OBJECTIVES:

To compare the frequency of opioid and corticosteroid prescriptions dispensed for children with pneumonia or sinusitis visits on the basis of location of care.

METHODS:

We evaluated 2016 South Carolina Medicaid claims data for 5 to 18 years olds with pneumonia or sinusitis. Visits were associated with 1 of 3 locations: the emergency department (ED), urgent care, or the ambulatory setting.

RESULTS:

Inclusion criteria were met by 31 838 children. Pneumonia visits were more often linked to an opioid prescription in the ED (34 of 542 [6.3%]) than in ambulatory settings (24 of 1590 [1.5%]; P ≤ .0001) and were more frequently linked to a steroid prescription in the ED (106 of 542 [19.6%]) than in ambulatory settings (196 of 1590 [12.3%]; P ≤ .0001). Sinusitis visits were more often linked to an opioid prescription in the ED (202 of 2705 [7.5%]) than in ambulatory settings (568 of 26 866 [2.1%]; P ≤ .0001) and were more frequently linked to a steroid prescription in the ED (510 of 2705 [18.9%]) than in ambulatory settings (1922 of 26 866 [7.2%]; P ≤ .0001). In logistic regression for children with pneumonia, the ED setting was associated with increased odds of receiving an opioid (adjusted odds ratio [aOR] 4.69) or steroid (aOR 1.67). Similarly, patients with sinusitis were more likely to be prescribed opioids (aOR 4.02) or steroids (aOR 3.05) in the ED than in ambulatory sites.

CONCLUSIONS:

School-aged children received opioid and steroid prescriptions for pneumonia or sinusitis at a higher frequency in the ED versus the ambulatory setting.

What’s Known on This Subject:

Opioids and systemic corticosteroids are not indicated for pediatric bacterial pneumonia or sinusitis on the basis of pediatric guidelines, policy statements, and research revealing limited efficacy of either agent in these selected conditions.

What This Study Adds:

Despite recent guidelines for pediatric pneumonia and sinusitis, a clinically significant percentage of healthy children received opioids and systemic corticosteroids, and they were prescribed and dispensed at higher frequencies after visits to an emergency department than after an ambulatory visit.

There is a lack of evidence supporting the use of opioids and systemic corticosteroids to control symptoms of pediatric respiratory tract infections in the outpatient setting.15  Opioids reveal no benefit over placebo in managing acute cough.2  Studies supporting adjunctive corticosteroid treatment in pediatric pneumonia are limited to the inpatient setting and to patients with asthma.35  There are no trials in which systemic corticosteroids for bacterial sinusitis in children are evaluated. Medication errors involving opioids are common and may lead to significant adverse events and patient harm.69  Opioids can cause sedation, respiratory depression, and respiratory failure and can contribute to substance use disorder and addiction.1012  Although the majority of adverse effects of corticosteroids is associated with chronic use, short-term administration can cause behavioral effects, insomnia, hypertension, and hyperglycemia in children.13  Opioids and systemic corticosteroids are not indicated for pediatric bacterial pneumonia or sinusitis on the basis of guidelines, policy statements, and research.1418 

In a previous study of state-level Medicaid data, respiratory diagnoses composed 9% of the indications for opioid prescriptions in infants <6 months old.19  To our knowledge, there are no previous studies on the prescribing of opioids or steroids for children with respiratory conditions based on location of care. Our purpose for this study was to compare the frequency of opioid and systemic steroid prescribing and dispensing for children with pneumonia or sinusitis on the basis of the location of care. Our hypothesis was that the frequency of dispensed prescriptions of these 2 drug classes for pediatric pneumonia or sinusitis was greater in the emergency department (ED) and urgent care (UC) settings than in other ambulatory locations.

We used paid South Carolina Medicaid pharmacy and visit claims from calendar year 2016 (January 1, 2016 to December 31, 2016). We identified school-aged children (5–18 years old at the time of the index visit) with a primary diagnosis of pneumonia or sinusitis using International Classification of Diseases, 10th Revision (ICD-10) Clinical Classifications Software (CCS) provided by the Agency for Healthcare Research and Quality.20  Only the primary diagnosis of pneumonia or sinusitis during the first visit (or index visit) in calendar year 2016 for each patient was included. The following single-level CCS diagnosis categories were used: 122 (pneumonia) and 133 (other lower respiratory disease) for those that included the word pneumonia and 126 (other upper respiratory infections) for those that included acute or chronic sinusitis (see Supplemental Table 4).

The list of opioid medications used in this study was developed from multiple sources, beginning with the US Food and Drug Administration’s list of controlled substances. We included all medications in the American Hospital Formulary Service therapeutic classes “opiate agonists” and “opiate partial agonists.” We also included drugs in the therapeutic class “analgesic and antipyretic, miscellaneous,” which contained an opioid component. We employed known nomenclature conventions (ie, “HC” in a drug name may indicate hydrocodone) and published studies to obtain additional drug product names, similar to previous methods.6,21,22  We conducted additional searches to identify new opioids on the market using National Drug Codes. Nonoral formulations of opioid medications were excluded (ie, suppository, spray, patch, powder, ampule, etc). There were >22 000 opioid National Drug Codes used for this study. The list of systemic steroid medications was identified from the American Hospital Formulary Service pharmacologic therapeutic class “adrenals.” Nonoral formulations of systemic steroids were excluded (ie, intranasal, ophthalmic, and topical preparations).

The data were provided to the investigators in deidentified format, with limited demographic data available in the enrollee files and for individual clinic sites. The enrollee files were matched to the pharmacy and visit data by using a unique identifier provided by the South Carolina Office of Revenue and Fiscal Affairs. Demographic data included age, sex, and a race and ethnicity variable. Ethnicity is not entered separately in South Carolina Medicaid data, so Hispanic was used as a racial category.

We identified index visits (day 0) for pneumonia or sinusitis, but only among subjects who had at least 1 pharmacy claim for any medication on days 0 to 7 that could be linked to the visit, an approach taken to eliminate follow-up visits and laboratory-related claims with a qualifying diagnosis. We linked all medication claims dispensed on days 0 to 7 days from the pneumonia or sinusitis index visit. The same subject could appear in both diagnostic categories. Pneumonia and sinusitis index visits were assigned to 1 of 3 locations: ED, UC, or ambulatory office. ED and UC visits were classified on the basis of ED and UC billing codes. All other outpatient visits were classified as ambulatory office.

We excluded children with chronic conditions, as shown in the Consolidated Standards of Reporting Trials (CONSORT) diagram (Fig 1). Chronic conditions were defined by the pediatric complex chronic conditions (CCCs) classification system23  and the Pediatric Medical Complexity Algorithm (PMCA).24  Any child who matched a code from the CCC classification system or PMCA at any visit in 2016 was excluded. Additionally, acute secondary diagnoses with potentially appropriate indications for steroids (eg, asthma) were identified by key informed interviews with general pediatricians at our institution, and those children were excluded (conditions listed in Supplemental Table 5). Finally, children seen at >1 type of clinical location or hospitalized 0 to 7 days after the index visit were excluded.

FIGURE 1

CONSORT flow diagram. a Exclusion categories overlap.

FIGURE 1

CONSORT flow diagram. a Exclusion categories overlap.

We performed descriptive analyses of the remaining sample and completed bivariate comparisons of associations between age (in 1-year increments), race and/or ethnicity, and sex and opioid and steroid dispensed prescriptions by visit clinical location. Finally, multivariate logistic regression was used to examine differences in both opioid and steroid dispensed prescriptions for pneumonia and sinusitis by clinical location, adjusting for age, sex, and race. This study was deemed as not human research by the Medical University of South Carolina Institutional Review Board, and the South Carolina Office of Research and Fiscal Affairs approved the use of the data for this study.

The CONSORT flow diagram (Fig 1) reveals the selection of children included for analysis. Of the 5660 children with a primary diagnosis of pneumonia, 3520 were excluded for the following reasons: association with a chronic condition (n = 1569), association with an acute condition amenable to steroids (n = 1387), seen at multiple types of clinical sites (n = 1047), and hospitalization (n = 291). Of the 46 470 children with a primary diagnosis of sinusitis, 16 772 were excluded for the following reasons: association with a chronic condition (n = 9413), association with an acute condition amenable to steroids (n = 6630), seen at multiple types of clinical sites (n = 2256), and hospitalization (n = 66). Exclusion categories did overlap. Details of the CCC and PMCA chronic disease classification for excluded children are shown in Supplemental Table 6.

There were 31 838 children in the study sample, with 2140 children having a primary diagnosis of pneumonia and 29 698 having a primary diagnosis of sinusitis. The subject demographics by primary diagnosis and clinical location are displayed in Table 1. For all children with a primary diagnosis of pneumonia, the mean age was 8.6 years old, and 50.6% of the subjects were boys. Subject race and/or ethnicity among children with a primary diagnosis of pneumonia included 50.6% white, 31.3% African American, 9.3% Hispanic, and 5.7% other. For all children with a primary diagnosis of sinusitis, the mean age was 10.4 years old, and 46.7% of the subjects were boys. Subject race and/or ethnicity among children with a primary diagnosis of sinusitis were similar to the breakdown among those with pneumonia. Children with a primary diagnosis of pneumonia were seen most often in the ambulatory setting (n = 1590), followed by the ED (n = 542) and UC (n = 8). Children with a primary diagnosis of sinusitis were seen most often in the ambulatory setting (n = 26 866), followed by the ED (n = 2705) and UC (n = 127).

TABLE 1

Demographics of the Study Sample Divided by Patient Diagnosis With Bivariate Comparison of Children in the ED and Ambulatory Setting

CharacteristicTotal ChildrenChildren in EDChildren Ambulatory SettingChildren in UCP
Primary diagnosis of pneumonia      
 n 2140 542 1590  
 Age, y, mean ± SD 8.6 ± 3.6 8.9 ± 3.8 8.5 ± 3.5 9.3 ± 3.2 .07a 
 Race and/or ethnicity, n (%)     ≤ .0001 
  White 1083 (50.6) 197 (36.4) 879 (55.3) 7 (87.5)  
  African American 669 (31.3) 247 (45.6) 422 (26.5) 0.0 (0.0)  
  Hispanicb 199 (9.3) 47 (8.7) 151 (9.5) 1 (12.5)  
  Otherc 122 (5.7) 41 (7.6) 81 (5.1) 0 (0.0)  
  Missing 67 (3.1) 10 (1.8) 57 (3.6) 0 (0.0)  
 Sex, n (%)     .6a 
  Boys 1083 (50.6) 269 (49.6) 811 (51.1) 3 (37.5)  
  Girls 1056 (49.4) 273 (20.4) 778 (48.9) 5 (62.5)  
  Missing 1 (0.5) 0 (0.0) 1 (0.06) 0 (0.0)  
Primary diagnosis of sinusitis      
 n 29 698 2705 26 866 127  
 Age, y, mean ± SD 10.4 ± 4.1 10.5 ± 4.4 10.4 ± 4.0 9.1 ± 3.7 .0003a 
 Race and/or ethnicity, n (%)     .0001a 
  White 15 340 (51.7) 982 (36.3) 14 268 (53.1) 90 (70.9)  
  African American 9930 (33.4) 1449 (53.6) 8458 (31.5) 23 (18.1)  
  Hispanicb 2291 (7.7) 120 (4.4) 2163 (8.1) 8 (6.3))  
  Otherc 1413 (4.8) 109 (4.0) 1299 (4.8) 5 (3.9)  
  Missing 724 (2.4) 45 (1.7) 680 (2.5) 1 (0.8)  
 Sex, n (%)     .06a 
  Boys 13 853 (46.7) 1205 (44.6) 12 592 (46.9) 56 (44.1)  
  Girls 15 842 (53.4) 1499 (55.4) 14 272 (53.1) 71 (55.9)  
  Missing 3 (0.01) 1 (0.04) 2 (0.01) 0 (0.0)  
CharacteristicTotal ChildrenChildren in EDChildren Ambulatory SettingChildren in UCP
Primary diagnosis of pneumonia      
 n 2140 542 1590  
 Age, y, mean ± SD 8.6 ± 3.6 8.9 ± 3.8 8.5 ± 3.5 9.3 ± 3.2 .07a 
 Race and/or ethnicity, n (%)     ≤ .0001 
  White 1083 (50.6) 197 (36.4) 879 (55.3) 7 (87.5)  
  African American 669 (31.3) 247 (45.6) 422 (26.5) 0.0 (0.0)  
  Hispanicb 199 (9.3) 47 (8.7) 151 (9.5) 1 (12.5)  
  Otherc 122 (5.7) 41 (7.6) 81 (5.1) 0 (0.0)  
  Missing 67 (3.1) 10 (1.8) 57 (3.6) 0 (0.0)  
 Sex, n (%)     .6a 
  Boys 1083 (50.6) 269 (49.6) 811 (51.1) 3 (37.5)  
  Girls 1056 (49.4) 273 (20.4) 778 (48.9) 5 (62.5)  
  Missing 1 (0.5) 0 (0.0) 1 (0.06) 0 (0.0)  
Primary diagnosis of sinusitis      
 n 29 698 2705 26 866 127  
 Age, y, mean ± SD 10.4 ± 4.1 10.5 ± 4.4 10.4 ± 4.0 9.1 ± 3.7 .0003a 
 Race and/or ethnicity, n (%)     .0001a 
  White 15 340 (51.7) 982 (36.3) 14 268 (53.1) 90 (70.9)  
  African American 9930 (33.4) 1449 (53.6) 8458 (31.5) 23 (18.1)  
  Hispanicb 2291 (7.7) 120 (4.4) 2163 (8.1) 8 (6.3))  
  Otherc 1413 (4.8) 109 (4.0) 1299 (4.8) 5 (3.9)  
  Missing 724 (2.4) 45 (1.7) 680 (2.5) 1 (0.8)  
 Sex, n (%)     .06a 
  Boys 13 853 (46.7) 1205 (44.6) 12 592 (46.9) 56 (44.1)  
  Girls 15 842 (53.4) 1499 (55.4) 14 272 (53.1) 71 (55.9)  
  Missing 3 (0.01) 1 (0.04) 2 (0.01) 0 (0.0)  

Bivariate analyses were completed between children in ambulatory care and the ED. UC was not included in bivariate comparisons because of the small sample size.

a

P values represent bivariate comparisons completed with χ2 analysis, except for age, which was analyzed by t test.

b

Hispanic is used as a racial category because in the South Carolina Medicaid Database, ethnicity is not entered separately.

c

Other race is defined as any race other than white, African American, and Hispanic.

All bivariate and regression analyses were performed comparing children seen in the ED and ambulatory settings, excluding UC because of the small sample size. P values represent χ2 comparisons, except for age, which was analyzed by t test. The bivariate analyses for age, race, and sex for children at each clinical location are also displayed in Table 1. Children with pneumonia who were seen in the ED were slightly older (8.9 years) compared with those seen in the ambulatory setting (8.5 years). White children were seen more often in the ambulatory setting compared with the ED, whereas African American children were more likely to be seen in the ED compared with the ambulatory setting.

Dispensed prescription rates of opioids or steroids by primary diagnosis and clinical location are illustrated in Table 2. For children with pneumonia, there were 58 associated opioid prescriptions and 302 associated steroid prescriptions. For children with sinusitis, there were 770 associated opioid prescriptions and 2435 associated steroid prescriptions.

TABLE 2

Bivariate Comparisons of the Study Sample by Opioid and Steroid Prescription

nOpioid PrescriptionSteroid Prescription
Primary diagnosis of pneumonia (n = 2132)a    
 Location    
  ED 542 34 (6.3%) 106 (19.6%) 
  Ambulatory setting 1590 24 (1.5%) 196 (12.3%) 
  Pb — ≤.0001 ≤.0001 
 Age 2302 12.4 ± 4.4 y 8.7 ± 3.8 y 
 P — ≤.0001 .7 
 Race (nonmissing)    
  White 1076 37 (3.4%) 139 (12.9%) 
  African American 669 11 (1.6%) 108 (16.1%) 
  Hispanicc 198 6 (3.0%) 31 (15.7%) 
  Otherd 122 3 (2.5%) 17 (13.9%) 
  P — .2 .3 
 Sex (nonmissing)    
  Boys 1080 27 (2.5%) 157 (14.5%) 
  Girls 1051 31 (3.0%) 145 (13.8%) 
  P — .5 .6 
Primary diagnosis of sinusitis (n = 29 571)e    
 Location    
  ED 2705 202 (7.5%) 510 (18.9%) 
  Ambulatory setting 26 866 568 (2.1%) 1922 (7.2%) 
  Pb — ≤.0001 ≤.0001 
 Age 29 996 13.1 ± 4.2 y 10.8 ± 4.2 y 
 P — ≤.0001 ≤.0001 
 Race (nonmissing)    
  White 15 250 470 (3.1%) 1308 (8.6%) 
  African American 9907 212 (2.1%) 844 (8.5%) 
  Hispanicc 2283 40 (5.3%) 122 (5.3%) 
  Otherd 1408 28 (3.7) 109 (4.6%) 
  P — ≤.0001 ≤.0001 
 Sex (nonmissing)    
  Boys 13 797 351 (2.5%) 1197 (8.7%) 
  Girls 15 771 419 (2.7%) 1235 (7.8%) 
  P — .5 .008 
nOpioid PrescriptionSteroid Prescription
Primary diagnosis of pneumonia (n = 2132)a    
 Location    
  ED 542 34 (6.3%) 106 (19.6%) 
  Ambulatory setting 1590 24 (1.5%) 196 (12.3%) 
  Pb — ≤.0001 ≤.0001 
 Age 2302 12.4 ± 4.4 y 8.7 ± 3.8 y 
 P — ≤.0001 .7 
 Race (nonmissing)    
  White 1076 37 (3.4%) 139 (12.9%) 
  African American 669 11 (1.6%) 108 (16.1%) 
  Hispanicc 198 6 (3.0%) 31 (15.7%) 
  Otherd 122 3 (2.5%) 17 (13.9%) 
  P — .2 .3 
 Sex (nonmissing)    
  Boys 1080 27 (2.5%) 157 (14.5%) 
  Girls 1051 31 (3.0%) 145 (13.8%) 
  P — .5 .6 
Primary diagnosis of sinusitis (n = 29 571)e    
 Location    
  ED 2705 202 (7.5%) 510 (18.9%) 
  Ambulatory setting 26 866 568 (2.1%) 1922 (7.2%) 
  Pb — ≤.0001 ≤.0001 
 Age 29 996 13.1 ± 4.2 y 10.8 ± 4.2 y 
 P — ≤.0001 ≤.0001 
 Race (nonmissing)    
  White 15 250 470 (3.1%) 1308 (8.6%) 
  African American 9907 212 (2.1%) 844 (8.5%) 
  Hispanicc 2283 40 (5.3%) 122 (5.3%) 
  Otherd 1408 28 (3.7) 109 (4.6%) 
  P — ≤.0001 ≤.0001 
 Sex (nonmissing)    
  Boys 13 797 351 (2.5%) 1197 (8.7%) 
  Girls 15 771 419 (2.7%) 1235 (7.8%) 
  P — .5 .008 

Bivariate analyses were completed between Ambulatory and ED children. UC was not included in bivariate comparisons because of small sample size. —, not applicable.

a

Opioid prescription: n = 58; steroid prescription: n = 302.

b

P values represent bivariate comparisons completed with χ2 analysis, except for age, which was analyzed by t test.

c

Hispanic is used as a racial category because in the South Carolina Medicaid database, ethnicity is not entered separately.

d

Other race is defined as any race other than white, African American, and Hispanic.

e

Opioid prescription: n = 770; steroid prescription: n = 2435.

Pneumonia visits to an ED were more often linked to a dispensed opioid prescription than pneumonia visits in ambulatory settings (34 of 542 [6.3%] vs 24 of 1590 [1.5%], respectively; P ≤ .0001), and a pneumonia ED visit was also more often linked to a dispensed steroid prescription than a pneumonia visit in the ambulatory setting (106 of 542 [19.6%] vs 196 of 1590 [12.3%], respectively; P ≤ .0001). Among children with pneumonia, increasing age was associated with dispensed opioid (P ≤ .0001) but not with dispensed steroid.

Sinusitis visits to an ED were more often linked to a dispensed opioid than sinusitis visits in ambulatory settings (202 of 2705 [7.5%] vs 568 of 26 866 [2.1%], respectively; P ≤ .0001). Steroid prescriptions were also more commonly linked to a sinusitis ED visit than an ambulatory visit (510 of 2705 [18.9%] vs 1922 of 26 866 [7.2%], respectively; P ≤ .0001). Among children with sinusitis, increasing age was positively associated with receipt of an opioid and steroid. White patients were more likely to receive opioid and steroid medications for sinusitis compared with patients of all other races. Boys with sinusitis were more likely to receive steroids compared with girls (8.7% vs 7.8%, respectively).

The logistic regression analyses used to evaluate associations among treatment location, patient age, race, and sex and receipt of a dispensed prescription for an opioid or a steroid are shown in Table 3. Children with pneumonia were more likely to receive a dispensed opioid prescription with each advancing year of age (adjusted odds ratio [aOR] 1.26; 95% confidence interval [CI] 1.18–1.34) or if they were seen in the ED versus ambulatory setting (aOR 4.69; 95% CI 2.68–8.21). Children with pneumonia were more likely to receive a dispensed steroid prescription if seen in the ED versus ambulatory setting (aOR 1.67; 95% CI 1.28–2.18). Children with sinusitis were more likely to receive an opioid prescription if they were older (aOR 1.17; 95% CI 1.15–1.20) or if they were seen in the ED versus ambulatory setting (aOR 4.02; 95% CI 3.38–4.78). Children with sinusitis were more likely to receive a steroid with each advancing year of age, if they were male, or if they were seen in the ED (aOR 3.05; 95% CI 2.74–3.41). Children of African American race were less likely to receive an opioid for either diagnosis. Both African American and Hispanic children were less likely to receive a steroid for sinusitis.

TABLE 3

Adjusted Associations Between Patient Variables and Opioid or Corticosteroid Prescription, 2016 Visits With a Primary Diagnosis of Pneumonia or Sinusitis

VariablePneumonia, aOR (95% CI), PSinusitis, aOR (95% CI), P
Opioid PrescriptionSteroid PrescriptionaOpioid PrescriptionSteroid Prescriptiona
Age 1.26 (1.18–1.34), <0.0001 1.01 (0.98–1.04), 0.7 1.17 (1.15–1.20), <0.0001 1.02 (1.01–1.03), <0.0001 
Race and/or ethnicity     
 White Reference Reference Reference Reference 
 African American 0.4 (0.2–0.8), 0.01 1.19 (0.90–1.57), 0.2 0.61 (0.52–0.72), <0.0001 0.89 (0.81–0.98), 0.02 
 Hispanicb 1.0 (0.41–2.59), 0.9 1.24 (0.81–1.89), 0.3 0.74 (0.53–1.02), 0.07 0.63 (0.52–0.77), <0.0001 
 Otherc 0.75 (0.21–0.26), 0.6 1.01 (0.59–1.75), 0.9 0.71 (0.48–1.05), 0.08 0.90 (0.73–1.10), 0.3 
Sex     
 Boys 0.93 (0.54–1.62), 0.8 1.07 (0.84–1.37), 0.6 1.07 (0.93–1.23), 0.35 1.15 (1.06–1.25), 0.001 
 Girls Reference Reference Reference Reference 
Clinical location     
 ED 4.69 (2.68–8.21), <0.0001 1.67 (1.28–2.18), 0.0002 4.02 (3.38–4.78), <0.0001 3.05 (2.74–3.41), <0.0001 
 Ambulatory setting Reference Reference Reference Reference 
VariablePneumonia, aOR (95% CI), PSinusitis, aOR (95% CI), P
Opioid PrescriptionSteroid PrescriptionaOpioid PrescriptionSteroid Prescriptiona
Age 1.26 (1.18–1.34), <0.0001 1.01 (0.98–1.04), 0.7 1.17 (1.15–1.20), <0.0001 1.02 (1.01–1.03), <0.0001 
Race and/or ethnicity     
 White Reference Reference Reference Reference 
 African American 0.4 (0.2–0.8), 0.01 1.19 (0.90–1.57), 0.2 0.61 (0.52–0.72), <0.0001 0.89 (0.81–0.98), 0.02 
 Hispanicb 1.0 (0.41–2.59), 0.9 1.24 (0.81–1.89), 0.3 0.74 (0.53–1.02), 0.07 0.63 (0.52–0.77), <0.0001 
 Otherc 0.75 (0.21–0.26), 0.6 1.01 (0.59–1.75), 0.9 0.71 (0.48–1.05), 0.08 0.90 (0.73–1.10), 0.3 
Sex     
 Boys 0.93 (0.54–1.62), 0.8 1.07 (0.84–1.37), 0.6 1.07 (0.93–1.23), 0.35 1.15 (1.06–1.25), 0.001 
 Girls Reference Reference Reference Reference 
Clinical location     
 ED 4.69 (2.68–8.21), <0.0001 1.67 (1.28–2.18), 0.0002 4.02 (3.38–4.78), <0.0001 3.05 (2.74–3.41), <0.0001 
 Ambulatory setting Reference Reference Reference Reference 

Multivariate logistic regression.

a

Only systemic corticosteroids were evaluated.

b

Hispanic is used as a racial category because in the South Carolina Medicaid database, ethnicity is not entered separately.

c

Other race is defined as any race other than white, African American, and Hispanic.

These data reveal that South Carolina Medicaid-insured, school-aged children and adolescents treated for pneumonia or sinusitis received opioid and corticosteroid prescriptions at higher frequencies when seen in the ED compared with the ambulatory setting, supporting our hypothesis. Each of the 828 opioid and 2737 systemic steroid prescriptions in the data set represent a potentially inappropriate prescription. These rates appear excessive given that the use of these medications is not supported by available research or recommended in national guidelines.2,3,1417  Restating our findings, among children seen in the ED, 1 in 16 children with pneumonia received an opioid prescription and 1 in 13 children with sinusitis received an opioid prescription. The findings suggest that focused improvement in prescribing opioids and steroids is needed in both the ED and ambulatory settings, but especially in the ED. Although more children who are acutely ill may be seen in the ED, national practice guidelines and research remain relevant for these patients.

Our findings of age-related greater opioid prescribing are consistent with previous research evaluating these medications as analgesics. Older children seen in EDs are more likely to receive opioids for pain compared with young children,2528  even with identical acute pain scores.26  Previous research has revealed that increasing age is associated with adverse drug events, hospitalizations, critical care use, and death due to opioids.11,21,29,30  These findings suggest the need for opioid stewardship for older adolescents and young adults, decreasing exposure to these potentially hazardous medications in a high-risk patient population.

The finding of lower adjusted odds of receiving an opioid for African American children echoes findings of other research evaluating analgesic treatment of racial and ethnic minorities.3134  Racial differences exist in opioid prescribing to children in the ambulatory setting, with white children more likely to receive opioids and minorities more likely to receive nonopioid analgesics on the basis of nationally representative data from the Medical Expenditure Panel Survey.33  In the ED setting, studies have revealed that African American patients were less likely to receive opioids and narcotic analgesics compared with white patients for abdominal pain and ED pain diagnoses codes, including arthritis or joint pain, back pain, cancer-related pain, chest pain, cholelithiasis, fractures, headache, injury excluding fracture, neck pain, nephrolithiasis, pelvic pain, sickle cell anemia, dental or jaw pain, fibromyalgia, and peripheral neuropathy.27,34  Data from the Medical Expenditure Panel Survey from 2003 to 2014 revealed that non-Hispanic African American children, Hispanic children, and Asian American children were all less likely to receive opioids compared with white children.33  Many factors likely contribute to the differential receipt of opioids by children of minority race or ethnicity, including patient cultural differences in concerns about opioids, as well as provider-level factors, such as bias in the risks of prescribing opioids to children of nonwhite racial or Hispanic backgrounds.34,35  Differences in steroid receipt by race may also be multifactorial but have been evaluated less.

There are several limitations to this study. The use of Medicaid data from a single year in 1 state limits generalizability. However, we examined recent data (from 2016) well after publication of relevant guidelines and after concerns of opioid prescribing had become widespread. Because the study is limited to children with Medicaid insurance coverage, we are unable to describe prescribing patterns for South Carolina children with private insurance, but Medicaid-insured children compose at least 50% of the children in South Carolina. Adult data reveal considerable variation in opioid prescribing among states and related to insurance status, but it is unclear whether there is similar variation in children.36 

Linking an opioid or steroid dispensed on days 0 to 7 to the index visit may not accurately represent what was prescribed at a particular visit. However, in previous studies, researchers have used this time interval and approach.7,21  To address this concern, we excluded any patient who had a visit to >1 type of site or was hospitalized between days 0 and 7. We also performed a post hoc analysis to further elucidate the time period from index visit to medication claim for an opioid or steroid. Although ∼80% of the prescriptions were dispensed on day 0 or 1, substantial percentages were filled up through day 7.

Our initial goal was to group acute care visits into more detailed groups, such as UC, after-hours care, and minute clinics; however, this was not possible because of limited billing code information. Furthermore, this lack of site-specific information may have led to some UC visits being grouped in the ambulatory category. The ambulatory category includes a mixture of subspecialist outpatient visits in addition to primary care visits. We were also unable to identify prescriptions from visits with adult versus pediatric providers on the basis of the information available in the Medicaid data set.

Lastly, although we eliminated potential appropriate acute indications for systemic corticosteroids, we were not able to eliminate potential acute indications for opioids as easily. We had difficulty in obtaining a comprehensive list of ICD-10 codes for acute indications for opioids because of the nature of ICD-10 codes. For example, there are 17 045 codes in International Classification of Diseases, 10th Revision, Clinical Modification related to fractures.37  We did perform a post hoc evaluation of the secondary diagnoses for children who received opioids by searching for “pain” and “fracture.” Of our patients with a primary diagnosis of sinusitis or pneumonia, there was only 1 patient who had a secondary diagnosis of fracture and 24 patients with a secondary diagnosis of pain, and none received an opioid. Thus, the small subset of patients who may have had secondary diagnoses that would warrant an opioid prescription would not have changed the overall results.

Our findings support the need for continued pediatric opioid and corticosteroid stewardship in children. Systemic corticosteroids should be prescribed judiciously to children, and providers should be cognizant that repeated or prolonged courses put children at risk for adrenal suppression and hypothalamic-pituitary-adrenal axis dysfunction.13  Although pain and symptomatic management is important, providers for children must also be aware of the trends in opioid abuse and diversion and must mitigate those risks while still providing adequate analgesia and symptom control.38,39  Providers must adhere to recommended pediatric guidelines for appropriate indications for prescribing opioids and steroids and use nonopioid alternatives and nonpharmacologic supportive care interventions when possible.1418  The opioid crisis in the United States is a multifaceted public health problem; all health care professionals play a role in preventing pediatric opioid misuse.

One potential area of focus moving forward is for pediatricians and pediatric pharmacists to collaborate with acute care providers, especially with ED providers, to reduce unnecessary opioid and systemic corticosteroid prescribing for pediatric pneumonia and sinusitis. Providers should have specific and ongoing updates regarding current national guidelines for pediatric patients and the potential for harm with these agents. Other potential strategies documented to improve care include evidence-based pathways and protocol-driven medication order sets created by interprofessional teams.40 

Children 5 to 18 years old with pneumonia or sinusitis were more likely to receive opioid and systemic corticosteroid medications in the ED versus the ambulatory setting. These findings suggest that providers are not following guidelines when prescribing these agents for treatment of pneumonia or sinusitis. Quality improvement and collaborative interventions with acute care providers may reduce unnecessary medication prescribing for pediatric pneumonia and sinusitis.

We thank the South Carolina Revenue and Fiscal Affairs Office, South Carolina Medicaid for providing study data. We thank the community pediatricians participating in the South Carolina Pediatric Practice Research Network for providing initiative for the study.

Drs Phang, Roberts, and Basco conceptualized and designed the study, drafted the initial manuscript, and reviewed and revised the manuscript; Mrs Ebeling performed all data management and analyses and assisted with manuscript preparation, including preparing tables, and final review of the manuscript; and Dr Garner provided content and expertise, contributed input on analysis approaches and findings, and assisted with manuscript preparation, including review of the final manuscript.

FUNDING: Supported by the National Institutes of Health National Center for Advancing Translational Sciences through grant UL1 TR001450. Funded by the National Institutes of Health (NIH).

     
  • aOR

    adjusted odds ratio

  •  
  • CCC

    complex chronic condition

  •  
  • CCS

    Clinical Classifications Software

  •  
  • CI

    confidence interval

  •  
  • CONSORT

    Consolidated Standards of Reporting Trials

  •  
  • ED

    emergency department

  •  
  • ICD-10

    International Classification of Diseases, 10th Revision

  •  
  • PMCA

    Pediatric Medical Complexity Algorithm

  •  
  • UC

    urgent care

1
Tobias
JD
,
Green
TP
,
Coté
CJ
;
Section on Anesthesiology and Pain Medicine
;
Committee on Drugs
.
Codeine: time to say “no”
.
Pediatrics
.
2016
;
138
(
4
):
e20162396
2
Goldman
RD
.
Codeine for acute cough in children
.
Can Fam Physician
.
2010
;
56
(
12
):
1293
1294
3
Ambroggio
L
,
Test
M
,
Metlay
JP
, et al
.
Adjunct systemic corticosteroid therapy in children with community-acquired pneumonia in the outpatient setting
.
J Pediatric Infect Dis Soc
.
2015
;
4
(
1
):
21
27
4
Weiss
AK
,
Hall
M
,
Lee
GE
,
Kronman
MP
,
Sheffler-Collins
S
,
Shah
SS
.
Adjunct corticosteroids in children hospitalized with community-acquired pneumonia
.
Pediatrics
.
2011
;
127
(
2
). Available at: www.pediatrics.org/cgi/content/full/127/2/e255
5
Stern
A
,
Skalsky
K
,
Avni
T
,
Carrara
E
,
Leibovici
L
,
Paul
M
.
Corticosteroids for pneumonia
.
Cochrane Database Syst Rev
.
2017
;(
12
):
CD007720
6
Basco
WT
 Jr.
,
Ebeling
M
,
Garner
SS
,
Hulsey
TC
,
Simpson
K
.
Opioid prescribing and potential overdose errors among children 0 to 36 months old
.
Clin Pediatr (Phila)
.
2015
;
54
(
8
):
738
744
7
Chung
CP
,
Callahan
ST
,
Cooper
WO
, et al
.
Outpatient opioid prescriptions for children and opioid-related adverse events
.
Pediatrics
.
2018
;
142
(
2
):
e20172156
8
Doherty
C
,
Mc Donnell
C
.
Tenfold medication errors: 5 years’ experience at a university-affiliated pediatric hospital
.
Pediatrics
.
2012
;
129
(
5
):
916
924
9
Paul
IM
,
Reynolds
KM
,
Green
JL
.
Adverse events associated with opioid-containing cough and cold medications in children
.
Clin Toxicol (Phila)
.
2018
;
56
(
11
):
1162
1164
10
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
11
Kane
JM
,
Colvin
JD
,
Bartlett
AH
,
Hall
M
.
Opioid-related critical care resource use in US children’s hospitals [published correction appears in Pediatrics. 2018;141(6):e20180935]
.
Pediatrics
.
2018
;
141
(
4
):
e20173335
12
McCabe
SE
,
West
BT
,
Veliz
P
,
McCabe
VV
,
Stoddard
SA
,
Boyd
CJ
.
Trends in medical and nonmedical use of prescription opioids among US adolescents: 1976–2015
.
Pediatrics
.
2017
;
139
(
4
):
e20162387
13
Lexicomp. Corticosteroids general statement.
2018
. Available at: http://online.lexi.com. Accessed September 9, 2018
14
Peters
AT
,
Spector
S
,
Hsu
J
, et al;
Joint Task Force on Practice Parameters, representing the American Academy of Allergy, Asthma and Immunology, the American College of Allergy, Asthma and Immunology, and the Joint Council of Allergy, Asthma and Immunology
.
Diagnosis and management of rhinosinusitis: a practice parameter update
.
Ann Allergy Asthma Immunol
.
2014
;
113
(
4
):
347
385
15
Bradley
JS
,
Byington
CL
,
Shah
SS
, et al;
Pediatric Infectious Diseases Society
;
Infectious Diseases Society of America
.
The management of community-acquired pneumonia in infants and children older than 3 months of age: clinical practice guidelines by the Pediatric Infectious Diseases Society and the Infectious Diseases Society of America
.
Clin Infect Dis
.
2011
;
53
(
7
):
e25
e76
16
Venekamp
RP
,
Bonten
MJ
,
Rovers
MM
,
Verheij
TJ
,
Sachs
AP
.
Systemic corticosteroid monotherapy for clinically diagnosed acute rhinosinusitis: a randomized controlled trial
.
CMAJ
.
2012
;
184
(
14
):
E751
E757
17
Venekamp
RP
,
Thompson
MJ
,
Hayward
G
, et al
.
Systemic corticosteroids for acute sinusitis
.
Cochrane Database Syst Rev
.
2014
;(
3
):
CD008115
18
Wald
ER
,
Applegate
KE
,
Bordley
C
, et al;
American Academy of Pediatrics
.
Clinical practice guideline for the diagnosis and management of acute bacterial sinusitis in children aged 1 to 18 years
.
Pediatrics
.
2013
;
132
(
1
). Available at: www.pediatrics.org/cgi/content/full/132/1/e262
19
Basco
WT
 Jr.
,
Roberts
JR
,
Ebeling
M
,
Garner
SS
,
Hulsey
TC
,
Simpson
K
.
Indications for use of combination acetaminophen/opioid drugs in infants <6 Months old
.
Clin Pediatr (Phila)
.
2018
;
57
(
6
):
741
744
20
Agency for Healthcare Research and Quality, Healthcare Cost and Utilization Project
. Clinical Classifications Software (CCS) for ICD-10-PCS (beta version). Available at: https://www.hcup-us.ahrq.gov/toolssoftware/ccs10/ccs10.jsp. Accessed September 9, 2018
21
Chua
KP
,
Shrime
MG
,
Conti
RM
.
Effect of FDA investigation on opioid prescribing to children after tonsillectomy/adenoidectomy
.
Pediatrics
.
2017
;
140
(
6
):
e20171765
22
Defalco
FJ
,
Ryan
PB
,
Soledad Cepeda
M
.
Applying standardized drug terminologies to observational healthcare databases: a case study on opioid exposure
.
Health Serv Outcomes Res Methodol
.
2013
;
13
(
1
):
58
67
23
Feudtner
C
,
Feinstein
JA
,
Zhong
W
,
Hall
M
,
Dai
D
.
Pediatric complex chronic conditions classification system version 2: updated for ICD-10 and complex medical technology dependence and transplantation
.
BMC Pediatr
.
2014
;
14
:
199
24
Simon
TD
,
Cawthon
ML
,
Stanford
S
, et al;
Center of Excellence on Quality of Care Measures for Children with Complex Needs (COE4CCN) Medical Complexity Working Group
.
Pediatric medical complexity algorithm: a new method to stratify children by medical complexity
.
Pediatrics
.
2014
;
133
(
6
). Available at: www.pediatrics.org/cgi/content/full/133/6/e1647
25
Alexander
J
,
Manno
M
.
Underuse of analgesia in very young pediatric patients with isolated painful injuries
.
Ann Emerg Med
.
2003
;
41
(
5
):
617
622
26
Kircher
J
,
Drendel
AL
,
Newton
AS
, et al
.
Acute pediatric musculoskeletal pain management in North America: a practice variation survey
.
Clin Pediatr (Phila)
.
2014
;
53
(
14
):
1326
1335
27
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
28
Van Winkle
PJ
,
Ghobadi
A
,
Chen
Q
,
Menchine
M
,
Sharp
AL
.
Association of age and opioid use for adolescents and young adults in community emergency departments
.
Am J Emerg Med
.
2019
;
37
(
8
):
1397
1403
29
Gaither
JR
,
Leventhal
JM
,
Ryan
SA
,
Camenga
DR
.
National trends in hospitalizations for opioid poisonings among children and adolescents, 1997 to 2012
.
JAMA Pediatr
.
2016
;
170
(
12
):
1195
1201
30
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
31
Joynt
M
,
Train
MK
,
Robbins
BW
,
Halterman
JS
,
Caiola
E
,
Fortuna
RJ
.
The impact of neighborhood socioeconomic status and race on the prescribing of opioids in emergency departments throughout the United States
.
J Gen Intern Med
.
2013
;
28
(
12
):
1604
1610
32
Sadhasivam
S
,
Chidambaran
V
,
Ngamprasertwong
P
, et al
.
Race and unequal burden of perioperative pain and opioid related adverse effects in children
.
Pediatrics
.
2012
;
129
(
5
):
832
838
33
Groenewald
CB
,
Rabbitts
JA
,
Hansen
EE
,
Palermo
TM
.
Racial differences in opioid prescribing for children in the United States
.
Pain
.
2018
;
159
(
10
):
2050
2057
34
Johnson
TJ
,
Weaver
MD
,
Borrero
S
, et al
.
Association of race and ethnicity with management of abdominal pain in the emergency department
.
Pediatrics
.
2013
;
132
(
4
). Available at: www.pediatrics.org/cgi/content/full/132/4/e851
35
Hollingshead
NA
,
Meints
SM
,
Miller
MM
,
Robinson
ME
,
Hirsh
AT
.
A comparison of race-related pain stereotypes held by white and black individuals
.
J Appl Soc Psychol
.
2016
;
46
(
12
):
718
723
36
Kuehn
BM
.
CDC: major disparities in opioid prescribing among states: some states crack down on excess prescribing
.
JAMA
.
2014
;
312
(
7
):
684
686
37
Carr
K.
No bones about it: fracture coding in ICD-10 is a challenge.
2014
. Available at: https://www.icd10monitor.com/no-bones-about-it-fracture-coding-in-icd-10-is-a-challenge. Accessed September 9, 2018
38
Dowell
D
,
Haegerich
TM
,
Chou
R
.
CDC guideline for prescribing opioids for chronic pain–United States, 2016
.
JAMA
.
2016
;
315
(
15
):
1624
1645
39
Harbaugh
CM
,
Gadepalli
SK
.
Pediatric postoperative opioid prescribing and the opioid crisis
.
Curr Opin Pediatr
.
2019
;
31
(
3
):
378
385
40
Monroe
K
,
Cohen
CT
,
Whelan
K
, et al
.
Quality initiative to improve time to antibiotics for febrile pediatric patients with potential neutropenia
.
Pediatr Qual Saf
.
2018
;
3
(
4
):
e095

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.

Supplementary data