Q3. The data for this question consists of a sample of 50 persons from the 1967-
ID: 1130807 • Letter: Q
Question
Q3. The data for this question consists of a sample of 50 persons from the 1967-1980 Evans County Study. There are two basic independent variables of interest: AGE and chronic disease status (CHR), where CHR is coded as 0 = none, 1 = chronic disease. A product term of the form AGE xCHR is also considered. The dependent variable is time until death, and the event is death. The primary question of interest concerns whether CHR, considered as the exposure variable, is related to survival time, controlling for AGE. The edited output of computer results for this question is given as follows: Model 1: Variable CHR Coef 0.8595 Chi-sq 7.61 Std. Err 0058 0.3116 Log likelihood =-142.87 Model 2: CHR AGE 0.3252 0.0193 Log likelihood- -132.45 6.13 19.63 0.8051 0.0856 0133 Model 3: CHR AGE CHRxAGE 0.0030 2.2556 0.0276 0.0345 Log likelihood132.35 0.20 10.01 0.01 6572 0016 9301 0.0874Explanation / Answer
d) This suggests that there is a difference in the slopes of the Hazard functions for people with and without the Chronic disease, as age varies. Hence, it is required to include the Interaction term CHR×Age in the regression. Thus, model 3 must be used.
e) Survival curve for people
With CHR = 1
0.8051 + 0.0856×Age
With CHR = 0
0.0856×Age
f) Overall, For people with Chronic disease, the average time until death is higher than those without a chronic disease, keeping everything else constant.
Related Questions
Navigate
Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.