The problem with causation

Chronic fatigue syndrome (CFS) patients are making a name for themselves these days, and not in a good way. The current brouhaha over new studies that have been submitted, then held, then whatever else is going on is bringing out the worst in the XMRV cheerleaders. I recently wrote about compassion for patients who are not in a good place to think critically, but even my own compassion is stretched to the limit, and frequently breaks, when it comes to the screaming impatience and conspiracy theorizing that can be seen on sites like Phoenix Rising.

Some CFS patients, let’s say the extremists, expect XMRV to be a combined Messiah/Holy Grail/panacea for all their ills. As I’ve mentioned before, some are even taking antiretroviral medication already, something that is highly dangerous and is not currently recommended by any reputable physician. The problem stems from an overall difficulty I see in chronic illness communities: the correlation fallacy. Which is to say, “correlation does not imply causation.” This simply means that just because two things appear to be connected does not necessarily mean that they are.

CFS, like many chronic illnesses, is a concatenation of unpredictable symptoms. While it’s common for exertion to push patients into a crash, we might also stay in bed all day, and the next day be crashed for no apparent reason. (In this context I use “crash” like many patients, to mean an especially disabling flare-up of symptoms.) This is one of the most annoying aspects of the illness. Quite a while ago I described exactly that situation to my doctor, and asked why it happened, and his answer was this: “You have CFS.” Right. That’s the way it goes.

In addition, like myself, many people with CFS have other conditions that either exacerbate our symptoms or result in fun new ones. So, although I must reiterate that there are specific criteria (well, half a dozen sets of criteria, but that’s another story) for diagnosing CFS from symptoms, most if not all of us experience symptom weirdness and ups and downs without rhyme or reason.

And yet, patients happily share stories of great success with dubious or even flat-out quack treatments. “I started using this herb, and the next day I started feeling better!” someone will chirp. Despite the fact that the selfsame person may recently have been bemoaning the capricious nature of our illness and how it’s so hard to make plans when you never know how you’re going to feel (true), now it’s become perfectly clear that taking this herb caused an upswing.

“Correlation does not imply causation” is one of the clarion calls of the critical thinker and the evidence-based physician. And yet these false correlations happen all the time. As you know, I try to come from a place of understanding, but the cognitive dissonance displayed in the above example really blows right past me. Whether it’s low-dose naltrexone (LDN), CoQ10, or even conventional pharmaceuticals, it’s usually pretty difficult to determine (and especially not in the space of a couple of days) whether any one treatment caused a major improvement. And that’s not even taking into account the placebo effect.

I don’t begrudge people feeling better. Nor do I discount the placebo effect for people who cannot get relief any other way. But what sometimes happens next is that other people who have had similar experiences will post about it as well. And then the other major fallacy found in chronic illness communities arises, where patient anecdotes are mistaken for hard data. Mostly, this isn’t too big of a deal. After all, science has so far failed us, so we do turn to others to find out what has helped them, and possibly try it ourselves. As I’ve said many times, I’m not entirely against this. The problem happens when people come to consider this information to be as valid as evidence-based medicine. Communities are wonderful for commiserating and sharing stories, but relying on these stories as a source of scientific data is dangerous.

Especially when it comes to XMRV. I’ve seen people deciding that everyone on a FB page or a forum should get tested for XMRV “and then we’ll have more results.” No. Those are still anecdotes, and given that the evidence-based studies can barely get it together on how to select a patient cohort and all the other complicated issues with testing CFS patients (such as severity of disease, which isn’t always taken into account), it’s ridiculous to imagine that half a dozen tests on people from all over the country or the world will contribute to a scientific body of evidence.

Finally, something I see missing from an enormous amount of XMRV discussion is the fact that even if it is shown to have a strong correlation with CFS, that does not imply causation. There isn’t even solid evidence of a correlation yet and people are jumping all over causation, with the antiretrovirals, rushing out to get expensive and useless tests, worrying themselves to death over whether they’ve spread the illness to others, and so forth. It’s perfectly likely that even if a correlation is found, the retrovirus may simply be a dormant infection to which people with CFS are especially susceptible. Or any number of other possibilities besides XMRV being a causative agent. I wish I were seeing less anxiety and more patience.

I get that some of the hoopla doesn’t even really have to do with medicine. It has to do with perception. The notion that a virus causes CFS is enormously attractive to patients who have been dismissed, invalidated, and marginalized for years. Proof of a somatic cause would mean even more than a cure — it would mean no more people trying to, say, cure CFS with neurolinguistic programming (NLP), as a fellow patient related to me the other day. The hope for that may be even stronger than the hope for a cure.

But again, hope, belief, fear…these do not aid in the calm assessment of the scientific process or of the data it (eventually) provides. Patients are falling back on emotion when they should be hanging on to rationality. No, it’s not fun. It’s easier and more comfortable to get together with other patients and howl instead of making the effort to think things through with a clear head. But that effort is absolutely crucial, especially considering the very negative views on our community that are starting to develop due to a visible and vocal segment of it. We’ve spent all this time trying to get attention, and now that we have it, we’re being seen by outsiders as a bunch of hysterical loonies. This is why I continue to plead for rationality and critical thinking in the CFS community.

Post to Twitter Post to Delicious Post to Digg Post to Facebook Post to Reddit Post to StumbleUpon

7 Responses to The problem with causation

  1. Pingback: Tweets that mention The problem with causation --

  2. Funny, I also posted an article today where I used the quote “correlation does not imply causation”.

    I do not know what XMRV will eventually mean to ME/CFS – the jury is still out, but the original paper by WPI is probably the most hyped science paper ever. Which I don’t consider to be a good thing. Science in general could use some more attention. ME/CFS deserves better research, science based research, and attention. Not the wishy washy psychobabble. But hype? No.
    Having a background in exact science I have a lot of questions regarding the research on ME/CFS and the way patients are treated (GET, CBT, between the ears, problems with disability, …), but the controversy surrounding the new studies and the conspiracy theories it spawned just turned me off. I decided I needed some time off, away from Facebook.

    TAM looks like fun, enjoy yourself. The London line-up for October looks very tempting with Alan Moore, Cory Doctorow and Stephen Fry … and James Randi & Richard Dawkins of course.

  3. Patients only want a cure. If XMRV is the cause for all or some of them, hopefully that will lead to a cure. So far only an association has been suggested. CFS patients come from all walks of life, and all have different opinions on this subject. They do not speak with one voice, except when asking for good biomedical research to be undertaken. They should never be spoken about or treated as if they are irrational, that’s an individual problem.

    The issue of the NIH/FDA XMRV paper is very different. Simply, the Government should not be interfering in scientific progress. Let the science speak for itself.

    • Absolutely I agree CFS patients do not speak with one voice. I’ve tried to make it clear that the people who worry me are a particular — and particularly vocal — subset of patients, but if I didn’t, then I apologize. I am very happy to know rational and critical thinking people with CFS and other chronic illnesses (such as your fellow commenters on this post!).

      It’s simply far too early to be assuming causation or even expecting it when there’s barely any correlation yet. But I am a skeptic not a cynic, and if the science winds up proving both, I will be thrilled to embrace that.

  4. Nice post, Joey! I’ve noticed similar behavior in other communities. As far as I can tell, no one really knows how anxiety, depression, and IBS correlate but only that they tend to come together. Perhaps anxiety can upset your stomach but also not being able to digest anything properly can surely make you depressed or anxious. We could use some “rationality and critical thinking” in just about every community but I definitely see your point here as well.

  5. Well said (or written).

  6. Pingback: A future I don’t want to see | Newly Nerfed