Is there a Link Between Vaccinations and Peanut Allergies?

I. Introduction

Is there any link between peanut allergies and childhood vaccinations? Peanut and tree nut allergies have become a common and potentially serious health problem in the United States. According to a recent case report and discussion from Harvard Medical School, "Peanut and tree nut allergies affect ~1.1% of the general population or about 3 million Americans. Peanuts … are the major cause of fatal and near-fatal anaphylactic reactions to food" (Lee and Sheffer 260). There has also been a rise in the incidences of peanut allergies in many Westernized countries. According to one 2003 report, the prevalence of reported peanut allergies in children under 18 years old increased from 0.4% in 1997 to 0.8% in 2002 (Sicherer, Muñoz-Furlong, and Sampson 1206).

Over the past decade, there has been a parallel increase in both the number of required immunizations and the rate of immunization compliance. According to the Centers for Disease Control and Prevention (CDC) records, the recommended childhood vaccination schedule for January-June 1996 included 15 immunizations for children under the age of 2 years (CDC 1996 943). The recommended vaccination schedule for 2006 represented an almost 50% increase in the number of immunizations, with 22 immunizations for children under the age of 2 years, not including annual influenza vaccines (CDC 2006 Q3). Compliance rates have also increased. In 1996, for example, the CDC reported a nationwide compliance rate of 82% among children ages 19-35 months for the 3-dose series of HepB vaccinations (CDC 1998 110). In 2005, the nationwide compliance rate for three or more doses of HepB was 96% (CDC 2006 1125).

Is the correlation between these two rising figures just a coincidence, or is there a causal relation between the increase in the number of vaccinations required from birth through two years of age and the incidence of peanut sensitivity and allergies? One of the currently favored hypotheses proposed to explain the increase in peanut allergies - often called the "hygiene hypothesis" - suggests that exposure to naturally occurring microbes might decrease the chances of developing an allergy or other autoimmune disease. But if this is the case, it also seems reasonable to think that challenging the immune system through exposure to vaccinations in the first two years of life (the period in which our immune systems develop) might increase the chances of developing an allergy or another problem with the immune system. In particular, it seems reasonable to think that exposure to vaccinations between birth and two years might increase the chances of developing peanut allergies.

This research project attempts to answer the question of (1) whether there is a difference between the incidence of peanut allergies in a group of children with a relatively low vaccination rate (80%) versus a group of children with a high vaccination rate (99%), and (2) whether there is a difference between the vaccination rates in a group of children with a high incidence (100%) of peanut allergies versus a group of children with a relatively low incidence (1%) of peanut allergies.

II. Anecdotal Evidence from Old-Order Amish in Lancaster, PA The first step in my research was to identify an undervaccinated population in the United States. Multiple studies suggest that the Old-Order Amish have low rates of vaccination. Although no comprehensive survey has been done, an August, 2006 report published in the CDC MMWR indicates that the vaccination rate among Amish communities may be as low as 28%. Another report published in the December, 2006 issue of Pediatric Infectious Disease Journal suggests that the vaccination rate may be as high as 84%, which is still relatively low compared to the national average rate of 95.8%.

Amish religious doctrine does not prohibit vaccination; however, it does not encourage it either. The Old-Order Amish practice separation from the world through group solidarity, caring for their own, and avoiding any form of dependence on government assistance. Health is considered primarily a gift from God, rather than the result of preventive medicine. Since there are no published studies regarding the incidence of peanut allergies among Amish children, my first step was to contact allergists practicing near the Old-Order communities living in and around Lancaster, PA. None of the allergists contacted had any record of an Amish child with peanut allergies, but several indicated that they didn't think the Amish would come to an allergist.

Most Amish families do not have phones or email addresses. So I decided to travel to Lancaster, PA in search of anecdotal information about Amish children with peanut allergies. There are approximately 2,000 Old-Order Amish families living in the Lancaster area. A nationwide survey published in the 2003 Journal of Allergy and Clinical Immunology indicates that 3% of families in the United States have one or more people with a nut allergy. Based on this statistic, I should have found as many as 60 children with nut allergies in the Lancaster area. I found evidence of only one child with a peanut allergy, at a clinic run by Dr. Holmes Morton in Strasburg, PA.

One problem with this anecdotal evidence from the Old-Order Amish in Pennsylvania - aside from the fact that it is not the result of a systematic survey - is the fact that the Old-Order Amish are a relatively genetically homogenous group who live in a rural farm environment. So the anecdotal evidence suggesting that there is low rate of peanut allergies among Amish children might be explained by either a genetic trait or the "hygiene hypothesis." For this reason, I decided to try to find a more diverse group of unvaccinated children.

III. Vashon-Ridgefield Case-Control Study
My control group is a cohort in Ridgefield, CT. I realized that I needed a group of children with a low vaccination rate in a community that is as similar as possible to Ridgefield. That is, I needed a group of largely unimmunized children living in a suburban or non-agricultural community. But, given the fact that a large percentage of the 4.2% of US children who are not immunized live in agricultural communities (where public health outreach programs are not as effective), is there any such group? Searching the internet for articles on immunization rates provided me with the answer: Vashon, WA.

Vashon Island, a quick 20-minute ferry ride across Puget Sound from Seattle, is a middle and upper-class community of approximately 10,000. It is a haven for alternative medicine, where therapies such as homeopathy and acupuncture are popular and more than 20% of the island's 1,600 school-age children have legally opted out of vaccination against childhood diseases by obtaining "philosophical exemptions" - that in Washington and several other states can be obtained by simply signing a form. (Ironically, Vashon also happens to be the home of Dr. William Foege, who served as the Director of the CDC from 1977-1983 and is known for his implementation of an international childhood vaccination initiative which resulted in a dramatic increase in worldwide immunization levels and the virtual eradication of diseases such as smallpox.)

I found the email addresses for the nurse at Chautququa Elementary School and McMurray Middle School (Kate Packard), and the parent organizer of a "Vashon Babes & Toddlers Meetup Group" (Jennifer Olsen), and contacted them both asking if they might be able to help with the research project. Kate Packard responded with the following data:

Chautauqua Elementary School
650 Children Enrolled 149 Exempt from Immunization 3 Peanut Allergies
McMurray Middle School
350 Children Enrolled 40 Exempt from Immunization 0 Peanut Allergies

She also noted that all 3 of the children with peanut allergies are immunized. Jennifer Olsen put me in touch with Lori Means, a Parent Educator at the Vashon Youth and Family Services, who in turn put me in touch with a local physician, Dr. Laurel Kuehl. Dr. Kuehl had the interesting idea of starting at an allergy clinic with a group of kids with nut allergies, and then finding out the percentages of vaccinated and unvaccinated kids in that group and comparing the numbers to the vaccine rate in that community. She also pointed out that I should look for variables such as "if the child was breast fed (less likely to have food allergies)" and "how early solids were introduced (more likely to have food allergies)". Both of these ideas were pursued in the "Peanut Allergy Survey" section of the research.

In Ridgefield, I contacted the nurses at East Ridge Middle School (Debra Butler) and Ridgebury Elementary School (Linda Miller). They responded with the following data:
Ridgebury Elementary School
464 Children Enrolled 8 Exempt from Immunization 12 Peanut Allergies

East Ridge Middle School
738 Children Enrolled 30 Exempt from Immunization 10 Peanut Allergies

All of the 22 children with peanut allergies in these two schools are fully immunized. As the data in these tables indicate, there is a 8/464 = 1.7% vaccination exemption rate at Ridgebury Elementary School in Ridgefield, CT, versus a 149/650 = 22.9% vaccination exemption rate at Chautauqua Elementary School in Vashon, WA. The number of children with peanut allergies, on the other hand, is 2.6% of the population at Ridgebury, and only 0.46% of the population at Chautauqua. These results are represented in the graphs below:
The results for middle school students were similar. There is a 0% vaccination exemption rate at East Ridge Middle School in Ridgefield, CT, versus a 14% vaccination exemption rate at McMurray Middle School in Vashon, WA. The number of children with peanut allergies, on the other hand, is 1.4% of the population at East Ridge, and 0% of the population at McMurray. These results are represented in the graphs below:
As noted above, 100% of the children with peanut allergies at all four schools were immunized.

IV. Online Peanut Allergy Survey To follow up on Dr. Laurel Kuehl's idea of "going about it backwards" - i.e., looking at a group of children who have nut allergies, and comparing the rate of vaccination in that group with the rate of vaccination in the community as a whole - I created an online survey using surveymonkey.com. The survey was designed to collect immunization information from a group with peanut allergies so that I could compare the vaccination rates in the group with the overall vaccination rates in the United States. It also asked questions about when the child was weaned and how early solid foods were introduced, as Dr. Kuehl recommended. The survey collected (and is still collecting) data from parents of children with peanut allergies via various peanut allergy web sites on the internet.

As of March 13, 2007, there were a total of 328 responses to the survey, all from parents or guardians of children with peanut allergies between the ages of 6 months and 16 years. (Please note that the data in this section is slightly different than the data presented on my research project board due to the increase in the number of respondents.) The results of the immunization questions on the survey indicate that there are only 6 children in the group with vaccination exemptions (due to medical, religious, or personal/philosophical reasons). Out of the 328 respondents, 312 are from the United States. So the rate of vaccination exemption is 6/312 = 1.9% in the group of children with peanut allergies, as contrasted with the national average 4.2% reported by the CDC. The following tables and charts represent additional data gathered through the surveys:
"Which of the following best describes the area you live in?"
Urban 24 Suburban 244 Rural 28
"When was the child weaned?" Not breast fed 0-1 month 1-3 months 3-9 months Older than 9 months Not sure 42 15 28 68 154 0 "At approximately what age were solid foods introduced?" 2-4 months 45 5 months 77 6 months 122 7-9 months 46 Older than 9 months 16 Not sure 5

V. Conclusion
This goal of this research project was to determine whether there is a difference between the incidence of peanut allergies in a group of children with a relatively low vaccination rate (80%) versus a group of children with a high vaccination rate (99%), and whether there is a difference between the vaccination rates in a group of children with a high incidence (100%) of peanut allergies versus a group of children with a relatively low incidence (approximately 1%) of peanut allergies.

My original hypothesis was that there would be a statistically significant difference, indicating a possible link between vaccines and peanut allergies. The data collected in all three parts of this study unambiguously support the hypothesis. In the case of the peanut allergy surveys, there was a greater than 2% difference between vaccination rate in the nationwide group of children with peanut allergies and the national vaccination rate as reported by the CDC. The results of the data collected in the Vashon-Ridgefield case-control study are even more dramatic. The elementary school and middle school population in Vashon currently has a vaccination compliance rate of 80.2%, and a peanut allergy rate of only 0.3%. An elementary and middle school population of roughly the same size in Ridgefield currently has a vaccination compliance rate of greater than 99.9%, and a peanut allergy rate of 1.8% - six times larger than that of Vashon.

The primary challenge I faced during the course of this research project was identifying a group of unimmunized children in an area that is geographically and demographically similar to Ridgefield, CT. Although aggregate statistics by state and county are available, there is no way to quickly determine whether a low immunization rate in a particular area is due to medical, religious, or primarily philosophical/personal exemptions. A national registry or data base of vaccination and basic health information in which names, addresses, and other sensitive information is concealed would be an invaluable resource to researchers and epidemiologists trying to understand the nature of disease.

I also encountered problems in trying to collect data from Old Order Amish communities. One way to extend this research project might be to do a more in-depth analysis of peanut allergies and vaccinations among the Amish. In particular, I would be interested in consulting with Amish community leaders (bishops and midwives) who might be able to provide me with information and guidelines on how to conduct a survey of Old Order Amish families that would respect their values and beliefs. Several studies have already been done on vaccination rates in particular Amish communities, but none have looked at the incidence of peanut allergies.

Scientists must be aware of the effects and impact of their research, and one thing I should emphasize is that the conclusion of this study does not suggest that children should not be immunized. Dr. Edward Rothstein, a Pennsylvania pediatrician who helps the American Academy of Pediatrics make immunization recommendations, puts it well: "I remember how the fear of polio changed our lives - not going to the swimming pool in the summer, not going to the movies, not getting involved with crowds. I remember pictures of wards full of iron lungs, hundreds in a room, with kids who couldn't breathe in them. It affected daily life more than AIDS does today." Similar stories can be told for other diseases which have become virtually unheard of; in the 1960s, for example, as many as 60,000 children were born with small heads, or deaf, or blind with cataracts because their pregnant mothers had been exposed to rubella. There are well-documented side effects and risks of vaccines, including the risk of children dying from vaccination, but these side effects and risks clearly do not outweigh the risks of the diseases themselves.

What this research study does suggest is that more research is required to understand the way in which early childhood vaccinations might affect the development of the immune system. Given that some autoimmune diseases such as Guillain-Barré syndrome, transverse myelitis, and multiple sclerosis are known or suspected to be exacerbated by vaccines, it is reasonable to think that vaccines might also act as an environmental trigger for other immune conditions - particularly when they are introduced in rapid succession during early childhood. This study doesn't reach any causal conclusions about the link between vaccines and what some health care professionals call the "peanut allergy epidemic." However, from a public health policy point of view, it might be prudent to consider spreading the immunizations out over a longer period of time, rather than trying to give 22 vaccines in the first 3 years of life - particularly now that we as a nation have achieved herd immunity. From a clinical standpoint, there may be reasons for children who have family histories of certain immune conditions to delay or even be exempted from vaccines. Although all 50 states allow medical exemptions for children who are immuno-compromised, none currently allow medical exemptions for children who have food allergies or who have a family history of auto-immune conditions. There are 17 states who do allow exemptions for personal or philosophical reasons, but the option to decline vaccinations is not typically well-publicized.

From Devi: "Thank you so much for posting my survey! Using the data I collected, I was able to win first place at the regional science fair and qualify for states. I couldn't have done any of this without your help and generosity. I've attached the research report for my project. Please feel free to share it with anyone who's interested."

Devi Lockwood wrote this in 2006

This paper was sighted in Nicole Smith Allergic Child's article
Possible Causes of Food Allergies (under Vaccinations)

Devi will graduated from Harvard University in 2014. Read more on what she is doing now, on her blog Devi K Lockwood I wish her the best in her career in the written word, many would have liked her to continue her interest in Science as a career, including myself.