Can the decision to refuse vaccination be truly ‘personal’?

“So, so tired of seeing children dying of measles. Our measles ward has overflowed and using an additional 15 beds. One to 2 kids dying every day from measles. Another one just started seizing as measles attacks his brain. It’s a truly terrible disease. I had forgotten since it has been uncommon in the US since my childhood.”

This poignant Facebook post came from my wife’s cousin, a physician volunteering in Papua New Guinea, which is suffering from a measles epidemic. It reminds me of a recent conversation with a friend. She mentioned that she and her husband have chosen not to vaccinate their daughter, and that “this was a very personal decision for us.”

This goes to the heart of how we think about evidence and about risk in an uncertain world. And is it possible for such a decision to be truly personal?

Let’s start with the evidence. Diagnoses of autism spectrum disorders have been on the rise. While partly a matter of an expanded definition and better diagnosis, this doesn’t really matter—no parents want to place their child at risk of autism. Vaccinations and the diagnosis of autism spectrum disorders can occur about the same time in the lives of young children. Coincident events—particularly around something as serious as autism—should begin a process of exploration. It is important to ask if the co-incidence is due to causation or if it is simply a “coincidence” without a causative link.

A study suggesting that the MMR vaccine was linked to autism was published in The Lancet, a British medical journal, in 1998. This important finding spurred other studies, but they did not confirm the connection claimed by surgeon and researcher Andrew Wakefield. A 2004 expose by the London Sunday Times uncovered evidence of a conflict of interest—Wakefield held a patent on a rival measles vaccine—and other problems with the study. In 2010, the British General Medical Council found Wakefield guilty of 36 counts of medical misconduct. He was barred from practicing medicine and The Lancet retracted the study.

In the interim, Wakefield’s study lit a firestorm of controversy over vaccines. Given the significance of the issue, the safety of vaccines has become one of the most studied issues in the history of public health. A readable history of the controversy appears in the journal PLoS Biology (see all the studies referenced here at http://goo.gl/x3dPxT).

At this point in time, there are claims & studies on both sides. How do you know whom to believe? The Institute of Medicine (IOM) was established to address matters like this. The IOM is the health arm of the National Academy of Sciences, which was chartered under President Abraham Lincoln in 1863. The IOM reviewed all of the studies, rated them for scientific validity and published findings in 2011. A follow-up study, published in Pediatrics this month, takes the IOM study further. The authors did an exhaustive search of the literature and summarized the findings of all of the reports that had gathered hard data and conducted statistically rigorous analysis.

The preponderance of the evidence shows no statistically significant association between vaccines and autism. I’m a numbers guy—I’ve taught statistics at the college level and use statistical tools in my work. To statisticians, nothing is ever proven. The question is always posed as “How likely is it that we would observe these data, if our hypothesis were true?” Here the hypothesis is that the association in time between an autism diagnosis and vaccination are due to causation instead of coincidence. Causation is more than “guilt by association”—lots of other factors influence the diagnosis and a reliable study will take all these related factors into consideration and have enough cases to “tease out” all of the relationships. The IOM and Pediatrics reports tell us that studies with a large enough sample and a rigorous methodology have failed to support causation. If vaccines do cause autism, the patterns observed in these careful tests are very unlikely.

Yes, there is always a chance that there is a connection that did not show up in these many studies. To a statistician, tomorrow’s sunrise is only highly probable. What we can say with confidence is that if the association were at all common, it would be revealed in these studies. Thus the association is either rare or nonexistent.

Let’s consider the alternative. Vaccinations DO prevent disease. We know that the MMR vaccine has dramatically reduced the incidence of measles, mumps and rubella. Before the vaccine, nearly everyone contracted measles before age 18. As the vaccine is, on average, 90% effective, widespread vaccination keeps isolated cases from spreading. We’ve forgotten that measles can cause serious complications, even death, as the post from my wife’s cousin attests. Before 1963, the United States saw 500 deaths per year. With widespread use of the vaccine, deaths from measles are unusual in the U.S.

Rubella (German measles) can be also very serious. About half of children born to a woman who contracts rubella in the weeks before conception or during the first trimester of pregnancy will be afflicted with “congenital rubella syndrome.” Miscarriage is also common. About half the children with CRS suffer from congenital heart disease, over half are deaf and 43% also have serious vision problems including blindness. Helen Keller’s deaf-blindness was the result of rubella. CRS is also associated with autism.

There was a time when parents could choose to leave their child unvaccinated, confident that the vaccinations of other children would be protective. As the claimed association between vaccines and autism persuades parents to avoid vaccines, we’re seeing these diseases re-emerge. A measles outbreak in France in 2011 involved over 15,000 cases. The number of measles cases in the United States remains low, although the Centers for Disease Control reported 17 outbreaks in the first half of 2014 with 554 cases, three times the total for all of 2013 and the highest since 1993. See http://www.cbsnews.com/news/measles-outbreak-in-orange-county-california-worst-in-decades/.

We don’t live in a world without risk. Without vaccines, children are at risk of a range of potentially serious diseases for their entire lives—a growing risk, if more and more parents refuse vaccines. And the risk of autism spectrum disorder from vaccines is undetectable by the best statistical analyses, thus is either rare or nonexistent. This should not be a difficult decision.

As leaders in their community, the decision of my friend and her husband will influence others to refuse vaccines. And even if their daughter suffers no harm from contracting, say, rubella, she could contribute to the spread of the disease to others who may be harmed. Nor is the risk limited to individuals whose parents refused vaccination. As the vaccine isn’t perfect, a small share of the vaccinated are also at risk in an outbreak—the European Centre for Disease Prevention & Control reported that 18% of those who contracted measles in had been vaccinated. Now back from Papua New Guinea, my wife’s cousin reports that 80% of the deaths were among babies too young to have been vaccinated.

The decision to refuse vaccination cannot be “personal,” in the sense that it affects only oneself or one’s family. As this misconception of the relative risk of vaccines spreads, the risk for all grows.