The Risk of Vaccinations from the Perspective of a Risk Analysis Nerd

Risk is something that I think about on a daily basis in my work. Every dam in our inventory goes through risk analyses at varying degrees of detail, and the results of those risk analyses are used in the overall program to rank which dams get the highest priority for further study or modification.

Snoozing yet? It's dry and it's what I'm typically complaining about on Twi.tter when I bemoan the fact that I'm writing yet another report. A lot of engineering judgement along with data goes into the final number, and dots on a chart are the result of groups of engineers and geologists gathering for several days to bash their heads against the table  wade through data and analyses, argue and ultimately come to a consensus. Good times. You know most engineers are introverts, right? It's an exhausting process.

Full disclosure of my inherent bias: I vaccinate my kids on the recommended schedule, and all mine and S's vaccination boosters are up to date.

Anyway, as we were stuck in a traffic jam of ridiculous proportions, I took the time to read through an FD.A vaccine insert and my brain started to flex its risk analysis muscle as percentages and number of cases per 1,000 jumped out at me on the reported studies included in said insert. I began to wonder:


What is the risk of adverse side effects of this particular vaccine compared to the risk of adverse effects related to another prescription drug? 

The comparison I'm using is between the M.M.R vaccine and amoxicillin, a widely used antibiotic.

M.M.R
From the package insertBecause clinical trials are conducted under widely varying conditions, adverse reaction rates observed in the clinical trials of a vaccine cannot be directly compared to rates in the clinical trials of another vaccine and may not reflect the rates observed in clinical practice. Vaccine-related adverse reactions reported during clinical trials were assessed by the study investigators to be possibly, probably, or definitely vaccine-related and are summarized below. 

I thought about going into the typical adverse reactions seen in children after the vaccine is administered (soreness at injection site, irritability, fever, rash, etc...), but am too lazy to recreate those tables for this post. Besides, those typical reactions for the ProQu.ad vaccine, which in clinical studies ranged from about 1.2% of kids had an adverse systemic reaction like diarrhea, upper respiratory infection, unspecified rash to about 2% who had a varicella or measles-like rash to 6% who had irritability and 21% who had a fever. Injection site adverse reactions danced from 2.3% for a rash to 22% for pain and tenderness at the site. These clinical studies included 4497 children who were vaccinated and 4424 who had safety follow-up. MM.R II and VAR.IVAX vaccines also had clinical trials, but with less subjects (2038, with 1997 follow-up). The results for these types of reactions are similar. The upshot is that these are the typical effects seen with these particular vaccines, and overall, it's not news.

Where people tend to focus on vaccine safety is in the fractions of percent of extreme reactions. To get to those reactions, we have to look at the post-marketing reports of adverse reactions, which are not the same as a controlled clinical trial where various variables are normalized to isolate the actual effect of the vaccine. These are the reported adverse reactions post-marketing: 

  • various infections including measles and varicella
  • blood and lymphatic disorders
  • immune system disorders (anaphylaxis, allergic reactions)
  • psychiatric disorders (agitation, apathy, nervousness)
  • nervous system disorders like afebrile convulsions or seizures, Bell's palsy, Guillain-Barre, encephalitis, to name a few
  • Eye disorders
  • Ear pain
  • Vascular disorders
  • Respiratory disorders like bronchial spasm, pulmonary congestion, rhinitis, sinusitis
  • Gastrointestinal disorders
  • Skin disorders
  • Musculoskeletal like arthritis, pain of hip leg or neck, swelling
  • Reproductive system and breast disorders - epididymitis (whatever that is)
  • General disorders and site conditions - the usual shots suck and hurt kind of reactions.
Note: this is not a detailed list, that would be way too much, but it hits all the major subcategories with a few of the reported reactions to make it clearer.

So, the scariest side effects listed in the insert:
  • Encephalitis and encephalopathy: these occur at approximately 1 per 3,000,000 doses. That's 3x10-7.
  • Death: This has been reported, but a causal relationship has not been established in healthy individuals. In other words, it's so rare, that they can't even come up with a 1 in x-million doses kind of number. They do have recommendations against getting vaccinated if the immune system is compromised in specific ways so as to avoid death.
  • Chronic joint symptons (rubella vaccine) specifically: In children 0-3% and short durations. The older you are, and if you are an unlucky woman, then it can be an issue for longer periods of time and (12 to 26% of adult women)
  • SSPE (subacute sclerosing panencephalitis) in children 1 case per 1,000,000. That's 1x10-7.
  • Aseptic meningitis has been reported to the Vaccine Adverse Event Reporting System. No causal relationship has been shown between the particular mumps vaccine used and this reaction. (there is a causal relationship between other strains of mumps vaccine and this reaction)
  • Thrombocytopenia (low blood platelet count) has been reported - tends to get worse in individuals who already have it, and if someone gets it because of the vaccine, then they may develop it more readily with follow-up doses. This is a case by case evaluation of an individuals reaction. No data is provided on how often it occurs, which leads me to believe that there isn't enough statistically relevant data to come up with a number.
  • Herpes zoster (shingles) related to VARI.VAX : does not appear to exceed that of previously determined occurrence in a study of healthy children who had been exposed to wild-type varicella.
An additional study was run to determine the incidence of febrile seizures when using the Pro.Quad vs. the MM.R II and VARI.VAX vaccines. That study, using 69,237 children and compared to 69,237 historical data which included the same age, gender and vaccination date matches showed that:
  • from 5-12 days after the vaccine the incidence of seizure was 0.7/1000 (7x10-4)
  • from 0-30 days after the vaccine, the incidence of seizer was 1.41/1000 (1.4x10-4)

Amoxicillin
I was hoping to find that the package insert for this drug had a similar litany of clinical studies and adverse reactions, alas, it is lacking in the juicy detail and statistical reports. However, it does include this:

SERIOUS AND OCCASIONALLY FATAL HYPERSENSITIVITY (ANAPHYLACTIC)
REACTIONS HAVE BEEN REPORTED IN PATIENTS ON PENICILLIN THERAPY.
ALTHOUGH ANAPHYLAXIS IS MORE FREQUENT FOLLOWING PARENTERAL THERAPY,
IT HAS OCCURRED IN PATIENTS ON ORAL PENICILLINS. THESE REACTIONS ARE MORE
LIKELY TO OCCUR IN INDIVIDUALS WITH A HISTORY OF PENICILLIN
HYPERSENSITIVITY AND/OR A HISTORY OF SENSITIVITY TO MULTIPLE ALLERGENS.
THERE HAVE BEEN REPORTS OF INDIVIDUALS WITH A HISTORY OF PENICILLIN
HYPERSENSITIVITY WHO HAVE EXPERIENCED SEVERE REACTIONS WHEN TREATED
WITH CEPHALOSPORINS. BEFORE INITIATING THERAPY WITH AMOXIL, CAREFUL
INQUIRY SHOULD BE MADE CONCERNING PREVIOUS HYPERSENSITIVITY REACTIONS
TO PENICILLINS, CEPHALOSPORINS, OR OTHER ALLERGENS. IF AN ALLERGIC
REACTION OCCURS, AMOXIL SHOULD BE DISCONTINUED AND APPROPRIATE
THERAPY INSTITUTED. SERIOUS ANAPHYLACTIC REACTIONS REQUIRE
IMMEDIATE EMERGENCY TREATMENT WITH EPINEPHRINE. OXYGEN,
INTRAVENOUS STEROIDS, AND AIRWAY MANAGEMENT, INCLUDING INTUBATION,
SHOULD ALSO BE ADMINISTERED AS INDICATED.

What Jumps out at Me
Now the really interesting thing to me in comparing these two package inserts (what I could find in online research) is that the adverse reactions cited in the studies and follow-ups on the vaccine are so small in number and for the most part are reactions that one can recover from. If your child is allergic to penicillin, it's less sure that they can recover from anaphylaxis. There are enough of these reactions that the FDA considers it prudent to all-cap an entire paragraph in the insert that this reaction can and does occur. 

The statistic I found for penicillin allergy is 1/10 of individuals are or remain allergic . That's a hell of a lot more than 1/3,000,000 cases of encephalitis due to a vaccine.

What About Everyday Risk Tolerance?
I work in the water resources business (specifically dam safety) and every project is tied to a risk analysis. The guidelines developed for risk analysis are based on typical societal tolerances for risk. In my industry it's typical to use a threshold of 1/10,000 as the threshold for individual risk (annual probability of a dam suffering a failure). That number isn't pulled out of the air exactly, but it can seem like it. It's based on historical data such as this chart from 2005. I realize this is old data, but had it readily available, plus it's not likely that it's changed a whole lot from then to now.

Basically, the probability of death from all causes per year for Americans ranges from 0.1 to 0.0001, or 1/10 to 1/10,000. This is the level of risk of death that we face every single day. 

Compare our average level of risk of death every day from any cause to the level of risk of contracting a severe adverse reaction from a vaccination. I'll compare risk of death picked off the chart in about the 1-2 year range. Turns out, it's pretty rough being a baby as the risk of death is roughly 1/200. The risk of contracting encephalitis (as an example) from the M.MR vaccine is 1/3,000,000.

I realize that not all reactions are recorded in the voluntary database, and that yes, there absolutely are outliers.


So?
I've either pissed you, the reader off completely with my quick and dirty analysis, or I've made a pretty good argument for calming down on the vaccination fight. Likely, I've only made my own head hurt in trying to decipher medical-ese in which I am wholly untrained.

The problem with risk tolerance is that one person's tolerance is different than another's. Because I work in the field every day, I've developed an internal monologue with myself when faced with irrational reactions to situations. I have managed to get almost-but-not-quite over my fear of being attacked by a shark during our infrequent beach vacations (1/60,000 lifetime risk) Yet, I'll get in a car and drive down the street without thinking and my risk of dying in a car accident over my lifetime is 1/84. This is a pretty cool article that shows the risks from various sources.

The point of this post is simply to point out that there is hard data available to analyze and mull over before making a vaccination decision. It seems to me that the anti-vaxxers and the pro-vaxxers spend a lot of time in shouting matches and name calling, which is never a great way to get a point across. Anti-vaxxers tend to go on and on about conspiracy between the F.DA and "big pharma" to shove horrible consequences under the rug, but rarely show evidence to back up their argument. Pro-vaxxers tend to treat anti-vaxxers like they are irresponsible and horrible parents. No wonder there can't be a rational debate. 

Here's my opinion: If a 1/3,000,000 or 1/1,000,000 chance of contracting a horrible side effect is not in your comfort zone of risk tolerance, then fine, don't vaccinate. I'll continue freaking out about sharks every time I get in the ocean.