Even substituting your line for my crappy freehand line, it still isn’t true. The risk exceeds that of 50 year olds in 1998 but not 60 year olds.
But the numbers are more revealing than that fact. The risk appears to have risen from 2 in a thousand to 5 in a thousand which is still a low risk.
But honestly, none of this matters if boosters are available and effective in unlimited repetitions. The problem is not whether or not 40 year-olds get the vaccine—but whether 40 year-olds are still protected when they’re 70 or 80 if they do so.
The issue is not that your line was “crappy” but not horizontal
I wrote “the rate among people aged 40–49 is now almost what it used to be in people aged 60–69 in 1998”. This is exactly what the chart in the US shows. Canada and Sweden show similar trends.
And do you know the concept of confidence interval? And trend extrapolation?
A few days before this paper came out, I decided to go ahead and get the Shingrix vaccine. What motivated me at age 69 is that my wife has a coworker around 40 years of age who just came down with a nasty case of shingles. I’d been thinking about getting the vaccine for the last year or two. This provided a good reminder and some motivation.
After the initial vaccination I felt fine for the first 12 hours and then like crap for the next 18 hours or so, but I’m glad I went ahead and got it done. It now appears there might even be some additional benefits beyond shingles prevention, which is an added plus.
Sorry I am a bit lost about why people think there will be less effectiveness if you take a vaccine early? Is there any reason to believe that a booster in 10 or 15 years would not “work” if you were already vaccinated? I’m not an immunologist, but I’ve never heard of this idea before.
Agreed. I was under the impression that the opposite was the case: a booster after prior vaccination exposure would be more responsive, as it already has some trace immunological memory, so it’s not starting from scratch. But my impression might be mistaken, so I’m eager to be enlightened.
It is actually not possible to fully control for this. This is one of the known limitations of observational research. We are measuring both the treatment effect and the differences between groups.
Some studies do a better job than others at least trying to control for this. For instance, in some vaccine work of this sort they will look at off-target results such as death by car accident. Sometimes the “effect size” is similar - the vaccinated group gets in fewer car accidents as well. In which case, we may be measuring mostly healthy user bias - those who get vaccinated tend to be healthier, wealthier, and wiser than those who don’t. Or whatever.
It is for this reason that this work is generally considered to be great at generating hypotheses, but not good at proving them out.
Of course exceptions exist, such as smoking with its enormous effect size (up to 3,000% increased risk of certain cancers as opposed to this study’s modest 23% risk of MACE).
One can look up things like the the Bradford-Hill criteria on wikipedia to get a better sense of when to consider epi work to be determinative. Bradford Hill criteria - Wikipedia
The study itself lists this in the limitations section: “Third, despite adjusting for a wide range of covariates, residual confounding factors still exist that could influence the results.”
It seems to have an extensive set of appendices which explain the statistical methods used, but I did not look at them. In general, I would have appreciated seeing in the main body of the paper 2 things:
What the raw, unadjusted effect looked like. Perhaps that showed a 50% reduction in MACE and the adjustments reduced it to 23%? Whatever it was, that would be interesting to see.
Some health outcomes believed to be off-target to the vaccine, and how those looked on both an unadjusted and adjusted level. Say if they looked at cancer or car accidents or broken bones. What we’d hope to see is that the difference between groups goes to zero after the statistical adjustments for covariates has been made. That would suggest the adjustments were pretty solid.
Anyway, I personally got vaccinated for Shingles a couple years ago, right after I turned 50. I hope it gives me lots of benefits but I won’t put too much stock in one epi paper on the subject.
