In 2008, New York City became the first place in the country to require fast-food restaurants and coffee chains to post calorie counts for their food and drinks. The antiobesity measure, officials hoped, would allow consumers to be better informed about their food choices and respond by cutting back on their overall calories.

Research has shown that consumers fare poorly when estimating calories on their own, but that they read calorie displays in restaurants. A new study by researchers at NYU Langone, however, reinforces the idea that the added information alone may not be enough to change consumers’ eating habits.

Brian Elbel, PhD, associate professor of population health and medicine, led earlier efforts that cast doubt on the impact of New York City’s calorie-labeling policy about a year after it went into effect. For a study published in a recent issue of the journal Health Affairs, the same group took a longer-term view of the policy. “Five to six years after labeling started, we still didn’t see any change in total calories purchased,” Dr. Elbel says.

The new study also suggested that the number of people who reported seeing and using the calorie labels was declining over time. “I think what that shows us is that while a subset of people are still seeing and using this information, we haven’t seen any population-level shift in the number of calories purchased as a result of labeling policies at fast-food restaurants,” Dr. Elbel says.

For the study, research assistants asked 7,699 people entering McDonald’s, Wendy’s, KFC, and Burger King restaurants to return their itemized receipts and answer a few follow-up questions in exchange for $2. The objective measure of calories purchased helped the researchers calculate what the consumers likely ate in all. The study compared calories consumed at the fast-food restaurants in New York City to the same restaurant chains in nearby New Jersey cities that lacked calorie displays. After tallying the numbers, the researchers found no significant difference between the two sets of consumers over time.

Dr. Elbel says the partial disconnect between people who still reported seeing and using the information but didn’t alter their behavior might be due to social desirability—providing the answer they think others want to hear. Alternatively, the behavioral effect may have been too small to be detectable, or the consumers may have mistakenly thought they read the displays or bought fewer calories than they really did.

As part of the Affordable Care Act, a similar calorie display rule is set to take effect throughout the country by December 2016, but Dr. Elbel says his research suggests that labeling alone will likely be insufficient to spur a change toward healthier behavior. “If we’re really going to come up with larger, sustained, population-level solutions to obesity, we’re going to have to think beyond these more informational sorts of approaches like calorie labeling. I think that’s pretty clear,” he says.

A previous modeling study by Dr. Elbel’s group concluded that regulating the size of sodas available to consumers might be an effective strategy. New York City’s proposed ban on the sale of supersized sodas, however, was invalidated by the courts before it went into effect in 2013. Other research, however, has provided evidence that food pricing, marketing, and availability also may play a role in determining what people eat—and how much of it. “I think the biggest message here is that it’s not going to be one thing that makes a difference, it’s going to a combination of things,” Dr. Elbel says.