This year marks the tenth anniversary of the establishment of the National Cancer Data Base (NCDB).1 The development of the NCDB began in 1988 when the American Cancer Society provided an initial four-year planning grant to the Commission on Cancer of the American College of Surgeons. The concept for the NCDB originated from the realization that the increasing use of computerized cancer registries in the 1980s would make it possible for data from local registries to be compiled to form a national data base. This provided an annual mechanism for the continuing assessment of cancer care at the local, state, and national levels.

“Clinical Highlights from the National Cancer Data Base, 1998,” which is featured in this issue of CA, reports on 689,714 cases of cancer from 1,227 hospitals that participated in the call for data in 1994.2 These cases represent 57% of the cancer cases expected in that year. In the first year of its existence, 1988, the NCDB received data from 501 hospitals including 232,577 cases, which represented 24% of the cancer cases diagnosed in the United States in 1985. It is anticipated that in the year 2000, more than 1 million cases will be accessioned into the NCDB annually, representing 80% of the cancer cases in the United States.

This remarkable growth is a result of the partnership forged between the American Cancer Society and the Commission on Cancer of the American College of Surgeons. This is yet another example of collaboration between these two institutions, which have been working together since 1913. The NCDB program is jointly governed by representatives from both organizations.

The major purposes of the NCDB at its founding were to (1) provide a scientific resource suitable for assessing patient care and outcome nationally and to disseminate this information to the medical community; (2) enhance ongoing cancer programs among hospitals approved by the Commission on Cancer and other cooperating hospitals by providing annual reports of national, regional, and hospital data; and (3) improve the process by which cancer patient care and research advances are translated into physician practice.3

Clearly, the NCDB has succeeded in fulfilling its stated purpose. Annual reports assessing cancer care at local and national levels have been provided to each participating institution, allowing each to compare its patterns of care and survival with national and state patterns. This process has led to enhancement and improvement of local cancer programs. Examples of the use of the NCDB to improve local care have been documented.4,5

More than 117 manuscripts from data generated by the NCDB have been published and presented in national and local meetings, leading to the effective transfer of advances in patient care into physician practice.

As pointed out in the “Highlights,” one strength of the NCDB is the large number of cases available for evaluation. Having such a large number allows for subset analysis of data and provides adequate numbers for analysis of rare cancers, such as those of the eye and nasopharynx.

The authors of “Highlights” also note an increase in breast-conservation surgery from 31% to 54% nationwide. One reason for this increase is undoubtedly the result of local programs comparing their patterns of breast-conservation surgery with national and regional data, thus increasing the appropriate utilization of conservative surgery in their institutions.

The NCDB is an important data base for pediatric cancers, comparing favorably with population-based survival evaluations in many childhood cancers. The authors of “Highlights” conclude that for reasons of representativeness and institutional coverage, the NCDB will be an important data base for pediatric cancers when national patterns are studied.

Also presented in this report is the discrepancy between patterns of care among African American and white populations in both ovarian and prostate cancer. Investigations into the reason for these differences remain a ripe source for further research by behavioral scientists and epidemiologists.

Of considerable interest also is the report comparing 1992 data from the Surveillance, Epidemiology and End Results (SEER) program on breast, colorectal, lung, and prostate cancer cases with NCDB data for the same year. The patient and disease treatment characteristics and patterns of care differed between the two data bases by only a few degrees.

One criticism of the NCDB has been that it is a convenience sample rather than a population-based sample. However, as the percentage of cancer cases accessioned into the NCDB increases, this difference should, and probably already is beginning to, become much less problematic. Certainly, the South is better represented by NCDB data because this area, which includes 16.8% of the United States population, is not represented in SEER at all (Figure in “Clinical Highlights from the National Cancer Data Base, 1998,” page 145).2

Lack of representation could have significant implications for populations such as rural African Americans, who live in large numbers in the South and are thus underrepresented as a special population in the SEER data. This problem has been noted in a previous publication.6

A valid criticism of the NCDB relates to quality control. The “Highlights” contains a report that continues the 10-year tradition of quality improvement of data entered into the NCDB. Opportunities for improvement, such as occurred with the development of new surgical codes (an endeavor that was completed this year), always exist and will be implemented by the NCDB as appropriate.

There is every reason to believe that the NCDB will continue to be an important, if not the most important, source in the United States for evaluation of patterns of cancer care. The American Cancer Society, through the development of its Infonet System, has shown a commitment to provide the public with information on patterns of cancer care on an institutional basis.

Much is left to be done before this information can be provided reliably, such as the development of severity indexes to be included in the data base. However, little doubt exists that the next millennium will see this type of information provided to purchasers and consumers of health care. The NCDB will play a significant role in this process as it continues to grow and continues its commitment to decreasing cancer-related morbidity and mortality.


  1. Top of page
  2. References
  • 1
    Menck HR, Garfinkel L, Dodd GD: Preliminary report of the National Cancer Data Base. CA Cancer J Clin 1991; 41: 718.
  • 2
    Menck HR, Bland KI, Eyre HJ, et al: Clinical highlights from the National Cancer Data Base, 1998. CA Cancer J Clin 1998; 48: 134145.
  • 3
    SteeleGDJr, WinchesterDP, MenckHR, et al (eds): National Cancer Data Base Annual Review of Patient Care 1993. Atlanta, American Cancer Society, 1993, pp 12.
  • 4
    PloetzBM, FremgenA, MenckHR (eds): Success Stories: Using Comparative Data to Improve Cancer Patient Care. Chicago, American College of Surgeons, 1995.
  • 5
    StrukusW, BlankenshipC, MenckHR (eds): Success Stories II: Cancer Liaison Physicians Using NCDB Data to Influence Patient Care. Chicago, American College of Surgeons, 1997.
  • 6
    Weiss HL, Soong SJ, Partridge EE, et al: Cancer incidence among predominantly black, rural-poor populations in southern states. South Med J 1997; 90: 986992.