Development and Evaluation of a Planetary Health Diet Index: The Mediators of Atherosclerosis in South Asians Living in America Study.

backgroundIn 2019, the EAT-Lancet Commission introduced the Planetary Health Diet (PHD), aiming to enhance both human health and global environmental sustainability. Although a few indexes have been developed based on the PHD for different populations, there is currently no index tailored specifically for the South Asian population.

objectiveThis study aimed to develop and validate a Planetary Health Diet Index (PHDI) based on the EAT-Lancet PHD in a prospective South Asian cohort based in the United States.

designAnalysis was conducted using cross-sectional dietary data obtained from a validated, ethnic-specific semiquantitative food frequency questionnaire administered to adults in the Mediators of Atherosclerosis in South Asians Living in America study at a baseline visit between 2010 and 2013. PARTICIPANTS AND

settingA total of 906 participants were enrolled at baseline, recruited from Chicago and the San Francisco Bay Area, in this community-based prospective cohort. Eligible participants were individuals of South Asian ancestry (defined as having 3 or more grandparents born in India, Pakistan, Nepal, Bangladesh, or Sri Lanka) with no history of cardiovascular disease, aged between 40 and 84 years, and proficient in English, Urdu, or Hindi. The analytic sample included a total of 891 participants without missing data or implausible energy intakes.

main outcome measuresThe main outcome measure was the PHDI score and its associations with assessments of construct validity and internal reliability. STATISTICAL ANALYSES: Linear regression models were used to examine the construct validity of the PHDI, as well as correlations between the index and the Alternative Healthy Eating Index and the dietary carbon footprint. Analysis of variance was used to evaluate the concurrent-criterion validity and principal components analysis to assess multidimensionality. Pearson correlations were estimated between the PHDI components and energy, and Cronbach's coefficient α was determined to assess internal consistency.

resultsThe PHDI was composed of 15 food components with intake ranges as prescribed by the 2019 EAT-Lancet PHD and scored proportionally 0 to 10 for a total possible score between 0 and 140. PHDI scores ranged from 51 to 119 out of 140 possible points, with a mean score of 89. PHDI was positively associated with plant-derived nutrients such as vegetable protein (P < .0001), total fiber (P < .0001), vitamins A (P < .006) and C (P < .012), and negatively associated with animal-sourced nutrients such as animal protein, cholesterol, and vitamin B12 (P < .0001 for all), amongst others. Concurrent criteria validity revealed significant differences among subgroups: current smokers scored lower (85) than nonsmokers (89) (P = .03); those with bachelor's degrees scored higher (89.3) than those without (85.4) (P < .0001); higher family income correlated with higher scores (89.7 vs 86.7; P < .0001); and participants with type 2 diabetes had higher scores (90.8) than those with prediabetes (88) (P = .002) or normal glycemia (88.4) (P < .008). Each 5-unit increase in PHDI score was associated with a higher Alternative Healthy Eating Index score (partial correlation: r = 0.7; P < .0001) and a 7% lower dietary carbon footprint (P < .001). Cronbach's α for internal reliability was .68.

conclusionsThe findings provided support for both the construct validity and reliability of the PHDI in this cohort of American South Asians. This index is suitable for assessing diets in relation to the 2019 EAT-Lancet dietary guidelines.