Case Study Essay
Table 8.1 Clinical Trial Data for Problem 1
a. Perform two different tests (Breslow-Day, and likelihood ratio test) for whether the drug effect on curing an infection is the same over eight centers.
i) Breslow-Day test:
The B-D test of homogeneity tests for whether the drug effect on curing an infection is the same over eight centers gave a chi2 = 8.0 with df=7 and p-value = 0.333. Therefore since the p-value is > 0.05, we fail to reject the null that the drug effect on curing an infection is the same over the eight centers. The drug effect …show more content…
D= (-2lm) – (-2Ls)
= (-2x- 22.18071) –(-2*-22.18071)
Therefore, in testing the fit of model m, we fail to reject the null hypothesis that all the parameters that are in the saturated model but not in model M equals 0.
2. Table 8.2 shows results of a study about Y = whether a patient having surgery with general anesthesia experienced a sore throat on waking (1 = yes) as a function of D = duration of the surgery (in minutes) and T = type of device used to secure the airway (0 = laryngeal mask airway, 1 = tracheal tube).
Table 8.2 Data for Problem 2 on Sore Throat after Surgery
a. Fit a main effects model using these predictors. What is the predicted probability function? Interpret the signs of the parameter estimates.
Logit[P(Y=1)] =β0 + β1(D) +β2(T)
Logit[P(Y=1)] = -1.41734 + 0.06868 (D=1min) -1.65895(T=1) or
P(soreThroat/D,T)= Exp (-1.41734+0.06868D-1.65895T)[1+Exp(-1.41734+0.06868D-1.65895T)
The coefficient of duration of surgery is positive meaning the