- 8 10 Table 8 20 Refers To A Clinical Trial For The Treatment Of Small Cell Lung Cancer Patients Were Randomly Assigned 1 (280.08 KiB) Viewed 19 times
8.10 Table 8.20 refers to a clinical trial for the treatment of small-cell lung cancer. Patients were randomly assigned to two treatment groups. The sequential therapy administered the same combination of chemotherapeutic agents in each treatment
Response to Chemotherapy Gender Progressive Disease No Change Partial Remission Complete Remission Therapy Sequential Male Female 28 4 45 12 29 5 26 2 Alternating Male Female 41 12 44 7 20 3 20 1 source: W. Holtbrugge and M. Schumacher, Appl. Statist. 40: 249–259, 1991. cycle; the alternating therapy had three different combinations, alternating from cycle to cycle. a. Fit a cumulative logit model with main effects for therapy and gender. Interpret effect estimates. h. For the therapy effect, compare ß, to the estimate obtained when the model is fitted to the binary response obtained by combining the first two response Categories and combining the last two response categories. What property of the model does this reflect? For the collapsing in (b), compare B1/SE to the ratio obtained for the uncollapsed response, (Usually, a disadvantage of collapsing ordinal responses is that the significance of effects diminishes.) 4- Ft the model to the uncollapsed data that also contains an interaction term. Does it fit better? Explain--why it is equivalent to using the four Therapy combinations as levels of a single factor.