Do Changes in BMI during the COVID-19 Pandemic Persist in the Post-Pandemic Period in a Pediatric Population Attending Health Care Clinics? A Longitudinal Study

Obesity is a global health challenge whose incidence has increased worldwide over the last decades [1]. The World Obesity Federation recently predicted a global increase in the prevalence of childhood obesity, by 100% during 2020–2035 [2]. The development of childhood obesity is influenced by a complex interplay of factors: genetic, behavioral (physical inactivity, television viewing, soft drink consumption) [3‒5], and socioeconomic (eating out with peers, increasing exposure to mass media and advertising) [6]. Childhood obesity is associated with various immediate and long-term illnesses such as sleep apnea, hypertension, type 2 diabetes, heart disease, stroke, osteoarthritis, and certain types of cancer [7]. Children with obesity are also likely to become adults with obesity, which is associated with increased risks of morbidity and mortality [8]. Many societies have not succeeded in countering increases in childhood obesity.

Data of several countries showed increases in body mass index (BMI) during the COVID-19 pandemic in young populations [9‒12]. We reported this most substantially for children aged 2–6 years [13]. The disruptions experienced around the world, in daily routines of life, including lockdowns and quarantines, may have contributed to an obesogenic lifestyle, such as reduced opportunities for physical activity, irregular sleep patterns, and extensively prolonged screen times [14, 15]. Changes among children in dietary patterns during the pandemic [16] included increased consumption of highly processed food, which tends to be high in saturated fat, sugar, and salt [17, 18]. Quarantine also had a negative psychological impact on children and their families [19]. The unfamiliar and prolonged stressful situation could have further promoted unhealthy food intake through stress-related eating, thereby leading to obesity and other health problems [20]. While a few studies have addressed changes in body weight in the post-COVID-19 pandemic period [21‒24], long-term effects of the pandemic on body weight have not been sufficiently analyzed. Therefore, the aim of this longitudinal study was to analyze changes in BMI during the COVID-19 pandemic, and in the subsequent period, in a large pediatric population in Israel attending health care clinics using a representative database.

This observational retrospective population-based cohort study was conducted using the electronic database of Clalit Health Services (CHS), an Israeli payer-provider integrated health care system that serves about 54% of the Israeli population. The database is accumulated by continuous real-time input from physicians and health service providers and includes patient demographic, socioeconomic, and clinical characteristics, hospital discharge and outpatient clinic diagnoses, laboratory test results, medical treatments, and medication dispensation information. Data were extracted from CHS using the Clalit Research Data sharing platform powered by MDClone (https://www.mdclone.com).

Included were children and adolescents aged 2–16 years at their first height and weight measurements during the pre-pandemic period (January 1, 2018, to February 28, 2020). Inclusion criteria were at least one measurement of height and weight in the pandemic period (March 1, 2020, to March 30, 2021) and one in the post-pandemic period (January 1, 2022, to September 30, 2023). Measurements were done during the attendance at the health care clinics.

Notably, during the pandemic period there were three lockdown periods in Israel: March–May 2020, September–October 2020, and December 2020–March 2021. During most of 2020, nurseries and schools were closed or functioned only partially, and social distancing was recommended.

Excluded from the study were children and adolescents with genetic syndromes associated with obesity (such as Prader Willi syndrome and Bardet Biedel syndrome), with monogenetic obesity, or with endocrine disorders associated with obesity (Cushing syndrome, craniopharyngioma). The performance of bariatric surgery during the study period and the use of medications that may impact weight (oral/systemic steroids, antipsychotic medications) were also exclusion criteria. Individuals with malignancies, chronic illness (type 1 diabetes, chronic renal failure, chronic heart disease, inflammatory bowel disease, and cystic fibrosis), and eating disorders were not included. Nor were those with improbable measurements (−5> BMI-standard deviation scores [SDSs] >5) included. Figure 1 shows the flowchart of the individuals included in the study.

The study was approved by the local institutional Ethics Committee in keeping with the principles of the Declaration of Helsinki. Written informed consent was waived due to the anonymous collection of data.

The following data were collected from patient medical files: demographic parameters (sex, ethnicity, age), socioeconomic position (SEP), and all the anthropometric parameters recorded within the study time period (height and weight and calculated BMI). The SEP index of the Israel Central Bureau of Statistics is an adjusted calculation of 14 variables that measure social and economic levels in the domains of demographics, education, standard of living, and employment. SEP by home address was based on classification of towns and neighborhoods according to the socioeconomic level of the population in 2015 [25]. The Israel Central Bureau of Statistics’ SEP clusters are ranked on a scale of 1–10, with 1 representing the lowest SEP. In the database of CHS, SEP was categorized into three levels based on the Israel Central Bureau of Statistics’ 10 clusters: low SEP (clusters 1–4), medium SEP (clusters 5–6), and high SEP (clusters 7–10).

BMI was calculated as weight (in kilograms) divided by height (in meters) squared. When more than one BMI measurement was available for each of the study periods, the median BMI was used for the analysis. To compare BMI values across age-groups by sex, BMI-SDSs were calculated using the growth chart percentiles of the Centers for Disease Control and Prevention [26].

Data from the pre-pandemic period were compared to data of the pandemic and post-pandemic periods. A change in BMI-SDS ≥0.25 was considered statistically significant [27]. Two age-groups at the pre-pandemic period: 2–6 years and 6.1–16 years were analyzed. BMI values <5th percentile (BMI-SDS <−1.645) defined underweight, the 5th–84th percentiles (−1.645≤ BMI-SDS ≤1.036) normal weight, the 85th–95th percentiles (1.036< BMI-SDS ≤1.645) overweight, and ≥95th percentile (BMI-SDS >1.645) defined obesity.

Statistical analyses were performed using SPSS software, version 29 (SPSS, Inc., Chicago, IL). Continuous variables with normal distributions were reported as means and standard deviations, continuous variables with skewed distributions as medians and interquartile ranges, and categorical variables as numbers and percentages. Demographic and anthropometric characteristics were compared between males and females using the independent-samples t test or the Mann-Whitney U test, for variables with normal or skewed distribution, respectively, or Pearson’s chi-square test for categorical variables.

General linear models repeated-measures analysis was conducted to evaluate changes in BMI-SDS between the periods. The models were specified with a within-group factor of time (the year of BMI-SDS evaluation), a between-group factor (according to sex, ethnicity, age category, SEP status, and pre-pandemic BMI category), and the interaction of group with time. The data are expressed as estimated marginal means and standard errors.

To evaluate differences in delta BMI-SDS (pandemic/post-pandemic minus pre-pandemic values) between categories (of sex, ethnicity, age, SEP status, and BMI in the pre-pandemic period), the independent-samples t test or 1-way ANOVA and post hoc Tukey were conducted. To evaluate differences between categories, in the proportions of individuals with delta BMI-SDS ≥0.25, Pearson’s chi-square tests were conducted. Significance was set at p ≤ 0.05.

Table 1 presents characteristics of the study cohort. Included were 106,871 individuals (52.1% males): 79.6% of Jewish ethnicity and 20.4% of Arab ethnicity. The median age at the pre-pandemic period was 8.2 years: 39.4% were aged 2–6 years and 60.6% 6.1–16 years. The mean time between the pre-pandemic and the pandemic BMI measurements was 1.9 ± 0.7 years (range 0.2–3.2) and between the pandemic and the post-pandemic BMI measurements, 2.1 ± 0.6 years (range 0.8–3.6).

In the pre-pandemic period, 13.4% of the cohort were with overweight and 12.8% with obesity. The respective proportions were 14.3% and 15.4% for the pandemic period and 13.3% and 15% for the post-pandemic period.

Table 2 presents the general linear models repeated-measures analysis for the trends in BMI-SDS from the pre-pandemic to the pandemic and the post-pandemic periods. During the entire study period, the estimated mean BMI-SDS was higher for females than males (p_group <0.001). For both sexes, BMI-SDS differed significantly between the three periods (p_group <0.001). For both sexes, the BMI-SDS increased significantly in the pandemic period, and then decreased in the post-pandemic period (the decrease was greater among females, p_time*group <0.001), while remaining significantly higher than in the pre-pandemic period (Fig. 2).

Throughout the study period, the BMI-SDS was lower in the younger (age 2–6 years) than the older group (p_group <0.001). For the former, the BMI-SDS increased during the pandemic, and decreased thereafter, while remaining higher than the pre-pandemic BMI-SDS (p_time <0.001). In the older age-group (6.1–16 years), the slope differed from that of the younger group (p_time*group <0.001), and BMI-SDS steadily increased from the pre-pandemic to the pandemic and post-pandemic periods (p_time <0.001).

During the entire study period, the BMI-SDS was higher among those of Arab than Jewish ethnicity (p_group <0.001); however, the slopes differed (p_time*group <0.001). Among those with Jewish ethnicity, the BMI-SDS increased during the pandemic, and subsequently decreased, while remaining higher than the pre-pandemic BMI-SDS (p_time <0.001). Among those with Arab ethnicity, the BMI-SDS steadily increased in the pandemic and post-pandemic periods (p_time <0.001).

During the entire study period, BMI-SDS was higher among those in the lowest than in the two other SEP clusters (p_group <0.001), and the slopes differed by SEP categories. During the pandemic, the BMI-SDS increased in all the SEP clusters (p_time <0.001). In the post-pandemic period, the BMI-SDS slightly decreased in the low and medium SEP groups and slightly increased in the high SEP group (p_time*group <0.001).

Among children with underweight or normal weight in the pre-pandemic years, BMI-SDS increased significantly in the pandemic and post-pandemic periods. Among those with overweight or obesity in the pre-pandemic years, BMI-SDS decreased steadily during the pandemic and post-pandemic periods (p_time*group <0.001).

Table 3 shows the changes in BMI-SDS and the proportions of individuals with increases in BMI-SDS ≥0.25 during the pandemic and post-pandemic periods, stratified by sex, age category, ethnicity, SEP category, and the BMI category in the pre-pandemic period. The increase in BMI-SDS was higher among females than males, in both the pandemic (p < 0.001) and the post-pandemic periods (p = 0.002). However, for a greater proportion of males than females, BMI-SDS increased by ≥0.25 (p = 0.001).

BMI-SDS increased more among children aged 2–6 than 6.1–16 years, in both the pandemic (p < 0.001) and the post-pandemic periods (p < 0.001). For a higher proportion of the younger than the older children, the BMI-SDS increased by ≥0.25 during the pandemic (p < 0.001) and post-pandemic periods (p < 0.001).

BMI-SDS increased more among children with Arab than Jewish ethnicity, in the pandemic (p < 0.001) and the post-pandemic periods (p < 0.001). For a greater proportion of the former than the latter, the BMI-SDS increased by ≥0.25 during the post-pandemic period (p < 0.001).

In the pandemic period, the BMI-SDS increased more among those in the lowest SEP category (p = 0.005). In the post-pandemic period, the increase was comparable and higher among those from the low and high SEP groups compared to the medium SEP group (p < 0.001). For greater proportions of the low and high compared to the median SEP group, BMI-SDS increased by ≥0.25 during the post-pandemic period (p < 0.001).

During the pandemic and post-pandemic periods, BMI-SDS increased more among individuals with underweight than with normal weight in the pre-pandemic period and decreased more among those with overweight and obesity in the pre-pandemic period (p < 0.001). For greater proportions of individuals with underweight than in the other body weight categories during the pre-pandemic period, the BMI-SDS increased by ≥0.25 during the pandemic (p < 0.001) and post-pandemic periods (p < 0.001).

Table 4 presents transitions in BMI category proportions, from the pre-pandemic to the pandemic and post-pandemic periods. For 55.2% and 59.8% of the individuals with underweight in the pre-pandemic period, weight status improved to normal weight during the pandemic and post-pandemic periods, respectively. Moreover, 34.6% and 41.7% of those with overweight in the pre-pandemic period were with normal weight during the pandemic and post-pandemic periods, respectively. Among those with obesity in the pre-pandemic period, normal weight was recorded for 9.5% in the pandemic and 14.7% in the post-pandemic period and overweight for 22.3% and 19.1% in the respective periods. Among the individuals with normal weight in the pre-pandemic period, weight status deteriorated to overweight for 8.3% during the pandemic and for 9.4% after the pandemic. Moreover, 5.3% and 4.5% of those with normal weight in the pre-pandemic period were with obesity during the pandemic and post-pandemic periods, respectively. Further, 22.5% and 25.7% of those with overweight in the pre-pandemic period were with obesity during the pandemic and post-pandemic periods, respectively.

In our pediatric cohort, BMI-SDS increased significantly during the pandemic period, and tended to decrease in the post-pandemic period (after a mean 2.1 ± 0.6 years), but still remained above the BMI-SDS in the pre-pandemic period. The overall prevalence of obesity increased by 2.6% during the pandemic and was 2.2% higher in the post-pandemic than the pre-pandemic period. The increased prevalence of obesity during the COVID-19 pandemic is similar to the 2% (95% CI: 1%, 3%) reported in a meta-analysis that included 12 pediatric studies [28].

We previously reported weight gain during the COVID-19 pandemic, in a smaller pediatric cohort (n = 36,837) [13]. However, that study was based on data of children from three districts in Israel ensured by the CHS. The current cohort includes all the districts ensured by CHS and strengthens our previous findings. It also corroborates reports worldwide of increased rates of overweight and obesity during the COVID-19 pandemic [12, 28‒30]. These trends may be attributed to the pandemic-related lifestyle changes described in previous literature, including decreased physical activity [31], changes in diet composition [32], increased indoor sedentary behaviors, restrictions in movement, and closures of nurseries and schools.

Most previous studies examined only short-term changes in body weight during the COVID-19 pandemic. Only a few pediatric studies have addressed the long-term effects of the pandemic and the subsequent period on weight status. Qin et al. [21] found that the BMI-SDS of children and adolescents (6–14 years old) significantly increased in 2020 (the pandemic period) compared to 2019 (p < 0.01) and decreased in 2021 and 2022 (the post-pandemic period). Irschik et al. [22] reported effects of COVID-19 restrictions in Austria on pediatric BMI, with very little effect of socioeconomic background. They reported an increased rate of obesity, from 6.4 to 12.1%, throughout the pandemic, reaching a maximum of 15.2% during the restrictions. Overall, age-adapted BMI z-scores increased significantly, by 0.22 during the restrictions, and remained increased by 0.19 compared to pre-pandemic levels.

Greve et al. [24] conducted a longitudinal study on BMI changes in a cohort of 1,298 children aged 2–18 years in Indianapolis, tracking their progress for up to 30 months following the onset of the pandemic. The study revealed a significant rise in BMI from the pre-pandemic period to the early stages of the pandemic. Although the rate of BMI increase stabilized from the early to late pandemic periods, the findings suggest that children continued to experience a slower yet persistent rise in BMI after the pandemic began.

For females compared to males, we report higher BMI-SDS throughout the study period, and greater increases in BMI-SDS, both in the pandemic and the post-pandemic periods. In the pre-pandemic, pandemic, and post-pandemic periods, the proportions of females versus males with overweight or obesity were 26.8% versus 25.6%, 31.2% versus 28.8%, and 29.3% versus 27.3%, respectively. This may be explained by the general tendency of a more active and less sedentary lifestyle in males than females. However, our findings contrast to studies that reported significantly greater increases in BMI among males than females during the COVID-19 pandemic [33, 34]. This discrepancy may be due to the smaller cohort of those studies compared to ours.

Interestingly, in both the pandemic and the post-pandemic periods, the increase in BMI-SDS was greater among children aged 2–6 than 6.1–16 years; and the increase in BMI-SDS was ≥0.25 for a greater proportion of the younger than the older group. A greater impact of the pandemic on BMI in younger ages was previously reported [13, 33, 35]. We speculate that among the younger children, increased snacking and more meals prepared and consumed at home may have promoted weight gain. Nevertheless, BMI changes differed between the age-groups. Among the younger children, the BMI-SDS increased during the pandemic, and subsequently decreased, yet remained greater than the pre-pandemic BMI-SDS. Among the older children, the BMI-SDS increased steadily, from the pre-pandemic to the pandemic and post-pandemic periods. We speculate that for the younger compared to the older children, the parents/caregivers may have had more positive influence on food choices and physical activity in the post-pandemic period. This underscores the importance of obesity prevention and management efforts during and following the COVID-19 pandemic, including increased efforts to promote healthy behaviors.

During the entire study period, BMI-SDS was higher among those with Arab than Jewish ethnicity; and the increase in BMI-SDS was greater for the former. This ethnical difference was previously reported in a study conducted prior to the COVID-19 pandemic [36]. Among our children with Jewish ethnicity, the BMI-SDS increased during the pandemic, and decreased thereafter, yet remained higher than the pre-pandemic BMI-SDS. Among those with Arab ethnicity, the BMI-SDS steadily increased during the pandemic and thereafter. Possibly, families in Arab communities may be less aware of the presence and implications of overweight and obesity in children. This may be related to sociocultural beliefs of ideal body image, and an acceptance of larger body size, and may also be due to lower SEP. Our findings raise concerns that a national trend of ethnic differences in pediatric health may have widened during the pandemic and thereafter.

During the study period, BMI-SDS was higher among those in the lowest than the two other SEP clusters. Lower SEP was previously reported to be associated with a higher prevalence of obesity in Israel [37]. Coping with financial strain may negatively impact health behaviors, such as by leading to unhealthy food choices and decreased physical activity. Indeed, Jenssen et al. [35] previously reported a higher rate of weight gain during the COVID-19 pandemic among children of lower socioeconomic status.

An interesting finding in our cohort was that following the increase in BMI-SDS across all SEP groups during the COVID-19 pandemic, in the post-pandemic period, BMI-SDS slightly decreased by 0.004 SD in the low SEP group and slightly increased by 0.007 SD in the high SEP group. These changes are statistically significant due to the large sample size but clinically negligible. In the medium SEP group, BMI-SDS decreased by 0.021 SD, though still a small change. This decrease in the medium SEP group can be attributed to several positive changes. As economies recovered, financial stability in the medium SEP group improved, allowing families to afford healthier food options with reduced reliance on cheaper, calorie-dense foods. Also, greater engagement in physical activities became more feasible.

BMI-SDS increased significantly in the pandemic and post-pandemic periods among children with underweight or normal weight in the pre-pandemic years. However, BMI-SDS decreased steadily during these periods among those with overweight or obesity. For a greater proportion of individuals with underweight than in the other weight categories during the pre-pandemic period, the increase in BMI-SDS was ≥0.25 during the pandemic and post-pandemic periods. Our findings corroborate our previous findings [13] and those of Weaver et al. [9] and Azoulay et al. [38] who found that children with normal weight, but not those with overweight or obesity, experienced significant acceleration in BMI-SDS change. Our findings contrast to a longitudinal study that showed sharp increases in BMI rates during the COVID-19 pandemic among persons aged 2–19 years in the USA, especially among those with overweight or obesity [12]. Also, Knapp et al. [30] found that children with BMI in the obesity compared with the healthy weight range were at higher risk for excess BMI gain during the pandemic. Irschik et al. [22] reported that all the children in their cohort experienced significant weight loss after the restrictions were lifted, but children with obesity continued to gain weight without any sign of the onset of normalization.

The improved weight status in our cohort, of both individuals with underweight and those with overweight or obesity in the pre-pandemic period, may be explained by changes in family practices during the pandemic. Specifically, during the pandemic, many parents spent more time at home and were more involved in their children’s daily lives. This could have led to greater parental control over meals frequency and content and activities, resulting in healthier behaviors that persisted after the pandemic.

Martinko et al. [39] observed less pronounced decreases in physical fitness during COVID-19 restrictions among individuals with obesity compared to their peers with normal weight. This interesting finding suggests that children with obesity may have been more resilient to the negative effects of the pandemic on physical fitness. Another possible explanation relates to public health messages warning that excess weight could increase risks of worse outcomes of COVID-19. Such messages may have motivated individuals with overweight or obesity to make additional efforts to improve their weight status, particularly given that participants in our cohort were in contact with health care teams during attendance of health care clinics. However, as no behavioral data were collected, these interpretations remain speculative.

The increased rate of eating disorders reported during the COVID-19 pandemic [40] may also explain the findings. Although we excluded individuals with confirmed eating disorders, some of those included may have had an undiagnosed eating disorder that impacts weight status.

To the best of our knowledge, this is the first study to follow the course of weight status after the pandemic and to analyze the long-term impact on children’s BMI in Israel. The follow-up is important, as the COVID-19 public health crisis led to societal changes, and only few studies have examined the lasting effects on body weight status. Moreover, hypotheses based on prior data cannot be generated.

A recent study of the entire Israeli population evaluated changes in BMI during the COVID-19 pandemic [41]. However, each year’s data represented a cross-section of the Israeli population, and the same individuals were not tracked from year to year. Therefore, the strengths of our study include the long-term longitudinal evaluation, using objective measurements of BMI, of a large cohort of the same children and adolescents, analyzed by age-group, ethnicity, and SEP clusters. Yet some limitations exist. First, only children and adolescents who attended health care clinics and had anthropometric measurements in the pre-pandemic, pandemic, and post-pandemic periods were included. Possibly, the parents of those included were more interested in their children’s growth status. Therefore, the observed changes in BMI might have been underestimated compared to those for the general population and may not be representative of population level data. Also, the lack of information on diet and physical activity limits comparing outcomes between the periods.

The COVID-19 pandemic correlated with an overall increase in BMI-SDS among children and adolescents in Israel. Although BMI-SDS decreased post-pandemic, it remained above pre-pandemic levels. This trend was evident across various demographic and socioeconomic groups, highlighting the widespread impact of the pandemic on pediatric BMI. The most significant BMI-SDS increases were seen in children with underweight and normal weight, while decreases occurred among those with overweight and obesity. The persistent elevation in BMI-SDS post-pandemic, particularly among certain age-groups and socioeconomic clusters, has significant public health implications given the known relationship between childhood obesity and long-term health outcomes. Policies targeted to effective intervention for preventing increasing obesity rates are of utmost importance.

The study was reviewed and approved in keeping with the principles of the Declaration of Helsinki by the local Ethics Committee (the Rabin Medical Center Ethics Committee, Rabin Medical Center, Approval No. RMC-24-0060). The need for informed consent was waived by the local Ethics Committee (Rabin Medical Center Ethics Committee, Rabin Medical Center, Petach Tikva, Israel).

The authors declare that they do not have any financial or other relations that might lead to a conflict of interest related to this work. Shlomit Shalitin is an associate editor of Hormone Research in Pediatrics, and Moshe Phillip is an editorial board Member of Hormone Research in Pediatrics.

No financial assistance was received in support of this study.

Shlomit Shalitin: substantial contribution to conception and design of the study, acquisition of data, interpretation of data, drafting the article, and final approval of the version to be published. Moshe Phillip: substantial contributions to revising the article critically for important intellectual content and final approval of the version to be published. Michal-Yackobovitch-Gavan: substantial contributions to acquisition of data, data analysis and interpretation of data, revising the article critically for important intellectual content, and final approval of the version to be published.

The data that support the findings of this study are not publicly available since their containing information could compromise the privacy of research participants but are available from the corresponding author (S.S.).