OBJECTIVES:Firearm injuries are a leading cause of pediatric mortality, but may not be adequately identified by International Classification of Diseases (ICD) codes. Natural language processing (NLP) could improve identification. We derived an NLP model from medical narratives and compared the sensitivity and specificity of ICD codes and the NLP models to identify pediatric acute firearm injury (AFI). METHODS:Retrospective study of ED visits by patients (aged 0 to 17 y) presenting to 7 pediatric emergency departments (ED) with a potential AFI between 2011 and 2019. A medical narrative review was performed to identify AFI narratives. These narratives, plus randomly selected narratives, underwent language feature extraction with logistic regression to identify topics significantly associated with the categories. Two Random Forest models were created: model 1 on all narratives associated with a potential AFI and model 2 using the confirmed AFI narratives. Model 2 was then applied to a new, unreviewed selection of random narratives. A second round of medical narrative review was performed on the random narratives, including those identified by model 2 as AFI. Sensitivity and specificity of ICD and NLP for the identification of AFI were calculated. RESULTS:A total of 2407 ED visits for potential AFI were identified. Of those, 1183 narratives were annotated by narrative review as AFI. In the randomly selected narratives (n=1150), 47 additional AFI were identified, 16 by NLP and 31 by narrative review, for a total of 1230 AFI. ICD codes correctly identified 1171 (95.2%) AFI, with a sensitivity of 95.2% (93.8% to 96.3%) and specificity of 93.5% (92.6% to 94.4%). NLP correctly identified 1041 (84.6%) AFIs in the final data set with a sensitivity of 84.6% (82.5% to 86.6%) and specificity of 94.7% (93.8% to 95.5%). CONCLUSIONS:ICD and NLP both had acceptable sensitivity and specificity for identifying pediatric AFI. Future directions include evaluating newer NLP models to address false positives and the identification of the intent of injury.
OBJECTIVES:This study identified demographic and experiential factors associated with caregiver anxiety in a pediatric emergency department (PED). METHODS:We conducted a cross-sectional survey of caregivers for children brought to our PED. Participant anxiety was assessed with the short-form Spielberger State-Trait Anxiety Inventory (STAI). State anxiety reflects how one feels at the moment; trait anxiety indicates how one generally feels. Each is scored using a 5-item Likert Scale. Primary outcomes were continuous state and trait scores. We also examined the impact of prior ED visits and caregiver characteristics on STAI scores. RESULTS:Five hundred two caregivers participated in the study and completed the survey. Caregivers with more children showed lower anxiety levels (mean state anxiety: 6.9 vs. 7.7, P = 0.03; mean trait anxiety: 6.5 vs. 7.2, P = 0.04), as did those with children who had prior PED visits (mean trait anxiety: 6.5 vs. 7.1, P = 0.02). Caregivers with a primary care provider (PCP) for their children reported higher state anxiety than those without a PCP (mean: 7.4 vs. 6.3, P = 0.04), although trait anxiety was not significantly different. Anxiety was not linked to length of stay, caregiver education level, or day of the week (all P >0.10). Only 2.6% (n = 13) of caregivers exceeded the short-form STAI anxiety thresholds. CONCLUSIONS:Our study showed that caregiver anxiety was more closely linked to caregiver experience, such as prior visits to the PED and the number of children, than to their child's illness severity. Triage acuity level and ED length of stay were not associated with anxiety levels.
OBJECTIVES:Determining concentrations of injury can guide responses to priority areas; understanding the intersection of pediatric and adult injury informs community-based interventions. We aimed to compare geospatial patterns of injury in Milwaukee County for children and adults. METHODS:Trauma registry data from Milwaukee County's level I pediatric and adult trauma centers were queried from 2016 to 2022. Injury addresses were geocoded for children (n = 1979) and adults (n = 10,607) who resided in and sustained injuries in Milwaukee County by fall, firearm, and motor vehicle crash mechanisms. Hotspots were calculated by mechanism within each population, then compared. Temporal patterns were analyzed. RESULTS:Children and adults had similar locations for firearm injuries (global similarity value = 0.67). Motor vehicle crashes were less concordant (0.598), and falls were disparate (0.163). Child and adult firearm hotspots coincided with motor vehicle crash hotspots. Fall injuries increased in the pediatric (r2 = 0.64, P = 0.02) and adult registries (r2 = 0.80, P = 0.006) over time. Likewise, firearm injuries increased in children (r2 = 0.68, P = 0.02) and adults (r2 = 0.58, P = 0.04). Motor vehicle crashes in adults increased (r2 = 0.57, P = 0.04), but not in children (r2 = 0.18, P = 0.33). CONCLUSIONS:Consistent injury hotspots exist for firearm-injured children and adults, with overlap in motor vehicle crash and firearm injury hotspots. There is a need for public health intervention to foster safer communities. This is the first geospatial comparison of injury across pediatric and adult populations in Milwaukee County.
OBJECTIVES:To evaluate the pediatric femoral nerve block (FNB) utilization for the state of Western Australia, with comparison of a pediatric emergency department (PED) versus general ED settings. To report on contemporary PED FNB procedure and complications for all evaluated cases of FNB. METHODS:A retrospective observational study of cases of femoral fractures presenting directly to or transferred to a PED, from January 1 to December 31, 2022. Variables analyzed included patient demographics, PED episode features, and FNB in transferred patients. Procedural variables included ultrasound (US) guidance, anesthetic agent, dose (mg/kg), pain scales, and adherence to documentation. All cases of FNB were evaluated for complications within 48 hours of admission. RESULTS:In all, 83 cases of femoral fractures were evaluated; 24 (80%) patients received FNB initially in the PED compared with 24 (45.28%) in a general ED before transfer, an odds ratio of 0.21 (95% CI: 0.07, 0.59), P=0.003 for FNB in a general ED compared with a PED was demonstrated. A total of 47 PED FNB procedures were evaluated; median age 3.9 years, median weight 17.0 kg, 35 (74.5%) with US-guidance, ropivacaine was used in all recorded cases, median dose of 1.8 mg/kg, all captured post-procedure pain scales of zero, and documentation adherence was 44.7%. No significant complications were detected. CONCLUSION:There was high adoption of FNB in a PED but poor utilization outside this setting, which cannot be explained by access to a FNB guideline, staffing profile, training or skill mix. Features of the PED FNB procedure included: US-guidance, EM proceduralists, conservative dosing of ropivacaine, and conferring a safe and effective approach. Barriers to practice in general ED settings require addressing, including evidence translation, training, timing, and confidence.
OBJECTIVES:Emergency department (ED) boarding for youth in mental health crisis has increased nationally, straining capacity and delaying definitive care. Telepsychiatry may reduce ED length of stay (LOS), but limited data exist to describe factors associated with ED LOS for pediatric populations receiving telepsychiatry consultation. METHODS:This retrospective study analyzed telepsychiatry consultations on patients aged below 18 years across 11 hospitals in an academic health system (July 2022 to July 2024). Demographic, clinical, and encounter-level variables were extracted from the electronic health record. ED LOS was defined as the time from ED arrival to ED discharge. Linear regression was used to evaluate independent predictors of ED LOS. A subanalysis compared the 50 longest LOS encounters with 50 controls randomly selected from the lowest 2 quartiles. RESULTS:Among 1187 consultations, median ED LOS was 34.9 hours (IQR=14.8 to 68.5). Older age (P=0.003) and public insurance (P=0.01) were associated with longer LOS, while female sex was associated with shorter LOS (P=0.007). Regression analyses identified diagnosis category as the strongest independent predictor of LOS. Extreme outlier cases (median LOS=666.5 h) differed significantly from controls in age, diagnosis, disposition, violence history, legal barriers, and antipsychotic use. CONCLUSIONS:Pediatric patients evaluated by telepsychiatry experienced prolonged ED LOS, with demographic, diagnostic, and social factors contributing to delays. Extreme outlier cases highlight the multifactorial drivers of extreme boarding, including aggression, legal constraints, and custodial issues. Early identification of high-risk patients may support targeted intensive case management and interagency coordination to reduce ED LOS.
OBJECTIVE:Pediatric drowning remains a major global public health crisis, requiring rapid and precise risk stratification upon hospital presentation. While clinical signs are critical, objective biological markers are needed to optimize triage. This study aimed to assess the prognostic value of the Szpilman classification, blood lactate levels, and systemic inflammatory indices, including the systemic immune-inflammation index (SII), neutrophil to lymphocyte ratio (NLR), and platelet-to-lymphocyte ratio (PLR) in predicting mortality among pediatric drowning victims. MATERIAL AND METHODS:A retrospective analysis was conducted on pediatric patients who presented to a tertiary pediatric emergency department due to drowning between 2020 and 2024. A receiver operating characteristic (ROC) analysis was performed to evaluate the ability of inflammatory indices, the Szpilman classification and blood lactate levels to predict mortality. RESULTS:A total of 91 pediatric patients (78 nonfatal and 13 fatal) were included in the final analysis. The mean age of the patients was 9.6±5.0 years, with a male predominance (72.5%). All evaluated clinical and laboratory diagnostic models demonstrated high statistical significance in predicting fatal outcomes (P<0.001). On the basis of ROC curve analysis, the calculated prognostic parameters were the Szpilman classification (area under the curve (AUC): 0.978), PLR (AUC: 0.957), lactate levels (AUC: 0.935), SII (AUC: 0.843), and NLR (AUC: 0.817). Crucially, while these indices demonstrated significant predictive capacity, fatal outcomes were explicitly linked to lower baseline values of the PLR, SII, and NLR. CONCLUSION:In our study, unlike typical trauma patterns, fatal outcomes in paediatric drowning cases were found to be significantly associated with lower baseline levels of systemic inflammation markers such as SII, NLR, and PLR. This may be related to the extreme acute metabolic and immunologic collapse characteristic of severe drowning cases. We believe that low inflammatory index values should be interpreted as an indicator of a poor prognosis.
INTRODUCTION:Nerium oleander is well known for its toxicity due to the presence of cardiotoxic glycosides. This shrub is often involved in unintentional ingestion by children. To improve knowledge and assist health care professionals, we conducted a 12-year retrospective study in France on unintentional exposure to N. oleander in children aged 0 to 12 years. METHODS:Data were extracted from the National Poisoning Case Database. The distribution of children according to the circumstances and the route of exposure, in addition to the type of symptoms and the clinical outcome, were specifically analysed. RESULTS:In total, 2748 cases were included. Most accidents occurred in the home. Of these, 224 (8.15%) children developed symptoms. Oral and buccal exposures were mainly associated with digestive disorders and oropharyngeal irritation, all of minor severity. The irritant effects of N. oleander were also observed following ocular and cutaneous exposure. CONCLUSION:Nerium oleander poisoning is relatively common in France. Our study showed that in unintentional exposures, symptoms were generally of minor severity. Nevertheless, thorough decontamination of the contact area is essential because of the plant's irritant properties.
Introduction: Nerium oleander is well known for its toxicity due to the presence of cardiotoxic glycosides. This shrub is often involved in unintentional ingestion by children. To improve knowledge and assist health care professionals, we conducted a 12-year retrospective study in France on unintentional exposure to N. oleander in children aged 0 to 12 years. Methods: Data were extracted from the National Poisoning Case Database. The distribution of children according to the circumstances and the route of exposure, in addition to the type of symptoms and the clinical outcome, were specifically analysed. Results: In total, 2748 cases were included. Most accidents occurred in the home. Of these, 224 (8.15%) children developed symptoms. Oral and buccal exposures were mainly associated with digestive disorders and oropharyngeal irritation, all of minor severity. The irritant effects of N. oleander were also observed following ocular and cutaneous exposure. Conclusion: Nerium oleander poisoning is relatively common in France. Our study showed that in unintentional exposures, symptoms were generally of minor severity. Nevertheless, thorough decontamination of the contact area is essential because of the plant’s irritant properties.
BACKGROUND:Acute otitis media (AOM) is among the most common pediatric infections and a leading indication for antibiotic prescribing. Although current guidelines recommend 5 to 7 days of therapy for children 2 to 5 years of age with uncomplicated AOM, variation in antibiotic duration remains common. We evaluated whether a 5-day antibiotic course was associated with similar short-term outcomes compared with a 7-day course. METHODS:We conducted a retrospective multicenter cohort study using electronic health record data from 44 pediatric emergency departments, urgent care centers, and primary care practices within an integrated health care system. Children 2 to 5 years of age with uncomplicated AOM treated with a single oral antibiotic during 2022 were included. The primary analysis compared children prescribed 5 versus 7 days of therapy. Treatment failure was defined as an AOM-related return encounter requiring a new systemic antibiotic prescription within 7 days after completion of therapy (EOT+7). Secondary analyses evaluated outcomes by antibiotic class and prespecified sensitivity analyses. RESULTS:The final analytic cohort included 7528 children; 906 received 5-day therapy, and 1799 received 7-day therapy. Treatment failure occurred in 0.44% and 0.61%, respectively (absolute risk difference: -0.17%; 95% CI: -0.73 to 0.39; risk ratio: 0.72; 95% CI: 0.23-2.26). Findings were consistent across antibiotic-specific and sensitivity analyses. CONCLUSIONS:Among children 2 to 5 years of age with uncomplicated AOM, a 5-day antibiotic course was not associated with an increased risk of short-term treatment failure compared with a 7-day course, supporting guideline-concordant prescribing and outpatient antimicrobial stewardship.
Background: Acute otitis media (AOM) is among the most common pediatric infections and a leading indication for antibiotic prescribing. Although current guidelines recommend 5 to 7 days of therapy for children 2 to 5 years of age with uncomplicated AOM, variation in antibiotic duration remains common. We evaluated whether a 5-day antibiotic course was associated with similar short-term outcomes compared with a 7-day course. Methods: We conducted a retrospective multicenter cohort study using electronic health record data from 44 pediatric emergency departments, urgent care centers, and primary care practices within an integrated health care system. Children 2 to 5 years of age with uncomplicated AOM treated with a single oral antibiotic during 2022 were included. The primary analysis compared children prescribed 5 versus 7 days of therapy. Treatment failure was defined as an AOM-related return encounter requiring a new systemic antibiotic prescription within 7 days after completion of therapy (EOT+7). Secondary analyses evaluated outcomes by antibiotic class and prespecified sensitivity analyses. Results: The final analytic cohort included 7528 children; 906 received 5-day therapy, and 1799 received 7-day therapy. Treatment failure occurred in 0.44% and 0.61%, respectively (absolute risk difference: −0.17%; 95% CI: −0.73 to 0.39; risk ratio: 0.72; 95% CI: 0.23-2.26). Findings were consistent across antibiotic-specific and sensitivity analyses. Conclusions: Among children 2 to 5 years of age with uncomplicated AOM, a 5-day antibiotic course was not associated with an increased risk of short-term treatment failure compared with a 7-day course, supporting guideline-concordant prescribing and outpatient antimicrobial stewardship.
BACKGROUND:Postpartum depression (PPD) is a common complication of childbirth with significant implications for maternal and child health. Standard screening practices may miss mothers with later-onset symptoms. The pediatric emergency department (PED) may provide an opportunity to extend PPD screening beyond routine outpatient care. OBJECTIVES:To evaluate the feasibility and utilization of Edinburgh Postnatal Depression Scale (EPDS) screening for PPD among mothers of children 12 months or younger presenting to a PED and to describe the demographics of screened patients. METHODS:We conducted a retrospective descriptive study of EPDS screening implemented as a quality improvement initiative in an urban academic PED between May 15, 2023 and November 30, 2023. Biological mothers of children 12 months or younger were screened at triage using the EPDS in English or Spanish. Positive screens were defined as a score ≥9 or endorsement of suicidal ideation. Demographic data, Emergency Severity Index (ESI), insurance status, EPDS results, and documented interventions were abstracted from the electronic medical record. RESULTS:A total of 734 EPDS screens were completed. Forty-one mothers (5.6%) screened at risk or positive. Of these, 17 (41%) had infants aged 6 to 12 months. Maternal mental health resources were offered to 37 (90.2%) mothers with positive screens. All mothers with positive screens had public insurance. The screened population was predominantly white (41.1%) and Hispanic (41.1%). CONCLUSIONS:PPD screening in the PED for mothers of children up to 12 months of age is feasible and may identify at-risk mothers. PED-based screening may help address gaps in maternal mental health care for underserved populations.
Objective: To assess the feasibility and parental acceptability of delayed consent for research involving blood samples collected from febrile infants aged 60 days or younger during emergent care. Methods: This was a substudy of a pilot trial at a tertiary pediatric ED to detect biomarkers of serious bacterial infections among febrile young infants. Families were provided an introductory letter indicating that, after clinically necessary blood tests, a small volume of additional blood would be collected and stored for research for which they could opt out at any time. After hospital discharge, all families received telephone follow-up for a delayed consent discussion, allowing them to accept or refuse participation in the main biomarker trial. A 6-month consecutive sample of parents was invited to complete a questionnaire regarding the acceptability of research blood collection with delayed telephone consent. Results: Among 345 consecutive infants, 22 (6.4%) families opted out of research blood collection during ED care, and 136 (42.1%) had research blood collected after clinical sampling. Nearly all families (333/345; 96.5%) had telephone follow-up, and questionnaires were completed by 312/333 (90.4%). Most families found delayed consent for research blood collection acceptable (271/312; 86.9%) or somewhat acceptable (34/312; 10.9%), and 81.5% provided consent to use their infants’ research blood samples. Conclusions: Delayed consent for research is feasible andhighly acceptable to parents of febrile infants in the ED. This methodology may minimize stress during emergent care, enhance inclusivity, and facilitate timely, ethical pediatric research when there are barriers to standard prospective consent. Broader implementation should consider study risk, context, and ethical oversight.
Objectives: Highly toxic medications are thought to lead to fatal overdose even after single-dose ingestions. Recent evidence has questioned some of the assumptions underlying this concept. In this context, this study aimed to evaluate the incidence, management, and outcomes of accidental ingestions involving highly toxic drugs in young children presenting to pediatric emergency departments (EDs). Methods: We performed a descriptive and analytical study using a prospective, nationwide, multicenter pediatric poisoning registry including 58 Spanish EDs from 2008 to 2024. We analyzed accidental ingestions of therapeutic drugs in children under 8 years of age with a predefined subgroup of infants under 1 year. Highly toxic drugs were defined according to established pediatric toxicology criteria. Clinical presentation, ED management, hospital disposition, and outcomes were compared with other therapeutic drug ingestions. Results: Among 3577 poisoning episodes, 1189 (33.2%) involved accidental ingestion of therapeutic drugs in children under 8 years; 61 cases (5.1%) involved highly toxic drugs. Median age was 3 years (interquartile range 1 to 6), and 26 children (46.6%) were under 1 year. Cardiovascular drugs (37.7%) and opioids (24.5%) were the most frequent agents. Only 4 patients (6.8%) were symptomatic at presentation. No child required advanced airway or circulatory support, and no fatalities occurred. Hospital admission was more frequent after ingestion of highly toxic drugs. Conclusions: Accidental ingestion of highly toxic medications represents a small proportion of pediatric poisonings in EDs and is generally associated with favorable short-term outcomes. This supports a more evidence-based approach to risk stratification in the management of exploratory ingestions.
Background: Drug-induced acute dystonia is an adverse drug reaction that is concerning in pediatric patients but resolves rapidly with appropriate treatment. In children, data on risky drugs, clinical patterns, and the management of dystonia in the emergency department are limited. This study aimed to evaluate the demographic and clinical findings and treatment outcomes of children presenting to the pediatric emergency department with drug-induced acute dystonia. Methods: This retrospective observational study includes children aged 1 month to 18 years who were diagnosed with acute dystonia in a tertiary pediatric emergency department between October 2022 and March 2025. The diagnosis was made by a pediatric emergency subspecialist based on clinical findings. Patients were classified according to clinical phenotype as focal/segmental dystonia (group I) and multifocal/generalized dystonia (group II). Demographic data, drug exposures, clinical characteristics, and treatment responses were analyzed. Results: A total of 79 patients were included in the study. The median age was 11 years (IQR: 7 to 16) in group I and 10 years (IQR: 6 to 16) in group II. The most commonly associated drug groups were antipsychotics (55.6%), antiemetics (26.6%), and psychostimulants (20.3%). Focal dystonia is the most common clinical pattern, affecting the head and neck muscles in 61% of cases. The use of metoclopramide was significantly higher in group I (OR: 0.21; 95% CI: 0.04-0.99). All patients were treated with parenteral biperiden. Conclusions: Antipsychotics and antiemetics are the main triggers of drug-induced acute dystonia in children. Dystonia usually appears within the first 72 hours after starting the drug. It can develop even at therapeutic doses. Dystonias associated with antiemetic drugs often show focal or segmental distribution. Parenteral biperiden is a fast and effective treatment option. Acute dystonia can mimic serious etiologies in the emergency department. Obtaining a detailed drug history can facilitate the diagnostic process.
A 22-month-old male with profound microcytic anemia presented with a limp, refusal to walk, and several days of vomiting. Dietary history and focused examination prompted further workup that revealed two rare but related diagnoses that may be easily missed in the emergency department but are treatable with early identification.
Background: There is limited evidence on the ability of point-of-care ultrasound (POCUS) to quantify and characterize pediatric parapneumonic effusions. This study aimed to evaluate the agreement between lung POCUS and radiology ultrasound (RADUS) or chest x-ray (CXR) in determining effusion size and complexity in children presenting to the pediatric emergency department (ED). Methods: We studied patients aged younger than 18 years diagnosed with parapneumonic effusions at the Hospital for Sick Children between June 1, 2018, and December 1, 2024. Included patients had undergone a lung POCUS demonstrating a pleural effusion and a RADUS or CXR within 36 hours of their POCUS examination. Patients with suboptimal POCUS images were excluded. Using previously described criteria, we determined whether an effusion was “drainable” or “non-drainable” using the number of involved intercostal spaces as an estimate of effusion size. Our primary outcome was the rate of agreement between the POCUS estimation of effusion size to that of RADUS (or CXR if RADUS was unavailable) using Cohen Kappa statistic. We also compared the rate of agreement between POCUS and RADUS for effusion complexity and the time taken to complete both imaging modalities. Results: Eighty-five cases were included in this study. POCUS agreed with RADUS substantially in its ability to quantify (κ=0.76) and characterize (κ=0.65) parapneumonic effusions. There was also substantial agreement between POCUS and CXR for estimating effusion size (κ = 0.71). The median time to complete POCUS was significantly shorter compared with RADUS (6 vs. 14 min; P =0.003). Conclusions: Using a simple method of counting involved intercostal spaces, our results support the use of lung POCUS for quantifying and characterizing parapneumonic effusions in the pediatric ED. Future research is needed to evaluate the generalizability of these findings and their impact on clinical outcomes.
Background: Each year in the US, approximately 600,000 (8.1 per 1000 child population) children are victims of maltreatment. In 2023, rates among Black/African American children were 47% higher than the overall national rate. It is unclear whether similar disparities exist in national emergency department (ED) data, as the Nationwide Emergency Department Sample (NEDS) just began including race and ethnicity in data in 2019. Objective: To evaluate racial disparity and disproportionality among a nationally representative sample of ED visits involving child maltreatment concerns. Methods: International Classification of Disease, 10th Revision, Clinical Modification (ICD-10-CM) codes were used to identify ED visits for confirmed (T74) and suspected (T76) child maltreatment concerns in patients under 18 years old from 1/1/2019 to 12/31/2020. NEDS sampling weights were applied to produce national estimates. Racial disparity and disproportionality were calculated using the Disparity Index (DI) and the Disproportionality Representation Index (DRI), respectively. Results: A total of 29,667 ED visits (130,981 weighted) involving child maltreatment concerns were identified. The DRI revealed that non-Hispanic Black children had maltreatment codes applied at a rate that was 1.8 greater times what was expected given their proportion in the general population. The DI showed that non-Hispanic Black children were twice as likely to have a maltreatment-related diagnosis code applied when compared with their non-Hispanic White counterparts. Conclusions: This research exposes racial differences in the application of diagnosis codes for child maltreatment concerns in a nationally representative ED sample. Future research should explore factors which may influence the application of maltreatment-related diagnosis codes.
The Pediatric Emergency Medicine Collaborative Research Committee (PEM CRC) is the research subcommittee of the American Academy of Pediatrics (AAP) Section on Emergency Medicine. Volunteer-led, the committee has a long history of supporting and facilitating high-impact multicenter pediatric emergency medicine research as well as nurturing future pediatric emergency medicine researchers. Currently, the 76 contributing sites across the United States, Canada, and Spain provide an avenue to investigate topics related to the emergency care of children with adequately powered, generalizable studies. We describe the history of PEM CRC, the impact on the field of pediatric emergency medicine, and future goals.
Objective: To identify factors associated with the utilization of procedural sedation during pediatric facial laceration repair. Methods: We performed a retrospective cross-sectional study at a tertiary care pediatric ED from 2016 to 2020 of patients aged 0 to 18 with facial lacerations repaired with sutures. Our primary outcome was utilization of procedural sedation (ketamine, propofol alone or plus fentanyl or ketamine, midazolam plus fentanyl), with a secondary outcome of utilization of anxiolysis (intranasal midazolam). We measured the association of covariates, including provider subspecialty and patient age, sex, race and ethnicity, insurance status, time and day, laceration characteristics, mechanism, and layers of repair in a multivariate analysis and multinomial logistic regression. Results: We identified 4943 patient visits, with 3859 (78.8%) laceration repairs performed by emergency medicine providers and 1084 (21.9%) performed by plastic surgery. Procedural sedation was used in 373 (7.6%) visits. Lacerations repaired by plastic surgery were 25 times more likely to involve procedural sedation compared with those repaired by emergency medicine providers [adjusted odds ratio (aOR): 25.0, 95% CI: 17.1-36.5]. Additional factors included: laceration length (≥5 vs. <2.5 cm, aOR: 3.87, 95% CI: 1.88, 7.94), laceration complexity (complex vs. simple aOR: 2.50, 95% CI: 1.75, 3.57), and time of presentation (overnight vs. day, aOR: 2.58, 95% CI: 1.36, 4.91). Conclusion: Provider specialty was the strongest independent predictor of procedural sedation use, even after adjustment for wound characteristics. These findings highlight substantial practice variation and suggest an opportunity to develop standardized, clinically driven approaches to sedation decision-making for pediatric facial lacerations.
