I commented some time ago that the age difference between 70 and 80 is large enough that it would raise concerns for me about the risk of an older candidate developing cognitive problems during 4 to 8 years on office, even if perfectly healthy at inauguration. This diary is in response to someone who asked me to write a diary on this subject.
My qualifications to discuss this: I’m a mathematical statistician by training, distinguished professor emerita in biostatistics, with over 200 peer-reviewed publications in the medical literature. I have been doing research in Alzheimer’s and dementia for more than 30 years, including leading the analysis efforts for the first population-based studies of AD prevalence and incidence 30 years ago.
Here is the quick summary, followed by details and references.
Based on age, an average person with no cognitive impairment at age 80 would have about a 15% chance of developing cognitive impairment in the next 4 years, and almost a one-third chance in the next 8 years.
But an average person at age 70 would have a much lower chance of developing new cognitive impairment: about 5% after 4 years, and about 11% over 8 years.
These calculations are approximate, based on an average person. Factors like genetics, environment, lifestyle, and other medical complications could change the risk. But other things being equal, 10 years older leads to substantially higher risk.
A separate calculation from different data shows that a person 80 years old at baseline would have about a 3-fold greater risk of developing dementia over time, compared to someone identical in all other respects but 10 years younger.
I will still vote for the nominee of the Democratic party regardless of age. I have great respect for the service, experience, messages, and compassion of Biden and Sanders, our two oldest candidates, and I will vote for either one if nominated. But given my professional background, I’d prefer to vote for someone somewhat younger.
--------—
Details and references: for folks who want them.
The three leading candidates for the Democratic party nomination are also all over 70 years of age: Biden would be 78 on his first day of office in 2021; Sanders would be 79, and Warren would be 71. Given our concerns about Trump’s cognitive status, should we worry about our own candidates?
Let’s consider a hypothetical person in good cognitive health, but 80 years old at inauguration, and for comparison, a hypothetical 70-year-old, identical except for age:
- What is their risk, based on age alone, of becoming cognitively impaired during their first term?
- What is their risk, based on age alone, of becoming cognitively impaired by the end of a hypothetical second term?
Calculations:
Both Gillis and Gao looked at all studies with solid methodology and sufficient detail about numbers to do a “meta-analysis” combining all published information. Gillis looked at the likelihood of transition from cognitively healthy to mild cognitive impairment, while Gao looked at the transition from not having dementia to having dementia. I used Gillis’s numbers here, but Gao’s support the general approach and give similar estimates of how the risks go up with age.
From Gillis, starting around age 80, the annual risk of going from no impairment to mild impairment or worse is about 4% per year. To stay healthy for 4 straight years, the chance each year is 96%. Assuming that risks each year are similar and independent, a pretty standard “constant hazard” exponential model, the probability of staying healthy is
0.96 x 0.96 x 0.96 x 0.96 = 0.85. So the chance of not staying healthy is 1-0.85, or 15%.
For 8 years, you have to repeat the staying healthy for 4 more years. I’m giving the benefit of the doubt and assuming the risk stays the same each year, so it’s 0.85 x 0.85 or 0.66. The chance of NOT staying healthy, then, is 1-0.66 =0.34, about one in 3.
For our hypothetical 70-year old, it is a little trickier. Risk in early 70s is about 1% per year but this goes up to 2% per year in later 70s. 0.99 x 0.99 x 0.99 x 0.98 = 0.95, and 0.98 x 0.98 x 0.98 x 0.98 x 0.95 =0.89.
These numbers are generally consistent with a different calculation based on relative risk as a function of age. Hebert and colleagues examined a population-based study in Chicago, with subjects around 75 years old at baseline on average and followed for 11 years. They found that the odds of developing a new diagnosis of AD went up 1.116-fold for every year older at baseline. A 10-year age difference would lead to a tripling of the odds (1.116 taken to the 10th power). (A 7 or 8 year difference would still more than double the odds.)
I do not present these as definitive estimates of risk. I’ve done very simple calculations based on reasonable numbers in the peer-reviewed literature. You can take other factors into account or do more elaborate calculations. For example, a family history of dementia or medical conditions like inadequately treated hypertension would increase the risk, while some genotypes and healthy lifestyles may decrease risk. These estimates, to repeat, are for an average 70- or 80-year old.
But the bottom line remains: once you are past 65 or so, the older you are, the greater the risk of developing cognitive impairment or dementia, compared to someone exactly the same but years younger.
References:
Gillis C, Mirzaei F, Potashman M, et al. The incidence of mild cognitive impairment: A systematic review and data synthesis. Alzheimers Dement 2019 Mar 8;11:248-256. doi: 10.1016/j.dadm.2019.01.004. eCollection 2019 Dec. (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6416157/)
Gao S, Burney H, Callahan C, et al. Incidence of Dementia and Alzheimer Disease Over Time: A Meta-Analysis. JAGS 67:1361-1369, 2019.
Hebert L, Bienias J, Aggarwal N, et al. Change in risk of Alzheimer disease over time. Neurology 2010 Aug 31; 75(9): 786–791. (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2938969/)