Here’s a fact that I don’t think that a lot of people are aware of: the risks of developing dementia with aging have been falling pretty steadily for a long time now in the industrialized world. This trend has been holding for decades now. An 80-year-old today is much less likely to show signs of dementia (memory loss, confusion, etc.) than the people in their parents’ generation. (Some may be reminded of the similar Flynn Effect, which is likely to have some similar drivers behind it).
Now that is good news, but we scientific types are not content with that. No, we would very much like to know where this good news is coming from and why. That’s the focus of this new paper, which is looking into a fundamental question: do people reaching old age today have higher cognitive capacity to start with (presumably built up in their younger years), or are they holding on to it better as they age? Those aren’t mutually exclusive, either, of course. And we have to assume that these effects must be due to environmental and behavioral changes, since this is far, far too short a time to see much of a genetic-background change. No, we seem in general to be doing some things right, either early in life, or throughout our lifespans, or in old age itself. And quite possibly all three.
As the paper notes, the explanations really do make a difference, not least because they indicate which points in human lifespan need the most effort put into their improvement. The literature thus far has apparently been conflicted, both in determining “memory level” (cognitive capacity) and “memory slope” (the rate of decline with age). The authors pulled data on simple word-list recall cognitive tests, 783,000 of them over 219,000 patients, from five large decades-long longitudinal studies.
The improvement in memory function over time is clear in the data, but there are some subtle differences. Memory level improvement is more noticeable in the more recently born cohorts, but an advantage in memory slope seems largest for the older groups. In fact, that memory-slope advantage seems to have gone as far as it can go once you get to people born after around 1950, but the generational improvements in memory level have actually been getting great as time goes on. That’s certainly worth knowing.
The paper goes on to look at these effects if you spread the data out into tiers of the most-declining versus least-declining individuals. And that shows that the generational progress just mentioned is disproportionately coming from improvements in the worst decline trajectories. That said, severe decline in every cohort is mostly a function of increasing age, as you’d have expected. You can get even more fine-grained than that: individuals often show a pattern over time of periods of memory stability punctuated by bursts of faster decline. Looking through the data for these, the authors find that the stable periods become less common with aging in every individual cohort, but looking across the cohors you see that the more recently-born ones do show more time spent in the stable periods overall.
In general, the more recently-born populations of both Europe and the US have a higher memory capacity as they get older, and they lose it more slowly - i.e., both of the explanations mentioned above are operating at the same time. And that “losing it more slowly”, when you study it more closely, is mostly a function of staying in stable periods for a longer time as opposed to the intermittent periods of decline. Aging brains are still at risk, but their steady plateaus have become longer.
So if these improvements are down to environmental and lifestyle changes (as they really have to be, as mentioned above), what might these be? The advantages in memory level are widely believed to be set up earlier in life: better nutrition, more education, and overall better childhood health. But the “memory slope” changes are more likely to be things that can be addressed throughout adult life, such as lowering cardiovascular risk (hypertension, etc.), lifestyle factors like smoking and drinking, and lower exposure to particulate air pollution.
Improvements in level and slope are both operating, but it does seem like the early-life memory level improvements are more of a factor now. That means that if we want more mentally healthy older adults on a population level, we should start while they are children, while not backsliding on the improvements we’ve already been making across adulthood. It would be nicer, perhaps, if there were some more clear-cut courses of action for people who are already getting old, and I speak from the heart here as someone who’s 64. The people wanting to sell you stuff certainly think so, and it would be a lot simpler if there was some Single Bad Thing that we could all point at (aluminum cookware, not enough B12, something). But having had good care as a child and living a relatively clean life thereafter are probably the biggest factors, and there’s not as much to be done about those after a point!
This of course needed to be tracked down. The standard is generally 0.15% for a known impurity (one whose structure and properties have previously been worked out) and 0.10% for unknown ones. And this peak was getting too close to that cutoff - there were no guarantees that whatever-it-was might not come in across the threshold at some point, for some reason, and cause some real problems. Plus, just what was it anyway? Making sure of the structure could allow a switch to the 0.15% standard eventually.