I helped write English textbooks awhile back, and in doing so I learned about common mistakes made in presenting arguments. Among the most common—you see it in politics all the time—are incorrect cause-and-effect statements.
We assign causes regularly, but they’re not always valid. Just because two things co-occur doesn’t mean that one causes the other. For example, the “Low Fuel” light coming on doesn’t cause your car to run out of gas shortly thereafter. The two events may be associated with one another, but there is no cause-and-effect between them. And while that example is a simple one to understand and discount, the potential for confusion increases with complexity.
The concept is of particular importance in genomics research and its applicability to medicine. Researchers are coming up with a ton of associations between genes and disease states, but there are still very few proven causal relationships. The importance of this distinction was recently underscored by the mainstream media’s coverage of a paper showing an association between a father’s age and the incidence of autism in his children.
First, we have two things co-occurring. One, research indicates that the incidence of autism increases with a father’s age. Two, the number of mutations passed from father to offspring also increases with age, as shown by the recent paper in Nature that sparked the headlines. For raw numbers, the average dad (age 33) passed along 55 mutations to his offspring. But aging produced a pretty steep curve of increasing mutations—if age 20 is the baseline, a 36-year-old father will pass along twice as many mutations and a 70-year-old roughly eight times as many. And many of those mutations turn out to be associated with autism spectrum disorders (ASDs).
It makes a nice, tidy package, doesn’t it? Put two and two together and you have the conclusion that the mutations that accumulate with age cause the rise in ASD rates for older fathers. And it also correlates nicely with the increase in ASDs and the increasing average age of fathers over time.
Mainstream media jumped on it, and the headlines that came out played up the cause angle. Take, for example, CBS News: Study: Father's age a likely factor in autism. The online Daily Beast “Lifestyle” section goes a little farther: Study Says Older Fathers Linked to Autism, Gives Mothers a Break. Some general audience sites questioned whether young men should start banking sperm based on the report: Older men’s sperm linked to autism? And a post on at least one site, Slate, went as far as to advocate doing just that: Dude, bank your sperm (though that one is extreme enough to have me hoping that there’s some satire in there somewhere).
The trouble is, there is no proven causal link. There may be cause-and-effect at work to be sure, but research is in the early stages, and there are many other factors at play here. The media reports do use some hedge words, but for the general reader, the cause-and-effect implication is clear. Contrast this with Nature’s own coverage of the paper by Ewen Callaway: Fathers bequeath more mutations as they age. This is a fact, and it’s really all the paper has shown. Callaway goes on to speculate about the possible mutation/ASD link, but within the context of undefined correlations and the necessity of further inquiry. I look forward to seeing the associations clarified and to see whether the increase in mutations with age really does lead to cases of autism. Or if there might be something else entirely that plays the causative role.
As more samples are collected, genome sequences performed and analyses completed, there will be many more headlines about gene/disease links. Some of them will be about proven cause-and-effect relationships, but, especially in the near future, they will be in the minority. So as you sip your coffee (which may be good for you, bad for you, help prevent Alzheimer’s disease, etc.) during your morning news surfing, don’t choke if you see a provocative headline that suggests that you scurry down to your local sperm bank. In all probability, it’s taking an intriguing association and prematurely promoting it as a cause.