TY - JOUR
T1 - Mortality Associated with Healthy Eating Index Components and an Empirical-Scores Healthy Eating Index in a Cohort of Postmenopausal Women
AU - Prentice, Ross L.
AU - Aragaki, Aaron K.
AU - Van Horn, Linda
AU - Thomson, Cynthia A.
AU - Tinker, Lesley F.
AU - Manson, Jo Ann E.
AU - Mossavar-Rahmani, Yasmin
AU - Huang, Ying
AU - Zheng, Cheng
AU - Beresford, Shirley A.A.
AU - Wallace, Robert
AU - Anderson, Garnet L.
AU - Lampe, Johanna W.
AU - Neuhouser, Marian L.
N1 - Publisher Copyright: © The Author(s) 2022. Published by Oxford University Press on behalf of the American Society for Nutrition. All rights reserved.
PY - 2022/11/1
Y1 - 2022/11/1
N2 - Background: Studies of diet and chronic disease include a recent important focus on dietary patterns. Patterns are typically defined by listing dietary variables and by totaling scores that reflect whether consumption is encouraged or discouraged for listed variables. However, precision may be improved by including total energy consumption among the dietary variables and by scoring dietary variables empirically. Objectives: To relate Healthy Eating Index (HEI)–2010 components and total energy intake to all-cause and cause-specific mortality in Women’s Health Initiative (WHI) cohorts and to define and evaluate an associated Empirical-Scores Healthy Eating Index (E-HEI). Methods: Analyses are conducted in WHI cohorts (n = 67,247) of healthy postmenopausal women, aged 50–79 y, when enrolled during 1993–1998 at 40 US clinical centers, with embedded nutrition biomarker studies. Replicate food-frequency assessments for HEI-2010 ratio variables and doubly labeled water total energy assessments, separated by ∼6 mo, are used as response variables to jointly calibrate baseline dietary data to reduce measurement error influences, using 2 nutrition biomarker studies (n = 199). Calibrated dietary variables are associated with mortality risk, and an E-HEI is defined, using cross-validated HR regression estimation. Results: Of 15 dietary variables considered, all but empty calories calibrated well. Ten variables related significantly (P < 0.05) to total mortality, with favorable fruit, vegetable, whole grain, refined grain, and unsaturated fat associations and unfavorable sodium, saturated fat, and total energy associations. The E-HEI had cross-validated total mortality HRs (95% CIs) of 0.87 (0.82, 0.93), 0.80 (0.76, 0.86), 0.77 (0.72, 0.82), and 0.74 (0.69, 0.79) respectively, for quintiles 2 through 5 compared with quintile 1. These depart more strongly from the null than do HRs for HEI-2010 quintiles, primarily because of total energy. Conclusions: Mortality among US postmenopausal women depends strongly on diet, as evidenced by a new E-HEI that differs substantially from earlier dietary pattern score specifications. J Nutr 2022;152:2493–2504.
AB - Background: Studies of diet and chronic disease include a recent important focus on dietary patterns. Patterns are typically defined by listing dietary variables and by totaling scores that reflect whether consumption is encouraged or discouraged for listed variables. However, precision may be improved by including total energy consumption among the dietary variables and by scoring dietary variables empirically. Objectives: To relate Healthy Eating Index (HEI)–2010 components and total energy intake to all-cause and cause-specific mortality in Women’s Health Initiative (WHI) cohorts and to define and evaluate an associated Empirical-Scores Healthy Eating Index (E-HEI). Methods: Analyses are conducted in WHI cohorts (n = 67,247) of healthy postmenopausal women, aged 50–79 y, when enrolled during 1993–1998 at 40 US clinical centers, with embedded nutrition biomarker studies. Replicate food-frequency assessments for HEI-2010 ratio variables and doubly labeled water total energy assessments, separated by ∼6 mo, are used as response variables to jointly calibrate baseline dietary data to reduce measurement error influences, using 2 nutrition biomarker studies (n = 199). Calibrated dietary variables are associated with mortality risk, and an E-HEI is defined, using cross-validated HR regression estimation. Results: Of 15 dietary variables considered, all but empty calories calibrated well. Ten variables related significantly (P < 0.05) to total mortality, with favorable fruit, vegetable, whole grain, refined grain, and unsaturated fat associations and unfavorable sodium, saturated fat, and total energy associations. The E-HEI had cross-validated total mortality HRs (95% CIs) of 0.87 (0.82, 0.93), 0.80 (0.76, 0.86), 0.77 (0.72, 0.82), and 0.74 (0.69, 0.79) respectively, for quintiles 2 through 5 compared with quintile 1. These depart more strongly from the null than do HRs for HEI-2010 quintiles, primarily because of total energy. Conclusions: Mortality among US postmenopausal women depends strongly on diet, as evidenced by a new E-HEI that differs substantially from earlier dietary pattern score specifications. J Nutr 2022;152:2493–2504.
KW - Healthy Eating Index
KW - all-cause mortality
KW - cancer
KW - cardiovascular disease
KW - dietary variables
KW - hazard ratio
KW - measurement error
KW - multivariate calibration
UR - http://www.scopus.com/inward/record.url?scp=85144936519&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85144936519&partnerID=8YFLogxK
U2 - 10.1093/jn/nxac068
DO - 10.1093/jn/nxac068
M3 - Article
C2 - 35289897
SN - 0022-3166
VL - 152
SP - 2493
EP - 2504
JO - Journal of Nutrition
JF - Journal of Nutrition
IS - 11
ER -