Do people who shop at farmers markets eat more fruit and vegetables?

A farmers market could change what you eat in three ways: it puts fresh produce within reach, it can stretch food-assistance benefits, and a weekly trip may build a routine. Researchers have looked for the effect in surveys, in before-and-after studies and in trials.

Surveys of shoppers. Among 400 low-income women in eastern North Carolina, 42% of those who had ever shopped at a farmers market ate five or more servings of fruit and vegetables a day, against 24% of those who had not⁵. Among 645 market shoppers in New York City and rural North Carolina, those who bought produce at the market more often had higher self-reported intake and higher skin carotenoids, a non-invasive marker of produce intake².

Why surveys can mislead. Shoppers are not a random slice of the population. In rural North Carolina and Kentucky, more than half of market customers had a college degree, against about a third of people reached by random phone calls, and fewer used SNAP⁶. In a Utah matching program, SNAP recipients who went on to use the vouchers already ate more produce before they started⁷. People who seek out markets may already eat well.

Trials of price. The controlled evidence is mostly about money. At one Los Angeles WIC clinic, new mothers given weekly farmers market vouchers worth $10 for six months ate 1.4 more servings of fruit and vegetables per 1,000 calories than mothers at a comparison clinic, and the difference was still there six months after the vouchers stopped⁸. In a randomized trial of 30% rebates on produce bought with SNAP benefits, adults ate about a third of a cup more fruit and vegetables a day⁹. A randomized trial at 77 markets found that a $2 match for each $1 of SNAP raised intake by 0.16 cup a day, or 0.31 cup among shoppers who used it¹.

What did not work alone. A web lesson for 744 women in WIC improved their knowledge of markets but not their produce intake¹⁰.

What it means. Where money or access is the barrier, matching programs raise intake modestly. For everyone else, a market is a pleasant place to buy produce, with no proof yet that it changes how much you eat.

Surveys: do shoppers eat more?

A survey of 400 low-income women at a health department clinic in eastern North Carolina found that 114 had ever shopped at a farmers market; 42.1% of them ate five or more servings of fruit and vegetables a day, against 24.0% of the others (P < 0.001), and the pattern held among non-students⁵. A second study interviewed customers at markets in North Carolina (70) and Kentucky (102) and phoned random samples of household food shoppers (109 and 149). In the Kentucky customers and both phone samples, more frequent market use went with higher intake: in the phone samples, people who visited two or three times a month or more ate 1.0 and 1.3 more servings a day than those who went once a month or less. Body mass index was unrelated. But the customers reported fewer servings (4.3 and 3.7 a day) than the phone samples (7.2 and 7.3), and the two kinds of sample were recruited and weighted differently, so the gap may reflect method as much as behavior⁶.

Kelley and colleagues (2022) interviewed 645 shoppers at 17 markets in New York City and rural North Carolina and measured skin carotenoids. Buying produce at a market more often went with higher self-reported intake and higher skin carotenoids; spending more and buying more varieties went with higher self-reported intake only. Everyone in the sample was a market shopper, so this is a dose-response among shoppers, not a comparison with non-shoppers².

Why shoppers are not a fair comparison

In Utah, 212 SNAP recipients were told about a market matching program (Double Up Food Bucks) and followed for about five months. Only 34 used it. Users ate more fruit and vegetables than non-users at the start and midpoint (P = 0.02 each), and intake fell over the season overall; the authors note that income and distance were not accounted for⁷. In the rural survey above, 63% and 56% of market customers were college graduates, against 35% in both phone samples⁶. A systematic review of 49 articles found that market use is shaped by economic, service, spatial, social and personal factors, that only 39% included low-income shoppers, that fewer than 15% focused on racial or ethnic minorities, and that designs varied too much to compare results³.

Trials and before-and-after studies of price

Herman and colleagues (2008) did not randomize: 602 women enrolling for postpartum WIC services were assigned by clinic to $10 a week in vouchers at a farmers market, the same at a supermarket, or a minimal non-food incentive. After six months, intake was 1.4 servings per 1,000 calories higher at the market clinic and 0.8 higher at the supermarket clinic than at the comparison clinic, and the market gain was still significant six months later. The clinics differed at baseline and about a quarter of the women were lost to follow-up⁸.

Olsho and colleagues (2016) randomly assigned 7,500 SNAP households in Hampden County, Massachusetts, to a 30% rebate on targeted fruit and vegetables bought with SNAP benefits; 47,595 others kept usual benefits. Among 2,009 adults analyzed, intake of targeted produce was 0.24 cup a day higher (95% CI 0.13 to 0.34) and all fruit and vegetables 0.32 cup higher (0.17 to 0.48), the extra over targeted produce explained by 100% fruit juice, which earned no rebate. The rebate was not limited to farmers markets⁹.

Karpyn and colleagues (2022) randomized SNAP shoppers at 77 markets in 13 states and Washington, DC, to no extra incentive, a moderate one or the highest one. A $0.40 match led to average spending of $19.03 per transaction and a 2:1 match to $36.28. At 2:1, intake rose 0.16 cup a day across everyone assigned that level and 0.31 cup among those who used it, by self-report¹.

In a before-and-after study of Utah SNAP shoppers, median intake rose from 2.82 to 3.29 times a day (P = 0.002), but only 138 of 339 (40%) completed follow-up and there was no comparison group¹¹.

What a lesson alone did not do

In a four-arm randomized trial of 744 women in a New Jersey WIC agency, a web lesson on market shopping improved knowledge of the voucher program and of seasonal produce, but fruit and vegetable intake did not differ between arms¹⁰.

Food safety at markets

A knowledge synthesis of 83 studies found no difference in contamination of fresh produce from markets versus other sources, but limited and mixed evidence of more Campylobacter and Salmonella on market chicken. Vendor handwashing averaged 4% across five studies, and the outbreaks and case reports found for 1994 to 2016 caused 411 illnesses and two deaths⁴. Unpasteurized milk is a separate question; see our raw milk page.

Where the evidence is thin

A review of 37 articles (34 studies) on local food systems found that farmers markets were the best studied, produce intake the most measured outcome, and long-term health effects rarely measured¹². Most intake data are self-reported, most studies follow shoppers for a season or less, and we found no trial long enough to test body weight or disease.

Bottom line

Moderate confidence (3 of 5). Market shoppers report higher produce intake, but they are a self-selected group, and the strongest tests, price incentives, show gains of roughly 0.2 to 0.3 cup a day. For related questions, see fresh vs frozen produce and organic vs conventional.