Premium
A Validation of Automated Self‐Administered 24‐Hour Dietary Recalls(ASA24) Relative to Interviewer‐Administered Recalls using the Nutrition Data System for Research (NDSR)
Author(s) -
Mitchell Diane C,
Cheng Feon W,
Still Christopher D,
Jensen Gordon L
Publication year - 2016
Publication title -
the faseb journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.709
H-Index - 277
eISSN - 1530-6860
pISSN - 0892-6638
DOI - 10.1096/fasebj.30.1_supplement.43.3
Subject(s) - interview , medicine , test (biology) , demography , telephone interview , environmental health , gerontology , paleontology , social science , sociology , political science , law , biology
The ASA24, an innovative internet‐based tool widely used by researchers, may be a valid and reliable alternative in research when cost or time constraints do not allow for telephone 24‐hr dietary recalls or when food frequency data is insufficient for the level of detail needed. Few studies have been conducted to evaluate, validate or test the feasibility and cost‐effectiveness of the ASA24 relative to standard methods in different populations. Our objective was to assess the relative validity of the ASA24 in middle aged adults residing in rural Pennsylvania. Methods Adult men (n= 60) and women (n= 108) 50–65 years of age (mean age 57 ± 4.5 yrs.) were recruited and randomly assigned using a cross‐over design to complete both the internet‐based ASA24 and telephone interviewer‐administered NDSR assessments. Participants were either emailed or called on randomly selected, unannounced days to complete 1 weekend and 2 weekdays of intake. Of the 181 participants who were recruited, 3 participants dropped, 9 (5%) did not complete the ASA24 and 1 did not complete the NDSR assessments leaving 168 participants with complete assessments for both methods. Relative validity was assessed by Spearman correlation for energy and energy adjusted and deattenuated correlations for nutrients and food groups. Results The ASA24 yielded higher energy intakes (1966 vs. 1853 kcal) with a correlation of 0.67. Correlations were higher for women (r=0.58) than men (r=0.48). Correlations coefficients for nutrients ranged from 0.21 to 0.84 for all nutrients examined. Correlations were lowest for polyunsaturated fat (r=0.21) and for zinc (r=0.33) and highest for fiber (r=0.84), vitamin A (r=0.80), magnesium (r=0.84), vitamin C (r=0.78) and potassium (r=0.78). For food groups correlations were highest for whole fruit (r=0.99) and total fruit (r=0.92) and lowest for total protein foods (r=0.48), total grains (r =0.61), and vegetables (r=0.66). Conclusion This study provides inconclusive evidence of the validity of the ASA24 relative to standard telephone methods with relatively high correlations for some nutrients and food groups and low correlations for others. The use of the ASA24 may largely depend on specific research questions and on nutrient and food outcomes of interest. Further research will improve our understanding of the validity of ASA24 and its role in replacing standard interviewer‐administered methods. Support or Funding Information Supported by the United States Department of Agriculture, #8050‐51530‐012‐01A.