Epstein Barr and the cause of the cause – O’Reilly

One of the most intriguing news stories of the new year claimed that the The Epstein-Barr virus (EBV) is the “cause” of multiple sclerosis (MS), and suggested that antiviral drugs or Epstein-Barr vaccines could eliminate MS.

I am neither a doctor nor an epidemiologist. But I think this article forces us to think about the meaning of “cause”. Although Epstein-Barr is not a household name, he is extremely common. a good estimate is that 95% of the population is infected with it. It is a variant of herpes; if you have ever had mononucleosis, you have had it; most infections are asymptomatic. We hear a lot more about MS; I had friends who died of it. But MS is much less common: about 0.036% of the population has it (35.9 per 100,000).


Learn faster. Dig deeper. To see further.

We know that causality is not a one-time thing: if X happens, then Y always happens. Many people smoke; we know that smoking causes lung cancer; but many people who smoke do not get lung cancer. We agree with that; causation has been painstakingly documented in great detail, in part because the tobacco industry has gone to such lengths to spread misinformation.

But what does it mean to say that a virus that infects almost everyone causes a disease that affects very few people? The researchers seem to have done their job well. They studied 10 million people in the US military. 5% of them tested negative for Epstein-Barr at the start of their service. 955 of this group were eventually diagnosed with MS and had been infected with EBV prior to their MS diagnosis, indicating a 32 times higher risk factor than for people without EBV.

It’s certainly fair to say that Epstein-Barr is involved in MS, or that he contributes to MS, or some other phrase (which couldn’t be unreasonably called “weasel words”). Is there another trigger that only has an effect when EBV is already present? Or is EBV the only cause of MS, a cause that just doesn’t take effect in the vast majority of people?

This is where we have to think very carefully about causation, because as important as this research is, there seems to be something missing. An omitted variable, perhaps a genetic predisposition? Another triggering condition, perhaps environmental? The cigarettes were clearly a “smoking gun”: 10 to 20% of smokers develop lung cancer (not to mention other illnesses). The EBV can also be a smoking gun, but one that rarely fires.

If there are no other factors, we are justified in using the word “causes”. But this is hardly satisfying – and this is where the more precise language of causal inference clashes with human language. Mathematical language is more useful: maybe EBV is “necessary” for MS (i.e. EBV is necessary; you can’t get MS without it), but clearly not “sufficient” (EBV does not necessarily lead to MS). Although again, the precision of the math may be too much.

Biological systems are not necessarily mathematical, and there may not be a “sufficient” condition; EBV only leads to MS in an extremely small number of cases. In return, we must take this into account in decision-making. Does it make sense to develop a vaccine against a rare (albeit tragic, debilitating, and inevitably fatal) disease? Whether EBV is implicated in other diseases, possibly. However, vaccines are not without risk (or cost), and even if the risk is very low (as is the case with all vaccines we use today), it is not clear that it makes sense to take that risk for a disease so few get. How do you trade a small risk for a very small reward? Given the anti-vax hysteria surrounding COVID, requiring children to be vaccinated against a rare disease might not be bad public health policy; it could be the end of public health policy.

More generally: how to build software systems that predict rare events? This is another version of the same problem – and unfortunately the political decision we are least likely to make is not to create such software. The abuse of such systems is a clear and present danger: for example, AI systems that claim to predict “criminal behavior” based on everything from crime data to facial pictures, are already under development. Numerous are already usedand in strong demand law enforcement agencies. They will certainly generate far more false positives than true positives, stigmatizing thousands (if not millions) of people in the process. Even with carefully collected and unbiased data (which does not exist), and assuming some sort of causal link between past history, physical appearance and future criminal behavior (as in the discredited 19th century pseudoscience of physiognomy), it is very difficult, if not impossible, to reason from a relatively common cause to a very rare effect. Most people don’t become criminals, regardless of their physical appearance. Deciding a priori who will do it can only become an exercise in applied racism and prejudice.

Virology aside, the Epstein-Barr virus has one thing to teach us. How to think of a cause that rarely causes anything? This is a question we must answer.

About Margaret Shaw

Check Also

Ben Affleck, 49, cuts a youthful figure as he steps out showing off his buff arms while smoking

The JLo effect! Ben Affleck, 49, cuts a youthful figure as he steps out showing …