Childhood Vaccines: Part 1

The road to better EBM

This topic has been impossible to avoid. I will eventually do at least a couple more posts on this, but I still plan on picking up the atherosclerosis series soon!

I assume many readers have by now heard something of the controversy surrounding the candidacy of RFK Jr, especially with respect to some of his comments about childhood vaccines. Personally, I’m not particularly interested in specific claims he has made during these long form discussions. I want to address whatever the steel man version of the argument against our current vaccine schedule is.

My first thought to this end was, “Why do I assume our childhood vaccine schedule is good?” Dozens of easy answers to this question flooded into my mind, but here are a few:

• Doctors overwhelmingly endorse these products as safe and effective.

• Various diseases have basically been eliminated, or at least markedly attenuated, since the advent of vaccines.

• Due to the frequency and timing of administration (many vaccines, throughout first few years of life), developmental disorders (eg, autism) will always be recognized within the window of administration of these vaccines.

• There is this concept of herd immunity being required for vaccines to work. (Probably the source of extreme hostility toward any skepticism of these products.)

• There have been many studies published that failed to show any connection between vaccines and autism, autoimmune disease, allergies, etc.

• Wasn’t this whole hesitancy started by Wakefield’s work, which has been retracted and is said to have been fraudulent?

My next thought was, “How could I be wrong?” This required more mental effort, but plenty of possibilities came to mind here as well. Here’s some of my stream of consciousness:

• Doctors often fail to critically appraise studies, instead we are often just taught what they show

• Clearly our public health institutions are capable of strongly endorsing policy that is questionable re: some covid policies

• I also think Abrysvo has unproven safety and efficacy, and that got through VRBPAC. So not just covid unfortunately.

• I’ve never actually read the studies myself, so I really don’t know much.

• So what is the evidence base behind our current schedule?

• There are plenty of diseases that have basically disappeared in the US that we never vaccinated for.

So where to start? It would be truly daunting to assess each vaccine in the childhood schedule one by one, and I basically have one week before I’m back on rotations and can only do this on weekends. Thankfully, there is a great resource to understand the strongest version of the vaccine hesitancy argument.

“Turtles All The Way Down: Vaccine Science and Myth” was published a few years ago, and as I understand it, this book represents something like a steel-man case against the status quo of our childhood vaccine schedule. Any doc who wants to understand the concerns in this space need only order a copy, which I recommend.

There are a core set of legitimate arguments that underly vaccine hesitancy. When we don’t deal with these, instead choosing to point and laugh at the particularly outlandish claims that may appear in their proximity, we are copping out.

Note: This is about to be a side rant. If you would prefer to skip it just scroll until you see more bold text. This is me basically describing how docs don’t have time to read trials and fact check information that gets dumped on them. We are dependent on guidelines being based on high quality evidence.

As many readers know, medical school in the United States takes four years to complete. The traditional model is that the first two years are “pre-clinical”, during which students must learn an enormous amount of information about anatomy, physiology, and pathology of every system in the human body. The latter half of medical school is the “clinical years”, during which you spend your days in the hospitals, clinics, etc.

I think sometimes people assume that there is enough time for nuanced discussion and critical appraisal of trials during medical training. In my experience there is often not enough time. Most docs simply work hard and follow the guidelines. That is how it has to be. Let’s take an example:

When a student is told that guideline-directed medical therapy (GDMT) improves outcomes in patients with heart failure with reduced ejection fraction (HFrEF), they write that down. Now they have that fact, better not lose it.

A student’s primary concern after learning this fact might be memorizing the four pillars of GDMT, which include a beta blocker, an SGLT2 inhibitor, an ACEi/ARB/ARNI, and an MRA. You don’t want to look bad if the attending physician asks you about these pillars on rounds. You also had better know the specific medications in each of those drug classes, just in case you get asked what you would give. You also can’t just pick any beta blocker, only some of them have trial data supporting their use for HFrEF. And if you know that carvedilol is one of those medications, you also should know that “Coreg” is the brand name for carvedilol.

The point is that there are endless bits of information like this to memorize. The end result is that anyone can tell you that GDMT improves outcomes in HFrEF, but very few docs critically appraise the clinical trial data underlying this kind of routine medical intervention. We expect certainty from our doctors, who in turn expect certainty from our guidelines. This is all fine as long as the methods used to generate those guidelines are rigorous and unbiased.

There is broad acceptance that childhood vaccines are safe and effective. Furthermore, there is a strong taboo on questioning these products, possibly more so than any other pharmaceutical. If you asked doctors across the nation about their opinion on childhood vaccines, I bet 99% would confidently assert their safety and dismiss any fears about these products as baseless.

My contention is that most of those doctors haven’t “done their work”, as I like to refer to it. They have never looked at any of the studies that support the approval of those vaccine products. They were merely told that vaccines are safe, so they wrote it down. That doesn’t mean they are wrong, it is just worth taking into account.

The primary exposure they have to vaccine hesitancy is the occasional lay person, typically a mother of an unvaccinated child, who can’t even speak the language of medicine. She goes in the mental bucket of a victim of fear mongering, hopelessly naive.

Over time a strong mental schema forms:

Doubt about childhood vaccines —> uneducated conspiracy theorist; danger to themselves and others.

Click, whirr.

This thinking is reinforced by its ubiquity and convenience.

Rant Over

I’ve almost finished reading the book, and it really is worth checking out. It is data driven and presents substantive arguments against the status quo in the United States. Anyone seeking to understand people skeptical of vaccinations would do well to read it. I would like to address the first chapter in this post:

This chapter looks at the vaccines given to children before the age of 2. There is one core claim in this chapter, and then something of a side argument about the rotavirus vaccine trials.

Core Claim: The safety of these vaccines is unknown. This is true due to a couple facts:

1. “Placebo” groups in these trials receive either some other vaccine or the same vaccine without the antigen. It is impossible to assess baseline rates of adverse events if none of your trials use an inert placebo, like saline.

2. New vaccines are compared to old ones, and those to older ones, etc. The oldest versions were often never studied in a rigorous way, as RCTs barely existed when they came into use.

3. The design of the Rotavirus vaccine trials was unethical. The use of “Vaccine-sans-antigen” as a placebo is not just misleading, but dangerous.

My Thoughts

The points in this chapter are well taken. When someone reads that a “placebo” was used in a trial, they assume that means something like saline was used. Many would be surprised to learn that pharmaceutical companies are allowed to refer to their product absent the antigen as a placebo. I think the FDA and medical journals should make this explicit, as the word placebo generates a “click, whirr” effect for most readers. There should be strict criteria on what is allowed to be referred to as a “placebo”.

There are basically two situations that are brought up in this chapter.

1. You are testing a newer generation of a vaccine that already exists.

2. You are testing a novel vaccine

For the former, there is a strong ethical case that can be used to defend the lack of placebo groups. If these vaccines are safe and effective, it can be viewed as unethical to deliver a true placebo to any children in the studies. The downside is that you sacrifice any ability to assess baseline rates of adverse events when you do this. The trouble of course is that you have to be quite certain of the safety profile of the original vaccines, because you are only able to assess relative safety under this process.

Many of the original vaccines were studied in a way that only allowed assessment of efficacy. This has often been done by testing a novel vaccine product against some other vaccine. It’s easy to see how this works with respect to efficacy, but does nothing to help assess safety.

The ethical argument that allows limited knowledge re:safety for newer generation vaccines does not hold for novel vaccine products. I find the use of the vaccine-sans-antigen in the rotavirus trials difficult to defend. This is an oral vaccine, as opposed to injection, and the trialists should have used a genuine placebo. If you are inclined to defend the use of vaccine-sans-antigen with respect to determining efficacy, you simply need a third arm in your trial. If not, don’t claim to have determined safety. It really is as simple as that.

Note: Some might take issue with this claim, mainly due to the fact that the components of this particular placebo seem like they shouldn't be harmful. For GSK’s Rotarix, the list included: DMEM (cell culture media, AKA forbidden gatorade for the nerds out there), sucrose, dextran, sorbitol, and amino acids. This was reconstituted in a buffer containing calcium carbonate and xanthane. I don’t know whether taking this as a placebo during the trial increased the risk of intussusception above baseline. Neither does GSK, because they didn’t study it. That’s the issue. We are never as clever as we think we are, and these companies should not be allowed to cut corners when developing a product for children.

Keeping things in context: Limited safety data alone is of course not sufficient reason to throw out the whole enterprise here. We can at the very least examine the adverse effects observed in the clinical trials for a given product and set that as an upper bound for the risk profile of a given vaccine.

In case this isn’t intuitive, here’s my thought process:

• If you want to know how much harm comes from a given product, ideally you would run a large RCT with a genuine placebo.

• In the absence of that evidence, we can still just look at the adverse events in the trials that have been run to get some idea of what the burden of disease is in the treated group.

• This way you don’t know the baseline rate of adverse events, but you can reasonably claim that the harm caused by your product can’t be higher than what is observed in the treated group.

• This will by definition be an overestimation of harm, but it allows comparison with the benefit you expect to derive from your intervention.

• All this still assumes large trials with thousands of kids. The sample size needed is negatively correlated with the effect size.

Final thoughts for now: Let me know in the comments if it would be worth breaking down more of the chapters in the book. Overall I think some of my takeaways so far is this:

• Vaccines are medical products, subject to individual scrutiny just like any other drug. Claims like “vaccines are safe and effective” are just as meaningless as “drugs are safe and effective”.

• Many vaccines we give have limited safety data, lots of efficacy data. This does not mean we know that they are particularly harmful. I think the safety profile of many of these products, as well as the products in combination, have not been adequately studied. This can easily be remedied!

• In the wake of Covid we have something like a crisis with respect to faith in our public health institutions. The mistrust felt by many has been well-earned. There should be a massive push by the CDC/FDA/NIH to conduct studies that answer many of the questions raised in this book. Directly addressing these issues is the only way to restore public trust in our institutions.

• Treating childhood vaccines as a class leads to absurdities. The rationale behind vaccinating every child for Hep B (questionable) is dramatically different than that behind Measles (measles infections have effectively been eliminated by the vaccine.)

• We should avoid the cartoonish version of the vaccine efficacy argument that pretends infectious disease mortality wasn’t dramatically reduced by improved sanitation, hygiene, and antibiotics, often well before the vaccines were introduced.

• A sober analysis of benefits and risks should continually be encouraged for any medical product, whether it has sacred cow status or not