IDPH Cancer Assessment, Lake County, Illinois

In August of 2018, the United States Environmental Protection Agency (U.S. EPA) released the 2014 National Air Toxics Assessment (NATA) which identifies areas of the country that may have an increased cancer risk based on emissions estimates for toxic compounds and screening level air quality modeling.  The NATA identified the Waukegan-Gurnee area, in Lake County, Illinois, as potentially having elevated cancer risk, with ethylene oxide (EtO) identified as the predominant pollutant driving the elevated estimate.  Following the Illinois Department of Public Health’s (IDPH) investigation and release of a report about cancer incidence surrounding the Sterigenics facility in Willowbrook, Illinois, concerned community residents reached out to  IDPH with questions around residential EtO exposure.  IDPH began assessing cancer incidence in this community using data from the Illinois State Cancer Registry (ISCR).  The ISCR has received gold certification from the North American Association of Central Cancer Registries for 20 consecutive years.  Only those registries meeting the highest standards are awarded gold certification.

We cannot confirm that EtO caused cancer in Lake County.  This study is designed to look for correlations and CANNOT determine cause.  A correlation between an exposure and cancer does not prove that the exposure caused the disease.  Establishing the likelihood of a causative relationship between a chemical exposure and cancer requires multiple studies and multiple lines of evidence that consistently point to the link between the exposure and cancer.

Two groups of cancers were examined.  The first group included breast cancer and lymphohematopoietic cancers.  Certain lymphohematopoietic cancers have consistently been associate with EtO.  These include Non-Hodgkin’s lymphoma, Hodgkin’s lymphoma, myeloma, and lymphocytic leukemia.  The second group included more common cancers such as prostate, ovarian, and bladder cancers.  Elevations in cancer incidence or excess cancers were determined by examining the number of cancers in the study area (observed cancers) and comparing them to the expected number of cancers that would be seen if the study area had the same incidence rate as the comparison group(s).

Rates for certain cancers were increased among residents residing in areas surrounding the Vantage and Medline facilities, while others were not.  

None of the four lymphohematopoietic cancers that previous studies have found to be associated with EtO, namely, female breast, non-Hodgkin lymphoma, Hodgkin’s lymphoma, and lymphocytic leukemia, were found to be increased in either males or females.  Another lymphohematopoietic cancer, myeloma, showed a statistically significant elevation among females relative to the Lake County average, but the difference was reduced to non-significant when compared to the Cook County average.

No lymphohematopoietic cancer displayed an increasing trend over time (1998 to 2017).  Male lymphocytic leukemia cases were elevated at the beginning of the time period,   but then decreased to expected levels in subsequent time periods.  Hodgkin lymphoma in females showed a steadily declining trend over time.

A separate analysis of pediatric cancer revealed that no childhood cancers were elevated.  

Several other cancer sites that have never been documented to be related to EtO were found to be higher (i.e., stomach and colorectal cancers for females and lung cancer in males) and other sites were found to be lower (i.e., female skin and bladder, and male bladder) than comparable populations.

The study results should be treated with caution, as the results were not consistent (i.e., male vs. female, and myeloma vs. other EtO related cancers) and the study has several important limitations and constraints (e.g., using residency as measurement of exposure, tying cancer to the residence at the time of diagnosis, lacking other risk factors, etc.).  Additional studies in the future are needed to confirm these findings.

The study has several important limitations and constraints (for a full listing of limitations please see the report).  Some examples include using residency as a measurement of exposure, tying cancer to the residence at the time of diagnosis, and not considering other risk factors.  A fuller description is provided in the study document (see link above).  This study, as with any epidemiologic study, cannot answer the question of what caused an individual’s cancer, nor can it speak to an individual’s future risk of cancer.  Additional studies are needed to confirm these findings.

To select appropriate study areas, we needed to consider three factors: scope, population, sample size.  First, the scope of the exposure.  We wanted to capture the affected area fully, but at the same time, we did not want to stretch the area so far that we might ‘dilute’ meaningful findings.  Second, available population numbers to determine if there were more cancers than expected.  The only population source was the U.S. Census Bureau and its population data are only available at certain geographical levels such as county, census tract, or census blocks.  The census tracts were used in our study as the smallest building blocks to create the study areas.  Third, sample size.  We needed large enough numbers to do the analysis.  An appropriate study area must meet all these three factors. 

The Vantage and Medline facilities are approximately 3 miles apart.  Initial examination of two separate study areas revealed several census tracts that were shared, as well as somewhat small population figures that would lessen the study’s ability to detect differences in two separate study areas.  Given the proximity of the two facilities, a single study area was utilized that centered on the EtO emission sources and contained the entirety of the modeled exposure areas.

Your area may not have been included in the study area for a couple of reasons.  It may have been well outside of the air sampling/modeling area.  Creating a study area that is too expansive could lead to the inclusion of individuals who may not have been exposed to EtO from the facilities of interest.  

All cancer cases who lived in the study area at the time of cancer diagnosis were included in the study, including people who were diagnosed with cancer while living in the area but later moved away.  People who only worked or attended schools in the area but did not live in the area when cancer was diagnosed were not included.  

The study used data from the Illinois State Cancer Registry (ISCR), which is the only population-based source for cancer incidence information in Illinois.  Cancer cases are collected through mandated reporting by hospitals, ambulatory surgical treatment centers, non-hospital affiliated radiation therapy treatment centers, independent pathology labs, dermatologists, and through the voluntary exchange of cancer patient data with other (mostly nearby) states.  In June 2020, for the 2017 data, ISCR received Gold certification – the highest standard for registry certification from the North American Association of Central Cancer Registries. 

Interviewing individuals in study areas can be fraught with inconsistencies and inaccuracies due to errors in recollection, known as recall biases and errors.  For example, a person who thought they, or their neighbor, had colon cancer, actually might have stomach cancer.  Some studies compared what people recalled to what was in their medical records, and found a huge difference, sometimes as much as 60-80%.  This recall bias can also be applied to risk factors.  If a person suspected that something was causing their cancer, they might remember that better than someone who did not make that connection.  The cancer information for this assessment came from medical records and pathology reports.  

IDPH used data covering 20 years, from 1998 to 2017.  The 20 years of data cover the years of operation of Vantage and Medline.  It also allows for the typical cancer latency period, the time between when a person is exposed and when they are diagnosed.  The latency period for lymphohematopoietic cancers (found to associated with EtO exposure) is 4-10 years and 10-15 years for solid tumors.

We also wanted to use the best data we have in terms of data completeness and quality.  The data before 1995 were not as good because the cancer registry was just established, and the collected data were not certified by the North American Association of Central Cancer Registries (NAACCR).  All data years we used for this study, 1998-2017, received the Gold Standard certification.

IDPH is not making any recommendations about moving.  The observed cancer cases over the 20-year period was 3% higher than expected in females and 4% lower in males.  Many Illinois counties have cancer rates that differ from each other by more than 5%.   

Risk is defined as the probability that an event will occur.  Everyone is at risk for cancer to some degree.  Not all people develop the same disease for the same reason (i.e., no one factor determines whether an individual will develop a disease). It is the interaction of many factors that produce disease (e.g., for cancer this could be genetics, immunity, diet, occupation, hormones, viruses, socioeconomic, lifestyle, age, or physical environment). 

Cancer does not develop immediately after contact with a cancer-causing agent (carcinogen). The time between the exposure to a carcinogen and medical diagnosis of cancer, called latency period, is often 4-10 years for lymphohematopoietic cancers (found to associated with EtO exposure) and 10 to 20 years solid tumors. This makes it very difficult to pinpoint what caused the cancer.  Cancers are usually related to long-term lifestyle behaviors (e.g., smoking) or significant exposure to carcinogens for many years.

Please contact the IDPH Division of Epidemiologic Studies, email: ethyleneoxide-cancer@illinois.gov

Cancer is a common disease, perhaps more common than many people realize. In the U.S., 1 in 2 men has a lifetime risk of developing cancer. For women, the lifetime risk is 1 in 3. The number of people with cancer is increasing in most communities because more people are living to the ages of greatest cancer occurrence. 

Many people can reduce their chances of developing or dying from cancer by adopting a healthier lifestyle and by visiting their doctor regularly for cancer-related checkups. Screening examinations, conducted regularly by a health care professional, can result in the detection of cancers of the breast, tongue, mouth, colon, rectum, cervix, prostate, testis, and melanomas at earlier stages, when treatment is more likely to be successful. More than half of all new cancer cases occur in the nine screening-accessible cancer sites listed above. 

The causes of most cancers are not well understood. Current knowledge suggests that many cancers are influenced by a combination of factors, including heredity, environment, and behaviors related to how we live, called lifestyle behaviors. Lifestyle behaviors that increase cancer risk include cigarette smoking, alcohol use, diet, obesity, and lack of physical activity, and they account for the majority of all cancer deaths in the United States. Environmental and occupational exposures to cancer-causing chemicals, ionizing radiation, and other agents produced by humans also significantly contribute to cancer risk. A World Health Organization report (2006) concluded that 16% of cancers in men (other than lung cancers) and 13% in women were attributable to environmental and occupational exposures in developed countries. 

It is a common perception that cancer is a single disease. In fact, cancer is many different diseases, each with differing rates of occurrence, risks, causes, and chances of survival.