ms_cidatables.sas File Reference

This macro creates the output tables for type 1 and type 2 cohorts. More...

Detailed Description

This macro creates the output tables for type 1 and type 2 cohorts.

• Create separate prevalent and incident datasets.
  
  • Select all output levels relevant to t1_cida/t2_cida/t4cida.
    
  • Loop through groups.
    • To prevent unnecessary looping, only loop if AgeGroup differs between groups.
    • If prevalent datasets are requested only perform analysis for prevalent groups.
      
    • If all groups are running, grouplist is blank. Otherwise only aggregate data for current group and optional nvrexp group.
    • For type 4, &CIDATABLESGROUP. is associated with base pregnant cohort groups.
      • Create &GROULIST. with MIL groups associated with the base pregnant cohort groups to output MIL groups.
        
      • Select the MIL groups associated with the groupname to remove groups that do not need to be looped.
        
    • Compute numerators.
      
      • Square stratifications variables and assign the associated MIL groups.
      • Merge squared values with _mstr.
        
      • Summarize data.
        
    • The table to be generated in msoc will be the result of a merge between numerators and denominators.
      
      • Initialize missing variables to prevent merge warnings.
    • Combine with denominators.
      • Does not apply to type 4.
        
    • List groups without denominators.
    • Merge common level values between numerators and denominators, then merge in numerators only data.
      
    • Do not report denonminators if running surveillance mode or cohort defined by age anniversary or calendar date or if the group is unexposed.
      
    • Do not report followuptime, eps_wEvents, all_events when conducting signal detection analysis.
      
    • List covariates and put into macro variables.
    • Output cida table to dplocal and msoc.
      • The output tables will be stratified by the levels provided in the USERSTRATA inputfile.

Program inputs

• dplocal.&RUNID._denomcounts (Dataset (type 1 and type 2) with eligible members and member-days metrics for the T1_CIDA and T2_CIDA tables.)
• dplocal.&RUNID._mstr (Dataset for type 1 and type 2 analyses with one record per individual per index date for every GROUP.)
• dplocal.&RUNID._mstr_mi (Dataset for type 4 analyses produced for exposed and comparator/unexposed pregnant cohorts.)
• infolder.&&TYPE&TYPE.file (Dataset required for the T# cohort identification strategy.)
• infolder.&COHORTFILE. (Dataset used to define enrollment and demographic requirements, type of cohort identification strategy.)
• infolder.&USERSTRATA. (Dataset with ouput tables and stratifications specification.)

Program outputs

• dplocal.&RUNID._numcounts (Dataset for type 1 and type 2 analyses with cohort metrics for the T::_CIDA table.)

• msoc.&RUNID._t&TYPE._cida&DS_SUFFIX. (Dataset where &TYPE. is the type of analysis (1, 2, 4) and &DS_SUFFIX. is the parameter listed below.)

  ms_cidatables(INPUT_DS=_mstr_mi, DS_SUFFIX=, WHERE_CLAUSE=, FOLLOWUPTIMEVAR=followuptime);

Parameters

[in]	INPUT_DS	Name of the dplocal input dataset used to generate the output.
[in]	DS_SUFFIX	Suffix to add to output dataset - either will be missing or set to _prev for prevalent tables.
[in]	WHERE_CLAUSE	Where clause to restrict &runid._mstr for computations.
[in]	FOLLOWUPTIMEVAR	Followuptime variable - will only exist for type 2.

SAS Macros Dependencies

• ms_squaredtableshell.sas

Author

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