Final Assignment due: December 16, 1998



Yüklə 23.96 Kb.
tarix14.04.2016
ölçüsü23.96 Kb.
EPI 826

Fall, 1998


Final Assignment

DUE: December 16, 1998


Consider the enclosed data for 137 patients from a lung cancer trial. The variables in this data set are listed as follows:


Variable #

Variable Name

Code

Coding

1

Treatment

treat

standard=1, test=2

2

Large Cell

cell1

large=1, other=0

3

Adeno Cell

cell2

adeno=1, other=0

4

Small Cell

cell3

small=1, other=0

5

Squamous Cell

cell4

squamous=1, other=0

6

Survival Time

survtime

(days) integers

7

Performance Status

perform

0=worst, …, 100=best

8

Disease Duration

sicktime

(months) integers

9

Age

age

(years) integers

10

Prior Therapy

priotx

none=0, some=10

11

Status

cstat

0=censored, 1=died













  1. Run all univariate analyses for assessing the influence of each covariate on survival (survtime). You would use several invocations of proc phreg for this purpose.




  1. Summarize your results in the form of a table provided below.



Table 1: Single covariate regression models


COVARIATE

-2log L

CHI-SQUARE (p-value)

Hazard ratio

95% CI

none

1011.768










treat













cell1













cell2













cell3













cell4













perform













sicktime













age













priortx
















  1. Follow the model building strategy that we discussed in class. Table 1 was your first step (STEP 1) towards building a multivariable model. Notice that the cell type has four indicator variables. Use the cell1 (large cell) as your reference category in the multivariate model.




  1. Attach the (one page) printout of the multivariable model that you considered following STEP 1. Call this model-1.




  1. Now go to STEP 3, removing all variables that are not significant in the multivariable model. Use the chunk test to check your hypothesis. Provide the one page printout of this model. Call this model-2.




  1. Your final step, is to consider adding (one at a time) to model-2 each covariate not considered in STEP 1. This would involve a series of tests, which you should summarize into the following table.



Table 2: Mulitvariable models with variables added to model-2


Added Variable

-2log L

Chi-square (p-value)

Base model-2







+

+

+



+

+







Does this change your model-2?


Based on your analysis and your final model, answer the following questions:


  1. What is the hazard ratio that compares persons with adeno cell type with persons with large cell type? Explain your answer using the general hazard ratio formula for the Cox PH model.




  1. What is the hazard ratio that compares persons with adeno cell type with persons with squamous cell type? Explain your answer using the general hazard ratio formula for the Cox PH model.




  1. Is there an effect of treatment on survival time? Explain briefly.

Kataloq: course
course -> Your Questions and Your Answers Language Practices and Policies, 2/21/08 How do you balance the need to cover the curriculum with the need to meet el language needs?
course -> Implementing the Microsoft. Net framework with C# Introduction
course -> Pspice statements and dot commands used in this circuit
course -> 13. 7 (a) C1 is a coupling capacitor that couples the ac component of vS into the amplifier. C2 is a bypass capacitor. C3 is a coupling capacitor that couples the ac component of the signal at the drain to the output vO
course -> Examples of search strategy Topic: Government policies in Central Europe regarding the destruction and degradation of wetlands
course -> Chapter 5 Question 35
course -> Notes for avl adelson-Velski Landis
course -> Fundamentals of Computer Science Spring 2009 Burg Makefiles What is the make utility? make

Yüklə 23.96 Kb.

Dostları ilə paylaş:




Verilənlər bazası müəlliflik hüququ ilə müdafiə olunur ©azrefs.org 2020
rəhbərliyinə müraciət

gir | qeydiyyatdan keç
    Ana səhifə


yükləyin