SciELO - Scientific Electronic Library Online

 
vol.49Socioeconomic inequalities and changes in oral health behaviors among Brazilian adolescents from 2009 to 2012Users’ dissatisfaction with dental care: a population-based household study author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand

Journal

Article

Indicators

Related links

Share


Revista de Saúde Pública

Print version ISSN 0034-8910On-line version ISSN 1518-8787

Rev. Saúde Pública vol.49  São Paulo  2015  Epub Aug 07, 2015

http://dx.doi.org/10.1590/S0034-8910.2015049005590 

Artigos Originais

Gender differences in the association between tooth loss and obesity among older adults in Brazil

Diferenças de gêneros na associação entre perda dentária e obesidade entre idosos brasileiros

Ankur Singh I  

Marco Aurélio Peres I  

Karen Glazer Peres I  

Carla de Oliveira Bernardo II  

Andre Xavier II   III  

Eleonora D’Orsi II  

IAustralian Research Centre for Population Oral Health. School of Dentistry. The University of Adelaide. Adelaide, SA, Australia

II Programa de Pós-Graduação em Saúde Coletiva. Centro de ciências da Saúde. Universidade Federal de Santa Catarina. Florianópolis, SC, Brasil

IIIFaculdade de Medicina. Universidade do Sul de Santa Catarina. Tubarão, SC, Brasil

ABSTRACT

OBJECTIVE

To analyze if differences according to gender exists in the association between tooth loss and obesity among older adults.

METHODS

We analyzed data on 1,704 older adults (60 years and over) from the baseline of a prospective cohort study conducted in Florianopolis, SC, Southern Brazil. Multivariable logistic regression models were used to assess the association between tooth loss and general and central obesity after adjustment for confounders (age, gender, skin color, educational attainment, income, smoking, physical activity, use of dentures, hypertension, and diabetes). Linear regressions were also assessed with body mass index and waist circumference as continuous outcomes. Interaction between gender and tooth loss was further assessed.

RESULTS

Overall mean body mass index was 28.0 kg/m2. Mean waist circumference was 96.8 cm for males and 92.6 cm for females. Increasing tooth loss was positively associated with increased body mass index and waist circumference after adjustment for confounders. Edentates had 1.4 (95%CI 1.1;1.9) times higher odds of being centrally obese than individuals with a higher number of teeth; however, the association lost significance after adjustment for confounders. In comparison with edentate males, edentate females presented a twofold higher adjusted prevalence of general and central obesity. In the joint effects model, edentate females had a 3.8 (95%CI 2.2;6.6) times higher odds to be centrally obese in comparison with males with more than 10 teeth present in both the arches. Similarly, females with less than 10 teeth in at least one arch had a 2.7 (95%CI 1.6;4.4) times higher odds ratio of having central obesity in comparison with males with more than 10 teeth present in both the arches.

CONCLUSIONS

Central obesity was more prevalent than general obesity among the older adults. We did not observe any association between general obesity and tooth loss. The association between central obesity and tooth loss depends on gender – females with tooth loss had greater probability of being obese.

Key words: Aged; Tooth Loss, epidemiology; Obesity, epidemiology; Gender and Health; Socioeconomic Factors; Risk Factors; Cross-Sectional Studies

RESUMO

OBJETIVO

Analisar se há diferenças entre gêneros na associação entre obesidade e perda dentária em idosos.

MÉTODOS

Foram avaliados dados de 1.704 idosos (60 anos ou mais) da linha de base de um estudo de coorte prospectivo realizado em Florianópolis, SC. Modelos de regressão logística multivariáveis foram realizados para aferição da associação entre perda dentária e obesidade geral e central, ajustados por variáveis de confusão (idade, gênero, cor da pele, educação, renda, tabagismo, atividade física, uso de prótese dentária, hipertensão e diabetes). Na regressão linear, índice de massa corporal e circunferência da cintura foram tratados como variáveis contínuas. Foi avaliada também a interação entre gênero e perda dentária.

RESULTADOS

A média geral do índice de massa corporal foi 28,0. A média de circunferência da cintura foi de 96,8 cm para homens e 92,6 cm para mulheres. O aumento da perda dentária associou-se positivamente com o aumento do índice de massa corporal e da circunferência da cintura, após ajuste por variáveis de confusão. Edêntulos apresentaram chance 1,4 (IC95% 1,1;1,9) vez maior de apresentarem obesidade central quando comparados com aqueles com maior número de dentes presentes; entretanto, a associação perdeu significância estatística após o ajuste pelas variáveis de confusão. Comparadas com homens edêntulos, mulheres edêntulas apresentaram prevalência ajustada de obesidade geral e central duas vezes maior. No modelo de efeitos conjuntos, mulheres edêntulas tiveram uma razão de chances 3,8 (IC95% 2,2;6,6) vezes maior de apresentarem obesidade central, quando comparadas com homens com mais de 10 dentes presentes em ambas as arcadas. De maneira similar, mulheres com menos de 10 dentes presentes em pelo menos uma arcada tiveram uma razão de chances 2,7 (IC95% 1,6;4,4) vezes maior de terem obesidade central quando comparadas com homens com mais de 10 dentes presentes em ambas arcadas.

CONCLUSÕES

Obesidade central foi mais prevalente que obesidade geral em idosos. Não foram encontradas associações entre obesidade geral e perda dentária. Associação entre obesidade central e perda dentária depende do gênero – mulheres com perdas dentárias apresentaram maior probabilidade de serem obesas.

Palavras-Chave: Idoso; Perda de Dente, epidemiologia; Obesidade, epidemiologia; Gênero e Saúde; Fatores Socioeconômicos; Fatores de Risco; Estudos Transversais

INTRODUCTION

The World Health Organization (WHO) reports that the worldwide prevalence of obesity doubled between 1980 and 2008.27Identified as a growing global public health burden, obesity is associated with increased risk of cardiovascular diseases, diabetes, cancers and other chronic diseases.4

The prevalence of obesity is increasing in all age groups, including older populations throughout the world.10 When compared with the younger age groups, the absolute rise in mortality rates associated with obesity is greater among older adults. Reviews on obesity in older adults have reported its association with hypertension, diabetes, cardiovascular diseases, and stroke as well as lower cognitive skills, frailty, degenerative osteoarthritis, sexual dysfunction, urinary incontinence, and renal disease.26 Obesity in older adults is further associated with physical immobility,8 which can severely affect their quality of life by restricting movements in day-to-day life.

Ageing also leads to greater susceptibility to loss of teeth due to a cumulative effect of previous oral diseases.24 Because tooth loss reduces masticatory functions, older adults tend to adapt their dietary intake for ease of eating food items.14 Consequently, a decrease in fibrous fruits and vegetable and an increase in the intake of soft processed food may occur.15 Thus, older adults with tooth loss have higher intake of fat, saturated fat and cholesterol, which may lead to obesity.21,23

Studies have assessed and established associations between tooth loss and obesity among adults2,6,11,16 using different measurements.6,11,13,22,25 Central (or abdominal) obesity is widely measured by waist circumference (WC) or waist to hip ratio. Even with normal body mass index (BMI), elevated WC measures can mean a two to threefold increase in the risk of cardiovascular disease and premature death.20 Fewer studies have assessed the association between tooth loss and obesity using the measurement of WC or both WC and BMI.2,17,18

Patterns of weight and its progression to obesity differ in females and males.9 WHO statistics report that females are more obese than males in all WHO regions.27 This highlights the need for studies to consider gender wise variations while studying obesity and its risk factors. Gender wise differences have also been reported in tooth loss in Brazil5 and globally.18 In a Swedish study, women showed stronger association between edentulism and obesity than men.19 Despite Sweden and Brazil differ in cultural context, no Brazilian study has assessed the role of gender in the association between tooth loss and obesity in older adults. Furthermore, no other study has assessed the role of gender in this association for this age group in any lower and middle income country. As a study has previously assessed and reported the association between tooth loss and obesity in a younger sample in Florianopolis,2 and considering the significant gap in the literature, the current study aimed to analyze if differences according to gender exists in the association between tooth loss and obesity among older adults.

METHODS

In this cross-sectional study, we analyzed the baseline data of a population-based prospective cohort, EpiFloripa Aging, carried out with older adults aged 60 years or over in Florianopolis, Southern Brazil, from September 2009 to June 2010.

A two-stage cluster sample selection was drawn. Eighty out of 420 census tracts were ordered according to mean monthly income of the family head, and a systematic selection was conducted, allowing the selection of eight census tracts in each decile of monthly income. The number of households in these census tracts was counted and updated before data collection by fieldworkers, who estimated the number of households to be visited to achieve 1,911 individuals. The households were systematically selected and every older adult was invited to participate.

All individuals aged 60 years or over residing in the selected households were eligible to participate. Institutionalized older people were excluded from the study. Non-response was defined as either four unsuccessful interview attempts (due to, for example, people being away from home) or refusals. There were no substitutions. Out of the 1,911 invited individuals, a total of 1,705 interviews, health examinations and anthropometric measurements were completed. A detailed explanation about the study sampling has been published elsewhere.3

Thirty-four trained interviewers carried out data collection by applying a pre-coded structured questionnaire with personal digital assistants. A pilot study was conducted with 99 individuals within census tracts not selected for the study sample. Data quality control consisted of applying a short version of the questionnaire by telephone with approximately 10.0% of the sample (n = 150). Reliability measures were assessed using kappa statistics.

The main exposure was tooth loss. The number of self-reported natural teeth for each dental arch was recorded as follows: 10 or more natural teeth, less than 10 natural teeth, no natural teeth (edentate). We categorized the loss of teeth as: 10 or more natural teeth in both arches, less than 10 teeth in at least one arch, and edentate (no natural teeth in both arches).

The measures of obesity included BMI and WC. Weight (kg) was measured twice with portable scales of 100 g graduation calibrated before training and fieldwork. Height (m) was the mean of two measures obtained with a stadiometer of 1 mm graduation. Participants were asked to stand in front of the stadiometer without shoes and wearing light clothes. BMI was estimated as weight divided by square height. WC was assessed twice using an anthropometric inelastic tape with 1 mm markings (Sanny®), measured at the narrowest waist level, or, if this was not apparent, at the midpoint between the lowest rib and the top of the iliac crest. The examiners were instructed to make sure that the tape was not too tight or too loose and that it was lying flat and on the skin and horizontal when recording. The individual’s WC considered was the mean of both measurements.

We analyzed general obesity using BMI as a binary variable (obese: ≥ 30 kg/m2 and non-obese: < 30 kg/m2) and central obesity using WC (obese: ≥ 102 cm for males and ≥ 88 cm for females; non-obese: < 102 cm for males and < 88 cm for females). We used BMI and WC as continuous variables.2

Covariates included sex, age (60 to 64; 65 to 69; 70 to 74; 75 to 79; and ≥ 80 years); educational attainment (≤ 4; 5 to 8; 9 to 11; ≥ 12 years), equivalized monthly family income (divided into quartiles), and self-reported skin color/race (white, parda [mixed race involving African ancestry], black, Asian or indigenous). To assess equivalized income, the total household income was divided by the square root of the total number of household members and then divided into tertiles. The first tertile included individuals with incomes ranging from 0.00 to 452.00 BRL; the second, from 453.00 to 1,130.00 BRL; and the third, 1,130.00 BRL or more (1.00 USD corresponded to approximately 1.70 USD). The statistical models also included the following variables as covariates: self-reported use of dental prosthesis, diabetes, hypertension, and smoking status (never smoked; former smoker; current smoker); physical activity in leisure time measured by the long version of the International Physical Activity Questionnaire (IPAQ), adapted and validated for Brazilian older adults1 (inactive: less than 10 min weekly; insufficiently active: 10 to 149 min weekly; and active: active for up to 150 min weekly).

Unadjusted and adjusted linear and logistic regression analyses were used to assess the associations between tooth loss and obesity. Adjusted analyses were performed step by step. First, we included tooth loss and demographic variables (age and skin color/race); second, we added socioeconomic variables (educational attainment and income). Smoking and physical activity were included in the third step, followed by the addition of self-reported use of dental prosthesis in the fourth model. We further added self-reported hypertension and diabetes in the next model. The associations were adjusted for gender in the final model. In the multivariable analysis, p ≤ 0.20 in the bivariate association between each covariate and the outcome was used as an entry criterion. Interaction between gender and tooth loss was graphically displayed and assessed in the context of multivariable modeling, controlling for educational attainment, income, skin color/race, smoking, physical activity, use of dentures, hypertension and diabetes. The joint effect of tooth loss and gender was further estimated by fitting logistic regression models for the outcome of obesity with a composite variable of tooth loss and gender, and adjustment for the abovementioned confounders. Survey commands were used to control the design effect and perform weighted analysis. Statistical significance in the models was determined by Wald’s test and p < 0.05 was considered statistically significant. All analyses were done using Stata v.11. This study was approved by the research ethics committee at Universidade Federal de Santa Catarina (Process 352/08). All participants signed an informed consent form.

RESULTS

The mean equivalized family income per capita of the sample was 1,514.16 BRL (890.70 USD). Many participants had less than 10 teeth in at least one of the dental arches (39.7%), while 33.5% were edentate and 26.8% had 10 or more teeth in both dental arches. Loss of teeth was significantly associated with gender, age, skin color/race, educational attainment, income and hypertension (p < 0.05) (Table 1).

Table 1 Sample characteristics according to general and central obesity prevalence, Florianopolis, Southern Brazil, 2009 to 2010. (N = 1,704) 

Characteristic n
%*
General obesity
Central obesity
    % 95%CI % 95%CI
Gender            
Male 616 37.6 21.2 17.2;25.9 35.6 31.5;39.9
Female 1,088 62.4 37.1 34.3;39.9 65.3 61.6;68.8
Age years            
60 to 64 470 28.0 32.4 27.8;37.5 50.7 45.9;55.5
65 to 69 384 23.1 33.6 27.3;40.6 53.9 47.4;60.3
70 to 74 340 19.5 31.9 26.5;37.8 59.1 51.9;65.9
75 to 79 272 15.8 26.2 20.7;32.4 53.0 46.5;59.4
≥ 80 238 13.6 28.7 22.6;35.7 55.8 47.9;63.3
Self-reported skin color            
White 1,444 86.8 30.8 28.2;33.4 55.1 52.2;58.0
Parda 131 7.2 32.3 25.3;40.1 46.9 39.4;48.6
Black 84 4.3 28.7 17.7;43.1 41.3 28.6;55.3
Asian 12 0.7 19.7 4.2;58.0 13.7 1.8;57.5
Indigenous 17 1.0 35.2 15.3;62.0 71.0 42.9;88.9
Educational attainment (years of schooling)            
≤ 4 745 40.6 34.6 30.6;38.8 57.3 52.1;62.4
5 to 8 321 18.5 28.3 23.0;34.3 50.6 44.4;56.7
9 to 11 234 15.9 29.3 23.4;36.0 48.4 40.4;56.4
> 12 393 25.0 28.6 23.3;34.5 55.5 49.3;62.5
Income            
3rd tertile 568 35.5 29.2 24.8;34.1 53.6 48.0;59.1
2nd tertile 566 33.1 33.5 29.0;38.3 55.5 50.6;60.4
1st tertile 571 31.4 30.6 26.3;35.4 53.4 48.7;58.0
Number of natural teeth            
≥ 10 in both arches 395 26.8 29.1 22.9;36.2 49.8 42.5;57.0
< 10 in at least one arch 673 39.7 31.9 28.2;35.9 52.8 48.4;57.1
Edentate 582 33.5 32.0 28.2;36.1 58.6 53.4;63.6
Smoking            
Never smoked 1,038 59.6 34.1 30.9;37.5 58.5 54.2;62.6
Former smoker 523 32.0 29.5 24.6;34.9 50.9 47.2;54.5
Current smoker 141 8.4 15.8 10.0;24.1 35.8 27.0;45.6
Physical activity in leisure time            
Inactive 932 53.2 34.2 30.9;37.7 55.6 51.8;59.3
Insufficiently active 279 16.0 35.1 28.9;41.8 60.1 53.3;66.5
Active 494 30.8 23.6 19.6;28.2 48.6 42.4;54.9
Self-reported use of dental prosthesis            
Yes 944 51.6 30.9 27.8;34.2 55.4 52.0;58.7
No 760 48.4 31.3 27.2;35.6 52.8 48.4;57.2
Hypertension            
No 698 42.0 20.3 16.2;25.1 42.6 37.2;48.2
Yes 1,007 58.0 38.9 35.8;42.1 62.5 59.3;65.6
Diabetes            
No 1,329 78.4 26.7 24.0;29.5 49.6 46.1;53.1
Yes 376 21.6 47.1 40.6;53.6 70.7 64.9;75.9
General obesity            
Non-obese 1,179 68.9        
Obese 525 31.1        
Central obesity            
Non-obese 799 45.9        
Obese 905 54.1        

* All percentages are weighted.

Overall mean BMI was 28 kg/m2 and overall mean WC was 96.8 cm for males and 92.6 cm for females. We observed a higher prevalence of central obesity (54.1%) than general obesity (31.1%). Females were more obese than males and current smokers were less obese than former smokers as well as non-smokers. Central and general obesity were significantly less prevalent among those who reported to be physically active in leisure time when compared with those who reported insufficient activity or inactivity. Both central and general obesity were significantly higher among those who reported hypertension or diabetes and among those who were not using dentures, compared with those using dentures (Table 1).

The crude estimates from logistic regression showed that the odds of edentate individuals having general or central obesity were, respectively, 1.2 and 1.4 times higher than individuals with more than 10 teeth in each arch. We observed significant associations. Edentulousness was associated with central obesity (p < 0.05) but not with general obesity (p > 0.05). The association between central obesity and edentulousness remained significant after adjusting for demographic variables (age and skin color), but became marginally insignificant on inclusion of socioeconomic variables (income and educational attainment). The association regained its significance after adjustment for smoking and physical activity. However, this association became marginally insignificant after the inclusion of self-reported use of dentures in the next model. The association could not regain the statistical significance after the addition of hypertension and diabetes in the following model. The addition of gender reduced edentates’ odds of having central obesity (OR = 1.3, 95%CI 0.8;2.1) in the final model. The odds of individuals with less than 10 teeth in one arch to have general or central obesity were 1.1 times higher than those with 10 or more teeth in both arches. The associations were statistically insignificant (p > 0.05) and failed to gain statistical significance after adjusting for demographic, socioeconomic, and behavioral variables along with adjustment for chronic conditions and use of dentures (Table 2).

Table 2 Multivariable logistic regression models for the association of general obesity, central obesity and number of natural teeth. Florianopolis, SC, Southern Brazil, 2009 to 2010. (N = 1,704) 

Number of natural teeth Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7

OR 95%CI OR 95%CI OR 95%CI OR 95%CI OR 95%CI OR 95%CI OR 95%CI

General
≥ 10 in both arches 1   1   1   1   1   1   1  
< 10 in at least one arch 1.1 0.8,1.7 1.2 0.8,1.8 1.2 0.8,1.8 1.2 0.8,1.8 1.2 0.8,1.8 1.2 0.7,1.8 1.1 0.7,1.6
Edentate 1.2 0.8,1.7 1.3 0.9,1.9 1.1 0.7,1.8 1.2 0.8,1.8 1.2 0.7,1.9 1.1 0.6,1.9 1.0 0.6,1.7
Tooth loss*gender                           p < 0.001
Males ≥ 10 teeth in both arches                     1      
Females ≥ 10 teeth in both arches                     1.4 0.8,2.5    
Males ≤ 10 teeth in at least one arch                     1.0 0.5,1.9    
Females ≤ 10 teeth in at least one arch                     1.7 0.9,3.2    
Edentate males                     0.7 0.3,1.5    
Edentate females                     1.8 0.9,3.6    

Central
≥ 10 in both arches 1   1   1   1   1   1   1  
< 10 in at least one arch 1.1 0.8,1.6 1.1 0.8,1.7 1.1 0.8,1.7 1.2 0.8,1.6 1.2 0.8,1.8 1.2 0.7,1.8 1.0 0.7,1.5
Edentate 1.4 1.1,1.9 1.4 1.0,2.0 1.4 1.0,2.1 1.5 1.0,2.2 1.5 1.0,2.4 1.5 0.9,2.3 1.3 0.8,2.1
Tooth loss*gender                           p < 0.001
Males ≥ 10 teeth in both arches                     1      
Females ≥ 10 teeth in both arches                     2.1 1.3,3.4    
Males ≤ 10 teeth in at least one arch                     0.8 0.5,1.3    
Females ≤ 10 teeth in at least one arch                     2.7 1.6,4.4    
Edentate males                     0.8 0.5,1.5    
Edentate females                     3.8 2.2,6.6    

Model 1: Crude estimates; Model 2: Adjusted for demographic variables (age, skin color/race); Model 3: Adjusted for demographic variables and socioeconomic variables (wealth, educational attainment); Model 4: Adjusted for demographic variables, socioeconomic variables, smoking and physical activity; Model 5: Adjusted for demographic variables, socioeconomic variables, smoking, physical activity and use of dentures; Model 6: Adjusted for demographic variables, socioeconomic variables, smoking, physical activity, use of denture, hypertension and diabetes; Model 7: Adjusted for demographic variables, socioeconomic variables, smoking, physical activity, use of dentures, hypertension, diabetes and gender

The continuous outcomes of BMI and WC were associated with the number of natural teeth in the crude estimates from the multiple linear regression, showing increasing BMI and WC with the decrease in natural teeth throughout the models. However, the associations were insignificant for all models (p > 0.05) (Table 3). We also observed a significant association suggesting bidirectional relationship as those with central obesity were 1.3 times more likely to be edentate (OR = 1.3, 95%CI 1.0;1.7).

Table 3 Multivariable linear models for the association of general obesity, central obesity and number of natural teeth. Florianopolis, SC, Southern Brazil, 2009 to 2010. (N = 1,704) 

Number of natural teeth Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7

β Coefficient 95%CI β Coefficient 95%CI β Coefficient 95%CI β Coefficient 95%CI β Coefficient 95%CI β Coefficient 95%CI β Coefficient 95%CI
BMI                            
≥ 10 in both arches Ref   Ref   Ref   Ref   Ref   Ref   Ref  
< 10 in at least one arch 0.2 -0.6;1.0 0.5 -0.3;1.3 0.4 -0.5;1.2 0.4 -0.4;1.3 0.4 -0.4;1.2 0.3 -0.5;1.1 0.2 -0.6;0.9
Edentate 0.1 -0.6;0.9 0.6 -0.2;1.4 0.4 -0.4;1.2 0.5 -0.4;1.3 0.5 -0.4;1.3 0.2 -0.7;1.1 0.1 -0.8;0.9
WC                            
≥ 10 in both arches Ref   Ref   Ref   Ref   Ref   Ref   Ref  
< 10 in at least one arch 0.1 -1.9;2.1 0.4 -1.6;2.5 0.2 -1.8;2.2 0.1 -1.8;2.0 0.3 -1.6;2.0 0.0 -1.7;1.7 0.7 -1.0;2.5
Edentate 0.4 -1.9;2.6 1.1 -1.1;3.3 0.7 -1.6;2.9 0.4 -1.8;2.6 0.8 -1.5;3.3 0.5 -1.8;2.8 1.1 -1.1;3.3

BMI: body mass index; WC: waist circumference; Ref: Reference

Model 1: Crude estimates; Model 2: Adjusted for demographic variables (age, skin color/race); Model 3: Adjusted for demographic variables and socioeconomic variables (wealth, educational attainment); Model 4: Adjusted for demographic variables, socioeconomic variables, smoking and physical activity; Model 5: Adjusted for demographic variables, socioeconomic variables, smoking, physical activity and use of dentures; Model 6: Adjusted for demographic variables, socioeconomic variables, smoking, physical activity, use of denture, hypertension and diabetes; Model 7: Adjusted for demographic variables, socioeconomic variables, smoking, physical activity, use of dentures, hypertension, diabetes and gender

We observed significant interaction terms between tooth loss and gender for both categorical outcomes (Table 2). In comparison with males with at least 10 teeth present in each arch, females who were edentate or with less than 10 teeth in one arch had 3.8 and 2.7 times higher odds of being centrally obese, respectively (Table 2). The concomitant odds ratio (OR = 3.8, 95%CI 2.2;6.6) of edentulousness and female gender for central obesity was higher than the independent odds of either edentulousness (OR = 1.3, 95%CI 0.8;2.1) (adjusted for age, skin color, educational attainment, income, smoking, physical activity, use of dentures, hypertension and diabetes) (Table 1) or female gender (OR = 3.2, 95%CI 2.4;3.49, further adjusted for tooth loss; not reported in the table) alone. We observed marked differences in the adjusted prevalence (adjusted for age, skin color, educational attainment, income, smoking, physical activity, use of dentures, hypertension and diabetes) of obesity between males and females. Edentate females had more than double the prevalence of general and central obesity compared with males in the same category of tooth loss. Females with less than 10 teeth in any arch had a 1.9 times higher prevalence of central obesity in comparison with males in the same category of tooth loss (Figure).

Figure Adjusted prevalence* of general and central obesity among older adults according to gender and tooth loss. Florianopolis, SC, Southern Brazil, 2009 to 2010. 

DISCUSSION

Edentates had greater probability of being centrally obese, but the association was explained by confounders. The association between tooth loss and obesity varied according to gender. The significant interaction term between gender and edentulousness highlighted the differences by gender in the association between tooth loss and obesity. This association varied between presence of less than 10 teeth in at least one of the dental arch and edentulousness with central obesity.

Studies have assessed the association between tooth loss and obesity in adults2,11,17 and older adults.6,25 Edentulousness has been reported to be associated with underweight as well as overweight and obesity in Brazil.25 Marcenes et al11 reported similar but insignificant estimates for the association between edentulousness and obesity. Hilgert et al6 reported varying results: edentate participants who used both upper and lower dentures had an inverse association with obesity, while those wearing only upper dentures had higher odds of being obese. The findings of our study are inconsistent with evidence, since we did not observe any association between tooth loss and general obesity. The association between edentulousness and central obesity was partially explained by gender.

The current study also had contrasting results in comparison with the findings of a previous study2 on a younger sample from EpiFloripa with similar objectives. While the prevalence of central and general obesity was similar in the younger sample,2 it was different in the older adults from the same city in the current study. This difference was also far more pronounced in females than in males in this study. Furthermore, they highlighted age-wise variation in the association between tooth loss and obesity, but did not find any differences in this association between males and females; that contrasts our finding of significant gender wise variation in the association in a geriatric sample. These differences could reflect variation in job activities: the younger population may be more active and heterogeneous in terms of health behaviors depending on different jobs, while the older population is more homogenous as a great proportion of is retired or participates in inactive employment. Thus, age would play a key role among younger adults while gender differences are more important among older adults.

A Swedish study with older adults reported that the association between edentulousness and obesity was stronger for females than males,19 similarly to the current study. The gender differences in the association between tooth loss and obesity points towards the need to investigate the dietary behaviors and their determinants in the older population. The differences in dietary preferences between older males and females in Brazil could possibly explain this difference. According to the VIGITEL 2009 report,a males had significantly higher consumption of beans and more than double the consumption of meat with visible fat (rich sources of protein) when compared with females. Females had a higher consumption of fruits and vegetables than males and there were no significant differences between the consumption of soft drinks among older adults aged 65 years and above.7 Loss of teeth may lead to reduction in the intake of fruits and vegetables among older females.15 The consumption of other carbohydrate sources may increase as tooth loss can further limit females’ dietary intake of proteins. This can explain the higher prevalence of central obesity in comparison to general obesity, particularly among females. This has been highlighted in a study where substitution of carbohydrates with proteins was associated with reduced central obesity.12 Hence, we suggest further research to study the relation between dietary alterations caused by tooth loss and the two different outcomes of obesity.

This study had several strengths and some limitations. To our knowledge, this is the only population-based study carried out to assess the role of gender in the association between oral health and obesity in Brazil. Moreover, this study used standardized anthropometric measures, presenting highly reliable data. Few studies18 have assessed the association of tooth loss with both central and general obesity in older adults using BMI and WC. One reason for this could be that fewer studies are planned to assess such associations, which results in limited choice of investigators to use the variables available in the datasets. The other reason could be that, with the recent epidemiological transition and particularly emphasis on central obesity as a risk factor for cardiovascular disease independent of general obesity, recent studies tend to collect detailed data on multiple factors and type of obesity.

However, because of the cross-sectional design, we cannot establish a causal relationship. The linkages between tooth loss and obesity could not be completely explained because we did not have data on participants’ dietary intake. Also, tooth loss was self-reported, which can cause recall bias, and proxies were used for individuals having cognitive difficulties (n = 42). We used three wide categories of tooth loss as exposure, while other studies may have used more reliable measures of self-reported number of teeth.

A bidirectional relationship in this association cannot be ruled out, with the higher odds of the centrally obese being edentate. Hence, biological plausibility should be assessed with well-designed population-level studies with detailed information on the dietary intake of the participants. Future research from the EpiFloripa Aging cohort study may clarify the association between tooth loss and obesity.

The prevalence of central obesity was higher in comparison with the general obesity among older adults. The association between edentulousness and central obesity depended on gender, while no association was found between tooth loss and general obesity. Females with tooth loss had greater probability of being obese than males. The gender differences in the association may be addressed by population-based strategies to prevent and control behaviors leading to obesity. These strategies should be tailored to the vulnerability of the population groups and appropriately designed to the cultural context.

REFERENCES

1.  .Benedetti TRB, Mazo GZ, Barros MVG. Aplicação do Questionário Internacional de Atividades Físicas para avaliação do nível de atividades físicas de mulheres idosas: validade concorrente e reprodutibilidade teste-reteste. Rev Bras Cienc Mov. 2004;12(1):25-34. [ Links ]

2.  .Bernardo CO, Boing AF, Vasconcelos FAG, Peres KG, Peres MA. Association between tooth loss and obesity in Brazilian adults: a population-based study. Rev Saude Publica. 2012;46(5):834-42. DOI:10.1590/S0034-89102012000500010 [ Links ]

3.  .Corseuil MW, Schneider IJ, Silva DA, Silva KS, Borges LJ, Orsi E. Perception of environmental obstacles to commuting physical activity in Brazilian elderly. Prev Med. 2011;53(4-5):289-92. DOI:10.1016/j.ypmed.2011.07.016 [ Links ]

4.  .Danaei G, Singh GM, Paciorek CJ, Lin JK, Cowan MJ, Finucane MM, et al. The global cardiovascular risk transition: associations of four metabolic risk factors with national income, urbanization, and Western diet in 1980 and 2008. Circulation. 2013;127(14):1493-502. DOI:10.1161/CIRCULATIONAHA.113.001470 [ Links ]

5.  .De Marchi RJ, Leal AF, Padilha DM, Brondani MA. Vulnerability and the psychosocial aspects of tooth loss in old age: a Southern Brazilian study.J Cross Cult Gerontol. 2012;27(3):239-58. DOI:10.1007/s10823-012-9170-5 [ Links ]

6.  .Hilgert JB, Hugo FN, Sousa MLR, Bozzetti MC. Oral status and its association with obesity in Southern Brazilian older people.Gerodontology. 2009;26(1):46-52. DOI:10.1111/j.1741-2358.2008.00226.x [ Links ]

7.  .Iser BPM, Claro RM, Moura EC, Malta DC, Morais Neto OL. Risk and protection factors for chronic non communicable diseases by telephone survey-VIGITEL-2009. Rev Bras Epidemiol. 2011;14 Suppl 1:90-102. DOI:10.1590/S1415-790X2011000500010 [ Links ]

8.  .Jensen GL, Friedmann JM. Obesity is associated with functional decline in community-dwelling rural older persons. J Am Geriatr Soc. 2002;50(5):918-23. DOI:10.1046/j.1532-5415.2002.50220.x [ Links ]

9.  .Kimokoti RW, Newby PK, Gona P, Zhu L, McKeon-O’Malley C, Guzman JP, et al. Patterns of weight change and progression to overweight and obesity differ in men and women: implications for research and interventions.Public Health Nutr. 2013;16(8):1463-75. DOI:10.1017/S1368980012003801 [ Links ]

10.  .Kopelman PG. Obesity as a medical problem. Nature. 2000;404(6778):635-43. DOI:10.1038/35007508 [ Links ]

11.  .Marcenes W, Steele JG, Sheiham A, Walls AWG. The relationship between dental status, food selection, nutrient intake, nutritional status, and body mass index in older people. Cad Saude Publica. 2003;19(3):809-16. DOI:10.1590/S0102-311X2003000300013 [ Links ]

12.  .Merchant A, Anand SS, Vuksan M, Jacobs R, Davis B, Teo K, Yusuf S. Protein intake is inversely associated with abdominal obesity in a multi-ethnic population. J Nutr. 2005;135(5):1196-201. [ Links ]

13.  .Mojon P, Budtz-Jorgensen E, Rapin CH. Relationship between oral health and nutrition in very old people. Age Ageing. 1999;28(5):463-8. DOI:10.1093/ageing/28.5.463 [ Links ]

14.  .N’Gom PI, Woda A. Influence of impaired mastication on nutrition.J Prosthet Dent. 2002;87(6):667-73. DOI:10.1067/mpr.2002.123229 [ Links ]

15.  .Nowjack-Raymer RE, Sheiham A. Association of edentulism and diet and nutrition in US adults. J Dent Res. 2003;82(2):123-6. DOI:10.1177/154405910308200209 [ Links ]

16.  .Ostberg AL, Nyholm M, Gullberg B, Rastam L, Lindblad U. Tooth loss and obesity in a defined Swedish population. Scand J Public Health. 2009;37(4):427-33. DOI:10.1177/1403494808099964 [ Links ]

17.  .Östberg AL, Bengtsson C, Lissner L, Hakeberg M. Oral health and obesity indicators. BMC Oral Health. 2012;12:50. DOI:10.1186/1472-6831-12-50 [ Links ]

18.  .Österberg T, Johanson C, Sundh V, Steen B, Birkhed D. Secular trends of dental status in five 70-year-old cohorts between 1971 and 2001.Community Dent Oral Epidemiol. 2006;34(6):446-54. DOI:10.1111/j.1600-0528.2006.00298.x [ Links ]

19.  .Österberg T, Dey DK, Sundh V, Carlsson GE, Jansson JO, Mellström D. Edentulism associated with obesity: a study of four national surveys of 16.416 Swedes aged 55-84 years. Acta Odontol Scand. 2010;68(6):360-7. DOI:10.3109/00016357.2010.514721 [ Links ]

20.  .Pischon T, Boeing H, Hoffmann K, Bergmann M, Schulze MB, Overvad K, et al. General and abdominal adiposity and risk of death in Europe. N Engl J Med. 2008;359(20):2105-20. DOI:10.1056/NEJMoa0801891 [ Links ]

21.  .Sahyoun NR, Lin CL, Krall E. Nutritional status of the older adult is associated with dentition status. J Am Diet Assoc. 2003;103(1):61-6. DOI:10.1053/jada.2003.50003 [ Links ]

22.  .Schwahn C, Polzer I, Haring R, Dörr M, Wallaschofski H, Kocher, et al. Missing, unreplaced teeth and risk of all-cause and cardiovascular mortality. Int J Cardiol. 2013;167(4):1430-7. DOI:10.1016/j.ijcard.2012.04.061 [ Links ]

23.  .Shinkai RSA, Hatch JP, Sakai S, Mobley CC, Saunders MJ, Rugh JD. Oral function and diet quality in a community-based sample. J Dent Res. 2001;80(7):1625-30. DOI:10.1177/00220345010800070601 [ Links ]

24.  .Steele JG, Sanders AE, Slade GD, Allen PF, Lahti S, Nuttall N, et al. How do age and tooth loss affect oral health impacts and quality of life? A study comparing two national samples. Community Dent Oral Epidemiol. 2004;32(2):107-14. DOI:10.1111/j.0301-5661.2004.00131.x [ Links ]

25.  .Tôrres LHN, Silva DD, Neri AL, Hilgert JB, Hugo FN, Sousa MLR. Association between underweight and overweight/obesity with oral health among independently living Brazilian elderly. Nutrition. 2013;29(1):152- 7. DOI:10.1016/j.nut.2012.05.011 [ Links ]

26.  .Villareal DT, Apovian CM, Kushner RF, Klein S. Obesity in older adults: technical review and position statement of the American Society for Nutrition and NAASO, The Obesity Society. Am J Clin Nutr. 2005;82(5):923-34. DOI:10.1038/oby.2005.228. 2005 [ Links ]

.World Health Organization. World health statistics. Geneva; 2012. [ Links ]

Research supported by the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq – Process 569834/2008-2).

a Ministério da Saúde, Departamento de Análise de Situação de Saúde, Secretaria de Vigilância em Saúde. Vigilância de fatores de risco e proteção para doenças crônicas por inquérito telefônico, Vigitel, 2009. Brasília (DF): Ministério da Saúde; 2010.

Received: May 22, 2014; Accepted: April 17, 2015

Correspondence:Marco A Peres. Australian Research Centre for Population Oral Health. School of Dentistry. The University of Adelaide. 122 Frome Street, 1st floor. Adelaide SA. Australia 5005. E-mail:marco.peres@adelaide.edu.au

The authors declare no conflict of interest.

Creative Commons License  This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License, which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.