López-Toledo et al. conducted an interesting randomized, parallel-group clinical trial comparing the effects of flaxseed powder on various biochemical markers in adult patients with uncontrolled Type 2 diabetes who are currently taking metformin [1]. If correct, the authors’ data suggest that three months of consuming 16 g of flaxseed per day decreased glucose, total cholesterol, and triglyceride levels [1]. This suggests that flaxseed consumption has an important role in preventing or mitigating disorders of lipid metabolism and glucose disposal. While the results seem promising, these conclusions are not supported by the published analyses because they are based on within-group comparisons rather than between-group tests of statistical significance [2,3].
A Difference in Nominal Significance (DINS) error occurs when within-group analyses are used to draw conclusions about between-group differences, without directly comparing groups [4,5]. Stated simply, a statistically significant change from the baseline measurement (p < 0.05) was observed in the intervention group, but not in the control group (p > 0.05). The article authors then concluded that the intervention changed more than the control without directly comparing the changes. This approach can inflate the type one error rate of 0.05 to as high as 50 percent [2,3,4] with equal sample sizes per group and asymptotically up to 95% when sample sizes are unequal. A more robust approach to compare between group changes would involve the use of an ANCOVA, with the baseline values serving as the covariate [6]. ANCOVA typically has the highest statistical power for comparing changes from baseline [6].
The registered primary outcomes are presented in two separate figures stratified by experimental condition with insufficient detail to estimate effects. These figures depict identical point estimates in the control group and intervention group, which would result in a p-value of 1 for between group comparisons. We contacted the authors several times and requested the raw data so that we could test the reproducibility and evaluate the verifiability of their analyses. After submitting our initial draft of this letter to the editor, the authors provided us with the raw data and the results of their new analysis following our recommendations. We were able to replicate their new results. While, their new results, align with their original conclusions, it is still important to underscore the importance of directly comparing groups to avoid potential DINS errors in the future.
Funding
The investigators are supported in part by the National Institutes of Health, grant numbers R25DK099080, R25HL124208 and P20GM109096.
Conflicts of Interest
David Allison reports that he and or his institution received grants contracts payments, and in-kind donations from numerous government, not for profit, and for profit entities; including food, dietary supplements, and pharmaceutical companies, but none are related to this project. The views expressed are those of the authors and not necessarily the NIH or any other organization.
References
- López-Toledo, S.; Pineda De la Cruz, M.C.; Gutiérrez-Hurtado, I.A.; Gijón-Soriano, A.L.; Martinez-Martínez, E.; Valencia-Santiago, C.; Orellana-Centeno, J.E.; Ramírez-García, S.A.; Pacheco-Cruz, R. Flaxseed Improves Glucose and Lipid Metabolism in Mexican Subjects with Type 2 Diabetes: A Parallel Randomized Clinical Trial. Nutrients 2025, 17, 709. [Google Scholar] [CrossRef] [PubMed]
- Bland, J.M.; Altman, D.G. Comparisons against baseline within randomised groups are often used and can be highly misleading. Trials 2011, 12, 264. [Google Scholar] [CrossRef] [PubMed]
- Bland, J.M.; Altman, D.G. Best (but oft forgotten) practices: Testing for treatment effects in randomized trials by separate analyses of changes from baseline in each group is a misleading approach. Am. J. Clin. Nutr. 2015, 102, 991–994. [Google Scholar] [CrossRef] [PubMed]
- Gelman, A.; Stern, H. The difference between “significant” and “not significant” is not itself statistically significant. Am. Stat. 2006, 60, 328–331. [Google Scholar] [CrossRef]
- Allison, D.B.; Brown, A.W.; George, B.J.; Kaiser, K.A. Reproducibility: A tragedy of errors. Nature 2016, 530, 27–29. [Google Scholar] [CrossRef] [PubMed]
- Vickers, A.J. The use of percentage change from baseline as an outcome in a controlled trial is statistically inefficient: A simulation study. BMC Med. Res. Methodol. 2001, 1, 6. [Google Scholar] [CrossRef] [PubMed]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2026 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license.