I have commented on the difference between relative and absolute risk before, but this is an important distinction and really deserves an article of its own.
I was reading Dr. Brownstein’s blog today, which is a good one overall, and Dr. Brownstein loves to beat up on the vaccine industry, and today he posted yet another article on this, where he spoke about relative risk and how they use this to look to deceive us.
Vaccines are very bad news so it’s great he’s doing this, looking to get the word out more on this stuff, and this topic has indeed attracted a lot of attention over the past few years, and rightly so. More and more people are questioning the appropriateness, effectiveness, and safety of these vaccines and questioning is always a good thing, even though the vaccine makers and the conventional health authorities and conventional media they have in their back pocket do strongly discourage any of that.
This article isn’t about vaccines though, but I thought that his example of the way that this year’s flu vaccine is being promoted was a very good example of how relative risk and absolute risk paint a different picture, and now absolute risk is what we always need to be looking at.
Relative risk by the way is the percentage of the change in risk between two variables, for instance the difference in the risk of getting the flu, or any other undesirable outcome. So for instance if we compare the incidence of something and it is X%, and the percentage here doesn’t matter, and we see that X occurs at different multiple, then the relative risk has changed by that percentage.
Let’s use an example from diabetes. Let’s say one person in 50 gets a certain undesirable outcome with no intervention. We’ll say that this is some sort of complication that is associated with diabetes. Then we look at someone with high blood sugar, let’s say an A1C of 8, and now three people in 50 get this. So now we have a 300% increase in risk, the relative risk. So we tell people that you are 3 times more likely to get this than a normal person is, and this scares people, as it is intended to.
In this case though the risk is very low, as we’ve gone from 1 person in 50 to 3 people in 50, that’s not very significant actually.
If we are to measure the absolute risk, which is the correct way to do it, we’d see that the risk went from 2% to 6%, which means you are 4% more likely to get this complication if your A1C is that high, which sounds much less scary than 300%, but is the correct measure.
When we see charts showing complications, they are always expressed in relative risk, they don’t even mention what the absolute risk is, and since we’re dealing with conditions that non diabetics rarely get, this really matters. So if 1 person in 100 or 200 or whatever gets this, the fact that you are twice or three times more likely to get this doesn’t mean much at all really.
This trick is used in a lot of studies and it sure does make things more dramatic, and much more so than it should be. So if they say under these circumstances you are twice as likely to have something bad happen to you, that does sound pretty alarming, until you look at how often it happens normally, the control in other words.
Dr. Brownstein’s example in this post was the CDC claiming that this year’s flu vaccine reduced the relative risk of patients going to the doctor by 48%, which they call a modest reduction, and of course they are recommending our getting this vaccine. So you might think, well getting the vaccine sure sounds like it reduces the risk here, maybe I should get it, although we’re not sure why going to the doctor for the flu should be seen as a big enough deal to inject a cocktail of known toxins into our veins, ones that have been shown to produce a number of nasty side effects, and by all reasonable accounts, the treatment is worse than what it is seeking to prevent.
In any case though, Dr. Brownstein explains that the absolute risk reduction here is only 1%, meaning that for every 100 people who get the vaccine, it will only prevent one from getting bad enough to want to see a doctor.
This is because we know that among these 100 patients, there will only be a few who get that bad anyway, and there’s 1% less that need to go now.
It’s important to look to quantify risks accurately and not use tactics such as relative risk reduction to look to deceive people. Many in the medical and scientific community though shamelessly use this measure, and while it’s perfectly fine to do so, you can’t neglect expressing the risk in absolute terms, because people will very often mistakenly assume that the relative risk and the absolute risk are the same thing, and without knowing what the absolute risk is, relative risk isn’t even meaningful.
There’s another important variable here, and that’s comparing the changes in risk with other interventions, and Dr,. Brownstein mentions taking lots of Vitamin C as an alternative here. So if this vaccine reduces the absolute risk by 1%, something else, like Vitamin C, may reduce it even more.
If you’re looking to promote something, you can create a better picture if you just ignore alternatives, and we love to do this in the diabetes industry, particularly when it comes to medications. We love to do that with medications period actually.
So for instance, if it were the case that a certain med reduced the risk for something, there may be something else we can do, like changing our diet, getting more exercise, or taking certain supplements, that may work as good or better at reducing the risk, without introducing further risks.
So we need to be looking at overall risk here, net risk reduction in other words, which includes any risk reduction of the treatment, but also accounts for additional risks incurred from the treatment itself. People using insulin therapy to look to prevent microvascular damage but dying of heart attacks from the macrovascular damage this treatment increases, as we have seen in studies, is a good example here.
To even pretend to be able to do this calculation though, we need to know what the changes in absolute risk are. When we’re talking blood sugar reduction, this is especially important, lest we cause more risks than we prevent. That’s pretty much what happens in real life actually.