Ayush Malhotra
Age 14 | Waterloo, Ontario
The purpose of this study is to better understand the relationship between income, obesity, and life expectancy. It was predicted that a higher GDP per capita would lead to higher life expectancy, a higher GDP per capita would lead to a lower obesity rate, and a higher obesity rate after controlling for GDP per capita would lead to a lower life expectancy. The World Bank and NDC Risk panel data was formatted in Stata mp14 and analyzed using linear panel regression to estimate the association between obesity, life expectancy, and GDP per capita in 202 countries over the course of 44 years. Based on the regressions, on average, a US$ 1,000 increase in GDP per capita in a given country would increase one’s lifespan by 4 years. It was also found that, if a given country’s GDP per capita increased by US$ 1,000, the number of women who are obese would decrease by .02%. Interestingly, the findings among men did not replicate suggesting that annual GDP per capita would increase obesity rates for men. The final regression results suggest that as the obesity rate of a given country increases, the life expectancy decreases, but affects men approximately 100% more.
Keywords: obesity, life expectancy, income
“More die in the United States of too much food than of too little”
–John Kenneth Galbraith, economist
Increasing numbers of people around the world are obese, which is defined as having a body mass index (BMI) of 30 or more. In the United States alone, more than one in every three persons is obese (National Institute of Health, 2017). The worldwide prevalence of obesity nearly tripled from 1975 to 2016 (World Health Organization, 2020). While obesity is most evident among middle-aged people, levels of obesity are also rising among children. One in six children in the United States is obese or overweight (Centers for Disease Control and Prevention, 2019; Mehta & Chang, 2009). This is alarming as the medical literature shows obesity is related to multiple chronic diseases, such as heart disease, stroke, and type 2 diabetes (Kearns et al., 2014).
From a policy perspective, it is important to understand how a country’s wealth impacts levels of obesity among its citizens. While studies have looked at this, findings have been mixed. Earlier studies showed that obesity was a disease of the socioeconomic elite (Monteiro et al., 2004). Those who were wealthier had access to more “luxury” food. In contrast, recent studies show a negative correlation between high socioeconomic conditions and obesity levels (Monteiro et al., 2004). A limitation of these studies is that they rely on a small sample of countries. The purpose of this study is to better understand the relationship between annual GDP per capita, obesity, and life expectancy across a universal sample of countries. More specifically, this study attempts to answer three important questions: 1. How does obesity affect life expectancy; 2. How do the country's income levels affect both obesity and life expectancies? 3. Is the effect of obesity on life expectancy is moderated by increasing income levels in a given country? To answer these questions, I run a panel analysis on 202 countries over a period of 41 years, making this (to the best of my knowledge) one of the largest sample studies in this research area.
BACKGROUND RESEARCH
A review of multiple studies (Monteiro et al., 2004) concluded that obesity was a disease of the socioeconomic elite. It was believed that those who were wealthier, had access to more food, and as a result, had higher chances of being obese. However, a 2010 study (Kuntz, Benjamin & Lampert, 2010) showed contrasting results. The results showed a negative correlation between those with high socioeconomic conditions and their obesity levels. This is likely because people with lower income may gravitate towards dense foods, such as those with more refined grains, added sugars, or fats, that represent the lowest cost food option to the consumer. Many farms that grow healthy stocks sell their food products at higher prices. When stores and restaurants purchase these healthy stocks, they sell them for even higher prices to make a profit. On the other hand, when factories produce artificial foods using cheap ingredients, they can be retailed at more affordable prices than healthier foods. Those with lower income are more likely to choose artificial foods or non-healthier food products, increasing their chances of getting diagnosed with obesity.
Obesity has been found to influence a person’s life expectancy. An eye-opening 2014 study (Printz, 2014.) found that obesity may shorten one's lifespan by up to fourteen years. This study analyzed 20 studies conducted in the United States, Australia, and Sweden. The researchers found that the risk of dying from most major health causes, such as cancer, grow significantly among those in class three obesity (extreme obesity). More specifically, those in the BMI range of (40 – 44.9), (45 – 49.9), (50 – 54.9), and (55 – 59.9) lost 6, 9, 10 and 14 years respectively from their average lifespan.
Alongside obesity, life expectancy is also significantly correlated with GDP. Data clearly show that countries with higher income experience a higher life expectancy than countries with lower income. Chetty et al. (2016) studied the association between income and life expectancy on a sample of individuals aged between 40 and 76 years. The income data was collected from 1.4 billion tax records through the years 2001–2014. To obtain further data including the mortality results of those in the study, the researchers used the records from the Social Security Administration death records. Once the data was cleaned and analyzed, the results showed that the difference between the life expectancies of those who were in the top 1% of income and those who were in the poorest 1% of the income class differed by approximately 14.6 years; that is, people in the top 1% income bracket lived approximately 14.6 years more than those in the bottom 1% of the income bracket. All results obtained in this study showed a gradual upwards relationship between income and life expectancy.
Another study published in the Harvard Gazette (Reuell, 2016) showed that being poor in the United States can give you the same life expectancy as those who live in some of the most underdeveloped countries in the world. In addition, this study found that low-income families in wealthier neighborhoods had on average higher life expectancies than low-income families in poor neighborhoods. Although studies such as these have been very useful, there is little work on these results compare to other countries.
HYPOTHESES
If people have higher income, then they will have access to better hospitals, better doctors, and better gym facilities, because of which they will be healthier, and hence will enjoy a longer life expectancy. Research has shown that in a country with a higher average income such as the United States, healthcare is taken for granted by most of the population. While in a country with a meager average salary such as India, more than 75% of citizens do not have access to healthcare (Iyengar & Dholakia, 2012; Kashyap, 2018). Healthcare is vital to maintain a healthy life, thus, allowing those with higher income to live a healthier life. Along with healthcare, money is also critical for nourishment and housing. Studies show that 0.5% of the population in the US is homeless while a staggering 5% of the Indian population is homeless. Those with higher incomes will be able to afford better healthcare, get proper nourishment, as well as adequate housing, therefore, will enjoy a higher life expectancy. Thus, it was hypothesized:
Hypothesis 1: The countries with a higher annual GDP per capita will have a higher average life expectancy while countries with a lower average income will exhibit a lower average life expectancy.
Studies show those who struggle with poverty often turn to fast food, as they are cheaper than healthier options (Drewnowski & Specter, 2004). Consuming fast food has been shown to increase obesity (Drewnowski & Specter, 2004). In the case that the “cheap food” is not fast food, even then it may be food that is conveniently cooked with unhealthy ingredients. Often, these foods have more significant portions of fats and calories. If these eating habits continue, the chances of obesity grow greater. Since obesity mainly arises from the consumption of unhealthy food, and those with higher incomes will be able to afford healthier food, it was hypothesized:
Hypothesis 2: The countries with a higher average income will have a lower obesity rate, while the countries with a lower average income will have a higher obesity rate.
An estimated 2.2 million people across the globe die from obesity every year. Mainly middle-aged people suffer from obesity (Molarius et al., 2000). This would suggest that the deaths from obesity would harm a country's’ life expectancy. Thus, it was hypothesized:
Hypothesis 3: Controlling for a countries’ development (GDP), a country with higher obesity levels will have a lower life expectancy.
METHOD
To compile the sample, three different data sets were used. Obesity panel data for both men and women was collected from NCD Risk Factor Collaboration (NCD-RisC) (NCD Risk Factor Collaboration, n.d.), while panel data for annual GDP per capita and life expectancy was collected from The World Bank (GDP per capita (current US$), n.d.; Life expectancy at birth, total (years), n.d.). The obesity data shows the prevalence of BMI>30 for both men and women. Data was first cleaned and organized in Excel according to the data layout preferences of Stata mp14 (a data analyzing software). For example, data from “The World Bank” included many countries that the “NCD Risk” data did not include. As a result, various countries that only appeared in one or two data sets were removed. The final sample included data on 202 countries across 41 years from 1975 to 2016.
Panel regressions were run to test each of the three hypotheses, by using the “areg” command in Stata. Time and country effects in each of these regressions were controlled for by including a dummy variable for every one of the years and the countries. This is important as technological advancements have occurred over the years, and these advancements could create bias in the study. The regression model takes the following form:
(1) Yij = βxij + year control + country control + ε
Where Yij is the dependent variable – the life expectancy or the obesity levels for country i in year j, and xij is the independent variable – the obesity levels and the GDP per capita.
RESULTS
Figures 1 and 2 are descriptive charts obtained through preliminary analysis. Figure 1 shows that BMI has gone up alarmingly over the last 41 years across 202 countries. Similarly, life expectancy has also increased over the last 41 years (Figure 2).
Figure 1: Mean BMI Across Countries
Figure 2: Mean Life Expectancy Across Countries
Table 1 shows the results of the panel regressions. Model 1 shows that the relationship between annual GDP per capita and life expectancy is positive and significant. In other words, as a countries’ annual GDP per capita increases, the countries’ mean life expectancy increases as well. This relationship is significant at p < 0.01. Based on the regression coefficient for annual GDP per capita, a $1,000 USD increase in GDP per capita in a given country would increase one’s lifespan by 4 years. This result suggests that H1 is supported.
Table 1: Panel Regression Results for 202 Countries across 41 Years
Model 2 shows that the 𝛽 coefficient for annual GDP per capita (-0.000000181) is negative and significant (p < 0.01), suggesting that as the Annual GDP per capita of a given country increases the obesity rate among women in that country would be lowered. Based on the 𝛽 value, on average, if a given country’s GDP per capita increased by $1,000, the number of women who are obese would decrease by .02%. Thus, H2 is supported for females. Model 3, in contrast to Model 2, shows a positive and significant relationship between annual GDP per capita and men’s obesity rate (p < 0.01). Thus, while an increase in the annual GDP per capita decreases obesity among women, it increases obesity rates for men. Figure 3 shows these contrasting results. It highlights how the change in GDP per capita affects the obesity levels for me and women, based on the regression coefficients. This suggests that Hypothesis 2 does not hold for males. Model 4 shows the effect of obesity rates on life expectancy. The 𝛽 coefficients show significant evidence that high obesity rates have a negative effect on a countries’ life expectancy at birth (p < 0.01). Interestingly, the difference in the 𝛽 coefficients (-11.10 for women and -24.29 for men) seems to suggest that among men this correlation is stronger while for women it may be less strong. Thus, these results support H3.
Figure 3: Effect of GDP on Obesity for Men and Women
DISCUSSION & CONCLUSION
In this study, it is found that higher annual GDP per capita leads to a higher life expectancy, a higher annual GDP per capita leads to lower obesity rates for women, but the opposite for men and higher obesity rates lead to a lower life expectancy. The study’s results can be helpful to researchers, the general population, and governments across the world. To elaborate, the data implies that if a given government wanted to provide money to those who were struggling with obesity, they would see a more significant improvement if they invested it in females, while they could see the opposite among men. These results can prove to be vital because they suggest that the current worldwide belief on obesity (obesity is a disease of the lower class) is false in terms of men. Currently, various approaches are already in practice against obesity. To elaborate, the idea of food stamps was thought to be allowing those who struggle with poverty a chance to obtain access to “healthier” foods. Though recently, Baum (2011) found that food stamps tend to cause a rise in obesity rates. Rather than solutions like this, other approaches may need to be taken such as less costly gym memberships, or as some companies have started doing currently, incentivizing weight loss.
This study had one main limitation. In the data, factors such as “country”, “year”, and “GDP per capita” were controlled for but there still could be other variables that can play a role in the outcome of the analyses. For example, a recent study shows that soft drink consumption is directly related to global obesity rates. In fact, the study found that soft drink consumptions are linked with obesity rates even in low- and middle-income countries (Basu et al., 2013). Since soft drink consumption rates change every year, controlling for “country” will not eliminate this variable. Not controlling for soft drink consumption could lead to bias in the regressions.
To conclude, obesity is significantly linked to life expectancy and annual GDP per capita worldwide. The results gained in this study can provide an important resource for governments, researchers, and medical facilities around the world.
REFEERENCES
Basu, S., McKee, M., Galea, G., & Stuckler, D. (2013). Relationship of soft drink consumption to global overweight, obesity, and diabetes: A cross-national analysis of 75 countries. American Journal of Public Health, 103(11), 2071-2077. https://doi.org/10.2105/AJPH.2012.300974
Baum, C. L. (2011). The effects of food stamps on obesity. Southern Economic Journal, 77(3), 623-651. http://www.jstor.org/stable/40997278
Biddle, S., Cavill, N., Ekelund, U., Gorely, T., Griffiths, M., Jago, R., Oppert, J-M., Raats, M., Salmon, J., Stratton, G., Vicente-Rodriguez, G., Butland, B., Prosser, L., & Richardson, D. (2010). Sedentary behaviour and obesity: review of the current scientific evidence. Department of Health; Department for Children, Schools and Families. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/833151/dh_128225.pdf
Centers for Disease Control and Prevention (2018, January 29). Obesity. https://www.cdc.gov/healthyschools/obesity/facts.htm
Chetty, R., Stepner, M., Abraham, S., Lin, S., Scuderi, B., Turner, N., Bergeron, A., & Cutler, D. (2016). The association between income and life expectancy in the United States, 2001-2014. JAMA, 315(16), 1750-1766. https://doi.org/10.1001/jama.2016.4226
Drewnowski, A., & Specter, S. E. (2004). Poverty and obesity: The role of energy density and energy costs. The American Journal of Clinical Nutrition, 79(1), 6-16. https://doi.org/10.1093/ajcn/79.1.6
Fontaine, K. R., Redden, D. T., Wang, C., Westfall, A. O., & Allison, D. B. (2003). Years of life lost due to obesity. JAMA, 289(2), 187-193. https://doi.org/10.1001/jama.289.2.187
Foreyt, J. P., Goodrick, G. K., & Gotto, A. M. (1981). Limitations of behavioral treatment of obesity: Review and analysis. Journal of Behavioral Medicine, 4(2), 159-174. https://doi.org/10.1007/BF00844268
GDP per capita (current US$). (n.d.). The World Bank. https://data.worldbank.org/indicator/ny.gdp.pcap.cd
Iyengar, S., & Dholakia, R. H. (2012). Access of the rural poor to primary healthcare in India. Review of Market Integration, 4(1), 71-109. https://doi.org/10.1177%2F097492921200400103
Kashyap, K. (2018, January 11). How startups are trying to overcome India's healthcare challenges. Forbes. https://www.forbes.com/sites/krnkashyap/2017/09/25/how-startups-are-trying-to-overcome-indias-healthcare-challenges/#221c41ee1548
Kearns, K., Dee, A., Fitzgerald, A. P., Doherty, E., & Perry, I. J. (2014). Chronic disease burden associated with overweight and obesity in Ireland: The effects of a small BMI reduction at population level. BMC Public Health, 14, 143. https://doi.org/10.1186/1471-2458-14-143
Kravis, I. B., Heston, A. W., & Summers, R. (1978). Real GDP per capita for more than one hundred countries. The Economic Journal, 88(350), 215-242. https://doi.org/10.2307/2232127
Land, K. C., Guralnik, J. M., & Blazer, D. G. (1994). Estimating increment-decrement life tables with multiple covariates from panel data: The case of active life expectancy. Demography, 31(2), 297-319. https://doi.org/10.2307/2061887
Life expectancy at birth, total (years). (n.d.). The World Bank. https://data.worldbank.org/indicator/sp.dyn.le00.in
Mehta, N. K., & Chang, V. W. (2009). Mortality attributable to obesity among middle-aged adults in the United States. Demography, 46(4), 851-872. https://doi.org/10.1353/dem.0.0077
Molarius, A., Seidell, J. C., Visscher, T. L., & Hofman, A. (2000). Misclassification of high‐risk older subjects using waist action levels established for young and middle‐aged adults—results from the Rotterdam Study. Journal of the American Geriatrics Society, 48(12), 1638-1645. https://doi.org/10.1111/j.1532-5415.2000.tb03876.x
Monteiro, C. A., Moura, E. C., Conde, W. L., & Popkin, B. M. (2004). Socioeconomic status and obesity in adult populations of developing countries: A review. Bulletin of the World Health Organization, 82, 940-946. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2623095/
National Health Institute. (2017, August 1). Overweight & obesity statistics. https://www.niddk.nih.gov/health-information/health-statistics/overweight-obesity
NCD Risk Factor Collaboration. (n.d.). Adult body-mass index data visualizations. http://ncdrisc.org/data-visualisations-adiposity.html
Olshansky, S. J., Passaro, D. J., Hershow, R. C., Layden, J., Carnes, B. A., Brody, J., ... & Ludwig, D. S. (2005). A potential decline in life expectancy in the United States in the 21st century. New England Journal of Medicine, 352(11), 1138-1145. https://doi.org/10.1056/NEJMsr043743
Peeters, A, Barendregt, J.J.M, Willekens, F, Mackenbach, J.P, Al Mamun, A, & Bonneux, L.G.A. (2003). Obesity in adulthood and its consequences for life expectancy: a life-table analysis. Annals of Internal Medicine, 38(1), 24-32. https://doi.org/10.7326/0003-4819-138-1-200301070-00008
Reuell, P. (2016, April 11). For life expectancy, money matters. The Harvard Gazette. https://news.harvard.edu/gazette/story/2016/04/for-life-expectancy-money-matters/
Sobal, J., & Stunkard, A. J. (1989). Socioeconomic status and obesity: A review of the literature. Psychological Bulletin, 105(2), 260. https://doi.org/10.1037/0033-2909.105.2.260
Whiteman, H. (2014, May 29). Worldwide obesity rates see 'startling' increase over past 3 decades. Medical News Today. https://www.medicalnewstoday.com/articles/277450.php
World Health Organization (2020, April 1). Obesity and overweight. https://www.who.int/news-room/fact-sheets/detail/obesity-and-overweight
ABOUT THE AUTHOR
Ayush Malhotra
Ayush Malhotra is a grade 9 student currently attending the ib program at Cameron Heights Collegiate Institute in waterloo, Ontario. Ayush became passionate about research ever since he noticed how much of a toll obesity has on millions worldwide. Ayush enjoys traveling and experiencing different cultures as well as playing tennis at the provincial level. Currently, Ayush is working on an app for mental wellness during the COVID-19 pandemic called talkhAPPi and hopes to continue developing it come 2021.