Abstract
This study uses Pediatric Health Information System database data to compare hospitalizations in US children’s hospitals early in the COVID-19 pandemic (March-August 2020) vs the same period in 2017-2019, overall and for respiratory, chronic, nonrespiratory, and other conditions.
Measures to mitigate the COVID-19 pandemic affected children’s access to health services and their physical and mental health. Reductions in hospitalizations for children occurred in 2020 compared with prior years.1 Little is known about the reasons for the decline and whether it varied by patient characteristics.
Children’s hospitals provide inpatient care for the most diverse, high-severity, and complex illnesses2 and are located in large urban areas, which were particularly affected by COVID-19 outbreaks. Centralization of pediatric inpatient care into children’s hospitals was urged to free beds in non–children’s hospitals for adult COVID-19 patients. We compared hospitalizations in US children’s hospitals before and during the pandemic.
Methods
We conducted a retrospective study of admissions for children aged 0 to 18 years in 42 US freestanding children’s hospitals in the Pediatric Health Information System database. Admissions were categorized by spring (March 15 to May 31) and summer (June 1 to August 31). Using quantile regression, the median weekly volumes of admissions during spring and summer were compared between 2017-2019 and 2020 for all-cause admissions and the most common condition-specific reasons for admission.3 We also assessed patients’ demographic and clinical characteristics. Analyses were performed with SAS version 9.4 (SAS Institute Inc). To account for multiple comparisons, Sidak correction was applied to all P values. The threshold for statistical significance was a 2-sided P < .05. The study was approved by the Boston Children’s Hospital Institutional Review Board with a waiver of consent.
Results
There were 1 699 911 admissions included in the study; 54% were male. Weekly all-cause hospitalizations decreased in the spring from a median of 12 830 (interquartile range [IQR], 12 468-13 095) in 2017-2019 to 7033 (IQR, 6187-8231) in 2020 (48.3% [95% CI, 38.6%-58.0%] decrease) and in the summer from a median of 11 697 (IQR, 11 495-11 947) in 2017-2019 to 9178 (IQR, 8739-9358) in 2020 (23.5% [95% CI, 19.2%-27.8%] decrease) (P < .001 for both). The least percentage change per week was in summer 2020 among adolescents aged 15 to 18 years (−9.6% [95% CI, −17.5% to −1.8%]; P = .47) and the greatest change was in spring 2020 among those with the lowest illness severity (−56.4% [95% CI, −65.1% to −47.7%]; P < .001) (Table 1). Decreases in hospitalizations occurred in all demographic and clinical subgroups.
Table 1. Weekly Number of Admissions to Children’s Hospitals Before and During the COVID-19 Pandemic by Patient Characteristics.
| Characteristics | Spring (March 15 to May 31) | Summer (June 1 to August 31) | ||||
|---|---|---|---|---|---|---|
| No. of admissions per wk, median (IQR) | Adjusted change, % (95% CI)a | No. of admissions per wk, median (IQR) | Adjusted change, % (95% CI)a | |||
| 2017-2019 | 2020 | 2017-2019 | 2020 | |||
| All admissions | 12 830 (12 468-13 095) | 7033 (6187-8231) | −48.3 (−58.0 to −38.6) | 11 697 (11 495-11 947) | 9178 (8739-9358) | −23.5 (−27.8 to −19.2) |
| Age at admission, y | ||||||
| <1 | 2990 (2796-3070) | 1961 (1840-2081) | −37.1 (−44.9 to −29.3) | 2723 (2616-2784) | 2003 (1934-2129) | −27.8 (−32.8 to −22.8) |
| 1-4 | 3355 (3203-3469) | 1452 (1248-1744) | −64.3 (−73.9 to −54.8) | 2840 (2757-2966) | 1964 (1858-2008) | −35.0 (−41.2 to −28.8) |
| 5-9 | 2344 (2268-2421) | 1089 (979-1389) | −52.5 (−64.7 to −40.3) | 2115 (2029-2192) | 1552 (1516-1620) | −27.0 (−33.8 to −20.1) |
| 10-14 | 2453 (2377-2532) | 1435 (1186-1664) | −41.1 (−53 to −29.1) | 2339 (2281-2375) | 2004 (1836-2074) | −17.2 (−23.4 to −11.0) |
| 15-18 | 1650 (1613-1714) | 1096 (916-1238) | −36.1 (−48.6 to −23.5) | 1674 (1627-1726) | 1597 (1440-1649) | −9.6 (−17.5 to −1.8) |
| Sex | ||||||
| Male | 6903 (6703-7067) | 3692 (3290-4319) | −49.9 (−59 to −40.7) | 6284 (6212-6445) | 4809 (4661-4932) | −26.5 (−31.5 to −21.4) |
| Female | 5937 (5748-6054) | 3339 (2900-3863) | −46.5 (−56.2 to −36.8) | 5364 (5259-5491) | 4336 (4075-4498) | −22.6 (−27.6 to −17.7) |
| Payer | ||||||
| Public | 6965 (6817-7133) | 3673 (3260-4269) | −50.3 (−60.7 to −39.9) | 6201 (6082-6383) | 4741 (4533-4872) | −26.2 (−31.4 to −21.0) |
| Private | 5006 (4897-5122) | 2742 (2386-3287) | −46.3 (−57.1 to −35.5) | 4681 (4515-4717) | 3607 (3455-3670) | −24.6 (−28.4 to −20.8) |
| Race/ethnicityb | ||||||
| Non-Hispanic White | 6323 (6104-6463) | 3539 (3086-4229) | −45.2 (−56.1 to −34.3) | 5804 (5716-5940) | 4677 (4427-4806) | −20.5 (−24.7 to −16.4) |
| Non-Hispanic Black | 2517 (2365-2596) | 1281 (1166-1496) | −53.9 (−63.4 to −44.4) | 2193 (2117-2287) | 1693 (1662-1725) | −26.9 (−32.1 to −21.6) |
| Hispanic | 2496 (2442-2536) | 1369 (1187-1540) | −48.7 (−58.6 to −38.8) | 2257 (2169-2309) | 1735 (1641-1755) | −27.7 (−33.1 to −22.2) |
| Asian | 359 (337-379) | 190 (173-216) | −52.4 (−64.7 to −40) | 337 (316-351) | 240 (224-253) | −35.9 (−43.7 to −28.1) |
| Other | 1125 (1076-1202) | 653 (583-707) | −47.1 (−57.5 to −36.8) | 1089 (1040-1125) | 817 (750-856) | −31.2 (−37.7 to −24.8) |
| Hospital region | ||||||
| Midwest | 3360 (3269-3440) | 1657 (1480-1990) | −54 (−64.1 to −43.8) | 3025 (2888-3116) | 2279 (2181-2343) | −22.0 (−28.6 to −15.3) |
| Northeast | 1754 (1687-1806) | 851 (751-1100) | −51.9 (−65.6 to −38.1) | 1582 (1517-1613) | 1289 (1189-1337) | −19.8 (−25.8 to −13.8) |
| South | 5137 (4978-5320) | 2899 (2542-3388) | −50.1 (−60.2 to −39.9) | 4785 (4597-4910) | 3703 (3506-3768) | −28.5 (−33.4 to −23.5) |
| West | 2546 (2419-2605) | 1642 (1443-1749) | −35.5 (−45.3 to −25.7) | 2322 (2242-2405) | 1897 (1851-1944) | −25.2 (−29.3 to −21.1) |
| Severity of illnessc | ||||||
| 1 (Lowest) | 3310 (3097-3427) | 1393 (1261-1555) | −56.4 (−65.1 to −47.7) | 2661 (2574-2789) | 1740 (1695-1808) | −36.1 (−43.3 to −28.9) |
| 2 | 3249 (3153-3326) | 1662 (1417-1969) | −46.6 (−57.3 to −35.8) | 2915 (2831-2993) | 2278 (2175-2309) | −22.4 (−27.2 to −17.6) |
| 3 | 3137 (3008-3316) | 1930 (1674-2209) | −49.4 (−60.2 to −38.7) | 2910 (2848-3000) | 2479 (2406-2511) | −19.1 (−23.7 to −14.5) |
| 4 (Highest) | 3090 (2999-3169) | 2043 (1850-2457) | −37.6 (−49.4 to −25.8) | 3197 (3056-3253) | 2683 (2503-2818) | −17.1 (−22.9 to −11.4) |
| Chronic conditions, No. | ||||||
| 0 | 7666 (7348-7842) | 3934 (3398-4508) | −49.1 (−57.8 to −40.5) | 3233 (3131-3312) | 2386 (2349-2423) | −28.9 (−33.8 to −23.9) |
| 1 | 3027 (2965-3122) | 1776 (1597-2172) | −49 (−58.8 to −39.2) | 3450 (3387-3572) | 2658 (2497-2706) | −24.7 (−30.3 to −19.2) |
| 2-3 | 1831 (1771-1873) | 1140 (1049-1299) | −43.4 (−55.8 to −30.9) | 3430 (3403-3514) | 2879 (2672-2945) | −19.1 (−25.1 to −13.2) |
| ≥4 | 305 (287-328) | 196 (175-219) | −46.5 (−56.5 to −36.5) | 1552 (1495-1591) | 1274 (1167-1324) | −22.4 (−28.5 to −16.2) |
| Hospital transferd | ||||||
| No | 11 231 (10 856-11 461) | 5940 (5187-7087) | −49.1 (−59.3 to −38.9) | 6284 (6212-6445) | 4809 (4661-4932) | −23.6 (−28.0 to −19.3) |
| Yes | 1647 (1480-1736) | 1075 (993-1136) | −48.1 (−56.9 to −39.3) | 5364 (5259-5491) | 4336 (4075-4498) | −32.3 (−39.3 to −25.3) |
Abbreviation: IQR, interquartile range.
Percent changes in median volume of admissions per week, estimated from quantile regression with a fixed effect for year; P < .01 for all percent changes except for age group 15 to 18 years in summer 2020 (P = .47); P values were adjusted for multiple comparisons using the Sidak correction.
Assessed to determine whether trends in admissions varied by race and ethnicity before vs during the COVID-19 pandemic. Race and ethnicity were self-reported by patients and families using classification options defined by hospitals.
Severity of illness determined from Hospitalization Resource Intensity Score for Kids (HRISK) quartiles.
Indicates that a patient was transferred from another acute care hospital for admission to the children’s hospital in the study cohort.
The largest decrease in weekly condition-specific hospitalizations occurred in spring 2020 with respiratory failure (from a median of 296 [IQR, 253-464] in 2017-2019 to 87 [IQR, 85-134] in 2020; 167.7% [95% CI, 129.6%-205.9%] decrease; P < .001) (Table 2). Decreases also occurred in spring 2020 with nonrespiratory conditions, including cellulitis (from a median of 221 [IQR, 211-239] in 2017-2019 to 120 [IQR, 111-142] in 2020; 45.4% [95% CI, 32.6%-58.1%] decrease; P < .001) and epilepsy (from a median of 585 [IQR, 561-606] in 2017-2019 to 300 [IQR, 239-373] in 2020; 48.0% [95% CI, 31.1%-64.9%] decrease; P < .001). Although not significant, the least percentage change per week was in summer 2020 for diabetic ketoacidosis (from a median of 138 [IQR, 125-146] in 2017-2019 to 150 [IQR, 142-162]; change, −0.7% [95% CI, 12.0% to −13.7%]; P > .99). Suicide/intentional injury was the only hospitalization with a (nonsignificant) percentage increase in summer 2020 (from a median of 93 [IQR, 87-102] to 109 [IQR, 93-128]; change, 11.8% [95% CI, −10.8% to 34.5%]; P > .99).
Table 2. Weekly Number of Admissions for the 20 Most Common Conditions at Children’s Hospitals Before and During the COVID-19 Pandemic.
| Condition-specific admissions | Spring (March 15 to May 31) | Summer (June 1 to August 31) | ||||||
|---|---|---|---|---|---|---|---|---|
| No. of admissions per wk, median (IQR) | Adjusted change, % (95% CI)a | P value | No. of admissions per wk, median (IQR) | Adjusted change, % (95% CI)a | P value | |||
| 2017-2019 | 2020 | 2017-2019 | 2020 | |||||
| Respiratory conditions | ||||||||
| Viral infection | 122 (110-133) | 47 (39-60) | −70.8 (−90.3 to −51.3) | <.001 | 120 (104-131) | 62 (55-64) | −61.1 (−75.0 to −47.2) | <.001 |
| Respiratory failure | 296 (253-464) | 87 (85-136) | −167.7 (−205.9 to −129.6) | <.001 | 188 (148-225) | 94 (87-97) | −93.6 (−125.2 to −62) | <.001 |
| Pneumonia | 352 (304-405) | 71 (53-126) | −82.8 (−112.1 to 53.5) | <.001 | 192 (158-220) | 42 (40-47) | −84.1 (−108.6 to 59.7) | <.001 |
| Bronchiolitis | 456 (373-553) | 26 (15-121) | −121.8 (−159.8 to 83.9) | <.001 | 174 (145-218) | 24 (13-27) | −98.6 (−143.5 to −53.6) | .001 |
| Asthma | 577 (528-630) | 59 (51-73) | −79.4 (−91.6 to −67.1) | <.001 | 261 (219-341) | 80 (55-98) | −68.1 (−105.2 to −31.1) | .01 |
| Chronic conditions | ||||||||
| Diabetic ketoacidosis | 125 (114-135) | 132 (105-139) | 3.2 (−19.9 to 26.3) | >.99 | 138 (125-146) | 150 (142-162) | −0.7 (−13.5 to 12.0) | >.99 |
| Suicide/intentional injury | 139 (122-159) | 116 (112-122) | −14.4 (−36.1 to 7.3) | .98 | 93 (87-102) | 109 (93-128) | 11.8 (−10.8 to 34.5) | .99 |
| Major depressive disorder | 238 (215-253) | 128 (113-140) | −52.5 (−64.1 to −41) | <.001 | 156 (140-175) | 143 (132-167) | −20.6 (−45.6 to 4.5) | .89 |
| Epilepsy | 586 (561-606) | 300 (239-373) | −48 (−64.9 to −31.1) | <.001 | 596 (573-612) | 500 (471-525) | −14.8 (−22.3 to −7.3) | .004 |
| Sickle cell crisis | 97 (86-113) | 52 (43-58) | −50 (−72.3 to −27.7) | .001 | 84 (74-92) | 66 (54-77) | −36.3 (−57.1 to −15.5) | .02 |
| Nonrespiratory infections | ||||||||
| Appendicitis | 340 (322-350) | 296 (271-333) | −12.5 (−26.1 to 1.1) | .77 | 342 (322-352) | 351 (321-367) | −1.8 (−10.5 to 7) | >.99 |
| Septicemia | 136 (102-157) | 107 (99-114) | −65.4 (−87.7 to −43.1) | <.001 | 118 (101-134) | 116 (98-129) | −32.3 (−54.2 to −10.5) | .09 |
| Urinary tract infection | 145 (136-156) | 106 (89-112) | −26 (−42.4 to −9.5) | .05 | 152 (137-162) | 138 (129-145) | −14.5 (−28.7 to −0.2) | .62 |
| Cellulitis | 221 (211-239) | 120 (111-142) | −45.4 (−58.1 to −32.6) | <.001 | 250 (237-265) | 175 (155-184) | −25.9 (−34.8 to −16.9) | <.001 |
| Gastroenteritis | 459 (413-507) | 154 (138-179) | −81.8 (−101.7 to −61.8) | <.001 | 361 (329-391) | 222 (210-228) | −55 (−64.6 to −45.4) | <.001 |
| Other conditions | ||||||||
| Neoplasms and chemotherapy | 297 (285-313) | 263 (249-277) | −6.7 (−14 to 0.5) | .75 | 288 (280-298) | 259 (246-269) | −13.2 (−19.5 to −6.9) | .002 |
| Extremity fracture | 313 (268-340) | 231 (198-251) | −26.2 (−43.8 to −8.6) | .08 | 300 (281-312) | 249 (236-275) | −19.4 (−30.2 to −8.5) | .02 |
| Constipation | 122 (109-132) | 43 (29-63) | −67.9 (−95 to −40.8) | <.001 | 101 (92-117) | 84 (68-91) | −21.4 (−41.9 to −0.9) | .57 |
| Headache | 96 (83-103) | 27 (24-30) | −59.9 (−77.4 to −42.4) | <.001 | 62 (56-74) | 43 (34-47) | −44.4 (−71.3 to −17.4) | .03 |
| Hypertrophy of tonsils and adenoids | 211 (194-230) | 43 (20-103) | −69.8 (−98.6 to −41.1) | <.001 | 225 (206-243) | 157 (137-167) | −38.7 (−53.1 to −24.2) | <.001 |
Abbreviation: IQR, interquartile range.
Percent changes in median volume of admissions per week, estimated from quantile regression with a fixed effect for year; P values were adjusted for multiple comparisons using the Sidak correction.
Discussion
In spring and summer 2020, during the COVID-19 pandemic, all-cause admissions and many condition-specific admissions for acute and chronic health problems decreased in US children’s hospitals. These decreases exceeded those reported with prior initiatives to prevent hospitalizations.4 In addition to social distancing, potential reasons include more watchful waiting for children with symptoms and increased thresholds for emergency department and hospital care, especially for lower-severity illnesses.
Diabetic ketoacidosis and suicide hospitalizations did not significantly change. Increased diabetic ketoacidosis during the COVID-19 pandemic has been reported in Germany.5 Increased depression, anxiety, and suicidal ideation and planning in children have occurred during the COVID-19 pandemic.6
Limitations include the absence of data from non–children’s hospitals. Transfers of pediatric patients from non–children’s hospitals to children’s hospitals did not increase, suggesting that changes in the location of where children received care did not influence the results. Data on the resurgence of COVID-19 in the fall and winter of 2020 were not available.
Section Editor: Jody W. Zylke, MD, Deputy Editor.
References
- 1.Wilder JL, Parsons CR, Growdon AS, Toomey SL, Mansbach JM. Pediatric hospitalizations during the COVID-19 pandemic. Pediatrics. 2020;146(6):e2020005983. doi: 10.1542/peds.2020-005983 [DOI] [PubMed] [Google Scholar]
- 2.Berry JG, Hall M, Cohen E, et al. Hospitals’ diversity of diagnosis groups and associated costs of care. Pediatrics. 2021;147(3):e2020018101. doi: 10.1542/peds.2020-018101 [DOI] [PubMed] [Google Scholar]
- 3.Gill PJ, Anwar MR, Thavam T, Hall M, Rodean J, Mahant S. Pediatric clinical classification system for use in inpatient settings. JAMA Pediatr. Published online March 1, 2021. doi: 10.1001/jamapediatrics.2020.6007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Coller RJ, Nelson BB, Sklansky DJ, et al. Preventing hospitalizations in children with medical complexity: a systematic review. Pediatrics. 2014;134(6):e1628-e1647. doi: 10.1542/peds.2014-1956 [DOI] [PubMed] [Google Scholar]
- 5.Kamrath C, Mönkemöller K, Biester T, et al. Ketoacidosis in children and adolescents with newly diagnosed type 1 diabetes during the COVID-19 pandemic in Germany. JAMA. 2020;324(8):801-804. doi: 10.1001/jama.2020.13445 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Xie X, Xue Q, Zhou Y, et al. Mental health status among children in home confinement during the coronavirus disease 2019 outbreak in Hubei Province, China. JAMA Pediatr. 2020;174(9):898-900. doi: 10.1001/jamapediatrics.2020.1619 [DOI] [PMC free article] [PubMed] [Google Scholar]