Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Comment
  • Published:

Post hoc subgroup analysis and identification—learning more from existing data

European Journal of Clinical Nutrition volume 77, pages 843–844 (2023)Cite this article

Subjects

Currently, trials in nutrition often attempt to detect relatively small benefits in the presence of relatively large and often somewhat unpredictable between-participant variation.

Therefore, unsurprisingly, in some cases such trials will lead to non-significant overall results (“null results”) even though they were designed and indeed powered to be large enough to detect a significant effect. For instance, one study found no significant difference in metabolic syndrome score between a Nordic diet and standard Western diet (823 school children at different schools in Denmark) and another study found no significant difference in three-year incidence of type 2 diabetes between moderate and high protein diets (2223 participants in several European countries) [1, 2].

This is a preview of subscription content, access via your institution

Relevant articles

Open Access articles citing this article.

Access options

Buy this article

  • Purchase on Springer Link
  • Instant access to full article PDF
Buy now

Prices may be subject to local taxes which are calculated during checkout

References

  1. Damsgaard CT, Dalskov S-M, Laursen RP, Ritz C, Hjorth MF, Lauritzen L, et al. Provision of healthy school meals does not affect the metabolic syndrome score in 8-11-year-old children, but reduces cardiometabolic risk markers despite increasing waist circumference. Br J Nutr. 2014;112:1826–36.

    Article  CAS  PubMed  Google Scholar 

  2. Raben A, Vestentoft PS, Brand-Miller J, Jalo E, Drummen M, Simpson L, et al. PREVIEW—Results from a 3-year randomised 2 x 2 factorial multinational trial investigating the role of protein, glycemic index and physical activity for prevention of type-2 diabetes. Diabetes Obes Metab. 2021;23:324–37. https://doi.org/10.1111/dom.14219.

    Article  CAS  PubMed  Google Scholar 

  3. Holzapfel C, Waldenberger M, Lorkowski S, Daniel H. Genetics and epigenetics in personalized nutrition: evidence, expectations, and experiences. Mol Nutr Food Res. 2022;66:2200077. https://doi.org/10.1002/mnfr.202200077.

    Article  CAS  Google Scholar 

  4. Alosh M, Fritsch K, Huque M, Mahjoob K, Pennello G, Rothmann M, et al. Statistical considerations on subgroup analysis in clinical trials. Stat Biopharmaceut Res. 2015;7:286–303. https://doi.org/10.1080/19466315.2015.1077726.

    Article  Google Scholar 

  5. Ondra T, Dmitrienko A, Friede T, Graf A, Miller F, Stallard N, et al. Methods for identification and confirmation of targeted subgroups in clinical trials: a systematic review. J Biopharmaceut Stat. 2016;26:99–119.

    Article  Google Scholar 

  6. Wang R, Lagakos SW, Ware JH, Hunter DJ, Drazen JM. Statistics in medicine—reporting of subgroup analyses in clinical trials. N Engl J Med. 2007;357:2189–94.

    Article  CAS  PubMed  Google Scholar 

  7. Ruberg SJ, Chen L, Wang Y. The mean does not mean as much anymore: finding sub-groups for tailored therapeutics. Clin Trials. 2010;7:574–83.

    Article  PubMed  Google Scholar 

  8. Goldstein BA, Rigdon J. Using machine learning to identify heterogeneous effects in randomized clinical trials—Moving beyond the forest plot and into the forest. JAMA Netw Open. 2019;2:e190004. https://doi.org/10.1001/jamanetworkopen.2019.0004.

    Article  PubMed  Google Scholar 

  9. Royston P, Sauerbrei W. Interactions between treatment and continuous covariates: a step toward individualizing therapy. J Clin Oncol. 2008;26:1397–9.

    Article  PubMed  Google Scholar 

  10. Ritz C, Astrup A, Larsen TM, Hjorth MF. Weight loss at your fingertips: personalized nutrition with fasting glucose and insulin using a novel statistical approach. Eur J Clin Nutr. 2019;73:1529–35.

    Article  CAS  PubMed  Google Scholar 

  11. Ferrario PG, Watzl B, Ritz C. The role of baseline serum 25(OH)D concentration for a potential personalized vitamin D supplementation. Eur J Clin Nutr. 2022;76:1624–9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  12. Matsouaka RA, Li J, Cai T. Evaluating marker-guided treatment selection strategies. Biometrics. 2014;70:489–99.

    Article  PubMed  PubMed Central  Google Scholar 

  13. Fu H, Zhou J, Faries DE. Estimating optimal treatment regimes via subgroup identification in randomized control trials and observational studies. Stat Med. 2016;35:3285–302.

    Article  PubMed  Google Scholar 

  14. Vistisen D, Witte DR, Tabák AG, Herder C, Brunner EJ, Kivimäki M, et al. Patterns of obesity development before the diagnosis of type 2 diabetes: the Whitehall II cohort study. PLoS Med. 2014;11:e1001602.

    Article  PubMed  PubMed Central  Google Scholar 

  15. Aris IM, Rifas-Shiman SL, Li LJ, Kleinman KP, Coull BA, Gold DR, et al. Patterns of body mass index milestones in early life and cardiometabolic risk in early adolescence. Int J Epidemiol. 2019;48:157–67.

    Article  PubMed  PubMed Central  Google Scholar 

  16. Cai T, Tony Cai T, Guo Z. Optimal statistical inference for individualized treatment effects in high-dimensional models. J R Stat Soc Ser B. 2021;83:669–719.

    Article  Google Scholar 

  17. Lu W, Zhang HH, Zeng D. Variable selection for optimal treatment decision. Stat Methods Med Res. 2013;22:493–504.

    Article  PubMed  Google Scholar 

  18. Tian L, Alizadeh AA, Gentles AJ, Tibshirani R. A simple method for estimating interactions between a treatment and a large number of covariates. J Am Stat Assoc. 2014;109:1517–32.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  19. Du Y, Chen H, Varadhan R. Lasso estimation of hierarchical interactions for analyzing heterogeneity of treatment effect. Stat Med. 2021;40:5417–33.

    Article  PubMed  Google Scholar 

  20. Zhang J, Cavallari JM, Fang SC, Weisskopf MG, Lin X, Mittleman MA, et al. Application of linear mixed-effects model with LASSO to identify metal components associated with cardiac autonomic responses among welders: a repeated measures study. Occup Environ Med. 2017;74:810–5.

    Article  PubMed  Google Scholar 

  21. Wen Y, Lu Q. Multikernel linear mixed model with adaptive lasso for complex phenotype prediction. Stat Med. 2020;39:1311–1327.

    Article  PubMed  PubMed Central  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. National Institute of Public Health, University of Southern Denmark, Copenhagen, Denmark

    Elizabeth Mannion & Christian Ritz

  2. Institut für Physiologie und Biochemie der Ernährung, Max Rubner-Institut, Karlsruhe, Germany

    Paola G. Ferrario

Authors
  1. Elizabeth Mannion
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Christian Ritz
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Paola G. Ferrario
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

EM: Revision of the manuscript. PGF: Comments to the manuscript. CR: Initiation, comments and revision of the manuscript. All authors have read and approved the manuscript.

Corresponding author

Correspondence to Christian Ritz.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Mannion, E., Ritz, C. & Ferrario, P.G. Post hoc subgroup analysis and identification—learning more from existing data. Eur J Clin Nutr 77, 843–844 (2023). https://doi.org/10.1038/s41430-023-01297-5

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s41430-023-01297-5

This article is cited by

Search

Advanced search

Quick links