These results show that a Ca:P ratio above 0.57 in the habitual diets of subjects was not associated with global obesity (BMI) but was negatively associated with central obesity based on WHtR, which is considered risk factor for cardiovascular diseases [32, 33]. Habitual dietary calcium intake above the median value (485.4 mg) was negatively associated with central obesity based on WHtR becoming a protective factor. The same association occurred regarding the consumption of dairy products, although with a greater protective effect.
No study relating the Ca:P ratio in the habitual diet with obesity was found in the literature. However, a higher dietary calcium intake has been identified by some researchers as a protective factor for overweight [6, 7, 42–44]. The consumption of phosphorus, in addition to being associated with lower calcium absorption [5, 45, 46], was positively associated with global obesity and central obesity based on WC in the study by Beydoun et al . We should mention that this was the only study found that related phosphorus intake to obesity. The Ca:P ratio median in this study was used as the cutoff point, from which a protective effect for central obesity was observed, and this median was much smaller than the value suggested in the literature for bone health promotion [15, 21].
In the above study , which reported calcium intake as a protective factor and phosphorus as a risk factor for obesity, the mean Ca:P ratio in the habitual diet was 0.65. This average value was calculated by us from data on average calcium and phosphorus intake measured during the study; however, this ratio was not discussed in that study as a factor that affects obesity.
Kemi et al.  studied the effects of this ratio on calcium metabolism and serum parathyroid hormone and also found values lower than those suggested [15, 21] (average 0.74). They also observed that values below 0.65 were deleterious whereas values above 0.65 were not. Although this study focused on bone health in a population with calcium intake that was sufficient or above the requirements, the ratio found in both studies was well below that suggested in the literature, leading us to believe that even low Ca:P ratios have a beneficial effect. This beneficial effect was observed in both the bone health study and the present study on obesity. It is noteworthy that among the subjects we studied, none had a Ca:P ratio greater than or equal to the value suggested in the literature of 1.3:1.0 [15, 21].
With regard to calcium, approximately 84% of the subjects did not meet the requirements for this nutrient. A higher proportion (99%) was found in a study conducted in Brazil with individuals over 40 years of age . Several researchers have also found insufficient intake of this mineral in other countries [8, 10, 47]. In a study assessing the calcium intake of a population over 2 years of age, it was found that only 32.30% of subjects showed adequate intake of this nutrient. In the adult population, a similar proportion was found: 38.80% for the population aged 19 to 30 years and 33% for the population aged 31 to 50 years .
It is important to note that the DRIs used in this study are lower than those used in the previously mentioned studies because of recent updates in the requirements for this nutrient, thereby reducing the percentage of individuals who did not meet these needs in relation to previous studies.
The opposite situation has been found for phosphorus intake, which is 2 to 3 times higher than required [14, 15]. Beydoun et al.  associated the consumption of dairy products and nutrients with obesity, central obesity and metabolic syndrome and found that, for each daily intake of 100 mg of phosphorus, the prevalence of global obesity (BMI) and central obesity (WC) increased by 7% and 6%, respectively. In the present study, average phosphorus intake was 1.6 times greater than the requirements, and the highest daily amount ingested was 5 times higher than the requirements. However, despite its high consumption, phosphorus was not related to any obesity diagnostic parameter, suggesting that the protective effect achieved by the high Ca:P ratio (which could be achieved by a higher calcium intake and lower phosphorus intake) was because calcium alone was protective for central obesity based on WHtR. A similar relationship occurred between dairy consumption and central obesity based on WHtR, which corroborates the above discussion because this food group is not only a good source of phosphorus but also of calcium, unlike many other foods that contain large amounts of phosphorus but low amounts of calcium.
Most researchers who have linked the consumption of calcium and dairy products with central obesity based on waist circumference, which is the most widely used parameter, found an inverse relationship [8, 43, 48–50]; however, in the present study and the study by Brooks et al.  this relationship was not found.
This study also investigated the association between food intake variables and the waist-to-height ratio as a parameter for diagnosing central obesity and found a relationship between Ca:P ratio in the habitual diet, calcium intake and dairy products. Although the WHtR is considered a parameter for diagnosing central obesity, which when present represents a risk factor for cardiovascular diseases [32, 33, 52, 53], no studies were found in the literature relating this parameter with the intake variables mentioned. This finding is another new piece of evidence provided by this study.
The consumption of skim milk was positively associated with global obesity. However, the vast majority of the study population did not consume this food, resulting in a consumption median equal to zero, which prevented further statistical analyses. Studies with a larger number of consumers should be performed.
This study has the following limitations: the absence of information on comorbidities of individuals and its cross-sectional design, which does not allow us to establish a cause-and-effect relationship. However, this type of study model determines risk factors and identifies previously unknown relationships in groups or in the general population. Other limiting factor was the exclusion of 17.67% of the sample for dietary analyses due to incomplete data. Finally, the individuals studied did not reach a Ca:P ratio greater than or equal to values suggested in the literature, thus preventing a comparative analysis with individuals who have met these recommendations.
Regarding its strengths, this study has a stratified and systematic sampling design and uses recent data, allowing the representation of the adult study population. This study also used a validated handbook for the study population, with measures of food prepared at home, thus facilitating a more accurate quantification of portion sizes consumed. Another positive factor was the coordination of home visits by master’s students and weekly meetings of all staff with the research coordinators throughout the training and data collection periods, facilitating the standardization of the methodology used throughout the study.