Lack of Association Between Pandemic School Mode Policy and Pediatric Body Mass Index Percentile Over Time

Abstract

During the 2020–2021 academic year, schools across the country were closed for prolonged periods. Prior research suggests that children tend to gain more weight during times of extended school closures, such as summer vacation; however, little is known about the impact of school learning mode on changes. Thus, the aim of this study was to measure the association between school mode (in-person, hybrid, remote) and pediatric body mass index (BMI) percentile increases over time. In this longitudinal, statewide retrospective cohort study in Massachusetts, we found that BMI percentile increased in elementary and middle school students in all learning modes, and that increases slowed but did not reverse following the statewide reopening. Body mass percentile increases were highest in elementary school aged children. Hispanic ethnicity and receipt of Medicaid insurance were also associated with increases. Additional research is needed to identify strategies to combat pediatric body mass percentile increases and to address disparities.

Background

BMI percentile in children increases during periods of extended school closure, such as summer break.^1,2 Pediatric BMI percentile, which should remain stable for any given individual over time, increased during the early COVID-19 pandemic, particularly among elementary school aged children.^3–5 School closures are theorized to be a contributor;⁶ however, little data are available to evaluate the impact of school mode on observed increases and if changes reversed after schools returned to full-time in-person learning.

In Massachusetts, during the initial reopening period in fall 2021, school mode was determined at the district level and varied statewide. During the early reopening period, the majority of public schools opened in a hybrid or remote learning mode. Although in theory there was the possibility for different learning modes for elementary, middle, and high schools, in practice, learning mode choice was highly conserved within public school districts.⁷ In April 2021, a statewide policy mandated that all public schools open for full-time in-person learning.

The aim of this study was to measure the impact of school learning mode (in-person, remote, hybrid) on change in BMI percentile in a longitudinal cohort of pediatric patients, comparing measurements from the pre-pandemic period, the 2020–2021 when school modes were variable, and the 2021–2022 academic school years, when all schools were open for full-time in-person learning.

Methods

Data were obtained from the Reliant Medical Group database, a 500-provider, multispecialty group practice primarily located in Central Massachusetts. The database includes 14 pediatric and family practice sites that represent an ethnically and socioeconomically diverse population of children. As part of routine rooming procedures, all parents in the practice were asked to report school mode option during the 2021–2022 school year. Longitudinal data available included: age at the time of the clinical visit, BMI percentile, race and ethnicity collected from patients or parents during the new patient registration process, Medicaid status, school learning model, and type of school (public, private, home school). School age of the child was classified by age at the time of the visit (elementary school age: 5–11 years at the time of the visit, middle school age: 11–14 years at the time of the visit, high school age: older than 14 years at the time of the visit).

BMI percentiles were split into three periods (pre-COVID-19: from November 1, 2019, to April 30, 2020; initial school reopening: from November 1, 2020, to April 5, 2021; and post-full reopening: from November 1, 2021, to April 5, 2022) and per-person change was calculated. Individuals with at least one BMI measurement during each period and school mode and type available were included. Impact of school mode was evaluated with a difference-in-differences approach⁸ using a multivariable linear mixed-effects regression with a random intercept and standard errors clustered within patients and stratified by school age of student (elementary, middle, high). This study was approved by the VA Boston Research and Development committee prior to data collection and analysis.

Results

During the pre-COVID period, ∼12,000 patients aged 5–16 years received a physical; complete longitudinal data were available for 3650 patients (1519 elementary, 1257 middle, 874 high school). More than 90% of students reported attending public school; remote was the most common learning mode for elementary aged children (43.9%), whereas hybrid (with variable amounts of in-person time) was the most common learning mode for middle school (50.9%) and high school aged children (54.8%) (Refer Supplementary Table S1 for details of included patients. Details of patients excluded are included in Supplementary Table S2).

At baseline, the mean BMI percentiles were 0.631, 0.629, and 0.623 for elementary school, middle school, and high school aged children, respectively. Unadjusted mean BMI percentiles increased between the pre-COVID-19 and initial reopening time periods for elementary school aged and middle school aged children but not for high school aged children (Fig. 1); no association between school mode and change in BMI percentile was identified among any age group. In our multivariable regression model, BMI percentiles increased by 4.5 percentile points among elementary school aged children (95% confidence interval [CI]: 3.4–5.6) and 3.3 percentile points among middle school aged children (95% CI: 1.6–4.1) from pre-COVID-19 to the initial school reopening period (Table 1). In the regression model, BMI percentile increases were also identified among Hispanic elementary school aged children (5.5% increase, 95% CI, 1–10) and high school aged children (7.2% increase, 95% CI, 0.3–14), and among elementary school aged children (5.7%, 95% CI, 2.4–8.9) and middle school aged children (5.9%, 95% CI, 2.4–9.4) receiving Medicaid insurance.

Figure 1.

Unadjusted mean BMI percentiles by time period (pre-COVID, early school reopening, and post-full reopening) and school mode, stratified by school type (elementary, middle, high). None of the differences in school mode is significant using a cutoff of p < 0.05.

Table 1.

Results from Linear Mixed-Effects Regression of Patient Characteristics on BMI Percentile

Independent variables	Elementary school				Middle school				High school
	Coefficient^a	p	95% CI		Coefficient	p	95% CI		Coefficient	p	95% CI
	Coefficient^a	p	LL^b	UL^c	Coefficient	p	LL	UL	Coefficient	p	LL	UL
Time period
Pre-COVID (ref) (period 1)	—	—	—	—	—	—	—	—	—	—	—	—
Early school reopening (period 2)	0.045	<0.001	0.034	0.056	0.033	<0.001	0.023	0.043	−0.010	0.142	−0.023	0.003
Post-full reopening (period 3)	0.029	<0.001	0.016	0.041	0.006	0.345	−0.006	0.018	−0.021	0.008	−0.037	−0.006
School mode
Remote (ref)	—	—	—	—	—	—	—	—	—	—	—	—
Hybrid	0.009	0.613	−0.025	0.042	−0.008	0.650	−0.044	0.027	0.035	0.096	−0.006	0.076
Traditional	0.026	0.330	−0.026	0.077	−0.015	0.674	−0.086	0.055	−0.019	0.673	−0.108	0.070
School mode × policy period
Remote × pre-COVID (ref)	—	—	—	—	—	—	—	—	—	—	—	—
Hybrid × period 2	0.002	0.827	−0.013	0.017	−0.009	0.221	−0.022	0.005	−0.002	0.837	−0.019	0.015
Traditional × period 2	0.009	0.409	−0.013	0.031	0.013	0.346	−0.014	0.039	0.016	0.295	−0.014	0.046
Hybrid × period 3	−0.007	0.407	−0.024	0.010	0.006	0.452	−0.010	0.023	0.004	0.737	−0.017	0.024
Traditional × period 3	−0.003	0.844	−0.029	0.024	0.052	0.001	0.021	0.083	0.049	0.012	0.011	0.087
School type
Public (ref)	—	—	—	—	—	—	—	—	—	—	—	—
Private	−0.057	0.121	−0.129	0.015	−0.005	0.901	−0.086	0.076	0.075	0.088	−0.011	0.161
Home school	−0.090	0.049	−0.180	0.000	0.023	0.785	−0.141	0.187	−0.018	0.768	−0.140	0.103
Hispanic	0.055	0.018	0.010	0.100	0.040	0.123	−0.011	0.092	0.072	0.040	0.003	0.140
Medicaid	0.057	0.001	0.024	0.089	0.059	0.001	0.024	0.094	0.030	0.161	−0.012	0.073
Female	−0.010	0.510	−0.038	0.019	0.010	0.513	−0.021	0.042	0.016	0.402	−0.021	0.053
Race
White (ref)	—	—	—	—	—	—	—	—	—	—	—	—
Black	0.052	0.092	−0.008	0.112	0.072	0.068	−0.005	0.149	0.137	0.005	0.042	0.233
Asian/Pacific Islander/Native American	−0.053	0.038	−0.104	−0.003	0.001	0.974	−0.054	0.056	−0.064	0.087	−0.137	0.009
Unknown	0.018	0.316	−0.017	0.054	0.022	0.269	−0.017	0.061	−0.003	0.913	−0.051	0.045

Data are separated into three time periods: pre-COVID/period 1, which spans from November 1, 2019, to April 30, 2020; early school reopening/period 2, which spans from November 1, 2020 to April 5, 2021; and post-full reopening/period 3, which spans from November 1, 2021 to April 5, 2022.

Coefficients represent the absolute change in BMI percentile.

Lower limit of the 95% CI.

Upper limit of the 95% CI.

CI, confidence interval.

Among elementary school aged students, some differences by school type were identified. Among the small number of elementary school students who reported home schooling (65, 3.1%), there was a slight decrease in BMI percentile during the study period when compared with students attending public school (−9%, 95% CI, 0–18). There was also a trend toward a reduction among elementary school aged students attending private school, although this was not statistically significant.

Relative to the first time period, BMI among elementary school aged children increased in both the second and third time periods, with a smaller increase observed in the third time period; longer term impacts in middle and high school students were variable.

Discussion

Our finding of a significant increase in BMI percentile over a 1-year period spanning the beginning of the COVID-19 pandemic is consistent with earlier reports⁵ and expands upon previous research to evaluate the impact of school mode at an individual, rather than community, level, which has consistently found that elementary school aged children were most impacted.^3,4 During the time period following lifting of restrictions, including school closures, increases in BMI percentiles slowed but did not reverse, suggesting that ongoing efforts to mitigate the impacts of pandemic weight increases on children are needed.

Consistent with prior work by Beck et al,⁹ our study also identified significant disparities; Hispanic children and children with Medicaid insurance had significantly higher BMI percentile than other groups, suggesting that specific outreach and strategies may be needed to prevent long-term negative health outcomes in these populations that have been impacted by systematic disparities and lack of access to resources. Schools are an important source of meals and nutrition for many children, particularly for those living in poverty. During the pandemic, food insecurity increased substantially,¹⁰ particularly among children of color, and many children were not able to access meals typically available through the school system.¹¹ Additionally, children of color and children living in poverty disproportionally lack access to physical activities and outdoor spaces, which may also have contributed to the higher BMI percentile increases seen in these populations.¹²

Limitations include the relatively small cohort, particularly for traditional school mode attendance, which limits power to detect a small but significant difference. Small sample size also limited our ability to gather precise estimates about post-policy changes among middle and high school students. The study was conducted in a single state and relied on individually reported school mode, which may be subject to bias. School mode was strongly correlated with type of school, which may have impacted our ability to identify true differences due to collinearity. Private schools were more likely to be fully open during the time period than public schools, which primarily operated in either a hybrid or remote learning mode for most of the study period.

This may have suggested residual confounding if we had identified differences between the different learning types; however, no impact was found. Home schooled students may have been less likely to be impacted by closures due to less disruption to their normal routines explaining the reduction in BMI percentiles found among the relatively small number of home schooled elementary school aged children.

Conclusions

Given the well-documented negative health effects of obesity and strong correlation between childhood and adulthood BMI,¹³ the growing body of evidence suggests that the COVID-19 pandemic will have detrimental health impacts on elementary and middle school children far into the future if strategies are not rapidly implemented to counter pandemic-associated pediatric weight increases. More studies are needed to identify factors associated with pandemic weight gain in children and to identify strategies, such as school-based physical activity programs,¹⁴ to limit the long-term health impacts of weight increases on children, particularly among Hispanic children and children receiving Medicaid.

Footnotes

Authors' Contributions

W.B.-E., L.F., Z.E., and R.E.N. conceptualized and designed the study, drafted the initial article, and participated in review and revision. E.O. and E.M.S.-P. participated in the analysis and interpretation of the data and reviewing and revising the final article. All authors approved the final article as submitted and agree to be accountable for all aspects of their work.

Disclaimer

The views expressed are those of the authors and do not necessary represent those of the U.S. Department of Veterans Affairs or the U.S. Federal Government.

Access to Data and Data Analysis

R.E.N. and Z.E. had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Funding Information

No funding was received for this article.

Author Disclosure Statement

W.B.-E. is the site PI for a study funded by Gilead pharmaceuticals (funds to institution). E.M.S.-P. receives funds from the Gilead FOCUS program. The remaining authors report no competing interests.

Supplementary Material

References

von Hippel

, Workman

. From kindergarten through second grade, US children's obesity prevalence grows only during summer vacations. Obesity, 2016; 24(11):2296–2300; doi: 10.1002/oby.21613

Von Hippel

, Powell

, Downey

, et al. The effect of school on overweight in childhood: Gain in body mass index during the school year and during summer vacation. Am J Public Health, 2007; 97(4):696–702; doi: 10.2105/AJPH.2005.080754

Woolford

, Sidell

, Li

, et al. Changes in body mass index among children and adolescents during the COVID-19 pandemic. JAMA, 2021; 326(14):1434–1436; doi: 10.1001/jama.2021.15036

Lange

, Kompaniyets

, Freedman

, et al. longitudinal trends in body mass index before and during the COVID-19 pandemic among persons aged 2–19 years—United States, 2018–2020. Morb Mortal Wkly Rep, 2021; 70(37):1278; doi: 10.15585/mmwr.mm7037a3

, Aris

, Hivert

M-F

, et al. Association of changes in obesity prevalence with the COVID-19 pandemic in youth in Massachusetts. JAMA Pediatr, 2022; 176(2):198; doi: 10.1001/jamapediatrics.2021.5095

Rundle

, Park

, Herbstman

, et al. COVID-19-related school closings and risk of weight gain among children. Obesity (Silver Spring), 2020; 28(6):1008–1009; doi: 10.1002/oby.22813

Ertem

, Schechter-Perkins

, Oster

, et al. The impact of school opening model on SARS-CoV-2 community incidence and mortality. Nat Med, 2021; 27(12):2120–2126; doi: 10.1038/s41591-021-01563-8

Columbia University School of Public Health. Difference in Difference Estimation 2022 [Updated May 6]. Available from: https://www.publichealth.columbia.edu/research/population-health-methods/difference-difference-estimation [Last accessed: May 6, 2022].

Beck

, Huang

, Lendzion

, et al. Weight gain during the COVID-19 pandemic in a high-risk cohort of children in San Francisco, CA. Child Obes, 2022; 18(2):143–146; doi: 10.1089/chi.2021.0128

10.

Parekh

, Ali

, O'Connor

, et al. Food insecurity among households with children during the COVID-19 pandemic: Results from a study among social media users across the United States. Nutr J, 2021; 20(1):73; doi: 10.1186/s12937-021-00732-2

11.

Gupta

, Gonzalez

, Waxman

Forty Percent of Black and Hispanic Parents of School-Age Children Are Food Insecure. Urban Institute, Washington, DC; 2020.

12.

Watson

, Whitfield

, Huntzicker

, et al. Cross-sectional study of changes in physical activity behavior during the COVID-19 pandemic among US adults. Int J Behav Nutr Phys Act, 2021; 18(1):91; doi: 10.1186/s12966-021-01161-4

13.

Simmonds

, Llewellyn

, Owen

, et al. Predicting adult obesity from childhood obesity: A systematic review and meta-analysis. Obes Rev, 2016; 17(2):95–107; doi: 10.1111/obr.12334

14.

Yin

, Moore

, Johnson

, et al. The impact of a 3-year after-school obesity prevention program in elementary school children. Child Obes, 2012; 8(1):60–70; doi: 10.1089/chi.2011.0085

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB

0.01 MB