Before learning about the “birds and the bees” we may have been told how the stork brought us, as a little baby, to our parents. Even with a minimal interest in the animal kingdom of storks, birds and bees, we likely started to question this curious story.
That is until we heard this news about Denmark: Post-1960 there was a significant decline in the number of nesting storks in Denmark. Also, beginning in the late 1960s, Denmark started recording its lowest average number of childbirths per woman. In short: fewer storks = fewer babies.
Here rests one of the fundamental errors in debates, research and uncritical thinking: confusing correlation with causation. So powerful are spurious relationships that they can sometimes have significant public policy implications. Consider the story of autism and vaccines.
In 1998, The Lancet, a respected medical journal, published Dr. Andrew Wakefield’s research claiming a link between autism and the MMR (measles, mumps, rubella) vaccine. Ever since, people in the autism community have raised concerns about live-virus vaccines and their children’s health. Fueled by the popular media, in particular the Internet, Dr. Wakefield’s research has resulted in a decline in vaccinations and, some say, a resulting increase in childhood diseases like measles.
However, on Jan. 28, Britain’s General Medical Council concluded that Dr. Wakefield acted dishonestly, unethically and irresponsibly when carrying out his research. And on Feb. 2, The Lancetsaid “we fully retract this paper from the published record.”
Peter Nardi discusses how to use our critical skills to avoid scams, respond to rumors and debunk questionable research.
Although it’s reasonable to have some concerns about the many ingredients that go into vaccines and other medications, it’s still important to look more closely at the specific issues raised by the MMR vaccine and autism research, and use our critical-thinking skills in understanding what is going on. When assessing research, it’s important to evaluate several elements: the sample, the quality of the data collection process (such as survey item wording or interview style), and how the data are analyzed (appropriate statistics and charts).
Let’s begin with the sample: The original Wakefield study took blood samples from only 12 English children who were attending his son’s birthday party. They were each paid the equivalent of around $8. Already, we begin to question the quality of the research when such a small sample is used. It’s also important at this point to consider any ethical questions about paying the children studied and how they may have been affected by having invasive blood samples taken.
For research to carry any weight, replication is essential, and studies with larger and better samples have not demonstrated a correlation between vaccines and autism.
Furthermore, before cause and effect can actually be declared from a correlation, a timeline must demonstrate that the cause came before the effect. For example, students who study more tend to have higher grades. But does studying lead to higher grades, or do those students who have higher grades (maybe who are smarter to begin with) tend to study more to ensure continuation of a high GPA?
When reviewing how researchers collected the data, assessing which data occurred when is important. In many cases, it turns out that autism appeared before the vaccinations were administered.
In analyzing the data collected, in order to claim a cause and effect, review how the research eliminated alternative explanations. Do changes in industrialization and urbanization in Denmark, for example, connect to a decline in the stork population as well as to changes in family life and fertility? Spurious correlations are easily addressed by searching for a third explanation. The appearance of autism tends to occur between the ages of 2 and 5, the same period when vaccines are administered.
Just because there is a societal increase in autism rates coinciding with an increase in the distribution of the MMR vaccine, it does not indicate a cause and effect relationship, especially if autism rates continued after thimerosal (the mercury-based preservative hypothesized to be at fault) was removed from the vaccine.
Increases in autism rates could be due to other explanations such as changing definitions of autism and better diagnosing techniques, thus illustrating how other variables can create the illusion of a correlation between immunizations and autism. Other studies also indicate that boys are about four times as likely to have autism despite similar rates of vaccination.
Finally, a major study in 2002 of almost half a million Danish children found no difference in immunization records between those children with and without autism. To date, there is no scientific evidence in the published literature of a causal connection between immunization vaccines and autism. And thanks to Denmark we have the research on this spurious relationship between autism and the MMR vaccine — and, of course, on storks and childbirth.
Sign up for our free e-newsletter.
Are you on Facebook? Become our fan.
Follow us on Twitter.
