19 January 2007. The hypothesis that more years of formal education protect the brain against the ravages of dementia may have just gotten a higher grade. In this month’s Archives of Neurology, Anthony Fotenos and colleagues at Washington University, St. Louis, Missouri, report that older adults with higher education and occupational achievement are better able to weather loss of brain volume, and even Alzheimer disease (AD) pathology, without showing signs of dementia. But there’s a twist. In the same journal, Washington University researchers also report that highly educated individuals who do get AD are likely to have earlier onset. This second finding may seem counterintuitive at first, agreed Cathy Roe, lead author on the study. “We believe it implies that people with more education may be involved in activities, hobbies, or jobs which might highlight to a person when he or she starts to have very subtle cognitive impairment,” she said in an interview with ARF. “It says nothing at all about education and risk for AD,” she added.
Yet a third report indicates that education may have a limited impact on survival in people with AD. In the January 10 British Medical Journal online, Carol Brayne and colleagues at the University of Cambridge, England, report results of a study to determine survival time post-dementia diagnosis. They found that the median survival time was 4.5 years. Though the study was not designed to specifically determine the effect of education on survival, they found more education tended to shorten survival time. However, the effect was not robust.
More years of formal education are equated with cognitive reserve, a concept that Fotenos, coauthor Randy Buckner, and colleagues define as a delay in time between the appearance of AD pathology and disease expression. Research suggests that people with more cognitive reserve are better able to compensate for ongoing brain pathology, such as amyloid-β and tau toxicity (see Bennett et al., 2003 and ARF related news story). The same appears true in non-demented older people, since cross-sectional data indicate that those with higher socioeconomic status (SES), i.e., more education and better jobs, appear to have greater age-related loss in brain volume, suggesting that they may be more successfully compensating for ongoing disease than their less educated counterparts (see Coffey et al., 1999). But as Fotenos and colleagues point out, “alternatively, there could be a relation between SES and brain aging without any involvement of preclinical pathology.” While their findings support the cognitive reserve hypothesis, they don’t completely discount the second possibility.
To test whether greater brain volume loss in older, well-educated, non-demented individuals is related to ongoing pathology, Fotenos and colleagues used MRI to measure whole brain volumes (adjusted for head size) in 362 non-demented volunteers. One hundred of the older subjects (aged 65-93 years) with clinical dementia ratings (CDR) of zero were also evaluated for dementia and SES, and 91 of them were followed up with clinical assessments of dementia over a 3-year interval. Of those, 33 volunteers also received follow-up MRI scans for brain volume measurements, and a cohort of 58 CDR 0 volunteers also were imaged for amyloid with Pittsburgh Compound B (PiB).
As in previous studies, Fotenos and colleagues found that adjusted whole brain volume (aWBV) decreases as people age (mean yearly loss of 0.24 and 0.2 percent for men and women, respectively). They also found that in the 100 older adults, those with highest SES rank had both significantly lower aWBV and a significantly faster rate of decline over the 3-year follow-up. So are these losses related to pathology?
Several observations suggest that the answer is yes. First, Fotenos and colleagues found that some of those high SES individuals were indeed preclinical because during the 3-year follow-up, their CDR ranking increased to 0.5, indicative of progression to mild dementia. In addition, those preclinical individuals had even greater rates of aWBV loss, contributing to more rapid declines in the high SES group as a whole. Also, imaging revealed that 16 percent of the non-demented individuals had high PiB binding, indicating the presence of amyloid plaques. Elevated PiB binding was also associated with reduced aWBV.
All told, the findings suggest that higher education helps older adults compensate for decreasing brain volume and ongoing pathology, supporting the cognitive reserve hypothesis. But the authors also point out that their findings do not exclude the possibility that other factors are at work. They note that it is difficult to fully account for the magnitude of the SES/aWBV interaction unless SES offers greater protection or CDR 0 pathology is more severe than is generally accepted. “We thus conclude that reserve likely explains some, but perhaps not all, of the novel association reported here between SES and structural brain aging,” they write.
The second Archives of Neurology study was spawned by earlier work from Alison Goate’s lab at Washington University. Pau Pastor and colleagues found that in a large familial AD kindred in South America, age of onset was later in those with less formal education (see Pastor et al., 2003). Now, Cathy Roe, who was coauthor on that original study, John Morris, and colleagues have replicated those findings using data collected by the National Alzheimer’s Coordinating Center (21,880 subjects) and the Washington University Alzheimer’s Disease Research Center (1,449 subjects).
The researchers compared three different educational levels: less than 8 years, between 8 and 11 years, and over 12 years of formal education. In both the national and Washington University data sets, reported age of onset of dementia was earliest in those with highest education and latest in those with lowest education. Reported age of onset is when the patients or their spouses/caregivers first noticed symptoms. The study corroborates similar findings by other groups (see Del Ser et al., 1999 and Moritz et al., 1993).
Does earlier onset in more educated people say anything about the underlying pathology of the disease? In a previous study, Moritz and colleagues found that less educated people generally had more severe pathology on presentation. Roe and colleagues found exactly the same thing in both their data sets. But they also found that the time between first reported symptoms and presentation at Alzheimer Disease Centers was the same, regardless of education. “That tells us that it is not that [people with lower education] notice the symptoms and are taking their time to go to the ADC; it seems they are noticing the symptoms later,” said Roe.
The combined evidence suggests that people with higher education are more attuned to subtle cognitive changes. It also has practical implications, particularly for diagnosis, because it suggests that those with lower education could potentially be diagnosed earlier. “It may not be enough to ask if patients are having memory problems. Diagnosis may be better tailored to activities that people are involved with,” suggested Roe.
The survival study reported by Brayne and colleagues was based on a 14-year follow-up of 438 patients who were part of a larger cohort (13,004 people) of older adults (>65 years) taken from primary care registries in England and Wales. First author Jing Xie and colleagues found that the median time to survival after diagnosis (defined as the midpoint between first follow-up where dementia was detected and the last dementia-free follow-up) was 4.1 years for men and 4.6 years for women. As with previous studies on survival, Xie and colleagues found that age at onset had a major effect on survival. Those diagnosed before age 69 survived on average 10.7 years, falling to 5.4, 4.3, and 3.8 years in those aged 70-79, 80-89, and 90 or over, respectively (see ARF related news story).
The researchers found a relationship between years of education and survival. Those with less education had longer survival, but they note that this effect was not strong and they did not take other variables into consideration. On this relationship between education and survival, they note that other studies have been equivocal. A French study found that education had no effect (see Helmer et al., 2001), while a study in the Netherlands found that higher education led to higher mortality rates, but only if patients had either low Mini-Mental State Exam scores or depression (see Geerlings et al., 1999).—Tom Fagan.
Roe CM, Xiong C, Grant E, Miller JP, Morris JC. Education and reported onset of symptoms among individuals with Alzheimer disease. Arch Neurol. 2008 January;65:108-111. Abstract
Fotenos AF, Mintun MA, Snyder AZ, Morris JC, Buckner RL. Brain volume decline in aging: evidence for a relation between socioeconomic status, preclinical Alzheimer disease, and reserve. Arch Neurol. 2008 Jan;65(1):113-20. Abstract
Xie J, Brayne C, Matthews FE, Medical Research Council Cognitive Function and Ageing Study collaborators. Survival times in people with dementia: analysis from population based cohort study with 14 year follow-up. Brit. Med. J. online. 2008. Jan 10. Abstract