Are Food Deserts a Mirage?

Related articles

Food deserts are areas frequently in urban settings where it is difficult to find stores providing fresh foods, especially fruits and vegetables. Food deserts have long been thought to contribute to poor nutrition because the food people need is just not available. A study in JAMA Pediatrics suggests that this long-held thought may be a mirage.


“Although supermarkets stock a variety of inexpensive unhealthy foods, previous research suggests that expenditures on and consumption of fruits and vegetables are higher among individuals primarily shopping in supermarkets, potentially due to higher availability and lower price of fresh produce relative to other store types.”

Rising levels of obesity remain problematic, and this is especially true among pediatric populations, our children. Whether the presence or absence of a supermarket makes a difference has yielded mixed results, but New York City developed the Food Re-

tail Expansion to Support Health (FRESH) program, designed to lower “the costs of owning, leasing, and renovating supermarket retail space via tax and zoning incentives.” [1] The JAMA Pediatrics article provides some follow-up health outcomes on the $140 million provided by taxpayers.


All NYC public school pupils living within a half-mile of a FRESH-subsidized supermarket and having a BMI measurement before and after the supermarket's opening were the treatment group; those living outside that range acted as the controls. Of the roughly 2 million public school students between 2009 and 2016, approximately 11,000 met the requirements for the treatment group, and 43,000 served as controls. [2] The students were predominantly Hispanic, Latino, or Black, and 85% were eligible for free or reduced-price lunches. The treatment group was slightly more obese, 24.3%, than the control group at 23.3%. Obesity was defined as being in the 95th percentile or higher for BMI for their age and gender. The mean age was between 10 and 11.

Over a 3 to 12-month period, there was no change in the percentage of obesity within the control group. There was a 1% decrease in obesity in the treatment group, moving from 24.3% to 23.3%. BMI also showed a more significant reduction in the treatment group than control, but it was minimal. The most significant changes were in our youngest children, K through grade 5, with essentially no difference among teens (grade 9-12). The ‘tweens were, in fact, in between these two groups.

Hope springs eternal

Results from this quasi-experimental study suggest that the establishment and renovation of government-subsidized supermarkets vis à vis zoning and financial incentives may contribute to a small decrease in childhood obesity risk.”

I applaud the researchers for publishing the data, but why is it so difficult to say there was no significant clinical impact? They do mention that experts believe that reductions in outcomes 3 to 5 larger than they found were felt to be clinically meaningful. They note that studies in other cities have shown no change in obesity with the introduction of new supermarkets and that fresh fruits and vegetables may cost more irrespective of government-subsidized supermarket construction.  

“…changes in infrastructure, in turn, may result in the presence of healthy, albeit unaffordable, food options, which may undermine the program’s success. Therefore, future supermarket interventions might be more successful if paired with financial incentive programs, such as subsidies for produce.”

As we have seen in other health concerns, the availability of a product or service is contingent on its affordance. The editorial accompanying the article makes a similar point and notes that the effect size of new supermarkets on obesity is essentially a rounding error.  I will leave the last words to the editorial.

“In summary, our reading of the results of this study is that they are far more consistent with trivial to zero effects than promising “effects.” Given that, where should we go from here? We may be reminded of the streetlight effect, which is an old tale where one individual helps another search for their lost keys under a streetlight (“where the light is”), despite losing the keys elsewhere (where it is dark). This story suggests that when seeking truth, we often look for truth in the easiest or most convenient places rather than where the truth may actually be. We should thus begin considering more novel factors that might provide more effective levers for change, however less obvious than the community and public health ideas considered to date, the factors that may explain these relationships: taking special care to focus on areas in the dark than continuing our futile examination under the streetlight.”


[1] Supermarkets must provide “at least (1) 5000 sq ft of retail space for grocery products, (2) 50% of retail space for food products, (3) 30% of retail space for perishable goods, and (4) 500 sq ft for fresh produce21 and must be located in an eligible area.”

[2] Ineligible students recorded no BMI values, lived outside the FRESH areas, or moved during the study period.


Source: Association Between a Policy to Subsidize Supermarkets in Underserved Areas and Childhood Obesity Risk JAMA Pediatrics DOI: 10.1001/jamapediatrics.2022.1153

Taking a Hard Look at the Empirical Evidence for Popular Community-Based Interventions in Obesity JAMA Pediatrics DOI:10.1001/jamapediatrics.2022.115