Older adults assigned daily walnuts scored modestly lower on a dietary inflammation index

Assignment is the strength of the design. The paper presents these results as ancillary to a randomized controlled trial ^[12], so differences at two years are less likely to come from people choosing walnuts for themselves.

The measurement is weaker. Diet came from recall: on five occasions over two years, participants described everything they ate and drank in the preceding 24 hours ^[6]. Five recalls of 24 hours each cover five days of eating ^[13], which is under 1 percent of the roughly 730 days of the trial ^[14]. The participants had been told either to eat walnuts or to avoid them ^[3][4], and beyond those instructions they were not asked to overhaul how they ate ^[5]. They knew what the study expected of them. Walnuts are also part of what the index scores ^[8], so we'd expect some of the gap to come from the nuts themselves.

The thing this doesn't tell you is how big the shift was. Discover Magazine reports only that the walnut group scored modestly but statistically significantly lower ^[7], and its account omits the scores, the between-group difference and any confidence interval. Senior investigator Sujatha Rajaram said in a press release: "The important message is that incorporating walnuts into the diet was associated with a shift toward a less inflammatory dietary pattern." ^[10]

In our view the result supports a narrow statement. In a randomized comparison, a daily walnut habit moved the estimated inflammatory potential of older adults' reported diets by a small, significant amount. Discover's own summary is that the findings suggest potential benefits worth investigating ^[16].