cancel
Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for 
Search instead for 
Did you mean: 
Banke
‎08-30-2025
  • 120 Posts
  • 47 Likes Given
  • 1 Solutions
  • 2 Likes Received

IDENTIFYING TREATED AND UNTREATED PATIENTS

by Pyrite | Level 9 in New SAS User
‎09-18-2022 05:13 PM
‎09-18-2022 05:13 PM
Hi all, i have two datasets:  1. dataset1 = disease. it contains patients with disease (dx = 1) and no disease (dx = 0) 2. dataset 2 = medication. it contains patients taking medication for the disease (t2D = 1) and those not taking medications(t2D = 0)   i  need to get 1. patient with disease and taking med 2. patient with disease not taking med 3. patients without disease taking med my code is too long,beacuse i had to create new datasets for each category of patients and used inner join.  I /***QUESTION 1**/ /**TAKES EVERY PATIENT WITH THE MEDICATION THAT HAVE DIABETES**/ DATA DXMED; /*1,759,104*/ MERGE T2DMED2 (IN = A) /*1,759,104*/ T2DM2 (IN = B) /**/; BY PATIENT_ID; IF A =1 OR B = 1; RUN; PROC FREQ DATA = DXMED; TABLES DX*T2D; RUN; **Q2**/ /**PATIENTS TREATED WITHOUT A DIAGNOSIS OF DIABETES**/ DATA MYHUMANA.T2DXXX; /**6,807,243**/ /*DISTINCT PATIENTS WITH NO T2D**/ SET MYHUMANA.T2DX; /*DISTINCT PATIENTS WITH T2D AND NO T2D, 7,436,258**/ IF T2D = 0; RUN; /**432,513**/ DATA Q2; /**INNER JOIN**/ MERGE MYHUMANA.T2DXXX (IN = A) /*DISTINCT PATIENTS WITH NO T2D /**7,436,258**/ MYHUMANA.MED(IN = B); /**DISTINCT PATIENTS TAKING ADM**/ /**1,759,104**/ BY PATIENT_ID; IF A=1 AND B=1; RUN; /***CROSS-CHECK WITH ALTERNATE DATA STEP***/ DATA Q2A; /**LEFT JOIN SO ALL PATIENTS TAKING ADM IS IN DATASET Q2A**/ /**1,759,104**/ MERGE MYHUMANA.MED(IN = A) /**1,759,104**/ /*DISTINCT PATIENTS TREATED*/ MYHUMANA.T2DX(IN = B); /**7,436,258**/ /*DISTINCT PATIENTS WITH T2D AND NO T2D**/ BY PATIENT_ID; IF A=1; RUN; /**T2D = 0 = 432,513**/ PROC FREQ DATA = Q2A; TABLES T2D; RUN; /**Q3**/ /**PATIENTS WITH DIABETES AND RECEIVED AN ANTIDIABETIC MEDICATION**/ /**367,471**/ DATA Q3; MERGE MYHUMANA.T2DXX (IN = A) /**WITH T2D ONLY**/ /**629,015**/ MYHUMANA.MED(IN = B); /**TAKING ADM ONLY**/ /**1,759,104**/ BY PATIENT_ID; IF A=1 AND B=1; RUN; s there an efficent way i can do this please? THANK YOU ... View more

Re: SUBSTRING TO IDENTIFY PATIENTS WITH A DISEASE

by Pyrite | Level 9 in New SAS User
‎09-17-2022 04:39 PM
‎09-17-2022 04:39 PM
I MEANT COLON(:) ... View more

Re: SUBSTRING TO IDENTIFY PATIENTS WITH A DISEASE

by Pyrite | Level 9 in New SAS User
‎09-17-2022 04:38 PM
‎09-17-2022 04:38 PM
Thank you Koen, I did not use the semicolon initially, I used it because one of my colleagues who got a similar number used it. I got the same number whether I used or not ... View more

SUBSTRING TO IDENTIFY PATIENTS WITH A DISEASE

by Pyrite | Level 9 in New SAS User
‎09-17-2022 04:20 PM
‎09-17-2022 04:20 PM
Hello everyone,    I ran the attached code and although it worked, the distribution of my DM  = 0  AND DM  =1 (6,807,243 AND 629,015) is different from what my colleagues are getting and our codes are very similar. I have attached a picture of one of my colleagues. I don't DATA WORK.MED; SET HUMANA.Deid_2014_2016_v1_MED; KEEP PATIENT_ID ADMISSION_DIAGNOSIS_CODE MEDICAL_DIAGNOSIS_CODE_2 MEDICAL_DIAGNOSIS_CODE_3 MEDICAL_DIAGNOSIS_CODE_4 MEDICAL_DIAGNOSIS_CODE_5 MEDICAL_DIAGNOSIS_CODE_6 MEDICAL_DIAGNOSIS_CODE_7 MEDICAL_DIAGNOSIS_CODE_8 MEDICAL_DIAGNOSIS_CODE_9 MEDICAL_PRIMARY_DIAGNOSIS_CODE; RUN; /***B. IDENTIFYING PATIENTS WITH TYPE 2 DIABETES MELLITUS BY SUBSTRACTING THOSE WITH TYPE 1 DIABTES FROM THOSE WI SUBSTRING FUNCTION IS USED TO EXTRACT THE FIRST THREE OBSERVATIONS OF "250" AND THE LAST 2 DIGITS CORESSPONDING TO THE DIAGNOSIS FOR ICD9 TYPE I DIABETES, FOR ICD10, E11 IS THE FIRST 3 DIGITS FOR ALL T2D DIAGNOSIS****/ /**INITIATE DIAGOSIS VARIABLE “T2D” AT 0, PATIENTS WITH T2D WILL HAVE A SCORE OF 1**/ DATA DIABETES; /**804,378,068**/ SET MED; *T2D = "0"; /**a new variable for type II diabetes**/ ARRAY DIAG (*) $ ADMISSION_DIAGNOSIS_CODE MEDICAL_PRIMARY_DIAGNOSIS_CODE MEDICAL_DIAGNOSIS_CODE_2 - MEDICAL_DIAGNOSIS_CODE_9; DM = "0"; DO I = 1 TO DIM (DIAG); IF SUBSTR (DIAG(I),1,3) IN:('250') AND SUBSTR(DIAG(I),4,2) NOT IN: ('03', '11', '13', '21', '23', '31', '33', '41', '43', '51', '53', '61', '63', '71', '73', '81', '83', '91', '93') THEN DM = "1"; IF SUBSTR (DIAG(I),1,3) IN: ('E11')THEN DM = "1"; IF DM = "" THEN DM = "0"; LABEL DM =' Type II Diabetes '; END; RUN; /**C. CREATING A DATASET FOR DISTINCT OBSERVATIONS WITH DM AND NO DM **/ PROC SORT DATA = DIABETES /***7,436,258***/ OUT = T2D; BY PATIENT_ID; RUN; /**CHECK FOR UNIQUE PATIENTS WITH T2D**/ PROC FREQ DATA = T2D; TABLES DM /**DM, 1 = 629,015 ; 0 = 6,807,243**/ RUN; know where I'm missing it. I will appreciate your help please. I cant share the file because it is copyright   Thanks ... View more

Re: substring

by Pyrite | Level 9 in New SAS User
‎09-15-2022 04:40 PM
‎09-15-2022 04:40 PM
DATA MYHUMANA.MED_DIAGX; SET MYHUMANA.MED; T2D = 0; /**a new variable for type II diabetes**/ ARRAY DIAG (*) ADMISSION_DIAGNOSIS_CODE MEDICAL_PRIMARY_DIAGNOSIS_CODE MEDICAL_DIAGNOSIS_CODE_2 - MEDICAL_DIAGNOSIS_CODE_9; DO I = 1 TO 10; IF SUBSTR (DIAG(I),1,3) IN ('250') AND SUBSTR(DIAG(I),4,2) NOT IN ('03', '11', '13', '21', '23', '31', '33', '41', '43', '51', '53', '61', '63', '71', '73', '81', '83', '91', '93') OR SUBSTR (DIAG(I),1,3) IN ('E11')then T2D = 1; LABEL T2D ='Type II Diabetes'; END; RUN; Thank you so much, the "NOT IN" function was what i was looking for, thnks for spotting the error in my code too. I adjusted my code and it worked ... View more

substring

by Pyrite | Level 9 in New SAS User
‎09-15-2022 08:56 AM
‎09-15-2022 08:56 AM
Hello everyone,    Please, I want to get patients with type II diabetes by subtracting those with Type I diabetes from the total patients with diabetes (the general code for diabetes is 250 but sub-types are identified by the last 2 digits). I am doing this because the coding for type I is more correct than type 2 and I was instructed to do so. My code doesn't work probably because I DATA MYHUMANA.MED_DIAG; SET MYHUMANA.MED; T2D = 0; /**a new variable for type II diabetes**/ ARRAY DIAG (*) ADMISSION_DIAGNOSIS_CODE MEDICAL_PRIMARY_DIAGNOSIS_CODE MEDICAL_DIAGNOSIS_CODE_2 - MEDICAL_DIAGNOSIS_CODE_9; DO I = 1 TO 10; IF SUBSTR (DIAG(I),1,3) IN ('250') AND SUBSTR NE (DX1(I),4,2)IN ('03', '11', '13', '21', '23', '31', '33', '41', '43', '51', '53', '61', '63', '71', '73', '81', '83', '91', '93') THEN T2D = 0; /**the codes in the DX1(I),4,2 function are type I diabetes**/ OR SUBSTR (DIAG(I),1,3) IN ('E11')then T1D = 1; LABEL T2D ='Type II Diabetes'; END; RUN; can't use the NE function with substr. Please advice.   Thank you ... View more

Re: CONVERTING NUMERIC TO CHARACTER

by Pyrite | Level 9 in New SAS User
‎09-07-2022 09:44 AM
‎09-07-2022 09:44 AM
Thank you, I had initially tried the "put" function, then I started using input as "trial and error"..😖. This code worked on my SAS windows but I am working on the project on an official SAS server. I don't know why the server keeps giving me errors. I have converted on my regular SAS and copied it to the server ... View more

Re: CONVERTING NUMERIC TO CHARACTER

by Pyrite | Level 9 in New SAS User
‎09-07-2022 09:40 AM
‎09-07-2022 09:40 AM
LOG 17 data WORK.ndc_code; 18 /*infile "~/ndc code.csv" dlm=";" dsd truncover firstobs=2;*/ 19 infile "C:\Users\HP SPECTRE\Desktop\Humana\ndc.csv" dlm=";" dsd truncover firstobs=2; 20 21 length 22 dx 8 23 ndc $11 24 ; 25 input 26 dx 27 ndc 28 ; 29 run; NOTE: The infile "C:\Users\HP SPECTRE\Desktop\Humana\ndc.csv" is: Filename=C:\Users\HP SPECTRE\Desktop\Humana\ndc.csv, RECFM=V,LRECL=32767,File Size (bytes)=18889, Last Modified=07Sep2022:07:00:45, Create Time=07Sep2022:07:44:12 NOTE: Invalid data for dx in line 2 1-14. RULE: ----+----1----+----2----+----3----+----4----+----5----+----6----+----7----+----8----+----9----+----0 2 1,00054014225, 14 dx=. ndc= _ERROR_=1 _N_=1 NOTE: Invalid data for dx in line 3 1-14. 3 1,00115115201, 14 dx=. ndc= _ERROR_=1 _N_=2 NOTE: Invalid data for dx in line 4 1-14. RULE: ----+----1----+----2----+----3----+----4----+----5----+----6----+----7----+----8----+----9----+----0 4 1,00115115202, 14 dx=. ndc= _ERROR_=1 _N_=3 NOTE: Invalid data for dx in line 5 1-14. 5 1,00378282277, 14 dx=. ndc= _ERROR_=1 _N_=4 NOTE: Invalid data for dx in line 6 1-14. 6 1,16252052501, 14 dx=. ndc= _ERROR_=1 _N_=5 NOTE: Invalid data for dx in line 7 1-14. 7 1,23155014901, 14 dx=. ndc= _ERROR_=1 _N_=6 NOTE: Invalid data for dx in line 8 1-14. 8 1,42291013290, 14 dx=. ndc= _ERROR_=1 _N_=7 NOTE: Invalid data for dx in line 9 1-14. 9 1,47781034201, 14 dx=. ndc= _ERROR_=1 _N_=8 NOTE: Invalid data for dx in line 10 1-14. 10 1,50419086251, 14 dx=. ndc= _ERROR_=1 _N_=9 NOTE: Invalid data for dx in line 11 1-14. 11 1,51862021201, 14 dx=. ndc= _ERROR_=1 _N_=10 NOTE: Invalid data for dx in line 12 1-14. 12 1,64380076006, 14 dx=. ndc= _ERROR_=1 _N_=11 NOTE: Invalid data for dx in line 13 1-14. 13 1,69543012210, 14 dx=. ndc= _ERROR_=1 _N_=12 NOTE: Invalid data for dx in line 14 1-14. 14 1,69543012211, 14 dx=. ndc= _ERROR_=1 _N_=13 NOTE: Invalid data for dx in line 15 1-14. 15 1,76439012210, 14 dx=. ndc= _ERROR_=1 _N_=14 NOTE: Invalid data for dx in line 16 1-14. 16 1,76439012211, 14 dx=. ndc= _ERROR_=1 _N_=15 NOTE: Invalid data for dx in line 17 1-14. 17 1,00054014025, 14 dx=. ndc= _ERROR_=1 _N_=16 NOTE: Invalid data for dx in line 18 1-14. 18 1,00115115001, 14 dx=. ndc= _ERROR_=1 _N_=17 NOTE: Invalid data for dx in line 19 1-14. 19 1,00115115002, 14 dx=. ndc= _ERROR_=1 _N_=18 NOTE: Invalid data for dx in line 20 1-14. 20 1,00378282077, 14 dx=. ndc= _ERROR_=1 _N_=19 NOTE: Invalid data for dx in line 21 1-14. WARNING: Limit set by ERRORS= option reached. Further errors of this type will not be printed. 21 1,16252052301, 14 dx=. ndc= _ERROR_=1 _N_=20 NOTE: 1180 records were read from the infile "C:\Users\HP SPECTRE\Desktop\Humana\ndc.csv". The minimum record length was 14. The maximum record length was 14. NOTE: The data set WORK.NDC_CODE has 1180 observations and 2 variables. NOTE: DATA statement used (Total process time): real time 0.01 seconds cpu time 0.01 seconds data ndc_code; /*infile "~/ndc code.csv" dlm=";" dsd truncover firstobs=2;*/ infile "C:\Users\HP SPECTRE\Desktop\Humana\ndc.csv" dlm=";" dsd truncover firstobs=2; length dx 8 ndc $11 ; input dx ndc ; run; Thank you so much. I tried your code, please see the errors it gave me, where did  go wrong please?  ... View more

CONVERTING NUMERIC TO CHARACTER

by Pyrite | Level 9 in New SAS User
‎09-07-2022 07:42 AM
‎09-07-2022 07:42 AM
I have a simple excel file that I have to convert to CSV to import to my SAS server. Every attempt to make SAS read the variable titled "NDC" as a character variable DATA MYHUMANA.NDC2; SET MYHUMANA.NDC; NDC = INPUT(NDC, BEST12.)); RUN; DATA MYHUMANA.NDC2; SET MYHUMANA.NDC; NDC = STRIP(PUT(NDC, BEST12.)); RUN; DATA MYHUMANA.NDC2; SET MYHUMANA.NDC; NDC_new = STRIP(PUT(NDC, BEST32.)); RUN; has been futile. I converted to text in the original excel file and it did not work, I also tried many SAS codes to convert to a character variable but it did not work. I don't seem to understand where I am wrong, I am so confused. I have out some of the codes I used and the excel file, please help. Thanks ... View more

Re: Is there a way to customize/change the label for the standardized ...

by Pyrite | Level 9 in SAS Procedures
‎07-14-2022 12:52 PM
‎07-14-2022 12:52 PM
I found this post helpful. I however noticed that labels with space in them e.g. 'Government insurance'n did not work because of the space in between. If I put a hyphen  '_'  between, it worked bu that does not fit the purpose ... View more

Re: Modifying space between forestplot data

by Pyrite | Level 9 in Graphics Programming
‎06-07-2022 03:22 PM
‎06-07-2022 03:22 PM
I'm running the program on SAS 9.4 windows. I tried modifying the resolution and size, but it was still the same.   Thank you ... View more

Re: Modifying space between forestplot data

by Pyrite | Level 9 in Graphics Programming
‎06-04-2022 08:50 PM
‎06-04-2022 08:50 PM
Thank you very much for the clarification🙈 and for moving it to the right section ... View more

Modifying space between forestplot data

by Pyrite | Level 9 in Graphics Programming
‎06-03-2022 04:41 PM
‎06-03-2022 04:41 PM
Hello everyone,   I have been trying to create a simple forest plot in SAS for my data. The ouptput looks good in my result viewer but its resolution is not great. I have tried severally to save it a picture format directly into my system but the space between the individual plots gets wider and the picture looks distorted. I have tried so many adjustments to the code such as changing the design height, the offsetmin and max but did not work. I will appreciate your help. Also adding a label "higher mortality risk" and "lower mortality risk" did not work so I ran only the proc template to generate the result attached Thank you AS shown in my SAS result viewer When i save on my PC using ODS listing/************************************************************/ proc format; value unicodeLabel 1 = "(*ESC*){unicode '2190'x} Lower mortality risk -" 2 = "- Higher mortality risk (*ESC*){unicode '2192'x}" ; run; data bottomLabel; attrib textLeft format=unicodeLabel. textRight format=unicodeLabel.; labXLeft = 0.9; labXRight= 1.1; labY = 27; textLeft = 1; textRight = 2; output; run; data forestfinal; merge tcr.forest bottomLabel; run; /******************did not work******************************************/ proc template; define statgraph forestAxisTable; dynamic _bandColor _headerColor _subGroupcolor; begingraph / designheight= 1.0in designwidth = 1.0in; discreteattrmap name='text' / trimleading=true; value '1' / textAttrs=(color=blue size=5 weight=bold); value '2' / textAttrs=(color=blue size=5 weight=normal); enddiscreteattrmap; discreteattrvar attrvar=txtDAV var=id attrmap='text'; layout lattice / columns=1; /*--Column headers--*/ sidebar / align=top; layout lattice / columns=3 rowweights=uniform columnweights = (0.3 0.42 0.28) backgroundcolor=_headerColor opaque=true; entry textAttrs=(size=8 weight=bold) hAlign=left "Subgroup" hAlign=right "Hazard Ratio (95% CI)"; entry textAttrs=(size=8 weight=bold) " "; entry textAttrs=(size=8 weight=bold) hAlign=right "P-value"; endLayout; endsidebar; /* Single cell with inner margins for left and right tables */ layout overlay / xAxisOpts=(display=(line ticks tickvalues) tickValueAttrs=(size=6 weight=bold) labelAttrs=(size=6 weight=bold) linearOpts=(tickValuePriority=true tickValueList=(0.0 0.5 1.0 1.5 2.0)) offsetMin=0.3 OFFSETMAX=0.5) yAxisOpts=(reverse=true display=none) wallDisplay=none; /* Left-side table */ innerMargin / align=left gutter=0.3in; axisTable y=obsId value=SUBGROUP / textGroup=txtDAV indentWeight=indentwt display=(values); axisTable y=obsId value=GROUP / display=(values); endInnerMargin; /* Hazard Ratio plot */ referenceLine y=ref / lineAttrs=(pattern=solid thickness=15 color = white); scatterPlot y=obsId x=mean / markerAttrs=(symbol=diamondFilled color = blue) xErrorLower=LOW xErrorUpper=HIGH errorBarCapShape=none; referenceLine x=1; /* better indicator */ textPlot y=labY x=labXLeft text=textLeft / strip=true position=left contributeOffsets=(yMin); textPlot y=labY x=labXRight text=textRight / strip=true position=right contributeOffsets=(yMin); /* Right-side table */ innerMargin / align=right; %let commonRightTableOpts = display=(values) showMissing=false labelAttrs=(size=7 weight=bold); /*axisTable y=obsId value=PCIGroup / &commonRightTableOpts;*/ axisTable y=obsId value=p_value / &commonRightTableOpts; /**axisTable y=obsId value=p_value / &commonRightTableOpts pad=(left=9%) display=(values);**/ endInnerMargin; endLayout; /* overlay */ endLayout; /* lattice */ endgraph; end; run; ods listing image_dpi=200; ods graphics / reset width=7.5in height=4.5in imagename='FIGURE4A'; proc sgrender data=tcr.forest2 template=forestAxisTable; dynamic _bandcolor='white' _headercolor=''; run;   ... View more

Re: Obtaining P-value from bootstrap samples

by Pyrite | Level 9 in Statistical Procedures
‎05-02-2022 06:22 AM
‎05-02-2022 06:22 AM
Thank you, I will try this. Yes, my hypothesis is OR = 1 ... View more

Obtaining P-value from bootstrap samples

by Pyrite | Level 9 in Statistical Procedures
‎04-30-2022 12:16 PM
‎04-30-2022 12:16 PM
Hello everyone,  I ran proc logistic on a bootstrapped sample, does anyone please know how I can obtain p-values? I will appreciate your help, thanks  proc surveyselect data=melanoma.nomiss NOPRINT seed=12345 out=Bootout method=urs /*resample with replacement*/ samprate=1 /*each bootstrap sample has N observations*/ reps=1000; /*generate 1000 bootstrap resamples */ %macro ODSOff(); /* Call prior to BY-group processing */ ods graphics off; ods exclude all; ods noresults; %mend; %macro ODSOn(); /* Call after BY-group processing */ ods graphics on; ods exclude none; ods results; %mend; %ODSOff; *B. run log reg on bootstrap samples; proc logistic data=Bootout plots=ROC outest = estimate; /** N = 1372**/ BY REPLICATE; title "multivariable logistic regression-bootstrapped sample"; class sexcat (param=ref ref = "2") racethc(param=ref ref = "0") insurcatb (param=ref ref = "2") location (param=ref ref = "0") povertycat (param=ref ref = "0") cci (param=ref ref = "0") surgcat(param=ref ref = "0") hormcat (param=ref ref = "1") radicat (param=ref ref = "0") chemocat (param=ref ref = "0") histology (ref = "amelanotic") primary (ref = "head_neck");; model immunocat (event='1') = agenum sexcat racethc insurcatb povertycat intervalb cci location surgcat hormcat radicat chemocat histology primary dxdateyearnum/ rl selection = backward include=6 clodds=pl; ods output CLoddsPL=CL_boot_Mort_mod_1 ; run; proc univariate data=cl_boot_mort_mod_1 noprint; class Effect; var OddsRatioEst; output out=WidePctls1 pctlpre=P_ pctlpts=2.5 97.5 mean=Mean Std=Std P ; run; proc print data=WidePctls1 noobs label; format Mean Std P_2_5 P_97_5 6.4; label Mean="BootMean" Std="BootStdErr" P_2_5="95% Lower CL" P_97_5="95% Upper CL"; run; %ODSOn; proc print data = WidePctls1; run; Output ... View more
Contact Me
Online Status
Offline
Date Last Visited
‎08-30-2025 07:33 PM
Likes given to
Likes from
View All
Quick links