Pediatric Antibiotic Prescribing Patterns From Emergency Department Visits

Abstract

OBJECTIVES

Antibiotics are frequently prescribed for pediatric emergency department (ED) patients having respiratory or genitourinary infections. Professional guidelines and antibiograms exist to help guide appropriate antibiotics for these infections, but it is unknown how well ED providers adhere to these recommendations.

METHODS

From January 1, 2017 to December 31, 2022 using a national electronic medical record database, we retrospectively reviewed de-identified pediatric ED patients who were discharged home with an antibiotic. Antibiotics and infections were grouped for analysis, and results were compared yearly.

RESULTS

We identified 85,026 pediatric ED patients with infectious diseases discharged home with 24,363 antibiotic prescriptions. Over the 6-year study period, penicillin-based antibiotics were prescribed more than macrolides or cephalosporins for respiratory tract infections, rising from 76.5% in 2017 to 81.3% in 2022 (p < 0.001). For genitourinary tract infections, cephalosporins were prescribed more than 5 other types of antibiotics, rising from 62.7% in 2017 to 73.9% in 2022 (p < 0.001).

CONCLUSIONS

Over this 6-year period, ED providers in our region were fairly adherent to and improved for guideline recommended respiratory tract infections and antibiogram guided treatment for genitourinary tract infection in pediatric patients.

Introduction

In the United States, there is an estimated 29 million emergency department (ED) visits by pediatric patients less than 18 years old.¹ Of these there are approximately 8.5 million ED visits for infectious causes² and 7 million of these patients are prescribed antibiotics yearly.¹ The majority of these prescriptions are for acute respiratory illness (including acute otitis media and acute pharyngitis). The second and third most common conditions—skin and soft tissue infections and urinary tract infections—are both far less frequent than acute respiratory illnesses.

Antibiotics are frequently overprescribed for children for unnecessary viral conditions. Estimates of this ranges between 2.5%³ to 23%⁴ to 33%¹ to 51%⁵ for various conditions or clinical environments. Macrolide antibiotics (for instance: azithromycin) are the most commonly overprescribed antibiotic class in children and are often used unnecessarily for acute otitis media or bronchitis/bronchiolitis. However, the American Academy of Pediatrics (AAP) does not recommend azithromycin for the treatment of acute otitis media because of the high incidence of Streptococcus pneumoniae resistance,⁶ or for acute bronchitis/bronchiolitis^7,8 because of the high likelihood the infection is viral.

To combat this overuse of antibiotics in children, national organizations such as the US Centers for Disease Control and Prevention (CDC) and specialty-specific organizations have published antibiotic stewardship initiatives^9,10 and clinical practice guidelines.^7,8,11,12 Our main objective for this project is to study the antibiotic prescribing pattern in our nationwide healthcare system and learn if ED providers are becoming more compliant with the clinical practice guidelines (Appendix 1).

Materials and Methods

To perform this study, we retrospectively accessed the Enterprise Data Warehouse (EDW) at HCA Healthcare, located in Nashville, TN. HCA Healthcare operates 186 hospitals and more than 2000 other sites of care (free standing EDs, surgery centers, urgent care centers and physician clinics) in the United States and the United Kingdom. The EDW stores all the electronic health data from every patient encounter at HCA Healthcare. For our project, we only included the HCA Healthcare facilities in our division (12 hospitals in 3 states, 100+ sites of care). The data extraction out of the EDW was performed with Safe Harbor de-identification techniques so that no protected health information would be abstracted out for analysis.

From the EDW, we selected all pediatric patients aged 3 months to less than 18 years old that were seen in an ED visit during the 6-year period from January 1, 2017 to December 31, 2022. We excluded any patient that was admitted to the hospital or transferred to another facility, yielding only a population of patients that were discharged home. From this population we collected all their diagnoses codes (ICD-10)¹³ and divided them into disease categories (see Appendix 2 for disease category definitions). We collected all of their discharge prescriptions and divided them into prescription categories (see Appendix 3 for prescription category definitions). In this manner we could estimate adherence or compliance with the clinical practice guidelines identified in Appendix 1.

All data preprocessing and analysis was performed using R Statistical Software (v.4.4.1, R Core Team 2024). Figures and tables were created with Excel. Descriptive statistics were used to evaluate patient demographics. Pearson χ² tests were done to determine significance between groups, using significance at or below the 0.05 level.

Results

We identified 85,026 pediatric patients discharged home with an infectious disease diagnosis from EDs in our division during this 6-year period. Their demographical information can be seen in the Table, along with their infectious disease category incidences. 22,655 of these patients (26.6%) received an antibiotic prescription upon discharge home. A total of 24,363 antibiotic prescriptions were discovered, showing that up to 1708 patients (2.0%) received more than 1 antibiotic prescription. Similarly, there were 86,405 infectious disease diagnoses codes discovered in these 85,026 unique patients. This suggests up to 1379 patients (1.6%) had multiple infections.

Table.

Subject Characteristics

Subject Characteristics
Number of subjects	85,026
Age (mean yr ± SD)	7.15 ± 5.67
Sex (no., %)
Female	44,747 (52.6%)
Male	40,279 (47.4%)
Race (no., %)
African American	37,092 (43.6%)
Caucasian	35,568 (41.8%)
Other	12,366 (14.5%)
Year (no., %)
2017	5374 (20.0%)
2018	5192 (19.3%)
2019	5441 (20.2%)
2020	2848 (10.6%)
2021	3462 (12.9%)
2022	4602 (17.1%)
Respiratory tract infections (no., %)	56,345 (66.3%)
Genitourinary infections (no., %)	9279 (10.9%)
Skin and soft tissue infections (no., %)	6210 (7.3%)
Viral infections (no., %)	14,571 (17.1%)

Next, we separately analyzed infectious categories. 56,345 patients were diagnosed with a respiratory tract infection (Table) and 17,041 antibiotic prescriptions were written. In this group, the most prescribed class of antibiotics were penicillin-based antibiotics. The frequency of these rose from 76.5% of respiratory tract infection in 2017 to 81.3% in 2022. Whereas, cephalosporin-based antibiotics went from 15.3% to 11.8%, and macrolide-based antibiotics went from 6.7% to 5.2% (χ² 91.886, df 30, p < 0.001). Line graphs of these 3 antibiotic classes for all 6 years are shown in Figure 1.

Figure 1.

Frequency of antibiotic prescriptions for ED respiratory tract infections in children.

Similarly, we then analyzed genitourinary infections. 9279 patients were diagnosed with a genitourinary infection (Table) and 6620 antibiotic prescriptions were written. In this group, the most prescribed class of antibiotics were cephalosporins. The frequency of these rose from 62.7% of genitourinary infections in 2017 to 73.9% in 2022. The next most common antibiotics was sulfamethoxazole-trimethoprim, and this declined from 26.1% to 10.7% in the same time period. Lastly, penicillin-based antibiotics rose from 6.9% to 7.8% (χ² 325.444, df 30, p < 0.001). Line graphs of these 3 antibiotic classes for all 6 years are shown in Figure 2.

Figure 2.

Frequency of antibiotic prescriptions for ED genitourinary infections in children.

Discussion

This study provides insight into the antibiotic prescribing patterns of ED providers (physicians, resident physicians, physician assistants, and/or nurse practitioners) in our hospital system for pediatric patients diagnosed with respiratory and genitourinary infections over a 6-year period (2017–2022). The data collected show clinical practice guideline-compliant prescribing regarding antibiotic choice within the HCA Healthcare EDs that were studied. These findings align with the goals of antibiotic stewardship initiatives in the current times of rising antibiotic resistance.

One of the key findings of this study is the decreasing rates of macrolide and cephalosporin prescribing with increased usage of penicillin for pediatric acute respiratory tract infections (ARTIs). This is particularly encouraging as it aligns with the AAP’s recommendation against macrolides, such as azithromycin, for the treatment of ARTIs.^6–8 Additionally, the decreasing rates of macrolide usage translates to a decline in broad spectrum antibiotic usage which was previously noted to be more common in the southern United States.⁴ A previous study showed that antibiotics were prescribed in greater than 20% of ambulatory pediatric visits with broad-spectrum antibiotics accounting for ∼50% of these. Furthermore, broad spectrum antibiotic prescribing was found to be most likely in respiratory conditions in which no antibiotic was indicated.⁴

An increased rate of cephalosporin usage for treatment of pediatric genitourinary infections was observed. This trend aligns with current practicing guidelines and AAP recommendations (Appendix 1)^7,14 to guide treatment based on local or regional antibiograms which show better sensitivities towards cephalosporins (see published South Carolina antibiograms¹⁵). Results of these data can be appropriately generalized to other facilities and practices throughout the southeastern United States if they have similar local antibiograms. However, further generalization to facilities in other regions would require knowledge of their antibiograms for uropathogens.

Results of this study shows pretty good adherence of ED providers within HCA Healthcare to clinical practice guidelines and does not support a need for change in practice at this time. These encouraging findings may represent increased awareness of antibiotic stewardship initiatives as these have been shown to positively impact antibiotic prescribing patterns.^16,17

Despite the positive findings, we do recognize limitations of our study which may require future investigation. First, our study is limited to a single healthcare system within a specific region of the United States. Although HCA Healthcare has a wide reach, these findings may not be fully generalizable to other systems or regions. Second, the retrospective design of our study limits the ability to assess the reasoning or potential confounding factors that may influence the providers choice of treatment or lack thereof. Related to the previous point, some patients may have received parenteral antibiotics while in the ED and discharged without further antibiotic coverage, falsely being assigned to the data set of pediatric patients who did not receive antibiotics. Additionally, many patients had multiple infections (86,405 infectious disease diagnosis codes in 85,026 unique patients), leading to potential contamination of data. Due to the design of the study, we were unable to distinguish for which infection the antibiotic was prescribed or if it was chosen to cover multiple infections. This is further complicated by the cases in which more than 1 antibiotic was prescribed (1708 patients). Finally, there are inherent limitations of data collection during the COVID-19 pandemic in 2020–2022 which may have influenced changes to the patient populations and the infectious presentations prompting their ED visits. Evidence of this shift can be seen in the significantly decreased incidence of many common pediatric illnesses, such as AOM,¹⁸ that has been attributed to social distancing practices and increased emphasis on good personal hygiene during the pandemic.

Further research is needed to examine the appropriateness of antibiotic treatment as our study focused on the appropriateness of the antibiotic class, not necessarily whether there was an indication for antimicrobial treatment. Additionally, further research evaluating how prescribing practices are impacted by antibiotic stewardship initiatives may provide insight into the best actions for further improving antibiotic prescribing practices.

Conclusions

This study demonstrates compliance with clinical practice guidelines regarding antibiotic prescribing for pediatric ARTIs and genitourinary infections within HCA Healthcare facilities in the southeast United States. Although the data do not demonstrate the need for a change in practice within HCA Healthcare facilities, the data does reflect the potential positive impacts of antibiotic stewardship initiatives. ED providers should continue this evidence-based practice of increasing first-line penicillin usage for ARTIs and local antibiogram-guided treatment of genitourinary infections. Continued investment in antibiotic stewardship initiatives, both at the institutional and national level, will be essential in further educating providers on appropriate antibiotic prescribing practices. This is especially important in pediatric populations where antibiotics are frequently overprescribed.¹ By reinforcing evidence-based guidelines and antibiotic stewardship initiatives, healthcare systems can support the responsible use of antibiotics and help preserve the effectiveness of these crucial medications for future generations.

ABBREVIATIONS

AAP

American Academy of Pediatrics;

AOM

acute otitis media;

ARTI

acute respiratory tract infection;

CDC

US Centers for Disease Control and Prevention;

ED

emergency department;

EDW

Enterprise Data Warehouse;

IRB

institutional review board;

ICD-10

International Classification of Diseases, Tenth Revision;

URI

upper respiratory tract infection