Cocoa: Stopping dementia in its tracks (or possibly not)

This morning I heard of a study which had found that drinking hot chocolate regularly could stop older people from developing dementia. The study, run by a team from Harvard, suggests that cocoa improved blood flow to the brain, resulting in better scores on tasks designed to assess memory. I duly had a look at the paper, published in Neurology.

The study was designed to investigate the effect of cocoa on neural coupling, which refers to the relationship between neural activity and cerebral blood flow. Sixty people were included in the study: half were given two cups of high-flavanol (an antioxidant) cocoa each day and the rest were given low-flavanol cocoa. A number of cognitive assessments were carried out, many of which are also used by clinicians in cases where dementia is suspected. The Mini Mental State Examination (MMSE), for example, assesses cognitive functioning in a range of domains, as does the Trail-Making Test. After 30 days of prescribed cocoa-drinking, the cognitive tests were repeated and a significant improvement was observed on one of them (the Trail-Making Test B). There were no differences reported between the groups, i.e. Trails B scores improved regardless of the quantity of flavanol ingested and no other significant improvements were observed in either group.

Like all studies, this one has its limitations. Whilst it’s a positive that the authors split the sample in half and gave them different quantities of flavanol, it’s usually the case that you have an ‘intervention’ group and a ‘control’ group. In this case, you would have had those who took cocoa for 30 days, perhaps splitting them to take different quantities of flavanol, but you would also have had a similar group – matched for age, gender and health – who took no cocoa. They would be your ‘controls’. That way, you can compare scores across all these individuals at baseline and the end of treatment (in this case end of cocoa) and compare them. The advantage of doing this is that you can see if the improvement was associated with the cocoa or if it was simply a spontaneous improvement which was also observed in the control group.

In this study, there was no control group, so it’s difficult to presume that the observed improvement was due to taking cocoa. In any case, an improvement on one test of cognitive function may be interesting, but it won’t necessarily have any real-world impact. The Trail-Making Test assesses a range of functions, but the test is the sort of thing we use clinically, it’s not something you would have much cause to do in daily life. It’s a bit like the Brain Training games which teach you to remember long lists of words – your skills in that domain may improve, but realistically that’s only useful if you want to be able to recite your shopping list verbatim at any moment. I’d be more excited if the improvement had been seen on the MMSE since that is a much more global test of functioning. As it is, this was a small study which didn’t have a control condition and that seriously limits the extent to which we can draw conclusions from the findings. If you’re considering stocking up on the Green and Blacks simply to stave off dementia, I’d think again.