Cash-Back Program for Healthy Groceries Improves Diets

March 25, 2013

JOHANNESBURG — A large insurance company's grocery-purchase rebate program, with participation from a nationwide supermarket chain, motivated shoppers to buy healthier foods like fresh produce and fewer sweets and snack foods[1].

While it worked for a segment of the population in South Africa, in its current form, it would be hard to apply nationwide and in other countries like the US, according researchers of the program, led by Dr Roland Sturm (RAND Corporation, Santa Monica, CA). Their analysis was published last week in the American Journal of Preventive Medicine.

Cash-register price-scanner data linked to 170 000 households, including families taking part and those not in the rebate program, were used to return as cash 10% and 25% of the cost of specific healthy food items. Participants started out earning the lower rebate and could progress to the 25% level with greater involvement in the program.

The 25% rebates, as it happened, proved to be the more effective incentive, not only for buying healthy foods overall but for their subset of fruits and vegetables.

Changes in Proportion of Prespecified Food Types to Total Foods in Grocery Purchases From 2009 to 2012, by Level of Cash Rebate for Healthy Food Items

Change in ratio of expenditures on: 10% rebate 25% rebate
Healthy foods to total foods (%) +6.0 +9.3
Fruits/vegetables to total foods (%) +5.7 +8.5
"Less desirable"* foods to total foods (%) -5.6 -7.2

*"Those that are high in saturated fats, trans-fatty acids, added sugar, salt, or refined starch," including "sweets, chocolates, ice cream, sugary foods, chips, sugar-sweetened beverages, and fried items"

The rebates applied to a list of >6000 food items classified as healthy by a panel of experts, based on international guidelines, that included fruits and vegetables, whole grains, and nonfat dairy items and excluded "most items with added sugars or salt," according Sturm et al.

Coauthor Ruopeng An (RAND) urged caution in interpreting the findings. They suggest, he told heartwire , that "providing cash discounts is definitely effective in changing people's dietary behavior, but the cost is also pretty high."

The healthy-food rebates are only one part of a broad insurance-company sponsored program that also includes incentives for engaging in physical activity, health screenings, preventive-care services, and other wellness behaviors, according to An. The goal is to cut the cost of paying for healthcare overall.

"The program has been successful in terms of attracting more people to be enrolled," he said, "but we haven't looked into the data yet to see if the program is cost-effective--whether it saves money for insurers in the long run."

And whether it would work in other countries is unknown. "I do think there are some hurdles if you want to translate our results to the United States," An said. "It's even difficult to generalize the results to the South African general population."

A key reason is that the participating families were members of a private health plan, who are disproportionately "healthy and wealthy, with a much lower obesity and overweight rate compared with the general population," he said.

"I think we should be quite conservative in generalizing the results to the US," and any contribution to public health would likely be small. "Even with this huge discount, we can probably close only a very small proportion of the gap between what people are eating and what is recommended by the American dietary guidelines."

The authors had no disclosures.