Safari Books Online is a digital library providing on-demand subscription access to thousands of learning resources.
Various versions of data from the Stanford Heart Transplant study have been reported in a number of publications. The data set used in Chapter 5, "Estimating Cox Regression Models with PROC PHREG," appeared in Crowley and Hu (1977). The sample consisted of 103 cardiac patients who were enrolled in the transplantation program between 1967 and 1974. After enrollment, patients waited varying lengths of time until a suitable donor heart was found. Patients were followed until death or until the termination date of April 1, 1974. Of the 69 transplant recipients, only 24 were still alive at termination. At the time of transplantation, all but four of the patients were tissue typed to determine the degree of similarity with the donor. The data set contains the following variables:
DOB
is the date of birth.
DOA
is the date of acceptance into the program.
DOT
is the date of transplant.
DLS
is the date last seen (death date or censoring date).
DEAD
has a value of 1 if the patient is dead at DLS; otherwise, DEAD has a value of 0.
SURG
has a value of 1 if the patient had open-heart surgery before DOA; otherwise, SURG has a value of 0.
M1
is the number of donor alleles with no match in the recipient (1 through 4).
M2
has a value of 1 if the donor and recipient mismatch on the HLA-A2 antigen; otherwise, M2 has a value of 0.
M3
is the mismatch score.