Poll: Many Americans claim to be improving their diets

NEW YORK While many consumers claim to have adjusted their diets to include more healthy foods and beverages, new data suggested that there may be a disparity between what is considered "good" by consumers and by experts.

Anew poll by Harris Interactive, conducted between Sept. 14 and 20 among 2,620 adults, found that the majority of all adults claimed that they:

• Frequently or somewhat often eat healthier at home compared with when dining out (79%);


• Drink water as opposed to another type of beverage at meals (74%);


• Choose healthy snacks (72%);


• Eat a balanced diet (72%);


• Read nutritional information on packaged food products before buying it (68%);


• Attempt to eat smaller portions (64%); and


• Exercise regularly (57%).

All of the responses were in line with what doctors and nutrition experts recommend for people to maintain a healthy weight.

Harris Interactive, however, noted that some of these results may reflect what consumers think they should be doing rather than what they actually are doing. For example, while there were some differences in their replies to this question among those who are and are not overweight or obese, the differences are not very large. Most of those who are obese or even morbidly obese claimed to be doing the same healthy things that those who are not overweight claimed they were doing.

Relatively few people are regularly (five or more times per week) eating a full breakfast (22%); a full or well-balanced lunch (21%); or a full or well-balanced dinner (37%). Overall, 32% of consumers surveyed felt they were heavier than they should be but also felt they generally were healthy and content. Among the 32%:

• 10% were normal weight;


• 41% were overweight;


• 55% were obese; and


• 30% were morbidly obese.

Responses to the poll were analyzed by Americans' body mass index.

This Harris Poll was conducted online within the United States between Sept. 14 and 20, 2010, among 2,620 adults (aged 18 years and over). Figures for age, sex, race/ethnicity, education, region and household income were weighted where necessary to bring them into line with their actual proportions in the population. Propensity score weighting also was used to adjust for respondents' propensity to be online.