About lisahoward

lisahoward · ‎04-26-2015

Thank you Jacob I created an offset variable and then included it in my procedure: proc genmod data=IRsforCohort; class Exposure &covariates ; model outcome=Exposure &covariates Age_At_Index IndexYear / dist = poisson offset=myoffset link=log type3; Estimate ' 1 Vs 0' Exposure 1 -1 ; ods output Estimates=RR&infile._adj; title1 "Proc GENMOD for Adjusted &infile &time"; run; I changed the IR variable I had created in a previous data step and now use the Outcome variable instead so have outcome=Exposure. I get the following results. In your opinion you think I should use the Mean rather than the LBeta variables to show relative risk for Treatment Difference in Headache rate during 90 days follow-up period for treated and untreated group Contrast Estimate Results Mean Mean L'Beta Standard L'Beta Chi- Label Estimate Confidence Limits Estimate Error Alpha Confidence Limits Square Pr > ChiSq 1 Vs 0 0.9714 0.5534 1.7049 -0.0291 0.2870 0.05 -0.5916 0.5335 0.01 0.9194

lisahoward · ‎04-25-2015

Sorry to bother you again. I put the syntaxwith outcome=EXPOSURE instead or IR=EXPOSURE to the left of =, But i think i have the 'offset=log(PersonYrs'' bit wrong as I get an error. Please can you help, I have created PersonYrs so this variable exists in the IRsforCohort dataset. thanks. proc genmod data=IRsforCohort; class EXPOSURE &covariates ; model outcome=EXPOSURE &covariates Age_At_Index IndexYear / dist = poisson offset=log(PersonYrs) link=log type3; Estimate ' 1 Vs 0' EXPOSURE 1 -1 ; ods output Estimates=RR&infile._adj; title1 "Proc GENMOD for Adjusted &infile &time"; run; ERROR: Variable LOG not found. ERROR 22-322: Syntax error, expecting one of the following: ;, A, AGGREGATE, ALPH, ALPHA, CICONV, CL, CLCONV, CODING, CONVERGE, CONVH, CORRB, COVB, D, DIAGNOSTICS, DIST, DSCALE, ERR, ERROR, EXPECTED, ID, INFLUENCE, INITIAL, INTERCEPT, ITPRINT, LINK, LOGNB, LRCI, LRCL, MAXIT, MAXITER, NOINT, NOLOGNB, NOPRINTCL, NOSCALE, OBSTAT, OBSTATS, OFFSET, P, PRED, PREDICTED, PSCALE, R, RESIDUAL, RORDER, SCALE, SCORING, SINGULAR, TYPE1, TYPE3, TYPE3WALD, WALD, WALDCI, WALDCL, XVARS.

lisahoward · ‎04-25-2015

Great thank you so much I am very appreciatl=ive of the time you have taken to answer my questions. I already created person years in a previous data step so that I could calculate the Incidence rate of the outcome. I use the IR in the model statement: IR=Exposure . So I will include the Person years variable in the offset option. I presume I am correct in ssuming it should be IR=Exposure and not Outcome=Exposure as the spec wording I am going off in the protocol is : Unadjusted model: Headace rate = Exposure; Adjusted model: AMI rate =Exposure Group + Age+ indexyear + baseline risk factors I am assuming Headache rate is Incidence Rate and not Outcome .?

lisahoward · ‎04-25-2015

Hi , How would I go about creating offset=log(personyears). This is my current syntax: ods listing; proc genmod data=IRsforCohort; class TestRx &covariates ; model IR=TestRX &covariates Age_At_Index IndexYear / dist = poisson link=log type3; Estimate ' 1 Vs 0' testrx 1 -1 ; ods output Estimates=RR&infile._adj; title1 "Proc GENMOD for Adjusted &infile &time"; run; Do I just add (in bold): ods listing; proc genmod data=IRsforCohort; class TestRx &covariates ; model IR=TestRX &covariates Age_At_Index IndexYear / dist = poisson offset= personyears link=log type3; Estimate ' 1 Vs 0' testrx 1 -1 ; ods output Estimates=RR&infile._adj; title1 "Proc GENMOD for Adjusted &infile &time"; run;

lisahoward · ‎04-25-2015

Dear Jacob, Thank you so much for your help. I will modify the syntax to include all categorical variables in the class statement only and then in the model statement i will include the categorical variables + age and year of index. I appreciate your help. Could I ask your advice on the offset option please. Should I include this , I am unsure what and how this will effect the RR. The question I am trying to answer or the RR I am trying to get is based on the following: Treatment Difference in the Diagnosis (Headache) Incidence rate , during the follow-up period (defined as 90 days) between treated and untreated group. I am doing both unadjusted and adjusted Relative Rates. Also should I be displaying the LBETA estimate value or the mean estimate value in my report to represent the RR value? I apologise I am not a statistician so any advice is truly appreciated. Many thanks in advance.

lisahoward · ‎04-25-2015

Thank you so I would put the &covariates in the class statement as well? I had understood I only needed to put continuous variables (such as Age and Year in the class statement) and then any other covariates that were categorical did not need to be listed in the class statement. The &covariates contains Age and Year as well as all the 0/1 answered variables as well.

lisahoward · ‎04-25-2015

Hello. I am trying to calculate the RR using Poisson regression for a cohort of patients (treated=1, untreated=0) who have an outcome of X using the following Syntax. i have calculated the IR rate by exposure in a previous datastep as I need to call it in to use in the model data line in this procedure. The model is adjusted with some specified 0/1 covariates which I call in using the &covariates statement. I also need to include two other variables as covariates Age_At_Index and Index_Year. My question is do I need to put them in the Class statement? The model looks at Incidence Rate = Exposure. Many thanks in advance for any help. ods listing; proc genmod data=IRsforCohort; class Exposure IndexYear Age_At_Index; model IR=Exposure &covariates / dist = poisson link=log type3; estimate ' 1 Vs 0' Exposure 1 -1 ; ods output ParameterEstimates=RR&infile._adj; title1 "Proc GENMOD for Adjusted"; run;

lisahoward · ‎04-24-2015

Apologises for the time delay with regards to my response but I have a quick question regarding the offset statement which =Person. What is the variable 'Person representing' Is this variable holding the total number of patients who had an outcome for that particular exposure? Many thanks.

lisahoward · ‎04-02-2015

I have been asked to calculate the Standardized difference for continuous and categorical variables. I understand how to do it for the continuous variables but am unsure how to do it for the binary categorical variables. This is the code I use for the continuous variables where x and y are the means of the two variables for the two groups Y =treated X = Untreated : I use proc means to get the mean for the continuous variable in the untreated and treated groups first and then run the following code: data Stan_Diff; set Cohortfile; if Var_Type='Cont' then do; length stddiff diff 8; format stddiff 6.2 diff 8.4; denomSD = sqrt((y_&varname_Treat**2 + x_&varname_UnTrt**2)/2); diff = x_&varname_Treat - y_&varname_UnTrt; if denomSD gt 0 then do; stddiff = 100*(nx_&varname_Treat - ny_&varname_UnTrt)/denomSD; end; else stddiff = .; end; run; For the Binary variables I am using proc freq to get total counts and percentages not proc means. Should I be using proc means. I am unsure how to get the SD without creating a mean value first. My data set looks like this: For the continous variables Age and Medications I use proc means to produce a mean. For the catagorical variables Diabetes Asthma etc I use proc freq to generate the count and percentage for my output table. So this is where I get stuck as with out a mean how do i produce the SD? Patient Exposure Age Medications Diabetes Asthma Smoking Osteo Fracture 1 1 50 6 1 0 1 0 1 2 1 54 4 0 0 1 0 1 3 0 42 11 0 1 1 1 1 4 1 35 0 1 1 0 1 0 5 0 70 9 1 0 1 0 1 6 0 55 12 1 1 0 1 0 The table I am trying to create looks like this: Before propensity score match (N=xxx) Baseline Characteristics Treated Untreated Standardized Difference (N=xx) (N=xx) Demographics Age – Mean xx.x (xx.x) xx.x (xx.x) x.xxx Number of unique medications – Mean xx.x (xx.x) xx.x (xx.x) x.xxx Baseline Comorbidities Diabetes– count (%) xx.x (xx.x) xx.x (xx.x) x.xxx Asthma – count (%) xx.x (xx.x) xx.x (xx.x) x.xxx Smoking – count (%) xx.x (xx.x) xx.x (xx.x) x.xxx Osteoporosis – count (%) xx.x (xx.x) xx.x (xx.x) x.xxx Fracture – count (%) xx.x (xx.x) xx.x (xx.x) x.xxx Many thanks in advance, Lisa.

lisahoward · ‎03-27-2015

Hello, thank you so much for your response. There is one exposure group ( flag = 1) and one untreated group (flag =0). Age is a continuous variable but i may have to add some discrete variables into the adjusted model such as various comorbids e.g. Diabetes etc ( all have a 1 or 0 flag). Would I just add these into the class line of the code and the model line of the code for the adjusted part? May I ask what the line of code does (in particular the -1 bit? estimate ' 1 Vs 2' Exposure 1 -1; *Adjusted RR I really appreciate all your help and you taking the time to respond. Many thanks, Lisa

lisahoward · ‎03-25-2015

I am trying to create the following table below and am struggling on where to start: Mock Table1: Treatment Difference in the Headache rate during 90 , 180 and 270 days follow-up period between treated Untreated group (using Poisson Regression) Headache Incidence Rate Unadjusted Adjusted Treated Not Treated RR 95% CI P value RR 95% CI P value N=xxx N=xxx n % n % xx x.xx xx x.xx xx.xx xx.xx-xx.xx .xxxx xx.xx xx.xx-xx.xx .xxxx * there will be three rows in total one for 90 days one for 180 days and one for 270 days. N=number of Subjects; RR=risk ratio (treated vs. untreated); CI=confidence interval for the RR; RR, 95% CI and p value are from the poisson model. Unadjusted model: Headache rate = Exposure Group; Adjusted model: Headache rate =Exposure Group + Age+ index year I know (or think) I have to use Proc Genmod , but I have never used this procedure before and I am unsure of the syntax used to get the RR CI and P values, or the adjusted and unadjusted stats. The dataset I am using has the all subjects their exposure, age, index year , flag for headache (0/1) and headache diagnosis date. I am not bothered about the cosmetics of the table format tI can figure that bit out, it is the information inside the table that I need help with Any help would be gratefully received . Many thanks in advance, Lisa.

lisahoward · ‎02-10-2015

Hello, I am trying to keep all dates in a table that occur between -365 to -30 days prior to Index date. Any dates occurring within 30 days of the Index are not required. I know this is probably quite straightforward code but I can't seem to figure out how to do this. Any help on the syntax would be appreciated. Many thanks.

lisahoward · ‎01-26-2015

Thank you so much for your quick response.

lisahoward · ‎01-26-2015

Hello, I wondered if anyone could me in coming up with some SAS code help to randomly assign an index date to a patients diagnosis, (lets say diabetes), within a 6 month calender block? Normally the index date would be taken as first occurrence of diagnosis, in this instance though the index date is is assigned by randomly selecting one of the patients physician visits within the 6 month calender block that the first diagnosis occurred. For example patient X is diagnosed 02Jan2000 with Diabetes. He has 100 Physician visits and dates in the healthcare database , 90 of these dates/visits occur in the same 6 month block of time 01jan2000 - 30june2000 as the first diagnosis of diabetes. 1 of these 90 visit dates has to be randomly chosen and assigned to the patient as the index date. The other 10 visit dates can't be used as they fall outside of the cohort accrual 6 month block (either before or after 01jan2000 - 30june2000). Does anyone have some code that would help me find and assign index dates this way please? The over all cohort of patients spans 2004-2013 so there would be 20 blocks of 6 months to work through. Index are only assigned/matched within each block so if a diagnosis occurred in block 20 (1JUL2013-31DEC2013) then only visits for that patient also falling within the same time block are used to randomly find and assign an index date for that patient. So far I have created a dataset with first diagnosis and have created a variable Index_Range that I have populated as follows: PatID Dx Date Index_Range 1 20Jan2005 1JAN2005 - 30JUN2005 2 18JUN2006 1JAN2006 - 30JUN2006 3 02SEP2012 1JUL2012 - 31DEC2012 I next search and pull out all physician records for patients in dataset created above and assign a similar Index_Range type variable using the physician visit date. I presume I then merge all visits in Diagnosis dataset and Physician visit Dataset matching on PatiD and Index_Range variable. What I don't know is how to code after, when i have merged my datasets and have all the visits matched via index range to the patients diagnosis what is the code I use to randomly select only one visit per patient and assign that visit as the Index date. Any thoughts on helping me write the code to 'randomly select' would be great. Thank you in advance.

lisahoward · ‎10-02-2014

Following on from that and trying to explain it a bit better if a patient had a diagnosis of 12apr2012 and their enrollment start date was 12apr2010 and their enrollment end date of 18may 2012 i want to calculate the number of days of enrollment just in the year of diagnosis so only want to return the number of days of enrollment in 2012 and not include any days prior to 2012 or post 2012.

Online Status	Offline
Date Last Visited	‎09-01-2015 07:11 AM

Re: Statistical significance differences syntax

Re: Statistical significance differences syntax

Statistical significance differences syntax

SAS proc report to PDF , truncating text

Re: Hazard Ratio Plot

Hazard Ratio Plot

Re: Selecting colours for plot lines by grouping variable

Re: Selecting colours for plot lines by grouping variable

Re: Selecting colours for plot lines by grouping variable

Re: Selecting colours for plot lines by grouping variable

Re: Leading Zeros

Re: Proc GENMOD Class question

Re: Proc GENMOD Class question

Re: Proc GENMOD Class question

Re: Proc GENMOD Class question

Re: Proc GENMOD Class question

Re: Proc GENMOD Class question

Proc GENMOD Class question

Re: Proc Genmod Syntax question

Standardized Differences binary categorical variable

Re: Proc Genmod Syntax question

Proc Genmod Syntax question

Sas - keep only observations between 2 dates

Re: Randomly assigning an Index Date to a patient

Randomly assigning an Index Date to a patient

Re: Calculating Days within a given year.