Adults with epilepsy are recruited into a clinical trial to compare the effect of two classes of anticonvulsant drugsenzyme-inducing and non-enzyme-inducing on serum lipids. The outcome of interest is a new diagnosis of hyperlipidemia. Use the output to answer the following questions.

1. 1. What procedure was used to generate the results in SAS?
2. 2. Is there sufficient evidence to suggest that drug class is associated with hyperlipidemia? Justify your response.
3. 3. Are any of the independent variables included in the model categorical? If so which ones? Justify your response.
4. 4. Are age and sex associated with hyperlipidemia? Justify your response.

Model Information Data Set Response Variable hyperlipidemia Number of Response Levels 2 Model binary logit Optimization Technique Fisher's scoring Number of Observations Read 150 Number of Observations Used 150 Response Profile Ordered Total Value hyperlipidemia Frequency 1 No 125 2 Yes 25 Class Level Information Class Value Design Variables sex Female 0 Male 1 1 drug Inducing Non-Inducing 0 Model Convergence Status Convergence criterion (GCONV=1E-8) satisfied. Model Fit Statistics Intercept Intercept and Only Covariates Criterion AIC 137.168 135.365 SC 140.179 147.408 -2 Log L 135.168 127.365 Testing Global Null Hypothesis: BETA=0 Test Chi-Square DF Pr > Chisq Likelihood Ratio 7.8030 0.0503 Score 7.5715 0.0557 Wald 7.1492 0.0673 Type 3 Analysis of Effects Wald Effect DF Chi-Square Pr > Chisq drug 4.1269 0.0422 age 2.6323 0.1047 1.0963 0.2951 1 sex 1 Analysis of Maximum Likelihood Estimates Parameter DF Estimate -3.7306 Intercept 1 Standard Wald Error Chi-Square Pr > Chisq 1.1866 9.8835 0.0017 0.4734 4.1269 0.0422 0.0252 2.6323 0.1047 drug Inducing 1 0.9617 age 0.0409 sex Male -0.4904 0.4684 1.0963 0.2951 Odds Ratio Estimates Point Estimate 2.616 Effect drug Inducing vs Non-Inducing age sex Male vs Female 95% Wald Confidence Limits 1.034 6.616 0.992 1.095 0.245 1.534 1.042 0.612 Association of Predicted Probabilities and Observed Responses Percent Concordant 68.3 Somers' D 0.373 Percent Discordant 31.0 Gamma Percent Tied 0.8 Tau-a 0.104 0.376 Pairs 3125 c 0.686 Model Information Data Set Response Variable hyperlipidemia Number of Response Levels 2 Model binary logit Optimization Technique Fisher's scoring Number of Observations Read 150 Number of Observations Used 150 Response Profile Ordered Total Value hyperlipidemia Frequency 1 No 125 2 Yes 25 Class Level Information Class Value Design Variables sex Female 0 Male 1 1 drug Inducing Non-Inducing 0 Model Convergence Status Convergence criterion (GCONV=1E-8) satisfied. Model Fit Statistics Intercept Intercept and Only Covariates Criterion AIC 137.168 135.365 SC 140.179 147.408 -2 Log L 135.168 127.365 Testing Global Null Hypothesis: BETA=0 Test Chi-Square DF Pr > Chisq Likelihood Ratio 7.8030 0.0503 Score 7.5715 0.0557 Wald 7.1492 0.0673 Type 3 Analysis of Effects Wald Effect DF Chi-Square Pr > Chisq drug 4.1269 0.0422 age 2.6323 0.1047 1.0963 0.2951 1 sex 1 Analysis of Maximum Likelihood Estimates Parameter DF Estimate -3.7306 Intercept 1 Standard Wald Error Chi-Square Pr > Chisq 1.1866 9.8835 0.0017 0.4734 4.1269 0.0422 0.0252 2.6323 0.1047 drug Inducing 1 0.9617 age 0.0409 sex Male -0.4904 0.4684 1.0963 0.2951 Odds Ratio Estimates Point Estimate 2.616 Effect drug Inducing vs Non-Inducing age sex Male vs Female 95% Wald Confidence Limits 1.034 6.616 0.992 1.095 0.245 1.534 1.042 0.612 Association of Predicted Probabilities and Observed Responses Percent Concordant 68.3 Somers' D 0.373 Percent Discordant 31.0 Gamma Percent Tied 0.8 Tau-a 0.104 0.376 Pairs 3125 c 0.686