There is a lot of discussion on the internet these days on whether of not kids should get vaccines. Depending which blog you read or what page you visit, vaccines can either save lives or cause autism. Considering I have a 6 month at home, the topic is of interest and since a lot of the information on the internet is very subjective or not well researched, I felt it was worth a good discussion. I want to preface this by saying I have I will do my best to present qualitative and quantitative analysis and limit my own judgement. I doubt I will succeed 100%, but lets see how this goes.
As an engineer, I like to worth with numbers and data. So lets start with some numbers on vaccines in the United State. Per the Center for Disease Control (CDC) between January of 2006 and December of 2015, over 2 billion vaccines were given (2,845,946,816 to be exact). This included numbers for 31 types or varieties of vaccines, therefore the numbers may be higher if we consider some non-typical vaccines (typical ones include, but not limited to MMR, TD, influenza, HPV, Hep A/B/HIB). Lets keep this number in mind as we continue the discussion.
With a baseline of vaccines given over a period of time, lets now consider how many court cases have been reported regarding issues or deaths with vaccines. This should give us an idea of how many major issues have occurred where at least the plaintiff identified vaccines as the cause. Not perfect, but we have to start somewhere. Per data presented by the National Vaccines Injury Compensation Program (VICP) between 2006 and 2015 4724 cases were filed. While many cases were dismissed, lets consider this number in full so we have a conservative look (at least based on this first set of data). Looking at the numbers of cases filed versus the number of vaccines given during that period, the percentage that a single vaccines would have caused a condition severe enough to file a case is 0.00017% (or around 1.7 incidents out of 10,000 vaccines). While nothing is certain and there is always risk, a 0.00017% change of having an issue from a vaccines is a pretty safe bet. Even is we assume the actual number of double, or even tenfold, that percentage is still pretty favorable. Of course this is just one data source, so lets continue to explore.
Autism is another “hot” vaccine topic. It has been claimed that vaccines have (and continue) to contribute to elevated caused of autism. I spent some time looking through medical journals via my wife and I’s graduate access to various library. That research yielded no medical studies or reports that link vaccines (MMR getting the most scrutiny) to autism. The only medical journal article that claimed a link between the two (Lancet by Wakefield printed in 1998) was later retracted by 11 of the 13 authors and was scrutinized heavily by the medical community in general. The retraction was discussed in details in multiple journal articles, stating a lack of evidence and proof. Now does this lack of journal articles mean there is no link between the two? Of course not. It does however reduce the probability. Any graduate student will tell you that if there is a topic you can write about and even loosely prove, you can publish it. The lack of articles then at least indicates on some level that any proof of a link of vaccinations to autism is very hard to scientifically prove, even loosely. This is of course more qualitative analysis. No numbers involved. So lets go back to some numbers.
Per the CDC, 1.5% of kids are diagnosed with some level of autism (based on data of 8 years old kids in 11 communities). The CDC also indicates that depending on the vaccine that somewhere between 72-93% of children get vaccinated (depending on type of vaccination and age). Therefore if there were 100 kids, 72-93 of them would be vaccinated and of those 1.5 would have some level of autism. Assuming the low end of the vaccinations to give us a conservative number, of the 72 kids, 2% would have autism. Now of course it is not reasonable to say that all kids with autism got it from vaccines. Therefore that 2% is not a true number and the real percentage would be a fraction of that. Of course, that is using conservative numbers (instead of a probability distribution function or PDF) and assumes there is some link between autism and vaccines. Again, this analysis becomes difficult because we have no solid data on any actual link. We can only speculate based on individual cases that there may be a link, but the probability of that link cannot be boiled down to a number. Any internet site or blog that indicates otherwise should be questioned. As I will discuss later, studies are a hard thing to do right and generally most websites fall far short of the mark.
One last thought here on autism is that one major argument on the vaccine link is the increase in autism over time. In 2012 the number was in the neighborhood of 1 in every 68 kids. Go back to 2000, that drops to 1 in every 150 kids. Depending what you read that number was 1 in a few thousand back in the eighties. While the argument is valid since vaccines have continued to increase over time, what should also be considered is how we assess autism. The methods for diagnosing autism have evolved greatly over the last few decades and are subjective and heuristic. What does that mean? It means we are able to diagnose children with autism better now when we could even five years ago and that the diagnosis methods leave much in the hands of physicians. Therefore since we are able to diagnose better and physicians have better knowledge and the ability to use their own expert judgement, it should be expected that autism rates rise. One interesting point to consider is that mental retardation and mental disability diagnosis have fell as autism has risen. If you consider prior diagnosis for something such as mental retardation it can be seen that many of the diagnosis key points are now associated with autism. What that should indicate to you is what we used to diagnose as a mental illness, such as retardation, we have reclassified under autism. Again does this mean there is not a link between autism and vaccines, of course not. This does however provide a possible explanation to a topic that drives that discussion. When you don’t hard hard empirical data, you are stuck with quantitative analysis (basically subjective). One could interpret the data above and come to conclusion that autism is on the rise and there is no link to the drop in other metal illnesses. Another could come to the conclusion I have come to above. Again, the conclusion is subjective to a persons thoughts, knowledge, and usual stake in the topic. The analysis presented above is done without a stake in the answer and is based on numerous journals and my own thoughts on probability. Take it with a grain of salt of course. We could continue with more topics against vaccines, but lets change gears and discuss why vaccines may not be a bad idea.
When talking vaccines, the important things that should be considered is the concept of herd immunity. Simply, if a large percentage of the population is immune or treated for a disease, the probability of having an outbreak is small. As mentioned before CDC has at least 31 vaccines that they track. These are diseases that have been researched and fought over the years. I wont go into detail here, but the history of vaccinations is fascinating. There is also a book on the history of Bellevue hospital in NY that discusses this topic a bit that can really be interesting. Boiled down, vaccines have in part helped us to live longer lives. We don’t have outbreaks of measles or mumps much anymore and when was the last case of polio you remember in your community? The Medium Corporation aggregated data on the topic of vaccines and created some great graph showing the decline in deaths due to certain virus (The Polio graph on the top of the blog post was also developed by them). The impact is clear, especially in graphical form, that vaccines do make a difference in deaths. Here is a sample graph from that site and a link if you want to see the rest.
Nothing is certain of course and cases do happen, but probabilities are small compared to decades ago. Why I mentioned the idea of herd is that choosing not to get vaccinated is a choice that affect your kids of course, but is also affects the population. While the effect is not completely known, it had been hypothesized that if the herd immunity drops to far, chances of outbreak continue. Even though who have gotten shots may be at risk because nothing is 100% and vaccines are not always effective in all kids. This is not a statement to push you to get vaccinated, but it is something to consider. Remember everything comes with risk.
One semi-related point I want to make before I summarize this whole discussion is on the “studies” that can be found on the internet and how to digest them. Lets say you are looking up a medical topic, good chance you can find dozens of studies on the internet. The question you should ask however before taking any of them to heart is, how valid are they? There is a saying in programming, “garbage in-garbage out” and it applies great to many internet studies. A study is actually difficult to do right and data can be easily manipulated. For example its easy to pick 100 or 300 people to study and have the results favor your hypothesis. All it takes it a good choice of population. Therefore if you don’t start with valid data, your results cannot be considered valid. If you see a single study, not published in any medical journals, that has a stark difference in data then others, question it. Especially if that study is found on a website that is promoting a certain cause or outcome. I guess what I am advocating is, remember there is no checks and balances on the internet, anyone can post anything. If you are want data and studies that have been checked and reviewed and at least have been through some checks and balances, visit your library. University libraries have access to hundreds if not thousands of journals world wide. There is a wealth of info there. Also consider the author and his sources. I always recommend looking them up and checking sources. Its easy to put sources that don’t exist.
I know this post was long and I appreciate you sticking with me (if you made it this far). The bottom line of all this is that the data presented right now does not conclusively show any link between vaccines and autism. The link cannot be completely dismissed, but probabilities at least show the chances should be very small if there is a link. In the end, its up to the parents. If you don’t feel comfortable, then its your right to not do it. A word of caution though, vaccines were designed for a reason and if your kiddos come down with something you could have been protected for, that is on you. Also remember that you are not only accepting that risk for you kiddo, but you may be increasing the risk for others. Not a guilt trip or anything, just something you should consider. No choice here comes with zero risk. Personally I like my chances on the vaccine side and will continue to make that choice. As always, my wife and I will stay informed and if things change, then perhaps so will our minds.