Lunch in the Big Apple didn't get "lighter" after New York City started requiring fast food chains to post calorie content, Tamara Dumanovsky and colleagues reported online in BMJ.
But the one in six customers surveyed who did use the calorie counts cut 106 calories from their lunch purchase on average compared with those who either didn't see the calorie information or who simply ignored it, wrote Dumanovsky, an independent consultant who used to work for the New York City Department of Health and Mental Hygiene.
Saving that many calories on a daily basis would mean losing (or not gaining) one pound a month, explained coauthor Cathy A. Nonas of the NYC Department of Health and Mental Hygiene.
Read this story on www.medpagetoday.com.
On a public health scale, it could translate to 150,000 fewer cases of obesity in New York City alone, she said in an interview monitored by media relations for the department.
"Small changes make a big difference," Nonas told MedPage Today.
New York City rolled out legislation in 2008 requiring any restaurant with 15 or more franchises across the country to post calorie information on its menus.
Chain restaurants across the nation will have to follow suite once provisions of the 2010 Affordable Care Act go into effect. Some areas, including Seattle, Philadelphia, and all of California, already have similar regulations in place.
Evidence for efficacy is "good news," according to Marion Nestle,a food politics expert at New York University in New York City.
"Calorie labeling is working for people who pay attention to it. That's all anyone can ask," she wrote in an email. "The next trick is to get more people to pay attention."
In the study, which surveyed New Yorkers who purchased food at 168 randomly-selected locations of top fast food chains in the city during lunchtime, just 15 percent of customers reported using the calorie information in 2009.
Nonas urged physicians to join the effort to boost these numbers by talking to patients about using the calorie labelling.
While a good first step, tackling obesity may require more, argued Susan A. Jebb in an editorial accompanying the BMJ paper.
People make food choices in a wider context of taste, price, and social norms, so further long-term advances may require "a transformation of the food supply too," noted Jebb, co-chair of a voluntary British program for calorie labeling in restaurants.
After the New York City program went into effect, register receipts overall showed no change in calories purchased for lunch (mean 828 in 2007 versus 846 in 2009).
Women, older adults, and customers in the wealthiest areas were more likely to have relied on the calorie labelling for their lunch decisions.
After adjusting for these factors, type of purchase, restaurant chain, and inflation, there was a small effect across the board as average calories fell from 847 in 2007 to 827 in 2009.
More substantial calorie cutting occurred after the calorie counts were posted at three chains, accounting for 42 percent of the lunchtime crowd surveyed:
McDonald's purchases dropped from an average of 829 to 785 calories
KFC diners reduced their calories from 927 to 868
Au Bon Pain purchases declined in average calorie content from 555 to 475
Subway was the only chain that bucked the trend, going from an average of 749 calories for lunch in 2007 to 882 in 2009.
The researchers speculated that menu changes and marketing may have had something to do with these trends as well.
For example, Au Bon Pain brought in a portion-controlled menu in 2008, while KFC introduced grilled chicken to its menu in 2009.
But Subway started promoting a "super-sized portion -- the '$5 foot foot-long'" in 2008, which boosted caloric averages there, the group noted.
"As much as calorie posting is really good, you can't take the responsibility off the restaurants," Nonas argued. "We must put pressure on them to make formulation changes for smaller portions, for healthier options, and for lower calorie food."
Restaurants aren't likely to be hurt financially, she noted.
Customers who reported using calorie information bought fewer items and fewer drinks, but didn't spend less overall than those who didn't see or use the labeling.
The researchers cautioned that generalizability to areas outside New York City or to breakfast or dinner decisions isn't clear.
Other limitations included potential bias from reliance on self-reporting for use of the calorie labels, lack of data on whether people actually ate everything they bought, and lack of a control group given the economic downturn that may have shifted fast food consumption patterns, the investigators added.