When it’s time to decide about vaccines for their children, parents are often wrought with angst. They want to protect their children and do “the right thing,” but they worry that vaccination could inadvertently harm the child. These fears are often made worse if a parent seeks out information on the internet or social media. Indeed, even in casual conversations with other parents, vaccine safety concerns may be magnified.

Often the most concerning ideas are those that are vague; for example, the notion that vaccines are replacing infectious diseases with chronic diseases. This idea is validated in parents’ minds by the fact that they have never seen or experienced the infectious disease that the vaccine is aimed at, but they have seen or experienced the chronic condition being put forth as an example.

Asthma is one such example. Asthma is a chronic condition in which airways in the lungs suffer prolonged, low-level inflammation, causing the individual to have periodic episodes of breathing difficulties and wheezing when the inflammation is aggravated.

• Find out more about changes in the lungs due to asthma in this video from the American Lung Association.

While the causes of asthma continue to be studied, several factors are known to contribute to the likelihood of developing this condition. These include family history, other allergies, infections during childhood that are caused by certain respiratory viruses, exposure to particular environmental conditions (e.g., dust, mites, smoke, chemicals, molds and smog), and obesity.

Recently, a paper was published that suggested a potential association between the amount of aluminum children were exposed to in vaccines during the first two years of life and their likelihood of being diagnosed with asthma between 2 and 5 years of age. Like concerns about vaccines causing chronic conditions, concerns about aluminum in vaccines have been around for some time, which is why these researchers completed the study. Because of this interest, we wanted to work through these findings together with our readers.

Study

Daley MF, Reifler LM, Glanz JM, et al. Association between aluminum exposure from vaccines before age 24 months and persistent asthma at age 24-59 months. Acad Pediatr. 2023 Jan-Feb;23(1):37-46.

Brief summary

This study analyzed data from the Vaccine Safety Datalink (VSD), a U.S.-based research network designed to monitor vaccinations for the purposes of vaccine safety. Three aspects of this study are worth looking at more closely to understand what it does and does not tell us about aluminum and asthma: study population, aluminum processing, and disease caused by aluminum.

Study population

The study population included 326,991 children born between January 2008 and December 2014 who received care from one of seven VSD sites. Of these, 7,546 were identified as having asthma based on the study definition. For each child, immunization records were used to determine their aluminum exposure from vaccines during the first two years of life with the goal of comparing that exposure with the likelihood of developing asthma between 3 and 5 years of age. Two population-based choices made by the researchers are important given the goals of this study:

• First, children were not included if they were diagnosed with asthma before 24 months of age. The authors indicated that they wanted to ensure that the exposure to aluminum in vaccines was distinct in time from the development of asthma.
• Second, the data were subdivided based on whether a child had been diagnosed with eczema prior to 12 months of age. This choice was made because children with eczema are at increased risk for developing asthma.

As a result of these two choices, more than 6% of children (n = 25,188) were excluded from the study sample (because they were diagnosed with asthma before 24 months of age) and more than 4% (n = 14,337) were separated and compared with the remaining cohort (because they were diagnosed with eczema before 12 months of age). Given the number of children excluded because they developed asthma prior to “completing” their exposure to aluminum, we need to be cautious about our interpretation of these data, particularly as related to causality. It is notable that more children were excluded because they had the condition being studied than children in the subgroup considered to be at increased risk for developing the condition.

Aluminum processing

The goal of this study was to evaluate exposure to aluminum and development of asthma. However, this aspect of the study also has important considerations to keep in mind when considering the findings:

• First, the authors chose to measure total aluminum exposure from vaccines in the first two years of life and compare that with development of asthma in the next three years of life. This is like saying that the ice cream you consumed in elementary school is associated with your weight gain in middle school. It’s problematic because so many other things, including ice cream you ate while in middle school, would also affect your weight. And, like ice cream, aluminum does not simply accumulate in the body. Once aluminum reaches the bloodstream, most of it (90%) is bound to a protein, called transferrin, and removed through our kidneys. About half is eliminated within 24 hours and about three-fourths, within two weeks. As such, aluminum exposure in the first two years of life would be difficult to associate with the development of a disease in the third to fifth year of life — a time when most of the aluminum from the vaccines is long gone.
• Related to this, and equally important, vaccines are not the only source of aluminum exposure in the first five years of life. Indeed, aluminum is introduced daily through the environment, food and water. And, while most aluminum that is consumed does not enter our bloodstream, some does, so only looking at one source of aluminum does not present a complete picture of aluminum exposure.
• Further, studies of aluminum processing have demonstrated that the amount of aluminum from food and water that enters our bloodstream varies based not only on the quantities in each food, but also on the chemical environment at the time. For example, if you eat an orange at lunch with your turkey and cheese sandwich, you are likely to absorb more aluminum from the cheese and bread in your sandwich than if you eat a banana with your sandwich. Does this mean you should not have that orange? Of course, not. We consume milligrams of aluminum every day and only a small portion of that aluminum makes it out of our digestive tract and into our body. But this example demonstrates the larger point that understanding what happens following aluminum exposure is more complex than simply comparing a quantity of aluminum and a particular disease outcome.

By opting to evaluate exposure to aluminum based solely on vaccine content in the first two years of life, the study established an arbitrary assumption. And, while one could argue that this was a low threshold to which all children were exposed, we have no information about the other quantities to which children were exposed. The authors tried to evaluate breastfeeding during the first two years of life, but the data were limited, and there was no information related to which formulas the non-breastfed infants were fed (aluminum concentrations vary by product and preparation), daily volumes they consumed, etc. The authors also failed to control for exposure to environmental pollutants or a family history of asthma or allergies. This critical lack of controls severely limited the authors’ ability to isolate the effect of the one variable they were interested in studying — aluminum intake. Additionally, all participants continued to consume food and water during years three to five, when they were diagnosed, and here too, there were no data to evaluate their exposures. In short, while it is difficult to assess the effects of exposure in a situation like this, applying arbitrary assumptions that exclude ongoing exposures and biological processing variability does not really help us understand anything about the association between these two variables.

Disease caused by aluminum

Aluminum-containing ingredients are included in some vaccines because they serve as adjuvants, which are ingredients that increase the immune response to a vaccine. The authors theorized that the aluminum content in vaccines could change the type of immune response generated by the vaccine (T helper 1 (Th1) vs T helper 2 (Th2)). Because the latter (Th2) induces chemicals that cause low levels of inflammation, they hypothesized that a disease based on long-term inflammation, like asthma, was theoretically possible. While understanding the mechanism by which an adjuvant increases the immune response is an important undertaking, the type of study the authors undertook will not answer that question. Further, if we evaluate the larger context, some observations are important:

• The aluminum in vaccines is processed and removed, so even if the type of immune response shifts when aluminum is present, it would not induce long-term generation of the chemicals that cause inflammation — especially one to three years later.
• When people become ill from exposure to high levels of aluminum, they most often experience conditions related to bone density because when remaining aluminum accumulates in our bodies, most of it accumulates in our bones.
• And, while the lungs have been found to contain aluminum, it is typically in a different form because it is inhaled when we breathe, rather than deposited from our bloodstream.

For these reasons, an analysis to understand the hypothesis that an adjuvant changes the way the vaccine is processed would be better answered using alternative research approaches.

What this study did and did not show

The primary conclusion these authors presented was that “a positive association was found between vaccine-related aluminum exposure and persistent asthma.”

Unfortunately, this primary conclusion fails to provide additional context:

1. Of the more than 300,000 children included in this study, just over 7,500 were diagnosed with asthma (2.3%). On the other hand, about 70% of the children were exposed to the highest quantities of aluminum in vaccines (4 mg or more).
2. The average age when the children in both the eczema and non-eczema groups were diagnosed was between 44 and 45 months of age — almost two years after the aluminum exposure period being measured, which if you recall from the earlier section, aluminum breaks down over time. Likewise, if aluminum exposure from vaccines was associated with the development of asthma and a subgroup of the children were predisposed to developing asthma, one might expect that the children with eczema would develop the disease sooner.
3. In both groups, other measures were also positively associated with the diagnosis of asthma, including male sex, non-Hispanic Black race/ethnicity, having food allergies, and two measures of healthcare utilization (outpatient and emergency department visits). In some categories, early-life severe bronchiolitis and being born prematurely were also associated with development of asthma. Indeed, many of these measures were more positively associated with development of asthma than cumulative aluminum exposure.

Likewise, as described earlier, choices made in designing this study can’t be dismissed because the real world does not operate in closed systems. Deciding to only look at aluminum delivered by vaccines while ignoring the everyday exposures from other sources is not representative of what happens. Further, using cumulative measures of aluminum introduced by vaccines as if they are static over a four- to five-year period ignores everything we know about aluminum chemistry and processing.

Takeaway

This study did not prove a causal relationship between aluminum exposure in vaccines and the development of asthma. Like the Indian folktale about the blind men and the elephant, this study provides one bit of information that needs to be considered in the context of a much larger and more complex system. While more studies will be done to further explore the hypothesis that this study has generated, parents should continue to follow the recommended immunization schedule to protect their children from very real, potentially life-threatening, diseases.

• Watch this video by Dr. Offit in which he answers, “Is Aluminum in Vaccines Associated with Asthma?”

Claims have circulated suggesting that COVID-19 mRNA vaccines are causing huge increases in cases of colon cancer that develop quickly after vaccination. The headline-grabbing claims cite “turbo cancers” and 500-fold increases in cases. These claims point to a paper by Akkus and colleagues that was published in June 2024. Because of important weaknesses in the study and the lack of basis for these claims, we wanted to work through the findings with our readers.

Study

Akkus E, Karaoglan B, Akyol C, Ünal AE, Kuzu MA, Savaş B, Utkan G. Types and Rates of COVID-19 Vaccination in Patients With Newly Diagnosed Microsatellite Stable and Instable Non-Metastatic Colon Cancer. Cureus. 2024 Jun 6;16(6):e61780. 

Brief summary

This study investigated whether receipt of the BNT162b2 (Pfizer BioNTech) mRNA COVID-19 vaccine was associated with diagnosis of a particular type of colon cancer, known as deficient mismatch repair colon cancer (dMMR). 

Study population

This study included 76 people newly diagnosed with colon cancer who had been vaccinated against COVID-19 at least three months before their diagnosis. Participants were divided into two groups (study and control) based on their type of colon cancer. Those with dMMR cancer were the study group. During the two years of the study, 38 people qualified for the study group. The control group consisted of 38 of 242 people diagnosed with a different type of colon cancer, known as microsatellite stable cancer (MSS) during the same period. The 38 individuals in the control group were randomly selected from the larger group of 242.

Notably, the two types of colon cancer are different, so the choice of study populations was flawed. A better way to answer this study question would have been to compare people newly diagnosed with dMMR who did and did not receive COVID-19 mRNA vaccine, but even then, the premise of this study makes it difficult to learn much. Keep reading to learn more. 

What this study did and did not show

As described in the paper, colon cancer takes about 10-15 years to develop, so this study does not tell us anything about an association between receipt of the COVID-19 mRNA vaccine and developing cancer because everyone in the study already had colon cancer when they got vaccinated as the study authors noted: “the time interval (three months) is short and probably the patients already had the malignant lesion at the time of the vaccine.” 

Second, as described in the “study population” section, the two types of cancer are different, so some differences in the findings are more likely the result of the type of cancer and not receipt of the vaccine. For example, the data showed that people with dMMR colon cancer who got the COVID-19 mRNA vaccine had higher levels of a protein known as C-reactive protein (CRP) than people with MSS colon cancer who got a non-mRNA formulation of COVID-19 vaccine. However, several points are important here:

1. CRP is a protein made in the liver. When someone has inflammation, the liver releases CRP. Some studies, not related to vaccination, have shown that people with dMMR cancer have higher levels of CRP than people with MSS cancer, so this finding does not tell us anything about the impact of the vaccine type.
2. As described above, to evaluate this difference, the authors compared people with dMMR cancer who got the mRNA vaccine to people with MMS cancer who got a non-mRNA vaccine. However, most participants in both groups received mRNA COVID-19 vaccines, so to make this comparison, they compared small subgroups of the already small study population (15 vs 14 people, respectively).
3. Importantly, there were other statistically significant differences between the groups, such as lower BMI in the dMMR cancer group. This finding is not unexpected because numerous studies (including ones conducted before COVID-19) have shown that people with dMMR colon cancer typically have a lower BMI than people with MSS colon cancer. 

Overall, this study is not sufficient to make any claims about an association between dMMR colon cancer and receipt of COVID-19 mRNA vaccines due to the study design and sample size.

Takeaway

The main conclusion of the paper was that a higher percentage of participants with dMMR colon cancer received COVID-19 mRNA vaccines compared with participants with MSS colon cancer. However, the study was not sufficient to draw conclusions about a relationship between receipt of COVID-19 mRNA vaccines and subsequent diagnosis of dMMR colon cancer. 

Some parents wonder whether having a gene known as MTHFR is a reason for their child to forgo some vaccinations. The answer is no. Children with the MTHFR gene can get all routinely recommended vaccines. 

This concern about vaccination stems from a study related to the smallpox vaccine that has been misinterpreted and misrepresented. In fact, the study authors published a follow-up letter to the editor to indicate that their study was being improperly interpreted.

Study: Reif DM, McKinney BA, Motsinger AA, Chanock SJ, Edwards KM, Rock MT, Moore JH, Crowe JE. Genetic basis for adverse events after smallpox vaccination. J Infect Dis. 2008 Jul 1;198(1):16-22.

Brief summary: This study investigated whether certain genetic differences in the MTHFR and IRF1 genes were associated with adverse events after receiving the Aventis Pasteur smallpox vaccine. 

Study population: Participants across two studies who had no prior immunity to smallpox received the Aventis Pasteur smallpox vaccine and were monitored for adverse events. Samples were collected from participants who also consented to analyze their genetic profiles with the goal of understanding whether people who experienced adverse events following vaccination had genetic differences compared with those who did not experience adverse events. The study population included 131 people — 40 experienced adverse events and 91 did not. 

What this study did and did not show

• This study focused on adverse events following receipt of a smallpox vaccine that had a high rate of side effects. Because the adverse events were specific to that vaccine, this study cannot be generalized to other vaccines.
  • This study investigated the Aventis Pasteur smallpox vaccine, which is not used in the U.S at this time. Smallpox was eradicated, meaning it does not infect anyone anywhere in the world. As such, we do not vaccinate against smallpox. Supplies of this vaccine are maintained in the government stockpile in the event of a bioterrorist event.
  • This vaccine is administered differently than most vaccines. Getting this vaccine involves putting a drop of vaccine on the skin and inserting it into the skin with a two-pronged needle (also called a bifurcated needle).
  • There was no placebo group in these studies, so we also do not know if some adverse events, such as skin reactions, were due to the vaccine administration method rather than the vaccine components.
• This study was relatively small (total of 131 participants), and there have been no additional studies on this topic. One isolated study does not provide sufficient evidence to make broad claims about vaccine safety.
• Likewise, this study did not provide sufficient evidence to recommend MTHFR mutation screening prior to vaccination.
• The study is outdated. In an interview with The Atlantic, the lead author on the paper said, “It’s just not even a valid study by today’s methodology.”

For these reasons, this study is not sufficient to make any claims about an association between MTHFR gene mutations and vaccine adverse events.

Thimerosal, an ethylmercury-containing preservative, was removed from routinely used childhood vaccines in the U.S. by 2001. However, multidose vials of influenza vaccine for older children and adults still contain small quantities of thimerosal, and in many other countries, particularly resource-limited countries, multidose vials of vaccines remain common. In multidose vials, preservatives like thimerosal ensure that the vaccine does not become contaminated when needles are repeatedly stuck into the vial to draw up doses of vaccine. Thimerosal in vaccines, even at the quantities used before 2001 in routinely recommended childhood vaccines in the U.S., has been studied extensively and shown to be safe. Unfortunately, some people continue to suggest that this ingredient is unsafe — despite scientific evidence contrary to their personal beliefs. The authors of the paper being reviewed herein offer such an example. You can read more about their long and personal history of questioning vaccine safety in this article from New York Magazine (published 7/17/25). 

The article being reviewed below is known as a “review article.” Review articles aim to understand the findings of a body of literature about a single topic in order to collect and synthesize what is known about a topic in one place. In this article, the authors were examining the current research about whether thimerosal can get into the brain through a layer of cells known as the blood-brain barrier (BBB). Because this is a review article, the authors did not do any new experiments. They analyzed information available from other research papers.

Review articles should accurately represent what we know about a topic. Two aspects of this are important to understand. First, the way that papers are identified is important in these types of articles because if the way articles are identified is biased or incomplete, the findings will also be biased or incomplete. Second, how the authors interpret the findings is important for evaluating the strength of the study. For example, if half of the research on a topic says one thing and the other half finds the opposite, then it is important to include both perspectives in the review article and discuss possible reasons for the discrepancies. However, if almost all the research says one thing, it is important for the review article to focus on the findings that are the most consistent and not over-represent outlier findings.

Study: Kern JK, Geier DA, Homme KG, Geier MR. Examining the evidence that ethylmercury crosses the blood-brain barrier. Environ Toxicol Pharmacol. 2020 Feb;74:103312.

Methods: How were papers identified?

Some of the issues with this article come from the papers that were included. First, while the abstract suggests a time period and number of articles included, there is no description of methodology to support that information in the body of the paper. A well-written abstract should support the content of the paper. Second, the authors cite their own work multiple times. Citing one’s own work is not necessarily a problem, nor is citing several papers from any author or group of authors. However, it is something to consider when evaluating the findings of a review article. When multiple studies are from the same author or authors, it might indicate that other people are not finding the same results or that the findings by that author or authors are being over-represented in the analysis. Third, many of the articles in this review are quite old. Using old articles is not always bad, especially if the authors are talking about the history of an issue, but review articles should also include the most recent research because that gives a better representation of the current state of knowledge in a field.  

Findings/interpretations: What did the authors conclude?

The stated goal of this paper was “In this current review, the evidence supporting the notion that ethylmercury-containing compounds cross the BBB will be presented and discussed.” However, the authors fail to mention that crossing the blood-brain barrier (BBB) does not necessarily make something harmful. The BBB mainly filters based on molecule size and structure. Many medications and substances can, and do, cross the BBB, including caffeine, acetaminophen (Tylenol), and antidepressants. So, simply proving something crosses the BBB does not automatically mean that it is causing harm. 

In addition to the issues with the methodology and the premise of the study, two other examples demonstrate the weakness of this paper for drawing accurate conclusions from it:

1. Inaccurate or misrepresentation of individual studies: The authors inaccurately stated and misrepresented the findings of several papers. For example, they claimed that a CDC-sponsored study by Thompson and colleagues found an association between thimerosal exposure and tic disorders in children. This study actually found no causal association between thimerosal exposure and brain function because they found an almost equal number of positive and negative effects. In fact, higher thimerosal exposure in the first month of life was associated with a lower likelihood of tics in girls.
2. Accurate representation of the body of evidence: When there is conflicting evidence in a field, it is important to represent that fairly in a review article. However, the authors overstated how much disagreement exists related to this topic. Dozens of studies by different research groups have found no impact of vaccines or thimerosal on brain development. The authors claimed that thimerosal is harmful, citing their own research, which is at odds with the overarching body of literature on this topic. In fact, a separate review article about thimerosal and brain development by Azevedo and colleagues noted that almost all the research supporting a causal link between thimerosal and autism was completed by the authors of this review article. The Azevedo study was published three years after the Geier review article. It included almost three times as many papers as the Geier review, making it clear that this review was limited, and possibly biased, in the approach used to evaluate this body of literature.