Socioeconomic status and obesity in the Awajún of Perú
Obesity is the main cause of death in every country in Latin America (Kain and Albala, 2003), thus it is important to elucidate factors that may contribute to it. This project explored the association between obesity and socio-economic status (SES) in the Awajún, an indigenous group in Perú, currently undergoing a nutritional transition from traditional to western diets (Berlin and Markell, 1977). Individual demographic information was collected via surveys (N=225, 107 men and 118 women), and anthropometric measurements were recorded. More than 30% of the Awajún were classified as overweight based on BMI; 42.0% of women were overweight compared to 24.6% of men. SES showed a statistically significantly positive association with BMI, an outcome of obesity, in this community when controlled for sex and age. When stratified by gender, the relationship between SES and obesity was only statistically significant for men. This may be because the shift toward sedimentary jobs and less physical activity associated with high SES affects men more than women due to cultural reasons. Men may be shifting to sedentary, high status occupations, such as students or professors, while women continue to work in the home because this is more culturally respectable. More women were found to be overweight compared to men, but this did not seem to be associated with SES indicating a more complex or multifactorial relationship between SES and BMI in women.
Latin American countries were once associated with malnutrition, but now they are experiencing a growing obesity problem that is reaching epidemic proportions (Jacoby et al., 2003). Obesity is one of the leading risk factors for heart disease, cancer, Type 2 Diabetes, and other metabolic diseases (Pampel et al., 2012) and is currently the principal cause of death in every country in Latin America (Kain and Albala, 2003). The World Health Organization has noted increasing levels of obesity in countries emerging from poverty, especially in more urban areas (Kain and Albala, 2003).
Globalization, urbanization, and an increasing gradient in socio-economic status (SES) are fueling a “nutritional transition” in Latin America such that traditional diets high in grains, fruits, and vegetables are being substituted for more “Western” diets high in protein, saturated fat, and processed sugars (Kain and Albala, 2003). This shift in diet, along with a decline in energy expenditure associated with more sedentary lifestyles, has increased the threat of obesity in low and middle-income countries (Popkin et al., 2011, Kain and Albala, 2003, Uauy et al., 2001). The economic burden of metabolic, non-communicable diseases, as well as the increasing prevalence of these diseases in countries already struggling with malnutrition, makes research into the underlying causes of obesity in these settings necessary.
The degree of economic and social development toward “Western” ideals of success and civilization has been shown to play a role in the rise of obesity in developing countries (Monteiro et al., 2004). In some cases, the relationship between SES, obesity, and gender (a socio-cultural construct), can be determined by the country’s economic status (Kain and Albala, 2003). In higher-income countries, such as the United States, Monteiro et al. (2004) found obesity to increase with lower SES, while the inverse is true for developing countries. Overall, obesity tended to become more prevalent in lower SES populations as the country’s gross national product (GNP) increased (Monteiro et al., 2004). In developing countries, high SES has been linked to increasing rates of being overweight and obese (Martorell et al., 1998), while lower SES is associated with malnutrition due to economic disparities and inequality (Pampel et al, 2012). The relationship between SES and obesity has been described as such: in poorer countries, there is a positive relationship between obesity and SES; in middle-income countries, there is no relationship; and in wealthy countries there is a negative relationship between the two (Martorell et al., 1998).
Some studies have found the relationship between SES and obesity to be gendered. Monteiro et al. (2004) found that the shift to a negative relationship between SES and obesity took place at an earlier stage of economic development for women than men, occurring at a GNP of US $2500 per capita. By comparing 14 related studies, seven low- or middle-income countries were found to have a statistically significant positive relationship between SES and obesity for men, while the other seven low- or middle-income countries had no association. For women, 10 countries out of the 14 studies showed a statistically inverse relationship between SES and obesity (Monteiro et al., 2004). Although there was no clear pattern for all the researched countries, women in the studies were more likely to have a positive association between obesity and SES as compared to men. In another study by Popkin et al. (2011), it was observed that men with higher SES had higher rates of being overweight than men with lower SES in developing countries, and the relationship was not proven statistically significant with women.
Perú is a developing South American nation where more than half of the population (53 percent) has been reported to be overweight, which is defined as having a Body Mass Index (BMI) over 25 kg/m2 (Filozof et al., 2008). According to the Centers of Disease Control and Prevention (CDC), a BMI greater than 24.99 kg/m2 is classified as overweight. A BMI greater than 30 kg/m2 is classified as obese. The Awajún are a Peruvian indigenous group that traditionally subsisted on a cassava-based diet with occasional hunting and fishing (Roche et al., 2007). Although they were once isolated, the construction of a nearby highway has influenced their cultural norms and lifestyles. This investigation aims to explore the association between obesity and SES in the Awajún, a population undergoing a social, economic and nutritional transition (Berlin and Markell, 1977). The raw data used in this investigation was collected in 2013 by Paula Tallman, a biological anthropology graduate student at Northwestern University. The main questions this project aims to address are: what is the prevalence of obesity in the Awajún of Perú, is socio-economic status a predictor of obesity, and is this relationship gendered?
High SES will be a predictor of obesity for men and women in the community, as the Awajún are currently undergoing a nutritional and social transition from traditional to Western diets and from more active to more sedentary occupations. This study, however, will not look at energy expenditure and food consumption pathways directly.
The 2012 census of Perú recorded 900 people in Shushug, Mente, Paanki, and Boca de Lobo, the major centers where the Awajún are located (Geohive, 2012). The total sample size was 225 individuals from those locations, 107 women and 118 men. The age range was from 18 to 65 years old, with a mean age of 33.38 years. The average number of children per family was 3.65, and the maximum was 15 children.
Anthropometric measurements of height (cm), weight (to the nearest 0.1 kg), visceral fat (number generated by scale ranging from 0 to 18), percent body fat (percent), suprailiac skinfold (mm), subscapular skinfold (mm), tricep skinfold (mm), bicep skinfold (mm) and waist circumference (inches) were used to establish BMI (kg/height m2) as an indicator of adiposity and obesity. Height was measured using a portable stadiometer, waist circumference was recorded with a tape measure, skinfolds were measured with calipers, and other measurements were gauged using an OMRON HBF-516B composition monitor and scale. A BMI < 18.5 kg/m2 was coded as underweight=0; 18.5 to 24.9 kg/m2 as normal=1; 25.0 to 29.9 kg/m2 as overweight=2; and 30 kg/m2 or above as obese=3, following CDC guidelines.
For the purposes of this study, age was coded in years, and gender was coded as women=0 and men=1. Income was recorded as the average monthly personal earnings in soles, the Peruvian currency. Occupation was coded such that un-employed=0, agriculturalist=1, wage laborer=2, government worker=3, student=4, teacher/nurse=5 and higher professional=6. Students of any age were rated higher than government or wage laborers due to the special status of students in the community. Education was coded as none=0, unfinished primary school=1, finished primary school=2, un-finished secondary school=3, finished secondary school=4, unfinished higher education=5 and finished higher education=6.
SES was defined in this study as a combination of income, education, and occupation. Each score for income, education, and occupation was scaled to 1 and summed to create a summary score for SES. Multivariate linear regression analysis (STATA 10) was used to assess the relationship between SES, gender and BMI. Statistical significance was defined as p < 0.05.
It was found that the average monthly household income in the Awajún community was 565.24 Peruvian soles (equivalent to US$191.15). The overall national average monthly income in 2014 was 1600 soles (equivalent to US$541.09), which shows the lower economic development of the community compared to the country overall (Trading Economics, 2014). Maximum income within the community was 4,500 soles with a large standard deviation of 685.24 soles, suggesting an economic divide between the rich and the poor. Most of those surveyed (62.50 percent) were agriculturalists, 15.63 percent were students, 13.39 percent were teachers or nurses, and 0.89 percent were professionals. Regarding educational attainment, half of the people in the survey (50.45 percent) had not finished secondary school, 20.09 percent had finished secondary school, and 16.96 percent had completed higher education.
BMI was determined as the main outcome of obesity in this community (Table A). Most studies analyzing obesity in Latin America have also used BMI as the major health outcome. BMI was positively associated with other measurements of adiposity as seen in Table A, which displays the results of a linear regression using BMI as the dependent variable and visceral fat, percent body fat, waist circumference, and other measurements as the independent variables. BMI was established as an outcome of obesity in the Awajún of Perú through a shared p-value of 0 with visceral fat, subscapular skinfold, bicep skinfold, and waist circumference. There was no significant link between BMI and percent body fat, suprailiac skinfold, or triceps skinfold. The adjusted r-squared value for the data was 0.7731; therefore, obesity is accurately represented through these variables.
The prevalence of obesity in the Awajún was measured using BMI to determine the nutritional status of the community (Table B). BMI was scaled on a measure of 0-3, where 0=underweight, 1=normal, 2=overweight, and 3=obese, according to CDC categories for BMI. More than 30 percent of the sample was overweight, below the national average of 53 percent reported by Filozof et al. (2008). Only 2.67 percent of the community was obese, while most of the community (66.67 percent) had a normal BMI.
BMI was then determined for women (Table C) and men (Table D) to show the differences between genders. Figure 1 shows the distribution of BMI categories between men and women with 42.0 percent of women classified as overweight, compared to 24.57 percent of men. Therefore, there was a marked difference in nutritional and health status between men and women among the Awajún.
Once BMI was established as an outcome of obesity, and obesity was recognized as a health issue among the community studied, multivariate regression analysis was used to determine if BMI was associated with SES controlling for sex and age. There was a statistically significantly positive association between BMI and SES (B=0.83, P=0.04), where increasing SES in the Awajún led to a 0.83 increase in BMI (Table E). Gender also had a statistically significant association with BMI, shown through a p-value of 0.002. A coefficient of -1.14, where men=1 and women=0, presents women as having a higher tendency toward obesity than men. The standard error was small enough to support the accuracy of this claim. Sex and age were included as controlled variables. The adjusted r-squared value was 0.05, indicating a lack of association between SES, sex and age.
Gender was shown to be a significant predictor of obesity (Table E). The relationship between SES, BMI and sex was then analyzed to see if this difference in BMI between sexes was due to SES in this community. There was no statistically significant association between BMI and SES in women (Table F), as the p-value was 0.102 (and not below 0.05). Therefore, obesity found in women among the Awajún was not a result of SES. The adjusted r-squared value was 0.02, supporting lack of association between the variables.
Next, it was determined whether SES played a role in obesity among men in the Awajún. There was a statistically significant positive association between BMI and SES among men (Table G), with a p-value of 0.07 and a coefficient of 0.89. Therefore, as SES increased, BMI increased by 0.89 for men in this community. The adjusted r-squared value was found to be 0.05.
More than 30 percent of the Awajún were found to be overweight, such that 42.05 percent of women were overweight compared to 24.57 percent of men (Figure 1; Table C; Table D). This finding is in agreement with data by Monteiro et al. (2004), who found that the prevalence of obesity for women was double that of men at the national level in Brazil, another developing country in South America. A study in Peruvian cities by Jacoby et al. (2003) also showed an increase in obesity in women compared to men (40.28 percent of women were overweight compared to 44.93 percent of men; 22.85 percent of women were obese, compared to 15.98 percent of men). The Awajún may be less overweight than their urban counterparts due to their early stage in the nutritional and lifestyle transition to more Western diets, values and occupations, which have been linked to obesity in many developing countries.
Additionally, SES was proven to have a statistically significant association with BMI, an outcome of obesity, in the community. A possible explanation of this trend could be that higher SES leads to consumption of a more “Western” diet and avoidance of physically demanding tasks or occupations in globalizing countries. Pamel et al. (2012: 1079) noted “problems of malnutrition are replaced by problems of overconsumption that differentially affect SES groups”. It has been shown that upwardly mobile consumers in globalizing countries prefer imported products high in fat, carbohydrates, and sugar (Uauy et al., 2001). Those of higher SES are more likely to eat imported “Western” food due to an availability of money and a shifting “cultural value” toward less traditional diets (Berlin and Markell, 1977). Lower-status occupations tend to involve more physical activity, which can offset adiposity, compared to high status occupations, such as managerial or professional jobs, which are more sedentary (Wardle et al., 2002). Access to technology through globalization has also reduced energy expenditure in labor-intensive jobs such as farming and mining (Popkin et al., 2011). Activity patterns have been changing as rapidly as diet in developing countries, and the difference in energy expenditure can be significant (Kain and Albala, 2003).
The cultural valuation of adiposity can also be a factor in obesity prevalence (Wardle et al., 2002). Surveys showed the Awajún to equate heavier figures to health; this may play a role in activity and diet patterns for those of high SES due to their income or status in the community (Jacoby et al., 2003). Those of lower SES would not be able to buy more food than needed for aesthetics or reputation due to limited financial resources.
It is also now established that the development of obesity and chronic diseases such as Type 2 Diabetes can be influenced by maternal nutrition from conception to adulthood (Popkin et al., 2011). According to Popkin, nutritional deficiencies during fetal growth can trigger “anatomical, hormonal, and physiological” changes that increase the survival of the fetus by enhancing metabolic efficiency (2011:6). Thus, children born in “resource-poor” environments will be able to utilize the energy from their diet much more efficiently due to adaptations in-utero, which can lead to increased risk of adiposity later in life. The tendency toward adiposity due to biological adaptations in-utero can be heightened by economic and social transitions toward higher SES and more “Western,” energy-rich diets (Popkin et al., 2011). Therefore, this could be a possible explanation for the association between SES and BMI in the Awajún, who are currently undergoing a social, economic, and nutritional transition due to globalization.
However, men were shown to have a statistically significantly positive association between SES and BMI, while women were not (Tables F and G). Awajún men traditionally spent most of their time hunting and fishing (Berlin and Markell, 1977). Women, on the other hand, had sedentary jobs working in the home and caring for children. As stated above, a change in energy expenditure due to shifting physical activity patterns has led to growing adiposity in developing countries. I argue that the shift toward more sedimentary jobs and less physical activity due to globalization affects men more than women due to gendered occupation status among this community. Men may be more eager to become students or professors because there is a high status associated with these occupations; women typically do not achieve high educational levels and do not become students or professors since it is more culturally acceptable for them to work in the home.
Weight gain in poorer countries can also symbolize higher status, and among men it can be indicative of physical prowess and power (Pampel et al., 2012) so that heavier men may be seen as more successful. Therefore, men will be shifting toward less energy expenditure and weight gain with higher SES, while women’s energy expenditure may not change much. Although more women were found to be overweight compared to men, this did not seem to be associated with SES, indicating a more complex or multifactorial relationship between SES and BMI in women. A bio-cultural approach is necessary to study human nutrition issues such as obesity in indigenous populations so that health outcomes can be understood as manifestations of culture change and gender differences.
Further research could explore the relationship between energy expenditure and obesity for men and women in the community. The association between SES and obesity between genders may change over time with increasing GDP levels based on previous studies. A follow-up investigation could provide interesting information on the effect of GDP on SES and obesity in this community. Additionally, more factors need to be considered in a larger sample to determine the source of the gender difference in obesity found in the Awajún.
Odette is a rising senior majoring in Anthropology and minoring in Global Health, following a pre-medical track. She is half-Guatemalan and interested in pursing a career in medical anthropology as a clinician and researcher.
Berlin EA, Markell EK. 1977. An assessment of the nutritional and health status of an Aguaruna Jívaro community, Amazonas, Peru. Ecology of Food and Nutrition 6: 69-81.
Centers for Disease Control and Prevention. About BMI for Adults. (2014, July 11). Retrieved December 4, 2014, from http://www.cdc.gov/healthyweight/assessing/bmi/adult_bmi/index.html
Filozof C, Gonzalez C, Sereday M, Mazza C, Braguinsky. 2001. Obesity prevalence and trends in Latin-American countries. Obes Rev 2(2): 99-106.
GeoHive. Peru General Information. (2012, June 30). Retrieved December 4, 2014, from http://www.geohive.com/cntry/peru.aspx
Jacoby E, Goldstein J, Lopez A, Nunez E, Lopez T. 2003. Social class, family, and life-style factors associated with overweight and obesity among adults in Peruvian cities. Prev Med 37(5): 396-405.
Kain J, Vio F, Albala C. 2003. Obesity trends and determinant factors in Latin America. Cad Saude Publica, 19(S1): S77-S86.
Leatherman T. 2005. A space of vulnerability in poverty and health: political-ecology and bio-cultural analysis. Ethos 33: 46-70.
Martorell R, Khan LK, Hughes ML, Grummer-Strawn LM. 1998. Obesity in Latin American women and children. J Nutr 128(9): 1464-1473.
Monteiro CA, Moura EC, Conde WL, Popkin BM. 2004. Socioeconomic status and obesity in adult populations of developing countries: a review. Bulletin of the World Health Organization 82(2): 940-946.
Pampel FC, Denney JT, Krueger PM. 2012. Obesity, SES, and economic development: a test of the reversal hypothesis. Soc Sci Med 74(7): 1073-1081.
Popkin BM, Adair LS, Ng SW. 2011. Global nutrition transition and the pandemic of obesity in developing countries. Nutrition Reviews 70: 3-21.
Trading Economics. Peru average monthly wages. (2014, December 6). Retrieved December 6, 2014, from http://www.tradingeconomics.com/peru/wages/forecast
Uauy R, Albala C, Kain J. 2001. Obesity trends in Latin America: Transiting from Under- to Overweight. J Nutr 131: 893S-899S.
Wardle J, Waller J, Jarvis MJ. 2002. Sex differences in the association of socioeconomic status with obesity. Am J Public Health 92(8): 1299-1304.