Determining statistical significance
To confirm the existence of a cluster, investigators must show that the number of cancer cases in the cluster is statistically significantly greater than the number of cancer cases expected given the age, sex, and racial distribution of the group of people who developed the disease. If the difference between the actual and expected number of cancer cases is statistically significant, the finding is unlikely to be the result of chance alone. However, it is important to keep in mind that even a statistically significant difference between actual and expected numbers of cases can arise by chance.
Determining the relevant population and geographic area
An important challenge in confirming a cancer cluster is accurately defining the group of people who should be considered potentially at risk of developing the specific cancer (typically the total number of people who live in a specific geographic area). When defining a cancer cluster, there can be a tendency to expand the geographic borders as additional cases of the suspected disease are discovered. However, if investigators define the borders of a cluster based on where they find cancer cases, they may alarm people about cancers that are not related to the suspected cluster. Instead, investigators first define the population and geographic area that is “at risk” and then identify cancer cases within those parameters.
Identifying a cause for a cluster
A confirmed cancer cluster—that is, a finding of a statistically significant excess of cancers—may not be the result of any single external cause or hazard (also called an exposure). A cancer cluster could be the result of chance, an error in the calculation of the expected number of cancer cases, differences in how cancer cases were classified, or a known cause of cancer, such as smoking. Even if a cluster is confirmed, it can be very difficult to identify the cause. People move in and out of a geographic area over time, which can make it difficult for investigators to identify hazards or potential carcinogens to which they may have been exposed and to obtain medical records to confirm the diagnosis of cancer. Also, it typically takes a long time for cancer to develop, and any relevant exposure may have occurred in the past or in a different geographic area from where the cancer was diagnosed.
|Where can people get more information about cancer clusters?
In addition to state and local health departments and cancer registries, the following agencies may have more information about cancer clusters.
Agency for Toxic Substances and Disease Registry (ATSDR)
The CDC’s ATSDR conducts public health assessments of potentially hazardous waste sites, performs health consultations on specific hazardous substances, designs and conducts health surveillance programs, and provides education and training about hazardous substances. Information about public health assessments conducted by ATSDR can be found on its Public Health Assessments and Health Consultations page. Reports can be searched by state or U.S. territory. Contact information for ATSDR regional offices is available online.
The CDC’s NCEH works to promote healthy and safe environments and prevent harmful exposures. The NCEH website includes general information about cancer clusters, links to resources, and answers to frequently asked questions.
National Institute for Occupational Safety and Health (NIOSH)
The HHE Program of CDC’s NIOSH investigates potentially hazardous working conditions, including suspected cancer clusters. Employees, authorized employee representatives, and employers can request these evaluations. HHE reports are available on the NIOSH website.
Office of Occupational Medicine
OSHA’s Office of Occupational Medicine performs workplace-related case evaluations and cluster investigations, including medical record reviews, employee interviews, and medical screening activities.
1. Centers for Disease Control and Prevention. Investigating Suspected Cancer Clusters and Responding to Community Concerns: Guidelines from CDC and the Council of State and Territorial Epidemiologists. Morbidity and Mortality Weekly Report 2013;62(RR08):1-14. [PubMed Abstract]
Source: National Cancer Institute.