NCDB National Cancer Database Participant User Files - onetomapanalytics/Meta_Data GitHub Wiki

NCDB - National Cancer Database Participant User Files (PUF)

General description

  1. Database primary purpose - Review and advance the quality of care delivered to cancer patients through analyses of cases reported to the NCDB
  2. Overall data type - Health outcomes
  3. Dataset type - Longitudinal
  4. Data source - Registry
  5. Data level - Patient level
  6. Geographic location of the data collection sites - United States
  7. Sponsor, manager, or home institution - Commission on Cancer of the American College of Surgeons and the American Cancer Society
  8. Date range - 2004 - 2014
  9. Geolocation data - Facility location based on the U.S. Census division
  10. Dates - Year of diagnosis
  11. Hospital identifiers - Facility de-identified ID
  12. Longitudinal tracking - Track patients and facility through the de-identified case and facility PUF ID
  13. Clinical areas of interest - Cancer
  14. Number of records - Data is collected in more than 1,500 Commission on Cancer-accredited facilities (1), including 31 million records for patients diagnosed between 1985-2015 (2)
  15. Variables that are uniquely present in this dataset - Data regarding cases submitted to the Commission on Cancer’s (CoC) NCDB

Applicable methods

  1. Association methods, such as Cox proportional hazards regression models (3, 4, 5), logistic regression (6, 7, 8), multivariate analysis (9)
  2. Inferential tests (10)
  3. Time to event (11, 12)
  4. Difference-in-difference (13, 14)
  5. Propensity scores (15, 16)

High-impact designs

  • Describe the use of the NCDB to study cancer care (17)

  • Compare the number of incident cancer cases in the NCDB with other datasets, such as the United States Cancer Statistics data (2), SEER-Medicare (11, 18), and the Duke University registry (19)

  • Evaluate time to surgery and outcomes (11, 12)

  • Propose a revised pathologic staging classification and examine its prognostic value (10)

  • Compare results from observational cancer registry data with those of randomized clinical trials (3)

  • Evaluate overall survival outcomes following a surgical treatment (20, 21)

  • Compare differences in surgery pre- and post-Medicaid expansion (13)

  • Describe patterns of metastasis and treatment (22)

  • Determine the effect of hospital affiliation and volume on mortality (23, 24)

  • Identify the incidence, risk factors, and impact on survival associated with refusal of treatment (25)

  • Evaluate disparities on surgical treatment based on sociodemographic factor (26, 27, 28), geographic region (29), between Veteran and non-Veteran patients (30), race (31)

  • Compare distinct approaches in relation to outcomes and survival rates (32, 33)

  • Determine temporal trends in chemotherapy use (34)

  • Examine the impact of radiation, chemotherapy, and immunotherapy on outcomes and survival (19, 35, 36)

  • Quantify internally inconsistent and anomalous radiation therapy data and determine their association with overall survival 5

  • Determine whether circulating tumor cells status is predictive of radiotherapeutic benefit in early-stage cancer (37)

  • Assess the role of facility type on survival (38)

Data dictionary

To access the NCDB PUF data dictionary, click here

Variable categories

  1. Facility (e.g., type and location)
  2. Patient demographics (e.g., age, sex, race, primary payor, income, education, Charlson score)
  3. Cancer identification (e.g., primary site, laterality, histology, behavior, grade, size)
  4. Stage of disease (e.g., diagnostic and staging procedure, pathologic stage group, site-specific factor)
  5. Treatment (e.g., surgical procedure, approach, radiation, chemotherapy, palliative care)
  6. Outcomes (e.g., discharge, readmission, mortality, vital status)