In this large cohort of school-aged Finnish children, the burden of oral diseases was associated with increased risk of developing excess weight. Importantly, the association was independent of socioeconomic status, diet, and physical activity, suggesting a direct or indirect role of oral diseases in weight gain. Although non-significant, caries experience and gingivitis/calculus separately displayed slightly lower effect sizes with the same direction, thus increasing credibility of the findings. To our knowledge, this is a novel finding in this age group.

In previous studies, both caries and periodontal diseases have been associated with excess weight and central obesity in children [5, 6, 34, 35]. Here, we observed an increased risk between the burden of oral diseases (that is, having both caries experience and gingivitis/calculus) and developing excess weight during early adolescence, which may be a clinically relevant finding. In an older Finnish cohort, poor oral health in childhood (measured in 1980) associated with increases in BMI and waist circumference alongside other cardiometabolic conditions and risk factors such as a metabolic syndrome, increased blood pressure, and adverse lipid profiles during follow-up to adulthood [10, 11]. Diet and physical activity were not considered as confounders in those studies. However, caries alone was not associated with the development of BMI in that study, but initial weight status was not considered in the modeling. In a study on Chinese adolescents, a high level of caries experience predicted the development of central obesity in a 3-year follow-up period during late adolescence, but not during early adolescence [8]. In a 7-year follow-up study conducted in England, caries in early school-age was not associated with excess weight at adolescence [9]. That study did not consider the baseline weight status nor any health behaviors as covariates. In our study, the significant association between the burden of oral diseases and adiposity emerged after adjusting for covariates, e.g., shared risk factors. Then again, the estimates of the association were only marginally affected by the adjustment. The mean follow-up time of 2.3 years in our study was rather short, although it partially covers the quick growth stage from late childhood through early adolescence. With a longer follow-up time, the number of participants who developed excess adiposity may have been larger, likely strengthening the relationship we observed here.

Finns comprise a low caries population, which was evident in our study. Overall, 74% of children had no caries experience in permanent dentition, which is higher than the 61% found in a national sample of 12-year-olds in 2019 [36]. Then again, only one-third of the participants were free of gingivitis or dental calculus. Every fifth participant had both caries experience and gingivitis/calculus. The rather good oral health of our participants is likely explained by good oral hygiene and free-of-charge dental care for minors [37]. Fluoride toothpaste is commonly used in Finland [38], and 68% of Finnish fourth and fifth graders brushed their teeth twice daily [39]. We dichotomized the DMFT variable as the group sizes would otherwise be very uneven and small, hampering statistical power. This limited us from focusing on those children with multiple cavities, warranting further research in pediatric populations with a high number of cavities. Similarly, we dichotomized the CPITN variable. Deepened periodontal pockets (CPITN > 2) are rare in children [40] which limits the analysis to gingivitis and calculus. In the dental examination, the highest value in each sextant was reported, and the highest value of all the sextants determined the final CPITN value, meaning that those with the value of 2 were assigned as having calculus although they simultaneously could have had the value of 1 (gingivitis). We acknowledge that calculus is not a sign of gingivitis per se, yet they correlate [41]; thus, calculus was used as a proxy for gingivitis. Although calculus is not considered a disease, we used the word “disease” collectively for caries experience and gingivitis/calculus.

The strength of this study lies in the substantial number of children for whom data on a multitude of oral health and anthropometric measurements were available. Oral health data was obtained from a national register. However, DMFT and CPITN scores are relatively crude markers of caries experience and gingivitis/calculus and may be to some extent vulnerable to subjectivity of the professional. We limited our sampling to the dental check-ups performed by dentist or dental hygienist to overcome a part of the subjectivity. Nevertheless, the dental examinations between examiners were not calibrated. Moreover, as we cannot be certain that the teeth registered as missing were in fact unerupted permanent teeth, we repeated the analyses using the DFT index. These results were in line with the DMFT results although they did not reach statistical significance. The dental examinations were performed within 1 year before or after the Fin-HIT baseline data collection. Although some variation in the timing of the dental examinations existed, they took place before the follow-up on anthropometric measures, supporting temporal association.

We adjusted the analysis for self-reported baseline consumption of sweet treats and plants as well as for leisure-time physical activity as potential confounders possibly explaining the association to adiposity. In particular, the role of sugar is relevant and should be considered when examining the relationship between caries and excess weight [42]. The primary dietary determinant of caries is frequent consumption of sugar that cariogenic bacteria ferment in the oral cavity, leading to acidification of the dental biofilm and eventually to caries lesions if not controlled by remineralization enhancing activities especially tooth brushing with fluoride toothpaste [14]. In contrast, for youth obesity outcomes, the evidence of sugar seems to limit to sugar-sweetened beverages [43]. In our study, sweet treat consumption did not associate with anthropometric outcomes (data not shown), but regardless, residual confounding may exist. In addition, the Fin-HIT cohort likely reflects a sample with a slightly higher SES compared with the overall Finnish population [44] which may indicate a more health-conscious sample with lower rates of excess weight and better oral health compared with socially deprived samples. Loss to follow-up is common in prospective studies and was witnessed here as 46% of the initial cohort missed the follow-up data collection. Of the participants with oral health data available, 59% were either lost to follow-up or excluded due to missing questionnaire or anthropometric data. This group exhibited higher proportion of excess weight, central obesity, caries experience, and burden of oral diseases than those who were included in the current sample (Table S2). They also had lower maternal SES. Taken together, these limitations may decrease the generalizability of our findings to the broader population. However, given the direction of our findings, the associations we observed would probably be stronger in a more representative sample. In addition to the large proportion of participants lost to follow-up, a further limitation might be the rather low number of participants who developed excess weight or central obesity. This can result in an underpowered study and type II error. However, type II error means not rejecting the null hypothesis when it in fact is false. Even though not statistically significant, all hazard ratios in our analysis pointed toward a positive association of oral diseases with developing excess weight. This suggests that findings should be replicated with larger sample sizes or longer follow-up times. Moreover, we were limited to the use of maternal occupation at the time of child’s birth as the indicator for SES. Parental education and income levels might be more informative factors to express family’s SES than occupation alone.

To conclude, only a quarter of participants had caries experience whereas gingivitis/calculus were prevalent in two-thirds of the sample. Every fifth participant had both caries experience and gingivitis/calculus. The burden of having both caries and gingivitis/calculus predicted the development of excess weight but not that of central obesity when moving from late childhood to early adolescence. Having burden of oral diseases without excess weight in early adolescence could imply future weight gain and, thus, could be used to identify individuals at risk of gaining excess weight. Oral diseases and excess weight share risk factors, highlighting that our findings warrant further research to explore whether oral diseases and weight gain merely share risk factors or if their relationship is of causal nature.