Relationship between mastication and malnutrition in community-dwelling older adults: a meta-analysis.

Only three studies were included and all were cross-sectional, so the review can show that chewing problems and malnutrition occur together but not that one causes the other. The studies disagreed with each other (94% heterogeneity), and with so few studies the authors could not run a sensitivity analysis to work out why. The authors rated the certainty of the evidence very low and say themselves the results should be read cautiously. Publication bias could not be checked. Some of the included studies did not account for confounding factors such as socioeconomic status, which affects both access to dental care and nutrition. Chewing was measured four different ways across the studies and there is no agreed standard, which makes the studies hard to compare. Everyone studied was over 60 and living at home; people in care homes or hospital, and studies set in community health centres or primary care, were excluded.

The authors state that there was no financial support for the work and nothing to declare. No competing interests are reported. Because there were too few studies to check for publication bias formally, the authors widened their search to grey literature and consulted an expert for unpublished work instead.

Do not read the 2.21 figure as proof that chewing problems cause malnutrition, or that fixing someone's chewing would improve their diet. The range around that figure runs from 0.93 to 5.26, which includes no association at all, the three studies pointed in different directions, and the authors rated the certainty of the evidence very low.