Author: Ting Lei
1. Introduction
Obesity is one of the biggest health problems across the United States. According to the data from the National Health and Nutrition Examination Survey:
- More than 1 in 3 adults are considered to be obese.
- More than 1 in 20 adults are considered to have extreme obesity.
- More than 1 in 6 teenagers age from 6 to 19 are considered to be obese.
Obesity is usually accompanied with various other diseases, which combined kill millions of people per year. These diseases include type 2 diabetes, heart disease, high blood pressure, nonalcoholic fatty liver disease, osteoarthritis, cancer, stroke and various others. The goal of this article is to provide a spatial overview of communities’ ability to access healthy food and quality diet, and to explore some environmental factors that are leading causes of overweight and obesity.
2. Analysis
2.1. Software Used
I used R to pre-process the dataset into a more suitable structure for analysis. Then, Tableau is used to provide data visualization and analysis.
2.2. Dataset
Our data is acquired from United States Department of Agriculture (USDA). The dataset is the August 2015 version of USDA Food Environment Atlas that includes all states in the U.S. and their 200 attributes such as population, fast food store count, recreation & fitness facilities count, diabetes rate, obesity rate, National School Lunch Program participants rate, Household food insecurity rate, etc. These data are from a variety of sources and cover several years and geographic levels.
2.2. Approaches
In order to validate the dataset, we generated a simple scatterplot between diabetes rate and obesity rate for each state in the Figure 1 below. The reason we plotted such two variables was the well-established relationship between diabetic and obese. The strong positive correlation shown in the graph below gave us the confidence that the dataset agreed with our expectation.
Again, we can see from Figure 1 that obesity clearly put Americans at higher risk of chronic diseases like diabetes.
Figure 1: Obesity rate by State vs. Diabetes rate by State
We are also interested in the adult obesity rate for different states in the United States, which is shown in figure 2. Red represents high obesity percentage whereas green vice versa. From this figure, we find that Mississippi (36.70%) has the highest rate of adult obesity in the country, placed ahead of Louisiana (35.55%), Alabama (35.08%), and South Carolina (34.65%). Hawaii (19.22%) has the lowest rate of obesity, with Colorado (20.68%), Massachusetts (23.39%), California (24.22%), and Connecticut (24.35%) being the four states with less than 25 percent of obesity.
Figure 2: Obesity Rate across USA
To summarize, the southeastern United States have the highest rates of obesity in the country compare to others. And below we will try to explore the possible factors that are causing the high obesity rates in these states.
Ethnics and Racial Disparities
According to stateofobesity.org, African-Americans are nearly 1.5 times more likely to be obese compared with Whites and Latino adults. Approximately 47.8% of African Americans are obese while the rate for Whites is 32.6 %. Figure 3 below clearly shows that the southeastern states are the states with highest African-American population, which may be one of the reasons for the high obesity rate among these states.
Figure 3: African American Percentage Across USA
Income Level and food insecurity
Another interesting factor related to obesity is the income level, since income level determines what kind of food one can acquire.
Figure 4: Income level vs. Obesity Rate
Figure 4 above shows that low-income and poverty have a strong correlation with an increase in obesity rate. This could be explained by the fact that most affordable or less expensive food are either less nutritious or calorie-dense. Also, people in poverty may not have consistent access to healthy food since they have limited access to supermarkets and fresh produce. On the other hand, gas stations and convenience stores are their main food sources, but these stores mostly sell sodas, candy, and junk foods only.
This phenomenon is what we call Food Insecurity. And with such less healthy eating pattern, it can be a contributing factor to weight gain and obesity. Figure 5 below proves that increase in Food Insecurity results in the increase of obesity rate with R-squared equal to 30%. It is also worth noticing that Mississippi, the highest obesity rate among the US, also has the highest food insecurity rate at 20.9%.
Figure 5: Food Insecurity vs. Obesity Rate
Recreation & fitness facilities
Achieving healthy energy balance also requires being physically active regularly. And one reason for people not engaging in sufficient amount of physical activity is the lack of places for recreation and fitness facilities. Figure 6 below shows the count of recreation & fitness facilities for each state. Orange indicates the states with the lowest obesity rate across the United States, whereas blue indicate the states with the highest obesity rate across the United States.
Figure 6: Fitness Facilities count / State
In the figure, we found that higher count of fitness facilities is more likely to support people with healthy lifestyle habits. In fact, it strongly influences the amount of physical activity. Four out of 5 states with higher numbers of fitness facilities are associated with the lowest obesity rate across the United States.
3. Results and Conclusion
In this article, we demonstrate the use of Tableau and R to run data visualization regarding to obesity in the United States. As a result, obesity is an actual and serious health problem in United States especially in southeastern states. The wide availability of junk food and its low price have maximized the chances of impulse purchases for families in poverty. Also, the lack of places to recreation discourage people to engage in physical activity.
Therefore, the government may consider increasing their food security, enforcing state food policy on junk food, and increase affordable gyms that make people to be physically active in order to decrease the obesity rate.
Reference
Flegal KM, Carroll MD, Kit BK, Ogden CL. Prevalence of obesity and trends in the distribution of body mass index among US adults, 1999–2010. Journal of the American Medical Association. 2012; 307(5):491–97.
“Maximizing The Impact of Obesity-Prevention Efforts In Black Communities: Key Findings and Strategic Recommendations.” Special Report: Racial and Ethnic Disparities in Obesity. N.p., Sept. 2014. Web.
I really like this topic, and obesity is really a increasing healthy problem nowadays in the United States, so this one actually helps people see what is going on.
I strongly agree with the correlation between food insecurity and obesity. On the other hand, the effect of fitness level can be argued. Many people are active and are still not able to maintain proper weight level. Excercise seems to affect general health and good feeling more than body weight
I think this is a very interesting choice of topic and really like the application of statistical analysis to a dietetics problem. However, given the first set of statistics outlined in the introduction, it appears as though another hypothesis test could be done to see if the true proportion of obesity is different than the 1 in 3 adults/1 in 6 children/1 in 20 extremely obese.
Very interesting analysis. The variables you included clearly impact obesity rates. Considering the map of the US and the wide variation in obesity levels across states, I’d be curious to further explore what differences between the states drive the most difference in obesity levels. It might be interesting to include census data and all of the granular demographic information it includes to compare against obesity levels to further analyze this.
Interesting analysis Ting. I find it strange that you called people of lower income getting fast food an impulse purchase. Often times this is the only food they can afford or have time to get between working day and night jobs, leaving them little time to prepare their own meal of higher quality. It would make sense for the government to make healthy food more affordable for these low income families so that their health can improve.
Also in regards to the analysis of a lack of recreation areas in the southeastern US, I would argue that since it is humid subtropical climate it is much easier to get outside each and every day with relative comfort. That is compared to the midwest where there are days where it painful to go outside due to the weather.
I found this analysis very interesting. Coordinating the accessibility of a fitness facility with obesity rates is interesting and intuitive. You mentioned that you did an analysis of 2015 data, it would be interesting to track this data as it has changed over time and seeing trends in how obesity has maybe grown as an issue in this country at a potentially alarming rate.
Very interesting analysis. Another factor that may be worth considering is the typical diets of the inhabitants of various states. Outside of what is available at fast food restaurants, different foods are typically consumed by different regions. For example, many popular southern dishes are extremely calorie dense and often contain large amounts of fried foods. I would think that this has some correlation to the fact that southeastern states have the highest rates of obesity.
This is a very interesting and important research topic to look at. You talk a bit about the various diseases associated with high obesity in a population and eliminating high obesity would hopefully limit the prevalence of these diseases. I would be very interested in comparing the US and other countries around the world to understand whether the same factors are in place. Is there some sort of an equivalent to the southeastern US states in France somewhere? Are the same types of factors at play? This would be a much more involved project but I believe a similar analysis to what you’ve already done could be used. Nice work on this article.
Interesting topic and one that needs to be looked more into. It would be interesting to see how obesity in the younger generation has changed since michelle obama started the campaign to get kids more exercise
This is exactly what I thought of! I’d like to see how her campaign periods have impacted obesity levels in specific areas after the campaigns have ended. It would be interesting to even take it a step further and look at specific school districts.
Interesting analysis. I wonder if building more fitness centers would decrease obesity levels. I think the relationship potentially runs in the other direction: that populations that are more health conscious and affluent, such as those in Connecticut, are more likely to seek fitness centers. I bet big national chains like Anytime Fitness are pretty conscious of demand, especially since they’ve likely already saturated markets such as Connecticut. My guess would be that if they aren’t building more fitness centers in Mississippi, it’s because they don’t predict that the demand is there.
This is an extremely interesting analysis of n issue affecting millions of people. I thought the income analysis and food security analysis was particularly interesting, showing a trend that I would expect. I was shocked that the African-American obesity rate is 1.5 times the White obesity rate. As you state the correlation between the state’s obesity rate and African-American population, I wonder how much the income level correlation with obesity has to do with the African-American obesity rate being 1.5 times greater. I know the African-American poverty rate is much higher than the White poverty rate. Great post!
Very interesting topic! I particularly liked how you gave a visualization to which states had the highest obesity rate. It was eye opening to see that the country was almost divided in half by having a high obesity percentage versus a low one. It makes you wonder maybe what effect the environment you live in had on your weight or eating habits.
Very interesting! Small states as Rhode Island or Massachusetts seems to have better results than bigger states, it could be interesting to perform a geographical analysis with cities or smaller regions in order to know if theres is a big difference between big cities and the country side.
it would be also interesting to analize university areas where people is usually practising more sport versus rural areas.
Definitely a topic of concern. I think it is very discouraging that healthy foods tend to cost more. The premise makes senses but is a concerning fact.
We often talk about price differences of healthy vs non healthy food and if you are more poor, you tend to eat worse due to this. To be able to actually quantify this fact and see the data in front of you is really interesting. I have never heard of the term “food insecurity” but after reading this but it obviously makes a lot of sense with the analysis. Even though the r^2 value is low, there is a clear trend.
This is a very interesting analysis. It would be very interesting to analyze the role that culture and reception plays on specifically the analysis of healthy food and number of recreational facilities. For example, is there any reverse causation taking place with the recreational facilities. Perhaps the reason there are more recreational facilities in Massachusetts is because they value the culture values activity more and that’s why more were built. It would be interesting to see which factor came first. A similar analysis could be done with the number of fast food restaurants and lack of healthy food.
Thank you for sharing your analysis. I thought that it made sense that the more fitness centers, the lower the obesity rate in the state. I did find it interesting however that the highest obesity rates did not necessarily have the lowest number of fitness centers though. I would be interesting in seeing how obesity changes over the next few years based on new healthy school lunches being implemented, more education to the public about health eating, and implementation of transportation sharing.
As someone really interesting in healthcare I know that obesity related deaths is the number one cause of death in the United States. It’s truly a public health crisis, and I really like the way you quantified and visualized the issue. I hadn’t thought of obesity being a monetary issue, but there is a clear correlation between higher rates of obesity and worse off areas. Not only can people not afford good food, but they can’t afford gyms, workout equipment, etc.
Great topic to cover in the United States, where obesity is more of a problem than anywhere else in the world. I thought the authors did a good job of preprocessing the data and validating its significance showing that obesity leads to life-threatening conditions. I found it interesting that the 4 states with the lowest obesity rates also had the highest prevalence of fitness centers, a statistic I don’t think I have ever seen before (and made much clearer by Tableau!). I also liked the way that the article was structured and ended by offering suggestions on how we could possibly improve national obesity rates.
Thank you for sharing your work. It was informative to see the relationship between obesity and income level. It shows that obesity is caused by many underlying factors including income. On the last graphic it was interesting to see some states had a ton of workout facilities, but it may have been slightly misleading since it was not adjusted per capita.
Thank you for sharing your work. It is kind of quite interesting to see the negative correlation between income and obesity as conventional wisdom would lead us to think otherwise. Also the I appreciate the graphics use as it shows the linear trend and the variation of the data.
I think that considering “age-group” into your analysis would be very interesting. This would complement income level very well. As families have less money to spend food on, they usually resort to fast-food, which in turn affects the kids the most because they will grow up eating fast-food and will be at much higher risk of diabetes. Another note to add is that kids who grow up mainly eating fast-food may continue this trend into their future, regardless of income. I wonder if those fast food fanatics will keep eating unhealthy even if they have money to spend money on more healthy options, which is more expensive and less efficient (cooking time).
Obesity and diabetes have been a hot topic in US health new for quite some time now and this post is no exception. While states in the Southeastern US suffer the most from these health issues, the states that have the lowest rates of obesity and diabetes have some of the highest rates of recreational centers and food security. While increasing rec centers and food security in the southeastern United States may help, I think if would be interesting to see where people suffering from obesity and diabetes are living (City, Suburbs, rural, etc).
Thanks for sharing your finding. I am wondering if commuting time would contribute to obesity level. There are multiple transportation options for people living in big cities, buses, subway, bike or even walk. It becomes more difficult for people have to commute by car since the work location is not easily accessible through other methods of transportation. It would be interesting to see results and maybe more availability of public transportation would help reduce obesity. We can also compare US obesity results to other countries where public transportation is more readily available.
I found this very interesting as someone who used to be on government assistance (food stamps in the US) and now at an income level where I can afford to get what I want, health or not. This report is spot on when it talks about the correlation between not having money to by healthy food and having money to buy healthy food. When we were on assistance in my 20’s it was easy to focus on substance to survive, which is not always (most of the time is not) healthy. But, when you get to a point that you have the ability to buy what you want you may not chose health because that person is fixated on food (having it) and even though it is health food they usually over eat and stay obese.
On another note I think the BMI standard is set to tight and it depends on your body type. Not everyone has a genetically small frame and it also is where you feel comfortable. For example if I was to drop 25lbs I would have a constant headache but do not at the weight I am now, so I am comfortable where I am.
Good points, David. While I have been fortunate enough to never be on government assistance, there are numerous reports (as evidenced by this submission) that show strong evidence for correlation between income and food choices. However, as you also pointed out, the presence of “expendable” income does not automatically translate to better eating choices: that is often at the discretion of the buyer, and their preferences, knowledge about healthy eating, and general lifestyle. I think you’re absolutely correct that many of the people with higher incomes tend to overeat “healthier” food items, leading to sustained obesity even when they can afford to eat “clean”.
I’m not sure I agree with your statement that the BMI standards are too tight, though. Taken directly from the CDC’s website (https://www.cdc.gov/healthyweight/assessing/bmi/index.html):
“Body Mass Index (BMI) is a person’s weight in kilograms divided by the square of height in meters. A high BMI can be an indicator of high body fatness. BMI can be used to screen for weight categories that may lead to health problems but it is not diagnostic of the body fatness or health of an individual.”
This seems to indicate that BMI is not used as a serious diagnostic tool for determining how healthy a person is, but is rather used as an indicator that an individual may have high levels of body fat, which has direct ties to a number of health issues.
One other interesting note: my current research effort for ISyE 412 examines tobacco consumption, and one of our analyses was regarding the consumption of cigarettes regionally in the US. Our data seems to indicate very similar trends to those shown in this report: namely that the highest prevalence of smoking occurs in the southern and midwestern states. Perhaps, as we look further into these datasets, we can find a common causal factor between smoking and obesity (could be tied to education level, as southern states generally rank low in studies of the overall intelligence of a state’s population).
Great topic. I really appreciate Figure 6 and your inclusion of the importance of recreational activities. I would attribute much of our obesity to the overabundance of low quality food and lack of physical activity. While somewhat off topic, if you want to see something that will shock you in a healthy way, I recommend a documentary that I once watched, The Science of Fasting.
The percentage of obesity by state is almost exactly what I would have guessed it to be. Interesting that the south and the midwest are very similar considering the difference in weather patterns.
It was surprising to see that the deep south has the highest obesity rate. I would have predicted a high obesity rate in the south, but not higher than the regions further north in the Midwest. This is because cold weather would have seemed like a good indicator of high obesity rates where it is harder to be active outside. However, wealth is certainly important as well. As a low income college student, I have noticed the quality of my diet is lesser than it was at home with my parents. It would be interesting to look at how average household income by state correlates to the state’s obesity rate. I predict a positive correlation given figure 5’s positive correlation without filtering the data by state.
Very interesting analysis! I think its really fascinating that you decided to look at recreation and fitness facilities, and the correlation between them seemed intuitive. I also agree with your analysis that low-income and poverty seem to have a strong correlation with an increase in obesity rate. Another interesting factor to consider is profession/career. Are people with desk jobs more likely to have higher obesity rate? Stress is also often correlated with weight. Does stress level at job affect obesity rates?
Obesity is a huge problem plaguing the United States and I think this article highlights important concepts. I thought it was smart to see the relationship between diabetes and obesity.
I completely agree with the analysis of income level and the food security. I think this article really touches an important topic in obesity as it really is a huge threat to humans in the modern world.
Thank you for your analysis. I think that your analysis really highlights the extent to the problem obesity has been in the United States.
I like the topic because obesity is a serious health problem around the world and specifically in the United States; I almost did my 412 project on obesity. Your graphs and data show that obesity has many underlying factors which is why it’s been such a hard issue for health groups and governments to tackle. I am curious if there is a correlation between the type of cuisine in a particular area and obesity. For example, in the South, their cuisine or more traditional dishes are often fried which may be one of the factors as to why they have the highest obesity rates. Another idea would be to compare a states obesity to outside temperature. While there are indoor gyms, in some areas like the South or the North it may be too hot or too cold to workout outside most of the year. Overall, this was a thoroughly executed project.
Your insights on obesity rates are very interesting because of their reliance on eating habits. All of your points line up to prove that food insecurity is a main cause of obesity, which is a very strong case for your results. For example, lower income level leads to higher food insecurity (qualitatively, for now), which leads to higher rates of obesity. This agrees with the data since lower income results in higher rate of obesity, generally. This could be further proved by correlating these factors to food insecurity directly. For instance, analyzing the effect on race on food insecurity. It can be shown that African-Americans have lower food insecurity and this is why they have a higher rate of obesity, and this will further prove your point that we need to increase food security as a nation.
I’m curious what the different colors represent on figure 6.
I found this topic very interesting. The map of obesity by state perfectly showed how the “flyover states” are more obese than the coasts. After reading I was curious about whether gender played a role in obesity.
I found this topic extremely interesting, however the statistics do not surprise me. I think these statistics show us that we really are what we eat and reinforces how unhealthy eating habits can lead to lifelong diseases like diabetes. The government should make an effort to increase healthy food rations in places with high rates of obesity. This will cause a shift in the type of food consumed by people. Ultimately, this is a public health crisis and should not disproportionately affect people because of the color of their skin or their income.
Interesting to see that the obesity rate is lower for the western part of the United States.