National Cancer Institute Home at the National Institutes of Health | www.cancer.gov

Methods Used in DevCan

Current Methods

  • Fay MP, Pfeiffer R, Cronin KA, Le C, Feuer EJ. Age-conditional probabilities of developing cancer. Stat Med 2003;22(11):1837-48. [Full TextExternal Web Site Policy]
  • Fay MP. Estimating age conditional probability of developing disease from surveillance data. Popul Health Metr 2004 Jul 27;2(1):6. [http://www.pophealthmetrics.com/content/2/1/6External Web Site Policy]
    This report corrects and updates Technical Report 2003-03 and replaces Technical Report 2003-03-A. [View Technical Reports]

Additional details:

  • To calculate the age-conditional probability of dying of a specific cancer we use standard competing risks methodology [View PDF].
  • To calculate the age-conditional probability of developing or dying from a specific cancer in the absence of other causes, we simply take the calculations in the presence of other causes and set the counts for other causes of death equal to zero. Probabilities which cover age periods through the end of one's life are undefined in this method [get more details].
  • When calculating the age conditional probability of developing a particular cancer, the same follow-back year is used for every person in the data set to prevent biases for certain sub-groups [get more details].

Previous Versions

Prior to DevCan version 5.0, the method to calculate the age-conditional probability of developing cancer used the method described in:

  • Feuer EJ, Wun L-M, Boring CC, Flanders WD, Timmel MJ, Tong T. Lifetime Risk of Developing Breast Cancer. J Natl Cancer Inst 1993;85:892-97. [View AbstractExternal Web Site Policy]
  • Wun L-M, Merrill RM, Feuer EJ. Estimating Lifetime and Age Conditional Probabilites of Developing Cancer. Lifetime Data Analysis 1998;4:169-86. [View AbstractExternal Web Site Policy]

Comparison of Old and New Methods

In general, the old and new methods give similar results. For example, in estimating lifetime risk of developing specific cancers for different sex and race combinations, the two methods generally agree to within about 2 percent. For details see:

  • Fay MP, Pfeiffer R, Cronin KA, Le C, Feuer EJ. Comparison of Two Methods for Calculating Age-Conditional Probabilities of Developing Cancer. Statistical Research and Applications Branch, NCI, Techical Report # 2002-01. [View PDF]
  • Fay MP. Estimating age conditional probability of developing disease from surveillance data. Popul Health Metr 2004 Jul 27;2(1):6. [http://www.pophealthmetrics.com/content/2/1/6External Web Site Policy]