We observed, using a representative sample of people ≥70 years old who resided in one district of Mexico City, that the perception of having worse oral health than others of the same age and the failure to utilize dental services were associated with an increased probability of being frail. This is the first study that has explored frailty in the Mexican elderly and included oral health components. The multidisciplinary approach of this project allows us to explore the relationships between frailty and many other variables
There are few previous data on the association between oral health and some components of the frailty syndrome, but not with the frailty syndrome overall
There was an association between not using dental services during the past year and a higher probability of being frail and (OR
2.1, 95% CI 1.2–3.8); Overall, the utilization of dental services among the elderly is lower
] (49% in this study) than the utilization of medical services (85%)
]. According to Kiyak
], elderly people who believe that declining health is a part of aging and that recovery is not possible, reflect that attitude by not utilizing dental services. Taking this into account, the observed lack of utilization of dental services might also reflect compromised general health, which in turn may be correlated with frailty.
Another possible explanation could be constriction of life space, which has been recognized as a risk factor for frailty
]. Constriction of life space is attributable to loss of mobility, weakness, and slowness, which result in difficulties with moving and travelling independently to receive dental care. This, in addition to low expectations about oral health, could contribute to this population’s failure to utilize dental services.
The probability of being frail was 2.2 times higher for those who considered their oral health worse than others of the same age; this could be explained by the positive association between oral health and general health
]. We can assume that a bad perception of oral health reflects declining general health.
Oral health is recognized as a component of general health
], and it is unusual for people to identify oral health conditions as their only health problems when those issues are so advanced that they affect feeding ability, speaking, chewing, physical appearance, and social life, frequently producing pain and favoring depression
]. Also, when oral health is poor, inflammation markers are increased, which in turn can alter the metabolism of other organs; even though self-perception of oral health is subjective, it can represent a risk indicator for frailty.
Oral health expectations among the elderly could also be influenced by age, education level, socioeconomic situation, and social support
], variables that have been associated with frailty.
Even though the presence of severe periodontitis did not surpass the threshold of statistical significance as a predictor of frailty (OR
3.9, 95% CI 0.98–15.6) in the final logistic model, the physiological process related to the development of periodontal infection could be related to the development of frailty. It has been shown that severe periodontitis is linked to energy imbalance, which in the elderly has been associated with loss of mobility and strength alterations that end with the development of frailty syndrome. It has been shown that higher concentrations of pro-inflammatory cytokines could favor the altered inflammation state observed in frailty syndrome
]. It is important to point out that periodontitis requires the presence of teeth; in the logistic regression model, we included edentulous subjects in the “no severe periodontitis” group in order to preserve the sample size, but in a later analysis considering only completely or partially dentate (1–32 teeth) subjects, we observed that those with severe periodontitis have 5.3 times the risk for frailty than those without severe periodontitis (95% CI 1.3–22.2). The participation of severe periodontitis in the development of frailty should be explored in the longitudinal analysis.
Other studies have explored the relationship between dental variables and frailty: Avlund et al.
] reported that having few teeth increases the risk of fatigue in the elderly; also, Semba et al.
] reported that frail edentate women with complete dentures who complain about chewing problems have higher mortality rates than frail edentulous women who reported no chewing problems. In this study, we did not find an association between the number of teeth and frailty or between chewing problems and frailty; we must consider that in our sample the prevalence of overweight and obesity was 75%, which could dissipate the effect of tooth loss on frailty among the elderly in Coyoacán. On the other hand, elderly people living in Mexico who have few (1-9) teeth and who wear nonfunctional dental prostheses tend to modify their diets, increasing the consumption of low-nutrient foods. This could explain the fact that many of them are overweight and obese, and it could hide the effects of having few or no teeth on the risk of frailty and mortality
]. It is important to mention that obese sarcopenia could be a confounding factor in the evaluation of frailty
Several researchers have measured dental variables with physical activity. Takata and cols., explored the association between chewing ability and number of teeth with physical activity using hand grip strength, leg extensor strength, leg extensor power, stepping rate, and one-leg standing time as indicators. These are specific measurements of physical activity, one of the components of frailty. However, comparisons with our study are difficult because we used The Physical Activity Scale for the Elderly (PASE)
], an instrument designed for screening for low physical activity in the elderly. Also, we measured frailty as a whole and not considering the components individually. Even when the associations of chewing ability and number of teeth with frailty were statistically significant in the univariate model, in the logistic regression model the associations disappeared. Similarly, Takata did not find association with number of teeth; but when comparing the number of chewable foods they observed an increase in isokinetic leg extension power and in one-leg standing time, association that we did not observed with frailty. The type of measurement for physical activity used in this study might be a limitation, it is worth considering in future studies a more detailed measurement
It is important to mention that this study included clinical evaluation of dental variables as well as questions about self-perception of oral health and self-reports of utilization of dental services during the last year. None of these variables has been considered in other reports on frailty. However, the inclusion of several oral health conditions represents a challenge because of the mutually exclusive characteristics of some dental variables (e.g., periodontitis and edentulism).
We recognize the need for a global oral health indicator, and we question, “What would be an optimal oral health indicator in the elderly, and how can we measure it?” A global oral health indicator should be able to classify those people who are properly rehabilitated (using functional dental prosthesis) and have no periodontal problems. It should also be able to classify those with complex oral conditions (e.g., nonfunctional dental prosthesis and periodontal disease in the remaining teeth, or presence of remaining roots), which could increase the risk of developing other chronic conditions, favoring energy imbalance and creating clinical and therapeutic challenges.
The covariates measured in this analysis were consistent with those measured in previous reports; we observed a higher risk of frailty among those who had been hospitalized during the previous year, those who consumed higher number of drugs on a daily basis, those who reported incontinence, and those who had a history of myocardial infarction
Due to the cross-sectional nature of the study design, causal inferences cannot be made. In order to deepening in the relationship between oral health status and frailty, it is necessary to perform longitudinal studies.
Even though the non-response rate was 25% for the dental clinical evaluation, many characteristics of those who accepted and those who did not accept the oral clinical evaluation were similar. Similarities between responding and non-responding groups were found with respect to sociodemographic characteristics such as age and gender, medical variables, dental services utilization, and self-perception of oral health. The number of years of education completed was higher among those who did not accept the oral clinical evaluation. As a whole, these results imply that there are no differences between subjects who were clinically evaluated and those who were not. Certain oral health-related variables, such as coronal and root caries and oral mucosal lesions, were not evaluated because of time constraints on the clinical evaluations. Furthermore, it is possible that the mutually exclusive relationships between some clinical dental data (e.g., periodontitis and edentulism) interfere with the inclusion of the variables in the analysis. Another limitation could be the type of measurements used for frailty; we used epidemiological screening methods recommended for the elderly population. In order to have more accurate results it can be considered the utilization of more detailed measurements for the components of frailty in future studies.