History of the Surveillance, Epidemiology, and End Results (SEER) Program (2024)

Search for other works by this author on:

Article history

Abstract

There is a building momentum propelled by the Cancer Moonshot Initiative to achieve bold public health goals such as reducing the cancer death rate by half within 25 years. The optimal path forward to achieve the goal is preventing the disease and its intermediary outcomes such as progression and recurrence. Thus, perfecting methods to assess the progress toward primary and secondary prevention has been and will continue to be of paramount importance. The Surveillance, Epidemiology, and End Results (SEER) Program established in 1973 was the first, and remains the premier, laboratory for experimenting with new methods for cancer data collection and translating the data into cancer statistics. The history of the SEER Program has been a narrative of discovery about how cancer impacts population health, a history of geographic expansion coupled with forays into the details of every cancer case, and a history of developing infrastructure and methods for supporting the war on cancer.

Early years

When the National Cancer Act (NCA) was passed in 1971, it brought substantial attention and funding to the National Cancer Institute (NCI) with a renewed and refocused mission to address the cancer epidemic in the United States (1). It also included language that ultimately led to the establishment of the SEER Program. Section 409 of the NCA called for the director of the NCI to “establish programs as necessary for cooperation with state and other health agencies in the diagnosis, prevention, and treatment of cancer” (1). With that statutory language, the groundwork for the establishment of SEER was put in place. As SEER celebrates its 50th anniversary as the preeminent national cancer surveillance program in the United States, this article will highlight the changes to the SEER Program since its inception, while other articles in this monograph will address how SEER has contributed to the cancer research community and improvements in cancer outcomes.

Prior to the enactment of the NCA and SEER, cancer surveillance conducted by NCI consisted of 2 programs: the End Results Program and the Third National Cancer Survey (1). The passage of the NCA provided an opportunity to refine these 2 programs through consolidation into SEER with a clear set of programmatic goals and objectives, which at the time included

1. developing and reporting estimates of cancer incidence and mortality on a periodic basis for the total United States;

2. monitoring annual cancer incidence trends to identify unusual changes in specific forms of cancer occurring in population subgroups defined by geographic, demographic, and social characteristics and providing insight into their etiology; and

3. promoting studies designed to identify factors pertaining to the environment, occupation, socioeconomic status, tobacco, diet, screening practices, patterns of care, and determinants of the length and quality of patient survival that are amenable to cancer control interventions.

The SEER Program officially began with the collection of cancer cases starting on January 1, 1973, with participation of state central cancer registries, including Connecticut, Iowa (also celebrating its 50th anniversary in 2023), New Mexico, Utah, and Hawaii, along with regional registries in Detroit and San Francisco-Oakland. Geographic areas were selected, in part, based on their ability to operate and maintain a high-quality population-based cancer reporting system and for their epidemiologically significant population subgroups. The SEER Program, at its inception, represented almost 8% of the US population, with approximately 120 000 new cancer cases annually. In 1974, the New Orleans metro area, a 13-county region of Seattle-Puget Sound, and the Atlanta metro area were included. In 1978, 10 predominantly Black rural counties in Georgia were added. Native American people residing in Arizona were added in 1980. In 1983, four counties in New Jersey were added. Puerto Rico participated in the SEER Program until 1989 and spurred an effort to increase representation of minority populations that included Los Angeles County and 4 counties in the San Francisco Bay area, both of which joined in 1992. By 2000, SEER continued to represent 14% of the US population with approximately 160 000 new cancer cases annually (2).

The first major report published using data from the newly established SEER Program was first authored by Dr Charles Smart (3). The paper discussed the implication of findings from population-based data for the management of breast cancer. Following Dr Smart’s report, data collected by registries participating in the SEER Program have been used to describe and understand cancer trends, support cancer control decisions, and compare population-based cancer statistics by geographic areas (4). In addition, evidence derived from the SEER data has been used to support tobacco-related legislation and to identify population groups with higher risk of developing cancer or dying from cancer (5). The SEER Program provided the data for the analysis presented in the inaugural edition of the Annual Report to the Nation on the Status of Cancer, with the first edition including a special section on lung cancer and tobacco smoking (6).

SEER in the new millennium (2000–2016)

Accumulation of scientific knowledge in the field of oncology resulted in the era of genomic characterizations of cancer and the practice of precision medicine. These paradigmatic changes translated into increased opportunities for new classifications based on the phenotypic expression of the newly understood molecular mechanism. NCI SEER was actively involved with the development of a major update to the International Classification of Diseases for Oncology, the ICD-O-3 (7). Two SEER experts, Ms Constance Percy and Ms April Fritz, were among the editors of the ICD-O-3 manual and led the implementation of the new classification in the United States—implementation that was completed by SEER central registries in 2000. Changes to morphology coding were quickly followed by updates to the SEER Summary Stage classification (with the introduction of a new staging system in 2001, the SEER Summary Staging Manual 2000) (8) and the retirement of the Extent of Disease staging system in 2004 at the same time when the Collaborative Staging data collection system was established by a group of cancer surveillance partners (9). The quality assurance efforts to develop a reliable set of rules for determination of new primaries and for the accurate coding of rare and complex histologic types have always been a priority of the program. These efforts have materialized in the publication of the Multiple Primary and Histology Coding Rules manual, a set of rules adopted by all cancer registries, both central and hospital based, and by several other countries (10). At the same time, SEER worked in conjunction with the national partners to expand the collection of prognostic factors, adopting the categorizations of tumors by schemas that are set up at the national level by the North American Association of Central Cancer Registries (NAACCR) and the American Joint Committee on Cancer and then collecting and releasing more than 150 prognostic variables, often referred to as Site Specific Factors (or Site-Specific data items starting in 2018) (11).

From an operational perspective, the new millennium brought the development of the SEER Data Management System (SEER*DMS), a platform that allows cost-efficient and reliable registry operations such as visual editing, consolidation, linkages, and follow-back (12). As of 2023, all but 3 central registries have migrated to the SEER*DMS platform, which fully supports all central registry operations. It is anticipated that the remaining registries will complete the system migration in 2024.

This coordinated system of cancer registries represented urban and rural geographic areas as well as a range of education and poverty indicators throughout the covered populations (13). Registries in the SEER Program are engaged through contract mechanisms that enable a collaborative relationship with the federal government. SEER registries, as contractors, have been responsible for maintaining cancer information reporting systems; abstracting records for resident cancer patients (regardless if the diagnosis and treatment occur outside of the registry catchment area); abstracting all death certificates where cancer is listed as the cause of death; searching records of laboratories, nursing homes, and other health service units providing diagnostic services to ensure that the patient record is complete; registering all in situ and malignant neoplasms (with certain skin cancer exceptions); and recording data on newly diagnosed patients, including select patient demographics, primary site, morphology, extent of disease, and first course of cancer-directed therapy (14). The SEER standard for case ascertainment is 98%.

Collection of this data, particularly patient demographics, requires taking steps to ensure that patient confidentiality is protected while moving to automate some of the tasks conducted by SEER registries (9). The SEER Program has an information technology contractor supporting SEER Program efforts in these areas. In the late 1990s, the SEER Program developed and implemented the SEER*DMS. SEER*DMS is a database that meets federal information technology system requirements for maintaining personally identifiable information and is constructed to allow for the SEER registries to meet data submission requirements but utilize secure enclaves for each registry. Data access and data use through re-release processes are restricted to role-based access and signing an applicable data use agreement. This process is constructed such that federal government personnel cannot randomly access this data but must follow the same process as those outside the SEER Program. SEER*DMS allows for the use of standardized data files that meet contract requirements and registry community standards as established by the NAACCR. It enables the automation of tasks that had been previously identified as labor intensive or those with a high error rate. This results in the registry staff being able to focus on the completion of cancer case collections and to ensure that collection is of the highest quality.

The SEER Program receives data twice annually, and SEER registry submissions are assessed against numerous criteria, including completeness and quality. An underlying aspect of the SEER Program in managing the various sources and types of data is a robust quality improvement program that occurs with the direct involvement of the SEER registry community. The registries identify issues that need to be addressed as well as best practices to be shared across the SEER community.

The SEER Program also leverages several inter-agency agreements with other agencies or groups to support the SEER registries as cases are abstracted and reconciled. Use of data from the National Death Index, Centers for Medicare & Medicaid Services, and others supports these efforts and contributes to the completeness and quality of SEER data, and most of the agreements providing such data were established in 2011 with updates and renewals occurring to continue supporting the SEER registries. As treatment for cancer shifts from inpatient settings to home or other outpatient settings and as treatment modalities changed (rise in oral chemotherapeutic agents), coordination with other types of cancer data providers became essential. This shift has, in part, precipitated the need to further evolve the SEER Program. In recent years, the program collaborated directly with pharmacy, medical claims, electronic medical health records, and genomic data providers to further expand data sources collected by SEER registries (15). Furthermore, collaboration with the National Program of Cancer Registries, the American Cancer Society, and NAACCR resulted in several Annual Reports to the Nation on the Status of Cancer. These widely cited reports outline advancements in addressing cancer care needs and identify ongoing areas for improvement (16,17).

Recent advancements in the SEER Program (2017–2023)

Although these contributing factors have impacted the available datasets, another relevant aspect has been the ongoing expansion of the SEER Program through the addition of new registries. This is correlated with the release of new datasets beginning with SEER 8 in 1975 and includes SEER 12 in 1992, SEER 17 in 2000, and SEER 22 in 2000. The addition of new registries to the SEER Program has occurred through the federal government acquisition process and sought to increase the representation of the US population overall while increasing representation of key minority and underserved populations (18). The most recent expansion through this process occurred in 2018, which resulted in a reconfiguration of the SEER Program (see Figure 1). This reconfiguration created 2 types of SEER registries: core registries, which contribute data and expertise to the SEER Program, and research support registries. Research support registries enable the SEER Program to have an additional pool of registries available for contract awards as projects are developed. Such examples include the National Childhood Cancer Registry and the Virtual Pooled Registry. This enables the SEER Program to continue to leverage the expertise of these research support registries without the larger financial and personnel commitment that the core registries have to the SEER Program.

As of July 2021, when the final set of registries were brought into the SEER Program, there are 18 core registries (Figure 1) representing 48% of the US population, contributing approximately 850 000 new cancer cases annually. Moreover, the program supports 3 additional registries that collect cancer data for American Indian and Alaskan Native populations, depicted in the figure as Alaska Natives, Arizona Indians, and Cherokee Nation. These registries follow the same rigorous collection and reporting procedures as all core registries and adhere to Tribal Government policies regarding data release. Additionally, there are 10 research support registries, some of them participating in various projects of the NCI such as the National Childhood Cancer Registry (19). Adding registries to the core and research support categories, the SEER Program is in position to continue to move toward several of its revamped goals. These include, but are not limited to, collecting longitudinal data that can contribute to the clinical trials enterprise and inform newly diagnosed patients, collecting comprehensive genomic data by which to better characterize cancers, being able to collect and provide information on metastasis and recurrence, identifying comorbidities by which to better provide context for therapies and outcomes, and engaging in real-time case ascertainment to reduce the time in making this data available to the research community. This configuration of SEER Program registries and the varied array of data sources result in a national cancer population–based surveillance program that remains the gold standard and positions itself for another 50 years of contributions to addressing the cancer burden in the United States and internationally (20).

For the past 50 years, the SEER Program has pioneered methods for data collection and analysis of population-based surveillance data. Many of these methods have been adopted by national and international organizations as part of data standards, coding manuals, and classifications use for reporting cancer patterns and trends. In parallel, the program built an architecture of methods and tools for population-based cancer statistics, with SEER*Explorer as the most recent and most comprehensive online product for population-based cancer statistics. The future of the SEER Program will be closely aligned to the overall goals of the war on cancer as identified by the Cancer Moonshot (21). Among other goals, it will include collection and analysis of longitudinal treatment data coupled with a comprehensive genomic data characterization of cancers affecting the population in SEER jurisdictions. This new infrastructure would support researchers across the United States and internationally to identify best approaches to prevent and control the disease, improve the life of cancer survivors, and ultimately cut cancer mortality by more than half within 25 years.

Data availability

No new data were generated or analyzed for this commentary.

Author contributions

Steve Friedman, MHA (Project administration; Writing—original draft; Writing—review & editing) and Serban Negoita, MD, DrPH (Methodology; Supervision; Writing—original draft).

Funding

Funding provided by the National Cancer Institute.

Monograph sponsorship

This article appears as part of the monograph “50th Anniversary Issue of the National Cancer Institute’s SEER Program: A Half-Century of Turning Cancer Data into Discovery,” sponsored by the National Cancer Institute.

Conflicts of interest

SF declares no conflicts of interest. SN is an associated editor of JNCI. The author was not involved in the editorial review or decision to publish the manuscript. The author declares no other conflicts of interest.

References

Published by Oxford University Press 2024.

Issue Section:

MONOGRAPH

Email alerts

Citing articles via

More from Oxford Academic