As people close to me will know, I recently dove into the task of researching vaccines. This is in honor of the upcoming birth of my first child! I feel that the possibility of vaccine injury is real. Based on that feeling, it is my duty to do my due diligence to understand any and all documentation of risks. I have already met much resistance from friends and family on this topic. Why is it that questioning the safety and efficacy of vaccines draws such an emotional response in nearly every case? I will save that topic for another post. For now, I am focusing my intention on two objectives. First, I want to find the best information to use in making my decision on when and how to vaccinate our daughter. Second, I want to share any information which I find in hopes that it will be a positive resource to other parents going through the same process.
Today, after posting a vaccine related link on facebook, a friend replied with a link to a study on the MMR vaccine. From what I understand, this is the vaccine surrounded by the most controversy. After reading the study’s abstract I felt compelled to purchase and digest the full text. For the price of two books, I received 4 pages documenting this scientific endeavor. The title of the study is:
“Autism and measles, mumps, and rubella vaccine: no epidemiological evidence for a causal association”
I found the title to be a little misleading. In this study, it appears to me that they have endeavored to statistically compare medical records of children in a particular region of England. They gathered up a number of children diagnosed with Core Autism, Atypical Austism, and Asperger’s Syndrome and analyzed the cases on a couple of metrics:
1. Overall number of occurrences of the three conditions
2. Temporal relation of three dates: vaccination, parental concern, diagnoses
On the first point, they find no statistical change in the fitted curve that they applied to the sampled data. They graphed the occurrences of each disorder for children born from 1979 to 1992. I believe the objective here was to prove that the introduction of the MMR vaccine in the UK in 1987 was not significant enough to change the rate of disorder incidence. However, since the number of subjects in their study was less than 500, and most of them were born after 1987, I’m not sure it would have been possible to see a large difference in the rates of these disorders in their data.
On the second point, the researchers say that there is essentially no statistically significant relationship between the three dates of vaccination, parental concern, and diagnosis. They do mention that around 5 or 6 months there was a lot of parental concern showing up. However they attribute that to the fact that peak vaccinations occurred at 13 months and the peak parental concern occurred at 18 months. However, this correlation appears to be dismissed by the researchers. Why it is dismissed, I am not sure.
My own conclusions after reading the study
This study appears to have been conducted in order to refute the controversial claims made by ‘Wakefield and colleagues’. I have not read the Wakefield paper on MMR and so cannot comment on it. This study by Taylor and others does not, in my mind, refute the idea that the MMR vaccine could cause injury to a child and that this injury could later lead to autism. They state in their findings that the time between vaccination, parental concern, and diagnosis does not show a reliable relationship. I’m not sure that injury due to a vaccine would have to be predictably occurring at a certain number of months after administration. I think that it is completely possible that the impact of vaccination in children who are injured by it could be sub-clinical for some time and that this time could be different for each child.
About the expectation of a large increase in the rate of these disorders (core autism, atypical autism, and aspergers syndrome) at the introduction of the MMR vaccine, I think that the amount of increase in rate would have to be vary large to be detected by this study. Statisticians out there, please correct me if I’m wrong, but I’m thinking that if the rate of autism is 1 in 1,000 (which would be higher than the rate cited in this study from 1999) you would have to have many thousands of subjects in your study to detect a doubling in the rate from 1 in 1,000 to 2 in 1,000. Unless I missed it, this study does not cite how many individuals were analyzed in the study. In any statistical analysis, wouldn’t they clearly state the number of individuals that they had to examine in order to find this number of cases? Or perhaps they were unable to determine the total number of births in the geological region of the study? I would think it should be easy to tabulate the number of births and simply say how many were included per year overall. In fact that brings up a good point that I had not considered until now. They graph the number of cases found each year, but not the rate of cases. They are graphing the raw number of children with these conditions, but not the relation of the number of children with the condition to the number of healthy children. I wonder why they made that choice. It seems like a significant difference to me. If we are examining the number of cases diagnosed over a certain period, shouldn’t we also account for population size and growth?
Another limitation of the study that I think is worth mentioning is that they do not include any other parameters such as the other vaccinations that a child was given. Since we’re analyzing vaccine affects here, I think that a complete analysis should include some comment on the other vaccinations received by the children.
I hope that anyone reading this article will find my discussion to be interesting and informative. I am practicing my ability to critically read these studies in order to more fully understand the science around vaccine safety.