What Americans Eat Is a Mystery
What are Americans eating? Who knows. That’s because the U.S. Department of Agriculture, which coordinates one of the nation’s leading surveys of dietary practices, has fallen down on the job, says the General Accounting Office.
A report released recently found the USDA’s most recent food consumption survey riddled with mismanagement, poor design, reporting bias and, most glaring, only a 34% response rate, which the GAO said “may not be representative of the U.S. population.”
Results of the survey, conducted every 10 years since 1936, are considered critical information for multibillion-dollar federal food assistance programs and key regulatory and nutrition education programs. They are also used by food marketers, researchers, agricultural producers and everybody else who wants to know who’s eating what.
Many health experts, who have long been dubious about the reliability of USDA’s surveys, are concerned because nobody knows how useful or useless the 1987-88 food consumption survey is. As a result, major public health programs that rely heavily on the information could be thrown askew.
“No one will ever know with certainty how good the data are,” said Jerry Killian, who was GAO’s assistant director of food and nutrition projects when the report was compiled. “They may be perfectly good. Or they may be perfectly lousy.”
Sue Ann Ritchko, administrator of USDA’s Human Nutrition Information Service, said the GAO’s assessment of the survey was accurate and the agency has taken steps to rectify problems. Nevertheless, Ritchko said, while the survey has limitations, it is still “a very robust source of data.”
Booz, Allen & Hamilton, the parent company of National Analysts, which conducted the survey under contract from the USDA, said the GAO’s findings were valid and that it had launched an internal investigation of the problems identified by the agency, which included:
* Aside from the extremely low response rate, National Analysts did not give details on those who didn’t respond--crucial in determining how they differ from those who did. The firm said it lost the documents during an office move.
* The two-part survey was burdensome and time-consuming. The first section required the household’s main meal preparer to complete an 89-page questionnaire about foods consumed or disposed of during a week’s period. The second section required each household member to keep a three-day food diary. Each participant was paid only $2, and the interviewers sometimes just showed up at respondents’ doors without advance notice.
* Administration of the contract was lax and mismanaged at nearly every stage, wasting federal funds. Results were two years overdue and $1.4 million over budget.
Of particular concern is the impact the questionable information will have on the Environmental Protection Agency, which uses consumption data to calculate the maximum residues of pesticides that can remain on foods. Very different risk levels can result from plugging different consumption figures into the equations. The EPA has chosen to stick with the dietary patterns identified by the USDA’s 1977-78 survey, which are clearly outdated, says Rick Borchelt of the House Committee on Science, Space and Technology, which commissioned the GAO report.
Nutritionists also worry that targeted educational efforts could be hampered. For example, it may become unclear to what extent certain people need to increase iron or decrease fat or sodium. The 1987-88 survey also grossly under-surveyed children younger than age 2 and working mothers, according to Ritchko. Those two groups have distinct dietary patterns and needs.
What’s more, if we don’t know what people are eating, research priorities could be upset. “What are the research questions we need to ask to improve the diet? We shouldn’t spend money on nutrition research we don’t need,” said a staff aide for the House committee.
While the GAO’s report highlighted problems with the USDA’s 1987-88 survey, diet experts have been disenchanted with the agency for years. “People in the nutrition community have questioned if the USDA’s surveys have ever been representative of the U.S. population,” said Nancy Chapman, director of public policy for the Society for Nutrition Education.
