Unexpected food sources & disparities

Over the past month I gave two talks on unexpected food provision and disparities in urban communities. One talk was at the American Public Health Association (APHA) annual meeting in Denver. The other talk was at the North American Primary Care Research Group (NAPCRG) meeting in Colorado Springs. A combination slide set is available here. A rough slide-by-slide transcript might go something like this:

Slide 1 – Introduction

Sean Lucan is an Associate Professor of Family and Social Medicine, a practicing family physician, a former Robert Wood Johnson Foundation Clinical Scholar, and a fellow at the National Academy of Medicine (formerly Institute of Medicine). Dr. Lucan’s research focuses on how different aspects of urban food environments may influence what people eat, and what the implications are for obesity and chronic diseases, particularly in low-income and minority communities.

I’m going to start off today giving a fair bit of background, so you can understand my work in the context of patient care and other research in the field.

Slide 2 – Disclosures

I have no real or perceived conflicts of interests or disclosures relevant to this presentation.

Slide 3 - Doctors promote healthful behavior & manage disease

As noted, I am a family doctor. And like other family doctors, and primary care providers more generally, I promote healthful behaviors--ideally to prevent chronic diseases from occurring or, if not, to manage them when they occur.   I actually spend a good deal of my professional time counseling patient on nutrition and healthful dietary intake. Unfortunately, the great efforts I can expend in clinic often get minimized …

Slide 4 - Food environments challenge doctor advice / patient intent

… or overwhelmed, because when patient leave my office, they have to navigate this and this, and this, and this.  These are just few pictures of streets in the Bronx, where I practice and where my patients live.  They show that food environments can really challenge doctors’ advice for healthful eating and patients’ best intentions to eat healthfully.  So I love this quote from Harvard’s David Williams from a NAPCRG plenary a few years ago: “What if we treat illness and then send them back to the same conditions that made them sick in the first place?”

Slide 5 - Changing Environmental Context

The point is that environments matters. Here is a figure from a 2010 paper in the American Journal of Public Health (AJPH) by former commissioner of health for New York City and current Director of the Centers for Disease Control and Prevention, Tom Frieden.  What the figure suggests is that rather than trying to intervene on behavior at the level of the individual--as in doctors’ offices, counseling and educating patients one-on-one, where we put in a lot of effort and generally have very little population impact--maybe we would do better to address environmental context. 

Slide 6 - Food-Environment Studies, 1990 - 2015

Well certainly Tom Frieden and I are not the only ones to have this idea.  This is a plot showing studies focusing on food environments that have been published over time. What you can see is an exponential increase in the research in this area starting about 10 years ago.  In 2015, the rate was about 200 papers per year and rising.   So a lot of research is being done. The problem though …

Slide 7 - Limitations of food-environment research

… Well actually there are several. In fact in 2014, I published a review of the literature and commentary that detailed some of the most important limitations in this exploding field. The one I want to focus on today is this …

Slide 8 - Food-Environment Studies, Jan ‘09 - May ‘10

… Now this is not my data (it actually comes from colleagues at NCI) and the slide is a little bit dated. But I think the findings still hold today and the essential point is that most of the literature on food environments has focused very narrowly and disproportionately on food stores (particularly supermarkets) and restaurants (particularly fast-food outlets), to the exclusion of nearly all other sources of food. This is a problem, because there are so many other food sources out there.

Slide 9 - Other food sources in local environments

For instance, in the Bronx neighborhoods where my patients live (and in many other settings across the country, both urban and rural), there are mobile food vendors--i.e., street carts, trucks, and stands selling food. I have discussed mobile food vending in the past and here is a list of the papers that came out of that work (paper 1, paper 2, paper 3). I have also presented and published on farmers’ markets in the past, which are another source of food in the Bronx and in other urban and rural communities.

Slide 10 - Non-intuitive storefront food retail

And then there are places like pharmacies … and dollar stores … and sports stores … and quickie lubes … and laundromats … and all the various other storefront businesses that, while not normally considered ‘food stores’, are very often sources of food.  And those foods tend to be highly-processed, pre-packaged, sugary snacks, sweetened drinks, and frozen confections.

Slide 11 - Ubiquity of energy-dense snack foods

In fact another paper, also from 2010, also in the American Journal of Public Health, and also by a guy named Tom who also just happened to be a commissioner of health in New York City (this time Tom Farley), the ubiquity of energy dense snack foods was obvious. Looking at retail outlets in 19 US cities, Farley et al showed that substantial percentages of storefronts--from pharmacies and gas stations, to hotels and salons, to furniture store and apparel shops--offered some kind of food or drink. Specifically, they offered candy, sugar drinks, salty snacks, and/or baked or frozen sweets.  

So that pretty well characterizes the situation for storefront businesses. But the

Farley-et-al study didn’t additionally consider non-storefronts like mobile food vendors and farmers markets. It also didn’t consider food and beverage items beyond those listed here and it didn’t assess for potential difference in food offerings by neighborhood/neighborhood characteristics.

Slide 12 - Study objectives

So for the current study, our objectives were: (1) to assess *all* food-and-beverage items available through *all* businesses in distinct urban communities, and (2) to assess differences by neighborhood, specifically the Bronx (lower-income, minority communities) compared to the Upper East Side (UES) of Manhattan (an affluent and mostly white community).

Slide 13 - Observing all business: storefronts and sidewalk sellers

For our methods, investigators observed *all* businesses (which wound up being well over 1,000) on a sample of street segments in the Bronx (n = 155) and in the UES (n= 51).  Businesses included storefront retail (fast-food and table-service restaurants, various stores, etc.) and sidewalk sellers (mobile food vendors and farmers’ markets).

Slide 14-15 - Maps of the study area

Just for orientation for those unfamiliar with the study areas, this is a map of New York City, which includes the island of Manhattan as well as the 4 surrounding boroughs of Staten Island, Brooklyn, Queens, and the Bronx (where I practice and where most of my patients live).  Across the bridge (as least when Governor Christie keeps it open) is New Jersey and up top here is Westchester County, NY.  Now if I zoom in a little on the areas of interest ….

Slide 16 - Blow up of the study area

… the highlighting represents the streets that were sampled and you can see we tried to get a representative slice across the regions. 

Slide 17 - Assessments and analyses

On selected streets, researchers systematically assessed all business for offered foods, including ‘healthier’ items (i.e., fruits, vegetables, whole grains, nuts) and ‘less-healthful’ items (e.g., the candy, chips, and baked and frozen sweets from the Farley-et-al study). They also assessed for beverages: ‘healthier’ drinks (water, milk), ‘less-healthful’ drinks (sodas, alcohol), and ‘other’ drinks (diet drinks, juices).

Descriptive analyses considered ‘food businesses’, that is outlets primarily focused on food provision (e.g., super-markets/groceries, restaurants, farmers’ markets, green carts) versus ‘other businesses’ (e.g., clothing stores, beauty salons, gas stations, hardware stores, department stores, news stands, etc.)

Analyses also considered the Bronx vs. the UES

Slide 18 - Typical streets revealed some characteristic differences

This is what we found. These photos represent typical streets in the UES and the Bronx and reveal characteristic differences. The UES had lower retail density and fewer businesses offering food or drink. Additionally, for restaurants, those in the UES tended be sit-down with table service whereas those in the Bronx tended to be fast-food or take-out/take-away outlets …

Slide 19 - ‘Other businesses’

… which is the first bullet of this slide. Another key difference between neighborhoods related to ‘other businesses.’ Interestingly, ‘other businesses’ outnumbered ‘food businesses’ in both the UES and the Bronx almost precisely by a ratio of 3:1. Several ‘other businesses’ were found in both the UES and the Bronx (e.g., banks, doctor’s offices, hardware stores, pharmacies), but some were unique to the UES (e.g., art galleries, fashion boutiques, spas, tea shops), whereas others were unique to the Bronx (e.g., auto shops, dollar stores, laudromats, and pawn shops).

Slide 20-21 - Offering any food or drink or only ‘less-healthful’ items

Considering all business, in the UES, more than 1 in 4 offered some kind of food or beverages (Note: business offering *only* water--in bottles, by fountain, or in coolers--were excluded to be conservative, although including these businesses did not meaningfully inflate proportions).  In the Bronx, the number offering food or drink was much higher--not quite 1 in 2 but definitely more than 1 in 3.

Now if we exclude ‘food business’ (the supermarket, grocery stores, various restaurants and other businesses primarily focused on food provision) so as only to consider ‘other businesses’, the percentages are much lower. Still, you could find something to eat or drink from places such as newsstands, home furnishing stores, and veterinarian offices in the UES, and from many more kinds of places in the Bronx and in almost 3 times as many cases.

Notably, no businesses in either neighborhood sold ‘healthier’ items to the exclusion of ‘less-healthful’ items.  But there were business in both areas that sold only ‘less-healthful’ foods. In the UES, the percentage was nearly 10%; in the Bronx it was about twice as high.

If only considering ‘other businesses’, the percentage selling ‘less-healthful items only’ in the UES was more than a third and in the Bronx nearly half.

The bottom line is that in the Bronx there were many more businesses offering any food or drink, and higher percentages offering only ‘less-healthful items without healthier alternatives. 

Slide 22 - Example: dollar store

I want to end with some photos just to show you what some of this food and beverage provision actually looks like in the Bronx. Here is a dollar store, which is essentially like a low-end grocery or convenience store. Perhaps it is not surprising to find sugary drinks and snacks here.

Slide 23 - Example: locksmith

But at a locksmith?! This is a lock, safe, and gate retailer and inside the store you see a vending machines for sugar-sweetened beverages, candy, and a variety of salty refined snack chips. Unexpected.

Slide 24 - Example: laundromat

But perhaps not as unexpected as this. This is a laundromat--a public place to wash clothes. One might expect the sale of detergent and fabric softener here. But you probably don’t expect to find a full sandwich counter as in this example.

Slide 25 - Example: medical office

But perhaps most surprising, horrifying, and sad is this example. This is an actual medical office in the Bronx, and you can see a vending machine for sugar-sweetened beverages right in the waiting room!

Slide 26 - Discussion

To conclude, I think what these findings tell us is that local food environments include businesses well beyond just ‘food stores’ and restaurants; ‘other businesses’ offer foods and drinks very often too (and failing to consider these food sources really calls into question the findings, conclusions, and implications of much of the existing literature on food access, food environments, and ‘food deserts’).

Additionally--as is the case with most issues related to health--unhealthful food provision was not evenly distributed; many more businesses offered foods and drinks in the Bronx (lower-income, minority communities) than in the UES (affluent white community) and items tended to be less-healthful with fewer healthier options. So rather than continuing to treat illness and then send them back to the same conditions that made them sick in the first place, maybe these findings can increase awareness and direct activity to make some changes. Physicians, patients, and public health advocates might work with local businesses that offer foods and beverages to encourage their offering more-healthful options. Non-perishable shelf-stable items might have particular potential here (e.g., nuts, seeds, dried fruits, and whole-grain products like granolas, chips, or crackers)

Slide 27 - Acknowledgments

These are the people and institutions I’d like to thank, in particular the amazing students who collected the data and my cartographer and spatial analyst Andrew Maroko. I’d also like to thank Montefiore, Einstein, Lehman College, and NIH/NICHD for funding this work.