Both oral and general health status depends on a dynamic interplay, including the individual’s behaviour, particularly eating behaviours. Behavioural and lifestyle-related factors are significant contributors to dental caries initiation and development [1].

Despite improvements in preventive measures, dental caries remains highly prevalent, especially among specific groups. As a chronic disease, dental caries reflects the cumulative effect of harmful socio-economic, biological, and behavioural factors that occurred during the lifespan of a person [2, 3].

Based on the impact of dental health status on food choice [4–6], the association between dental caries and nutritional status was previously explored [7–10]. The exact underlying mechanism for the relationship is not clear. Untreated dental caries may cause changing eating patterns, but, eating patterns may increase the individual risk of developing more dental decay.

Until now, two hypotheses regarding the association between dental caries and BMI have been suggested: a positive association [5, 7–9] and a null association [10].

Most of these studies have used only a non-parametric approach [5, 7–10], by treating BMI as a categorical variable. However, Royston, et al. pointed out a number of disadvantages with this approach [11]. The most significant drawback is the loss of information and strength through what is equivalent to rounding.

Additionally, most studies have focused on adolescents and ignored the cumulative effect of unfavourable eating behaviour among adults.

Based on the assumption that untreated dental caries might decrease as well as increase energy intake, a U-curved association in which people with underweight and overweight report more untreated dental caries compared to people with a normal weight, may be expected. Therefore, the purpose of this study was to investigate whether a non-linear association may be expected between the number of untreated dental decay and BMI by maintaining BMI as continuous variable.

In addition, we examined whether socioeconomic factors may moderate this association among adults attending a tertiary health care centre in Morocco.

The mean age of participants was 27 ± 8. Of the 101 participants, 69 (68.31%) were female. The BMI of 35 (34.65%) of subjects was below normal and the overall mean BMI was 22.31 ± 4.47.

The mean of teeth affected by dental caries was 5.37 ± 3.66, the median (IQR) of filled teeth was 2 (0–6).

In univariate analysis, it was found that low and medium educational level (PR = 1.65, 95% CI 1.09–1.68; 1.49, 1.23–1.81 respectively), low and medium income (PR = 1.65, 95% CI 1.30–2.07; 1.25 95% CI 1.02–1.52), gingival index (GI) (PR = 1.19, 95% CI 1.06–1.33), age at beginning tooth brushing (PR = 1.16, 95% CI 1.08–1.24), underweight (PR = 1.58, 95% CI 1.29–1.94), overweight and obesity (1.28, 95% CI 1.02–1.60) frequency of tooth brushing (0 or 1/day and 2/day) were risk factors associated with the increased rate of untreated dental caries. At the same time, an appropriate motion of brush (PR = 0.65, 95% CI 0.54–0.77) and planned visit (PR = 0.73, 95% CI 0.54–0.94) were protective factors.

Adjusted multivariate analysis revealed that only age at beginning tooth brushing (PR = 1.15, 95% CI 1.05–1.25), BMI < normal (PR = 1.66, 95% CI 1.30–2.12), BMI > normal (PR = 1.30, 95% CI 1.03–1.65), SDI (PR = 0.61, 95% CI 0.50–0.75) and GI (PR = 1.59, 95% CI 1.30–1.94) remained associated with the increase of the number of dental caries.

Figure 1 shows the adjusted rate ratios and confidence interval over the BMI values.

A significant quadratic effect between categorical BMI and the rate of untreated dental decay (the p values for non-linearity was <0.001 and for the overall effect p < 0.001) was observed.

This means that the number of untreated dental decay is a positive quadratic trend in the association with BMI values. So these data suggest a U-shaped trend in the association between untreated dental caries and BMI.

The present study indicated that adults with both low and high BMI have poorer dental health as manifested by more untreated caries compared with normal weight patients. This association was observed in both univariate and multivariate analyses and shown U-shaped distribution.

The association between unhealthy lifestyle and chronic disease continues to be a greater focus in public health because it is associated with substantial morbidity and mortality [15, 16].

Data from several population-based studies regarding the association between dental caries and BMI are conflicting. Patterns of previous reported relationships included no association [10] between BMI and dental caries, a positive association [17] and an inverse relationship [9, 18].

It should be stated that a majority of studies reporting positive association explored the association between obesity and dental caries [17, 19]. Most studies focus on linear (positive–negative) association.

Some studies suggest that dental caries is associated with both high and low body mass index [20, 21]. The present finding of a U-shaped association of BMI with dental caries suggest that this pattern of association is plausible.

We propose that association between BMI and dental caries may be explained thorough common explanatory factors. Unhealthy diets with frequent consumption of refined sugars and added sugars and low in fruit and vegetables may increase the level of both BMI [21] and dental caries. On the other hand, decreased food intake may be attributable to a high level of untreated dental caries and then explain frequency of untreated dental caries [9] among low-BMI adults.

BMI shares eating behaviours and social etiological factors with dental caries. Socio-economic factors are also indicators of individual behaviour regarding the types of food intake and may influence the dietary preference [22, 23].

Therefore, eating behaviours may mediate the link between BMI and dental health. Individuals with high risk should be encouraged to undertake lifestyle modifications and improvements in a dental health status diet.

An established relationship has been reported between brushing behaviour, educational level and the rate of dental caries. Findings of the present study corroborate these results [24].

Our results should be interpreted in light of both the study’s strengths and its limitations. High levels of dental decay were detected in the present study, which indicated a high-risk study population. Nonetheless, the associations detected may be generalizable to other study populations even if geographical inequalities in dental caries experience were reported [2, 24].

Unhealthy dietary patterns are prevalent among high-income countries and have emerged as public health conditions among low- and middle-income countries.

Such studies may then be crucial for estimating whether reductions in the prevalence of dental caries among specific groups may affect some populations’ morbidities like high and low BMI. Moreover, demand for better health systems throughout the world should integrate the possible co-existence of some pathologic conditions for common preventive measures.

Another limitation is the study design, which did not allow for assessment of temporality of associations. However, to our knowledge, this is the first study exploring the non-linear association between untreated dental caries and BMI as a categorical variable.

Future studies should explore more potential factors associated with unmet dental needs (dental insurance, barriers to access, psychological factors, etc.). In addition, dietary habits (composition and intake frequency) need to be considered.

In conclusion, a U-shaped trend in the association between dental decay and BMI was found. This means that an increased rate of untreated tooth decay was associated with both under- and overweight.