How does global and central obesity influences on arthritis or rheumatism? Results from the Brazilian National Health Survey, 2013
Previous studies suggest an influence of weight excess on joint diseases, such as the knee joint.1 This influence can be explained by the biomechanical effect of overweight2 or by the metabolic effects of adipokines, especially leptin,3 increased levels of C-reactive protein (CRP) and interleukin-6 (IL-6).1, 3 To our knowledge, no study which estimated the influence of global and central obesity simultaneously on arthritis or rheumatism has been conducted in a Brazilian population-based study, a developing country. Furthermore, the prevalence of overweight and obesity is increasing in the Brazilian adult population, reaching 55.7% and 19.8% in 2018,4 respectively, which reinforces the need for a more detailed study of obesity and increased waist circumference effects in this population.
To better understand whether global and central obesity are associated with the prevalence of arthritis or rheumatism in adulthood and the elderly, we conducted a cross-sectional study with 44,861 participants from the Brazilian National Health Survey.5 This is the most complete survey on health and its determinants ever carried out in the country, carried between 2013 and 2014 by the Brazilian Institute of Geography and Statistics (Instituto Brasileiro de Geografia e Estatística-IBGE), the Brazilian Health Ministry and Oswaldo Cruz Foundation (Fiocruz). The target population of the survey was comprised of adults aged 18 or older, living in urban or rural areas throughout the country, excluding the special census sectors.5 Using logistic regression, we made sequential adjustments for the region of the country, age, sex, educational attainment, health-related behaviours and health conditions. Finally, we added mutual adjustment for Body Mass Index (BMI) and Waist Circumference (WC) to estimate the independent relationship of biomechanical effect of weight excess and inflammatory effects of central adiposity on arthritis or rheumatism. For all analyses, we estimated odds ratios (OR) and their 95% confidence intervals (95% CI).
Most participants in this study were women (53.3%, 95% CI: 52.2-54.1), residents of the southeast region of the country (45.3%, 95% CI: 44.4-46.1), aged 35-54 years (50.4%, 95% CI: 49.6-51.2) and did not have completed elementary school (46.5%, 95% CI: 45.7-47.3). Regarding health-related behaviours, 11.9% (95% CI: 11.4-12.4) were abusive alcohol consumers, 15.9% (95% CI: 15.3-16.5) smokers and 83.2% (95% CI: 82.6-83.8) did not practice leisure physical activities. The prevalence of diabetes was 9.0% (95% CI: 8.5-9.5), hypertension was 28.6% (95% CI: 27.8-29.4) and high cholesterol was 17.8% (95% CI: 17.1-18.5). Finally, the prevalence of arthritis or rheumatism in this population was 6.7% (95% CI: 6.3-7.1). Table 1 shows the associations BMI and WC with arthritis/rheumatism adjusted for confounders. After full adjustments for sociodemographic characteristics, health-related behaviours, and health conditions (Model 4), overweight (OR: 1.36; 95% CI: 1.14-1.63), obesity (OR: 1.81; 95% CI: 1.49-2.19) and central obesity (OR: 1.68; 95% CI: 1.45-1.96) were associated with the presence of arthritis or rheumatism, compared with that adequate weight or non-central obesity. And, after the mutual adjustment for BMI and WC (Model 5), the associations for global obesity (OR: 1.43; 95% CI: 1.15-1.79) and central obesity (OR: 1.41; 95% CI: 1.18-1.69) had a decrease in the magnitude of the association, but they remained statistically important. This suggests that both the biomechanical effect of being overweight and the metabolic or inflammatory effects related to central obesity play a role.
| Self-reported arthritis/rheumatism | ||||||
|---|---|---|---|---|---|---|
| Model 0 | Model 1 | Model 2 | Model 3 | Model 4 | Model 5 | |
| OR (95% CI) | OR (95% CI) | OR (95% CI) | OR (95% CI) | OR (95% CI) | OR (95% CI) | |
| Body mass index | ||||||
| Normal weight | Reference | Reference | Reference | Reference | Reference | Reference |
| Underweight | 0.94 (0.57-1.55) | 0.66 (0.38-1.14) | 0.65 (0.37-1.13) | 0.65 (0.37-1.13) | 0.78 (0.43-1.42) | 0.86 (0.47-1.57) |
| Overweight | 1.41 (1.21-1.65) | 1.48 (1.25-1.74) | 1.47 (1.25-1.74) | 1.47 (1.24-1.73) | 1.36 (1.14-1.63) | 1.18 (0.97-1.42) |
| Obesity | 2.26 (1.92-2.66) | 2.20 (1.85-2.63) | 2.18 (1.83-2.61) | 2.19 (1.84-2.61) | 1.81 (1.49-2.19) | 1.43 (1.15-1.79) |
| Waist circumference | ||||||
| Non-central obese | Reference | Reference | Reference | Reference | Reference | Reference |
| Central obesitya | 3.16 (2.78-3.59) | 1.99 (1.73-2.30) | 1.98 (1.71-2.29) | 1.98 (1.72-2.29) | 1.68 (1.45-1.96) | 1.41 (1.18-1.69) |
Note
- Bold indicates significant values (P < 0.001).
- Associations were estimated by logistic regression. Model 0: adjusted for the region of Brazil Model 1: Model 0 + sex and age. Model 2: Model 1 + educational attainment. Model 3: Model 2 + abusive alcohol consumption, smoking and leisure physical activity. Model 4: Model 3 + hypertension, diabetes, and high cholesterol. Model 5: Model 4 + mutual adjustment for both exposures.
- Abbreviations: 95% CI, 95% Confidence Interval.
- a > 88cm for women, and ≥102 cm for men.
Mechanical tension can influence chondrocyte activity, which can lead to changes in the composition, structure, metabolism and mechanical properties of articular cartilage, bone tissue and other joint tissues.6 However, chronic low-grade inflammation, characteristic of obesity, may be related to the inflammation of joints that do not suffer from weight overload.1, 3 One issue to be discussed is the use of BMI as a diagnostic parameter for obesity, since it only considers total weight.7 The use of central adiposity measures, such as waist circumference, has been used to reduce this bias, for being more associated with metabolic disorders resulting from obesity.8 Similar to our results were observed in a study in the United Kingdom9 and in the Third National Health and Nutrition Examination Survey (NHANES III).10 However, it is important to note that in our study the biomechanical effect of weight was observed only amongst those with obesity, since after the mutual adjustment of global and central adiposity, the category overweight lost statistical significance.
The results of this research suggest that an increase in BMI and WC is associated with higher prevalence of arthritis or rheumatism. Health policies such as the National Food and Nutrition Policy – PNAN and the Clinical Protocol and Therapeutic Guidelines for the management of rheumatoid arthritis do not contemplate the relationship between weight and waist gain and the risk of developing arthritis or rheumatism. Likewise, the therapeutic protocol recommended by the Ministry of Health of Brazil could also consider obesity as a problem associated with health conditions, incorporating it in its treatment.
Some limitations need to be considered in the interpretation of our analyses, as it is a cross-sectional study, we have to consider reverse causality. In addition, we are unable to discriminate between arthritis or rheumatism or their body locations that can contribute to understanding the role of the effect of being overweight. However, the large sample size reflects the Brazilian population in terms of socioeconomic characteristics, body size, and health behaviours and conditions, making it possible to clearly identify the independent effect of global and central obesity on this relationship, even though we have not seen a significant change in magnitudes after adjustment inclusions.
ACKNOWLEDGEMENTS
The authors thank the Ministry of Health of Brazil, in partnership with the Brazilian Institute of Geography and Statistics (Instituto Brasileiro de Geografia e Estatística – IBGE) to realise the National Health Survey and for the important contributions to the knowledge of the health conditions of the Brazilian population.
CONFLICTS OF INTEREST
The authors declare that there are no conflicts of interest.
