A new study from the University of Iowa finds that children with autism spectrum disorder (ASD) are more than twice as likely to suffer from a food allergy than children who do not have ASD.
Wei Bao, assistant professor of epidemiology at the UI College of Public Health and the study’s corresponding author, says the finding adds to a growing body of research that suggests immunological dysfunction as a possible risk factor for the development of ASD.
“It is possible that the immunologic disruptions may have processes beginning early in life, which then influence brain development and social functioning, leading to the development of ASD,” says Bao.
The study analyzed the health information of nearly 200,000 children gathered by the U.S. National Health Interview Survey (NHIS), an annual survey of American households conducted by the U.S. Centers for Disease Control and Prevention. The children were between the ages of 3 and 17 and the data were gathered between 1997 and 2016.
The study found that 11.25 percent of children reportedly diagnosed with ASD have a food allergy, significantly higher than the 4.25 percent of children who are not diagnosed with ASD and have a food allergy.
Bao says his study could not determine the causality of this relationship given its observational nature. But previous studies have suggested possible links—increased production of antibodies, immune system overreactions causing impaired brain function, neurodevelopmental abnormalities, and alterations in the gut biome. He says those connections warrant further investigation.
“We don’t know which comes first, food allergy or ASD,” says Bao, adding that another longitudinal follow-up study of children since birth would be needed to establish temporality.
He says previous studies on the association of allergic conditions with ASD have focused mainly on respiratory allergy and skin allergy, and those studies have yielded inconsistent and inconclusive results. The new study found 18.73 percent of children with ASD suffered from respiratory allergies, whereas only 12.08 percent of children without ASD had such allergies, and 16.81 percent of children with ASD had skin allergies, well above the 9.84 percent of children without ASD.
“This indicates there could be a shared mechanism linking different types of allergic conditions to ASD,” says Bao.
Bao says the study is limited in that the NHIS depends on respondents to voluntarily self-report health conditions, so the number of children with ASD or allergies may be misreported by those taking the survey. But he says the large number of respondents and ethnic and gender cross-representation of the survey are major strengths.
The study, “Association of Food Allergy and Other Allergic Conditions with Autism Spectrum Disorder in Children,” was published online in the June 8 issue of JAMA Network Open. The first author is Guifeng Xu, PhD candidate in the UI College of Public Health and graduate research assistant in the UI Roy J. and Lucille A. Carver College of Medicine. Additional co-authors include Linda G. Snetselaar, professor of epidemiology in the UI College of Public Health; Jin Jing, professor of maternal and child health at the Sun Yat-Sen University in China; Buyun Liu, postdoctoral researcher in the UI College of Public Health; and Lane Strathearn, professor of pediatrics in the Carver College of Medicine.
This story originally appeared in Iowa Now
Additional Media Coverage
American Journal of Managed Care
The number of children in the United States diagnosed with autism spectrum disorder may be significantly higher than previously thought, according to a new University of Iowa analysis published in the Journal of the American Medical Association (JAMA).
The analysis of data from the U.S. Centers for Disease Control and Prevention (CDC) suggests that 2.4 percent of American children between the ages of 3 and 17—or 1 in 41—have been diagnosed with autism, higher than most earlier estimates of about 1.46 percent or 1 in 68 children. Wei Bao, assistant professor of epidemiology in the UI College of Public Health and corresponding author of the analysis, says the higher number shows the need for officials to think about reallocating health care resources to care for significantly more people with autism.
“Previous thinking about autism is that it is very rare, but this study tells us that it is no longer something that is very rare,” says Bao. “This should cause us to reconsider what our future priorities in research, service, and policy should be regarding children who have autism spectrum disorder. Clearly, we need more people to care for children with autism.”
The UI analysis used nationally representative data from the CDC’s National Health Interview Survey from 2014 to 2016, which collects data on a broad range of health topics through thousands of in-person household interviews each year. As part of the interview, CDC survey-takers ask respondents if the randomly sampled child living in the household has ever been diagnosed with autism.
Bao says the previous estimate of 1.46 percent was derived from the CDC’s Autism and Disabilities Monitoring Network (ADDM), which collects data from the health and special education records of 8-year-old children at 11 selected sites across the United States. The UI analysis was based on a nationally representative sample of children ages 3 to 17, and Bao cautions that these methodological differences in study design make direct comparisons between the two databases difficult.
The analysis also confirms earlier findings about gender and racial/ethnic disparity of autism in U.S. children, that it is much more frequent in boys (3.54 percent) than girls (1.22 percent), and less frequent in people of Hispanic origin (1.78 percent) than in non-Hispanic whites (2.71 percent) or non-Hispanic blacks (2.36 percent).
It found the highest prevalence of autism in Northeast states, at 3.05 percent. The Midwest was at 2.47 percent, the West at 2.24 percent, and the South at 2.21 percent. Bao speculates that rates are lower in the South and West because higher percentages of the population living in those states are Hispanic, a population that tends to have a lower prevalence of autism overall.
Bao says the limitation of the analysis is that the data is self-reported by the household respondent to the CDC survey-taker and is not subject to any third-party adjudication.
The analysis does not identify a cause for the increasing number of autism spectrum disorder cases. Bao says greater awareness among parents and health care providers might be the cause of some of the increase, but environmental and genetic factors likely are responsible for a large part of the gap. He points to previous studies—including one of his own—that show children are at greater risk of autism if their mothers have diabetes before or during pregnancy; since diabetes is often caused by obesity, the increase could be linked to the increasing weight of Americans.
“Autism is a highly complex disease caused by multiple genetic and environmental factors,” Bao says. “Maternal diabetes could be one of those factors, but it is not the only one. We need to find more about the underlying driving factors.”
The paper, “Prevalence of Autism Spectrum Disorder Among US Children and Adolescents, 2014–2016,” was published in the January 2018 issue of JAMA. Its first author is Guifeng Xu, PhD candidate in the UI College of Public Health and graduate research assistant in the UI Roy J. and Lucille A. Carver College of Medicine, and is co-authored by Buyun Liu, postdoctoral researcher in the College of Public Health, and Lane Strathearn, professor of pediatrics in the Carver College of Medicine.
This story originally appeared in Iowa Now.
Additional Media Coverage
Iowa Public Radio
Los Angleles Times (also published in the San Francisco Chronicle and New York Daily News)
WTNH-TV, Hartford, Conn.
New research from the University of Iowa shows that polychlorinated biphenyls (PCBs), chemicals known to cause cancer in humans, are present in older schools and that the source of the PCBs is most likely outdated building materials, such as window caulking and light ballasts.
The multi-year study by the Iowa Superfund Research Program at the UI is the largest yet to examine airborne PCBs in schools. It shows that though the presence of PCBs can vary from school to school and even classroom to classroom, children’s exposure rates are roughly the same in rural and urban areas. It also shows that exposure to PCBs by inhalation may be equal to or higher than exposure through diet, a finding that surprised researchers.
Researchers collected indoor and outdoor air samples at six schools in Iowa and Indiana from 2012 to 2015. And though none of the schools had PCB levels high enough to meet federal standards for immediate remediation, researchers say the study is important because it shows that reduction of airborne PCBs in schools could be accomplished by removing old caulk around windows and modernizing light fixtures.
The study, published in Environmental Science & Technology, comes at a time of increasing concern over PCBs in schools. In 2014, a school in Lexington, Massachusetts, was shuttered after high levels of PCBs were found inside the building. New York City officials recently completed a multi-year program to replace PCB-laden light ballasts in more than 800 schools. And last year, parents in Malibu, California, won a publicized court battle to remove PCBs in schools.
“Due to the presence of PCBs in the environment, humans are easily exposed to them,” says Rachel F. Marek, assistant research scientist at the UI College of Engineering and lead author of the study. “Exposure of school-aged children to PCBs is of particular concern because these are compounds we know impair memory and learning and cause cancer in humans.”
Besides PCBs, researchers looked for the first time at OH-PCBs, chemical compounds similar to PCBs, in schools. Although there is still much to learn about OH-PCBs and their potential health risks, some scientists believe that OH-PCBs could be more toxic than PCBs.
PCBs are man-made chemical compounds that were used in industrial and commercial applications from 1929 to 1979, when they were banned. However, PCBs are still present in the environment, especially in areas with high concentrations of heavy industry and in buildings constructed in the ’50s, ’60s, and ’70s, when PCBs were used in various building materials. Many public schools were built with PCB-laced materials, and public health experts estimate that as many as 25,920 schools nationwide still have window caulking that contains the chemical compounds.
Schools are not required to measure PCBs; however, the United States Environmental Protection Agency provides guidelines for acceptable PCB levels. UI researchers studied PCB levels in four urban schools in East Chicago, Indiana, and two rural schools in Columbus Junction, Iowa. PCB levels inside the schools were below the current EPA action level of 500 nanograms per cubic meter, with a maximum reading of 194 ng/m3. The highest PCB levels were reported at two schools in East Chicago. Both were built before the PCB ban in 1979.
“Our nation’s schools must provide a safe and healthy environment for growing and learning,” says Peter S. Thorne, professor and head of occupational and environmental health in the UI College of Public Health and principal investigator of the study. “In addition to protecting children from risks such as asthma and obesity, schools need to be free of elevated exposures to persistent pollutants, including lead and PCBs.”
The exact level of PCB exposure at which harm is caused to children is unknown. This is a question that the Iowa Superfund Research Program is trying to answer through its research.
The four schools in East Chicago are not far from the Indiana Harbor and Ship Canal, an industrial waterway that is a known source of airborne PCBs. Researchers hypothesized that the schools’ proximity to the canal would influence PCB levels inside the schools. However, though indoor air samples from two of the schools resembled air samples collected near the canal, indoor air samples from the other two schools presented very different results, which researchers concluded were caused by the presence of PCBs typically linked to window caulking and light ballasts.
Tests of indoor and outdoor air at two schools in Columbus Junction, which has no known source of industry-related PCB contamination, presented further evidence of PCB contamination via old building materials. Paint pigments also were found to be a likely source of PCBs in the schools.
“This is the first time we’ve been able to pinpoint the source of PCBs inside schools,” says Keri C. Hornbuckle, professor of civil and environmental engineering at the UI College of Engineering and project leader for the Iowa Superfund Research Program. “This study shows that the indoor air is contaminated, and that contamination is due to materials that remain in use in the school buildings.”
UI researchers are planning their next PCB project, one that will measure PCB levels in different rooms in the same pre-1979 building. They also plan to offer free indoor PCB testing to schools.
“The results of our study are really quite chilling,” says Hornbuckle. “To put it simply, any school that hasn’t been remodeled since the 1970s may have high levels of PCBs in the air, which children breathe day in and day out.”
The Iowa Superfund Research Program (ISRP) at the University of Iowa studies the sources, exposures, and toxicities of PCBs in the environment. This program, which is funded by the National Institute of Environmental Health Sciences, brings together scientists with expertise in toxicology, engineering, microbiology, public health, and chemistry.
The goal of the Airborne Exposure to Semi-Volatile Organic Pollutants (AESOP) Study, part of ISRP, is to measure PCB exposure levels indoors and out and to track PCB exposure among 345 children and their mothers. The study, which began in 2006, already has provided new insight into airborne PCB exposure and challenged prevailing views on how humans are exposed to PCBs.
This story originally appeared in Iowa Now.
Additional Media Coverage
Study: Replace aging building materials in schools
Four East Chicago schools Have PCB levels
The Times of Northwest Indiana
U of I study finds PCBs in Columbus Junction Schools
The Muscatine Journal
UI finds PCBs in schools
Des Moines Business Record
One of the most effective ways to improve the health of large numbers of people is through public health policy. Examples of successful policy initiatives with far-reaching impact include seat belt laws, tobacco regulations, and banning lead from paint.
To help faculty enhance their skills in translating public health research into practice and policy, the College of Public Health recently established a new Policy Fellow Program. Offered through the college’s Iowa Institute of Public Health Research and Policy (IIPHRP), the year-long program creates opportunities for primary faculty to attend training workshops, connect with experts, interact with policymakers and stakeholders, and develop and implement a policy-related project.
“As researchers, we have a desire to impact our community, and that means translating evidence into policy or practice change. Building stakeholder coalitions and creating effective dissemination plans are strategies that work, but are not skills most researchers have been formally trained in,” says Vickie Miene, IIPHRP interim director. “The Policy Fellow Program provides a supportive and individualized learning environment for faculty to experiment and learn these important skills while simultaneously translating their research into public health practice.”
Each fellow develops and implements an “action learning project” that focuses on a critical public health topic. The project includes at least one stakeholder meeting and a final product, such as a policy brief, proposed legislative language, or a how-to guide, to disseminate at the end of the fellowship.
“Our goal is to select fellows who are enthusiastic, who address significant public health issues, and who propose a project that will likely generate an effective example of translating research into policy or practice,” says Miene.
The 2016-2017 Policy Fellows are Mary Charlton, assistant professor of epidemiology; Cara Hamann, associate in epidemiology; and Kelli Ryckman, associate professor of epidemiology. Their projects, described in the following pages, are currently underway and will wrap up at the end of summer 2017, with final summaries posted on the IIPHRP web site.
Modernizing Cancer Reporting
Mary Charlton, assistant professor of epidemiology, is also an investigator with the Iowa Cancer Registry (ICR). Her project recommends policy changes to modernize Iowa’s cancer reporting laws.
The ICR is a population-based cancer registry that has served the State of Iowa since 1973. The registry is funded by the National Cancer Institute as part of its Surveillance, Epidemiology, and End Result’s (SEER) Program. Through its registries, the SEER Program routinely collects cancer data and patient demographics and is the authoritative source of information on cancer incidence and survival in the U.S.
According to SEER, “The ICR provides accurate and thorough reporting of cancer disease. The ICR has been consistently recognized for its extremely high quality data. However, Iowa has some of the weakest reporting laws compared to other states. Iowa also has one of the lowest rates of e-path (electronic) reporting of all SEER registries.”
In Iowa, about 25,000 cancer abstracts per year are collected from hospitals, pathology laboratories, cancer treatment centers, and physician practices. Two-thirds of the abstracts are collected by the 14 Iowa hospitals with accredited cancer centers. Trained staff employed by the ICR collect cancer data from the other 104 hospitals and non-hospital sites of diagnosis. The ICR maintains the confidentiality of the patients, physicians, and hospitals providing data.
To ensure continued funding from NCI, it is necessary to address the low rate of electronic reporting and Iowa’s weak cancer reporting requirements, which require a disproportionate amount of NCI funds to be spent on cancer data collection compared to other SEER registries.
“Existing Iowa cancer reporting requirements were developed long before widespread use of electronic record systems and at a time when nearly every cancer patient received treatment in a hospital,” Charlton says. “Cancer reporting requirements must catch up with technology.”
Charlton’s policy recommendations are to:
- Require all reporting entities to provide data electronically whenever possible.
Benefits include more timely case reporting, enhanced patient privacy, more complete reporting, and significant labor savings.
- Add provisions for cost-sharing mechanisms.
Nearly $2 million of the ICR budget is spent on cancer data collection by trained staff, which results in Iowa having the largest cost per case of all SEER registries.
- Clarify the definition of who is required to report.
Currently, as the Iowa Code is written, it is often interpreted that only hospitals have to report cancer cases. This leads to underreporting of cancers that can often be treated in a clinic or outpatient setting.
Improving Newborn Screening
Kelli Ryckman, associate professor of epidemiology, has long-standing research interests in genetics and the complications of preterm birth.
“My passion is finding ways to improve the care of preterm, low-birth-weight, and sick newborns in the Neonatal Intensive Care Unit,” Ryckman says about her motivation to apply to the Policy Fellow Program.
Her project is to refine the newborn screening policy for babies in the NICU that will result in fewer false-positive screens.
“Through the program, I hope to build consensus based on the evidence and disseminate a uniform set of guidelines for newborn screening in the NICU that allows for better use of resources in Iowa and beyond,” Ryckman says.
State newborn screening programs are critical public health services aimed at screening every child born for specific genetic, endocrine, and metabolic conditions that, if left untreated, can lead to severe disability or death.
There are about 39,000 births in Iowa each year, and approximately 10 percent of all births in Iowa are preterm, low-birth-weight, or transferred to the NICU. This 10 percent also makes up over a third of the follow-up workload required for the newborn screening program due to the high false-positive rate in this population. Transfusions, parental nutrition, medications, and timing of the test in the NICU can affect the validity of the newborn screening test.
“It’s well-recognized that premature, low-birth-weight, or sick newborns are more likely to falsely screen positive for one or more of the newborn screening tests,” Ryckman says. “False positives have adverse impact on parents and require additional testing, which can add to the stress on the baby.”
To address these challenges, the Clinical and Laboratory Standards Institute, a non-profit organization dedicated to improving clinical laboratory testing quality, developed a guideline for screening preterm, low-birth-weight, and sick newborns in 2009. Recent surveys have shown that despite the recommendations, many physicians caring for newborns in the NICU are unaware of its existence and that only 25 percent of states have adopted or were planning to adopt the recommendations, the American Academy of Pediatrics reports. Iowa has not yet adopted the recommendations.
Ryckman’s policy recommendation is to develop and disseminate a uniform set of guidelines to be implemented across Iowa. These guidelines will align with the national recommendations, provide for better utilization of resources, and decrease false positives in this special population of newborns.
Increasing Bicycle Safety
Cara Hamann, an associate in the Department of Epidemiology, developed an issue brief to highlight bicycling safety research, a topic that has received attention recently due to an increasing number of automobile and bicycle crashes in Iowa. In collaboration with the Iowa Bicycle Coalition, she also held a bicycle safety policy action forum of stakeholders in December 2016 and produced a forum recap document containing statewide key issues and action items.
“There were 340 bicycling injuries and 8 fatalities as a result of bicycle–motor vehicle collisions in Iowa in 2016,” says Hamann. “That’s up from 3 deaths in 2014, and 5 in 2015.”
Bicyclists have higher crash-related risk of injury and death due to their lack of physical protection, slow speed, and size differential compared to other roadway users. In alignment with the Iowa Bicycle Coalition 2017 legislative initiatives, Hamann supports several changes in Iowa code to make bicycling safer.
Require motorists to change lanes when passing bicyclists
Motorists hitting bicyclists from behind have accounted for about half of fatal bicycle crashes in Iowa over the past five years, Hamann says. Penalties assessed to motorists at fault in these fatal crashes are generally a traffic ticket with an enhanced penalty of a $1,000 fine and six-month license revocation. A measure to require vehicles to change lanes to pass bicyclists passed both the Senate and House sub-committees (HF 513, SF 450), but did not advance further.
Increase penalties and make distracted driving a primary offense
Most Iowa drivers don’t receive charges in bicycle–motor vehicle crashes, regardless of fault. Two bills that increase penalties were passed during the 2017 legislative session and signed into law by Gov. Branstad. One bill (SF 234) changes hand-held electronic communication (writing, sending, or viewing electronic messages) from a secondary to a primary offense and will go into effect starting July 1. However, this new law still allows drivers to use hand-held devices to make phone calls or check GPS directions. The second bill (SF 444) increases penalties to allow drivers to be charged with reckless driving when their hand-held device use causes a fatality. This type of charge is a Class C felony.
Increase appropriations for bicycling infrastructure
Results from a study of 294 sites in Iowa suggest that bicycle lanes and shared lane markings can reduce crash risk by as much as 60 percent. In a sample of Iowa drivers, shared lane markings improved driver positioning during overtaking, especially among older drivers.
“I was motivated to apply to this program to develop skills to translate my research into tangible products to be used for policy change and real impacts on safety,” says Hamann. “I want to use this opportunity to draw attention to bicycle safety and complement the current grassroots efforts underway in the bicycling community.”
Portraits by John Choate
This story originally appeared in the spring 2017 issue of InSight.