Education conceals dementia signs

Better educated people with dementia suffer a more rapid memory loss than those who have had less education, newspapers reported on October 23 2007. The Daily Mail reported that people who have had more years of education have an initial delay in the onset of memory decline associated with dementia, but that once this memory loss began, “those with more education saw their rate of decline accelerate 4 per cent faster for each additional year of education”.

These stories are based on a study that monitored elderly people in New York, and looked at whether the 117 people who went on to develop dementia had different patterns of memory decline based on the number of years of education they received. Although the study does have some strengths, such as its prospective design and long period of follow up, its small size means that the results should be considered preliminary and that further studies are needed.

In lieu of further studies, it should be pointed out that this study was in 117 caucasian, middle-class people in New York, born between 1894 and 1908. While the results may be representative for this particular group of people, they may not be applicable to people from different ethnic or socioeconomic backgrounds or to people who were born and educated in more recent times.

Where did the story come from?

Dr Charles Hall and colleagues at the Albert Einstein College of Medicine, and Arizona State University carried out this research. The National Institute of Aging in the US funded the study. The study was published in the peer-reviewed medical journal Neurology.

What kind of scientific study was this?

This was an analysis of a subset of people enrolled in a prospective cohort study: the Bronx Aging Study. Healthy elderly adults (aged 75 to 85 years old) without dementia enrolled in the study between 1980 and 1983 and have been followed up until 2007. This current study analysed data for the 117 participants who were cognitively normal at enrolment, but developed dementia during follow up, and who provided information about their education and completed the study memory tests.

At enrolment, participants answered questionnaires about how many years of education they received and their medical history. Medical history information was confirmed with their spouse or a family member if possible.

To test participants’ memory, researchers asked participants to read a list of 12 words on index cards and then try to remember as many of these as possible in a two minute period. The participant was then reminded of the words they had forgotten and asked to repeat the entire list of words again. This process of reminding was repeated a maximum of six times if participants could not remember the full list, and participants scored on how well they had performed. This memory test (the Bushke Selective Reminding Test) was used annually from year three of the study.

Researchers assessed participants’ neuropsychological abilities using a panel of tests when they enrolled, and repeated these tests on a yearly basis. If these tests indicated that the person might have dementia, they were given a brain scan and blood tests to rule out other possible causes of their problems. If they could find no other cause, dementia diagnoses were made by a panel of experts based on a set of standard criteria.

For those who developed dementia, researchers used computer modelling to identify the point at which the rate of memory decline had increased, and how fast this decline was before and after the selected point.

Researchers then compared the point at which memory decline increased, relative to the point at which dementia was diagnosed and the rate of decline between people with different levels of education.

What were the results of the study?

Researchers found that people with more education took longer to reach the point where their memory began to decline rapidly than those with less education.

For every extra year of education a person had, accelerated memory decline was delayed by about two and a half months. After this point was reached, however, the memories of more educated people declined faster than those with less education.

What interpretations did the researchers draw from these results?

The researchers concluded that these results show that people with more education have a delayed onset of cognitive decline before developing dementia, but that once their memory begins to decline, it declines faster than in people with less education.

They say that these results support the “cognitive reserve hypothesis”, which postulates that individuals with higher education have a greater ability to compensate for the changes that occur in the brain early in dementia, but that once signs of dementia begin to show decline is more rapid because the disease is more advanced.

What does the NHS Knowledge Service make of this study?

The strengths of this study are its prospective collection of data and its long period of follow up. However, the study is limited by the fact that it only analysed a relatively small number of people. Replication of these results in a larger population would increase the reliability of these results. Other points to consider when interpreting the study are:

• The cause of dementia can only be confirmed on autopsy and only 23 of the study's participants had an autopsy. These revealed a range of diagnoses, including Alzheimer’s disease and vascular dementia. It is possible that including people with a range of diseases such as these may mask different patterns of memory decline between people with different diseases. The researchers found that when they looked only at those people with an autopsy-confirmed diagnosis of Alzheimer’s disease, although each additional year of education did delay memory decline by about 1.8 months, this result was not statistically significant.
• In order to carry out their computer modelling, the researchers had to make certain assumptions, such as assuming that the rate of memory decline remains steady up to a point and then accelerates. If these assumptions are not correct, this would mean that the results are not likely to be correct.
• The study used only one measure of memory, using more than one test might have given a more complete picture or people’s memory decline.
• This study only assessed the effect of additional years of education, and did not consider the quality of the education or of people’s actual academic achievements.
• As with all studies of this type, other differences (not related to the length of time spent in education) may be responsible for the differences in onset and rate of decline in memory seen. People with more education did have higher memory scores to start with, tended to be diagnosed with dementia at an older age, and participated in more follow up visits than people with less education. Differences between the groups in access to health care or general measures of health could also have affected the results.
• Most of the people in this cohort were caucasian and middle class, though only 13 had had a college education, this may limit the generalisability of these results to groups with different ethnic or socioeconomic groups. In addition, most people in this sample were born between 1894 and 1908, and they may not be representative of people born and educated in more recent times.