|SHAMS-WHITE, MARISSA - Tufts University|
|CHUI, KENNETH - Tufts University|
|DEUSTER, PATRICIA - Uniformed Services University|
|MCKEOWN, NICOLA - Jean Mayer Human Nutrition Research Center On Aging At Tufts University|
|MUST, AVIVA - Tufts University|
Submitted to: Nutrients
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/16/2019
Publication Date: 1/23/2019
Citation: Shams-White, M.M., Chui, K., Deuster, P.A., McKeown, N.M., Must, A. 2019. Investigating items to improve the validity of the five-item Healthy Eating Score compared with the 2015 Healthy Eating Index in a military population. Nutrients. 11(2):251. https://doi.org/10.3390/nu11020251.
Interpretive Summary: To quickly assess the overall diet quality of military personnel, military researchers currently utilize five questions within a large questionnaire called the Global Assessment Tool to estimate healthy eating scores (HES-5), with a higher score indicating better diet quality. This study aimed to improve the HES-5's performance compared to the 2015 Healthy Eating Index (HEI-2015) in 333 active duty military personnel by adding various SSB intake, breakfast, and recovery snack consumption questions alone and in combination to the HES-5. The best improvement to all participants' HES-5 scores was seen by adding both a SSB question (8-oz servings) and post-exercise recovery fueling snack question to the HES-5. Future work should consider scoring mechanisms for the items, serving sizes, and question wording.
Technical Abstract: Military researchers utilize a 5-item healthy eating score (HES-5) in the Global Assessment Tool (GAT) questionnaire to quickly assess the overall diet quality of military personnel. This study aimed to modify the HES-5 to improve its validity relative to the 2015 Healthy Eating Index (HEI-2015) in active duty military personnel (N=333). A food frequency questionnaire was used to calculate HEI-2015 scores and to assess sugar-sweetened beverage (SSB) intake in 8-oz (SSB-8) and 12-oz servings. GAT nutrition questions were used to calculate HES-5 scores and capture breakfast and post-exercise recovery fueling snack (RFsnack) frequencies. Two scoring options were considered for the highest RFsnack category: "4" vs. "5" (RFsnack-5). Potential candidates were added alone and in combination to the HES-5 and compared to the HEI-2015 with a Pearson correlation coefficient. Scores with the highest correlations were compared via a Z-score equation to identify the simplest modification to the HES-5. Correlations between HES-5 and HEI-2015 scores in total participants, males, and females were 0.41, 0.45 and 0.32, respectively. Correlations were most significantly improved in total participants by adding RFsnack-5, SSB-8, RFsnack-5+SSB-8, and RFsnack-5+SSB-8+breakfast, though the addition of SSB-8+RFsnack-5 performed best (r=0.53). Future work should consider scoring mechanisms, serving sizes, and question wording.