Business Context
A pharmaceutical company is conducting a clinical trial to evaluate the efficacy of a new drug compared to a placebo. The goal is to analyze the survival time of patients and determine if the treatment significantly reduces the hazard of the event (death), adjusting for age and gender.
Data Preparation
Creation of a simulated clinical trial dataset with 100 patients, including treatment group, age, sex, survival time, and censoring status.
Copied!
| 1 | |
| 2 | DATA mycas.clinical_trial; |
| 3 | call streaminit(123); |
| 4 | DO i = 1 to 100; |
| 5 | Treatment = (rand('uniform') > 0.5); |
| 6 | Age = 50 + floor(rand('normal') * 10); |
| 7 | Sex = ifc(rand('uniform') > 0.5, 'M', 'F'); |
| 8 | LinearPred = 0.5 * Treatment + 0.02 * Age; |
| 9 | Time = rand('exponential') / exp(LinearPred); |
| 10 | IF Time > 20 THEN DO; |
| 11 | Time = 20; |
| 12 | STATUS = 0; |
| 13 | END; |
| 14 | ELSE STATUS = 1; |
| 15 | OUTPUT; |
| 16 | END; |
| 17 | |
| 18 | RUN; |
| 19 |
Étapes de réalisation
1
Fit a basic Cox proportional hazards model.
Copied!
| 1 | |
| 2 | PROC CAS; |
| 3 | phreg.cox TABLE={name='clinical_trial'}, class={{vars={'Sex'}}}, model={depVars={{name='Time', event='Status(1)'}}, effects={{vars={'Treatment', 'Age', 'Sex'}}}}; |
| 4 | |
| 5 | RUN; |
| 6 |
2
Request Hazard Ratios to interpret the treatment effect.
Copied!
| 1 | |
| 2 | PROC CAS; |
| 3 | phreg.cox TABLE={name='clinical_trial'}, class={{vars={'Sex'}}}, model={depVars={{name='Time', event='Status(1)'}}, effects={{vars={'Treatment', 'Age', 'Sex'}}}}, outputTables={names={'ParameterEstimates'}}; |
| 4 | |
| 5 | RUN; |
| 6 |
Expected Result
The action should successfully fit the model and return the 'ParameterEstimates' table. This table will show the coefficients and hazard ratios for Treatment, Age, and Sex, allowing the researchers to see if the Treatment effect is statistically significant (p-value < 0.05).