I have a dataset labelled as Final1 of patients with 84 records which has INST_NO i.e. the institution where they received some care I want to compare the above dataset with another dataset (Inst_dsc) in which I have a list of Institution Nos labelled as INST_AT How would i write a code to check if the INST_No in the 1st dataset is present in the 2nd dataset  Basically want to flag the cases or records      data match (Keep=INST_No INST_AT); 1227  length inst_AT 8 ; 1228  if _N_ then do; 1229  declare hash h (dataset:"work.inst_dsc"); 1230  h.definekey ('INST_AT'); 1231  call missing (INST_AT); 1232  end; 1233  set Final1; 1234  rc=h.find(INST_AT); 1235  if rc=0 then output; 1236  run;   ERROR: Uninitialized keys for hash object at line 1234 column 4. ERROR: DATA STEP Component Object failure.  Aborted during the EXECUTION phase. NOTE: The SAS System stopped processing this step because of errors. NOTE: There were 1 observations read from the data set Final1. WARNING: The data set WORK.MATCH may be incomplete.  When this step was stopped there were 0 observations and 2 variables. WARNING: Data set WORK.MATCH was not replaced because this step was stopped. NOTE: DATA statement used (Total process time):       real time           0.01 seconds       cpu time            0.01 seconds     1237  proc print ; 1238  run; ... View more

proc sql; create table deno_2 as select HCNE, CIHI_key, inst_no, Fac_no, Facility_LHIN, patient_LHIN, POSTAL_CODE,Fiscal_QTR,a.ADM_DT,a.ADM_TM from DAD_D_4 as a union all select HCNE, CIHI_key, inst_no, Facility_no, Facility_LHIN, patient_LHIN, POSTAL_CODE, Fiscal_QTR,b.REG_DT,b.REG_TM from NAC_D_8 as b ; quit;   How would I combine the two datasets by including the last two variables from the 2 datasets which are different  ... View more

How would i modify the code below to capture the direct admisions i.e. FROM deno_2_EVTFacilities the patients that occur in this database i want to check if these patients appeared in any hospital (in either  cihi.dad1819q4 or cihi.nacrs1819q4) before the admission to the EVT hospital in EVT_fy2018q4  So if they did not appear in any hospital then it would be classified as a direct admission otherwise  transfer Any help in modifying the code below would be appreciated     PROC SQL; CREATE TABLE SelfDirectEVT AS SELECT DISTINCT A.*, N.REG_DT AS TRANS_REG_DT, N.REG_TM AS TRANS_REG_TM, D.ADM_DT AS TRANS_ADM_DT, D.ADM_TM AS TRANS_ADM_TM, COALESCE(N.INST_FROM,D.INST_FROM_NO) AS TRANS_INST_FROM, D.INST_NO AS TRANS_INST_DAD, N.INST_NO AS TRANS_INST_NACRS, D.CIHI_KEY AS TRANS_KEY_DAD, N.CIHI_KEY AS TRANS_KEY_NACRS, E.INST_NAME AS TRANS_INST_NAME, E.INST_ABB AS TRANS_INST_ABB, E.FAC_NUM_NEW AS TRANS_FAC, CASE WHEN MISSING(D.CIHI_KEY)=0 THEN 'DAD' WHEN MISSING(N.CIHI_KEY)=0 THEN 'NACRS' ELSE '' END AS TRANS_TYPE, MISSING(D.CIHI_KEY)=0 AS DAD_TRANSFER, MISSING(N.CIHI_KEY)=0 AS NACRS_TRANSFER, MISSING(D.CIHI_KEY)=0 OR MISSING(N.CIHI_KEY)=0 AS ANY_TRANSFER, C.EVT_DT, C.EVT_TM, MISSING(C.EVT_DT)=0 AS INDICATOR3 'Ischemic stroke case transferred from tPA site received EVT' FROM deno_2_EVTFacilities AS A LEFT JOIN cihi.dad1819q4(WHERE=(INST_NO IN (&EVT_INST_AT.))) AS D ON A.HCNE=D.HCNE AND A.HCNE~='9999999999' AND (DATEPART(A.REG_DT)<=DATEPART(D.ADM_DT)<=DATEPART(A.REG_DT)+1) LEFT JOIN cihi.nacrs1819q4(WHERE=(INST_NO IN (&EVT_INST_AM.))) AS N ON A.HCNE=N.HCNE AND A.HCNE~='9999999999' AND A.CIHI_KEY~=N.CIHI_KEY AND (0 <= INTCK('HOUR', A.REG_TM, N.REG_TM) <= 24) LEFT JOIN EVT_fy2018q4 AS C ON D.CIHI_KEY=C.CIHI_KEY LEFT JOIN INST_EVT AS E ON (D.INST_NO=STRIP(PUT(E.INST_AT,BEST32.)) OR N.INST_NO=STRIP(PUT(E.INST_AM,BEST32.))) ORDER BY TRANS_FAC, A.CIHI_KEY; QUIT; ... View more