical Writer

We thank AAP staff members Kathy Clark, MLIS, MA, Director of Library and Archival Services, and Pia Daniels, MPH, PMP, Manager, Committees and Sections.

AAP

American Academy of Pediatrics

AE

adverse event

CPG

clinical practice guideline

CI

confidence interval

EMBASE

Excerpta Medica Database

MeSH

medical subject heading

NSAID

nonsteroidal antiinflammatory drug

OUD

opioid use disorder

PICOT

population, intervention, comparison group, outcome, time frame

RCT

randomized controlled trial

SR

systematic review

STOMP

Scenario-Tailored Opioid Messaging Program