Changes in Adiposity as Measured by Modified BMI Z-scores Associated with the COVID-19 Pandemic Among Well Children and Adolescents Within a Large Urban Health System

Background: To control the spread of COVID-19 in NYC, a stay-at-home order was issued for March 22, 2020-June 13, 2020 followed by ongoing restrictions and only partial school reopening. The health impacts of these unprecedented restrictions on children are not yet known. Our objective was to examine whether BMI change during the COVID-19 period differed from the prior year among healthy children.

Methods: This is a multicenter retrospective observational study of children aged five to eighteen years who received care within the NYC public hospital system between 2018 and 2020 encompassing all the five boroughs of the city. Using the system’s electronic medical records, we identified patients who had a well-child visit between June 13, 2020 and December 31, 2020. Patients were included if they had annual well-child visits in the two previous years with anthropometric data recorded at all three visits. Patients with diagnoses that may affect adiposity (e.g. hypothyroidism) were excluded. We collected sociodemographic and anthropometric data for each year and calculated a modified body mass index z-score (mBMIz), which is a measure of relative BMI adjusted for age and sex. We compared change in mBMIz/month between the 2019 and 2020 visits (which included the pandemic) with the change in mBMIz/month between the 2018 and 2019 visits using generalized estimating equations (GEE). We examined whether the mBMIz/month change differed between the two periods and whether 2018 BMI category (underweight (5%), normal (5-85%), overweight (85-95%), obese (95-99%), extremely obese(>99%))modified this effect. Analyses were conducted using STATA software.

Results: Of 23,458 patients seen between June 13, 2020 and December 31, 2020, 7,575 (32.3%) met our inclusion and exclusion criteria. The mean mBMIz was 0.68 in 2018, 0.69 in 2019 and 0.90 in 2020 (Table 1). In GEE analysis, the change in mBMIz/month was statistically significantly higher in 2019-2020 than in 2018-2019 (mean difference in mBMIz/month change=0.0152, p<0.0001). While the change in mBMIz/month in 2019-2020 was lowest in patients in the extremely obese category, the difference in mBMIz/month between the two periods was greatest in this group (interaction effect of period and 2018 baseline category p<0.0001) (Table 2).

Conclusion: Our study suggests that the activity restrictions due to the COVID-19 outbreak are associated with greater excess weight gain in a diverse NYC population of healthy children as measured by mBMIz relative to the year prior to the outbreak. These findings correspond to the mean BMI percentile of our population changing from approximately 75th to the 82nd. These findings require further monitoring to better understand the specific reasons for the observed patterns across children in different BMI categories, assessment of longer-term health impact on this population and interventions to modify or reverse these worrisome trends.