The Obesity Epidemic in the US
Robert Perez
April 29th, 2018
CDC Obesity Trends In The US
Introduction
Obesity has become a public health crisis in the United States. Obesity is caused by a number of reasons such as lifestyle, environmental and genetic factors. Obesity is creating problems that can be linked to significant health and social difficulties for people. This research focuses on three key factors age, education and income and attempts to figure out wether individuals are more at risk of becoming obese based on their age, their level of education or the amount of money they earn. The graph below indicates the relationship between obesity rates in the US and the factors effecting these rates.
Literature Review
Despite growing recognition of the problem, the obesity epidemic continues in the US and obesity rates are increasing around the world. According to the the NCBI, “The latest estimates are that approximately 34% of adults and 15â“20% of children and adolescents in the U.S. are obese. Obesity affects every segment of the U.S. population.(Mitchell, N., Catenacci, V., Wyatt, H. R., & Hill, J. O. (2011)) More than one-third of adults in the US are obese.(CDC) The CDC says that among non-Hispanic blacks and Mexican American men, those with higher incomes are more likely to have obesity than those with a lower income. projection models show that by the year 2030, ~90% (86.3%) of all American adults would become overweight or obese and 51.1% of them would become obese. (Wang, Beydoun (2008)) Obesity increases the risk of many chronic diseases in children and adults.” A large amount of research is now directed toward better understanding and treating obesity, and substantial public health efforts are directed toward reducing obesity rates. To date, however, there is little evidence of success in reversing the epidemic in the U.S.
Research Questions
Are individuals more susceptible to becoming overweight at a certain age?
Is there a relationship between degree of education, or income, on obesity?
Furthermore, have the percentages of obese people amongst various groups changed at all in the past 5 years?
Data and Variables
The data for this study was obtained from the CDC. This dataset includes data on adult’s diet, physical activity, and weight status from Behavioral Risk Factor Surveillance System. This data is used for DNPAO’s Data, Trends, and Maps database, which provides national and state specific data on obesity, nutrition, and physical activity. The variables we examined from the dataset were Pct_obese(rate of obesity), state, age, education and income. The mean percent of obesity was found and compared with each of the factors mentioned. This was followed by an ANOVA test and then illustrated in a nice bar graph to further illustrate the relationship between the variables.
Description Of The Variables:
1.Question: Percentage of obese adults, percentage of overweight adults.
2.State: full name of state
3.R: Race/ethnicity of respondents
4.Pct_obese: Percentage of respondents who fit the questions above
5.Gender: Gender of respondents “male”“female”
6.Age: Age of respondents
7.Income: Yearly Income of Respondents
8:Education: Level of education Completed
Observing the mean obesity rates by year
| YearEnd | mean | stdev |
|---|---|---|
| 2016 | 30.09 | 7.04 |
| 2015 | 29.53 | 6.96 |
| 2014 | 29.34 | 7.00 |
| 2013 | 28.82 | 6.78 |
| 2012 | 28.00 | 6.73 |
| 2011 | 27.74 | 6.76 |
From the chart above we can see a steady increase in obesity rates over this five year period. While this overall increase of just over 2% may seem low. In the overall big picture this can be considered a devestating increase.
Observing highest obesity rates by state
By observing obesity rates by state we can see that states within the South region are the states with the highest overall mean obesity rates. The states in the Northeast region were at the bottom for over mean obesity rates.
Mean obesity rate by age interval
## # A tibble: 6 x 2
## Age mean
## <chr> <dbl>
## 1 18 - 24 16.4
## 2 25 - 34 27.2
## 3 35 - 44 32.4
## 4 45 - 54 33.8
## 5 55 - 64 33.6
## 6 65 or older 26.8In the table above one can see that obesity rates peak around the ages of 35-64. People ages 65 and older have a slightly lower mean obesity rate then the above interval. However, whats most apparent about the table above is that people in their early adult years have the lowest rate of obesity as opposed to individuals between the ages of 35-44 which have the highest rate of obesity.
Regression Model by age
##
## Call:
## lm(formula = Pct_obese ~ Age, data = age)
##
## Residuals:
## Min 1Q Median 3Q Max
## -14.8612 -2.8472 0.0388 2.7104 16.4009
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 16.3991 0.2352 69.71 <2e-16 ***
## Age25 - 34 10.7622 0.3327 32.35 <2e-16 ***
## Age35 - 44 16.0431 0.3327 48.22 <2e-16 ***
## Age45 - 54 17.4481 0.3327 52.45 <2e-16 ***
## Age55 - 64 17.1616 0.3327 51.58 <2e-16 ***
## Age65 or older 10.3559 0.3327 31.13 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 4.208 on 1914 degrees of freedom
## Multiple R-squared: 0.6759, Adjusted R-squared: 0.675
## F-statistic: 798.3 on 5 and 1914 DF, p-value: < 2.2e-16From the above we can see that the p-value is extremely low meaning that age has a statistically significant effect on obesity rates. We also notice that based on the r-squared value that 67% of the change in obesity is explained by age.
The graph above clearly illustrates that younger adults have a lower obese rate than individuals between the ages of 35 and 64.Mean obesity rate by education
## # A tibble: 4 x 2
## Education mean
## <chr> <dbl>
## 1 Less than high school 33.0
## 2 High school graduate 30.9
## 3 Some college or technical school 30.1
## 4 College graduate 22.8The table above shows that as an individuals education increases the obesity rates steadily decrease.
Regression model by education
##
## Call:
## lm(formula = Pct_obese ~ Education, data = educ)
##
## Residuals:
## Min 1Q Median 3Q Max
## -11.726 -2.591 0.067 2.574 10.690
##
## Coefficients:
## Estimate Std. Error t value
## (Intercept) 22.7912 0.2116 107.71
## EducationHigh school graduate 8.1484 0.2992 27.23
## EducationLess than high school 10.2350 0.2992 34.20
## EducationSome college or technical school 7.3191 0.2992 24.46
## Pr(>|t|)
## (Intercept) <2e-16 ***
## EducationHigh school graduate <2e-16 ***
## EducationLess than high school <2e-16 ***
## EducationSome college or technical school <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3.785 on 1276 degrees of freedom
## Multiple R-squared: 0.5104, Adjusted R-squared: 0.5093
## F-statistic: 443.5 on 3 and 1276 DF, p-value: < 2.2e-16From the model above we can see that eduction has a statistically significant effect on obesity rates at a 95% confidence level. The r-squared also suggest that 51% of change in obesity rates can be explained by the level of education an individual has attained.
The Graph above clearly illustrates that the lower education you have, the the more likely you are of being obese as opposed to those with a college education.Mean obesity rates by income
## # A tibble: 7 x 2
## Income mean
## <chr> <dbl>
## 1 Less than $15,000 33.7
## 2 $15,000 - $24,999 32.2
## 3 $25,000 - $34,999 30.6
## 4 $35,000 - $49,999 30.4
## 5 $50,000 - $74,999 29.7
## 6 $75,000 or greater 25.2
## 7 Data not reported 24.3The table above provides the mean rate of obesity by income level. As we can see individuals who had an income of $75,000 or greater had the lowest rate of obesity as opposed to those who made less than $15,000.
Regression model by income
##
## Call:
## lm(formula = Pct_obese ~ Income, data = income)
##
## Residuals:
## Min 1Q Median 3Q Max
## -13.1516 -2.7516 0.0481 2.7484 13.6191
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 32.2247 0.2277 141.515 < 2e-16 ***
## Income$25,000 - $34,999 -1.6728 0.3220 -5.195 2.24e-07 ***
## Income$35,000 - $49,999 -1.8731 0.3220 -5.817 6.87e-09 ***
## Income$50,000 - $74,999 -2.5437 0.3220 -7.899 4.37e-15 ***
## Income$75,000 or greater -6.9731 0.3220 -21.653 < 2e-16 ***
## IncomeData not reported -7.9097 0.3220 -24.562 < 2e-16 ***
## IncomeLess than $15,000 1.5194 0.3220 4.718 2.53e-06 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 4.073 on 2233 degrees of freedom
## Multiple R-squared: 0.3841, Adjusted R-squared: 0.3825
## F-statistic: 232.1 on 6 and 2233 DF, p-value: < 2.2e-16From the model above we can see that income has a statistically significant effect on obesity rates at a 95% confidence level. The r-squared also suggest that 38% of change in obesity rates can be explained by the level of income an individual earns.
Discussion
In order to understand how to head this country in the proper direction when it comes to obesity we must first understand the factors that are effecting this countrys rates of obesity. In the above tables we were able to look at a seperate demographic group at a time and see if it significantly effected the percentage of adults that were obese in our country. Only after understand the key factors making us obese can we than as a country put forward the effort to make a change. From the above results we see that age, education and income all were keys factors in the percentage of obese people in America. Seems that if we educate ourselves, and earn a large income we have a better chance at not being obese.
This study has its limitations and therefore this study does not guarantee that these factors are the direct cause of obesity.
Bibliography
Centers for Disease Control and Prevention. ( 2010). Nutrition.
Mitchell, N., Catenacci, V., Wyatt, H. R., & Hill, J. O. (2011). OBESITY: OVERVIEW OF AN EPIDEMIC. The Psychiatric Clinics of North America, 34(4), 717â“732.
Reichmann, Vanessa, “Does Fruit and Vegetable Intake Decrease Risk for Obesity in Children and Adolescents?” (2009). Undergraduate Honors Theses. Paper 8.
Young, L. R., & Nestle, M. (2002). The Contribution of Expanding Portion Sizes to the US Obesity Epidemic. American Journal of Public Health, 92(2), 246â“249.