I am of an age when my ability to not “hear” my wife is becoming less a spontaneous reaction and more a physiologic one. I just don’t hear as well as I did 10 or 15 years ago and I have been known to spend more than a few moments looking for my cell phone. So I was particularly entranced by a headline that said that hearing loss, among other “modifiable factors,” might be involved in the development of dementia. I was particularly interested because the Lancet report was entitled, Dementia prevention, intervention, and care – so perhaps there was some evidence that I might improve both my hearing and reduce my risk of Alzheimer's. It is with a heavy heart that I report that the hype far exceeded the knowledge.
First, the press reports  mentioned that 33% of dementia might be eliminated by acting upon several modifiable risk factors; and of all the factors, hearing impairment was associated with the greatest risk, 9%. Now the authors did say early on that the 33% reduction would be possible only if the risk factor were eliminated entirely, so right away that 33% it is going to be a lot lower. But I had hope for the hearing impairment anyway.
I will spare you most of the gruesome details, but they found 11 prospective observational studies of hearing impairment and dementia, 9 with positive associations and 2 with no association. They eliminated the two with no association for unknown reasons, but having looked at the studies I would agree that at least one was of low quality. Of the remaining positive studies that did a meta-analysis of 3, no information on why they were selected. Here is the chart that summarized that meta-analysis.
I was at least happy to see that the confidence intervals did not touch a relative risk of 1, they all seemed to be moving towards that positive association of hearing impairment and dementia. Of course, without those negative studies or in fact the other positive studies they did not use it was difficult to say what these numbers mean other than that they were carefully chosen. Now the investigators report that hearing impairment was a common problem, parenthetically as discussed by my colleague Dr. Wells, so that reducing hearing impairment provided a lot of potential benefits. I was hopeful that at least I might take some action until I reached this sentence
“The mechanism underlying cognitive decline associated with peripheral hearing loss is not yet clear; nor is it established whether correction, such as hearing aids, can prevent or delay the onset of dementia…. Experimental evidence on whether hearing aid use might alleviate some of these negative effects is not available”
No studies have been undertaken to study whether hearing aids lessen your risk. So calling hearing impairment a modifiable factor is somewhat of a stretch. Now I can understand why the Alzheimer's Association, whose international meeting is taking place this week in London would tell this story of science. Possible modifiable risk factors to reduce dementia by 33%, it provides hope, and it invites further funding and research - it is their business model. What I cannot understand is why science journalists reporting on the meeting would not dig a little deeper than the press release or abstract and actually report on what the paper said, not what they were told it said. With reporting of this quality, it may be better to be a bit hearing impaired.
 This headline is typical and comes from the Financial Times: "Lifestyle changes could prevent a third of dementia Treating hearing loss in middle age could cut instances of dementia by 9 percent." This lead from NBC in Boston is a bit more balanced: "Avoiding 9 Key Risks Could Potentially Stave Off Dementia: Panel The reasoning behind the group's unproven theory is that years of silent wear eventually leads to Alzheimer's"