Menopause Live - IMS Updates

Date of release: 21 April, 2014

Education and cognitive reserve

It is often observed that people who are more intelligent, better educated, have intellectually challenging occupations, and engage in mentally stimulating leisure activities face lower risk of Alzheimer’s disease. The concept of cognitive reserve is used as an explanatory mechanism for these phenomena [1]. The inference is that certain attainments and activities (e.g. literacy or professional employment) result in greater capacity, efficiency or redundancy in terms of brain areas and neural pathways used when a cognitive task is performed. In popular usage, cognitive reserve represents one aspect of the adage, 'use it or lose it'. There is no direct evidence that activities associated with enhanced reserve directly affect the pathological burden of Alzheimer’s disease [2]; rather the added capacity appears to act as a buffer against the accumulating pathologies of age-associated disorders that can eventuate in dementia. Interestingly, once symptoms of dementia have emerged, cognitive decline proceeds more rapidly among those with higher cognitive reserve, presumably because the pathological burden is by this time greater.
Rapp and colleagues [3] recently examined the cognitive reserve hypothesis using demographic, cognitive, and MRI brain imaging data from former participants in the Women’s Health Initiative (WHI) clinical trials, who were without dementia at the time of baseline testing. Education served as a proxy for cognitive reserve. Two predictions were derived from the cognitive reserve hypothesis. First, cognition would decline more rapidly among poorly educated women, particularly in the presence of brain pathology. Second, among women who reached threshold for cognitive impairment, the rate of subsequent decline among poorly educated women would actually be reduced in comparison to the rate among better educated women. As described below, study results provided modest support for these predictions.
Participants were from the Magnetic Resonance Imaging substudy of the WHI Memory Study (WHIMS-MRI), derived from the WHI Study of Cognitive Aging, in turn derived from the WHI Memory Study, derived ultimately from the WHI hormone trial cohort. Structural MRI scans were used to infer brain pathology. Scans were obtained once, on average 8 years after WHI enrollment, when women were 70–89 years of age and had completed their randomized hormone interventions.  
MRI imaging was used to define two aspects of brain pathology. The first was the volume of predominately white matter change, thought to reflect ischemic injury, sometimes viewed as non-specific but associated with hypertension and other cardiovascular risk factors. The second was brain atrophy, calculated as the difference between total intracranial volume and total brain volume. Global cognitive function was assessed annually with the 15-item Modified Mini-Mental State (3MS) examination.
To examine the first prediction, women were classified according to MRI findings. The high MRI-pathology group consisted of 381women with ischemia volumes above the median plus brain atrophy above the median; 380 women with below median values on both imaging measures were classified as having low MRI-pathology; and the remaining 629 women were in the intermediate MRI-pathology group. Within each MRI-pathology group, investigators then examined the slope of 3MS decline among the three education subgroups. These slopes differed within the high MRI-pathology group, where decline was steepest for women who failed to complete high school (p = 0.03). Rates of decline did not vary by educational level among women with low (p = 0.31) or intermediate (p = 0.34) MRI-pathology. The interaction between education and MRI-pathology groups was not itself statistically significant.
The second prediction focused on 461 women who converted from no cognitive impairment either to mild cognitive impairment or to dementia. These women were dichotomized by education (high school or less, versus more than high school). During the 3 years prior to conversion, rates of decline were similar in the two education groups. During the 3 years after conversion, the rate of decline was reduced in the low-education group (p < 0.001).  
Of incidental note, hormone therapy usage during the course of the WHI clinical trial was unrelated to MRI-pathology in these analyses.


The strengths of these analyses lay in the prospective design and well-validated methods for identifying mild cognitive impairment and dementia. As noted by study authors, structural MRI outcomes based on ischemic changes and overall brain atrophy may be relatively insensitive to Alzheimer changes or other pathologies strongly linked to cognitive impairment. Perhaps for this reason, the interaction between education and MRI-pathology groups was not statistically significant, nor were differences in the 3MS slopes of women who developed dementia (when analyzed separately from women with mild cognitive impairment). Study findings should thus be interpreted cautiously, but overall findings are consistent with predictions of the cognitive reserve hypothesis. These results add to prior research, where pathological burden was based on cerebral blood flow, amyloid brain imaging, or postmortem neuropathological findings.  
An important inference from these findings is that societal efforts devoted to encouraging higher levels of education will lower the rate of cognitive impairment decades later. The cognitive reserve hypothesis thus provides yet another justification for the slogan, 'education pays'.

Victor W Henderson, MD MS
Departments of Health Research & Policy and of Neurology & Neurological Sciences, Stanford University, USA


  1. Stern Y. Cognitive reserve in aging and Alzheimers disease. Lancet Neurol 2012;11:1006-12

  2. Wilson RS, Boyle PA, Yu L, Barnes LL, Schneider JA, Bennett DA. Life-span cognitive activity, neuropathologic burden, and cognitive aging. Neurology 2013;81:314-21

  3. Rapp SR, Espeland MA, Manson JE, et al. Educational attainment, MRI changes, and cognitive function in older postmenopausal women from the Womens Health Initiative Memory Study. Int J Psychiatry Med 2013;46:121-43