qrp_report

****************************************************************************************************

*                                           PROGRAM OVERVIEW

****************************************************************************************************

*

* PROGRAM: baseline_aggregate.sas  

* Created (mm/dd/yyyy): 10/20/2020

*

*--------------------------------------------------------------------------------------------------

* PURPOSE: The macro imports and concatenates both Level 1 and Level 2 baseline tables

*

*  Program inputs:                                                                                   

*   - Level 1 tables in the format: [runid]_baseline_[cohort]_[periodid].sas7bdat

*   - Level 2 tables in the format: [runid]_adjusted_baseline_[periodid].sas7bdat

* 

*  Program outputs:                                                                                                                                       

*   - Dataset where each row represents a baseline metric and each column represents a DP value.

*     Groups in the GROUPTABLE are stacked

* 

*  PARAMETERS:        

*   - dpsiteid: name of DP to import. Will be used as libname

    - dpnumber: number to assign to each DP after transpose

*   - level: 1 or 2

*   - grouptable: name of the dataset that contains groups, runIDs, and infomation pertaining to the 

*     baseline table name to include in the report.

*   - outdata: name of the aggregated dataset

*   - periodid: monitoring period #

*

*

*  Programming Notes:                                                                                

*   - This macro is designed to be called within a loop that loops through each data partner

*

*   - For level 1 requests, the baseline table file name will be dynamically derived using 

*     &dpsiteid as the libname and the file name as: [runid]_baseline_[cohort]_&periodid

*

*   - GROUPTABLE must contain the following variables: GROUP, RUNID, COHORT, and MERGEVAR

*     For level 2 requests, only GROUP and RUNID need to be populated. GROUP will represent the ANALYSISGRP.

*     For level 1 requests, COHORT is optional and will be used to correctly name the associated baseline 

*     table ([runid]_baseline_[cohort]_[periodid].sas7bdat), MERGEVAR is required and is the name of 

*     the key variable in the baseline table (group or analysisgrp)

*

*   - It is assumed the user intends to import and aggregate all groups in the GROUPTABLE input file.

*     The user should restrict this file to only groups with a baseline table prior to executing 

*     this macro.

*

*--------------------------------------------------------------------------------------------------

* CONTACT INFO: 

*  Sentinel Coordinating Center

*  info@sentinelsystem.org

*

***************************************************************************************************;

​

%macro baseline_aggregate(dpsiteid = ,

                          maskedid = ,

                          dpnumber = ,

                          level = ,

                          grouptable = ,

                          outdata = ,

                          periodid = );

​

    %put =====> MACRO CALLED: baseline_aggregate;

​

    /*********************************************************************************************/

    /* Level 1 baseline tables

    /*********************************************************************************************/

​

    %if %eval(&level.=1) %then %do;

​

        /*Use GROUPTABLE to create list of baseline tables and create dataset in statement*/

        data _temp_baseline_tablenames;

            set &GROUPTABLE.;

            format baselinetablename $40.;

            if missing(cohort) then do;

                baselinetablename = lowcase(cats("&dpsiteid..",runid, "_baseline_&periodid"));

            end;

            else do;

                baselinetablename = lowcase(cats("&dpsiteid..",runid, "_baseline_", cohort, "_&periodid"));

            end;

        run;

​

        proc sql noprint;

            select distinct baselinetablename into: baselinetables separated by ' '

            from _temp_baseline_tablenames;

            select count(distinct baselinetablename) into: num_unique_baseline_tables

            from _temp_baseline_tablenames;

        quit;

​

        %put Extracting baseline tables: &baselinetables.;

​

        /*Loop through each baseline table, import, stack groups*/

​

        %do b = 1 %to %eval(&num_unique_baseline_tables.);

            %let infile = %scan(&baselinetables., &b., ' ');