Biostatistical analysis of mortality data for cohorts of cancer patients.

The Hardin Jones principle states that for a homogeneous cohort of cancer patients the logarithm of the fraction surviving at time t has a constant slope. With use of this principle, the survival times of the members of a heterogeneous cohort can be analyzed to divide the cohort into subcohorts with different mortality rate constants. Probable values of the additional survival time can be estimated for members surviving at the closing date of a clinical trial, permitting them to be included in the biostatistical analysis of the results of the trial in a more significant way than through Kaplan-Meier renormalization.