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

Age-Conditional Probability of Developing or Dying of Cancer in the Absence of Other Causes

DevCan can calculate the age-conditional probability of developing or dying from a specific cancer either in the presence of other causes of death (sometimes called the "crude" formulation) or in the absence of other types of death (sometimes called the "net" formulation). Although the calculations assuming the presence of other causes of death more closely relate to the true state of nature, the calculations assuming the absence of other types of death can be useful for some comparisons. Because the probabilities calculated in the absence of other causes of death assume no deaths from other causes, these probabilities are better as a disease progress measure used for comparing populations with very different rates of other cause mortality. The probability of developing a specific cancer in the absence of other causes of death from birth to a specific age is similar to the cumulative rate (an approximation to the cumulative risk) that has been used to compare different countries with very different rates of other cause mortality (see Day, N.E., 1992 "Cumulative Rate and Cumulative Risk", Chapter 9 of Cancer Incidence in Five Continents, Volume VI, Parkin D.M. et al (editors), International Agency for Research on Cancer Scientific Publication No. 120.) DevCan's age-conditional probability of developing a specific cancer in the absence of other causes has several advantages over the cumulative rate:

  • it is more general in that we can condition on being cancer free up until a specific age
  • it uses only the first incidence of a cancer (see description of follow-back year),
  • the rates of death from cancer are incorporated into its calculations (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]),
  • the rates of first cancer incidence and mortality follow a more plausible model than constant rates over 5 year age groups (see Fay MP. Estimating age-conditional probability of developing cancer using a piecewise mid-age joinpoint model for the rates. Statistical Research and Applications Branch, NCI, Techical Report # 2003-03-A. [View PDF])