PURE Healthy Diet Score validated

View on the News

Despite validation, questions remain

Dr. Mente and his associates have validated the PURE Healthy Diet Score. However, it remains unclear whether the score captures all of the many facets of diet, and it’s also uncertain whether the score is sensitive to changes in diet.

Dr. Eva Prescott, cardiologist, Bispepjerg Hospital, Copenhagen Mitchel L. Zoler/MDedge News

Dr. Eva Prescott

The researchers developed the PURE Healthy Diet Score with data from PURE, a large, international study. Their findings were controversial when first reported in 2017. Controversy arose over at least three of their findings: Decreased mortality was linked with increased consumption of saturated fat from dairy and red meat; higher scores did not correlate with a significant effect on cardiovascular disease in the derivation study; and the benefit from fruits and vegetables in the diet hit a plateau with an intake of about four daily servings. Their finding that decreased mortality linked with an increased intake of saturated fat ran counter to expectations.

Another issue with the quintile analysis that the researchers used to derive the formula was that the spread between the median scores of the bottom, worst-outcome quartile and the top, best-outcome quartile was only 7 points on a scale that ranged from 7 to 35. The small magnitude of the difference in scores between the bottom and top quintiles might limit the discriminatory power of this scoring system.

Eva Prescott, MD, is a cardiologist at Bispebjerg Hospital in Copenhagen. She has been an advisor to AstraZeneca, NovoNordisk, and Sanofi. She made these comments as designated discussant for the report.



  • The ONTARGET and TRANSCEND studies (N Engl J Med. 2008 Apr 10;358[15]:1547-58), which together included diet and outcomes data for 31,546 patients with vascular disease. Diet analysis and scoring showed that enrolled people in the quintile with the highest score had a statistically significant 24% relative reduction in mortality, compared with the quintile with the worst score after adjusting for measured confounders.
  • The INTERHEART study (Lancet. 2004 Sep 11;364[9438]:937-52), which had data for 27,098 people and showed that the primary outcome of incident MI was a statistically significant 22% lower after adjustment in the quintile with the best diet score, compared with the quintile with the worst score.
  • The INTERSTROKE study (Lancet. 2016 Aug 20;388[10046]:761-75), with data for 20,834 people, showed that the rate of stroke was a statistically significant 25% lower after adjustment in the quintile with the highest diet score, compared with those with the lowest score.

Dr. Mente had no financial disclosures.

Next Article: