ABSTRACT
A systematic search process was undertaken of dietary factors in relation to a measure of adiposity. Searches of the PubMed database were conducted using the search terms listed in Table 2.1. The following limits were applied to all searches: publication date from January 1992 through March 2003, English, and human. The additional limits, “randomized controlled” or “clinical trial,” were applied to the searches for prevention trials using the terms “overweight prevention OR obesity prevention.” However, not all trials identified were randomized and/or controlled; studies did not have to be randomized, controlled trials to be included. Abstract lists generated by the searches were reviewed by at least one, and not more than three, reviewers (OPN members) to determine which studies would be included. Additional studies on children were obtained from a previously conducted review (Ritchie et al., 2001). Articles identified through the formal search process were also supplemented with studies referenced in the other articles or previously known to the reviewers. Previously unpublished data for secular trends were obtained directly from the source. Data regarding food supply trends were obtained from the United States Department of Agriculture (USDA) web site. Data from the National Food Consumption Survey/Continuing Survey of Food Intake by Individuals (NFCS/CSFII) were obtained directly from USDA, and food intake data from NHANES were obtained directly from CDC. In all cases, these data were nationally representative. Obesity prevention trials were also identified by Yale’s SIP-08 project: Evidence-Based Guidelines for Obesity Prevention and Control. Regardless of the source of studies, the same inclusion and exclusion criteria were applied in selecting articles for review.
