In 2013, a group of researchers published a paper with results on a randomized trial of the Mediterranean diet, finding that it appears to have significant health benefits. Today, they retracted their original paper because of statistical errors and published a re-analysis of their data that corrects those errors, and still finds significant health benefits to the Mediterranean diet.
However, it appears some experts are not persuaded; the New York Times had an article today quoting multiple researchers as saying that they remain skeptical and are not convinced the re-analysis fully addresses the statistical issues. (The article also has an interesting summary of some of the challenges and issues with the original paper.)
What are the remaining concerns with the statistical analysis of the updated paper?