A US‐Based Epidemiological Analysis Reveals Large Discordance Between Exclusive Dietary Recall‐Based Vegetarian Status Assessment and Self‐Identification in a Nationally Representative Population‐Based Survey
Maximilian Andreas Storz, Roman Huber, Christian KoederPurpose
The greater the flexibility in definitions and concepts, the less likely research findings are to be true. Vegetarian status assessment in epidemiological research is a matter of ongoing controversy, and reported vegetarian prevalence is highly variable depending on the employed assessment methods. A systematic comparison of these potential vegetarian status assessment methods is currently unavailable.
Methods
Using the National Health and Nutrition Examination Surveys (NHANES) data, we examined the vegetarianism prevalence in the United States (2007–2010) using various approaches, including 24‐h dietary recall data only, self‐identification by participants, and various combinations of these approaches.
Results
Large differences and discrepancies were observed: The estimated vegetarian prevalence based on a single 24‐h dietary recall was roughly 10 times higher than the estimated prevalence based on a combination of both self‐identification and 24‐h dietary recall data (14.28% [95%‐CI: 13.37–15.24] vs. 1.1% [95%‐CI: 0.77–1.58]). The estimated vegetarian prevalence based on self‐identification alone was twice as high as compared to a combination of both self‐identification and dietary recall data (2.2% [95%‐CI: 1.85–2.60]).
Conclusions
Our findings suggest that epidemiological studies which do not specifically query vegetarian status (and instead rely exclusively on 24‐h‐dietary recalls) may overestimate vegetarian prevalence and are likely subject to misclassification. While self‐identification alone may also result in vegetarian prevalence inflation, our data suggest that the combination of self‐identification and dietary recall data may deliver the most conservative vegetarian prevalence estimate in this scenario.