Q & A With Dr. Tiia Ngandu: The Dementia Risk Score

Last week, I wrote about a dementia risk score developed by researchers at the Karolinska Institute in Sweden. To put the risk score in perspective, I checked in with Dr. Tiia Ngandu, co-author of the risk score article published in the September issue of Lancet Neurology. Here are her answers to my questions:

Tiia Ngandu, Ph.D.
Karolinska Institute (Sweden) and University of Kuopio (Finland)

Q: Some background information: my dad died of a hemorrhagic stroke last November, and his autopsy showed severe cerebral amyloid angiopathy (CAA) along with probable Alzheimer's disease. He had few of the risk factors I think were involved in calculating the "dementia risk score": his cholesterol was always low, he was not at all overweight, his blood pressure was low, he exercised and had 16 years of education. So, in my father's case, anyway, he would have had a low score, and I think would have been a "false negative" in your study.

A: First of all, I would like to say that the aim of our risk score was not to say definitely that someone will develop dementia or someone will not. That would not be possible, since we are dealing with a disease that is as complex and multifactorial as Alzheimer's disease. Those persons who had more risk factors had a higher risk of developing dementia in 20 years time, but even among those who had the worst risk score, only 35 % developed dementia - that is to say that with our score it is not possible to "label" persons into future demented and nondemented. We hope that the risk score will be more used to inform both physicians and patients about the risk factors behind dementia in a very practical way. We believe that the persons with most risk factors and highest risk score might benefit from both lifestyle and pharmaceutical interventions.

Q: Was a fairly homogeneous population used for the study? [I asked this because results from one population may not apply to another.]

A: All the participants are from Eastern Finland, and they were fairly homogeneous population.

Q: Did you differentiate at all among the various types of dementia? What cognitive or other tests were used to detect dementia?

A: We differentiated between different types of dementia. The majority of patients had AD. The Mini Mental State Examination (MMSE) was used as a screening test, and after screening the patient went through both clinical and neuropsychological examinations. Due to the use of MMSE at the screening, we may have missed persons with mild dementia, especially early vascular dementia.

Q: Were there risk factors considered that were not found to be predictive in this study? Smoking? Family history of dementia? Tooth loss? Genetic status?

A: Risk factors that were considered but not found predictive were midlife smoking and diastolic blood pressure. We did not have information on family history of dementia or tooth loss. ApoE4 was predictive in this study and it was included in a second risk score model. (It was left out from the first one because we wanted to include only parameters that are easily available in primary care settings)

Q: What other risk factors may be considered while you refine the scoring process?

A: Other risk factors that should be considered in the future risk scores are at least diabetes and family history of dementia: those factors have quite a lot of evidence behind them at the moment.

Q: I saw an abstract of a presentation by researchers at Kings College London at the recent Alzheimer's conference in Madrid that says that in the Honolulu Asia-Aging Study, "cholesterol levels in men with dementia, and in particular those with Alzheimer’s disease, had declined at least 15 years before the diagnosis and remained lower than men without dementia throughout that period." Do you think this is true in other populations? How might this affect the cholesterol component of your risk score?

A: Some studies have shown that the cholesterol levels (as well as blood pressure levels) decline in demented patients, and that this decline begins already before the diagnosis of dementia can be made. This decline is probably due to the disease process rather than the decline in cholesterol increasing the risk of dementia. In the Honolulu-Asia Aging Study this kind of decline was observed already 15 years before diagnosis. Also the research group from the Honolulu study thought that the decline might be due to early stages of dementia. I don't see that these findings are in any way contradictory to our findings that high cholesterol levels at midlife would increase the risk of dementia. One should keep in mind the trajectories of cholesterol change with aging and in relation to chronic diseases.

Q: The audience for my blog is mostly middle-aged people, many of whom are caregivers for dementia patients. On a practical level, is there anything we should learn from the dementia risk score other than a confirmation that cardiovascular health and education levels appear to be linked to dementia?

A: What I believe that a middle-aged person could learn from our risk score is that there are several things one can do in order to reduce the risk of dementia. Especially the cardiovascular risk factors can be modified with both lifestyles and medications. I think it is quite good news for those who worry that they might become demented and would like to do something to prevent or postpone it. Of course there are still many other and also unknown factors influencing the development of dementia, but we believe that this is a step in the right direction.

Comments

[Gail Rae Hudson]

First, I want you to know something I always think when I'm reading your posts: I not only admire your tenaciousness in seeking out point people and interviewing them from an educated layman's perspective, I'm grateful for it, Mona. Thank you!
I find it curious that only 35% of the subjects who scored high for dementia risk developed dementia. I'm not well schooled on statistics so I have to assume that 35% is statistically significant. It certainly sounds so. But, turn this around and it means that 65% with "the worst risk score" didn't develop dementia. That seems even more statistically significant, to me.
And, I can tell you, having known people who lived in Finland, that the population is not just "fairly homogenous" but extremely homogenous; possibly one of the most homogenous populations in the world.
As well, it's hard for me to go along with Dr. Ngandu's final answer. In fact, from my point of view, the study tells us nothing that would allay the fears of those who wonder if they are at risk for dementia, and certainly doesn't indicate much in the way of prevention. Spin, spin, spin. Whoa!
Anyway, what I really appreciate about this is your doggedness in asking questions, your personalization of the questions and your ability to step out of the way and let what Dr. Ngandu said speak (or, rather, refuse to speak) for itself.

[Mona Johnson]

Thanks for the compliment, Gail!

From an individual perspective, I think you're right - it's encouraging that 65% of those people with the worst risk scores didn't develop dementia. I think the keys here are:

1. this risk score is evolving, and this was the first version
2. 35% is a higher rate for developing dementia in 20 years than you would expect (in the general population), so this indicates Dr. Ngandu and colleagues are on the right track.

Also, several points in Dr. Ngandu's defense:

1. "fairly homogenous" was the wording in my question, which is probably why she answered using that phrase.
2. I don't feel that she refused to speak for herself - I'm more grateful that she took the time to answer my questions about her research. I do think it's probably true that if the overall population's cardiovascular health were improved, dementia rates would go down.
3. I think that journalists, rather than researchers, are more often responsible for "spin" - maybe because readers and viewers don't have much patience with complex issues and even more complex answers. You know all too well that anything to do with dementia is intrinsically complex, so both journalists and researchers are stuck with a disconnect between the information they want to present and the audience that needs the information - at least that's my perspective.

My biggest worry with this kind of information, or the Alzheimer's Association's "Maintain Your Brain" campaign is that we'll blame dementia patients for their illnesses. Many smart, active, healthy people develop dementia, and there's nothing obvious they could have done to prevent it.

I'll be interested to see how this risk score evolves over time.

[Gail Rae Hudson]

Very good points, Mona. I'm also curious to see how the risk score evolves over time.
Something I thought of after commenting: I notice that diabetes was mentioned as a risk factor that has "quite a lot of evidence behind [it]". What I find interesting is that of all the people I know very or fairly well (there are several others of which I know almost nothing) through caregivers who are elderly and displaying dementia (including but not exclusive to those online), which would be 11:
2 are diabetic, diagnosed in old age (which I believe makes a difference) (one is my mother, one is the mother of a friend here in Prescott).
1 (Bailey's Mom), I'm not sure of regarding diabetic status.
8 are not (at this time) (or were not) diabetic.
Of the two who are definitely diabetic, both experience a much milder type of dementia than all but one of the 8 who are definitely not diabetic and milder than the one of whom I'm not sure.
One of the non-diabetics (who is now dead, not your father), displayed an even milder form of dementia than the two who are diabetic.
Also, a cousin of my mother's was a Type 1 diabetic since early childhood (I am not including him in the group above). He died a few years ago of complications of his diabetes. He was in his early 90's (I know...amazing, huh; he was only disabled by his diabetes for the last year of his life). He never experienced dementia.
So, there's room for maneuvering even when considering diabetes as a "quite [evidential]" risk factor.
I hope you enter this in a parade, Mona. I'd love to see what sort of comments you get from the medical community!