Table 1.
Summary of community-based fruit and vegetable prescription programs
| First author (year), country | Study design | Main aim/objective | Sample size | Prescription | Prescription provider | Produce provider | Education provided | Study population description | Outcomes measured | Results |
|---|---|---|---|---|---|---|---|---|---|---|
| Aiyer(26) (2019), USA | Pre–post mixed methods cohort study | Demonstrate feasibility, acceptability, cost, and impact of food prescription program | 242 | F&V up to 30 lbs in weight provided for fortnightly collection for 6 months | Community health centre | Food bank /pantry | Nutrition education booklets and recipe cards | Adults >18 years, majority female, Hispanic. Mean household 4⋅6. 24⋅2 % in food assistance programs | FI status Redemption rate |
94⋅1 % FI decrease v. baseline (P < 0⋅01) Average redemption 6⋅5 out of possible 12. |
| Basu(27) (2021), USA | Pre–post cohort study | Determine if F&V vouchers improve intake and diet quality in two different cities | 671 | 4 × $5 vouchers delivered monthly via mail over a 6-month period | Federal/State funding | Grocery stores and farmers’ markets | Adults >21 years, majority black, female. Median income below FPL | F&V intake | F&V intake +0⋅22 daily cup eq. v. baseline (P < 0⋅001). | |
| Black(28) (2013), Australiaa | Retrospective cohort study | Investigate short-term health impacts of F&V subsidy on child health in disadvantaged Aboriginal families | 143 | Weekly F&V boxes or vouchers, 1–4 children: $40, 5+: $60. Supplied over 12 months | Community health centre | Grocery stores | Cooking/nutrition education sessions led by nutrition professionals | Children < 17 years in low-income, Aboriginal families | Sick visits Antibiotic use Hb/iron status Well visits BMI |
Decrease in sick visits −0⋅6/year (P < 0⋅05), Decrease in oral antibiotics −0⋅5 prescriptions/year (P < 0⋅05) Increase Hb +3⋅1 g/l (P < 0⋅05) No significant change in frequency of preventative visits or BMI. |
| Black(29) (2013), Australiaa | Retrospective cohort study | Evaluate if participation in F&V subsidy program improves nutritional biomarkers and dietary behaviour | 115 | Weekly F&V boxes or vouchers, 1–4 children: $40, 5+: $60. Supplied over 12 months | Community health centre | Grocery stores | Cooking/nutrition education sessions led by nutrition professionals | Children < 17 years in low-income, Aboriginal families | Nutritional biomarkers F&V intake |
Significant increases (P < 05) in b-cryptoxanthin (28⋅9 nmol/l, +18 %), vitamin C (10⋅1 mmol/l, +21 %) and lutein–zeaxanthin (39⋅3 nmol/l, +11 %) No significant increases in F&V consumption. |
| Bowling(30) (2016), USA | Pre–post cohort study | Assess efficacy of healthy food intervention in families enrolled on government assistance program | 186 | Attendance reward: 40 % rebate on FM spend. Plus $20 bonus for every third visit up to $120 per family | NGO, non-profit, or charity | Farmers’ market | Cooking demos, nutrition education, recipe cards, and handouts | Families enrolled in food assistance program with at least 1 child <12 years | F&V intake Soda intake |
Increased vegetable consumption +0⋅28 times/day (P = 0⋅005) Reduced soda consumption −0⋅14 times/day (P = 0⋅005). |
| Briefel(31) (2020), USAb | Cluster randomised controlled trial | Investigate impact of food box and F&V voucher delivery program on child food insecurity | 2859 | Monthly food box delivery including $15 F&V voucher over 25 months | Federal/State funding | Grocery stores | Nutrition education handouts | Families with children > 4 years eligible for free school meals | FI status | No significant improvement in child FI v. control at 12- or 18-month follow-ups. Adult FI significantly reduced (P = 0⋅002) at 12 months (−2⋅8 % v. control), not sustained at 18 months. |
| Bryce(32) (2017), USA | Pre–post cohort study | Examine impact of F&V prescription program on patients with uncontrolled T2D | 65 | $10 per week for F&V purchases at FM over 13 weeks. Plus $5 participation incentive | Community health centre | Farmers’ market | Cooking demos | Adults >18 years with T2D or HbA1c >6⋅5 %. Majority low-income Hispanic, female participants | HbA1c Weight BP |
Decrease in HbA1c concentration 9⋅54 to 8⋅83 % (P < 0⋅001) No significant changes in BP and weight. |
| Bryce(33) (2021), USA | Randomised controlled trial | Assess effectiveness of F&V prescription program on diabetic patients | 112 | $10 per week to purchase F&V at FM, up to 8 visits over 15 weeks | Community health centre | Farmers’ market | Cooking demos, nutrition education | Adults >18 years with T2D and HbA1c >8⋅0 %. Majority low-income Hispanic, female participants | HbA1c BP, BMI |
Decrease in HbA1c concentration 9⋅69 to 9⋅15 % for intervention group (P = 0⋅006) No statistically significant changes in BP or BMI for either group. |
| Burrington(34) (2020), USA | Pre–post cohort pilot study | Examine impact of F&V prescription and education program on rural families | 10 | $15 weekly online credit for family of 3, $20 for 4, $25 for 5+. F&V delivered for local collection over a 5-month period | School | Community organisation | Cooking/nutrition education sessions led by nutrition professionals | Low-income families with at least one child at risk from chronic disease associated with obesity | F&V intake FI status |
Most participants reported consuming more F&Vs Reduction in FI before v. after No statistical analysis recorded. |
| Buscail(35) (2018), Francec | Randomised controlled trial | Determine whether F&V vouchers modify consumption of F&V in children from low-income households | 64 | Monthly vouchers (approx. $7 per person) via post to participating families. F2F questionnaires at baseline, 6 months and 1 year | French Ministry of Health | Grocery stores and farmers’ markets | Nutrition education sessions | Families with household income below poverty line | F&V intake | At 1-year child, F&V consumption higher in intervention v. control: 4⋅0 (95 % CI 1⋅4, 6⋅0) servings v. 2⋅2 (95 % CI 0⋅9, 5⋅0) P < 0⋅001. Adults 3⋅0 (95 % CI 0⋅5, 7⋅0) servings v. 1⋅9 (95 % CI 0⋅5, 7⋅0) P = 0⋅02. Proportion of children defined as low F&V consumers 29⋅4 % (95 % CI 14⋅1, 44⋅7) intervention v. control 66⋅7 % (95 % CI 49⋅9, 83⋅5) P = 0⋅005. |
| Buscail(36) (2019), Francec | Randomised controlled trial | Determine whether F&V vouchers improve food insecurity over a 1-year period | 64 | Monthly vouchers (approx. $7 per person) via post to participating families. F2F questionnaires at baseline, 6 months and 1 year | French Ministry of Health | Grocery stores and farmers’ markets | Nutrition education sessions | Families with household income below poverty line | FI status | Food security in intervention group significantly improved: baseline FI 85⋅3 % (29) v. 61⋅8 % (21) after 1 year P = 0⋅03. No difference in the control group. |
| Cabili(37) (2020), USAb | Cluster randomised controlled trial | Investigate whether food box with F&V voucher improves diet quality in children of low-income households | 2859 | Monthly food box delivery including $15 F&V voucher over 25 months | Federal/State funding | Grocery stores | Nutrition education handouts | Families with children > 4 years eligible for free school meals | F&V intake | F&V (daily cup eq.) and whole grain (oz eq.) intake significantly higher in treatment v. control group: F&V combined 2⋅35 v. 2⋅25; fruits 1⋅25 v. 1⋅20; vegetables 1⋅05 v. 1⋅0; whole grains 0⋅73 v. 0⋅67 (P < 0⋅001). |
| Cavanagh(38) (2017), USA | Retrospective cohort study | Determine if F&V prescription program is effective in reducing BMI | 108 | Treatment group received 13 × $7 weekly vouchers for redemption at mobile F&V market | Community health centre | Mobile produce market | Nutrition education sessions led by nutrition professionals | Low-income adults with diagnosis of obesity, hypertension, and/or diabetes | BMI | BMI significantly different pre- v. post-intervention in both treatment −0⋅74 kg/m2 (2⋅72) and control 0⋅35 kg/m2 (1⋅91) groups (P < 0⋅0001). Difference between groups also significant P = 0⋅02. |
| Ferdinand(39) (2017), USA | Cross-sectional survey | Assess effectiveness of incentive program on increasing F&V purchasing | 96 | 6 × $4 coupons per week provided to purchase locally grown F&V over a 6-month period | NGO, non-profit, or charity | Farmers’ markets | Adults >18 years receiving SNAP, FMNP, WIC, and/or Medicaid benefits. Majority female, African American participants | F&V purchase F&V intake |
63 % of all participants purchased more F&V after the program, 66 % increased variety. SNAP participants increased quantity by 70 % and variety by 63 %. 89 % of participants reported positive changes in consumption of F&V. No pre-intervention consumption data were reported. |
|
| Fertig(40) (2021), USA | Pre–post cohort study | Determine the effectiveness of F&V incentive and education program for low-income families | 120 | 3 intervention groups receiving F&V vouchers of $10, $15, or $20 weekly. Participants also enrolled on education program | NGO, non-profit, or charity | Grocery store | Grocery store tour, recipes, and cooking education sessions | Food pantry clients >18 years, majority female participants | F&V intake Attitudes to healthy eating |
Pre v. post fruit consumption increased by 0⋅79 cups (P < 0⋅001). No changes in vegetable consumption. No significant change in attitude towards assembling a healthy meal. |
| Freedman(41) (2013), USA | Repeated measures cohort study | Evaluate whether financial incentives are effective at increasing F&V consumption among low-income, rural, diabetics | 41 | $25 F&V vouchers given at baseline (T1) and midpoint (T2) of 22 weeks intervention | Community health centre | Farmers’ market | Adults >18 years with diabetes, majority female, African American participants from low-income households | F&V intake | Marginally significant increase T1 to T2 in mean F&V servings; 5⋅9 to 7⋅5 (P = 0⋅07). Participants categorised as increasers (≥0⋅5 servings/day at T2/3 v. T1) and non-increasers. Odds of being an increaser higher for those using only vouchers as payment (OR 38⋅8, 95 % CI 3⋅35, 445⋅0) and visiting FM more often (OR 2⋅07, 95 % CI 1⋅09, 3⋅95). | |
| Jones(42) (2020), USA | Pre–post cohort study | Investigate the impact of paediatric F&V prescription program | 122 | Families given weekly F&V vouchers worth $1 per person per day (up to max $5 per day) over a 6-month period | Community health centres | Grocery stores and farmers’ markets | Nutrition education sessions | Children ≤ 6 years in Navajo Nation with low or very low FS | F&V intake FI status BMI |
Mean (sd) F&V consumption increased from 5⋅2 (2⋅1) to 6⋅8 (2⋅2) servings/day (P < 0⋅001). Household FI decreased from 82 to 65 % (P < 0⋅001). Of children classified as overweight or obese at baseline (n 58), 38 % achieved healthy BMI z-score at end (P < 0⋅05), their mean BMI percentile decreasing from 95⋅6 (4⋅1) to 73⋅1 (28⋅7), (P < 0⋅001). |
| Kerr(43) (2020), USA | Prospective cohort study | Investigate whether F&V prescription improves cardiometabolic outcomes in adults with or at risk of T2D | 47 | 10 × weekly F&V prescriptions enough for 21 servings non-starchy veg for collection from community site | Community health centres | Wholesale direct from farms | Adults with or at risk of T2D | F&V intake WC BP Weight HbA1c Glycaemic control FI |
Proportion consuming ≥1 serving/day of vegetables increased 15 to 50 % pre v. post (P < 0⋅0001). WC decreased −0⋅77 cm (95 % CI −1⋅42, 0⋅12, P = 0⋅022). SBP decreased −2⋅42 mm Hg (95 % CI −4⋅56, 0⋅28, P = 0⋅037). Weight decreased −0⋅4 kg (−0⋅7 to –0⋅04, P = 0⋅029) in women. In participants with HbA1c > 7⋅0 %: HbA1c decreased −0⋅35 % (−0⋅8 to –0⋅1, P = 0⋅009) Participants with CGM data (n 40), time in range 70–180 mg/dl improved (97⋅4 to 98⋅9 %, P < 0⋅01) Participants with low or very low FS fell from 35 to 13 % (P < 0⋅001). |
|
| LaBarba(44) (2019), USA | Pre–post cohort study | Determine impact of F&V prescription program on dietary quality in low-income expectant mothers | 25 | Monthly $40 F&V vouchers on attending obstetric appointment, from first trimester to 6 weeks postpartum | Community health centres | Grocery store | Grocery store tour, nutrition education sessions | Low-income, pregnant women. Majority Hispanic participants | F&V intake | F&V intake increased baseline v. T1 (pre-delivery); fruit 2⋅0 to 2⋅61 daily servings (P < 0⋅05); vegetable 1⋅44 to 2⋅39 (P < 0⋅001). Baseline v. T2 (6 weeks postpartum); fruit increased 2⋅0 to 2⋅5 daily servings (P < 0⋅05) vegetable 1⋅44 to 2⋅28 (P < 0⋅01). |
| Orsega-Smith(45) (2020), USA | Pre–post cohort study | Evaluate F&V prescription program offered via paediatrician to low-income families | 41 | Monthly F&V boxes collected from doctors’ office containing 15–25 lb F&V per box | Community health centres | Food bank /pantry | Cooking and nutrition education sessions | Low-income families with ≥2 children or overweight adult. Majority white participants | F&V intake FI status |
Increases in F&V intake (servings/day) pre v. post: Adult vegetables 2⋅22 ± 1⋅24 to 2⋅44 ± 1⋅03 (P < 0⋅001), fruit 2⋅05 ± 0⋅97 to 2⋅46 ± 0⋅92 (P < 0⋅05). Children; fruit 2⋅51 ± 1⋅21 to 2⋅77 ± 1⋅16 (P < 0⋅05). No significant difference in vegetable consumption. Perceptions of FI improved |
| Palar(46) (2017), USA | Prospective cohort study | Assess impact of food intervention program to improve nutrition, mental health, and health behaviours | 52 | Healthy meals and snacks fulfilling 100 % daily caloric requirements for collection 2× per week over a 6-month period. | NGO, non-profit, or charity | Food bank/pantry | Adults >18 years with HIV and/or T2D, majority male participants | Diet quality FI status ART adherence |
Frequency of consumption of fats decreased (P = 0⋅003), frequency consumption F&V increased (P = 0⋅011). Among people with diabetes, frequency of sugar consumption decreased (P = 0⋅006) VLFS decreased 59⋅6 to 11⋅5 % (P < 0⋅0001) ART adherence ≥95 % increased 46⋅7 to 70⋅0 % (P = 0⋅046). |
|
| Richie(47) (2019), USA | Pre-experimental | Measure impact of FM voucher program on health outcomes in low-income population | 308 | F&V vouchers redeemable at FM $1 per day per person in HH, or $2 for study participation. Over a 6-month period | Community health centres | Farmers’ market | Low-income adults >18 years with diagnosis of Diabetes, hypertension and/or obesity | FBG WC, BMI, weight, BP |
Participants’ mean (sd) FBG decreased pre v. post −8⋅92 (5⋅36) mg/dl (P = 0⋅023) WC decreased −0⋅26 (0⋅17) inches (P = 0⋅001). No significant changes in BP, BMI, or weight. |
|
| Ridberg(48) (2018), USA | Pre–post retrospective cohort | Assess food security status before and after paediatric F&V prescription program | 578 | Families received $0⋅50 to $1⋅00 vouchers per person per day redeemable at local FM over 4–6 months | NGO, non-profit, or charity | Farmers’ market | Nutrition education sessions | Low-income families with children aged 2–18 years, at least one child to be classified as obese or overweight | FI status | Mean FS score increased 0⋅72–0⋅81 (P < 0⋅001). High or marginal FS status increased 58–76 % (P < 0⋅001), low FS decreased by 33–22 % (P < 0⋅001), and very low FS decreased by 9–1 % (P < 0⋅001). Families attending 5–6 visits had +0⋅7 FS change score v. those attending 1–2 visits (95 % CI 0⋅01, 0⋅14, P < 0⋅05). |
| Ridberg(49) (2019), USA | Pre–post retrospective cohort | Assess F&V intake before and after paediatric F&V prescription program | 883 | Families received $0⋅50 to $1⋅00 vouchers per person per day to be redeemed at local FM over 4–6 months | NGO, non-profit, or charity | Farmers’ market | Nutrition education sessions | Low-income families with children aged 2–18 years, at least one child to be classified as obese or overweight | F&V intake | Mean (sd) fruit intake pre v. post; 1⋅6–1⋅7 cups (0⋅13 (1⋅2), 95 % CI 0⋅05, 0⋅21, P < 0⋅001); vegetable intake 1⋅2–1⋅3 cups (0⋅13 (1⋅1) 95 % CI 0⋅06, 0⋅21, P < 0⋅001). Combined mean F&V intake increased 0⋅26 cups pre v. post (95 % CI 0⋅13, 0⋅39, P < 0⋅001). Pre v. post meeting recommendations; 78 % v. 86 %. +0⋅32 cups for each additional clinic visit (95 % CI 0⋅20, 0⋅45; P < 0⋅001). |
| Saxe-Custack(50) (2019), USA | Pre–post longitudinal cohort | Determine impact of F&V prescription program on fruit consumption in children | 114 | $15 F&V voucher prescribed at paediatrician visits, to be redeemed at FM | Community health centre | Farmers’ market | Children aged 7–18 years | F&V intake FI status |
Mean (sd) daily servings whole fruit increased baseline v. 6 months 0⋅62 (0⋅69), to 0⋅81 (0⋅64), P = 0⋅029. No statistically significant changes in total fruit, fruit juice, or vegetable consumption No significant change in FI status. |
|
| Slagel(51) (2021), USA | Non-randomised control trial | Assess impact of pilot F&V prescription program on low-income adults with chronic health conditions | 24 | Monthly F&V vouchers of $1 per day per household member for 6 months | NGO, non-profit, or charity | Farmers’ market | Cooking and nutrition education sessions | Adults with overweight/obesity, diabetes, prediabetes, hypertension, and/or hyperlipidaemia | F&V intake | Mean (sd) Pre–post F&V intake higher in Intervention v. control; 0⋅81 (0⋅91) servings/day v. −0⋅25 (0⋅99) P = 0⋅02. Increases intervention group pre v. post total vegetable intake, dark green vegetable intake, and HEI score (P = 0⋅005). Increased knowledge of F&V preparation (P = 0⋅02), increase in F&V purchases (P = 0⋅05). |
| Snailer(52) (2019), USA | Pre–post cohort study | Determine whether F&V prescription program lowers HBA1c levels in food insecure participants with T2D | 14 | Monthly F&V vouchers worth $1 per day per HH member for redemption at FM over 6-month program | Community health centre | Farmers’ market | Cooking and nutrition education sessions | Food insecure adults with T2D and HbA1c > 7⋅0 %. Majority white, female participants | HbA1c | Average decrease in HbA1c 1⋅85 % (95 % CI −2⋅69, −1⋅01) P = 0⋅0004. Linear model showed a $10 increase in average monthly redemptions associated with 1⋅4 % (95 % CI 0⋅5, 2⋅4) decrease in HbA1c P = 0⋅006. |
| Trapl(53) (2018), USA | Pre–post cohort study | Evaluate F&V redemption and consumption among food insecure adults with HTN | 224 | Weekly $10 F&V vouchers for 3 months; up to $120, redeemable at local FM | Federal/State funding | Farmers’ market | Nutrition education sessions | Food insecure and hypertensive adults. Majority black, female participants | F&V intake | Mean servings (sd) fruit increased pre v. post 1⋅6 (1⋅3) to 2⋅4 (1⋅2), P < 0⋅001. Vegetables increased 1⋅7 (1⋅1) to 2⋅5 (1⋅3), P < 0⋅001. Fast-food servings decreased 1⋅3 (1⋅4) to 0⋅7 (1⋅0), P < 0⋅001. |
| Xie(54) (2021), USA | Prospective cohort study | Assess produce prescription program utilisation, healthy food purchases, and health outcomes | 699 | $40 monthly added to SNAP EBT card or store card for WIC approved F&V over a 1-year period | NGO, non-profit, or charity | Grocery stores | SNAP eligible adults aged ≥18 years | F&V intake (purchasing data as proxy) HbA1c, BMI, BP |
Frequent spenders v. sometimes spenders: higher F&V spending (B=$8⋅77, P < 0⋅001), higher F&V expenditure share (B = 3⋅3 %, P = 0⋅007), and more unique F&V purchases (B = 2⋅52, P < 0⋅001). No significant relationships between program utilisation and health outcomes. |
|
| York(55) (2021), USA | Prospective cohort study | Assess feasibility and impact of farming for life pilot program among Latino adults with T2D | 21 | Weekly F&V box, value $31 for collection by participants over a 12-week period. | NGO, non-profit, or charity | Direct from local organic farms | Latino adults with T2D, majority female participants | BP FI status HbA1c, BMI, WC |
Reduction in systolic BP (P = 0⋅03) and diastolic BP (P = 0⋅01) FI improved in 57 % of participants. No statistically significant changes in weight, WC or HbA1c. |
F&V, fruit and vegetables; FI, food insecurity; FPL, federal poverty level; Hb, haemoglobin; BMI, body mass index; FM, farmers’ market; NGO, non-governmental organisation; T2D, type 2 diabetes mellitus; HbA1c, glycated haemoglobin; BP, blood pressure; F2F, face to face; SNAP, Supplemental Nutrition Assistance Program; FMNP, farmers’ market nutrition program; WIC, women, infants, children special supplemental nutrition program; FS, food security; WC, waist circumference; CGM, continuous glucose monitor; HIV, human immunodeficiency virus; VLFS, very low food security; ART, antiretroviral therapy; HH, households; FBG, fasting blood glucose; HTN, hypertension; EBT, electronic benefits transfer.
abcSuperscript letters denote studies carried out on the same intervention.