History: The National Cancer Act of 1971 mandated the collection, analysis, and dissemination of all data useful in the prevention, diagnosis, and treatment of cancer. The Act resulted in the establishment of the National Cancer Program under which the Surveillance, Epidemiology, and End Results (SEER) Program was developed. The SEER Program collects cancer data on a routine basis from designated population-based cancer registries in various areas of the country. A sequel to two earlier NCI programs -- The End Results Program and the Third National Cancer Survey -- the SEER Program was initiated in several geographic areas of the United States and its territories with case ascertainment beginning with January 1, 1973 diagnoses.
The initial SEER reporting areas were the states of Connecticut, Iowa, New Mexico, Utah and Hawaii and the metropolitan areas of Detroit, Michigan, San Francisco-Oakland, California and the Commonwealth of Puerto Rico. In 1974-1975, the SEER Program was expanded to include the metropolitan area of New Orleans, Louisiana, the thirteen-county Seattle-Puget Sound area in the state of Washington and the metropolitan area of Atlanta, Georgia. New Orleans left the program at the end of the 1977 data collection year. In 1978 ten predominantly black rural counties in Georgia were added. In 1980 American Indians residing in Arizona were added. In 1983, four counties of New Jersey were added with coverage retrospective to 1979. New Jersey and Puerto Rico participated in the program until the end of the 1989 reporting year. In 1992 the SEER Program expanded in order to increase coverage of minority populations, especially Hispanics.
The two new areas were in California: Los Angeles County, and four counties in the San Jose-Monterey area south of San Francisco. These expansion areas were selected primarily for their ability to operate and maintain a population-based cancer reporting system, and for their epidemiologically significant population subgroups. With respect to selected demographic and epidemiologic factors, these selected areas provide a reasonably representative subset of the United States population.
Data Collection: The NMTR maintains a cancer information reporting system for the entire resident population of New Mexico and the American Indian population residing in Arizona (the latter in collaboration with the Arizona Cancer Registry). NMTR circuit-riding abstractors visit New Mexico facilities, as well as Arizona Indian Health Service (IHS) facilities, on a schedule that reflects the caseload of each facility. All standard sources for case identification are reviewed to ensure complete coverage. In most facilities, pathology reports, patients logs, disease indices and outpatient records are reviewed.
Abstracting: The NMTR uses PC DaSH (Patient Cancer Data System for Hospitals), a PC-based software program that enables cancer information to be entered electronically at the data source, and includes all requisite SEER data items describing patient demographics and tumor characteristics, methods of diagnosis, first course of treatment, and follow-up information. NMTR personnel and hospital registrars abstract all cancer cases on PC DaSH using desktop and laptop computers. All incoming PC DaSH abstracts are uploaded to the NMTR database via direct connection, encrypted electronic file transfer, or encrypted e-mail. Currently, seven New Mexico hospitals are ACoS-accredited and maintain hospital-based tumor registries. These seven hospitals submit approximately 40% of the abstracts obtained annually by the NMTR. Circuit-riding abstractors from the NMTR visit the remaining New Mexico and Arizona IHS hospitals and facilities, which submit the remaining 60%.
Coding: Coding schemes for many data items including place of birth, race, ethnicity, and site-specific surgery, are maintained in PC DaSH; however, abstractors also use the SEER Program manuals to determine reportability, and for coding EOD. PC DaSH is updated as SEER codes are modified, providing greater accuracy in abstracting and coding. American Indian tribal information is included in the patient chart at IHS/PHS facilities and is collected during abstraction. NMTR editing staff code this information once the case has been transmitted. Death information is also coded by NMTR editing and follow-up staff following initial abstraction. Addresses that cannot be auto-coded are researched and are manually coded by the editing department. Census tract and county of residence are auto-coded at the NMTR during abstract processing. Additional data items required by the American College of Surgeons (ACoS) are collected by hospital tumor registrars and maintained in each hospital’s computerized database. When NMTR abstractors assist ACoS facilities, they abstract and code ACoS items using each facility’s guidelines. Data are entered according to protocols prescribed by SEER and the North American Association of Central Cancer Registries (NAACCR) data standards committee.
Follow-up: The NMTR uses a variety of methods to collect vital status follow-up information for cases not known to be deceased. The primary follow-up method involves linking the Registry case file to a death certificate file provided monthly by the New Mexico Department of Health - Bureau of Vital Records and Health Statistics (BVRHS). Other passive methods of monitoring vital status include matching Social Security numbers (SSN) with Center for Medicare/Medicaid Services (CMS) records to identify deceased patients. All patients not know to be deceased are submitted to the CMS to determine if benefits are being collected. CMS then notifies the NMTR if the patient’s SSN is still active or if the patient is deceased. Additional procedures involve matching with the Indian Health Service (which is explained in more detail in the next paragraph), Social Security Administration Death Master File, the New Mexico Motor Vehicle Department, Hospital Disease Indices, lists from hospital billing departments and other files, and periodic searches using the National Death Index and commercial programs such as Accurint.
In addition to these follow-up processes, the NMTR sends computer-generated letters to the physician associated with a cancer case or directly to the patient (if no physician is listed or if the physician requests that all inquiries be sent directly to the patient). Another line of follow-up is to ask hospital registrars and abstractors to review the patient’s medical record for the most recent date that the patient was known to be alive. Active follow-up, including letters or hospital registrar/abstractor review, is not conducted for in situ cervical cancer cases or benign brain cases. Procedures for following American Indians of Arizona and New Mexico are different because of our policy of not contacting American Indian patients directly. Instead, medical records are reviewed for evidence of subsequent visits to medical clinics. We also run a computer match program twice a year. The NMTR data set is then matched against billing disease index codes from IHS records. The proportion of American Indian cases lost to follow-up is slightly higher than that for other ethnic groups, but has recently improved using these passive methods. Routine linkages with records from the Arizona Cancer Registry and Arizona vital records (i.e., death certificates) improves the quality of NMTR data for American Indians in Arizona.
We also use a tracking system, CaseID, which assigns a unique identifier to each abstract, and is then combined with the hospital number to create a unique key or CaseID. For hospitals that have a tumor registrar, each CaseID is based on the hospital’s master file. For hospitals that do not have an onsite tumor registrar, a program that uploads the cases to the NMTR’s databased assigns the CaseID. The CaseID is used for all matching between the hospital database and the Cancer Information Management System (CIMS) database, updating cases, and case matching with CIMS to check for missing cases.
