Can dementia be prevented?

BMJ 2010; 341 doi: http://dx.doi.org/10.1136/bmj.c4201 (Published 05 August 2010) Cite this as: BMJ 2010;341:c4201
  1. Tobias Kurth, director of research1,
  2. Giancarlo Logroscino, associate professor2
  1. 1INSERM Unit 708—Neuroepidemiology, Hôpital de la Pitié-Salpêtrière, 47 Boulevard de l’Hôpital, 75651 Paris, Cedex 13, France
  2. 2Department of Neurology and Psychiatry, School of Medicine, University of Bari, Policlinico, Piazza Giulio Cesare 11, 70124 Bari, Italy
  1. tobias.kurth{at}upmc.fr

    Modifiable risk factors exist, but targeted public health programmes are not yet warranted

    Cognitive decline and dementia pose a substantial threat to individuals and to public health, and their burden is expected to increase worldwide.1 Because effective treatments are sparse, preventive strategies are needed to delay the onset of dementia or reduce its incidence. However, before such strategies can be implemented, we need to identify which factors should be targeted and in whom.

    In the linked cohort study (doi:10.1136/bmj.c3885), Ritchie and colleagues aim to identify factors related to the development of new cases of dementia and evaluate the impact of the theoretical elimination of such factors in elderly people from France.2 Potential factors for the prevention of dementia were identified by a literature search of population and clinical studies and were broadly divided into sociodemographic, clinical, and environment or lifestyle factors.

    The study enrolled 1433 community dwelling people who lived in the city of Montpellier, France, and who were aged 65 or older at baseline between 1999 and 2001. These people were followed for the development of mild cognitive impairment or dementia for an average of seven years. Factors identified by the literature search that by themselves were associated with new onset dementia in their cohort study were entered into a multivariable model in which only core factors were retained by using a statistical selection technique. Remaining predictors of dementia were crystallised intelligence (which is based on knowledge and experience—for example, vocabulary and general information), depression, fruit and vegetable consumption, diabetes, and the presence of the apolipoprotein E (ApoE) ε4 allele.

    In a final step, the authors calculated the effect of the theoretical elimination of these factors by estimating the attributable fraction at the population level. The results indicate that a theoretical elimination of depression and diabetes and an increase in fruit and vegetable consumption would reduce the total burden of dementia by about 21%. For comparison, a theoretical elimination of the ApoE ε4 allele would lead to a reduction of about 7%.

    The study provides interesting information about the potential contribution of modifiable risk factors to the development of dementia or mild cognitive decline. However, several questions remain, in particular whether prevention programmes should be developed that aim to eliminate these specific factors as well as to whom and for how long such prevention programmes should be applied.

    The attributable fraction is the amount of a disease that can be attributed to a specific risk factor. Thus, eradicating or modifying risk factors with a high attributable fraction should eliminate or substantially reduce the incidence of the disease. However, the authors have identified only hypothetical risk factors for dementia that have not yet been proved to cause the disease. Thus residual confounding is possible—for example, confounding of fruit and vegetable intake by other lifestyle habits and confounding of depression and diabetes by coexisting morbidities. In addition, we have no evidence that modifying these factors would reduce the burden of dementia.

    Also, the population attributable fraction applies to a specific population and time frame. As in many ageing related conditions, the estimation of risk and the association of specific factors with a disease strongly depend on age. This has been shown for the association of body mass index and high blood pressure with mortality.3 4 A risk factor may be harmful at younger ages but be beneficial at older ages.5 For example, hypertension and hypercholesterolaemia in midlife are risk factors for developing dementia, whereas in later in life dementia is associated with lower cholesterol concentrations and lower blood pressure.6 This is also true for non-modifiable risk factors—for example, Apo E genotype does not seem to influence whether subjects will develop Alzheimer’s disease if they survive to a very old age.7

    Ritchie and colleagues’ study may suggest that public health programmes that enhance crystallised intelligence, increase fruit and vegetable consumption, and eliminate or reduce depression and diabetes should be prioritised to reduce the incidence of dementia in people in their early 70s. However, the uncertainty of causality and lack of proof that intervention in these factors will reduce the burden of dementia should caution against strong conclusions. In addition, the outcome of intervention studies on younger or much older people may be quite different, and it is unclear how long such interventions should last. Because resources are limited, research should focus on whether interventions should be targeted at people at higher risk of dementia rather than the whole population. Identification of people at increased risk of cognitive decline in the general population is challenging,8 and Ritchie and colleagues’ results may help to identify people at increased risk of developing dementia, but it is too early to call for prevention programmes that target these specific factors.


    Cite this as: BMJ 2010;341:c4201


    • Research, doi:10.1136/bmj.c3885
    • Competing interests: All authors have completed the Unified Competing Interest form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare: In the past two years TK has received research funding from the French National Research Agency, the National Institutes of Health, Merck, the Migraine Research Foundation, and the Parkinson’s Research Foundation and GL has received research funding from the Amyotrophic Lateral Sclerosis Association, Italian Ministry of Health, the Apulia Region, and special funding from the Italian Ministry of Universities; TK is a consultant to i3 Drug Safety and World Health Information Science Consultants, LLC and has received honorariums from Genzyme, Merck, and Pfizer for educational lectures. GL has received honorariums from Pfizer, Novartis, Glaxo, and Lilly Pharmaceutical for speaking engagements. TK and GL declare no other relationships or activities that could appear to have influenced the submitted work.

    • Provenance and peer review: Commissioned; not externally peer reviewed.