NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Williams JW, Plassman BL, Burke J, et al. Preventing Alzheimer's Disease and Cognitive Decline. Rockville (MD): Agency for Healthcare Research and Quality (US); 2010 Apr. (Evidence Reports/Technology Assessments, No. 193.)

  • This publication is provided for historical reference only and the information may be out of date.

This publication is provided for historical reference only and the information may be out of date.

5Conclusions

Among the many factors examined in this review, only some are amenable to being evaluated in randomized controlled trials (RCTs), and only a subset of these have actually been studied in high-quality RCTs as potential interventions for preventing or delaying the onset of Alzheimer’s disease (AD) and cognitive decline. Effects of interventions in important subgroups, such as minority populations, were evaluated infrequently. A few of the factors considered in this report have shown potential promise in observational studies for both AD and cognitive decline, and in RCTs for at least one of the outcomes of interest. Moreover, several of the factors reviewed have demonstrated benefits beyond the potential of preserved cognition; that is, they promote overall health. Thus, there may be other reasons to recommend an intervention (e.g., increased physical activity) while further research is completed on its role in cognition.

The most general conclusions of this evidence report are summarized in Tables 74 and 75. These conclusions are based on a systematic review of the evidence for each factor, and on judgments about the quality of that evidence made using principles developed by the GRADE working group (www.gradeworkinggroup.org). For each factor examined, we considered the entire body of evidence and summarized the quality of that evidence as low, moderate, or high. The GRADE approach assigns an initial rating of “low” quality to observational studies and “high” quality to RCTs. These initial ratings may be modified by considerations relating to: detailed study design, consistency, strength of association, dose-response effect, directness, precision, and consideration of all plausible residual confounders that could reduce a demonstrated effect. Note that even within a given rating level, the quality of evidence may vary substantially; for example, there is considerable variability within the “low” quality level.

Tables 74 and 75 list, for AD and cognitive decline, respectively, the potential risk factors and interventions considered in this report, the associations observed between them and the outcome of interest (if any), and the quality of evidence supporting those associations. The tables also list factors for which the evidence was insufficient to establish whether or not an association exists. It is noteworthy that this last category includes many of the risk factors examined in this report.

In addition to sparse evidence, the extant research literature has other important limitations. Needed advances in study design and reporting include validated measures of exposure, pre-specified exposure categorizations, longer term trials, reporting of power calculations, and an agreed-upon battery of cognitive measures. Improving research design and reporting in these and other ways could improve confidence in observed associations and targeting of potential interventions. Conducting trials initially in those at high risk (e.g., those with mild cognitive impairment) would be an efficient approach. Well-designed, long-term cohort studies with robust measures of exposure and cognitive outcomes are needed to address the factors for which there is a strong biological mechanism or preliminary clinical evidence to suggest an important association.

Tables

Table 74Summary of findings on potential risk factors and interventions for AD

Direction of associationFactorsLevel of evidence
Increased risk
  • APOE e4 genotype
  • Conjugated equine estrogen with methyl progesterone*
Moderate
  • Some non-steroidal anti-inflammatory drugs*
  • Depressive disorder
  • Diabetes mellitus
  • Hyperlipidemia in mid-life
  • Traumatic brain injury in males
  • Pesticide exposure
  • Never married, less social support
  • Current tobacco use
Low
Decreased risk
  • Mediterranean diet
  • Folic acid
  • HMG-CoA reductase inhibitors (statins)
  • Higher levels of education
  • Light to moderate alcohol intake
  • Cognitively engaging activities
  • Physical activity, particularly high levels
Low
No association
  • Ginkgo biloba*
High
  • Vitamin E*
  • Cholinesterase inhibitors*
Moderate
  • Anti-hypertensive medication*
  • Conjugated equine estrogen
  • Omega-3 fatty acids*
  • Vitamins B12, C, beta-carotene
  • Homocysteine
  • Hypertension
  • Obesity
  • Metabolic syndrome
  • Early childhood factors
  • Occupational level
  • Lead
Low
Inadequate evidence to assess association
  • Saturated fat intake
  • Fruit and vegetable intake
  • Trace metals
  • High caloric intake
  • Memantine
  • Sleep apnea
  • Anxiety disorders
  • Resiliency
  • Non-cognitive, non-physical leisure activities
  • Agent Orange, Gulf War Syndrome
  • Solvents, aluminum
  • Genetic factors other than APOE
(Not applicable)
*

Data from observational studies and RCTs.

Abbreviations: APOE = apolipoprotein E gene; APOE e4 = epsilon 4 allele of the apolipoprotein E gene; HMG-CoA = 3-hydroxy-3-methylglutaryl-coenzyme A; RCTs = randomized controlled trials

GRADE criteria (see text)

Table 75Summary of findings on potential risk factors and interventions for cognitive decline

Direction of associationFactorsLevel of evidence
Increased risk
  • APOE e4 genotype
  • Low plasma selenium
  • Depressive disorder
  • Diabetes mellitus
  • Metabolic syndrome
  • Current tobacco use
Low
Decreased risk
  • Cognitive training*
High
  • Vegetable intake
  • Mediterranean diet
  • Omega-3 fatty acids*
  • Physical activity*
  • Non-cognitive, non-physical leisure activities
Low
No association
  • Vitamin C, Vitamin E, beta-carotene supplements*
  • Conjugated equine estrogen*
  • HMG-CoA reductase inhibitors (statins)*
High
  • Aspirin*
  • Dehydroepiandosterone*
  • Cholinesterase inhibitors*
  • Multivitamin supplement*
  • Vitamins B6, B12 and folic acid supplements*
Moderate
  • Alcohol intake
  • Non-steroidal anti-inflammatory drugs*
  • Anti-hypertensive medication*
  • Homocysteine
  • Hyperlipidemia
  • Anxiety disorders
  • Hypertension
  • Obesity
  • Early childhood factors
  • Higher levels of education
  • Social network, social supports
Low
Inadequate evidence to assess association
  • Trace metals
  • Fat intake
  • High caloric intake
  • Gingko biloba*
  • Memantine
  • Sleep apnea
  • Resiliency
  • Occupational level
  • Traumatic brain injury
  • Toxic environmental exposures
  • Agent Orange, Gulf War Syndrome
  • Genetic factors other than APOE
(Not applicable)
*

Data from observational studies and RCTs.

Not associated with decreased risk but may be associated with increased risk.

Abbreviations: APOE = apolipoprotein E gene; APOE e4 = epsilon 4 allele of the apolipoprotein E gene; HMG-CoA = 3-hydroxy-3-methylglutaryl-coenzyme A; RCTs = randomized controlled trials

GRADE criteria (see text)