A healthy diet should consider environmental impact

We appreciate the description by Mente et al. 1 of a dietary score comprised of fruits, vegetables, nuts, legumes, fish, and dairy foods that are inversely associated with adverse health outcomes.These findings align with other studies, including that associations are strongest in diets dominated by starchy staples typical of many low-income populations.In this context, modest amounts of animal-sourced foods can provide important nutritional value.However, the omission of harmful components in the dietary score, such as trans fat, high sodium, and sugary beverages, is a critical concern.Also, their implication that high intakes of red meat and dairy foods would be without adverse health consequences, especially if compared with plant protein sources, is inconsistent with considerable evidence. 2,3The diversity of the PURE populations is a strength, but in many of these populations, high intake of red meat is recent, and the experience of high-income countries shows that several decades of an atherogenic diet are needed to produce clinical disease (myocardial infarction before age 40 is rare).Also, within the PURE populations, marked variation in socio-economic status likely confounds some findings despite statistical adjustment.In addition, the range of many foods in their analysis was limited; for instance, in the top quintile of their score, intake of milk was only 186 g/day (one glass is ∼250 g), and most of this consumption was in the Americans and Middle East.Mente et al. misrepresent the Planetary Health Diet (PHD) as highly restrictive and low in legumes and fat; in fact, the PHD reference targets for milk equivalents (250 g/day) and legumes (75 g/day dry weight, excluding peanuts) exceed the top quintile of the PURE diet score. 4The targets for other animal-sourced foods provide approximately one serving per day, although red meat is about one serving per week.In the PHD nutritional analysis, total fat intake was 38% of energy, again higher than in PURE (27%), and intended to be flexible.Mente et al.'s scoring of the PHD was misleading because it used only the ranges for individual components, which describe the global variation that appears compatible with health.They ignored the target reference diet numbers, which provide a balanced dietary pattern within a given