Abstract
Study objective:
To evaluate for increases in the use and costs of respiratory viral testing in pediatric emergency departments (EDs) because of the COVID-19 pandemic.
Methods:
We performed a cross-sectional study using the pediatric health information system. Eligible subjects were children (90 days to 18 years) who were discharged from a pediatric ED and included in the pediatric health information system from October 2016 through March 2024. To evaluate for changes in the frequency and costs of respiratory viral testing, we performed an interrupted time series analysis across 3 study periods: prepandemic (October 1, 2016 to March 14, 2020), early pandemic (March 15, 2020 to December 31, 2023), and late pandemic (January 1, 2023 to March 31, 2024).
Results:
We included 15,261,939 encounters from 34 pediatric EDs over the 90-month study period. At least 1 viral respiratory test was performed for 460,826 of 7,311,177 prepandemic encounters (6.3%), 1,240,807 of 5,100,796 early pandemic encounters (24.3%), and 545,696 of 2,849,966 late pandemic encounters (19.1%). There was a positive prepandemic slope in viral testing (0.17% encounters/month; 95% CI 0.17 to 0.18). The early pandemic was associated with a shift change of 4.98% (95% CI 4.90 to 5.07) and a positive slope (0.54% encounters/month; 95% CI 0.54 to 0.55). The late pandemic period was associated with a negative shift (−17.80%; 95% CI −17.90 to −17.70) and a positive slope (0.42% encounters/month; 95% CI 0.41 to 0.42). The slope in testing costs increased from $5,000/month (95% CI $4,200 to $5,700) to $33,000/month (95% CI $32,000 to $34,000) during the early pandemic.
Conclusion:
Respiratory testing and associated costs increased during the COVID-19 pandemic and were sustained despite decreasing incidence of disease. These findings highlight a need for further efforts to clarify indications for viral testing in the ED and efforts to reduce low-value testing.
INTRODUCTION
Background
Respiratory viral testing can provide diagnostic specificity for children with febrile or respiratory infections. In several well-defined situations, respiratory viral testing can aid prognostication1 or clinical decisionmaking.2–6 However, viral testing is costly and uncomfortable, and, for most patients, the gain in specificity rarely benefits subsequent testing or treatment decisions.7–11 Nevertheless, prior work has suggested that respiratory viral testing is still commonly performed for children (43% to 63% for bronchiolitis),12,13 with inconsistent guideline adherence.14 The onset of the COVID-19 pandemic resulted in changes in both respiratory testing recommendations and the motivations of both clinicians and caregivers, at times contrasting with established practices.15,16 Moreover, especially once the Centers for Disease Control and Prevention (CDC) recommended testing for COVID-19 in patients with respiratory symptoms,17,18 respiratory viral testing patterns in children likely changed substantially.
Importance
The role of respiratory viral testing among children in the acute care setting is one that requires reevaluation following the emergence of COVID-19. Overtesting for viral respiratory pathogens was an issue of concern before the pandemic. Objective data on recent trends in pathogen testing will provide a starting point to allow for a more rigorous evaluation of current testing practices and their associated costs. Further, such information could guide revised and more nuanced recommendations for judicious deployment of viral testing.
Goals of This Investigation
In this study, we sought to evaluate changes in respiratory viral testing associated with the COVID-19 pandemic among children discharged from a pediatric emergency department (ED). We also sought to describe changes in the costs associated with viral testing patterns.
MATERIALS AND METHODS
Study Design and Data Source
We performed a cross-sectional study using data from the pediatric health information system (PHIS). PHIS is an administrative database with encounter-level data from children’s hospitals associated with the Children’s Hospital Association (Overland Park, Kansas). PHIS includes ED, inpatient, and ambulatory surgery data for each encounter. The Children’s Hospital Association and the contributing hospitals jointly manage efforts to optimize the integrity and quality of PHIS data, and PHIS is one of the most widely used databases in pediatric research. For this study, we included PHIS data for the 34 children’s hospitals that had complete data for the study period. The institutional review board at Ann & Robert H. Lurie Children’s Hospital in Chicago approved the protocol as exempt prior to study commencement. In writing the current report, we adhered to the STrengthening the Reporting of OBservational studies in Epidemiology reporting guidelines.19
Selection of Participants
We included encounters from PHIS for children (90 days through 18 years of age) who had an ED discharge between October 1, 2016, and March 31, 2024. We focused on ED discharges because institution-specific testing practices for children admitted to the hospital may vary widely and drive viral testing practices. We excluded encounters with in-ED mortality or a complex chronic condition.20 Children with complex chronic conditions are often excluded from clinical practice guidelines for common respiratory illnesses and may be among the specific populations who benefit from viral testing.
Methods of Measurements
We obtained encounter-level demographic and cost data from PHIS. Demographic data included age, sex, primary payer (public, private, or other), and principal diagnosis. We classified age into the following groups: 0 to 1 year (bronchiolitis more common), 2 to 4 years (other types of lower respiratory tract infections and childhood asthma more frequent), and 5 to 17 years. To identify relevant diagnostic subgroups, we used the following International Classification of Disease, 10th revision, clinical modification codes (eg, as either a principal or secondary diagnosis): upper respiratory tract infections, sinusitis, pharyngitis, tonsillitis, bronchiolitis, pneumonia, and named viral infections (rhinovirus, enterovirus, coronavirus, adenovirus, metapneumovirus, influenza, parainfluenza, and respiratory syncytial virus; Table E1, available online at http://www.annemergmed.com). These codes were included from any diagnosis code position (primary or secondary).
Outcome Measures
Our primary outcomes were the frequency of respiratory viral testing and the associated cost. Respiratory viral tests included antigen or antibody assay, DNA probe, PCR, or reverse transcriptase polymerase chain (Table E2, available online at http://www.annemergmed.com). Costs were computed as the line-item charge for the viral test multiplied by the hospital’s laboratory-specific cost-to-charge ratio, divided by the regional wage index. This approach to the calculation of costs is a common measure of approximating real expenditures.21 Because of inflation over the study period, we adjusted all costs to 2023 dollars using the consumer price index for medical care.22,23
Primary Data Analysis
We generated descriptive statistics for the study sample, stratified by whether viral testing was performed. We calculated the median proportion (IQR) of patients with respiratory viral testing performed per month across all hospitals. We examined for changes in the monthly median proportion of patients per month with viral testing over 3 time periods: pre-COVID-19, early pandemic, and late pandemic. The pre-COVID-19 period was from October 1, 2016, to March 14, 2020. We divided the COVID-19 pandemic into 2 phases: early (March 15, 202024 to December 31, 2022) and late (January 1, 2023 to December 31, 2023). The late pandemic phase encompassed the “triple” viral surge of 2023.25 We also created a box plot of the proportion of patients with viral testing by hospital by study period.
Our main analysis of the result of the pandemic on viral testing was an encounter-level interrupted time series linear mixed-effects model. Given broad fluctuations in presentations of children to the ED during the first year of the pandemic,26–28 we modeled the performance of respiratory testing as a proportion of all ED encounters per month. We used sine and cosine terms as model predictors to account for seasonal variations in viral testing29 and included random intercepts by the hospital (to adjust for clustering by the hospital). We analyzed for changes in the frequency of viral testing using slope changes for monthly viral testing and shifts at specific time points, ie, at the onset of the pandemic and the transition from early to late pandemic. The slope changes are defined as the monthly rate changes in the percentage of encounters that included a respiratory viral test, with one at the onset of the pandemic and the second at the start of 2023. A positive slope change indicates an increase in the rate of change in the percentage of encounters with respiratory viral testing at the given time points (pandemic onset, early to late pandemic). Conversely, a negative slope change indicates a decrease in the rate of change in the percentage of encounters with respiratory viral testing. In contrast to slope (which refers to changes in unit/time), shift changes can be understood as the “sudden” changes in the baseline percentage of testing at the interruption points. A positive shift change would, therefore, indicate a sudden increase in respiratory viral testing. A negative shift change would indicate a sudden decrease in testing at this time point. Models were expressed as coefficients with 95% CI.
We conducted several sensitivity analyses for the main interrupted time series analysis. First, considering that widespread COVID-19 testing was not available in the months immediately following the onset of the pandemic,30 we repeated the interrupted time series analysis with wash-out periods of 2 and 4 months. In doing so, we removed all data from the onset until 2 and 4 months after the pandemic started (on March 15, 2020). Second, to examine testing patterns for multiplex respiratory viral testing, we repeated the interrupted time series using 2 definitions for multiplex testing: ≥3 tests and the subgroup of multiplex viral tests that were specifically stated to include ≥6 viruses. We used ≥6 tests because multiplex samples of this size or greater could be ascertained from the data set and because this cutoff would likely be generally accepted as true multiplex testing (vs a test that included both COVID-19 and influenza). Third, we repeated the interrupted time series by limiting our sample to patients with a respiratory infection and children with a diagnosis of bronchiolitis, for whom specific viral testing guidelines exist.31 Finally, we repeated our interrupted time series analysis stratified by age to evaluate age-specific differences in testing patterns. For each sensitivity analysis, we recalculated the interrupted time series using the same approach described above.
To evaluate changes in cost over time, we aggregated daily charges and developed a seasonality-adjusted interrupted time series model (using the same trigonometric functions as above but without random effects). We repeated sensitivity analysis for multiplex viral testing, respiratory diagnosis, and age group. Analyses were all performed using the nlme (v3.1–162) package in R, version 4.3.2 (R Foundation for Statistical Computing, Vienna, Austria).
RESULTS
Characteristics of Study Subjects
We identified 15,923,795 eligible patient encounters in PHIS for the 90-month study period. After excluding encounters for children with a complex chronic condition (n=658,787) or with in-ED mortality (n=3,069), our final sample for analyses consisted of 15,261,939 encounters with ED discharge. A small percentage of patients had missing data for payor status (<3%) or sex (<0.01%). For the study sample, 7,311,177 (47.9%) encounters were prepandemic, 5,100,796 (33.4%) early pandemic, and 2,849,966 late pandemic (18.7%; Table 1). There was a slight male predominance (52.4%), and the median age was 5.4 years (IQR 2.0 to 11.1 years).
Table 1.
Characteristics of included encounters, stratified by the performance of viral testing. Numbers represent N (%) or median (IQR).
| Viral Testing Performed (N=2,247,329) |
Viral Testing Not Performed (N=13,014,610) |
|||||
|---|---|---|---|---|---|---|
| Period | Prepandemic (N=460,826) |
Early Pandemic (N=1,240,807) |
Late Pandemic (N= 545,696) |
Prepandemic (N=6,850,351) |
Early Pandemic (N=3,859,989) |
Late Pandemic (N=2,304,270) |
| Age, median (IQR) | 3.1 (1.2–7.0) | 3.9 (1.5–8.9) | 3.9 (1.4–8.2) | 5.4 (2.0–11.0) | 6.0 (2.3–12.0) | 6.3 (2.6–11.8) |
| Sex | ||||||
| Male | 244,502 (53.1) | 650,934 (52.5) | 287,506 (52.7) | 3,582,195 (52.3) | 2,013,729 (52.2) | 1,210,866 (52.5) |
| Female | 216,293 (46.9) | 589,734 (47.5) | 258,146 (47.3) | 3,265,764 (47.7) | 1,845,593 (47.8) | 1,093,032 (47.4) |
| Primary payor | ||||||
| Public | 321,690 (69.8) | 870,244 (70.1) | 368,823 (67.6) | 4,384,208 (64.0) | 2,426,974 (62.9) | 1,429,287 (62.0) |
| Private | 107,351 (23.3) | 312,404 (25.2) | 144,850 (26.5) | 1,870,020 (27.3) | 1,195,300 (31.0) | 725,931 (31.5) |
| Other | 26,160 (5.7) | 43,151 (3.5) | 27,518 (5.0) | 352,475 (5.1) | 170,557 (4.4) | 118,908 (5.2) |
| Respiratory diagnosis | ||||||
| Named viral infections | 155,156 (33.7) | 241,817 (19.5) | 124,326 (22.8) | 82,051 (1.2) | 73,604 (1.9) | 37,696 (1.6) |
| URI | 78,003 (16.9) | 256,626 (20.7) | 105,913 (19.4) | 621,678 (9.1) | 226,412 (5.9) | 147,917 (6.4) |
| Pharyngitis | 28,105 (6.1) | 71,380 (5.8) | 48,279 (8.8) | 298,872 (4.4) | 79,249 (2.1) | 93,123 (4.0) |
| Bronchiolitis | 32,336 (7.0) | 56,732 (4.6) | 31,955 (5.9) | 123,241 (1.8) | 45,983 (1.2) | 29,154 (1.3) |
| Pneumonia | 13,271 (2.9) | 16,128 (1.3) | 12,217 (2.2) | 58,545 (0.9) | 13,823 (0.4) | 13,796 (0.6) |
| Sinusitis | 1,299 (0.3) | 2,493 (0.2) | 1,701 (0.3) | 12,762 (0.2) | 4,815 (0.1) | 4,342 (0.2) |
| Tonsillitis | 813 (0.2) | 2,068 (0.2) | 1,349 (0.2) | 10,106 (0.1) | 4,056 (0.1) | 3,535 (0.2) |
URI, upper respiratory infection. Sex missing in 3,645 (<0.01%) and payor status in 366,088 (2.4%).
Main Results
During the prepandemic period, 460,826 (6.3%) encounters had any respiratory viral test, 100,261 (1.4%) had multiplex viral testing, and 54,035 (0.7%) had a multiplex panel of ≥6 pathogens. During the early pandemic, 1,240,807 (24.3%) had any respiratory viral test, 512,575 (10.0%) had multiplex viral testing, and 97,834 (1.9%) had a multiplex pathogen panel with ≥6 pathogens. During the late pandemic, 545,696 (19.1%) had any respiratory viral test, 361,597 (12.7%) had multiplex viral testing, and 91,625 (3.2%) had a multiplex pathogen panel with ≥6 pathogens.
When evaluating variation in testing within included hospitals, the increase in testing during both the early and late pandemic was observed among all sites. During the prepandemic period, the median percentage of encounters tested by the hospital was 5.2% (IQR 2.8% to 8.9%). During the early pandemic, the median percentage of encounters that received viral testing was 24.3% (IQR 21.0% to 30.0%). During the late pandemic, the median percentage of encounters that received viral testing was 18.8% (95% IQR 13.7% to 26.3%; Figure E1, available online at http://www.annemergmed.com).
Graphs for the pattern of viral testing before and during the COVID-19 pandemic are displayed in Figure 1. There was a single peak of viral testing each year, apart from the effects of the pandemic, corresponding to the respiratory viral season. Following the onset of the pandemic (early pandemic phase), there was a steep increase in the proportion of ED encounters in which any respiratory viral testing was conducted, with irregular or bimodal peaks. During the late pandemic, testing decreased but remained above prepandemic levels.
Figure 1.
Monthly utilization of respiratory viral testing in the primary analysis for A, all viral testing and B, multiplex viral testing. The dashed lines indicate the onset of the early (March 15, 2020) and late (January 1, 2023) pandemic.
In the interrupted time series model for all ED encounters, the prepandemic slope for viral testing was 0.17% encounters per month (95% CI 0.17 to 0.18; Table 2), suggesting that viral testing was already increasing each month before the onset of the pandemic. At the onset of the early pandemic, there was an initial rise in testing (shift change of 4.98%; 95% CI 4.90 to 5.07) and a positive slope change (increase in slope to 0.54% encounters per month; 95% CI 0.54 to 0.55). During the late pandemic, there was a negative shift change in viral testing (−17.80%; 95% CI −17.90 to −17.70), and the slope was attenuated to 0.42% encounters per month (95% CI 0.41 to 0.42).
Table 2.
Results of interrupted time series analysis to assess changes in viral testing before and after the COVID-19 pandemic (early and late periods). Results are expressed as a percentage of encounters, with numbers in parenthesis representing 95% confidence intervals.
| Sample | Outcome | Prepandemic Slope (Monthly Change) | Early Pandemic (March 15, 2020) |
Late Pandemic (January 1, 2023) |
||
|---|---|---|---|---|---|---|
| Slope (Monthly Change) | Shift Change | Slope (Monthly Change) | Shift Change | |||
| All ED encounters |
Any viral test | 0.17 (0.17, 0.18) | 0.54 (0.54, 0.55) | 4.98 (4.90, 5.07) | 0.42 (0.41, 0.42) | -17.80 (−17.90, −17.70) |
| Multiplex respiratory viral testing | ||||||
| All ED encounters |
Multiplex viral assay | 0.05 (0.05, 0.05) | 0.60 (0.60, 0.60) | −3.29 (−3.35, −3.23) | 0.48 (0.47, 0.48) | −10.17, (−10.24, −10.10) |
| All ED encounters |
Multiplex viral assay with ≥6 viruses | 0.02 (0.02, 0.02) | 0.07 (0.07, 0.07) | −0.47 (−0.50, −0.44) | 0.20 (0.20, 0.21) | −1.28 (−1.33, −1.25) |
| Respiratory infection subgroups | ||||||
| Suspected respiratory tract infection |
Any viral test | 0.44 (0.43, 0.45) | 0.82 (0.80, 0.83) | 12.77 (12.32, 13.21) | 0.47 (0.43, 0.51) | −22.55 (−22.01, −21.08) |
| Bronchiolitis | Any viral test | 0.36 (0.35, 0.38) | 0.85 (0.82, 0.89) | 10.72 (9.74, 11.71) | 0.21 (0.13, 0.30) | −17.63 (−18.59, −16.68) |
| Age-groups | ||||||
| 0–1 years | Any viral test | 0.26 (0.26, 0.27) | 0.70 (0.69, 0.70) | 5.21 (5.02, 5.40) | 0.51 (0.49, 0.53) | −19.90 (−20.12, −19.67) |
| 2–5 years | Any viral test | 0.20 (0.20, 0.20) | 0.69 (0.68, 0.70) | 2.89 (2.71, 3.08) | 0.43 (0.41, 0.44) | −20.45 (−20.67, −20.24) |
| 6–17 years | Any viral test | 0.12 (0.12, 0.12) | 0.41 (0.40, 0.41) | 5.90 (5.79, 6.01) | 0.35 (0.34, 0.36) | −15.00 (−15.12, −14.87) |
ED, emergency department.
Sensitivity Analyses
Findings from the interrupted time series analysis with wash-out periods of 2 and 4 months were essentially the same (Table E3, available online at http://www.annemergmed.com). With respect to multiplex viral testing ≥3 tests, the prepandemic trend was positive (0.05% encounters per month; 95% CI 0.05 to 0.05). There was an immediate decrease at the onset of the early pandemic (−3.29% encounters; 95% CI −3.35 to −3.23) with a positive slope (0.60% per month; 95% CI 0.60 to 0.60). During the late pandemic, there was a negative shift change (−10.17% encounters; 95% CI −10.24 to −10.10), with a positive slope (0.48% encounters per month; 95% CI 0.47 to 0.48). Findings were similar when using ≥6 pathogen cutoff. When evaluating trends of respiratory virus testing by the presence of respiratory disease and by age group, findings were similar to the main interrupted time series analysis (Figure E2, available online at http://www.annemergmed.com).
Cost
Over the 90-month study period, the cost of respiratory viral testing was $12.1 million per month (2023 dollars). For the 3 study periods (pre-, early, and late pandemic), the average monthly costs associated with respiratory viral testing were $4.6 million, $16.3 million, and $22.7 million, respectively. Patterns of cost variation were generally similar to those observed for respiratory testing (Figure 2 and Table 3). The final interrupted time series model indicated that there was a small positive slope in the costs associated with respiratory testing ($5,000 per month; 95% CI $4,200 to $5,700). At the onset of the pandemic, there was a shift toward decreased costs associated with respiratory testing (-$250,000; 95% CI -$270,000 to -$220,000), with a large increase in slope ($33,000 per month; 95% CI $32,000 to $34,000). The late pandemic was associated with a negative shift change (-$580,000; 95% CI -$620,000 to -$540,000) and a continued positive upward slope ($27,000; 95% CI $24,000 to $30,000). A similar finding was noted when using an outcome of multiplex viral testing (for both ≥3 or ≥6 tests) and in subgroups of infections of children based on diagnosis and age group, with an increase in slopes during the pandemic (Figure E3, available online at http://www.annemergmed.com).
Figure 2.
Costs associated with respiratory viral testing in the primary analysis for A, all viral testing B, and multiplex viral testing. The dashed lines indicate the onset of the early (March 15, 2020) and late (January 1, 2023) pandemic.
Table 3.
Results of interrupted time series analysis for respiratory viral testing before and during the COVID-19 pandemic (early and late periods). Results are expressed in US dollars (adjusted to 2023 dollars using the consumer price index), with numbers in parentheses representing 95% confidence intervals.
| Early Pandemic (March 15, 2020) |
Late Pandemic (January 1, 2023) |
|||||
|---|---|---|---|---|---|---|
| Sample | Outcome | Prepandemic Slope (Monthly Change) | Slope (monthly change) | Shift change | Slope (monthly change) | Shift change |
| All ED encounters |
Any viral test | 5,000 (4,200, 5,700) | 33,000 (32,000, 34,000) | −250,000, (−270,000, −220,000) | 27,000 (24,000, 30,000) | −580,000 (−620,000, −540,000) |
| Multiplex respiratory pathogen panel | ||||||
| All ED encounters |
Multiplex viral assay | 2,800 (2,100, 3,400) | 23,000 (22,000, 24,000) | −200,000 (−220,000, −180,000) | 31,000 (28,000, 34,000) | −370,000 (−410,000, −340,000) |
| All ED encounters |
Multiplex viral assay with ≥6 viruses | 1,600 (1,400, 1,800) | 3,300 (3,000, 3,600) | −59,000 (−66,000, −51,000) | 15,000 (14,000, 16,000) | −55,000 (−65,000, −45,000) |
| Respiratory infection subgroups | ||||||
| Suspected respiratory tract infection |
Any viral test | 3,000 (2,500, 3,500) | 19,000 (18,000, 19,000) | 190,000 (−200,000, −170,000) | 23,000 (20,000 25,000) | −380,000 (−400,000, −350,000) |
| Bronchiolitis | Any viral test | 390 (200, 480) | 2,400 (2,300, 2,500) | −29,000 (−32,000, −26,000) | 2,900 (2,500, 3,300) | −48,000 (−52,000, −43,000) |
| Age-stratified models | ||||||
| 0–1 years | Any viral test | 1,900, (1,700, 2,200) | 11,000 (11,000, 11,000) | −86,000 (−95,000, −78,000) | 9,800 (8,800, 11,000) | −200,000 (−210,000, −180,000) |
| 2–5 years | Any viral test | 1,300 (1,100, 1,500) | 9,800 (9,500, 10,000) | −80,000 (−87,000, −73,000) | 5,000 (4,300, 6,100) | −170,000 (−170,000, −160,000) |
| 6–17 years | Any viral test | 1,700 (1,400, 2,000) | 13,000 (12,000, 13,000) | −85,000 (−98,000, −73,000) | 12,000 (10,000, 13,000) | −220,000 (−240,000, −210,000) |
LIMITATIONS
Our study findings should be interpreted in the light of several limitations. First, this study was performed using administrative data, which is often subject to substantial inaccuracies in documentation and/or coding. For example, some of the encounters may have had a respiratory viral test that was not billed and, therefore, not included in PHIS. Second, due to the nature of PHIS, we are unable to comment on the appropriateness of testing for individual patients or to assess the performance of viral testing and specific clinical outcomes. Third, we used primary and secondary diagnosis codes to identify diagnoses of interest. This approach maximizes sensitivity for identifying encounters that have a given diagnosis, potentially at the cost of specificity. Fourth, our evaluation of multiplex viral respiratory testing may be limited by the accuracy or number of the laboratory codes used by PHIS. We accounted for this by using a second definition for multiplex viral testing, which was more specific for this outcome (≥6). Finally, as we only evaluated practices in the EDs of children’s hospitals, our findings may not generalize to other settings, especially nonpediatric EDs.
DISCUSSION
Among 34 children’s hospital EDs included in an administrative database, there was a substantial and sustained rise in respiratory viral testing among children discharged from the ED. The increase in viral testing started at the onset of the COVID-19 pandemic and was sustained into the late pandemic, albeit at a lower level. Similar findings were noted among children receiving multiplex viral testing and in subgroups of children with respiratory infections and by patient age. There was a corresponding and sustained increase in the costs attributed to respiratory viral testing. Increases in viral testing and associated costs were sustained despite wide variation in COVID-19 incidence.
At the onset of the pandemic, we identified a rise in the use of respiratory testing despite the lack of a diagnostic test for COVID-19 at the start of the pandemic and a decrease in respiratory infections due to stay-at-home orders and lockdowns.27 The changes attributable to the onset of the pandemic in March 2020 are likely because of necessary changes in clinical practice and national guidelines and recommendations. Guidance for COVID-19 testing in the United States from the CDC was first published in June 2020 and focused on nursing homes, long-term care facilities, and high-density critical infrastructural workplaces.24 Current guidance from the CDC recommends that all patients admitted to the hospital require a COVID-19 and influenza test if they have symptoms consistent with an acute respiratory viral infection.17,18 Specific testing for COVID-19 in the United States demonstrated the largest increases over the first several months of the pandemic (particularly during the latter months of 2020),24 though its availability was not uniform, with differences based on geography and demographic factors.32,33
Our findings of an increase in viral testing during the late pandemic represent a substantial deviation from prepandemic testing recommendations. Before the pandemic, national guidelines31 and expert opinion7,11 recommended against routine testing for children with respiratory viruses. During the early pandemic, testing was frequently mandated, even if unnecessary, to guide clinical management. Particularly for younger children, testing could not be done in other ambulatory settings outside of the ED. During parts of the pandemic, testing was required for children to attend school and daycares, as well as for travel, in order to mitigate the spread of the SARS-CoV-2 virus.34 These issues may have accounted for some of the differences noted between the early and late pandemic when testing requirements started to change and testing availability outside of the ED was broader.35 Other factors driving testing beyond clinical indication included the need for cohorting within the ED or anticipatory testing for children for whom admission was considered or who were receiving a procedure.36
Previous work has suggested that, in most contexts, viral testing contributes minimally toward guiding clinical management. One large, randomized control trial of 1,243 children with acute respiratory illnesses prior to the onset of the pandemic demonstrated that physician knowledge of multiplex testing results did not influence decisions to prescribe antibiotics or order other diagnostic tests.37 Similar findings have also been noted in one systematic review11 and another large study conducted in Europe.10 Some recent work has suggested a role for viral testing results in reducing downstream clinical testing and empirical treatments for some patient subgroups, such as in improving the risk stratification of febrile young infants.38,39 Other investigators have postulated the role of viral testing in screening for outbreaks.40 The increase in testing provided greater detail with respect to the cause of bronchiolitis using health system data,41 providing greater detail than previously reported through prospective observational research.
Our finding of a sustained increase in respiratory viral testing may also indicate a general change in practice and expectations, both among clinicians and caregivers, regarding the value of viral testing in otherwise healthy children with respiratory illness. Patient and caregiver factors may include both a heightened awareness of and vigilance for common respiratory illnesses, especially given the symptom overlap between COVID-19 and other viral infections.42,43 Clinician expectations may include a desire to minimize further evaluation or gain specific information to guide parental counseling and address parental anxiety.7 One single-center qualitative study published before the pandemic, which included hospitalists and emergency physicians, identified several reasons why clinicians may choose not to obtain a viral test, including perceptions that such testing may not change management and minimize cost and patient discomfort.44 Results from the study indicate that respiratory viral testing is used to validate clinical suspicions and alleviate self-doubt. The study found that tests may also be performed when they are believed to potentially alter medical management or enhance the perception that the clinician is taking proactive measures.
Our findings with respect to multiplex viral testing are notable in the context of the Choosing Wisely priorities for Pediatric Emergency Medicine published in December 2022.45 These priorities, which were codeveloped by the American Academy of Pediatrics Section on Emergency Medicine and the Canadian Association of Emergency Physicians, specifically recommend against multiplex viral testing for otherwise healthy children with a suspected common respiratory illness.45,46 The Choosing Wisely priorities may have influenced our study findings during the late pandemic when testing rates declined.
Given that rates of viral respiratory testing have increased above prepandemic levels, concerted efforts will likely be needed to promote the judicious use of viral testing, especially given the relatively low morbidity of COVID-19 and most other viral respiratory infections in children.47,48 Although the COVID-19 pandemic was a large driver in performing more respiratory tests, even prior to the pandemic, rates of viral testing for suspected bronchiolitis were higher than advised by national recommendations.12,13 Campaigns such as Choosing Wisely may improve awareness both among clinicians and caregivers of situations where viral testing is not warranted.45 One mixed-methods study that focused on improving deimplementation efforts in the ED identified that high-fidelity, multidimensional interventions that include education, partner engagement, audit and feedback, and clinical decision support when applied daily and sustained over time are most effective in discontinuing the use of low-value care in ED.49
In conclusion, this study of a large sample of pediatric patients discharged from the ED of United States children’s hospitals revealed a marked and sustained increase in the utilization and cost of respiratory viral testing since the start of the COVID-19 pandemic. This escalation persisted despite variation in disease prevalence and was similar by patient age and for multiplex testing. The sustained increase in viral testing reflects a further departure from prepandemic recommendations.
Supplementary Material
Editor’s Capsule Summary.
What is already known on this topic
Low-value care in children in the emergency department (ED) is a persistent problem, especially viral respiratory testing.
What question this study addressed
Did the COVID-19 pandemic and associated recommendations on viral testing patterns change diagnostic studies in children?
What this study adds to our knowledge
In an interrupted time series analysis of a national, multiyear sample of children in EDs, viral testing and associated costs more than quadrupled and stayed increased despite wide variation in the incidence of COVID-19.
How this is relevant to clinical practice
The COVID-19 pandemic exacerbated observations of low-value ED viral testing in children with respiratory infections.
Funding and support:
By Annals’ policy, all authors are required to disclose any and all commercial, financial, and other relationships in any way related to the subject of this article as per ICMJE conflict of interest guidelines (see www.icmje.org). The authors have stated that no such relationships exist. This study was supported by funding from the Stanley Manne Children’s Research Institute and by Pediatric Pandemic Network resources. The Pediatric Pandemic Network is supported by the Health Resources and Services Administration (HRSA) of the United States Department of Health and Human Services (HHS) as part of grant awards U1IMC43532 and U1IMC45814 with 0 percent financed with nongovernmental sources. The content presented here is that of the authors and does not necessarily represent the official views of, nor an endorsement by, HRSA. HHS, or the United State Government. For more information, visit HRSA.gov.
Footnotes
Supervising editor: Benjamin T. Kerrey, MD, MS. Specific detailed information about possible conflict of interest for individual editors is available at https://www.annemergmed.com/editors.
Data sharing statement:
The data sets generated and/or analyzed during the current study are not publicly available. Individuals who are interested in accessing the data may contact the Children’s Hospital Association.
REFERENCES
- 1.Ramgopal S, Cotter JM, Navanandan N, et al. Viral detection is associated with severe disease in children with suspected community-acquired pneumonia. Pediatr Emerg Care. 2023;1097:10–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Bradley JS, Byington CL, Shah SS, et al. Pediatric Infectious Diseases Society and the 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:e25–e76. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Burstein B, Dubrovsky AS, Greene AW, Quach C. National survey on the impact of viral testing for the ED and inpatient management of febrile young infants. Hosp Pediatr. 2016;6:226–233. [DOI] [PubMed] [Google Scholar]
- 4.Lacroix S, Vrignaud B, Avril E, et al. Impact of rapid influenza diagnostic test on physician estimation of viral infection probability in paediatric emergency department during epidemic period. J Clin Virol. 2015;72:141–145. [DOI] [PubMed] [Google Scholar]
- 5.Bonner AB, Monroe KW, Talley LI, Klasner AE, Kimberlin DW. Impact of the rapid diagnosis of influenza on physician decision-making and patient management in the pediatric emergency department: results of a randomized, prospective, controlled trial. Pediatrics. 2003;112:363–367. [DOI] [PubMed] [Google Scholar]
- 6.Rogan DT, Kochar MS, Yang S, Quinn JV. Impact of rapid molecular respiratory virus testing on real-time decision making in a pediatric emergency department. J Mol Diagn. 2017;19:460–467. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Gill PJ, Richardson SE, Ostrow O, Friedman JN. Testing for respiratory viruses in children: to swab or not to swab. JAMA Pediatr. 2017;171:798–804. [DOI] [PubMed] [Google Scholar]
- 8.Rao S, Lamb MM, Moss A, et al. Effect of rapid respiratory virus testing on antibiotic prescribing among children presenting to the emergency department with acute respiratory illness: A randomized clinical trial. JAMA Netw Open. June 1 2021;4:e2111836. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Schnell J, Schroeder L, Sinclair K, Patel L, Dowd D. The effect of early knowledge of respiratory syncytial virus positivity on medical decision making and throughput time within the pediatric emergency department. Pediatr Emerg Care. 2020;36:134–137. [DOI] [PubMed] [Google Scholar]
- 10.Tan CD, Hagedoorn NN, Dewez JE, et al. Rapid viral testing and antibiotic prescription in febrile children with respiratory symptoms visiting emergency departments in Europe. Pediatr Infect Dis J. 2022;41:39–44. [DOI] [PubMed] [Google Scholar]
- 11.Doan Q, Enarson P, Kissoon N, Klassen TP, Johnson DW. Cochrane Review: rapid viral diagnosis for acute febrile respiratory illness in children in the Emergency Department. Evid Based Child Health. 2010;5:709–751. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Emerson BL, Tenore C, Grossman M. An initiative to reduce routine viral diagnostic testing in pediatric patients admitted with bronchiolitis. Jt Comm J Qual Patient Saf. 2018;44:751–756. [DOI] [PubMed] [Google Scholar]
- 13.Mussman GM, Lossius M, Wasif F, et al. Multisite emergency department inpatient collaborative to reduce unnecessary bronchiolitis care. Pediatrics. 2018;141:141–142. [DOI] [PubMed] [Google Scholar]
- 14.Rivera-Sepulveda AV, Rebmann T, Gerard J, Charney RL. Physician compliance with bronchiolitis guidelines in pediatric emergency departments. Clin Pediatr. 2019;58(9):1008–1018. [DOI] [PubMed] [Google Scholar]
- 15.Shapiro DJ, Thurm CW, Hall M, et al. Respiratory virus testing and clinical outcomes among children hospitalized with pneumonia. J Hosp Med. September 2022;17:693–701. [DOI] [PubMed] [Google Scholar]
- 16.Ambroggio L, Cotter J, Hall M, et al. Management of pediatric pneumonia: A decade after the Pediatric Infectious Diseases Society and Infectious Diseases Society of America guideline. Clin Infect Dis. November 30 2023;77:1604–1611. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Centers for Disease Control and Prevention. Clinical guidance for hospitalized and non-hospitalized patients when SARS-CoV-2 and influenza viruses are co-circulating. Accessed May 3, 2024. https://www.cdc.gov/flu/professionals/diagnosis/testing-guidance-for-clinicians.htm
- 18.Centers for Disease Control and Prevention. Overview of Testing for SARS-CoV-2, the virus that causes COVID-19. Accessed December 31, 2023. https://www.cdc.gov/coronavirus/2019-ncov/hcp/testing-overview.html
- 19.Von Elm E, Altman DG, Egger M, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet. 2007;370:1453–1457. [DOI] [PubMed] [Google Scholar]
- 20.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. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Pasquali SK, Chiswell K, Hall M, et al. Estimating resource utilization in congenital heart surgery. Ann Thorac Surg. September 2020;110: 962–968. 10.1016/j.athoracsur.2020.01.013 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.US Bureau of Labor Statistics. Measuring price change in the CPI: medical care. Accessed January 8, 2024. https://www.bls.gov/cpi/factsheets/medical-care.htm
- 23.Russell H, Hall M, Morse RB, et al. Longitudinal trends in costs for hospitalizations at children’s hospitals. Hosp Pediatr. 2020;10:797–801. [DOI] [PubMed] [Google Scholar]
- 24.Centers for Disease Control and Prevention. CDC museum COVID-19 timeline. Accessed May 30, 2024. https://www.cdc.gov/museum/timeline/covid19.html
- 25.Madad S, Salway RJ, Raggi J, Silvestri D, Cotter T, Romolt C. The 2022–2023 USA respiratory viral “tripledemic”: healthcare lessons learned. Microbiol Infect Dis AMJ. 2023;1:35–39. [Google Scholar]
- 26.DeLaroche AM, Rodean J, Aronson PL, et al. Pediatric emergency department visits at US children’s hospitals during the COVID-19 pandemic. Pediatrics. 2021;147:e2020039628-e2020039628. [DOI] [PubMed] [Google Scholar]
- 27.Ramgopal S, Pelletier JH, Rakkar J, Horvat CM. Forecast modeling to identify changes in pediatric emergency department utilization during the COVID-19 pandemic. Am J Emerg Med. 2021;49:142–147. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Haddadin Z, Blozinski A, Fernandez K, et al. Changes in pediatric emergency department visits during the COVID-19 pandemic. Hosp Pediatr. 2021;11:e57–e60. [DOI] [PubMed] [Google Scholar]
- 29.Cox NJ. Speaking Stata: in praise of trigonometric predictors. The Stata Journal. 2006;6:561–579. [Google Scholar]
- 30.Pandemic oversight. Federal COVID-19 Testing Report. Data Insights from Six Federal Health Care Programs. Accessed May 24, 2024. https://www.pandemicoversight.gov/oversight/our-publications-reports/federal-covid-19-testing-report
- 31.Ralston SL, Lieberthal AS, Meissner HC, et al. Clinical practice guideline: the diagnosis, management, and prevention of bronchiolitis. Pediatrics. 2014;134:e1474–e1502. [DOI] [PubMed] [Google Scholar]
- 32.Rentsch CT, Kidwai-Khan F, Tate JP, et al. Patterns of COVID-19 testing and mortality by race and ethnicity among United States veterans: A nationwide cohort study. PLOS Med. 2020;17:e1003379. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Souch JM, Cossman JS. A commentary on rural-urban disparities in COVID-19 testing rates per 100,000 and risk factors. J Rural Health. 2021;37:188–190. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Moghadas SM, Fitzpatrick MC, Shoukat A, Zhang K, Galvani AP. Simulated identification of silent COVID-19 infections among children and estimated future infection rates with vaccination. JAMA Netw Open. 2021;4:e217097. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Rader B, Gertz A, Iuliano AD, et al. Use of at-home Covid-19 tests—united states, August 23, 2021-March 12, 2022. MMWR Morb Mortal Wkly Rep. 2022;71:489–494. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Wake RM, Morgan M, Choi J, Winn S. Reducing nosocomial transmission of COVID-19: implementation of a COVID-19 triage system. Clin Med (Lond). 2020;20:e141–e145. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Mattila S, Paalanne N, Honkila M, Pokka T, Tapiainen T. Effect of point-of-care testing for respiratory pathogens on antibiotic use in children: A randomized clinical trial. JAMA Netw Open. June 1 2022;5:e2216162. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Aronson PL, Louie JP, Kerns E, et al. Prevalence of urinary tract infection, bacteremia, and meningitis among febrile infants aged 8 to 60 days with SARS-CoV-2. JAMA Netw Open. 2023;6:e2313354. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Blaschke AJ, Korgenski EK, Wilkes J, et al. Rhinovirus in febrile infants and risk of bacterial infection. Pediatrics. 2018;141. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Zven S, Dorr M, Malloy AMW, Susi A, Nylund CM, Hisle-Gorman E. Predicting the RSV surge: pediatric RSV patterns of the COVID pandemic. Pediatr Infect Dis J. 2023;42:e349–e351. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Movva N, Suh M, Reichert H, et al. Respiratory syncytial virus during the COVID-19 pandemic compared to historic levels: a retrospective cohort study of a health system. J Infect Dis. 2022;226:S175–S183. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Watson G, Pickard L, Williams B, Hargreaves D, Blair M. ‘Do I, don’t I?’ A qualitative study addressing parental perceptions about seeking healthcare during the COVID-19 pandemic. Arch Dis Child. 2021;106:1118–1124. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Mowbray F, Woodland L, Smith LE, Amlôt R, Rubin GJ. Is my cough a cold or Covid? A qualitative study of COVID-19 symptom recognition and attitudes toward testing in the UK. Front Public Health. 2021;9: 716421. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Huang MZ, Rhee KE, Gist L, Fisher ES. Barriers to minimizing respiratory viral testing in bronchiolitis: physician perceptions on testing practices. Hosp Pediatr. 2019;9:79–86. [DOI] [PubMed] [Google Scholar]
- 45.American Academy of Pediatrics. Choosing wisely: five things physicians and patients should question in the practice of pediatric emergency medicine. Accessed December 31, 2023. https://www.aap.org/en/news-room/news-releases/aap/2022/choosing-wisely-five-things-physicians-and-patients-should-question-in-the-practice-of-pediatric-emergency-medicine/
- 46.Mullan PC, Levasseur KA, Bajaj L, et al. Recommendations for choosing wisely in pediatric emergency medicine: five opportunities to improve value. Ann Emerg Med. 2024;84:167–175. [DOI] [PubMed] [Google Scholar]
- 47.Jank M, Oechsle A-L, Armann J, et al. Comparing SARS-CoV-2 variants among children and adolescents in Germany: relative risk of COVID-19-related hospitalization, ICU admission and mortality. Infection. 2023;51:1357–1367. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Ward JL, Harwood R, Kenny S, et al. Pediatric hospitalizations and ICU admissions due to COVID-19 and pediatric inflammatory multisystem syndrome temporally associated with SARS-CoV-2 in England. JAMA Pediatr. 2023;177:947–955. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Gangathimmaiah V, Drever N, Evans R, et al. What works for and what hinders deimplementation of low-value care in emergency medicine practice? A scoping review. BMJ Open. 2023;13:e072762. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data sets generated and/or analyzed during the current study are not publicly available. Individuals who are interested in accessing the data may contact the Children’s Hospital Association.