ms_createdistbaselinetable.sas File Reference

This macro creates a baseline table, and an optional covariate profile table. More...

Detailed Description

This macro creates a baseline table, and an optional covariate profile table.

• Create baseline profile.
  
• Create baseline table.
  
  • Strip age group of special characters.
    
  • Square age for reporting.
    
  • Create age group dummies.
    
  • Create sex dummies.
    
  • Create race dummies.
    
  • Create hispanic dummies.
    
  • Create year dummies.
  • For type 4:
    
    • Create PrePostInd dummies.
      
    • Create preg_outcome dummies.
      
    • Get list of requested outcomes for squaring.
      
  • Create riskscore categories dummies.
    
    • Convert scores categories to cat1, cat2, ... catN.
      
  • Create lab covariates additional variables.
    
    • Because we do not know which covariates have additional lab variables, we need to get this information from the list of variables available in _RawData.
      
    • Sorting by name to make sure covar::lab_result variable is always before covar::lab_unit for the algorithm.
      
    • Loop through lab additional variables and get distinct values for each one.
      
      • Restrict to character variables. Numeric variables are linked to a character variable (i.e. covar::lab_unit1 is linked to covar::lab_result1).
        
      • If variable is character, loop through distinct values.
        
        • Remove forbidden characters from new variable names.
        • If the variable is linked to a numeric variable, add it to the list to compute mean and std.
  • Merge back dummies.
    
    • Create a new patient indicator to count distinct patients.
      
  • Process discrete and continuous variables and merge the two.
    
    • Square with missing variables for sex, race, hispanic, year and PrePostInd (for type 4).
      
  • Square riskscores categories (if applicable).
    
  • Add missing groups (if any).
    
  • Create switchstep indicator for when stacking datasets and remove dummy covariates (if any).
    

Program inputs

• dplocal.&RUNID._ads_&DATA._&I.(Dataset containing one record per GROUP, PATID, and INDEXDT.)
• infolder.&COHORTFILE. (Dataset used to define enrollment and demographic requirements, type of cohort identification strategy.)

Program outputs

• msoc.&RUNID._baseline&OUTCOHORT._&I. (Dataset including metrics for cohorts of interest during baseline.)
• msoc.&RUNID._profile_&I. (Dataset containing patients and episodes by their unique combination of covariates observed.)

Usage

%ms_createdistbaselinetable(WHERE=(&groupvar. in("&&prev&groupvar.s.") and previnc_def in("Both","Prev")),
                            T3OUT=);

Parameters

[in]	WHERE	Where clause to restrict DPLocal.&RUNID._ads_&data._&i. for computations.
[out]	T3OUT	For Type 3 analyses, it takes the value of _an and _an_censor for analysis and analysis_censor datasets. It is blank for all non-Type 3 analyses.

SAS Macros Dependencies

None.

Author

Sentinel Coordinating Center (info@.nosp@m.sent.nosp@m.inels.nosp@m.yste.nosp@m.m.org)