Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
PREVENTING CHRONIC DISEASEP U B L I C H E A L T H R E S E A R C H , P R A C T I C E , A N D P O L I C Y Volume 14, E107 NOVEMBER 2017
SYSTEMATIC REVIEW
Pricing Strategies to EncourageAvailability, Purchase, and Consumption of
Healthy Foods and Beverages: ASystematic Review
Joel Gittelsohn, PhD1; Angela Cristina Bizzotto Trude, MS1; Hyunju Kim, MPH1
Accessible Version: www.cdc.gov/pcd/issues/2017/17_0213.htm
Suggested citation for this article: Gittelsohn J, Trude ACB,Kim H. Pricing Strategies to Encourage Availability, Purchase,and Consumption of Healthy Foods and Beverages: A SystematicReview. Prev Chronic Dis 2017;14:170213. DOI: https://doi.org/10.5888/pcd14.170213.
PEER REVIEWED
Abstract
IntroductionFood pricing policies to promote healthy diets, such as taxes, pricemanipulations, and food subsidies, have been tested in differentsettings. However, little consensus exists about the effect of thesepolicies on the availability of healthy and unhealthy foods, onwhat foods consumers buy, or on the impact of food purchases onconsumer health outcomes. We conducted a systematic review ofstudies of the effect of food-pricing interventions on retail salesand on consumer purchasing and consumption of healthy foodsand beverages.
MethodsWe used MEDLINE, Embase, PsycINFO, Web of Science, Clinic-alTrials.gov, and the Cochrane Library to conduct a systematicsearch for peer-reviewed articles related to studies of food pricingpolicies. We selected articles that were published in English fromJanuary 2000 through December 2016 on the following types ofstudies: 1) real-world experimental studies (randomized con-trolled trials, quasi-experimental studies, and natural experiments);2) population studies of people or retail stores in middle-incomeand high-income countries; 3) pricing interventions alone or incombination with other strategies (price promotions, coupons,taxes, or cash-back rebates), excluding studies of vending-ma-
chine or online sales; and 4) outcomes studies at the retail (stock-ing, sales) and consumer (purchasing, consumption) levels. We se-lected 65 articles representing 30 studies for review.
ResultsSixteen pricing intervention studies that sought to improve accessto healthy food and beverage options reported increased stockingand sales of promoted food items. Most studies (n = 23) reportedimprovement in the purchasing and consumption of healthy foodsor beverages or decreased purchasing and consumption of un-healthy foods or beverages. Most studies assessed promotions offresh fruits and vegetables (n = 20); however, these foods may behard to source, have high perishability, and raise concerns aboutsafety and handling. Few of the pricing studies we reviewed dis-couraged purchasing and consumption of unhealthy foods (n = 6).Many studies we reviewed had limitations, including lack of form-ative research, process evaluation, or psychosocial and health as-sessments of the intervention’s impact; short intervention duration;or no assessment of food substitutions or the effects of pricing in-terventions on food purchasing and diets.
ConclusionPricing interventions generally increased stocking, sales, purchas-ing, and consumption of promoted foods and beverages. Addition-al studies are needed to differentiate the potential impact of selec-ted pricing strategies and policies over others.
IntroductionPricing strategies to encourage the availability, purchasing, andconsumption of healthy foods and beverages have received in-creased attention in the past decade, in the United States andworldwide. Various pricing strategies have been studied in differ-ent settings, including taxes and price manipulations of sugar-sweetened beverages (SSBs), high calorie–low nutrient foods orfoods high in added sugars or saturated fats, and subsidies of fruits
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health
and Human Services, the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
www.cdc.gov/pcd/issues/2017/17_0213.htm • Centers for Disease Control and Prevention 1
and vegetables. Despite these studies, little consensus exists aboutthe effectiveness of these pricing strategies in changing the avail-ability and affordability of healthy and unhealthy foods or their ef-fect on consumer outcomes (ie, foods purchased, foods consumed,and health). Furthermore, little consensus exists about how pri-cing strategies function, alone or combined with health behaviorinterventions or as part of multi-level interventions.
Reviews were conducted previously on related topics. Nine recentreviews (from 2010 through 2015) examined the effect of taxes,subsidies, or their pooled effect on food consumption, consumerpurchases, body weight, or diet-related chronic diseases (1–9).However, many of these reviews described laboratory-based orsimulation studies (6–8). Only one systematic review describedfield intervention studies (9) and focused on subsidies to increasepurchasing of healthy foods. Few focused on implementation andoutcomes of pricing interventions at both the supply (retail) anddemand (consumer) levels in actual communities.
Decision makers need a systematic review of the effectiveness ofpricing incentives and disincentive strategies on availability, pur-chasing, and consumption of healthy and unhealthy foods andbeverages at the consumer and retail levels. Therefore, our goalwas to answer the following questions: 1) How do pricing incent-ives and disincentives influence access, purchasing, and consump-tion of healthy and unhealthy foods and beverages among variouspopulations in high-income and middle-income countries? 2)What additional work is needed to enable communities, states, andcountries to identify the best combination of strategies?
MethodsData sources
We conducted a systematic review of English-language, peer-re-viewed articles describing studies that evaluated the effectivenessof pricing incentives and disincentive strategies on purchasing andconsumption of healthy and unhealthy foods and beverages inhigh-income and middle-income countries in various socioeco-nomic settings. We searched 6 electronic databases — MEDLINE,Embase, PsycINFO, Web of Science, ClinicalTrials.gov, and theCochrane Library — from January 2000 through December 2016for relevant studies.
We developed a search strategy based on medical subject heading(MeSH) terms and based on the text and key words of key articleswe identified a priori (Appendix). We used Boolean operators tocombine keywords and MeSH terms for a focused search. We de-veloped 3 topics based on our research question (incentive/disin-centive, food intake, and food purchasing), and we then includedkey words and MeSH terms representing each term. Search terms
were pricing strategies, incentive, reimbursement, commerce, dis-incentive, reward, taxes, monetary incentive, consumer behavior,marketing, cost savings, food purchasing, food supply, dietary in-take, eating behavior, food intake, food and beverages, and snacks.
Study selection
We selected the following types of human studies published inEnglish in peer-reviewed journals, from 2000 through 2016: 1) ex-perimental studies (randomized controlled trials, quasi-experi-mental studies, and natural experiments, excluding reviews andcross-sectional, qualitative, and simulation models studies); 2)population studies of people or stores in middle-income and high-income countries; 3) studies of pricing interventions conductedalone or in combination with other strategies (price promotions,coupons, taxes, or cash-back rebates), excluding studies of vend-ing-machine or online sales; and 4) outcomes studied at the retail(stocking, sales) or consumer (purchasing, consumption) levels.Additional criteria were that study outcomes were assessed at theretail level (stocking, sales) or consumer level (purchasing, con-sumption) and that the study was not an evaluation of a govern-ment program in schools (eg, a school-based food assistance pro-gram).
Two reviewers (A.C.B.T., H.K.) reviewed abstracts and full art-icles independently to assess eligibility for inclusion. H.K. con-firmed or corrected A.C.B.T.’s data abstractions for completenessand accuracy. We also conducted a reference list search on thestudies we selected for review and identified 5 eligible studies.Lastly, we identified all peer-reviewed publications associatedwith each study and cited only those that contributed to this re-view.
Data extraction
For the synthesis, we employed an adjudication approach. Weused a series of descriptive criteria to characterize each study:project name, target population, model or theory, study goal, foodsand beverages that were the intervention’s focus and its retail ven-ue, sample size, intervention strategies, study design, study dura-tion, formative research, feasibility assessment, process evalu-ation, impact measures and results, sustainability, quality of re-search, study limitations, and study recommendations (AppendixTable 1). Two reviewers (A.C.B.T., H.K.) analyzed each study in-dependently and provided a long and a short response for the de-scriptive criteria. A third reviewer (J.G.) reviewed the descrip-tions and agreed or disagreed to the study’s inclusion. Where therewas disagreement, the third reviewer broke the tie.
We organized data into the following categories: 1) a descriptionof each study; 2) a description of the intervention, pricingstrategies, and the study evaluation; and 3) main results and study
PREVENTING CHRONIC DISEASE VOLUME 14, E107
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY NOVEMBER 2017
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
2 Centers for Disease Control and Prevention • www.cdc.gov/pcd/issues/2017/17_0213.htm
implications. Data were then grouped by type of pricing interven-tion categories: 1) financial discount on healthy foods and bever-ages, 2) redeemable coupons or vouchers for healthy foods andbeverages targeting participants in food assistance programs, 3) re-deemable coupons or vouchers for healthy foods and beveragestargeting consumers not participating in food assistance programs,4) cash rebates, and 5) disincentive strategies for unhealthy foodand beverage purchases (eg, tax, alone or combined with astrategy promoting healthy foods).
ResultsSearches returned 2,076 articles, and 1,677 were screened after ex-cluding duplicates (ie, the same article in different research data-bases) and by refining the searches by year, language, and species.After elimination of 1,625 for not meeting our study criteria, 52were fully assessed for eligibility; 27 were excluded and 5 wereincluded after a reference list search. Thus, 30 distinct studies in63 articles were included in the final analysis (Figure). The num-ber of peer-reviewed publications per study varied from 1 to 7,with a median of 2 per study.
Figure. Selection process, systematic review of pricing strategies to encouragepurchasing and consumption of healthy foods and beverages, 2000–2016.
Description of studies
The 30 studies included in the review were conducted in 9 coun-tries: the United States (n = 17), Australia (n = 2), New Zealand (n= 2), France (n = 2), Canada (n = 1), the United Kingdom (n = 1),South Africa (n = 1), Denmark (n = 1), Belgium (n = 1), Peru (n =1), and Mexico (n = 1) (Table 1). The largest number of studies (n= 8) took place in the northeastern United States (Table 1). Moststudies (n = 18) did not report the use of a theoretical model. Ofthe 12 that did, social cognitive theory was most commonly men-tioned (n = 5), followed by the social ecological model (n = 5).The most common study design (n = 15) was a randomized con-trolled trial, with randomization either at the group or individuallevel.
At the consumer level, sample sizes ranged from 28 individualparticipants (What to Eat for Lunch study) to more than 50,000households (Mexico excise tax study), with a median of 454 indi-viduals sampled. Among the randomized controlled trials, the me-
PREVENTING CHRONIC DISEASE VOLUME 14, E107
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY NOVEMBER 2017
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
www.cdc.gov/pcd/issues/2017/17_0213.htm • Centers for Disease Control and Prevention 3
dian sample size per study arm was 100 participants. The samplesize or unit of randomization in some studies was based onclusters (eg, food stores) and not necessarily on individuals.
Almost all studies (n = 25) examined the impact of a pricing inter-vention alone or in combination with other strategies related to thestocking, sales, purchasing, or consumption of healthy foods. Moststudies targeted low-income, disadvantaged populations (n = 18).Many studies (n = 12) targeted a specific population that wasreached through the venue of the intervention (eg, a worksite,sports gym, school, swimming pool, hospital).
Interventions and strategies studied
Nearly all studies (n = 27) examined interventions that promotedhealthy foods (Table 2). The most common types of foods pro-moted were fruits and vegetables (n = 20), particularly fresh pro-duce, followed by low-sugar beverages (n = 10), and healthily pre-pared entrees and side dishes (n = 8). Only a few studies (n = 6)discouraged unhealthy foods, such as SSBs and foods high in sat-urated fat and sugar, as part of the intervention, generally by rais-ing the prices of these foods.
The types of food sources targeted varied and included grocerystores and supermarkets (n = 7), all retailers in a setting (eg, city,neighborhood) (n = 6), farmers markets (n = 5), worksite cafeteri-as and school cafeterias (n = 5), food delivery services (n = 2),carryout restaurants (n = 1), corner stores (n = 1), and other typesof retailers. The number of the food sources intervened in for eachstudy also varied, from one to many thousands, because somestudies implemented the strategy city-wide (median, 5 food sourcelocations). Pricing interventions also differed between studies.Nine studies emphasized price discounts on healthy foods andbeverages, ranging from 10% to 33%. Four studies providedcoupons or vouchers of $5 to $20 to food assistance recipients. Sixstudies provided coupons or vouchers of $1 to $22 to the generalpopulation. In 5 studies, the pricing intervention was a cash rebate.The amount of the rebate took many different forms, such as astraight percentage off or a price reduction up to a certain limit.Six studies tested price increases on unhealthy foods, half ofwhich included a price reduction on healthier foods. Three of thesestudies were of local or federal taxes, including taxes on SSBs.
Six of the 30 studies sought to change the availability of healthy orunhealthy foods. Only 2 of the studies changed the physical loca-tion of foods as a means of increasing their uptake by consumers.Eleven of the 30 studies implemented labeling to identify healthyversus unhealthy food choices. Most labeling approaches oc-curred in studies (n = 8) centered on the promotion of healthyfoods and beverages. Five of the 30 studies used a policy ap-proach, and 4 studies involved taxes at the city or national level.
Evaluation strategies
Most studies (n = 17) reported no formative research (AppendixTable 2). When formative research was conducted, it consisted ofqualitative information gathering (n = 3), structured survey datacollection (n = 4), or a pilot study (n = 6).
Most studies (n = 20) reported conducting a feasibility assessment,which consists of assessment of economic or cultural acceptabil-ity, operability, or perceived sustainability. However, feasibilityassessment varied greatly among studies in terms of rigor andscope.
Process evaluation assesses how well an intervention was imple-mented according to the study plan and is usually assessed interms of reach, dose delivered, and fidelity (75). Most studies (n =18) reported no process evaluation. Two studies reported conduct-ing extensive process evaluations that assessed reach, dose de-livered, and fidelity (10,19).
Most studies (n = 20) assessed the impact of the intervention at theretail level (Table 2). Of these 20 studies, 15 collected data onsales of specific promoted foods. Other studies looked at changesin revenues, food availability, purchasing data, and changes inprices, although these measures were used in only 2 studies(36,40).
We examined impact assessment at the consumer level in 3 differ-ent domains: psychosocial, behavioral, and health outcomes. Morethan half (n = 17) of the studies reviewed included no consumer-level psychosocial assessment. Of those that did, measures usedvaried and included knowledge, self-efficacy, intentions forhealthier behaviors, perceived healthfulness of the diet and afford-ability of healthy foods, perception of barriers to eating healthy,and food security.
Most studies (n = 24) described consumer-level behavioral assess-ment, most often measurements of food purchasing and consump-tion. Only 10 of the 30 studies reviewed measured any type ofconsumer-level health outcome. Most commonly, change in bodymass index (BMI [kg/m2]) was assessed (7 studies), followed byblood work (4 studies).
Study results reported and study implications
We found little consistency in study results reported for feasibilityand process measures (Appendix Table 2). Where reported, feasib-ility of pricing interventions was moderate to high. Pricing inter-ventions were acceptable and generally were implemented asplanned. In 1 study (21), the pricing intervention was not imple-
PREVENTING CHRONIC DISEASE VOLUME 14, E107
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY NOVEMBER 2017
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
4 Centers for Disease Control and Prevention • www.cdc.gov/pcd/issues/2017/17_0213.htm
mented at any site because of food managers’ concerns aboutprofit loss. Where extensive process results were reported (2 stud-ies), implementation quality was generally described as moderate(10,19).
Of the studies (n = 21) that measured an intervention’s impact atthe retail level, the most common effects reported were increasedsales of healthy foods (7 studies) (11,19,20,35,45,58,59, improvedrevenues or total profits (4 studies) (11,25,36,40), increased stock-ing of healthier foods (4 studies) (19,21,26,74, decreased sales ofunhealthy foods (3 studies) (25,64,68), and increased sales ofhealthy foods as a ratio to unhealthy foods (2 studies) (10,55) (Ta-ble 3). All 16 studies that reported effects at the retail level found apositive impact on either stocking or sales. In summary, sales ofunits of healthy foods and beverages increased from 15% (19) to1,000% (25), and sales of unhealthy foods and beverages de-creased from 5% (64) to 47% (69). Stocking of healthy foods in-creased from 40% (19) to 63% (26) in response to pricing inter-ventions (Table 3).
Only 13 studies reported any assessment of the impact of interven-tions on consumer psychosocial factors. Four studies found im-proved perceptions related to healthy eating (27,37,56,74). Threestudies indicated that consumers improved their perception ofhealthfulness or availability of fruits and vegetables (36,45,60).Two studies found that consumers were more likely to shop atfarmers markets (36,43).
Most (n = 23) studies assessed the impact of interventions on con-sumer behavior. Thirteen studies found increases in the consump-tion of healthy foods and beverages associated with the interven-tion (36–38,40,41,43,51,54,56,58,62,65,73), and 8 studies foundi n c r e a s e s i n p u r c h a s i n g o f h e a l t h y f o o d s(10,26,27,31,35,53,54,59). Four studies found a reduction in thepurchasing of unhealthy foods (52, 67, 71, 73). Four studies founda reduction in the consumption of unhealthy foods (55, 56, 64, 72).Two studies reported no effect on healthy food purchasing (44,45),and 1 found no impact on healthy beverage consumption (27).Overall, the pricing interventions, alone or in combination withother approaches, appeared to be successful in changing con-sumer behavior.
Although few studies (n = 8) assessed health-related outcomes atthe consumer level, 5 found no impact on weight (20,22,26,41,58);2 found no impact on various serum vitamin measures when com-paring control and intervention groups over time (41,51).
Of the 14 studies that reported on sustainability of the interven-tion, 10 stated moderate to high sustainability through statewide or
citywide implementation of the intervention (11,40), food policiesthat are still in progress (62,68), and continued interest of the par-ticipants (21,31,36,38,52,74).
The mean score for quality of research measures was 6.9 (stand-ard deviation, 2.0), on a scale of 0 to 10 points (Table 3). Random-ized controlled trials received higher scores than studies without acomparison group. Common study limitations included short inter-vention duration, possible biases in self-reporting, use of nonvalid-ated assessment tools, and lack of power and external validity ofthe findings.
DiscussionTo our knowledge, this is the first systematic review to evaluatethe effectiveness of pricing incentives and disincentive strategieson availability, purchasing, and consumption of healthy and un-healthy foods and beverages in various settings, including field in-tervention studies and natural experiments. The various pricing in-tervention strategies that sought to improve access to healthy foodand beverage choices were successful. This result has been repor-ted by other systematic reviews where subsidies on fruits and ve-getables increased the purchasing and consumption of healthyfoods (2,76,77). However, only one study evaluated the impact offruit and vegetable subsidies from the perspective of retailers (74).Findings that the pricing interventions generally increased stock-ing and sales of promoted foods and beverages are encouraging.There is a need to consistently demonstrate these effects (particu-larly in terms of sales and revenues), to build support from foodretailers and vendors. We recommend that additional studies beconducted to demonstrate beneficial effects of pricing interven-tions on sales, and especially on profits and total retail revenues.
Pricing intervention strategies appeared to positively affect con-sumer-level behavior, with most studies reporting increases in pur-chasing and consumption of healthy foods or beverages or de-creased purchasing and consumption of unhealthy foods or bever-ages. We found no strong pattern to indicate that one type of pri-cing intervention worked better than another — all appeared to beeffective. Additional studies and meta-analyses are needed to dif-ferentiate the potential impact of particular pricing interventionsand policies over others. Only 2 studies changed the placement offoods in a store or market to make healthy choices more evident(26,35). This strategy can be effective, particularly when coupledwith a pricing intervention and should be tested in future trials.
Most studies promoted fresh fruits and vegetables. However, thesefoods, especially for small retailers located in low-income settings,may be hard to source, have high perishability, and raise concerns
PREVENTING CHRONIC DISEASE VOLUME 14, E107
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY NOVEMBER 2017
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
www.cdc.gov/pcd/issues/2017/17_0213.htm • Centers for Disease Control and Prevention 5
about safety and handling (12,78). In addition, it is arguable thatfocusing on fresh fruits and vegetables alone is unlikely to make asubstantial dent in diet-related chronic diseases (79). Pricing inter-vention trials should be broadened to include a range of healthyfoods and beverages, including frozen, and even canned, foods.
Very few studies of food pricing interventions we reviewed dis-couraged unhealthy foods. Formative research has revealed that itis easier to convince food source owners to optimize the purchaseof healthy foods than to get them to discourage the purchase of un-healthy foods (80,81). However, without some emphasis on de-creasing consumer uptake of unhealthy foods and beverages, inter-ventions that focus on healthy foods presume a substitution effectthat may not exist. An exception to this concern are taxes on junkfood and SSBs that have been adopted in recent years. Additionalstudies are needed in real community settings, testing both sub-sidies of healthy foods and beverages and increased prices of un-healthy foods and beverages.
Labeling foods that are part of pricing interventions appears to bea low-cost and effective way to draw attention to these foods.However, few studies reported labeling unhealthy foods. We needadditional experimental trials in community food source settingsthat involve labeling both healthy and unhealthy foods and bever-ages.
Many of the studies we reviewed were small (ie, involved fewerthan 50 respondents per treatment group), which raises concernsregarding enough statistical power to detect the true effect of theintervention. Future studies should be powered to find statisticaldifferences between evaluation groups at the food source and con-sumer levels. Researchers can improve the transferability of theirfindings by disclosing how the sample size was determined.
Few studies included in the review attempted to assess the impactof pricing interventions on health outcomes. It may be unrealisticto hope to see the impact of policy and environmental interven-tions of this nature on health outcomes. However, natural experi-ments, given sufficient study duration, may be able to assess theimpact of some of the large city-based policy initiatives, such assoda taxes. The average intervention duration was less than 1 year,and most lasted only a few months. Pricing intervention studies oflonger duration are needed to track effects on health outcomes,and not just at the behavioral level.
The lack of formative research for most trials is of concern, espe-cially in those studies that targeted specific populations. Evenwhen formative research was conducted, it was minimal and notreported in any detail. Future pricing interventions should bebased on solid formative research, and these findings should be re-ported in the published literature. Process evaluation of any form
was rarely conducted in these studies. This is a major limitation ofthese studies, as it is of any intervention that neglects to collectprocess data (75). Failure to include process evaluation means thatwhether the failure of the intervention was because it was inher-ently flawed or because the intervention was not implemented asintended cannot be understood. Process evaluation data should becollected to assess implementation for all future pricing interven-tions. Several studies emphasized the importance of assessing thesubstitution effect (using savings from discounts to purchase otherless healthy foods) and the compensation effect (purchasing morehealthy foods but not reducing total energy intake) of pricing in-terventions on food purchases and dietary intakes (28,63,70).However, such assessment was not done in any of the studies re-viewed and remains a major gap in this literature. Finally, a majorgap in the studies reviewed is any type of uniform attention toconsumer psychosocial outcomes. We recommend developing acore set of psychosocial measures for these types of interventiontrials and recommend that they be based on theoretical frame-works.
This systematic review has several limitations. First, we focusedexclusively on peer-reviewed literature. It is possible that addition-al, unpublished trials have been conducted. Second, some of thecharacteristics of specific trials that we marked as “not assessed”may have been assessed (eg, conducting formative research, pro-cess evaluation, cost-effectiveness) but were not published in peer-reviewed literature. This information may have been available ingray literature reports, on websites, or in other unexamined docu-ments and thus were not included in this review. However, our useof only peer-reviewed literature helps to ensure a reasonable qual-ity level of the research reported. Third, the use of only peer-re-viewed literature may lead to publication bias, because studieswith negative or null outcomes are less likely to be published.Fourth, our quality of study criteria did not include a measure ofnumber of community venues for implementation sites. Neverthe-less, we used these criteria to ensure comparability to previousstudies (77). Finally, we did not conduct a meta-analysis to evalu-ate the pooled effectiveness of each pricing intervention strategy.Thus, the statement that one pricing strategy was no more effect-ive than any other is based on the synthesis of the results andshould be interpreted with caution. Because each pricing interven-tion strategy assessed different outcomes, it was challenging tocompare the effect sizes of the studies and assess treatment effect.We included many studies outside the United States to enhancegeneralizability.
Pricing incentives and disincentive strategies to affect access, pur-chasing, and consumption of healthy and unhealthy foods andbeverages in high-income and medium-income countries providean evidence-based approach to improve healthy food access at the
PREVENTING CHRONIC DISEASE VOLUME 14, E107
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY NOVEMBER 2017
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
6 Centers for Disease Control and Prevention • www.cdc.gov/pcd/issues/2017/17_0213.htm
retail level and consumer purchasing and consumption (individual-level) behaviors. Most studies reviewed promoted fresh produce,although few discouraged purchasing and consumption of un-healthy foods. Further research that uses robust study designs andmeasurements are needed in real community settings to simultan-eously test subsidies of healthy foods and beverages and the ef-fects of increased costs of unhealthy foods and beverages.
AcknowledgmentsThis research was supported by Healthy Eating Research, a na-tional program of the Robert Wood Johnson Foundation (RWJF).The content is solely the responsibility of the authors and does notnecessarily represent the official views of RWJF. RWJF had norole in study design; collection, analysis, and interpretation ofdata; writing the article; and the decision to submit the article forpublication. We thank Dr Mary Story for her helpful comments.
Author InformationCorresponding Author: Joel Gittelsohn, PhD, Johns HopkinsBloomberg School of Public Health, Department of InternationalHealth, Global Obesity Prevention Center and Center for HumanNutrition, 615 N Wolfe St, Baltimore, MD, 21205. Telephone:410-955-3927. Email: [email protected].
Author Affiliations: 1Johns Hopkins Bloomberg School of PublicHealth, Department of International Health, Global ObesityPrevention Center and Center for Human Nutrition, Baltimore,Maryland.
ReferencesPowell LM, Chriqui JF, Khan T, Wada R, Chaloupka FJ.Assessing the potential effectiveness of food and beveragetaxes and subsidies for improving public health: a systematicreview of prices, demand and body weight outcomes. ObesRev 2013;14(2):110–28.
1.
Thow AM, Downs S, Jan S. A systematic review of theeffectiveness of food taxes and subsidies to improve diets:understanding the recent evidence. Nutr Rev 2014;72(9):551–65.
2.
Andreyeva T, Long MW, Brownell KD. The impact of foodprices on consumption: a systematic review of research on theprice elasticity of demand for food. Am J Public Health 2010;100(2):216–22.
3.
Powell LM, Chaloupka FJ. Food prices and obesity: evidenceand policy implications for taxes and subsidies. Milbank Q2009;87(1):229–57.
4.
Wilde PE, Llobrera J, Valpiani N. Household foodexpenditures and obesity risk. Curr Obes Rep 2012;1(3):123–33.
5.
Eyles H, Ni Mhurchu C, Nghiem N, Blakely T. Food pricingstrategies, population diets, and non-communicable disease: asystematic review of simulation studies. PLoS Med 2012;9(12):e1001353.
6.
Mytton OT, Clarke D, Rayner M. Taxing unhealthy food anddrinks to improve health. BMJ 2012;344(may15 2):e2931.
7.
Shemilt I, Marteau TM, Smith RD, Ogilvie D. Use andcumulation of evidence from modelling studies to informpolicy on food taxes and subsidies: biting off more than we canchew? BMC Public Health 2015;15(1):297.
8.
An R. Effectiveness of subsidies in promoting healthy foodpurchases and consumption: a review of field experiments.Public Health Nutr 2013;16(7):1215–28.
9.
Lee-Kwan SH, Goedkoop S, Yong R, Batorsky B, Hoffman V,Jeffries J, et al. Development and implementation of theBaltimore healthy carry-outs feasibility trial: processevaluation results. BMC Public Health 2013;13(1):638.
10.
Lee-Kwan SH, Bleich SN, Kim H, Colantuoni E, Gittelsohn J.Environmental intervention in carryout restaurants increasessales of healthy menu items in a low-income urban setting. AmJ Health Promot 2014;29(6)357–64.
11.
Noormohamed A, Lee SH, Batorsky B, Jackson A, Newman S,Gittelsohn J. Factors influencing ordering practices atBaltimore City carryouts: qualitative research to inform anobesity prevention intervention. Ecol Food Nutr 2012;51(6):481–91.
12.
Lee-Kwan SH, Yong R, Bleich SN, Kwan NH, Park JH,Lawrence R, et al. Carry-out restaurant intervention increasespurchases of healthy food. J Hunger Environ Nutr 2015;10(4):456–66.
13.
Jeffries JK, Lee SH, Frick KD, Gittelsohn J. Preferences forhealthy carryout meals in low-income neighborhoods ofBaltimore city. Health Promot Pract 2013;14(2):293–300.
14.
Lee SH, Hoffman VA, Bleich SN, Gittelsohn J. Frequency ofvisiting and food dollars spent at carryouts among low-income,urban African American adults. J Hunger Environ Nutr 2012;7(4):459–67.
15.
Lee SH, Rowan MT, Powell LM, Newman S, Klassen AC,Frick KD, et al. Characteristics of prepared food sources inlow-income neighborhoods of Baltimore City. Ecol Food Nutr2010;49(6):409–30.
16.
Hoffman VA, Lee SH, Bleich SN, Goedkoop S, Gittelsohn J.Relationship between BMI and food purchases in low-income,urban African American adult carry-out customers. J HungerEnviron Nutr 2013;8(4):533–45.
17.
PREVENTING CHRONIC DISEASE VOLUME 14, E107
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY NOVEMBER 2017
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
www.cdc.gov/pcd/issues/2017/17_0213.htm • Centers for Disease Control and Prevention 7
Budd N, Cuccia A, Jeffries JK, Prasad D, Frick KD, Powell L,et al. B’More Healthy: Retail Rewards — design of a multi-level communications and pricing intervention to improve thefood environment in Baltimore City. BMC Public Health 2015;15(1):283.
18.
Budd N, Jeffries JK, Jones-Smith JC, Kharmats AY,McDermott AY, Gittelsohn J. Store-directed price promotionsand communications strategies improve healthier food supplyand demand: impact results from a Baltimore City store-intervention trial. Public Health Nutr 2016.
19.
Olstad DL, Goonewardene LA, McCargar LJ, Raine KD.Choosing healthier foods in recreational sports settings: amixed methods investigation of the impact of nudging and aneconomic incentive. Int J Behav Nutr Phys Act 2014;11(1):6.
20.
Linde JA, Nygaard KE, MacLehose RF, Mitchell NR, HarnackLJ, Cousins JM, et al. HealthWorks: results of a multi-component group-randomized worksite environmentalintervention trial for weight gain prevention. Int J Behav NutrPhys Act 2012;9(1):14.
21.
VanWormer JJ, Linde JA, Harnack LJ, Stovitz SD, JefferyRW. Weight change and workplace absenteeism in theHealthWorks study. Obes Facts 2012;5(5):745–52.
22.
VanWormer JJ, Linde JA, Harnack LJ, Stovitz SD, JefferyRW. Self-weighing frequency is associated with weight gainprevention over 2 years among working adults. Int J BehavMed 2012;19(3):351–8.
23.
VanWormer JJ, Linde JA, Harnack LJ, Stovitz SD, JefferyRW. Is baseline physical activity a determinant of participationin worksite walking clubs? Data from the HealthWorks Trial. JPhys Act Health 2012;9(6):849–56.
24.
Brown DM, Tammineni SK. Managing sales of beverages inschools to preserve profits and improve children’s nutritionintake in 15 Mississippi schools. J Am Diet Assoc 2009;109(12):2036–42.
25.
Wolfenden L, Kingsland M, Rowland BC, Dodds P, GillhamK, Yoong SL, et al. Improving availability, promotion andpurchase of fruit and vegetable and non–sugar-sweetened drinkproducts at community sporting clubs: a randomised trial. Int JBehav Nutr Phys Act 2015;12(1):35.
26.
Le HN Gold L, Abbott G, Crawford D, McNaughton SA,Mhurchu CN, et al. Economic evaluation of price discountsand skill-building strategies on purchase and consumption ofhealthy food and beverages: the SHELf randomized controlledtrial. Soc Sci Med 2016;159:83–91.
27.
Ball K, McNaughton SA, Le HND, Gold L, Ni Mhurchu C,Abbott G, et al. Influence of price discounts and skill-buildingstrategies on purchase and consumption of healthy food andbeverages: outcomes of the Supermarket Healthy Eating forLife randomized controlled trial. Am J Clin Nutr 2015;101(5):1055–64.
28.
Olstad DL, Ball K, Abbott G, McNaughton SA, Le HN, NiMhurchu C, et al. A process evaluation of the SupermarketHealthy Eating for Life (SHELf) randomized controlled trial.Int J Behav Nutr Phys Act 2016;13(1):27.
29.
Ball K, McNaughton SA, Mhurchu CN, Andrianopoulos N,Inglis V, McNeilly B, et al. Supermarket Healthy Eating forLife (SHELf): protocol of a randomised controlled trialpromoting healthy food and beverage consumption throughprice reduction and skill-building strategies. BMC PublicHealth 2011;11(1):715.
30.
Ni Mhurchu C, Blakely T, Jiang Y, Eyles HC, Rodgers A.Effects of price discounts and tailored nutrition education onsupermarket purchases: a randomized controlled trial. Am JClin Nutr 2010;91(3):736–47.
31.
Blakely T, Ni Mhurchu C, Jiang Y, Matoe L, Funaki-TahifoteM, Eyles HC, et al. Do effects of price discounts and nutritioneducation on food purchases vary by ethnicity, income andeducation? Results from a randomised, controlled trial. JEpidemiol Community Health 2011;65(10):902–8.
32.
Mhurchu CN, Blakely T, Funaki-Tahifote M, McKerchar C,Wilton J, Chua S, et al. Inclusion of indigenous and ethnicminority populations in intervention trials: challenges andstrategies in a New Zealand supermarket study. J EpidemiolCommunity Health 2009;63(10):850–5.
33.
Ni Mhurchu C, Blakely T, Wall J, Rodgers A, Jiang Y, WiltonJ. Strategies to promote healthier food purchases: a pilotsupermarket intervention study. Public Health Nutr 2007;10(6):608–15.
34.
Cárdenas MK, Benziger CP, Pillay TD, Miranda JJ. The effectof changes in visibility and price on fruit purchasing at auniversity cafeteria in Lima, Peru. Public Health Nutr 2015;18(15):2742–9.
35.
Lindsay S, Lambert J, Penn T, Hedges S, Ortwine K, Mei A, etal. Monetary matched incentives to encourage the purchase offresh fruits and vegetables at farmers markets in underservedcommunities. Prev Chronic Dis 2013;10:E188.
36.
Anderson JV, Bybee DI, Brown RM, McLean DF, Garcia EM,Breer ML, et al. 5 A Day fruit and vegetable interventionimproves consumption in a low income population. J Am DietAssoc 2001;101(2):195–202.
37.
Herman DR, Harrison GG, Afifi AA, Jenks E. Effect of atargeted subsidy on intake of fruits and vegetables among low-income women in the Special Supplemental Nutrition Programfor Women, Infants, and Children. Am J Public Health 2008;98(1):98–105.
38.
Herman DR, Harrison GG, Jenks E. Choices made by low-income women provided with an economic supplement forfresh fruit and vegetable purchase. J Am Diet Assoc 2006;106(5):740–4.
39.
PREVENTING CHRONIC DISEASE VOLUME 14, E107
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY NOVEMBER 2017
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
8 Centers for Disease Control and Prevention • www.cdc.gov/pcd/issues/2017/17_0213.htm
Freedman DA, Mattison-Faye A, Alia K, Guest MA, HébertJR. Comparing farmers’ market revenue trends before and afterthe implementation of a monetary incentive for recipients offood assistance. Prev Chronic Dis 2014;11:E87.
40.
Bihan H, Méjean C, Castetbon K, Faure H, Ducros V, SedeaudA, et al. Impact of fruit and vegetable vouchers and dietaryadvice on fruit and vegetable intake in a low-incomepopulation. Eur J Clin Nutr 2012;66(3):369–75.
41.
Bihan H, Castetbon K, Mejean C, Peneau S, Pelabon L,Jellouli F, et al. Sociodemographic factors and attitudes towardfood affordability and health are associated with fruit andvegetable consumption in a low-income French population. JNutr 2010;140(4):823–30.
42.
Weinstein E, Galindo RJ, Fried M, Rucker L, Davis NJ. Impactof a focused nutrition educational intervention coupled withimproved access to fresh produce on purchasing behavior andconsumption of fruits and vegetables in overweight patientswith diabetes mellitus. Diabetes Educ 2014;40(1):100–6.
43.
Smith C, Parnell WR, Brown RC, Gray AR. Providingadditional money to food-insecure households and its effect onfood expenditure: a randomized controlled trial. Public HealthNutr 2013;16(8):1507–15.
44.
French SA, Story M, Fulkerson JA, Hannan P. Anenvironmental intervention to promote lower-fat food choicesin secondary schools: outcomes of the TACOS Study. Am JPublic Health 2004;94(9):1507–12.
45.
French SA, Story M, Fulkerson JA, Gerlach AF. Foodenvironment in secondary schools: a la carte, vendingmachines, and food policies and practices. Am J Public Health2003;93(7):1161–7.
46.
Fulkerson JA, French SA, Story M, Nelson H, Hannan PJ.Promotions to increase lower-fat food choices among studentsin secondary schools: description and outcomes of TACOS(Trying Alternative Cafeteria Options in Schools). PublicHealth Nutr 2004;7(5):665–74.
47.
Neumark-Sztainer D, French SA, Hannan PJ, Story M,Fulkerson JA. School lunch and snacking patterns among highschool students: associations with school food environmentand policies. Int J Behav Nutr Phys Act 2005;2(1):14.
48.
Shannon C, Story M, Fulkerson JA, French SA. Factors in theschool cafeteria influencing food choices by high schoolstudents. J Sch Health 2002;72(6):229–34.
49.
Hamdan S, Story M, French SA, Fulkerson JA, Nelson H.Perceptions of adolescents involved in promoting lower-fatfoods in schools: associations with level of involvement. J AmDiet Assoc 2005;105(2):247–51.
50.
Burr ML, Trembeth J, Jones KB, Geen J, Lynch LA, RobertsZE. The effects of dietary advice and vouchers on the intake offruit and fruit juice by pregnant women in a deprived area: acontrolled trial. Public Health Nutr 2007;10(6):559–65.
51.
Stites SD, Singletary SB, Menasha A, Cooblall C, Hantula D,Axelrod S, et al. Pre-ordering lunch at work. Results of theWhat to Eat for Lunch study. Appetite 2015;84:88–97.
52.
Thorndike AN, Riis J, Levy DE. Social norms and financialincentives to promote employees’ healthy food choices: arandomized controlled trial. Prev Med 2016;86:12–8.
53.
An R, Patel D, Segal D, Sturm R. Eating better for less: anational discount program for healthy food purchases in SouthAfrica. Am J Health Behav 2013;37(1):56–61.
54.
Sturm R, An R, Segal D, Patel D. A cash-back rebate programfor healthy food purchases in South Africa: results fromscanner data. Am J Prev Med 2013;44(6):567–72.
55.
Olsho LEW, Klerman JA, Wilde PE, Bartlett S. Financialincentives increase fruit and vegetable intake amongSupplemental Nutrition Assistance Program participants: arandomized controlled trial of the USDA Healthy IncentivesPilot. Am J Clin Nutr 2016;104(2):423–35.
56.
An R. Nationwide expansion of a financial incentive programon fruit and vegetable purchases among SupplementalNutrition Assistance Program participants: a cost-effectivenessanalysis. Soc Sci Med 2015;147:80–8.
57.
Kral TVE, Bannon AL, Moore RH. Effects of financialincentives for the purchase of healthy groceries on dietaryintake and weight outcomes among older adults: a randomizedpilot study. Appetite 2016;100:110–7.
58.
Phipps EJ, Braitman LE, Stites SD, Singletary SB, WallaceSL, Hunt L, et al. Impact of a rewards-based incentive programon promoting fruit and vegetable purchases. Am J PublicHealth 2015;105(1):166–72.
59.
Phipps EJ, Kumanyika SK, Stites SD, Singletary SB, CooblallC, DiSantis KI. Buying food on sale: a mixed methods studywith shoppers at an urban supermarket, Philadelphia,Pennsylvania, 2010–2012. Prev Chronic Dis 2014;11:E151.
60.
Phipps EJ, Wallace SL, Stites SD, Uplinger N, BrookSingletary S, Hunt L, et al. Using rewards-based incentives toincrease purchase of fruit and vegetables in lower-incomehouseholds: design and start-up of a randomized trial. PublicHealth Nutr 2013;16(5):936–41.
61.
Falbe J, Thompson HR, Becker CM, Rojas N, McCulloch CE,Madsen KA. Impact of the Berkeley excise tax on sugar-sweetened beverage consumption. Am J Public Health 2016;106(10):1865–71.
62.
Falbe J, Rojas N, Grummon AH, Madsen KA. Higher retailprices of sugar-sweetened beverages 3 months afterImplementation of an Excise Tax in Berkeley, California. Am JPublic Health 2015;105(11):2194–201.
63.
Jensen JD, Smed S. The Danish tax on saturated fat — shortrun effects on consumption, substitution patterns and consumerprices of fats. Food Policy 2013;42:18–31.
64.
PREVENTING CHRONIC DISEASE VOLUME 14, E107
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY NOVEMBER 2017
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
www.cdc.gov/pcd/issues/2017/17_0213.htm • Centers for Disease Control and Prevention 9
Smed S, Scarborough P, Rayner M, Jensen JD. The effects ofthe Danish saturated fat tax on food and nutrient intake andmodelled health outcomes: an econometric and comparativerisk assessment evaluation. Eur J Clin Nutr 2016;70(6):681–6.
65.
Jensen JD, Smed S, Aarup L, Nielsen E. Effects of the Danishsaturated fat tax on the demand for meat and dairy products.Public Health Nutr 2015;19(17):1–10.
66.
Juhl HJ, Jensen MB. Relative price changes as a tool tostimulate more healthy food choices — a Danish householdpanel study. Food Policy 2014;46:178–82.
67.
Batis C, Rivera JA, Popkin BM, Taillie LS. First-yearevaluation of Mexico’s tax on nonessential energy-densefoods : an observa t iona l s tudy . PLoS Med 2016;13(7):e1002057.
68.
Colchero MA, Popkin BM, Rivera JA, Ng SW. Beveragepurchases from stores in Mexico under the excise tax on sugarsweetened beverages: observational study. BMJ 2016;352:h6704.
69.
Colchero MA, Salgado JC, Unar-Munguía M, Hernández-Ávila M, Rivera-Dommarco JA. Price elasticity of the demandfor sugar sweetened beverages and soft drinks in Mexico. EconHum Biol 2015;19:129–37.
70.
Colchero MA, Salgado JC, Unar-Munguía M, Molina M, NgS, Rivera-Dommarco JA. Changes in prices after an excise taxto sweetened sugar beverages was implemented in Mexico:evidence from urban areas. PLoS One 2015;10(12):e0144408.
71.
Darmon N, Lacroix A, Muller L, Ruffieux B. Food pricepolicies improve diet quality while increasing socioeconomicinequalities in nutrition. Int J Behav Nutr Phys Act 2014;11(66):66.
72.
Harnack L, Oakes JM, Elbel B, Beatty T, Rydell S, French S.Effects of subsidies and prohibitions on nutrition in a foodbenefit program: a randomized clinical trial. JAMA Intern Med2016;176(11):1610–8.
73.
Deliens T, Deforche B, Annemans L, De Bourdeaudhuij I,Clarys P. Effectiveness of pricing strategies on French friesand fruit purchases among university students: results from anon-campus restaurant experiment. PLoS One 2016;11(11):e0165298.
74.
Steckler A, Linnan L. Process evaluation for public healthinterventions and research. San Francisco (CA): Jossey-Bass;2002.
75.
Niebylski ML, Redburn KA, Duhaney T, Campbell NR.Healthy food subsidies and unhealthy food taxation: asystematic review of the evidence. Nutrition 2015;31(6):787–95.
76.
An RP. Effectiveness of subsidies in promoting healthy foodpurchases and consumption: a review of field experiments.Public Health Nutr 2014;17(8):1215–28.
77.
Chung J, Li D. The prospective impact of a multi-periodpricing strategy on consumer perceptions for perishable foods.Br Food J 2013;115(3):377–93.
78.
Ledoux TA, Hingle MD, Baranowski T. Relationship of fruitand vegetable intake with adiposity: a systematic review. ObesRev 2011;12(5):e143–50.
79.
Waterlander WE, de Mul A, Schuit AJ, Seidell JC, SteenhuisIHM. Perceptions on the use of pricing strategies to stimulatehealthy eating among residents of deprived neighbourhoods: afocus group study. Int J Behav Nutr Phys Act 2010;7(1):44.
80.
Langellier BA, Garza JR, Prelip ML, Glik D, Brookmeyer R,Ortega AN. Corner store inventories, purchases, and strategiesfor intervention: a review of the literature. Calif J HealthPromot 2013;11(3):1–13.
81.
Story M, Lytle LA, Birnbaum AS, Perry CL. Peer-led, school-based nutrition education for young adolescents: feasibility andprocess evaluation of the TEENS study. J Sch Health 2002;72(3):121–7.
82.
Budd N, Jeffries JK, Jones-Smith J, Kharmats A, McDermottAY, Gittelsohn J. Store-directed price promotions andcommunications strategies improve healthier food supply anddemand: impact results from a randomized controlled,Baltimore City store-intervention trial. Public Health Nutr2017; Epub ahead of print.
83.
PREVENTING CHRONIC DISEASE VOLUME 14, E107
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY NOVEMBER 2017
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
10 Centers for Disease Control and Prevention • www.cdc.gov/pcd/issues/2017/17_0213.htm
Tables
Table 1. Studies of Pricing Strategies to Encourage Purchasing and Consumption of Healthy Foods and Beverages, 2000–2016
Type of Study/Study Name Design Sample Size Duration Target Group/Dates
Model/Theory Goal or Purpose
Financial discounts on healthy foods and beverages
Baltimore HealthyCarryouts (10–17)
Quasi-experimental, 2 arms 8 Carryoutrestaurants, 186consumers
8 months Low-income blackresidents in Baltimore,Maryland, 2011
SCT, SEM, SM To improve healthy foodpurchasing in carryoutrestaurants offering areduced-price combinationmeal; to increase total salesof healthy foods and carryoutrevenue
B’More HealthyRetail Rewards(18,19)
RCT, 4 arms: 1) pricingincentive at wholesaler, 2)communications, 3) pricingand communications, 4)control
24 Corner stores,1 wholesaler, 360store customers
6 months Low-income black adultconsumers in Baltimore,Maryland, 2012–2013
SCT, SEM To assess the impact ofseparate and combinedpricing and communicationstrategies on food purchasingand on retailer stocking andsales
Not named (healthyfoods at swimmingpools) (20)
Quasi-experimentalsuccessive and additiveinterventions: signage, taste-testing, price reduction;qualitative and quantitativeobservations
2 Concessionstands, 650adults, 342children
5 months Pool patrons: children andadults living in Alberta,Canada, in 2012
BE To assess the comparativeand additive efficacy of 2nudges and an economicincentive in supportinghealthy food purchases
HealthWorks(21–24)
Group-randomized controlledtrial, 2 arms: 1) intervention,2) control
6 Worksites,1,672 adults
2 years Employees at 6 worksitesin Minneapolis–St Paul,Minnesota, 2006–2008
SCT To positively influence weightgain prevention
Not named(Mississippi HealthyBeverages) (25)
Quasi-experimental design:first year, no intervention;second year, intervention
15 Schools (noindividual-leveldata)
2 years School-aged children(K–12), various incomelevels, living in Mississippi,2005–2006
Nonereported
To improve purchase ofhealthy beverage choices;maintaining profit in schoolstores by increasingavailability, reducing prices,and labeling
Not named (multi-componentintervention in sportsclubs) (26)
Group-randomized controlledtrial, 2 arms: 1) intervention,2) control
85 Clubs, 1,394club members
2.5 years Adult sports club membersin New South Wales,Australia, 2009–2012
SEM To increase consumption,availability, and sales of non-SSBs and FV in sporting clubcanteens
Supermarket HealthyEating for Life(SHELf) trial (27–30)
RCT, 4 arms: 1) skill-building,2) price reduction, 3) skill-building and price reduction,4) control
642 Women;impact datareported on 574women
3 months Female, main householdfood shoppers, low-SESand high-SESneighborhoods inAustralia, 2011–2012
SEM, SCT To increase purchasing andconsumption of FV, reducepurchasing of SSBs, increasepurchasing of low-calorie softdrinks and water
Supermarket HealthyOptions Project(SHOP) (31–34)
RCT, 4 arms: 1) tailorednutrition education, 2) pricereduction, 3) combination oftailored nutrition and pricereduction, 4) control
1,104 Adults 6 months Diverse adult foodshoppers, including Maori/Pacific Islanders in NewZealand, 2007–2009
Nonereported
To test the effect of pricediscounts and nutritioneducation on supermarketfood and nutrient purchases
Not named (LimaUniversity cafeteriastudy) (35)
Quasi-experimental, 3phases: 1) location only, 2)location and signage, 3)location, signage, and pricereduction
150 Students;qualitativeinterviews, 12students
6 weeks Young adults, collegestudents in Lima Peru,2016
SM To improve fruit purchases ina university cafeteria
Abbreviations: BE, behavioral economics; BMI, body mass index; CBPR, community-based participatory research; ET, economic theory; FV, fruits and vegetables;IMB, information–motivation–behavioral skills theory; K–12, kindergarten through 12th grade; RCT, randomized controlled trial; SCT, social cognitive theory; SEM,social ecological model; SES, socioeconomic status; SM, social marketing; SNAP, Supplemental Nutrition Assistance Program; SSB, sugar-sweetened beverage;WIC, Special Supplemental Nutrition Program for Women, Infants, and Children.
(continued on next page)
PREVENTING CHRONIC DISEASE VOLUME 14, E107
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY NOVEMBER 2017
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
www.cdc.gov/pcd/issues/2017/17_0213.htm • Centers for Disease Control and Prevention 11
(continued)
Table 1. Studies of Pricing Strategies to Encourage Purchasing and Consumption of Healthy Foods and Beverages, 2000–2016
Type of Study/Study Name Design Sample Size Duration Target Group/Dates
Model/Theory Goal or Purpose
Redeemable coupons or vouchers for healthy foods and beverages targeting participants in food assistance programs
Farmers MarketFresh Fund IncentiveProgram (36)
Mixed-methods, pre–postrepeated measure design; nocomparison group
908 Participants;252 with longerfollow-up (1 y)
19 months Low-income urbanHispanic families in SanDiego, California,2010–2011
Nonereported
To examine the effect of adoubling incentive on numberof farmers market visits,consumer diets, andeconomic benefits to farmers
Project FRESH (FarmResourcesEncouraging andSupporting Health)(37)
Quasi-experimental, 4 arms:1) coupon, 2) education, 3)coupon and education, 4)control
455 Adults 4 months Low-income black womenand white womenreceiving WIC in GeneseeCounty, Michigan, 2011
Nonereported
To increase FV attitudes andintake through a couponintervention and educationcombined
Not named (LosAngeles economicsubsidy) (38,39)
Quasi-experimental, 3 arms:1) WIC site no. 1 receivedsupermarket voucher, 2) WICsite no. 2 received farmersmarket voucher, 3) WIC siteno. 3 was control
454 Adults 6 months Adult Hispanic womenreceiving WIC in LosAngeles, California, 2001
Nonereported
To increase FV intake throughan economic subsidy for FVfor postpartum WICparticipants
Shop N Save (40) Quasi-experimental, timeseries, no comparison
336 Adults 2011 and2012 farmersmarket season(40 totalmarket days)
Low-income,predominantly blackwomen in rural SouthCarolina, 2011–2012
CBPR To increase access to FV,increase use of foodassistance, and improverevenue trends at a farmersmarket through a pricingintervention
Redeemable coupons or vouchers for healthy foods and beverages targeting nonparticipants in food assistance programs
Not named (Frenchsupermarkets)(41,42)
RCT, 3 arms: 1) letter withsocial norm feedback, 2)letter plus financial incentive,3) control
2,672 Adults 12 months Low-income health-deprived adults in France,2007–2009
Nonereported
To evaluate the impact ofnutritional counseling alone,or counseling plus vouchers,on FV consumption andbiomarkers
Not named (NewYork City farmersmarkets) (43)
Case-control, nonrandomizedtrial, 2 arms: 1) rebate, 2)control
169,485Households
12 weeks Low-income overweightLatino women with type 2diabetes in New York City,2011
Nonereported
To improve intake andpurchasing of FV by acombined education andvoucher intervention
Spend Study (44) RCT, 2 arms: 1) rebate, 2)control
5,076 SNAPparticipants
4 weeks Low-income, food-insecurehouseholds with 1 child ormore aged ≤18 y in NewZealand, 2009–2010
Nonereported
To examine the effect ofadditional money(supermarket vouchers) onfood expenditures in food-insecure households withchildren.
Trying AlternativeCafeteria Options inSchools (TACOS)(45–50)
RCT, 2 arms: 1) intervention;2) control
54 male andfemale students
2 years Secondary schoolstudents inMinneapolis–St Paul,Minnesota, 2000–2002
SCT To increase availability andsales of low-fat food optionsin high school cafeterias
Not named (UnitedKingdom fruit juicedelivery) (51)
RCT, 2 arms: 1) rewardsintervention, 2) delayedintervention control
58 Adults 30 weeks Low-income pregnantwomen in the UnitedKingdom
Nonereported
To increase fruit and fruitjuice intake by pregnantwomen by using vouchers orcounseling
What to Eat forLunch study (52)
RCT, 2 arms: 1) intervention(2 phases: full interventionincluding voucher; partialintervention no voucher), 2)delayed treatment control
28 Adults 8 weeks Overweight/obese hospitalemployees, majority blackwomen, in Philadelphia,Pennsylvania, 2012
IMB To promote healthy lunchpurchases at work throughcombined mindful eating,initial price reductions, andonline pre-ordering
Abbreviations: BE, behavioral economics; BMI, body mass index; CBPR, community-based participatory research; ET, economic theory; FV, fruits and vegetables;IMB, information–motivation–behavioral skills theory; K–12, kindergarten through 12th grade; RCT, randomized controlled trial; SCT, social cognitive theory; SEM,social ecological model; SES, socioeconomic status; SM, social marketing; SNAP, Supplemental Nutrition Assistance Program; SSB, sugar-sweetened beverage;WIC, Special Supplemental Nutrition Program for Women, Infants, and Children.
(continued on next page)
PREVENTING CHRONIC DISEASE VOLUME 14, E107
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY NOVEMBER 2017
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
12 Centers for Disease Control and Prevention • www.cdc.gov/pcd/issues/2017/17_0213.htm
(continued)
Table 1. Studies of Pricing Strategies to Encourage Purchasing and Consumption of Healthy Foods and Beverages, 2000–2016
Type of Study/Study Name Design Sample Size Duration Target Group/Dates
Model/Theory Goal or Purpose
Cash Rebate
Not named (Bostonsocial norm andrebate study) (53)
RCT, 3 arms: 1) letter withsocial norm feedback, 2)letter plus financial incentive,3) control
2,672 Adults 6 months White high-incomehospital employees,Boston, Massachusetts,2012–2013
Nonereported
To increase healthy foodpurchases through socialnorm feedback with andwithout a financial incentive
Healthy Foodprogram (54,55)
Case-control, nonrandomizedtrial, 2 arms: 1) rebate, 2)control
169,485Households
5 months Members of a SouthAfrican health plan, SouthAfrica, 2009–2012
Nonereported
To examine the effect of pricereductions for healthy fooditems on food purchases(healthy and unhealthy)
Healthy IncentivesPilot (HIP) (56,57)
RCT, 2 arms: 1) rebate; 2)control
5,076 SNAPparticipants
12-monthsinterventionstaggered over3 waves
SNAP participants in rural,urban, and suburbancommunities in HampdenCounty, Massachusetts,2011–2012
ET To test the effect of a rebateon FV purchases on changein purchasing of FV
Not named(Philadelphiafinancial incentives)(58)
RCT, 2 arms: 1) treatment, 2)control
54 Men andwomen
3 months Low-income,predominantly black,middle-aged adults inPhiladelphia,Pennsylvania(year notprovided)
Nonereported
To test the effects of financialincentives for the purchase ofhealthy foods and beverageson purchasing of healthy fooditems, dietary intake,household food environment,BMI
Rewards study(59–61)
RCT, 2 arms: 1) rewardsintervention, 2) delayedintervention control
58 Adults 12-week pilotstudy of 26-week studyduration in 4phases
Low-income black adultsin Philadelphia,Pennsylvania, 2010–2011
SEM To assess impact of rewards-based incentives onpurchases of fresh and frozenFV
Disincentives for unhealthy food and beverage purchases, with and without incentives for healthy food and beverage purchases
Berkeley, California,excise tax on soda(62,63)
Natural experiment,pre–post with a comparisongroup (San Francisco,Oakland)
2,989 (differentsamples: pretax,1,048; posttax,1,941)
1 year Low-income black andHispanic population inBerkeley, 2014–2015
Nonereported
To evaluate the impact of theexcise tax ($0.01/oz) on SSBprices and consumption
Danish saturated fattax (64–67)
Natural experiment,pre–post assessments
2,577Households,1,293 totalretailers
2 years Danish consumers,2010–2012
Nonereported
To estimate the impact of asaturated fat tax onconsumption of saturated fatand other nontargeted dietarymeasures
Excise tax on SSBs inMexico (68–71)
Natural experiment,pre–post assessments
>50,000Households inMexico, 14,784with children
2 years Mexican households,especially households witha child aged 2–5 y,2012–2014
Nonereported
To determine the effect of the1 peso/L tax on SSBs on SSBpurchases
Not named (Frenchfood baskets) (72)
Experimental, controlled, 2conditions tested: 1) FVsubsidy only, 2) FV and otherhealthy food subsidy plusincreased price on unhealthyproducts
128 Women 1 day Low- and medium-incomewomen, main foodshoppers, in Grenoble andLyon, France, 2008
Nonereported
To explore the effect of foodprice policies (taxes,subsidies) on expendituresfor and nutritional quality ofthe food baskets chosen bylow-income and medium-income households
Not named(Minneapolisfinancial incentives)(73)
RCT, 4 arms: 1) FV incentive,2) restrictions, 3) both, 4)control
297 Adults 12 weeks New SNAP participants inMinneapolis, Minnesota,2013–2015
Nonereported
To determine if an FVfinancial incentive alone,prohibition of purchasingunhealthy foods with foodbenefits alone, both in
Abbreviations: BE, behavioral economics; BMI, body mass index; CBPR, community-based participatory research; ET, economic theory; FV, fruits and vegetables;IMB, information–motivation–behavioral skills theory; K–12, kindergarten through 12th grade; RCT, randomized controlled trial; SCT, social cognitive theory; SEM,social ecological model; SES, socioeconomic status; SM, social marketing; SNAP, Supplemental Nutrition Assistance Program; SSB, sugar-sweetened beverage;WIC, Special Supplemental Nutrition Program for Women, Infants, and Children.
(continued on next page)
PREVENTING CHRONIC DISEASE VOLUME 14, E107
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY NOVEMBER 2017
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
www.cdc.gov/pcd/issues/2017/17_0213.htm • Centers for Disease Control and Prevention 13
(continued)
Table 1. Studies of Pricing Strategies to Encourage Purchasing and Consumption of Healthy Foods and Beverages, 2000–2016
Type of Study/Study Name Design Sample Size Duration Target Group/Dates
Model/Theory Goal or Purpose
combination, or controlimproves diet quality
Not named (BrusselsUniversity cafeteriastudy) (74)
Mixed-methods study with 2phases (phase 1, Frenchfries followed by shortinterviews; phase 2, fruitintervention followed byshort interviews)
230 Students;sales datarecorded on2,300–2,930sales for phase 1and 3,235–3,802during phase 2.Qualitativeresearch: 230students
10 weeks in 2phases
University students eatingin on-campus cafeteria,Brussels, Belgium, 2015
Nonereported
To examine the effect of apricing intervention (tax onFrench fries and 10%–20%price subsidy on fruit) onstudents’ purchasingbehavior
Abbreviations: BE, behavioral economics; BMI, body mass index; CBPR, community-based participatory research; ET, economic theory; FV, fruits and vegetables;IMB, information–motivation–behavioral skills theory; K–12, kindergarten through 12th grade; RCT, randomized controlled trial; SCT, social cognitive theory; SEM,social ecological model; SES, socioeconomic status; SM, social marketing; SNAP, Supplemental Nutrition Assistance Program; SSB, sugar-sweetened beverage;WIC, Special Supplemental Nutrition Program for Women, Infants, and Children.
PREVENTING CHRONIC DISEASE VOLUME 14, E107
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY NOVEMBER 2017
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
14 Centers for Disease Control and Prevention • www.cdc.gov/pcd/issues/2017/17_0213.htm
Table 2. Intervention and Evaluation Strategies of Food and Beverage Pricing Studies, 2000–2016
Type ofStudy/StudyName
Type of Food orBeverage Source Intervention
Change ofAvailability, Location,
or Labeling Policy Other
ImpactMeasure:
RetailImpact Measure:
Consumer
Financial discounts on healthy foods and beverages
BaltimoreHealthyCarryouts(10–17)
Healthy preparedfoods (sidedishes, entrees)and beverages(reduced or nosugar)
8 Carryoutrestaurants
Reduced-pricehealthy combinationmeal
Increased stocking ofhealthy foods andbeverages; revisedmenu board tohighlight healthy foodsand beverages(labeling, photos)
None Nutritioneducation
Sales,revenues(carryout)
Psychosocial:none; behavioral:food purchasing;health: none
B’More HealthyRetail Rewards(18,19)
FV, low-sugarbeverages,nutrient-densefoods, low-fatsnacks, whole-grain products
24 Cornerstores, 1wholesaler
10%–30% Pricediscount on healthyfood items at pointof purchase fromwholesaler
Increased stocking ofpromoted healthyfoods; shelf labels andshelf talkersidentifying healthyfoods in store and atwholesaler
None Nutritioneducation,media, andstructuralchanges
Sales (retailerandwholesaler)and owner’spsychosocialfactors
Psychosocial: foodsecurity,knowledge, self-efficacy,intentions;behavioral: foodpurchasing,dietary intake;health: BMI, foodsecurity
Not named(healthy foodsat swimmingpools) (20)
Healthy itemsmeeting definitionof “choose mostoften” (AlbertaNutritionalGuidelines forChildren)
2 Concessionstands in anoutdoorswimming poolfacility
30% Discount onhealthy foods
Heathy menu itemsemphasized withphotographs,appealing names;signage large andclose to cashier; pricereduction for healthyfoods indicated insignage
None Staff training,taste testing ofhealthy foods
Sales,revenues,and grossprofits
Psychosocial:none; behavioral:purchasing;health: none
HealthWorks(21–24)
Low-calorie freshand preparedfoods
6 Worksitecafeterias,vendingmachines
15% Price reductionon calorie-smartfoods
Increased availabilityof healthy foods by atleast 50% of allcafeteria and vendingmachine offerings;smaller portion sizesas substitutes; labeledcalorie-smart items atpoint of purchase
None Media, weightand activityself-monitoring,and nutritioneducation
None Psychosocial:none; behavioral:stair use,absenteeism;health: BMI(measured heightand weight)
Not named(MississippiHealthyBeverages)(25)
Healthybeverages (water,100% fruit juice,sports drinks)
18 Schoolvendingmachines andstores
10%–25% Discounton healthy drinks
At least 50% ofbeverages sold inschool should bewater, 100% fruitjuices, and sportsdrinks; passivemarketing ofbeverages throughchanged facings anddisplay cases
None None None None
Not named(multi-componentintervention insports clubs)(26)
Fresh FV (fruit,salads, or saladsandwiches), non-SSBs
85 Sportingclub canteens
Combination mealswith FV products andwater packagedtogether at areduced price
Substituted higher fatand energy productswith lower fat andenergy products andintroduced productslower in energy, fat, orsodium; positionedpromoted foods at eyelevel, upper half of
Writtenfood andnutritionpolicy
Media, nutritioneducation, andtraining
Financialrecords ofcanteenrevenues
Psychosocial:none; behavioral:purchasing;health: none
Abbreviations: BP, blood pressure; BMI, body mass index; EBT, electronic benefit transfer; FMNP, Farmers Market Nutrition Program; FV, fruits and vegetables;HbA1c, hemoglobin A1c; HDL, high-density lipoprotein; LDL, low-density lipoprotein; NCD, noncommunicable disease; SES, socioeconomic status; SNAP, Supple-mental Nutrition Assistance Program; SSB, sugar-sweetened beverage; WIC, Special Supplemental Nutrition Program for Women, Infants, and Children.
(continued on next page)
PREVENTING CHRONIC DISEASE VOLUME 14, E107
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY NOVEMBER 2017
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
www.cdc.gov/pcd/issues/2017/17_0213.htm • Centers for Disease Control and Prevention 15
(continued)
Table 2. Intervention and Evaluation Strategies of Food and Beverage Pricing Studies, 2000–2016
Type ofStudy/StudyName
Type of Food orBeverage Source Intervention
Change ofAvailability, Location,
or Labeling Policy Other
ImpactMeasure:
RetailImpact Measure:
Consumer
refrigerator, or on thecounter; placedsignage and postersto draw attention toFV products and non-SSBs
SupermarketHealthy Eatingfor Life (SHELf)trial (27–30)
FV (fresh, frozen,canned, anddried), low-caloriesoft drinks, andwater
2 Grocerystores, 1 inlow-SES area, 1in high-SESarea
Reduce purchase ofSSBs; 20% pricediscounts on all FV,low-calorie softdrinks, and water
Not done None Nutritioneducation
Sales data(on healthyfats, FV,healthymeats, andmilk)
Psychosocial:none; behavioral:electronic salesdata measured asa proxy forpurchasing;health: none
SupermarketHealthyOptions Project(SHOP)(31–34)
Healthy corefoods andbeverages thatmet Tick programcriteria (total fat,saturated fat,trans fatty acids,sodium, addedsugar, fiber, andcalcium)
8 New Zealandsupermarkets,all part of thesame chain
12.5% Discount forhealthy foods thatmet Tick programcriteria
Not done None Nutritioneducation
Sales data(cereals,healthy fats,FV, healthymeats, andmilk)
None
Not named(LimaUniversitycafeteria study)(35)
Fresh fruits 1 Universitycafeteria
33% Reduction infruit price
Relocated fruit itemscloser to cash register;nutrition benefit signat fruit container;poster promoting theitem, and price tags
None None Cafeteriasales data(unit, ratio offruitpurchased)
Psychosocial:reasons for notpurchasing fruits;behavioral: none;health: none
Redeemable coupons/vouchers for healthy food and beverage items targeting recipients of food assistant programs
FarmersMarket FreshFund IncentiveProgram (36)
Fresh FV, healthypackaged foods(eggs, bread, andmeat)
5 Farmersmarkets
Fresh Fund incentivematch tokens (up to$20/month) forSNAP recipients
Not done None Mediacampaign
Revenue ofFresh Fundandnon–FreshFundpurchases
Psychosocial:perceivedhealthfulness ofdiet; behavioral:frequency of useof farmersmarket, moneyspent on FV perweek, dailyservings of FVconsumed;health: none
Project FRESH(FarmResourcesEncouragingand SupportingHealth) (37)
Fresh FV City farmersmarkets (nospecificnumber)
$20 Coupons forfarmers markets
Not done None Nutritioneducation
None Psychosocial,attitudes towardFV; behavioral,intake of FV (daily,perceivedchange); health,none
Not named(Los Angeleseconomicsubsidy)(38,39)
Fresh FV Citysupermarkets,farmersmarkets
$10/Weekredeemable voucher
Not done None None Sales ofbeverages(monthly)
None
Shop N Save Fresh FV 1 Farmers $5 Matching Not done None None Consumer None
Abbreviations: BP, blood pressure; BMI, body mass index; EBT, electronic benefit transfer; FMNP, Farmers Market Nutrition Program; FV, fruits and vegetables;HbA1c, hemoglobin A1c; HDL, high-density lipoprotein; LDL, low-density lipoprotein; NCD, noncommunicable disease; SES, socioeconomic status; SNAP, Supple-mental Nutrition Assistance Program; SSB, sugar-sweetened beverage; WIC, Special Supplemental Nutrition Program for Women, Infants, and Children.
(continued on next page)
PREVENTING CHRONIC DISEASE VOLUME 14, E107
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY NOVEMBER 2017
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
16 Centers for Disease Control and Prevention • www.cdc.gov/pcd/issues/2017/17_0213.htm
(continued)
Table 2. Intervention and Evaluation Strategies of Food and Beverage Pricing Studies, 2000–2016
Type ofStudy/StudyName
Type of Food orBeverage Source Intervention
Change ofAvailability, Location,
or Labeling Policy Other
ImpactMeasure:
RetailImpact Measure:
Consumer
(40) market coupons forparticipants of SNAP,WIC, Senior or WICFMNP
sales receipts
Redeemable coupons or vouchers for healthy food and beverage items targeting nonparticipants in food assistance programs
Not named(Frenchsupermarkets)(41,42)
Fresh FV 22Supermarkets
Supermarketvouchers, (10 Euros/person in thehousehold for freshFV)
Not done None Nutritioneducation
Availability ofcalorie-smartfoods
Psychosocial:none; behavioral:self-reportedvoucher use, FVconsumption;health: BMI, bloodpressure, bloodmeasures ofvitamin C, serumβ carotene
Not named(New York Cityfarmersmarkets) (43)
Fresh FV 1 Farmersmarket
$6 Vouchers forfarmers marketpurchases
Not done None Nutritioneducation
None Psychosocial:none; behavioral:frequency of useof farmersmarket, intake ofFV; health: BMI,LDL cholesterol,HbA1c, and BPmeasures
Spend Study(44)
FV, healthygrocery fooditems, and highlyprocessed foods
Supermarket(no specificnumber),identified asmost frequentlyused
$5 Supermarketvoucher perindividual in thehousehold (average$17/household/ wk;voucher was notspecific for any foodor food group)
Not done None Main foodpreparerreceivedreminder textmessages,emails, ortelephone calls
None Psychosocial:none; behavioral:food grouppurchases, totalhousehold foodexpenditure;health: none
TryingAlternativeCafeteriaOptions inSchools(TACOS)(45–50)
Low-fat foods (≤5g/serving)
20 High schoolcafeterias
Coupon (one-timefree low-fat food),raffle for each low-fatfood purchase
Increased availabilityof low-fat foods by30% relative tobaseline
None Nutritioneducation,mediacampaigns
Unit sales,total revenuefrom low-fatfoods sales
Psychosocial:perceivedenvironment,behavioralintentions;behavioral: self-reported foodchoices; health:none
Not named(UnitedKingdom fruitjuice delivery)(51)
Fruit, 100% fruitjuice
Delivery system Vouchers exchangedfor 100% fruit juiceby the local milkdelivery service;received 21 Euros/week for 30 weeks
Not done None Nutritioneducation
None Psychosocial:barriers to eatingfruit; behavioral:dietary intake(frequency of fruitconsumption inpast 7 days);health: serum βcarotene
What to Eat forLunch study(52)
Low-calorie/low-fat cafeteria lunch
1 Hospitalcafeteria
Vouchers to use onlunch purchases (20$1.25 vouchers for 4wk)
Online pre-orderingsystem that providedcalorie and fatcontent; defaultoption was lowercalories/fat
None Nutritioneducation
Foodpurchases;fat andcaloriecontent ofeach item
Psychosocial:mindful eatingquestionnaire;behavioral: none;health: BMI, bloodmeasures (HbA1c,
Abbreviations: BP, blood pressure; BMI, body mass index; EBT, electronic benefit transfer; FMNP, Farmers Market Nutrition Program; FV, fruits and vegetables;HbA1c, hemoglobin A1c; HDL, high-density lipoprotein; LDL, low-density lipoprotein; NCD, noncommunicable disease; SES, socioeconomic status; SNAP, Supple-mental Nutrition Assistance Program; SSB, sugar-sweetened beverage; WIC, Special Supplemental Nutrition Program for Women, Infants, and Children.
(continued on next page)
PREVENTING CHRONIC DISEASE VOLUME 14, E107
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY NOVEMBER 2017
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
www.cdc.gov/pcd/issues/2017/17_0213.htm • Centers for Disease Control and Prevention 17
(continued)
Table 2. Intervention and Evaluation Strategies of Food and Beverage Pricing Studies, 2000–2016
Type ofStudy/StudyName
Type of Food orBeverage Source Intervention
Change ofAvailability, Location,
or Labeling Policy Other
ImpactMeasure:
RetailImpact Measure:
Consumer
triglycerides, HDLcholesterol, LDLcholesterol)
Cash rebate
Not named(Boston socialnorm andrebate study)(53)
Healthy foods onmenu (FV, wholegrains, leanprotein, low-fatdairy); avoidingsaturated fat andcalories
1 Hospitalcafeteria
Monthly financialaward ($5–$30)based on proportionof green-labeledproducts purchased
Traffic-light systemmenu labeling
None Nutritioneducation
Sales datafromcafeteriacashregisters
Psychosocial:none; behavioral:purchases ofgreen-labeledfoods; health:none
Healthy Foodprogram(54,55)
FV, other healthyfoods, less-desirable foodsand beverages,and neutral foods
432Supermarkets
10%–25% Rebatefor healthy foodpurchases
In-store labelingidentifying eligiblefoods, which werealso marked on storereceipts
None None Scanner datafrom creditcard chargesat thesupermarket
Psychosocial:none; behavioral:consumption ofFV, whole grains,salt-added foods,salty and sweetsnacks; health:BMI measure
HealthyIncentives Pilot(HIP) (56,57)
Fresh, canned,frozen, and driedFV
All SNAP-authorizedretailers inHampdenCounty,includingsupermarkets,superstores,grocery andfood specialtystores,conveniencestores, andfarmersmarkets
30% Rebate on totalFV purchased usingEBT cards for 12-month period
Not done None Nutritioneducation
FV sales Psychosocial:attitudes andperceptions aboutFV; behavioral:diet (24-hourrecall, FVscreener); health:none
Not named(Philadelphiafinancialincentives) (58)
FV (fresh, canned,frozen), low-calorie beverages,and low-energy–densefoods
All food storesprovidingreceipts
Financial incentive($1) for everyhealthy food itempurchased over 3months, $100maximum
Not done None Nutritioneducation
None Psychosocial:attitudes towardgrocery stores(baseline only);behavioral:dietary intake (3-day food record),home foodenvironment;health: BMI, waistcircumferencemeasures
Rewards study(59–61)
Fresh or frozen FV(defined by WICguidelines)
1 Largesupermarket inapredominantlyblack censustract
50% Rebate ondollar amount spenton fresh or frozenFV, reduced to 25%during a taperingphase
Not done None Communicationmaterials,nutritioneducation
Point-of-saledata by familyprovided bysupermarket
Psychosocial:perceptions ofbehaviorsresulting fromstudy (post);behavioral: none;health: none
Disincentives for unhealthy food and beverage purchases, with and without incentives for healthy food and beverage purchases
Berkeley, SSBs All beverage Not done Not done SSB $0.01- None Change in Psychosocial:
Abbreviations: BP, blood pressure; BMI, body mass index; EBT, electronic benefit transfer; FMNP, Farmers Market Nutrition Program; FV, fruits and vegetables;HbA1c, hemoglobin A1c; HDL, high-density lipoprotein; LDL, low-density lipoprotein; NCD, noncommunicable disease; SES, socioeconomic status; SNAP, Supple-mental Nutrition Assistance Program; SSB, sugar-sweetened beverage; WIC, Special Supplemental Nutrition Program for Women, Infants, and Children.
(continued on next page)
PREVENTING CHRONIC DISEASE VOLUME 14, E107
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY NOVEMBER 2017
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
18 Centers for Disease Control and Prevention • www.cdc.gov/pcd/issues/2017/17_0213.htm
(continued)
Table 2. Intervention and Evaluation Strategies of Food and Beverage Pricing Studies, 2000–2016
Type ofStudy/StudyName
Type of Food orBeverage Source Intervention
Change ofAvailability, Location,
or Labeling Policy Other
ImpactMeasure:
RetailImpact Measure:
Consumer
California,excise tax onsoda (62,63)
retailers inBerkeley,California
per-ouncetax ondistributors
price pre- andpost-tax
awareness of tax;behavioral:change inconsumption ofSSBs and water;health: none
Danishsaturated fattax (64–67)
Foods containingsaturated fat (eg,butter, blends,margarine, oil,meat, sour cream)
1,293 Foodretailers inDenmark
Not done Not done 25% Tax onfoods highinsaturatedfats
None Monthly salesand revenues
Psychosocial:none; behavioral:purchase of food,food intake;health: modeledestimation ofmortality fromNCDs, BMImeasured
Excise tax onSSBs in Mexico(68–71)
SSBs,nonessential highenergy, highdensity foods
16,000 Foodretailers
Not done Not done 1 Peso perliter tax onSSBs
None Change inprices ofSSBs
Psychosocial:none; behavioral;purchases ofSSBs and othernonessentialfoods; health:none
Not named(French foodbaskets) (72)
43 Foods wereclassified as FV,24 as otherhealthy products,51 as neutral, 62as unhealthy
Real-lifelaboratory,online order(foodsreceived)
Subsidy for FV (30%discount), tax onunhealthy products(30% increase)
Price changeidentified on screen(old price crossed out)
Assessedimpact ofsubsidiesand taxeson foodbasketselections
None None Psychosocial:none; behavioral:purchasing offood baskets;health: none
Not named(Minneapolisfinancialincentives) (73)
FV, SSBs, sweetbaked goods,candies
All SNAPretailers
30% financialincentive for FVpurchased usingfood benefits;restriction (notallowed to buy SSBs,sweet baked goods,or candies) with foodbenefits
Not done None Training None Psychosocial: foodsecurity;behavioral: foodintake and dietquality; health:none
Not named(BrusselsUniversitycafeteria study)(74)
French fries(unhealthy foodproduct) and fruit(healthy foodproduct)
1 On-campus,BrusselsUniversityrestaurant
Total meal priceincreases of 10%and 20% whenchoosing Frenchfries, and 10% and20% meal pricedecreases whenchoosing fruit fordessert
Point-of-purchaseposters andinformation boards
None Social media French friesand fruitsales countsrelative tototal numberof items sold
Psychosocial: foodprice, foodpreference, foodknowledge, bodysatisfaction,perception ofavailability andaccess of food;behavioral: none;health: none
Abbreviations: BP, blood pressure; BMI, body mass index; EBT, electronic benefit transfer; FMNP, Farmers Market Nutrition Program; FV, fruits and vegetables;HbA1c, hemoglobin A1c; HDL, high-density lipoprotein; LDL, low-density lipoprotein; NCD, noncommunicable disease; SES, socioeconomic status; SNAP, Supple-mental Nutrition Assistance Program; SSB, sugar-sweetened beverage; WIC, Special Supplemental Nutrition Program for Women, Infants, and Children.
PREVENTING CHRONIC DISEASE VOLUME 14, E107
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY NOVEMBER 2017
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
www.cdc.gov/pcd/issues/2017/17_0213.htm • Centers for Disease Control and Prevention 19
Table 3. Findings from Pricing Intervention Studies, 2000–2016
Type of Study/StudyName
Retail Stocking andSales Consumer Psychosocial Consumer Behavioral
Consumer HealthOutcomes Sustainabilitya
Quality ofResearchb
Studies of financial discounts on healthy foods and beverages
Baltimore HealthyCarryouts (10–17)
296% Increase in units ofhealthy sides andbeverages sold; totalrevenues (healthy foodsonly: 39%, healthy foodsand beverages: 173%)among interventioncarryouts from baseline
—c 450% Increase inpurchasing of promotedfoods
—c Moderate 9
B’More Healthy RetailRewards (18,19)
40%–61% Increase instocking score in allintervention groups; 15%increase in sales ofsnack foods forcombined interventiongroup
—c —c —c —d 8
Not named (HealthyFoods at SwimmingPools) (20)
30% Increase in sales ofhealthy items duringdiscounted period in asubsample
—c Overweight or obesepatrons and males weremore sensitive to signageplus taste testing pluspricing intervention
—c —d 6
HealthWorks (21–24) 50% Increase inavailability healthy foods;no impact on pricing (notimplemented)
—c Increase in frequency ofself-weighing
No difference on weightoutcomes over the 2-year period
Moderate; allcomponents butpricinginterventionwere sustained.
9
Not named (MississippiHealthy Beverages) (25)
Increase in sales andprofits of 125% on unitsof water, 134% on unitsof sports drinks,>1,000% increase onunits of fruit juice in mostschools; 55% decreaseon unit sales of soda in 9schools
—c —c —c —d 5
Not named (multi-component interventionin sports clubs) (26)
63% Increase inavailability of FV
—c 60% Increase in clubmembers purchasing FV;13.4% increase in clubmembers purchasingnon-SSBs
—c —d 7
Supermarket HealthyEating for Life (SHELf)trial (27–30)
—c Increased perceptions forhealthy eating, cooking,and eating healthy in allinterventions; nodifferences amongintervention groups
35% Increase inpurchasing of FV and15% increase in non-SSBpurchases in pricereduction alone and priceplus behavior arms;impact not maintained at6-monthspostintervention; noeffect on water or low-
—c Low 10
Abbreviations: —, not assessed; BMI, body mass index; FV, fruits and vegetables; HbA1c, hemoglobin A1c; HEI, Healthy Eating Index; SNAP Supplemental NutritionAssistance Program; SSB, sugar sweetened beverage.a Sustainability was scored as low (no mention of continuing the pricing interventions after the study has ended), moderate (few components of pricing interven-tions remained after the study has ended), or high (most components of pricing interventions remained after the study has ended).b Ranked on a scale of 0 to 10.c No assessment of interests was mentioned.d Not reported.
(continued on next page)
PREVENTING CHRONIC DISEASE VOLUME 14, E107
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY NOVEMBER 2017
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
20 Centers for Disease Control and Prevention • www.cdc.gov/pcd/issues/2017/17_0213.htm
(continued)
Table 3. Findings from Pricing Intervention Studies, 2000–2016
Type of Study/StudyName
Retail Stocking andSales Consumer Psychosocial Consumer Behavioral
Consumer HealthOutcomes Sustainabilitya
Quality ofResearchb
calorie beverages
Supermarket HealthyOptions Project (SHOP)(31–34)
—c —c 10% Increase inpurchases of healthyitems (0.79 kg/wk) indiscount group at 6months; no effectreported on education-only group at 6 months
—c High; aftercessation ofpricing discount,discount groupmaintained FVand otherhealthy foodpurchasing at 12months
8
Not named (LimaUniversity cafeteriastudy) (35)
135% Increase in units offruit sales
Most common reason tonot purchase fruit waspreference for unhealthysnack foods
57% Increase inpurchasing of fruitsamong females andtripled among males
—c —d 2
Redeemable coupons or vouchers for healthy foods and beverages targeting participants in food assistance programs
Farmers Market FreshFund Incentive Program(36)
74% Increase in revenueof farmers marketvendors
Increase in perception ofhealthfulness andlikelihood to continueshopping at a farmersmarket
24% Increase in peopleconsuming 5 or more FVservings per day
—c High;participatingfarmers marketscontinued tooffer incentivesto consumers
6
Project FRESH (FarmResources Encouragingand Supporting Health)(37)
—c Increase in attitudes andbeliefs regarding FV inintervention group
140%–640% Increase inFV intake score inintervention groups
—c —d 7
Not named (Los Angeleseconomic subsidy)(38,39)
—c —c 30% Increase in FVintake in interventiongroups
—c High; increase inFV intakesustained 6months afterintervention
7
Shop N Save (40) 43% Increase in foodassistance revenues atfarmers markets
—c —c —c High; DoubleBuck program tobe implementedby SouthCarolina
5
Redeemable coupons or vouchers for healthy foods and beverages targeting nonparticipants in food assistance programs
Not named (Frenchsupermarkets) (41,42)
—c —c 33% Increase in FVintake in voucher groupand 32% for advice group
No change in serumvitamin C and βcarotene levels; nodifference in otherhealth measures bygroup
—d 6
Not named (New YorkCity farmers markets)(43)
—c Decrease in reporteddifficulty of affording FV
20% Increase in FVintake (servings/d)
Decreased BMI andHbA1c, but nosignificant difference bygroup
—d 7
Spend Study (44) —c —c No difference inexpenditures on otherfood groups (ie, FV, meat
—c —d 7
Abbreviations: —, not assessed; BMI, body mass index; FV, fruits and vegetables; HbA1c, hemoglobin A1c; HEI, Healthy Eating Index; SNAP Supplemental NutritionAssistance Program; SSB, sugar sweetened beverage.a Sustainability was scored as low (no mention of continuing the pricing interventions after the study has ended), moderate (few components of pricing interven-tions remained after the study has ended), or high (most components of pricing interventions remained after the study has ended).b Ranked on a scale of 0 to 10.c No assessment of interests was mentioned.d Not reported.
(continued on next page)
PREVENTING CHRONIC DISEASE VOLUME 14, E107
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY NOVEMBER 2017
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
www.cdc.gov/pcd/issues/2017/17_0213.htm • Centers for Disease Control and Prevention 21
(continued)
Table 3. Findings from Pricing Intervention Studies, 2000–2016
Type of Study/StudyName
Retail Stocking andSales Consumer Psychosocial Consumer Behavioral
Consumer HealthOutcomes Sustainabilitya
Quality ofResearchb
food groups (ie, FV, meatand poultry, dairy)
Trying AlternativeCafeteria Options inSchools (TACOS) (45–50)
Increase in sales of low-fat foods (at year 1,27.5%; at year 2, 33.6%);no significant change intotal revenues
Increased perception ofmore low-fat foodavailability; no change inenvironmental orbehavioral intentions
No impact on foodchoices
—c —d 10
Not named (UnitedKingdom fruit juicedelivery) (51)
—c Taste and appetite werebarriers to eating fruit
59.1% Increase in fruitjuice intake (netpercentage ofconsumption)
Increase in serumβcarotene
—d 8
What to Eat for Lunchstudy (52)
—c Increased scores formindful eating behaviors
8% Decrease in totalcalories and 6%decrease in total fat infood purchases usingpre-ordering only (novouchers) compared withbaseline and discountphase
Decreased weight,HbA1c, and lipid profilesfrom pre to post, but notsignificant
High; in partialinterventionphase withoutany financialincentives,participants stillbought foodswith lowercalorie and fatcontent
9
Cash rebate
Not named (Bostonsocial norm and rebatestudy) (53)
—c —c 2.2% Increase inpurchasing of green-labelfoods; no differencebetween interventionarms; increase in healthyfood choices in socialnorms and smallfinancial incentives
—c Low; interventioneffect did notpersist 3 monthsafter completionof trial
10
Healthy Food program(54,55)
—c —c 9.3% Increase in healthyfood to total foodexpenditures; 63%increase in consumptionof FV and 195% of whole-grain foods; 68%decrease in consumptionof unhealthy foods (high-sugar or high-salt foods,fried foods, processedmeats, fast food)
No effect on obesity —d 7
Healthy Incentives Pilot(HIP) (56,57)
Increase in FV sales inlarge grocery store(qualitative finding)
Increase in attitudetoward FV over time
40% Increase in FVintake; 10% decreasedintake of refined grain;increase in HEI–2010score
—c —d 9
Not named (Philadelphia,financial incentives) (58)
—c —c 10% Increase in proteinintake, 28% in calciumintake, and 60% in dailyvegetable intake;increase in household
Slight decrease in BMIin both groups
—d 6
Abbreviations: —, not assessed; BMI, body mass index; FV, fruits and vegetables; HbA1c, hemoglobin A1c; HEI, Healthy Eating Index; SNAP Supplemental NutritionAssistance Program; SSB, sugar sweetened beverage.a Sustainability was scored as low (no mention of continuing the pricing interventions after the study has ended), moderate (few components of pricing interven-tions remained after the study has ended), or high (most components of pricing interventions remained after the study has ended).b Ranked on a scale of 0 to 10.c No assessment of interests was mentioned.d Not reported.
(continued on next page)
PREVENTING CHRONIC DISEASE VOLUME 14, E107
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY NOVEMBER 2017
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
22 Centers for Disease Control and Prevention • www.cdc.gov/pcd/issues/2017/17_0213.htm
(continued)
Table 3. Findings from Pricing Intervention Studies, 2000–2016
Type of Study/StudyName
Retail Stocking andSales Consumer Psychosocial Consumer Behavioral
Consumer HealthOutcomes Sustainabilitya
Quality ofResearchb
food environment
Rewards study (59–61) —c Increased perception ofbuying more FV
25%–30% Increase inpurchasing of FVservings/wk (30%increase in vegetableand 25% increase in fruitservings/wk); effect notsustained when incentivewas reduced
—c Low; changesnot maintainedduring taperingperiod
7
Disincentives for unhealthy food and beverage purchases, with and without incentives for healthy food and beverage purchases
Berkeley, California,excise tax on soda(62,63)
9% Increase in SSB retailprices in Berkeley
Increased knowledgeabout the tax
21% increasedconsumption of SSBs;63% increased waterconsumption
—c High; tax is stillin effect andBerkeley CityCouncilallocated $1.5million to fundprogram toreduce SSBconsumption
6
Danish saturated fat tax(64–67)
5% Decrease in sales ofground beef and creams
—c 4% Increase in saturatedfat intake; 1% increase insalt intake; 9% increasein FV intake
Increase in deaths fromcardiovascular disease(modeled)
Low; tax is nolonger in effect
5
Excise tax on SSBs inMexico (68–71)
47% Decrease in sales oftaxed foods; no changein sales of untaxed foods
—c 12% Decrease in SSBpurchases; 5% decreasein nonessential foodpurchases
—c High; tax is stillin effect
5
Not named (French foodbaskets) (72)
—c —c 25%–30% Increase inquantity of FV purchasedamong low- and middle-income shoppers; 52%decrease in unhealthyfood expenditures amongmiddle-income shoppersfor nutrient profilecondition
—c —d 4
Not named (Minneapolisfinancial incentives) (73)
—c Increased food securityin all groups
2%–6% Decreasedenergy intake (incentive,2%; restriction, 6%;combined, 6%); 66%increased intake of fruit;8% increase in HEI–2010score
—c —d 9
Not named (BrusselsUniversity cafeteriastudy) (74)
Increase in availability ofhealthy foods (qualitativeresult: perception ofstudents that influencedtheir food choices)
10.9%–21.8% decreasein French friespurchases;25.1%–42.4% increaseFV purchases
—c Moderate;studentsbelieved thatfruit pricereduction couldbe sustained inthe long term
3
Abbreviations: —, not assessed; BMI, body mass index; FV, fruits and vegetables; HbA1c, hemoglobin A1c; HEI, Healthy Eating Index; SNAP Supplemental NutritionAssistance Program; SSB, sugar sweetened beverage.a Sustainability was scored as low (no mention of continuing the pricing interventions after the study has ended), moderate (few components of pricing interven-tions remained after the study has ended), or high (most components of pricing interventions remained after the study has ended).b Ranked on a scale of 0 to 10.c No assessment of interests was mentioned.d Not reported.
PREVENTING CHRONIC DISEASE VOLUME 14, E107
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY NOVEMBER 2017
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
www.cdc.gov/pcd/issues/2017/17_0213.htm • Centers for Disease Control and Prevention 23
Appendix. Pricing Strategies to Encourage Availability, Purchase, and Consumptionof Healthy Foods and Beverages: A Systematic ReviewThis appendix is available for download as a Microsoft Word document at
https://www.cdc.gov/pcd/issues/2017/docs/17_0213Appendix.docx [DOCX – 33KB].
PREVENTING CHRONIC DISEASE VOLUME 14, E107
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY NOVEMBER 2017
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
24 Centers for Disease Control and Prevention • www.cdc.gov/pcd/issues/2017/17_0213.htm