Intended for healthcare professionals

CCBY Open access

Physical inactivity, cardiometabolic disease, and risk of dementia: an individual-participant meta-analysis

BMJ 2019; 365 doi: (Published 17 April 2019) Cite this as: BMJ 2019;365:l1495

Linked BMJ Opinion

Reflections on null findings: the case of physical activity and risk of dementia

Post-modern epidemiology for an evidence-informed public health

Congratulation to Kivimäki et al for showing that there is probably no effect of physical activity on dementia risk [1]. On the one hand, this is of course a bad news for population health as there was hope that physical activity could help prevent dementia [2]; we will have to find other preventive strategies for dementia. On the other hand, this could be a good news for evidence-informed public health. Indeed, using an interesting data analysis approach, the author could limit the risk of reverse causation, that is, dementia causing a reduction of physical activity. With this approach, the authors could move from a prediction exercise to a true causal inference study. Observational epidemiology aiming to address causal relationship has suffered major drawbacks, notably in the domains of nutrition or environmental science, where numerous findings were eventually not confirmed [3, 4]; the null finding in the study by Kivimäki et al, if confirmed, could reveal a new drawback of observational studies. Such failures are a real danger for public health. To better inform evidence for public health, it is necessary to move toward a post-modern epidemiology, clearly distinguishing description, prediction, and causal inference in observational studies, and taking advantage of new methods in causal analyses [5].

1. Kivimäki M, Singh-Manoux A, Pentti J, Sabia S, Nyberg ST, Alfredsson L, Goldberg M, Knutsson A, Koskenvuo M, Koskinen A, Kouvonen A, Nordin M, Oksanen T, Strandberg T, Suominen SB, Theorell T, Vahtera J, Väänänen A, Virtanen M, Westerholm P, Westerlund H, Zins M, Seshadri S, Batty GD, Sipilä PN, Shipley MJ, Lindbohm JV, Ferrie JE, Jokela M; IPD-Work consortium. Physical inactivity, cardiometabolic disease, and risk of dementia: an individual-participant meta-analysis. BMJ 2019; 365: l1495.
2. Livingston G, Sommerlad A, Orgeta V, Costafreda SG, Huntley J, Ames D, Ballard C, Banerjee S, Burns A, Cohen-Mansfield J, Cooper C, Fox N, Gitlin LN, Howard R, Kales HC, Larson EB, Ritchie K, Rockwood K, Sampson EL, Samus Q, Schneider LS, Selbæk G, Teri L, Mukadam N. Dementia prevention, intervention, and care. Lancet 2017; 390(10113): 2673-34.
3. Chiolero A. Counterfactual and interventionist approach to cure risk factor epidemiology. Int J Epidemiol 2016; 45(6): 2202-3.
4. Ioannidis JP. Why most published research findings are false. PLoS Med 2005; 2(8): e124.
5. Hernán MA, Hsu J, Healy B. A second chance to get causal inference right: A classification of data science tasks. Chance 2019; 32(1):42-9.

Competing interests: No competing interests

19 April 2019
Arnaud Chiolero
Public health physician and epidemiologist
Institute of Primary Health Care (BIHAM), University of Bern