Disclaimer: I’m a family physician, and I prescribe birth control pills. Yaz® and Yasmin® are not my first pills of choice, but I have prescribed them on occasion. I have never been paid by the company or provided any monetary compensation or other considerations for what I present here, or in the past. All I have received from them is free sample packets of pills to give patients, which I do at no charge. I have not been paid speakers’ fees or research money from whoever makes these drugs.
Ok, I just saw a really irresponsible bit of medical fear-mongering. You can watch it on the Canadian CTV station.
The title is “More risks being discovered with birth control pills.”
In the first place, the title is misleading, if not outright dishonest. At best, the title ought to be, “Health Canada has received a very small number of reports that might have something to do with the birth control pill, but also might be utter coincidence.”
The claim is about Yaz and Yasmin, two relatively new birth control pills on the Canadian market. The piece opens by “reminding” listeners that these pills have “possible links to deadly blood clots” [0:15].
Now, this is stupid for at least three reasons:
1) The links are not “possible,” they are known. All birth control pills have a risk of blood clots. That is known, and Yaz and Yasmin are no exception.
2) The error is worsened, however, by the implication that Yaz and Yasmin are more risky. They aren’t. What CTV doesn’t tell you is that women are most at risk of blood clots during the first year in which they start the pill. This can make Yaz and Yasmin look worse, because they are newer pills. Therefore, for a while, all the women starting Yaz and Yasmin are new users, because they haven’t been on the market long enough to have people who have used them for ages! So, older birth control pills look “better” when compared to newer pills, because older pills have a higher percentage of long-term users than newer pills. And, long-term users have already (presumably) successfully passed the highest risk period for developing clots, and so are less likely to do so than Yaz or Yasmin’s new users.
3) The piece also fails to properly contextualize these risks. Adults of reproductive age who do not use the pill have a risk of clots of 0.3 to 0.6 per 10,000. Birth control pills increase the risk of clots, but by how much? To about 1-1.5 per 10,000 (this is an increase in risk of 3-4x). To understand this kind of risk, consider that being pregnant increases your risk of a clot to about 13 per 10,000. So, you are at least eight times safer from blood clots by being on the pill than you are being pregnant.
But, since this misleading intro takes less than 5 seconds to say, the audience comes away already worried and primed to believe the worst. This just isn’t fair, or accurate.
Note what I am not saying. I am not saying that Yaz and Yasmin must be better, or are no different. I am saying that there is, at present, no evidence that they are worse than any other combined hormone pill (or, if there is, CTV has not presented it).
Correlation and causation
We are then told that now “Health Canada is also receiving reports of adverse reactions related to the kidneys” [0:53]. What does this mean? Well, anyone can send in an adverse drug reaction to Health Canada. I’ve done it as a physician. Anyone reading this can do it too. This isn’t “risks being discovered” about pills (as the piece’s title claims) but instead people who are taking the pill who report kidney problems.
Think of it this way—how many women take the pill? Lots. How many of those women might develop kidney problems? Some small amount. The question is, did the pill cause it, or is it just coincidence? At this point, no one knows—least of all CTV.
They then quote a pharmacist Lidia Molinara: “We don’t always know if it’s correlated, but it’s very coincidental”[0:57].
This is likewise idiotic. (I’m going to assume she misspoke.) Why?
1) We do know it is correlated. Correlated means something like “one thing happens and another thing happens too, with at least the appearance of a relationship.” So, the people took the Pill and have kidney disease. That’s correlation. They took the pill, had the problem, and so it appears (on the surface, at least) that the two events may be linked. This is what CTV is claiming.
2) The second rather silly thing is to claim it is “very coincidental”—what does this mean? Coincidence means things that just happen to occur together, but don’t have any causative link. So, what she’s saying is that we don’t know if they’re correlated (if they go together) but we do know that they are coincidentally together (i.e., they go together). This is utterly confused—the two statements almost contradict each other.
What I suspect she meant to say is, “We don’t know if it’s causative. It could be a coincidental correlation, but we don’t know if there’s causation.”
There is an old saying in science and medicine: Correlation does not equal causation—just because two things go together, doesn’t mean one causes the other. The sun rising is correlated with roosters crowing. But, roosters crowing does not cause the sun to rise (if anything, it’s the reverse).
This graph shows that the global temperature is correlated with the number of pirates in the world—as pirate numbers go down, global temperature goes up. But, that doesn’t mean pirates have anything to do with global warming, or vice versa:
Bad to worse
Having completely confused us about both previous risks and whether there’s any actual evidence of harm, things get worse.
The piece uses the efforts of US legal firms to launch class action lawsuits about components of the pill causing kidney problems due to potassium issues.
What they don’t tell you is that such lawsuits have not always been bastions of scientific reasoning. For example, consider that class action lawsuits against Dow Corning lead to bankruptcy protection over the issue of silicon breast implants.
Yet, there was never—and there still is not– any scientific evidence that the implants caused problems and they are now (surprise!) back on the market because plastic surgeons and their customers prefer them. This wouldn’t be possible if there was evidence of the harm claimed in the class action suit. So, Dow Corning employees, owners, and stockholders were taken to the cleaners over bogus claims of harm.
The complete non sequetur
Then, they go from bad to worse. Having claimed that the kidney problems are supposedly due to potassium issues, they then tell us that two women in Edmonton have reported problems but don’t want to be on camera. Those women are smart—I wish the CTV reporter had been as wise, because then she wouldn’t look completely clueless.
These two women have reportedly “suffered kidney infections while on the Yaz birth control pill” [1:34].
Kidney infections have nothing, repeat nothing, to do with potassium problems. I can assure CTV that many young women (the sorts who are on the Pill) tend to get bladder and kidney infections. This is certainly a correlation—but, I doubt very much that the Pill can be blamed for causing those infections. What’s the mechanism of action for the Pill causing it? It certainly has nothing to do with the supposed link to potassium that US class action attorneys are giving us.
“However,” CTV says, “neither of them [the two women] are sure if there’s a link.” Exactly. Nor should they be sure, nor should anyone claim that there is a link. Two women, of the likely tens of thousands in Edmonton taking the Pill, got a kidney infection. I would bet my house that at least two women in Edmonton who were not taking the Pill also got kidney infections.
So, let’s summarize. We have:
- A misleading headline
- A misrepresentation about the blood clot risks of birth control pills generally, and Yaz and Yasmin in particular
- A muddled understanding of the difference between causation, correlation, and coincidence
- Claims about potassium that are not substantiated in any way except with claims from people suing for large damages
- Claims about infections that are completely unrelated to potassium, and are a common problem likely to be seen in this (and many other) populations of patients.
In short, disgraceful. They only useful information was the doctor telling people not to panic. But, after all you’d heard, would that be your conclusion?
Again, I’m not claiming there could not be matters worth looking at, or that all pills are the same. Drugs should be monitored for safety and unexpected complications.
But, this kind of reporting misses everything that matters in understanding these issues, is confused and muddled, and contributes not one jot to a rational discussion of the evidence.
All we can hope is that women discuss it with their doctor, and not someone from CTV.