Observational research rigour alone does not justify causal inference

• DOI: 10.1111/eci.12681
• Publication Date: 2016-10-13T01:08:16Z
• Authors: Keisuke Ejima; Peng Li; Daniel L. Smith; Tim R. Nagy; Inga Kadish; Thomas van Groen; John A. Dawson; Yongbin Yang; Amit Patki; David B. Allison
• Keywords: Rigour

Abstract Background Differing opinions exist on whether associations obtained in observational studies can be reliable indicators of a causal effect if the study is sufficiently well controlled and executed. Materials methods To test this, we conducted two animal that were rigorously executed beyond what achieved humans. In 1, randomized 332 genetically identical C57 BL /6J mice into three diet groups with differing food energy allotments recorded individual self‐selected daily intake lifespan. 2, 60 male ( CD 1) paired divided for 2‐week feeding regimen. We evaluated association between weight gain consumption. Within each pair, one was randomly assigned to an S group which animals had free access food. The second (R group) provided exactly same their partner ate day before. Results across all groups, found significant negative However, positive lifespan among ad libitum group: 29·99 (95% CI : 8·2–51·7) days per kcal. P = 0·003) (randomized vs. self‐selected)‐by‐food consumption interaction gain. Conclusion At least nutrition research, derived from may not effects, even most rigorous designs achievable.