Question

Suppose we conduct a study to determine the predictors of mortality among adults in an intensive care unit (ICU). We collect the vital status (STA) of 200 individuals; STA is coded as 0 if the individual “Lived” and 1 if the individual “Died”. Data on potential predictors were also collected including age (in years), presence of cancer (CAN, coded as 0=No and 1=Yes), CPR prior to ICU admission (CPR, coded as 0=No and 1=Yes), probable infection at ICU admission (INF, coded as 0=N0 and 1=Yes). We analyze the data using SAS's LOGISTIC procedure and observe the results shown in Table 2.

1) Identify the dependent variable and the independent variables in this study. Also, state the Omnibus Null and Alternative hypotheses.

Table 2: Output for Part Two The LOGISTIC Procedure Model Information Data Set WORK ICU Response Variable STA Number of Response Levels 2 Model binary logit Fished's scoring Testing Global Null Hypothesis: BETA-O Test Chi-Square DF Pr>ChiSq Likelihood Ratio 16.0796 4 0.0029 Score 15.9143 4 0.0031 Wald 14.0527 4 0.0071 Optimization Technique Number of Observations Read 200 Number of Observations Used 200 Response Profile Ordered Value STA Frequency Total Analysis of Maximum Likelihood Estimates Standard Wald Parameter DF Estimate Error Chi-Square Pr > ChiSq Intercept 1 3.3407 0.7470 20.0024 < 0001 AGE 1 00236 01104 5742 0.0325 CAN -02155 0.6075 0.1259 0.7227 CPR 1 -0.9591 0.5219 3.3767 0.0661 INF 1 -0.7803 0.3725 4.3889 0.0362 Probability modeled is STA- Model Convergence Status Convergence criterion (GCONV1E-8) satisfied Model F Statistics Intercept and Covariates Odds Ratio Estimates 95% Wald Effect Point Estimate Confidence Limits AGE 0.977 0.956 0.998 CAN 0.806 0.245 2.651 CPR 0.383 0.138 1.066 INF 0.458 0.221 0.951 Criterion Intercept Only 202161 205459 2 Log L 200 161 194081 210.573 184.081

Answer 1

Dependent variable = STA

it has two categories Yes or No

Independent variable = AGE , CAN , CPR INF

Hypothesis :-

H0 :- The mean (average value of the dependent variable) is the same for all groups.

v/s

H1 :- The mean (average value of the dependent variable) is not same for all groups.