Hi Everyone,   I was wondering if someone could help me remove/delete some observations on my dataset.    In my dataset, members can have multiple observations, and the number of observations vary.  I have this variable called "NEGHISDAYS", and it ranges between 1 to 365. I want to delete the member and all their observations from my dataset if any of their observations has a "NEGHISDAYS" between 1 & 105. The code I have right now only deletes the specific observations that falls between 1 & 105 but the other observations remain, but that is not what I want. I want the member and all their records removed.    Here's an example:   ID     NEGHISDAYS 1          . 1          25 1          107 1          .   The example has one member with four observations; one observation (observation 2) falls between 1&105. My current code removes observations 2 but keeps the other observations. However, since observation 2 falls between 1&105, I want the member and all their records (observations 1 -4) removed.     I am using SAS 9.2.                  ... View more

Hello Everyone!   I have a dataset with 6 years worth of patient healthcare claims. Patients appear multiple times within each year and also reappear in subsequent years. What I want to do is eliminate the duplicates within the year.    The code I used was:   proc sort data=dataset1 nodupkey out=dataset2; by patient_id; run;   While this code removes the duplicate patients within each year, it also removes the patients from subsequent years, which I do not want. In other words, patient 01 appears 10 times in 2009 3 times in 2010 7 times in 2011. The code I used removed 9 observations of patient 01 in 2009 but also completely removed all observations in the subsequent years which made it seem like patient 01 didn't have any health claims in 2010 or 2011.    Is there a way to remove duplicate patients within just one year without it impacting the other years? I do not want to split my dataset if I don't have to. Some additional information that might be helpful:   1) I use SAS 9.2 2) My years are in a variable (2009 through 2011) and are numeric 3) patient_id is a character variable       ... View more

Hi Everybody! I am fairly new to SAS and still have trouble with some things.    I have a dataset and want to get the distinct count of members based on two variables. Variable 1 (SDA) has 13 values and Variable 2 (FFYEAR) HAS 7 values. Instead of writing out the code 91 different times (13 X 7 = 91) for all the different combination of variables, is there a way I can write the code once but refer to all the different values in the variable?    I am using SAS 9.2 and here is the basic structure of my current code:   proc sql; select count (distinct member_id) as count from ELIGIBLE where sda='A' AND ffyear=2009; quit;    I want to avoid having to write the code over and over for each combination of variable values.  ... View more

Hi Everybody!   I recently came across acode online that works fantastically well in splitting a big dataset into smaller datasets. However, I was wondering if someone could help me modify it so the smaller datasets are more aligned with what I want.   Right now, the code (see below) splits the big dataset based on one variable (a character variable). What I want to do is split the big dataset by two different variables, the current variable in the code but also by a second variable (not currently in code).   For example, let's assume I have a dataset for the entire United States for six years (2009-2015). I want to split that dataset per state per year (Texas2012).   Here is the code I am using:    %LET TABLE=PQIEX.STELIGENROLL_KEYVAR; %LET COLUMN=SDA; PROC SQL NOPRINT; /* build a mini program for each value */ /* create a table with valid chars from data value */ SELECT DISTINCT CAT("SDA.PROFILE_",COMPRESS(&COLUMN.,,'KAD'),"(WHERE=(&COLUMN.=",QUOTE(&COLUMN.),"))") INTO :ALLSTEPS SEPARATED BY '' FROM &TABLE.; QUIT; %PUT NOTE:&ALLSTEPS.; DATA SDA.&ALLSTEPS.; SET &TABLE; RUN;   Some additional information that might be helpful:    1) SAS Version 9.2 2) Library name: PQIEX 3) Dataset: STELIGENROLL_KEYVAR 4) First Splitting Variable: SDA (character variable with 13 different values) 5) Second Splitting Variable (the one I am having trouble incorporating into the code): FFYEAR (numeric variable for the years 2009-2015)   I am fairly new with SAS and just have no clue how to modify this code in order to split the dataset by two variables. My mind starts getting confused on how to treat the numeric variable and how to incorporate it into the name of the new datasets. If y'all can please help me (and have patience with my ignorance), I would greatly appreciate it!   Muchas Gracias   ... View more

Hello Everyone,   I am trying to create a macro for this calculation I am doing for several datasets but just can't seem to get my code to work.  Here's some information you might need if you want to help me.   Library name: Diabetes Datasets = FFYEAR09 FFYEAR10 FFYEAR11 FFYEAR12 FFYEAR13 FFYEAR14 FFYEAR15   Here is the original code I want to use in a macro; it works when I do it individually for a dataset.   DATA DIABETES.FFYEAR09 (KEEP=RECIPIENT_NBR ALLMEDS TOTALQNT OCTMEDS OCTQNT); SET DIABETES.FFYEAR09; BY RECIPIENT_NBR NDC; RETAIN ALLMEDS TOTALQNT OCTMEDS OCTQNT; IF FIRST.RECIPIENT_NBR THEN DO; ALLMEDS = 0; TOTALQNT = 0; OCTMEDS = 0; OCTQNT = 0; END; IF FIRST.NDC THEN DO; ALLMEDS + 1; TOTALQNT + RX_QUANTITY; IF MONTH (RX_FILL_DT) = 10 THEN DO; OCTMEDS + 1; OCTQNT + RX_QUANTITY; END; END; IF LAST.RECIPIENT_NBR THEN OUTPUT; LABEL ALLMEDS = 'TOTAL NUMBER OF MEDICATIONS' TOTALQNT = 'TOTAL QUANTITY OF PILLS' OCTMEDS = 'TOTAL NUMBER OF MEDICATIONS IN OCTOBER' OCTQNT = 'TOTAL QUANTITY OF PILLS IN OCTOBER'; RUN;   I am fairly new to the world of SAS and this is my first attempt at a macro. This is the macro I developed based on some information I found online for macro beginners.   %MACRO TOTALMEDS; %DO I=1 %TO 7; DATA = DIABETES.FFYEAR&I (KEEP=RECIPIENT_NBR ALLMEDS TOTALQNT OCTMEDS OCTQNT); BY RECIPIENT_NBR NDC; RETAIN ALLMEDS TOTALQNT OCTMEDS OCTQNT; IF FIRST.RECIPIENT_NBR THEN DO; ALLMEDS = 0; TOTALQNT = 0; OCTMEDS = 0; OCTQNT = 0; END; IF FIRST.NDC THEN DO; ALLMEDS + 1; TOTALQNT + RX_QUANTITY; IF MONTH (RX_FILL_DT) = 10 THEN DO; OCTMEDS + 1; OCTQNT + RX_QUANTITY; END; END; IF LAST.RECIPIENT_NBR THEN OUTPUT; LABEL ALLMEDS = 'TOTAL NUMBER OF MEDICATIONS' TOTALQNT = 'TOTAL QUANTITY OF PILLS' OCTMEDS = 'TOTAL NUMBER OF MEDICATIONS IN OCTOBER' OCTQNT = 'TOTAL QUANTITY OF PILLS IN OCTOBER'; RUN; %MEND TOTALMEDS; MPRINT (TOTALMEDS): PROC PRINT DATA=DIABETES.FFYEAR&I; %TOTALMEDS   So something tells me I got the first statement after the "%Do I=1 …" wrong, and I am not sure where "MPRINT" is supposed to go for SAS to tell me exactly where I am messing up. Is there a more complicated macro I'm supposed to use?   If anyone can please shed some light on this, I would greatly appreciate your help!   I am using SAS 9.2     ... View more