whether the patient with NHL was neutropenic, whether proper components of the physical examination were performed, and whether chemotherapy was administered appropriately. Third, patient surveys can provide additional useful information. Patients can report on what happened during a clinical encounter and thereby provide information relevant to the processes of care. They can also rate their satisfaction with care and provide information on outcomes such as functional status. It is generally more expensive and time-consuming to collect information from medical records and from patients than from administrative data.

Cancer registries are also a potential source of information. They collect information on type of cancer, histology, stage at diagnosis, patient age, and initial course of treatment (whether the patient received surgery, chemotherapy, and radiation therapy that would normally be prescribed as part of the initial treatment plan). Registries exist at the regional, state, national, and international levels.

There are two main national registries: the Surveillance, Epidemiology, and End Results Program of the National Cancer Institute (NCI) and the National Cancer Data Base (NCDB) (Swan et al., 1998). The SEER program was established as a result of the National Cancer Act of 1971 to assemble, analyze, and distribute information on the prevention, diagnosis, and treatment of cancer. Cancer is the only chronic disease (aside from HIV/AIDS) for which a national surveillance program exists. The program routinely collects information from designated population-based cancer registries in different parts of the country. The different areas have been chosen for their capacity to maintain a cancer reporting system as well as for their ability to report epidemiologically significant population subgroups. Currently, 14 percent of the U.S. population is represented by the nine geographic areas that make up the SEER program's database. Goals of the SEER program include the following:

  • compiling (with the help of the National Center for Health Statistics) estimates of cancer incidence and mortality in the United States;
  • discovering trends and unusual changes in specific cancers based on their geographic, demographic, and social characteristics;
  • providing information about trends in therapy, changes in the extent of disease (stage at diagnosis), and changes in patient survival; and
  • promoting studies that identify the factors that can be controlled through intervention strategies.

Health service researchers have linked SEER to Medicare administrative files to evaluate patterns of care, the use of health services, and the costs of treatment (Potosky et al., 1993; Edwards, personal communication to Maria Hewitt, November 1998). Many locations outside of the SEER program's area maintain cancer registries. The National Program of Cancer Registries of the Centers for Disease Control and Prevention (CDC) is bolstering states' capabilities to monitor cancer trends (CDC, 1998).

The National Cancer Data Base is a joint project of the Commission on Cancer (COC) of the American College of Surgeons (ACoS) and the American Cancer Society (ACS) to facilitate community, hospital, state, and national assessment of care of patients with cancer (Menck et al., 1997). It began in 1989, and 1,600 hospitals currently report data on 600,000 new cases annually to the NCDB (an estimated 58 percent of new cancer cases) (NCDB, 1998; Swan et al., 1998). NCDB collects information on patient characteristics, tumor characteristics, first course of treat-



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement