Are the conclusions reached in these VAERS blog reports valid?

Let’s start off by reviewing the two different ways that vaccine reactions can be monitored: active and passive.

An active monitoring system tracks a patient’s entire history, starting before the vaccine is administered, and continuing even after a vaccine reaction. Follow-up interviews may be made after vaccination to increase thoroughness. Such systems are typically managed by a health-care provider and usually include the complete medical record of the patient.

A passive monitoring system (such as VAERS) is one in which reports of vaccine reactions are volunatarily submitted by patients, doctors, and pharmaceutical companies. No reports are filed for healthy people, very little past history is recorded, and few post-reaction follow-ups are available

Passive monitoring systems are less useful than active systems for a number of reasons:

• Underreporting. Because the reports are submitted voluntarily, many patients and doctors do not report vaccine reactions. Different estimates exist for the amount of underreporting and range from a factor of 10 to as much as a factor of 100 (meaning that the true number of vaccine reactions is between 10 and 100 times higher than what is reported to VAERS).
• Subject to bias. There are many issues that can bias a passive monitoring system. There might be more reports from new vaccines because people are worried about them. News articles might excite the public to submit more VAERS reports. It is possible that the kind of people who report reactions may be different than those who don’t recognize a reaction or ignore it, causing more bias. And of course, serious reactions might be reported more frequently (even though they account for only 14% of VAERS data).
• Inaccuracy. Although most reports are submitted by medical professionals (83%, according to the government) there are still VAERS reports that are filed by people who are unfamiliar with vaccine reaction reporting, and therefore have incorrect or insufficient information. Some have suggested that the system is “gamed” by people intentionally filing false reports, but this is unlikely because the CDC and FDA staff screens the data and does follow-ups. For example, all VAERS reports are examined by trained medical coders in order to ensure correct symptom terminology. The May 2011 blog entry has more to say about errors in VAERS data.
• Missing context. Although the VAERS reporting form asks for previous medications and other history, it is not as complete as a full medical record. Also, because the report ends at the vaccine reaction, it misses any future medical issues that may arise. This is a problem because it is important to correlate “similar” patients when evaluating vaccine safety. For example, if one patient gets a vaccine and has a reaction, while another patient does not get a vaccine or reaction, you do not have enough information to say that the vaccine caused the reaction. What you need to do, in order to say anything with confidence, is to “stratify” the data, organizing patients into groups that have similar histories so that you can eliminate every difference except for the vaccination. Then, those who did not get the vaccine form a useful “control group” and can be compared with those who did get the vaccine. This is possible in an active monitoring system, but nearly impossible in a passive system.
• Missing rates. It is important to know the rate of vaccine reactions, so that you can say what percentage of vaccinated people react. Although the MedAlerts VAERS search engine does have biometric data showing the number of vaccine doses that were distributed, this is subject to many inaccuracies and is not as good as the data from an active monitoring system (see the July 2011 blog entry).

So what good is a passive monitoring system such as VAERS? Most people agree that VAERS is useful as an early-warning indicator of possible vaccine problems. Trends that show-up in VAERS may indicate a problem, which can be further investigated using other data.

Do these blog entries make appropriate use of the VAERS data? Some of the entries do exactly what VAERS is good at: they look through the data and identify possible trends. But other blog entries try to verify a trend by looking at VAERS data, and when this happens, the blog is careful in its wording, saying things like “VAERS data seems to confirm this trend” or “you might want to talk to your doctor about this.” Some of the early blog entries (for example December 2009), do discuss issues that can affect the blog’s conclusion.

The bottom line is this: We use VAERS because it’s all we have. The only active vaccine monitoring system that exists in the U.S. is the Vaccine Safety Datalink, but it is government-private data that is available only to “qualified investigators”. Even CDC medical researchers (who do have access to the Vaccine Safety Datalink) often use VAERS data instead. If these blog entries are faulty, then so are those government VAERS analyses that “prove” vaccine safety.

So I will continue to make VAERS blog entries each month and look at all aspects of the data. I hope that I do not mislead, but my goal is to raise awareness of potential vaccine safety issues. I am not a medical researcher (as I state quite clearly). In the end, some will disregard these blog entries because of the problems with VAERS, but hopefully others will find the issues raised here to be useful.

Let’s search the VAERS database to see if the data agrees with this report.

Rather than asking for the results of any particular vaccine, let’s look at those VAERS events that list seizures as a symptom, and then examine the age groups and vaccines that were given. This avoids pre-judging any particular vaccine, and may turn up fresh results.

To identify VAERS events that have seizure symptoms, we use the MedDRA symptom classification (a 5-level organization of all known symptoms). The top-level of the MedDRA does not mention anything about seizures, but the next level down (High-Level Group Terms, or HLGT) has a symptom category called “Seizures (incl subtypes)”.

So search the data, check Expert Mode in the upper-right (to access different MedDRA levels), and make a Graph of Vaccine Type and Age. Also, open up Section 2 (Symptoms), check the MedDRA level HLGT and select “Seizures (incl subtypes)”. This will organize all of the seizure-related VAERS events according to the diseases that the vaccine prevents (the Vaccine Type) and the patient’s age.

The result is shown here:

The result seems to confirm the JAMA report. The five biggest spikes are for Diphtheria, Haemophilus, Pertussis, Polio, and Tetanus in children under 3 years of age. No other age group gets seizures with as high a frequency.

Of course, this raises another question: is the DTP-Polio-Hib combination-shot the problem, or are the individual vaccines also culpable? For example, is it safer to get a DTP and then separate shots for Polio and Hib? The spikes for the five separate serums suggest that each one is reactive, and when combined could create a vaccine that is more reactive than most other routinely recommended childhood vaccines.

The only way to know whether a given combination vaccine is more reactive than its individual components is to divide the number of VAERS events associated with a vaccine by the number of doses actually given. This gives us the rate of VAERS events. Unfortunately, dosage data for Pentacel isn’t available. However, dosage data is available for another combination vaccine: DTAPHEPBIP (known as Pediarix, a 5-in-1 shot with DTP, Polio, and Hepatitis B). Although this is not directly comparable to the JAMA article’s study data, it does allow us to examine the reactivity difference between combination vaccines and their component parts.

So let’s make a dose-adjusted graph, and look at the Vaccines, not the Vaccine Types. Once again, we look at seizure-related VAERS events, and this time we’ll limit it to children under age 1. This gives us the rate at which a particular vaccine is associated with childhood seizures. Keep in mind that dosage numbers are always an approximation (see my July 2011 blog). Also, Pediarix dosage data began in 2004, so let’s only look at data since then. Here is the relevant part of the graph (showing all of the DTP vaccines):

The results confirm that the Pediarix combination shot runs a noticeably-higher rate of VAERS events per dose than any of the other DTP vaccines (DTAP has a rate of 1 VAERS report per 155,000 doses given, but Pediarix has a rate of 1 per 93,000).

This data suggests that, when combined, vaccines are more reactive than when they are administered separately. My April 2010 blog showed that multiple vaccinations, given on the same day, cause more problems. It appears that multiple serums given in the same shot are also troublesome.

The graph shows a surprising result for 2011: it is the first year that had a noticeable decrease in the number of VAERS events. The slight decrease between 2009 and 2010 is nothing compared with the nearly 20% drop between 2010 and 2011 (from 36,806 events in 2010 to 29,908 events in 2011).

What changed? Let’s look at individual vaccines. Make a Graph of Year of Appearance and Vaccine Type (the Vaccine Type is the disease that the vaccine prevents).

Although there were fewer reports for nearly all of the vaccine types, one of them (Flu) went down significantly. In fact, the change in the number of Flu-related events (down from 16,742 in 2010 to 9,758 in 2011) more than accounts for the overall drop in VAERS events.

Of course, when looking at yearly Flu seasons it is important to note that they occur in the winter months, which span two different calendar years. So it is more useful to graph Month of Appearance instead of Year of Appearance. Doing that for the past few years confirms the decrease in Flu-related VAERS events during this current Flu season:

The trend is unmistakeable: Flu-related VAERS events are way down. In my September Blog Entry, I predicted a bad year for Flu-related VAERS events, but apparently I was wrong.

Of course, the number of Flu-related VAERS events rose significantly in the previous two years, no-doubt fueled by the H1N1 scare, so this may simply be the “normalizing” of an unusual spike. But Flu shots are still pushed hard during the winter months and many people still get the vaccination.

Is the decrease in the number of Flu-related VAERS reports caused by a safer Flu vaccine? Has the unusually late Flu season resulted in people avoiding the shot until 2012, leaving 2011 numbers low? Or are people intentionally avoiding getting the Flu vaccine?

This month I want to continue the investigation by asking what happens to the women in these pregnancy-and-HPV-related VAERS reports. Are the symptoms related to their pregnancy or are they simply general issues? And if they are pregnancy-related, are they serious? To find out, I will look at the symptoms in these reports and see what they indicate.

The best way to look at symptoms in the VAERS data is to consider the fact that they are organized using the 5-level MedDRA classification system. So for starts, let’s look at the top-level of the MedDRA, called the System Organ Class, or SOC. This breaks all symptoms into 26 different categories.

The two graphs below show the number of symptoms in the SOC categories. On the left is the entire VAERS database and on the right is just those VAERS events that mention an HPV vaccine and are pregnancy-related. Note that these graphs are made using Expert Mode (to access MedDRA) and then graphing Symptoms SOC (Highest).

The left graph shows the sort of things that would be expected in a typical VAERS report: a big spike in the broad SOC “General disorders” and other spikes involving injuries, investigations, and skin issues. It is perhaps interesting that “Nervous system disorders” also figure prominently.

The right graph focuses on the symptoms of pregnant women who got an HPV vaccine. It is not surprising that the biggest spike here is for Pregnancy conditions.

But although this graph indicates that the injuries to pregnant women are indeed pregnancy-related, it still doesn’t tell us what happens in any sort of detail. To do that, we need a new feature of MedAlerts that lets you see the MedDRA symptom hierarchy. So use Expert Mode and check Show Symptom Tree. You also have to narrow the search to pregnant women who got the HPV vaccine (as described in last month’s report). Here is a partial view of the symptom tree, showing the “Pregnancy” SOC as well as the next two levels of the hierarchy. There are two deeper levels in the tree that are not shown here.

Note that the numbers next to each symptom are not the number of VAERS reports that mention that symptom. Instead, they are the number of times that the particular point in the symptom tree appears in the pregnancy/HPV reports. Since each VAERS report has many symptoms, these points in the tree are visited many times. So the numbers are useful only for comparing relative frequency of a given point in the tree. See the MedAlerts Help page on MedDRA for more information about examining the MedDRA.

As you can see, the most common issue is “Foetal complications” followed by “Abortions and stillbirth” and “Maternal complications”. The graph continues deeper, showing more symptom detail. So it appears that, according to VAERS reports involving pregnant women who received the HPV vaccine, they indeed had substantial pregnancy-related difficulties and the pattern of symptoms appears to be different than what is generally reported to VAERS.

There is no reliable way to say whether a VAERS report involves a pregnant woman because it is not one of the questions asked by the government. However, a simple approach is to search the Write-up for the word “pregnant” (the Write-up is a multi-paragraph description of the report). Here is a Graph of Vaccines with the word “pregnant” in the Write-up field of Section 2 (Symptoms):

The worst offender, by a large margin, is Gardasil (HPV4). It’s important to note that many childhood vaccines are given before it is possible to be pregnant, so only those vaccines given to older people will show up here. Indeed, there are spikes around the Flu and Hepatitis vaccines which are given to older people. But even if we look at just Gardasil and Flu, the Gardasil spike is significant because many more people get Flu vaccines than Gardasil. And the comparison of Gardasil to Flu is even worse when you consider that this graph shows data from the entire 20+ years of VAERS reporting, but Gardasil has only existed for five years. This means that Gardasil, which is associated with 5 times as many pregnancy-related VAERS reports as the Seasonal Flu, in 1/4 the number of years, appears to be 20 times as dangerous as Seasonal Flu.

Also of interest is the fact that the number of VAERS reports is the same for both Seasonal Flu (FLU / FLUN) and H1N1 Flu (those with “H1N1″ in their name). But again, the Seasonal Flu data accounts for 20 years of reporting whereas the H1N1 Flu was identified separately in the VAERS data for just 1 year. Could the H1N1 Flu vaccine have done 20 times as much damage to pregnant women as the Seasonal Flu vaccine? My November, 2010 blog has something to say about that.

The results seem pretty damning for Gardasil and the H1N1 Flu vaccine. But what about the initial question concerning the validity of the search…does typing “pregnant” into the Write-up field really find pregnant women? To find out, there is a new analysis feature in MedAlerts which shows the context of a Write-up search, so you can see the surrounding words. To use this feature, check Expert Mode, make a Table of Write-Up Context (3 words), and search for the Write-up term “pregnant”. Also, select the Vaccine “HPV4″. This will analyze only those Gardasil-related reports that have “pregnant” in their Write-up.

The resulting table shows all of the context of the word “pregnant”. The top 20 results (in order of frequency of occurrence) are:
     that she was pregnant at the time (appears 8 times)
     the patient was pregnant at the time (appears 7 times)
     The patient was pregnant at the time (appears 7 times)
     the patient was pregnant when she received (appears 5 times)
     the patient became pregnant. The patient’s last (appears 5 times)
     that she was pregnant. No adverse effects (appears 5 times)
     who was not pregnant and who was (appears 4 times)
     the patient was pregnant. The patient’s last (appears 4 times)
     the patient became pregnant. The patient’s last (appears 4 times)
     out she was pregnant. Her LMP was (appears 4 times)
     was 5 weeks pregnant. The patient sought (appears 3 times)
     the patient was pregnant. The patient was (appears 3 times)
     the patient was pregnant. Her LMP was (appears 3 times)
     the patient was pregnant. At the time (appears 3 times)
     the patient was pregnant. The patient’s last (appears 3 times)
     the patient was pregnant when she was (appears 3 times)
     the patient became pregnant after receiving the (appears 3 times)
     the patient became pregnant after receiving GARDASIL. (appears 3 times)
     product, concerning a pregnant female patient who (appears 3 times)
     The patient was pregnant when she was (appears 3 times)

Only one of these shows that the patient was not pregnant, and that happens only 4 times in the data. Two other contexts refer to someone who became pregnant after the vaccination, but this may still be relevant. A complete examination of the results shows that 66 out of the 1290 matches suggest a patient who was not pregnant. This is just 5% of the results and is not significant-enough to invalidate the graph.

So if you are pregnant (or even think you might be pregnant) talk to your doctor and other health professionals before getting a vaccine, especially Gardasil or the current Flu vaccine which has the H1N1 serum in it.

As you know, “Flu season” starts in the fall and runs through the winter. Since most people get their Flu vaccinations early in the season, Flu-related VAERS reports start to appear in September and peak in October. Of course, it takes many months for all of the VAERS reports to accumulate (as I describe below), so a look at September data immediately after it has been released gives just a taste of what will happen.

Two years ago, when the H1N1 Flu “pandemic” was in the news, my blog entry showed that September 2009 had over 1000 VAERS reports related to Seasonal Flu vaccinations. By the end of the Flu season, that number had more than doubled, and the September/October total was 5423 (see “A” below).

Last year, with much less of a Flu scare, the numbers were even worse. My blog entry showed that September had 1262 Seasonal-Flu-related VAERS reports, and by the end of the season the September/October total was 6181 (see “B” below).

So what does it look like this year?

Already, September shows 1348 VAERS reports related to Seasonal Flu vaccinations (see “C”). This is more than has appeared in any previous year at this point in the data collection. There are even 365 reports in October, and these are people who were vaccinated, had reactions, reported them, and got them published, all in just 11 days!

For an idea of how the September vaccination numbers grow in the following months, let’s look at last September’s Flu-related VAERS numbers and see how they changed over the past year.

The graph below shows dates of CDC data releases. For example, the first entry is dated 9/22/10 which means that it accounts for VAERS data up to September 22, 2010 and therefore does not describe the entire month. At that point, there were over 500 Flu-related VAERS reports. By October 5, when September data was “complete,” the number was much higher, and by November 3 most of the reports were in for September. The numbers then grew slowly after that.

It is probably the case that the number of people filing Flu-related VAERS reports is high because the number of people are getting Flu vaccinations is high. Like last year, the current Flu serum contains three different strains: Influenza B (Seasonal Flu), H3N2 (Hong Kong), and the 2009 H1N1. It packs a punch, and people are filing more Flu-related VAERS reports than ever before.

To find out, we will use MedAlerts to look at serious VAERS events in these age groups. Serious VAERS events (those that are life-threatening, involve hospitalization, permanent disability, or death) are important measures of a vaccine’s safety.

To do the search, make a graph of Vaccine Type. Then, in Section 4 of the search form, check Serious in the upper-left and in Section 5 of the search form, click on Age (Range) and specify a range of ages in the From and To fields:

• To see children ages 0-6, set From to 0 and To to 7. Note that the “To” field is the upper bound, so making it 7 requests everything below 7, or 0-6.
• To see children ages 7-18, set From to 7 and To to 19.

The two graphs are shown below (with only the recommended vaccines shown).

As you can see, children age 0-6 years are reporting the most VAERS events following a DTP vaccination.

Children ages 7-18 years are significantly affected by the Human Papillomavirus (HPV) vaccination. Note that HPV is already associated with over 100 deaths, and is not limited to girls (there are three reports of boys who have died following an HPV vaccination). This is not the most widely given vaccination for 7-18 year olds, but it appears frequently in VAERS and is associated with serious adverse effects.

For example, we know that there have been 1822 Flu-related VAERS reports from the state of Illinois, and 1829 Flu-related VAERS reports from the state of Massachusetts. This information might make us think that the chance of having an adverse reaction from a Flu vaccine is the same in each state. But Illinois has about 12 million people and Massachusetts has about 6 million people, so in truth, Illinois has an adverse Flu vaccine reaction rate that is half that of Massachusetts. MedAlerts knows how to use state population information to give us a better understanding of the VAERS data (for more, see the May 2010 blog entry).

Another piece of related information that has been missing from MedAlerts (but is now available) is the number of doses of each vaccine that have been given. Without this information, it might appear that a widely-given vaccine, such as MMR, is more dangerous than a less-given vaccine (like Yellow-Fever) because there are many more VAERS reports that mention MMR. But if we consider the actual number of vaccinations that are given, it might turn out that the rate of injuries is much different.

How does MedAlerts know the number of vaccines that have been given? The CDC has actually been gathering this information for many years and making the data public (they call it “biologics surveillance”). Each year, pharmaceutical manufacturers report the number of doses distributed as well as the number of doses that are returned (hospitals and doctors can return unused doses to the manufacturer). The difference between the number of doses distributed and the number of doses returned is the net number of doses, and is considered to be the number of doses given. Of course, this figure is certainly higher than the actual number of doses given, because there must be doses that get destroyed, lost, or otherwise not actually given. Nevertheless, these figures are the best information available for determining the number of vaccines given. To see the latest dosage data that is used by MedAlerts, including references to the original CDC documents, click here.

And how can MedAlerts use this data? As usual, it is necessary to check Expert Mode in the upper-right of the search form, because seaches that use this information are specialized. Then, it is possible to apply this data by individual year, or by total reports across all years. To make a graph or table by individual year, select Year of Vaccination (dose-adjusted). To see a graph or table that considers all years, select Vaccines (dose-adjusted).

Let’s make a graph of each vaccine and look at the rate of VAERS reports associated with each (across all years). Make a graph of Vaccines (dose-adjusted). It looks like this:

First the good news: the numbers are quite small. The average vaccine is below the first vertical line (0.0010) which means that for every 1000 vaccines given, only one VAERS report is filed. But of course, we have already established that MedAlerts is using a vaccine dose count that is too high, and it is well known that the number of VAERS reports is low (by anywhere from a factor of 10 to 20 to a factor of 25 to a factor of 10 to 100). Both of these errors actually mean that the rate of vaccine injury is much higher than shown here.

The second thing that is clear from this graph is that some vaccines have an unusually high rate of VAERS reports. The worst offender is the Anthrax vaccine, which seems to be nearly 10 times as dangerous as any other vaccine. Could it be that this vaccine is unsafe, but because it is primarily given to military personnel, its danger is overlooked when compared with the other hazards of military life?

Two other vaccines show a high rate of VAERS reports: the Lyme and the Rotavirus vaccines. Both of these were given for only a few years and then withdrawn (Lyme ran from 1999 to 2001 and Rotavirus was available only in 1998). The rotavirus vaccine was withdrawn because of side-effects (bowel obstruction) and this graph supports the suggestion that it had significant side-effects. Removal of the Lyme vaccine is attributed to “poor sales” (see page 7 of this CDC document) but considering the large number of injury reports that appeared in a very short time, it would seem that there might have been another reason for withdrawing the vaccine.

There is much to be learned by combining multiple sources of information, and MedAlerts constantly strives to do this in order to further study issues of vaccine safety. The website’s new dosage data is just starting to be explored, and hopefully more useful analysis will come from it as time goes on.

Problem 1: Hospitalization. There are two outcomes in each VAERS report related to hospitalization: Hospitalized and Extended Stay in Hospital. Could it be that a VAERS report indicates an extended hospital stay but does not indicate hospitalization? Clearly, that would be a mistake. To find out, go to Section 4 of the search form and set Hospitalized? to “No” and Extended stay? to “Yes”. The result shows 627 VAERS reports with this mistake!

The ramifications of this mistake are actually significant, because the government often makes a distinction between those VAERS reports that are “serious” and those that are not. Non-serious VAERS reports are typically discarded as unimportant. The definition of “serious” includes hospitalization, life-threatening, disabled, or death, but it does not consider the extended-stay outcome. In fact, of those 627 reports with the hospitalization mistake, 505 of them are non-serious and would, if the mistake were fixed, become serious situations.

Problem 2: Gardasil underreporting. Each VAERS report lists the vaccinations that were given and also has a write-up that describes the problem. Is it possible that Gardasil is mentioned in the write-up but is not in the list of vaccinations that were given? To find out, check Expert Mode (needed to search for vaccines that were not given). Then in Section 2, set the Write-up field to “gardasil”. Finally, in Section 3, choose the “HPV” and “HPV4″ vaccines and check Excluded. HPV4 is Gardasil and HPV is a generic vaccine that could be Gardasil or Cervarix. By excluding both of these possibilities, we are finding VAERS reports that specifically do not list Gardasil.

How many reports match this search? 116! But let me immediately point out that this search does not necessarily indicate a mistake in the VAERS data. A careful reading of the write-ups in these reports reveals that many of them mention Gardasil without claiming that it was given to the patient. Some of the write-ups say that “Gardasil was not given” or that the patient intended to get Gardasil but was given a different vaccine instead. Some describe newborn infants who were exposed to Gardasil during their mother’s pregnancies. Of the 116 reports uncovered by this search, 43 of them follow one of these patterns and indicate that Gardasil was not given.

And what about the 73 reports that claim Gardasil was given? Why wasn’t it listed in the report? Some are clearly human error (either the VAERS submission form was filled-out incorrectly or a typo was made). Some appear to combine multiple medical situations into a single report and lose information in the process. Some list an “unknown” vaccine having been given, but it is clear from the writeup what that vaccine actually was.

Remember that I am not a health professional, so it is possible that some of these 73 VAERS reports are simply not Gardasil-related. In any case, here are the 73 reports that I think are mistakes. The numbers listed here are the VAERS ID numbers. Click on them to see the full report.

• Patients and relatives. Inexperienced people can easily make mistakes when filling out the VAERS submission forms.
• Doctors. Busy doctors filling-out forms at the end of a long day can make mistakes.
• Drug companies. Some reports are submitted by the drug companies, and they are dealing with third-hand data.
• VAERS. The people at the VAERS agency might make mistakes, too. To me, this is the least likely explanation as I discussed a few months ago. In fact, I’m hoping that this blog reaches the appropriate people, and that the VAERS data is fixed. If this happens, then people reading this blog years from now will not be able to reproduce the mistakes I’ve described.

Now if vaccination-related deaths happen uniformly, to people of all ages, then the two curves should look the same. If, however, vaccination-related deaths happen more predominantly to certain age groups, then this will stand-out. Here are the graphs:

The graph of all VAERS reports (on the left) is not surprising. It show that most VAERS reports involve young patients, which makes sense because most vaccinations are given to children. There is also a peak in the 17-44 category. But this is not a real peak; it is an anomoly caused by the unequal groupings. The 17-44 year range is the widest by a large margin, so it includes a large number of VAERS reports.

The graph on the right (VAERS events where the patient died) is surprising. It shows a much stronger peak among young children (below age 3). This peak is so strong that it even towers over the wide grouping of 17-44 year-olds.

There are really two problems with these graphs: (1) the age groupings are not uniform, making some groupings stand out unfairly, and (2) it would be helpful to analyze the 0-3 year old range in more detail to see exactly when these children are dying.

To solve this problem, MedAlerts lets you set different age groupings. It is another of the site’s “special features,” so to use it you must check the “Expert Mode” box in the upper-right. Then, instead of making a graph of Age, make a graph of Age, Custom. Custom Age graphs let you specify where you want the age breaks to occur. There are even a few buttons on the right side of the age break section that automatically fill it in with some “standard” schemes.

So let’s redo the above two graphs, but this time use the age breaks that CDC Wonder suggests (click “Set Wonder” in the Custom Age section). The results are here:

Now the graph of all VAERS events (on the left) does not show such a strong peak in the 17-44 range, but instead has most of its emphasis on people 18 years and younger. This is to be expected.

We also see in the graph of patients who died (on the right) that the peak for young people continues to be pronounced, and that this peak happens in the first 6 months of life.

Can we explore this at even more detail? Just set the age breaks to be even smaller. The graph below sets age breaks at 0.1 years, 0.2 years, 0.3 years, etc. This is the level of resolution that appears in VAERS reports (they seem to report age in tenths of a year, rather than twelvths of a year as you might expect). The graph below shows the age of patients who died in the first year of their life:

And now we see where it is happening: at about 3 months of age. No other age seems to die after a vaccination as much as these 3-month olds. It makes you wonder whether this is a bad time to get vaccinated.