Worldwide, the proportion of children being overweight has notably increased within the last decades (1). Overweight is defined as abnormal or excessive fat accumulation, which is associated with a higher chance of subsequent overweight, disability, and premature death in adulthood (2, 3). In addition, overweight children suffer from respiratory problems, hypertension, early signs of cardiovascular diseases, and psychological health problems (2, 4–6).

Overweight in childhood is considered one of the twentyfirst century's most serious public health challenges, which is arising from complex interactions between biological, behavioral, socio-environmental, and basic environmental factors (5–9). Despite major efforts to promote weight reduction, early childhood overweight has reached epidemic proportions in high-income countries (10). In 2019, ~38 million children under the age of 5 years were classified as overweight or obese worldwide (2). In Germany, a representative study has revealed that the percentage of overweight children is 10.8% for 3–6-year-old girls and 7.3% among boys. The prevalence of obesity among 3–6-year-old girls and boys is 3.2% and 1.0%, respectively (11).

In addition to the factors mentioned above, health in early childhood also depends on the socio-economic position (SEP) of the family, which is usually defined by parental education, occupation, and household income (12). In high-income countries, epidemiological studies have consistently shown that children with socio-economic disadvantages (i.e., low family SEP) have disproportionately poorer health outcomes than socio-economic advanced children and are more likely to be affected by childhood overweight (11, 13–16). Since health-related attitudes and behaviors formed at an early age are often carried into adulthood, health inequalities during childhood and adolescence might provide the foundation for health inequalities across the life course (17–20). Therefore, childhood and adolescence are particularly suitable time frames for health prevention and promotion (13, 21, 22).

One of the most effective interventions for childhood overweight is physical activity (23, 24). Preschool age is considered a critical window for the development of young children's physical activity habits (25). Scientists agree that children in these early years should be abundantly physically active through structured and unstructured play (26–28). Movement, play, and sports are of great importance in early childhood education and care (ECEC) centers as they function as a central socialization instance and have a formative influence on the health behavior of preschool children (29–32). Moreover, a sports focus of ECEC centers in the form of specific physical activity programs enjoys great popularity with parents and educators in Germany (33). Thus, ECEC centers represent feasible settings for health interventions, as 92% of children under 6 years of age are cared for in a daycare center in Germany (34). This study, therefore, aimed to examine the independent and interaction effects of family SEP and an ECEC center sports focus on the body mass index (BMI) of preschoolers. As significant gender differences could be expected in the relevance of these factors, all analyses were stratified by sex (35, 36).

Secondary data analysis was performed using data from the German National Educational Panel Study (NEPS) (37) of the Leibniz Institute for Educational Trajectories (LIfBi) at the University of Bamberg. The NEPS is a nationwide representative study with a multi-cohort sequence design. The main objective of the NEPS is the collection of life span data on the development of competencies, educational processes, educational decisions, and returns to education in different contexts. Surveys were carried out with children and their parents, as well as with educators and the institution heads of the ECEC centers. The clustering within ECEC centers makes multilevel analyses to consider the meso-level possible (38). In this study, we used the first wave of Starting Cohort 2 “Kindergarten” (SC2). Of the 2,996 children, 1,891 children (955 boys and 936 girls) from 224 ECEC centers had valid data on relevant variables (see below) and were included in our analyses.

Of the total 1,891 children from the 224 ECEC centers, 936 were girls and 955 were boys, representing a sufficient sample size for multilevel investigations (42, 43). Of these, 983 children were enrolled in an ECEC center with a sports focus. Children were 5 years old on average (mean = 4.98, SD = 0.34). Table 1 shows an overview of all demographic variables stratified by sex. Boys had a significantly higher mean BMI than girls (boys: mean = 15.59, SD = 3.52; girls: mean = 15.17, SD = 2.44; p = 0.0025). There were no significant differences in the other independent variables.

Table 2 depicts children's demographic variables stratified for each SEP tertile. BMI was the highest in the low-family SEP tertile and the lowest in the high-family SEP tertile. Slightly more than half of the examined children in the sample attended an ECEC center with a sports focus (51.98%). There were no differences in the attendance rate regarding the SEP (p = 0.643).

The results of the multilevel analysis of the main and interaction effects of family SEP and ECEC center focus on BMI are presented in Table 3. For boys, a significant main effect indicated a generally lower BMI in the middle SEP tertile and in the highest SEP tertile compared to the lowest SEP category. Another main effect showed that boys attending an ECEC center with a sports focus had a lower BMI than boys who do not attend a center with a sports focus. In addition, interactive effects between family SEP and ECEC center sports focus occurred (Table 3). Considering the predictive margins (Figures 1, 2), boys with low family SEP not attending a sports focus ECEC center had the highest BMI, while boys with low family SEP attending a sports focus ECEC center had the lowest BMI. For girls, a significant main effect revealed a generally lower BMI in the middle and high family SEP tertiles compared to the lowest SEP tertile. No association of ECEC center focus or interactive effect emerged for girls. Girls with high family SEP had the lowest BMI in both ECEC center types (with or without sports focus).

The analysis of 1,891 German preschool children revealed that the BMI of the preschoolers was related to the family SEP. Our results show that boys and girls from lower SEP families had a higher BMI on average. For boys, the sports focus of the ECEC center also played a role. We found that the association between SEP and BMI among boys was moderated by the sports focus of ECEC centers. Visiting an ECEC center with a sports focus appears especially health-promoting for boys from a weaker socio-economic background.

Our analysis showed a social gradient toward higher BMI for socio-economically disadvantaged children for both sexes. However, we found evidence for the relevance of ECEC center sports focus for BMI for boys only. Therefore, it might be speculated that for boys' factors outside the family might be relevant concerning BMI, whereas for girls, the family SEP seems to play a more important role (44). Our finding is all the more surprising considering that girls are known to have more sedentary lifestyles and lower levels of physical activity than boys (45). For this reason, it would be expected that girls within an ECEC will particularly benefit from a specific and consolidated curriculum for the promotion of physical activity.

Regarding socio-economic inequalities in health, surprisingly little is known about factors located at the meso-level since previous research on preschool children's BMI has focused predominantly on the micro level (46). One explanation might be the complexity and dynamics of the system studied (47, 48). As a consequence, there is little empirical knowledge about the role of institutions that form a central link between the individual and the macro level in the emergence of patterns of health inequalities in the developmental stages from early childhood to early adulthood (49, 50). This hinders the effective design of institutional interventions to promote better health, which is especially important, as childhood overweight is related to a higher risk of illness in adulthood (e.g., cardiovascular diseases or chronic illness), stigma, reduced self-esteem (51, 52), and a higher psychological stress perception (53). Certainly, the negative consequences of overweight on health do not occur as late as adulthood. Gender differences in the relationship between overweight and social inequality increase with age (54) and can be explained by sex differences (e.g., hormonal balance and neurological factors) and gender differences (e.g., gender-based stereotypes and related parental expectations that influence parenting) (55). Further studies should focus on how these biological and socio-cultural factors interact with the BMI and SEP of children already in preschool age and how ECEC centers can impact these effects in a positive way to prevent health and gender inequalities during the life course.

In addition to the sports focus of the ECEC center, other relevant factors might also be conceivable. For example, the composition of the group, the experience and expertise of the teachers, and the equipment available at the ECEC center might play relevant roles. Thus, further research might reveal other relevant ECEC characteristics.

In addition, further research might investigate which type of supply (e.g., sports courses, swimming, physical activity offers, enrichment of outdoor areas, equipment of the indoor area, or the sheer size of the facility) is related to the greatest increase in physical activity time. In this context, it would be also interesting to investigate whether physical aspects (e.g., area, number of playground equipment, and attractiveness of playground equipment) and social aspects (e.g., staff, qualification, and attitude of staff) interact.

In ECEC centers, a suitable physical activity environment should be created so that preschoolers can develop physical activities. In addition, evidence exists that the amount of physical activity in preschool age positively influences the amount of time people are active in adulthood (56). Wellqualified pedagogical staff promoting physical activity should therefore instruct the children's exercises (57).

It appears advisable that overweight prevention and treatment interventions should address the most disadvantaged groups to not further exacerbate inequalities in weight (58). Effectively tackling overweight, therefore, requires a “proportionate universalism” (59), whereby interventions are delivered at the level that meets the need across the social gradient. In our study, boys in the lowest family SEP tertile seem to benefit most from an ECEC center with a focus on sports. Accordingly, ECEC centers could play a critical role in addressing health inequalities regarding BMI, at least for boys.

All analyses were repeated with an alternative calculation for the BMI in children (60). The standard deviation score of the BMI (SDS BMI) was calculated based on representative data for Germany (61). The SDS BMI is standardized for age and gender and transformed to the value range of a standard normal distribution. This sensitive analysis generally yielded comparable results.

Several limitations have to be considered. Since this was a secondary data analysis, the possibilities for capturing individual and meso-level aspects were limited. In particular, further characteristics of the ECEC would be relevant here. These are usually differentiated into physical (sports equipment, movement areas, and architecture), economic (kindergarten fees), political (curriculum, rules, and timetables), and socio-cultural (attitudes and social norms) characteristics. Whether a child moves a lot or little in an ECEC center is then determined by a complex interplay of these characteristics (62). Another limitation refers to the outcome. The BMI represents a simple index of weight-for-height that is commonly used to classify overweight and obesity. Other measures might appear more suitable in the age of preschoolers, for example, skinfold thickness measurement or waist-to-hip ratio (2). Nevertheless, the BMI has the advantage that it is easy to measure and can therefore also be used with preschool children. However, this indicator does not consider the typical growth spurts in the preschool age group, which can lead to a statistical overshadowing of possible effects of physical activity promotion. In addition, improper body posture and body deformities, which are usually associated with increased BMI and obesity, should be considered in further studies. Furthermore, a selection effect cannot be ruled out; children who are already more active might be more likely to be enrolled in sports-focused ECEC centers. This means that the variable BMI is potentially endogenous which could bias the analysis. Another source of bias might be in the measurement of height and weight to calculate BMI. As in other large-scale population-based studies, this study used parent-rated data to assess height and weight. These estimates appear, however, less sensitive for underweight and overweight and might bias results (63, 64). Further studies are therefore needed to confirm the findings by applying professionally measured data for weight and height. In addition, it was not considered, whether the children examined lived with only one parent as there is evidence that children of single parents are more physically active and play outside a lot more (65). Future research could also take this aspect into consideration.

In conclusion, this study revealed the importance of daily physical activity for boys regarding their BMI in ECEC centers, especially for boys with low family SEP. Particularly boys from socio-economically disadvantaged families seem to benefit from visiting an ECEC center with a sports focus. However, for girls, no association of the ECEC center sports focus or interactive effects with SEP with BMI was found. Taken together, our analysis indicates that attempts to reduce the social gradient in BMI should take the gender as well as the characteristics of the ECEC center into account as they play an independent and interactive role.

Publicly available datasets were analyzed in this study. This data can be found here: https://www.neps-data.de/Datenzentrum/Datenzugangswege.

The studies involving human participants were reviewed and approved by a special data protection and security officer of the NEPS. Written informed consent to participate in this study was provided by the participants' legal guardian/next of kin.

RH analyzed the data. RH and AM interpreted the data. AM, RH, EW, and SS drafted the manuscript. SH, MB, KD, DJ, LS, TK, and MA critically revised the manuscript. All authors read and approved the final manuscript.