Produce Consumption in the United States—An Analysis of Consumption Frequencies, Serving Sizes, Processing Forms, and High-Consuming Population Subgroups for Microbial Risk Assessments

Journal of Food Protection, Volume 75, Number 2, February 2012 , pp. 328-340(13), DOI: 10.4315/0362-028X.JFP-11-313
K. HOELZER,* R. POUILLOT, K. EGAN, AND S. DENNIS
Center for Food Safety and Applied Nutrition, U.S. Food and Drug Administration, 5100 Paint Branch Parkway, College Park, Maryland 20740, USA

ABSTRACT

A great variety of fruits and vegetables are available in the United States. These items are produced in various geographic regions by a diverse industry. Produce has been increasingly identified as a vehicle for disease outbreaks. Changes in consumption may explain this increase, but analyses of produce consumption are limited. Comprehensive assessments of the public health risks associated with produce depend on quantitative consumption data, including the population fractions and subgroups of consumers, the quantities consumed by these individuals, and the processing that occurs before consumption. Here, we provide an analysis of nationally representative consumption estimates by estimating consumption frequencies, serving sizes, and processing forms for a variety of produce commodities based on 1999 through 2006 data from ‘‘What We Eat in America,’’ the dietary interview component of the National Health and Nutrition Examination Survey performed by the National Center for Health Statistics. Consumption patterns for fresh and heat-treated produce were assessed, compared with U.S. food availability estimates from the U.S. Department of Agriculture Economic Research Service (ERS), and combined with ERS data on temporal trends in food availability and nondomestic produce origins. To identify high-consuming population subgroups, we explored consumer habits and demographic predictors of fresh produce consumption (data available at www.foodrisk.org). Our analysis of common outbreak vehicles revealed limited temporal changes in food availability but frequent consumption as fresh commodities. In addition to providing quantitative consumption estimates for risk assessments, our data clearly show that produce consumption differs among fruits and vegetables, fresh and heat-treated foods, and demographic subgroups. These results are valuable for risk assessments and outbreak investigations and allow targeting of risk communication or interventions to those individuals at greatest risk.

Supplemental Materials

Supplemental Tables

Tables

Suppl. Table 1: Population fractions consuming commodity categories in any given food form.
Suppl. Table 2: Percentage of produce commodities consumed in each processing form, based on mean per-capita consumption.
Suppl. Table 3: Percentiles of serving size distributions (among eaters only) in gram/day for consumption in all food forms.
Suppl. Table 4: Percentiles of serving size distributions (among eaters only) in gram/day for consumption as fresh product.
Suppl. Table 5: Percentiles of serving size distributions (among eaters only) in gram/day for consumption as heat-treated product.
Suppl. Table 6: Percentiles of serving size distributions (among eaters only) in gram/day for consumption as dried product.
Suppl. Table 7: Percentiles of serving size distributions (among eaters only) in gram/day for consumption as frozen product.
Suppl. Table 8: Percentiles of serving size distributions (among eaters only) in gram/day for consumption as cured product.
Suppl. Table 9: Demographic predictors of fresh produce consumption
Suppl. Table 10: Varimax rotated factor patterns, based on principle factors. Consumption of fruits and vegetables was included in the analysis if commodity food form was consumed by at least 4% of the total population.

Supplemental Figures

Figures

Suppl. Figure 1: Fraction of commodity group consumed in each processing form, based on mean per-capita consumption
Suppl. Figure 2: Serving size distributions among consumers for commodities in all food forms, fresh and as heat-treated foods