fractionation: British Institute of Radiology Fractionation Studies
In DPpackage: Bayesian Nonparametric Modeling in R

Description Usage Format Details References Examples

The British Institute of Radiology (BIR) conducted two large-scale randomized clinical trials to assess the effectiveness of different radiotherapy treatment schedules for cancer of the larynx and pharynx. The cambined data come from 858 subjects with laryngeal squamous cell carcinomas and no involvement of asjacent organs. These data have been described and analyzed by Rezvani, Fowler, Hopewell, and Alcock (1993).

1	data(fractionation)

A data frame with 858 observations on the following 6 variables.

response: Three-year local control 1 (0 no control).
dose: Total dose (grays).
df: Total dose x dose/fraction.
time: Total time of treatment (days).
kt2: Tumor status (indicators for 2nd level of factor).
kt3: Tumor status (indicators for 3rd level of factor).

Three-year local control - meaning no detection of laryngeal carcinoma within three years after treatment - is the binary response, coded as 1 if local control is achieved and 0 otherwise. For this data set, three-year local control is achieved for 69 a total dose of radiation is administered in fractions over a treatment period. The dose per fraction df is measured in grays (Gy), the length of treatment period time is measured in days, and the number of fractions of the dose is nf. Tumors are classified by stage (i.e., the extent of invasion), into three groups. This categorical covariate is coded by two indicator variables kt2 and kt3, which are defined by kt2=1 (kt3=1) is the tumor is stage II (stage III) and zero otherwise.

Chappell, Nondahl and Fowler (1995) argued that the tumor stage, the total dose, the total time, and the interaction of the total dose per fraction are the relevant explanantory variables affecting probability of local control.

Chappell, R., Nondahl, D.M., and Fowler, J.F. (1995) Modelling Dose and Local Control in Radiotheraphy. Journal of the American Statistical Asso- ciation, 90: 829 - 838.

Newton, M.A., Czado, C., and Chappell, R. (1996) Bayesian inference for semiparametric binary regression. Journal of the American Statistical Association, 91, 142-153.

Rezvani, M., Fowler, J., Hopewell, J., and Alcock, C. (1993) Sensitivity of Human Squamous Cell Carcinoma of the Larynx to Fractionated Radiotherapy. British Journal of Radiology, 66: 245 - 255.