I’ve been fielding questions about the COVID-19 vaccine from family, friends, and some of you, so I thought I’d share some of what I know.
Disclaimer: I am only providing you with information. Do not construe anything in this post as medical advice. If you have medical questions specific to your situation, please contact your physician. If you have general questions about pharmacology or toxicology (or biostatistics, ethics, anime, manga, pizza), feel free to contact me.
Where Can You Find The Safety and Efficacy Data for the Vaccines?
Pfizer/BioNTech
On August 23, the US FDA approved the Pfizer/BioNTech vaccine, which is now known by its brand name, Comirnaty.
If you want to see the safety and efficacy information, please read the vaccine package insert.
Safety information starts in Table 1 (page 8) and goes through Table 4 (page 12).
Please keep in mind when you’re reading the safety information that some reactions are perfectly normal and expected for all vaccines, and some are expected in particular for this vaccine. For instance, getting a fever is normal and expected for many vaccines because that signals that your immune system is responding to the stuff in the vaccine (in this case, the stuff is the lipid particles that contain the mRNA that encodes the protein that will become the antigen — and if you want to know more about that, read the section on How Does This Vaccine Work, Exactly?).
A lot of what we call systemic effects are due to your immune system working overtime and addressing what it sees as an infection. Yes, you read that right. Read the section on How Does This Vaccine Work, Exactly? to find out more. If you don’t want to read that section here’s what you need to know — your body is acting like it’s infected to a low degree so that the immune system learns what the SARS-CoV2 spike protein looks like. In that process, the immune system is going to try to destroy wherever that foreign protein is. As part of that, you will likely get a fever. That fever will likely lead to dehydration. Dehydration leads to muscle aches and pains; likely due to a lack of hydration and electrolyte imbalances. Bottom-line, not feeling well is your body’s way of saying it’s fighting off something, and it is — it’s fighting off the SARS-CoV2 spike proteins being produced by your cells to train the immune system.
Efficacy information for Comirnaty can be found starting in Table 5 and ending at Table 6. The efficacy of this vaccine is amazing.
Moderna
Are you more interested in the Moderna vaccine? Then you want to read the FDA Decision Memorandum supporting the Emergency Use Authorization. These are pretty dense documents, and I tend to think of them as great non-prescription sleep aids, even for us pharmacologists and toxicologists.
NOTE: this vaccine is available under the Emergency Use Authorization.
If you want to see just the efficacy information and you don’t care anything at all about the demographics of the study population, then head straight to Table 9 (page 25). If you want vaccine effectiveness by demographic group, check out Tables 10–13 (pages 25–28). If you want to know more about who was in the study, and how well your particular set of demographics is represented, then Tables 4 and 5(pages 20–21) may be interesting to you, but really you want Tables 6–8 (pages 22–24).
Safety information that most people are interested in starts in tabular form with Table 23 (page 39). Table 24 is something else you should take a look at (page 40).
Table 25 begins to discuss unsolicited adverse events. Let me explain the term “unsolicited adverse events”. It means effects you don’t like that Moderna did not specifically ask the participants to report (i.e., Moderna gives the study participants a checklist and says, “Check off all the things you got or are experiencing”; if it’s not on that list, then it’s unsolicited). These may or may not be causally associated with the vaccine.
Okay, what does that mean? Not causally associated is the fancy way of saying, “yeah, these effects happened, but they were not caused in any way by the vaccine” — so, some of these may be due to the vaccine, and some of these may be due to something that has nothing to do with the vaccine. For instance, if you decided to eat undercooked meat and got diarrhea from it, you might not know that the meat caused the diarrhea, but you do know that you got that vaccine — so you report it, like you should. But that diarrhea had nothing at all to do with the vaccine, so it’s not causally associated. And since Moderna did not include diarrhea in its questionnaire, it’s an unsolicited adverse event.
Johnson & Johnson/Janssen
What is commonly referred to as the Johnson & Johnson, or J&J, vaccine is actually produced by Janssen Biotech, which is a J&J company. I might use all those terms interchangeably, so remember for our purposes Janssen = J&J = Johnson & Johnson.
If you’re interested in the J&J vaccine, then you’ll want to read FDA’s Decision Memorandum. Like the others, it’s dense, and yes, likely a great candidate as a non-prescription sleep aid.
NOTE: this vaccine is available under the Emergency Use Authorization.
Love demographics? Well you’ll enjoy Tables 6–9 (pages 21–25).
Efficacy data starts on page 25 and Table 10. In Tables 11–12 (pages 26–27) you can see the demographic breakdown of those who contracted COVID-19 following vaccination. In Table 13 (pages 29–30) you can see the breakdown of vaccine effectiveness for individuals with different underlying conditions. It is important to remember that this table can be kinda tricky to interpret as you need to also need to know what the prevalence of these various diseases are within the population to fully understand how underlying conditions may translate into risks of developing COVID-19 after getting the vaccine (in other words, this may be challenging for most people, including physicians, pharmacologists, and toxicologists, to fully understand and appreciate).
Table 14 (page 30) gives more information about vaccine effectiveness. The difference between Tables 10 and 14 is that Table 10 only includes COVID-19 cases where the primary lab used in the study (the central lab) confirmed the test. Table 14 includes COVID-19 confirmations from labs that were not the central lab. As you can see, the vaccine effectiveness rates aren’t a whole lot different between Tables 10 and 14.
Safety information begins on page 39. Table 23 (pages 39–40) give you the demographics and a high level summary of the safety information. Don’t be too concerned with the solicited local adverse reactions — this is what we would anticipate seeing with an actual vaccine, versus a placebo. The immune response to the vaccine at the injection site is going to drive a lot of this. Table 24 (page 42) gives more information on adverse events localized to the point of injection, while Table 25 (page 42) let’s ya know how long before someone noticed these events.
The systemic adverse events begins on page 43 in Table 26. Again, these are all symptoms we would expect of a fully functioning vaccine. Table 27 (page 44) gives the time to onset information (i.e., how many days before someone noticed these symptoms) and how long they lasted.
Table 28 (page 45) gives us the unsolicited adverse reactions. If you’re coming directly to this section and didn’t read the Moderna section, here’s how I like to describe unsolicited adverse reactions:
It means effects you don’t like that J&J did not specifically ask the participants to report (i.e., J&J gives the study participants a checklist and says, “Check off all the things you got or are experiencing”; if it’s not on that list, then it’s unsolicited). These may or may not be causally associated with the vaccine.
Okay, what does that mean? Not causally associated is the fancy way of saying, “yeah, these effects happened, but they were not caused in any way by the vaccine” — so, some of these may be due to the vaccine, and some of these may be due to something that has nothing to do with the vaccine. For instance, if you decided to eat undercooked meat and got diarrhea from it, you might not know that the meat caused the diarrhea, but you do know that you got that vaccine — so you report it, like you should. But that diarrhea had nothing at all to do with the vaccine, so it’s not causally associated. And since J&J did not include diarrhea in its questionnaire, t’s an unsolicited adverse event.
Table 29 (page 47) talks about a relatively rare adverse event called embolic and thrombotic events. These are embolisms and thrombosis events. Both are circulatory issues and both can be fatal or extremely devastating. An embolism is when something, could be a clot, could be a foreign object, is stuck in a blood vessel, leading to an obstruction of blood flow. A thrombus is a blood clot. So a thrombosis is when a blood clot forms in a blood vessel. A thromboembolism is when a blood clot forms in a vessel and obstructs the flow. Not fun. Anyway, the FDA did detect embolic and thrombotic events; however, they rightly concluded that there is not sufficient data to state these were caused by the vaccine due to the presence of existing conditions.
Table 30 (page 49) talks about tinnitus — that’s ringing in your ears. The FDA did detect a few cases of tinnitus, but they rightly concluded that these are confounded by existing conditions, so at this time the data are not sufficient to state they are caused by the vaccine.
Table 31 (page 51) lists specific serious adverse events identified by J&J that the FDA decided were mostly not related to the vaccine. There are 3 serious adverse events, though, that the FDA do believe are likely related to the vaccine. The first is a male that experienced diffuse urticaria — this is the fancy word for hives all over the body. The male did not self-report any previous allergies to vaccines or vaccine ingredients. The male did experience some swelling of the lips, but did not show any signs of breathing difficulties, thus it was not ruled as anaphylaxis. The FDA rightly concluded this was likely a hypersensitivity reaction to the vaccine. This does happen, and it is kinda scary, but there is not much you can do about it. One day you take a drug, eat a food, and all of a sudden you noticed that you are allergic. You weren’t before, but you are now. It is unfortunate, but it does happen and is not unexpected.
There is another case where a male was diagnosed with brachial neuritis. “Brachial” is the fancy word for arm, and “neuritis” is the fancy word for nerve inflammation (pro-tip: “-itis” is the suffix that means inflammation). Extensive evaluations were performed and there was no evidence of actual neuritis. There was also no evidence of a spinal injury. FDA rightly concluded the diagnosis of brachial neuritis is probably inappropriate given the evaluation data.
Another case is a male who presented with malaise, weakness, nerve pain, shortness of breath, headache, numbness and tingling in the extremities, chest pain and fever beginning on Day 2 following vaccination. The subject tested negative for influenza, COVID-19, and RSV. More than likely, this is just a normal systemic response to the vaccine (the fancy term is “systemic reactogenicity”).
There were a few less notable issues that the FDA also ruled out, and rightly so.
Death Is An Unsolicited Adverse Event — Should I Be Worried?
No, I wouldn’t worry about dying, at least not due to the vaccine. The percentage of people dying in these studies is exceedingly low. Keep in mind, human clinical trials are performed on humans, and humans die all the time (I don’t mean to be glib about this, it’s just a fact). If you are looking at a population of over 15,000 people with a wide age range, you have to expect some of them will die just due to chance over a period of several weeks. In Table 25 (page 41) you can see that 3 people died in the placebo group, while only 2 died in the vaccinated group. That does not mean that the vaccine prevents death; rather, those probabilities are just about equal, so there’s really no treatment effect here. It’s just all a roll of the dice.
How Does This Vaccine Work, Exactly?
The vaccine consists of mRNA that is surrounded by lipids (“lipid” is a fancy word for “fat”). So imagine you have a ball of fat — inside that ball of fat, there is an mRNA. mRNA is a molecule that cells use as the blueprint or instructions to make proteins.
So why is the ball of fat necessary? Because if we were to inject you with just mRNA, proteins in your blood will break it down. Those proteins are called RNAses — and they work to protect us from foreign RNAs.
The ball of fat is technically made up of fatty acids. Those fatty acids put the fats on the outside, and the acid head of the fat is on the inside of the ball, where the mRNA is. It just so happens that the fatty acids are similar to what is on the outside of your cells — the cell membrane. So the fatty acid ball looks kinda like a small cell. What’s really cool is that the fatty acid ball can merge with the cell membrane and deposit the mRNA into the inside of the cell.
Once that mRNA is inside the cell, the cell doesn’t realize the mRNA doesn’t belong. So cells will start to use the mRNA and begin to translate it into proteins.
It just so happens that the protein encoded by this mRNA encodes the spike protein for SARS-CoV2. So the cell will translate the mRNA into the spike protein, and then the spike protein will be deposited into the cell membrane, exposing the spike protein to the outside of the cell where the immune system can see it.
The immune system is constantly scanning the outside of cells, looking for things that don’t belong, like the spike protein. This is a gross simplification, but the immune system will begin to build antibodies that recognize a particular part of the spike protein — the part it recognizes is called the antigen. The way this works is essentially that a B-memory cell (T-cells do something similar) when it is activated will begin to divide (create new cells). These new cells are not identical copies — the DNA will be very different between them. B-cells have a much higher mutation rate in the region of DNA that encodes antibodies than most other cells in humans — that mutation rate allows the DNA that encodes antibodies to mutate, which results in antibody proteins that are all slightly different. The B-cells will continue to divide and undergo mutation in the antibodies until it creates a set of B-cells that have high specificity for the antigen. These B-cells will then further divide, become specialized cells (we call this differentiation) and release their antibodies. The antibodies will paint/tag the antigen, which targets the antigen (and whatever it is attached to) for destruction. It’s all very metal (in the rock music sense).
Once the infection dies down, which means the antigens aren’t seen anymore, the differentiated B-cells die off, and all that remains are the memory B-cells (something similar happens with T-cells). These memory B-cells and T-cells serve as the body’s database and reserve of antibodies. If the antigen is seen again, then the body’s database and reserve are activated, and the process repeats itself.
How Were They Able To Make The Vaccine So Quickly?
Although it appeared that they were able to make the vaccine really quickly, this was the result of several years of research and development. The liposomes, or lipid balls, have been researched as a drug delivery mechanism for a very long time. The idea of using mRNA as a vaccine has been around for a long time, too. Encapsulating mRNA in liposomes for delivery has also been researched and discussed for a very long time. And NIH and other biomedical funding agencies have been a lot more interested in coronaviruses, because two of the major viral intercontinental outbreaks, ignoring influenza, have been due to coronaviruses. So it just stood to reason that the major next pandemic was going probably going to be a coronavirus.
So we’ve had all the pieces available to make an mRNA vaccine for a very long time. But why now? Why hasn’t this been done before?
Well, timing is everything, and all of the tools were very mature now. Usually, vaccine development is a very slow process. And vaccine production is a very slow process. This means that research and development of a new vaccine is very expensive, both in terms of money and time.
Given the fact that we needed a vaccine quickly, the easiest and quickest way of developing one is to use mRNA encapsulated in a liposome. The problem with mRNA is that it is not stable for long periods of time at room temperature. Also, RNAses are all around us. I used to work in a lab and RNA was what I did. The joke in the lab was that if you even looked at your tube of RNA the wrong way it’d spontaneously degrade. RNA is not easy to work with.
So long-story short: we had everything we needed to make an RNA vaccine, we had the money, and we had a pressing and immediate need where the conventional means to manufacture a vaccine simply was not acceptable.
Is There Evidence A Booster Is Necessary?
Let’s take a close look at this. There are some that argue that there is evidence, and there are others in the scientific community who are skeptical (and remember: a good scientist is always skeptical). There are several sources of evidence for us to consider, so let’s go.
CDC HEROES-RECOVER Cohort
The CDC published in the MMWR on August 24, 2021 that in its HEROES-RECOVER cohort they saw a decline in vaccine effectiveness in frontline workers. Raw, unweighted estimates bear this out, too. The way to calculate vaccine effectiveness is slightly complicated, but easily doable in a spreadsheet or with some paper.
If you go down to the table at the bottom of the report, you can see the vaccine effectiveness pre-delta and post-delta variant. Their calculation is 91% effectiveness in their cohort pre-delta, and 66% post-delta. Doing this in a spreadsheet with just raw numbers (we’re not modeling this, and there is a lot to consider, so the raw estimates aren’t great, but they’re a decent ball-park estimate), I got 92% pre-delta and 74% post-delta.
What does this mean? This means that in this particular cohort, they are seeing a decrease in vaccine effectiveness. There could be any number of reasons for this, and we have to be careful in using this data by itself to speak for the entire US population.
For starters, this is a very small sample. The total number of participants is smaller than the town I grew up in. There could be sampling bias, there could be other constraints on this particular cohort.
In addition, these are front-line workers. Front-line workers have the highest exposure to the virus of anyone. Some of the people here are being exposed multiple times a day to SARS-CoV2 aerosols. Why is this relevant? Because infections with viruses and bacteria, just like toxicity due to chemicals, is driven by exposure. If you have a higher exposure, or more exposures, or more high exposures to an infectious agent you are more likely to become infected, symptomatic, sick and more likely to pass this along to someone else.
In other words — unless you are a front-line worker, these data are not particularly informative.
We know that the delta variant is more infectious and communicable than previous strains. Given that increased exposure, I would not at all be surprised that the vaccine effectiveness decreases — that just makes good sense.
So, bottom-line, does this report suggest the general US population needs boosters? No, it does not. However, it does suggest that due to the increased infectivity, people with higher exposures need to be more cautious.
Does this provide evidence that boosters will be efficacious? No, it does not.
Israeli Correlation Study of SARS-CoV2 Breakthrough Infections
This study has not yet been peer-reviewed (as of August 25, 2021), but it’s making the rounds.
What does the study say: In a nut-shell, it says that people who got the the second dose of vaccine in either January and February 2021 was more likely to have a breakthrough infection than someone who got their second dose in March or April 2021.
Unfortunately, they express this in terms of an odds ratio, which is not approachable by most people without advanced training and the ability to think in terms of both odds and ratios. So, first I’m going to show you their results, and then I’m going to explain odds ratios, and then I’ll try to put the results into some context.

Above you’ll see Table 1 from the Israeli Correlation Study. Under Model 1, they have broken down the data by age range. What they did was to identify everyone who had received their second dose within the months of January-April. The ones who received their second dose by the end of February are the early vaccinated. Else, they are the late vaccinated. The odds ratio ranges from 1.49 to 1.54 across the age ranges from 16 to above 60 years old. But what do these odds ratios mean?
First, we’ll talk about odds. The simplest way to think about odds is to think of it as the probability that something happens divided by the probability that it doesn’t happen. So let’s say the probability that you win a card game is 10%. That means the probability you don’t win is 1 – 0.10 = 0.90, or 90% probability that you don’t win. The odds that you win are 0.10/0.90 = 0.11, or 1:9 odds. Likewise, the odds that you lose are 9.0 or 9:1.
Here’s another example. The probability that you will roll a 6 with one 6-sided die is 1/6. The odds are 1:5 or 0.20. The odds you don’t roll a 6 are 5:1. Thus, you are 5x as likely to not roll a 6 as you are to roll a 6.
An odds ratio is where you are comparing the odds of one event against the odds of a different event. So, what is the odds ratio of rolling a 6 on 1 six-sided die to rolling a 6 on 1 twenty-sided die? We know the odds of rolling a 6 on 1 six-sided die is 1:5 or 0.20. The odds of rolling a 6 on a 20-sided die is 1:19 or 0.053. So the odds ratio of rolling a 6 on a six-sided die compared to the odds of rolling a 6 on a 20-sided die is: 0.20/0.053 = 3.78. This means you are 3.78x more likely to roll a 6 on a six-sided die compared to a 20-sided die.
The key here is that this doesn’t actually change your odds of rolling a 6 on a six-sided die. It just means that if you have to land exactly a 6, and you have your choice of using a six-sided die or a D20 (a 20-sided die) you should probably use the six-sided die.
The problem with odds ratios is that they don’t tell us anything about the actual probability. So, if you’re an Israeli in this particular study cohort, you are 1.53x more likely to develop COVID-19 if you received your second dose in January or February compared to if you received it in April. But that doesn’t mean that your actual risk of developing COVID-19 is 1.53x higher than it was before.
The risk of developing COVID-19 based on this study is closer to 0.3% for those who got their second dose in January or February and 0.2% for those who got their second dose of the vaccine later.

According to the Israeli Ministry of Health COVID Dashboard, as of August 25, 2021, there is an uptick in severely ill individuals who are vaccinated right now.

And these appear to mostly be those in the higher age ranges — their senior citizens. What this translates into is what we would expect — those who have the weakest immune systems are the ones who are the least likely to have good memory B and T cells, and thus, they are the ones who are the least likely to maintain immunity the longest. Same goes for those who are immunocompromised, because let’s face it, the elderly are also immunocompromised when compared to someone in their 20s.
What else drives someone to be immunocompromised? Stress. Lack of sleep. Being constantly barraged by virus particles. Who does that sound like? Front-line workers (see the section about the CDC HEROES-RECOVER cohort above).
So there appears to be an argument that a segment of the population may need a booster — the immunocompromised.
And oh, yeah, who was it that got the vaccines early on in the process? The immunocompromised and front-line workers. So, we have to take that into consideration when we see these data that say that there are these upticks in individuals who are vaccinated now developing breakthrough cases. We have to think critically about what else might be driving these trends. We need to look for additional causal factors and confounding factors.
So, is there evidence here that a booster is going to work? Nope. We need additional study for that — studies where people are given boosters and tracked. The Israeli government’s explanation for their booster campaign is pretty weak: “The benefit of a booster is known from other vaccines.” We can’t always assume data are transferable from one vaccine to another; the answer would be more honest if it simply said, “We don’t know if it will be efficacious, but we sure hope it will!” Because right now, hope is all we have.
Bottom-line on Boosters:
We have no evidence that boosters are safe and effective. We don’t have any evidence that they won’t be safe and effective. We simply don’t have any data right now. I hope that will change. There are reports that Pfizer has submitted data about boosters — so stay tuned!
Do we have evidence that the general population is at risk due to a lack of efficacy of the vaccines? Nope.
Do we have evidence that the general population needs boosters? Not at this time.
Do we have evidence that immunocompromised individuals may need boosters to avoid developing COVID-19? Yes, there is some evidence to support this.
Do we have evidence that front-line workers may need boosters to avoid developing COVID-19? There is limited evidence that some front-line workers who may have really high exposure rates may be experiencing breakthrough infections. It is unclear if there are other mechanisms that may be able to be put in place, certain risk management practices, that may help.
Conclusion
I will do my best to keep up and update this from time-to-time. If you’re interested in getting updates, please sign up for the newsletter (scroll down)!