Differences in the developmental pace of body mass index in children – new patterns discovered

Figure: LongITools

What was studied?

Body mass index (BMI) is a commonly used measure for assessing a person’s weight status. It is calculated as the ratio of a person’s weight to their height squared. In children, BMI is particularly useful in population studies. Since children’s height and weight continuously change as they grow, their age and sex must be considered in weight assessment. BMI measurements can be analysed as a curve or trajectory if they were available from several time points (or ages).

There are known differences in children's developmental rates, but typically these differences have only been studied at specific points of growth. In the current study, we aimed to determine if children can be grouped based on their developmental rate of BMI throughout childhood.

Who were studied?

We studied the BMI trajectories of children participating in the Northern Finland Birth Cohorts 1966 and 1986 from the age of 3 months to 16 years. Participants’ height and weight measurements from child health clinic visits and school health care were linked with the two cohorts’ data, increasing the number of measurements available per participant. This study was limited to those children who had regular height and weight measurements available, with no breaks of more than five years between subsequent measurements. Thus, we studied the BMI development of nearly 6 200 children out of the 20 000 children participating the Northern Finland Birth Cohort studies.

How was the study done?

In this study, we utilised a functional data analysis (FDA) framework, where the analysis objects are functions instead of observation points. This means that the children’s BMI trajectories were considered as observations of the underlying BMI functions. In the BMI functions, the x-axis reflects the child's age, and the y-axis reflects the child's BMI level.

For each pair of BMI functions (child 1 – child 2, child 1 – child 3, child 2 – child 3, etc.), a phase difference was calculated, which describes how well the functions are aligned with one another in terms of age. If two functions were perfectly aligned, the only visible difference between them would be the possible difference in BMI levels. For example, the peaks and valleys would coincide on the x-axis, but not necessarily rise as high or fall as low on the y-axis.

These phase differences were used in cluster analysis as a measure of the differences in BMI trajectories. The goal of cluster analysis is to form groups in which the objects within one group, in this case the BMI trajectories, are as similar as possible to each other and as different as possible between the groups.

What was found?

The BMI trajectory clustering resulted in three distinct groups for both girls and boys. Differences between these groups were related to the development of the BMI trajectory over time, rather than to the magnitude of BMI. In the largest group, the BMI trajectory followed a pattern that represents a typical childhood BMI trajectory. This pattern has been described in the research literature already in the 1980s: first, BMI increases rapidly until about 9 months of age, after which the trajectory starts to decrease. The decline stops around the age of 5–6 years and then starts to increase again towards the level of BMI in adulthood. About half of the children exhibited this BMI trajectory pattern.

In the next largest group, the increase in BMI during infancy and the subsequent decrease occurred more intensely than in the previous group. After this, the BMI typically 'dragged' in trough before the growth that ends in adulthood began again. Approximately 30–35% of children were found to have such a BMI trajectory.

In the smallest group, the same trajectory shape was observed as in the largest group, but the curves did not behave as regularly at different stages. About 15% of children were found to have such a BMI trajectory. The two smaller groups have not previously been reported in the research literature.

Why is this important?

The study described previously unreported patterns of childhood BMI trajectory that were relatively common in the studied data. In addition, there was a difference in the prevalence of the different patterns (representing the groups) between birth cohorts, which may indicate a generational difference in the development of BMI. So far, it is unclear whether growing on a certain BMI trajectory pattern affects later health.

BMI trajectory describes the development of body mass index over a longer period of time, which may be useful in understanding the underlying factors of obesity or its comorbidities. Changes in the body that affect weight gain begin before the weight increases to a point where, for example, overweight is observed. This could also be important for preventing the development of obesity later in life and for targeting interventions or treatments.

Author:

Anni Heiskala

The study was conducted as part of a European research project LongITools.

Research publication: Heiskala, A., et al. (2025) Timing based clustering of childhood BMI trajectories reveals differential maturational patterns; Study in the Northern Finland Birth Cohorts 1966 and 1986. International Journal of Obesity. https://doi.org/10.1038/s41366-025-01714-8