There has long been evidence that socioeconomic status has a positive and linear correlation with general health and even more specific medical conditions. It has also long been believed that some of the disparity in health is related to the quality of nutritional intake. In the past, poor people suffered some diseases because the variety of food nutrition was poor. And if you got gout or were obese, you were probably rich.
Today that has been inverted: if you are obese and eat a lot of fast food, you are probably poor, it is believed. This narrative has been trotted out casually by numerous lazy self-proclaimed food experts, such Mark Bittman in the NY Times: “The ‘fact’ that junk food is cheaper than real food has become a reflexive part of how we explain why so many Americans are overweight, particularly those with lower incomes.”
Except for most pundits it seems to be just a way to criticize Big Food, sell some books criticizing Big Food or promoting a new diet, and maybe starting a food company of their own.
In a new review upcoming in Economics and Human Biology, the authors analyzed 15 studies (utilizing information from national economy surveys) and found that the relationship between socioeconmic status and fast-food is not linear at all - the result was mixed. That means we can't just throw on more labels, put on higher taxes, or have government subsidize advertising agencies to create media campaigns about food and have people get healthier. Their definition of fast food is non-beverage items “that lend themselves to production line techniques” or where “customers pay before receiving their food, which is then eaten on site, delivered, or taken out.” Basically, it's convenient and designed to appeal to our taste for sugar, salt, and fat – “mak[ing] fast foods high in calories and sodium and low in nutrients, and thus unhealthy.” That criticism is why so many fast food places now serve lots of supposedly healthy stuff (salads) that most customers don't really want to buy.
The National Longitudinal Survey of Youth(NLSY) began in 1979 (the authors specifically considered results from 2008, 2010 and 2012) so they were looking at people in their 40’s and 50’s. The response to the question “In the past seven days, how many times did you eat food from a fast-food restaurant such as McDonald's, Kentucky Fried Chicken, Pizza Hut or Taco Bell?” was the variable of interest.
They made a distinction between income and wealth, which may seem a little picky. How many people with low income really have high wealth? Mostly those dodging taxes. But here is their definition: “Income is the flow of money received periodically, predominantly from employment. Wealth is the stock of financial resources stored in bank accounts, stocks, bonds, homes, possessions.” Basically, income is flow, wealth is accrued. Simple enough, and using that socioeconomic status can be measured by both wealth and income. A new physician has high income, but little wealth (those student loans add up) while a retired physician has less income, but usually greater wealth. The National Longitudinal Survey of Youth tried to quantify both income and wealth and the researchers used other variables to define leisure time as well as health and health interests 
Here is what they found:
- Nearly 80% ate fast foods, about 24% were ‘frequent’  fast-food eaters. (average 3.6 meals/3 weeks)
- Fast food eaters had lower incomes and wealth, but were not poor, with mean incomes four times above the poverty threshold for a family of four 
- Fast food eaters had higher BMIs, were more likely to be trying to lose weight, less likely to check labels for ingredients and more likely to drink sugary beverages.
- Fast food eaters were more likely to live in central cities and the South , more likely to work and work more hours than non-eaters (and therefore have less leisure time), and have cars (presumably increasing access to fast-foods – evidently GrubHub was not a factor)
This seems obvious. People who eat a lot of calories and don't exercise had poorer health. But they have cars and work hard, so they aren't all that poor. Framed one way, it support the belief by New York Times writers that fast-food causes lots of problems.
But when the authors plotted income and wealth against the percentage of fast food eaters things looked different - the poor don't eat the most fast food, the middle class do, the association isn't significant in either rich or poor people, and smokers, who also tend to be poor, we are told, had a negative association with fast food.  Nothing was impacted by readily available nutrition information. If you were in the store, you were there to buy fast food.
Then the authors altered the variables to judge their impact on the outcome - a sensitivity analysis
- Changing income or wealth from the poorest category to the richest meant in little change in fast food consumption, about 10% (from 5.1 to 4.6 meals/3 weeks)
- Moving into or out of areas with a high density of fast food restaurants had little impact
- Decreases in leisure time produced modest changes in fast-food consumption. Going from 46 weeks of employment to no job reduced the predicted number of fast food meals by 20% (5.1 to 4 meals/3 weeks)
- The health variable had the largest impact on fast-food consumption, reducing it by 51% (5.1 to 2.5 meals/3 weeks)
So here is the uncomfortable take home for Mark Bittman, Michael Pollan, Marion Nestle, and everyone else insisting if you read their book you can stop liking tasty food: Income and wealth are not drivers of fast-food consumption. Fast food is not the fate of the poor. It is driven by access, convenience, and personal interest in their health.
Burger King did not make you fat, nor did TV commercials, eating too much did.
That means creating zones where fast food is banned are just going to annoy people and make them drive farther. They are not going to start wanting a kale salad just because it's the only thing allowed nearby.
People want convenience. A solution to fast food is not to make it inconvenient, but to make alternatives just as convenient and tasty. Wealthy urban elites in billion dollar newspapers criticizing poor people is just virtue signaling for readers, it is not helping anyone.
And more nutritional information, like highlighting some ingredients for special concern, is probably a waste of time any money. As the American Council on Science and Health has long maintained, making the calorie number larger does far more good than breaking out sugar or salt or falt.
 The survey indicating work status and weekly hours worked which was used as a measure of leisure time. Health and health interests were based on a prior history of smoking, exercise, how often they read nutritional label or lists of ingredients, drinking sugary beverages, their self-reported BMI and whether they were trying to lose weight.
 Frequent was defined as 4 or more times in a week.
 Mean income for fast food eaters $90,887, for non-eaters $107,827, poverty level is $22,025
 There is a greater density of fast food sources in these areas.
 The authors thought this might represent using nicotine to reduce food cravings (smoking as a dietary aid)