Skip to main content
ARS Home » Northeast Area » Beltsville, Maryland (BHNRC) » Beltsville Human Nutrition Research Center » Food Components and Health Laboratory » Research » Publications at this Location » Publication #158411

Title: COMMERCIAL PORTION-CONTROLLED FOODS IN RESEARCH STUDIES: HOW ACCURATE ARE LABEL WEIGHTS?

Author
item Conway, Joan
item RHODES, DONNA - FSRG
item Rumpler, William

Submitted to: Journal Of The American Dietetic Association
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/11/2004
Publication Date: 7/1/2004
Citation: Conway, J.M., Rhodes, D.G., Rumpler, W.V. 2004. Commercial portion-controlled foods in research studies: how accurate are label weights? Journal Of The American Dietetic Association. 104:595-603.

Interpretive Summary: The purpose of this study was to evaluate the reliability of label weights of single-serving sized commercial portion-controlled foods under conditions of a research setting. Actual weights of 83 portion controlled (PC) food items and 21 portions of food from larger packaging were determined over 3 months and 2 years respectively. Comparison was made to the package label weights. The study was conducted at USDA's Beltsville Human Nutrition Research Center's Human Study Facility, which houses a metabolic kitchen. The difference between label and actual weights was tested by the appropriate statistical analyses. Compliance with federal guidelines of the US Food and Drug Administration (FDA) and the US National Institute for Standards and Technology (NIST) was also tested. There were no statistical differences between actual and label weights for only 37 food items. The actual weights of 15 PC food items were more than 1% below label weights, making them potentially out of compliance with the FDA guidelines. Single-serving sized cold cereal products had the highest error between actual and label weights. With advance planning and continuous monitoring, well-controlled feeding studies could incorporate PCs especially beverages, snack food, and confectionery products. Dietitians should encourage diabetics and others following strict dietary regimens to check actual weights of portion-controlled products carefully against package weights. This data is also important for food processors, manufacturers, and institutions that rely heavily on single-serving sized PC food items.

Technical Abstract: The purpose of this study was to evaluate the reliability of label weights as surrogates for actual weights in commercial portion-controlled foods used in a research setting. Actual weights of replicate samples of 83 portion controlled (PC) food items and 21 discrete units (DU) of food from larger packaging were determined over time. Comparison was made to the package label weights for the PCs and the per-serving weights for the DUs. The study was conducted at USDA's Beltsville Human Nutrition Research Center's Human Study Facility, which houses a metabolic kitchen and human nutrition research facility. The primary outcome measures were the actual and label weights of 101 food items consumed by human volunteers during controlled feeding studies. The difference between label and actual weights was tested by the paired t-test for those data that complied with the assumptions of normality. The Wilcoxon signed rank test was used in the remainder of the data. Compliance with federal guidelines for packaged weights was also assessed. There was no statistical difference between actual and label weights for only 37 food items. The actual weights of 15 PC food items were greater than 1% below label weights, making them potentially out of compliance with federal guidelines. With advance planning and continuous monitoring, well-controlled feeding studies could incorporate PCs and DUs, especially beverages, snack food, and confectionery products. Dietitians should encourage diabetics and others following strict dietary regimens to check actual weights of portion-controlled products carefully against package weights. This data is also important for food processors, manufacturers, and markerters.