Okay, I’m back. I really did fracture some ribs and I have a partially atelectatic lung. The prognosis is good as long as I keep breathing deeply, but it hurts like hell and I can’t sleep. I’ve been gone for about ten days and the world has gone berserk on the MMR vaccine & autism issue. Once more into the breech dear friends, once more, or close the wall up with our measles dead. As light background reading away from the fray, I recommend “Mercury, Vaccines, and Autism: One Controversy, Three Histories,” which was published in the American Journal of Public Health in the February 2008 issue. The author, Jeffrey P. Baker, MD, PhD, of Duke University, writes as a historian of the controversy in hopes that “…an understanding of this history provides important lessons for physicians and policymakers seeking to preserve the public’s trust in the nation’s vaccine system.” There seems to be a strong hint in Professor Baker’s paper that in early 2008 we were entering a cooling off period in which it might be possible to “take a first step toward transcending the powerful boundaries shaping today’s vaccines controversies.” No such luck.
I would like to dive right in and talk about the apparent rising frequency of autism, the so-called “epidemic.” I’m sure many of you know the arguments about how this autism epidemic might relate to the MMR vaccine. The simple argument goes like this: (1) There has been a drastic increase in the number of cases of autism over time; (2) this must be related to some new environmental exposure that didn’t exist before; (3) the only exposure that anyone’s come up with that seems to have coincided in time with the autism epidemic was (a) in the United Kingdom the MMR vaccine (b) in North America more specifically the preservative thimerosal in the MMR vaccine. There are a lot of problems with this line of reasoning. In this first post I’d like to show you that there probably has been little or no increase over time in autism.
There have been many epidemiological studies of time trends in autism rates among children born in the last three decades. All of the studies have reported increases. For each these studies, there seems to be one or more commentaries or reviews arguing that the apparent increase is an artifact due to changes in diagnostic practices, etc. In fact, I should make it clear that authors of the time trends studies themselves often concluded that the increases they found were due to such changes in diagnostic practices, etc. However, the most elegant paper I’ve come across on this topic was published in in the June 2007 issue of the Journal of the American Academy of Child and Adolescent Psychiatry. The three authors are Dr. Ashley Wazana of the Department of Psychiatry at Montreal Children’s Hospital, and Drs. Michaeline Bresnahan and Jennie Kline of Columbia University’s Department of Epidemiology. The authors looked at 12 studies that reported on time trends in the frequency of autistic disorder. Using a rather sophisticated modeling technique called prediction analysis, they concluded that observed increases in autistic disorder over time can be explained by three phenomenon:
- “Diagnostic criteria have changed over some part of the period during which increases have been observed.” The diagnostic criteria for autistic disorder were broadened over time.
- The average age of diagnosis for autistic disorder became younger (for example, in California, from about 7 years old for children born in 1987 to about 3 years old for children born in 1994 [Croen & Grether 2002]).
- Diagnostic substitution: “The efficiency of ascertainment (the probability that a true case is identified) has increased with greater awareness of the condition, introduction of new treatments and new resources, advocacy, broadening of diagnostic experience, and changes in diagnostic practices…”
Congratulations to Drs. Wazana, Bresnahan, and Kline for a most enlightening paper. The title is “The Autism Epidemic: Fact or Artifact*?”
Next post: Kudos to South African 24.com, for their story Rise in autism over-estimated? They seem to be the only media outlet to have reported on a paper in the May issue of Developmental Medicine and Child Neurology on Autism and diagnostic substitution.
*An artifact is (1) an error introduced by measurement methodology; or (2) a data error caused by the instrument of observation. In this case, measurements improved over time.