About SarahW13

SarahW13 · ‎06-13-2018

I thought I would check again to see if anyone has thoughts on this. Would cox regression even be appropriate in this case, or would I need other methods (eg, conditional logistic regression) to compare survival between the matched cases and controls? Any guidance would be much appreciated.

SarahW13 · ‎06-12-2018

Thank you, sld and PaulBrownPhD! Since posting my question, I was able to figure out what was wrong sld, as you suggested, the problem was with my predictor being coded as a character variable. My predictor (binary:1 = on treatment, 0 = not on treatment) was coded as a character variable instead of as a numeric variable, so SAS wanted me to include it in the class statement. I just changed the predictor from a character variable to a numeric variable, and now proc phreg is giving me appropriate Cox regression output for the time-varying predictor. I just thought I would post this in case anyone else runs into the same problem in the future.

SarahW13 · ‎06-12-2018

Hello all! I have a question about how to do Cox regression for matched cases and controls. In my example, I have 2 controls matched to each case. I used the code from this article to generate my dataset of matched cases and controls: http://www2.sas.com/proceedings/sugi29/173-29.pdf However, now I am stuck at figuring out how to actually run the analysis. I have my original dataset that has all of the covariates that I need, and I have the new dataset that has the id's of the controls, id's of the matched cases, and the variables that I used to match the cases & controls (in this case, age). In my original dataset, this is what I have: id age treatment survival survival_censor 1 57.1 0 3.41 1 2 38.1 0 2.05 1 3 42.5 0 1.95 0 100 57.4 1 3.10 1 101 57.0 1 5.38 0 120 38.2 1 0.95 0 125 38.9 1 4.41 0 105 43.9 1 3.41 0 132 41.3 1 2.37 0 Below is an example of what I have in the new dataset that has case:control matches (1:2): case_id control_id case_age control_age rand_num num 1 100 57.1 57.4 0.56778 1 1 101 57.1 57.0 0.56778 2 2 120 38.1 38.2 0.23116 1 2 125 38.1 38.9 0.23116 2 3 105 42.5 43.9 0.95096 1 3 132 42.5 41.3 0.95096 2 I would like to run a Cox regression comparing the survival between the matched cases and controls, but I have the actual case-control pairings in one dataset and the survival info (plus other covariates that I need to adjust for) in a separate dataset. Does anyone have any advice on the next steps I need to do? Thanks!

SarahW13 · ‎06-08-2018

Thanks, ballardw! Since posting my question, I was able to figure out what was wrong. My predictor (binary:1 = on treatment, 0 = not on treatment) was coded as a character variable instead of as a numeric variable, so SAS wanted me to include it in the class statement. I just changed the predictor from a character variable to a numeric variable, and now proc phreg is giving me appropriate Cox regression output even with the time-varying predictor. I just thought I would post this in case anyone else runs into the same problem in the future.

SarahW13 · ‎06-08-2018

Hi all! I have a question about the class statement in proc phreg. I am trying to run a cox regression model with a time-varying predictor. Whenever I run the model, I get completely blank output. I think it has something to do with the class statement. When I try to run the model with a time-nonvarying predictor (using only the first treatment that people received, instead of modeling treatment as time-varying), it only works if I include the predictor in the class statement. This is an example of the model with the time-nonvarying predictor: Proc phreg data=outcomes; Class treatment (ref="0"); Model Timetostatus*event1yes(0) = treatment1; Run; The code above does give me appropriate output. It does NOT work if I exclude the class statement. However, I actually want to make treatment a time-varying predictor like below: Proc phreg data=outcomes; Model Timetostatus*event1yes(0) = treatment; If timetostatus < regimen1_timetoend then treatment = treatment1; Else If timetostatus > regimen2_timetostart and timetostatus <= regimen2_timetoend then treatment = treatment2; Run; When I run the model above, the output is blank, and SAS shows as if no events occurred. I have carefully checked all the variables that I have used to make treatment a time-varying predictor; they have all been calculated correctly, and there are definitely events. So I am assuming that the reason why the output is blank is that I have not put treatment in the class statement, since I had to do that in the first model. Is there a way to include a time-varying predictor in the class statement? I have tried a model including the class statement with treatment; I also tried a model with treatment1 and treatment2 in the class statement. None of those worked. Any thoughts on this?

SarahW13 · ‎06-07-2018

Hi all! I have a question about the class statement in proc phreg. I am trying to run a cox regression model with a time-varying predictor. Whenever I run the model, I get completely blank output. I think it has something to do with the class statement. When I try to run the model as a time-nonvarying predictor, it only works if I classify the predictor in the class statement. This is an example of the model with the time-nonvarying predictor: Proc phreg data=outcomes; Class treatment (ref="0"); Model Timetostatus*event1yes(0) = treatment; Run; The code above does give me appropriate output. It does NOT work if I exclude the class statement. However, I actually want to make treatment a time-varying predictor like below: Proc phreg data=outcomes; Model Timetostatus*event1yes(0) = treatment; If timetostatus < regimen1_timetoend then treatment = treatment1; Else If timetostatus > regimen2_timetostart and timetostatus <= regimen2_timetoend then treatment = treatment2; Run; When I run the model above, the output is blank, and SAS shows as if no events occurred. I have triple-checked all the variables that I have used to make treatment a time-varying predictor; they have all been calculated correctly, and there are events. So I am assuming that the reason why the output is blank is that I have not put treatment in the class statement. Is there a way to include treatment in the class statement if it is a time-varying predictor? I have tried a model including the class statement with treatment; I also tried a model with treatment1 and treatment2 in the class statement. None of those worked. Any thoughts on this?

SarahW13 · ‎06-06-2018

Thank you so much! It worked! The only thing that I changed is that when I did proc sort, I did this: proc sort data=dataset; by id date; run; Not sure if it would have still worked with only sorting by id...but I sorted by both id and date since I wanted the Seq to be in the order of the dates. Thanks again!

SarahW13 · ‎06-06-2018

Hello all! I have a question about how to assign a sequential number to each subject based on another variable that is a date. This would be the start dataset: id date 1 1/10/07 1 3/23/07 1 5/23/08 2 2/10/07 2 2/01/08 3 8/01/12 3 1/23/13 3 2/10/14 3 7/10/15 This is what I want as the end result: id date seq 1 1/10/07 1 1 3/23/07 2 1 5/23/08 3 2 2/10/07 1 2 2/01/08 2 3 8/01/12 1 3 1/23/13 2 3 2/10/14 3 3 7/10/15 4 Any advice?

SarahW13 · ‎06-05-2018

Thank you so much, amatsu! This worked! 🙂

SarahW13 · ‎06-05-2018

I have a question about how to remove duplicate rows based on the value of a certain variable. I'm using the example below to explain what I am effectively trying to do. In this dataset, each id has many duplicate values for the variable med. In some of those rows, there is also a value for the variable flag. In those cases, I want to preserve only the rows where there is a value for flag. However, there are other rows for the id and med value where there is NO value for the variable flag (eg, where id = 1 and Med = B); in those cases, I want to just keep all of those rows intact. id med flag seq 1 A 1 1 1 A . 2 1 A . 3 1 B . 1 1 B . 2 2 A 1 1 2 A . 2 2 C . 1 2 C 0 2 2 C . 3 2 C . 4 2 D 1 1 This is the dataset that I want to have as the end result: id med flag seq 1 A 1 1 1 B . 1 1 B . 2 2 A 1 1 2 C 0 2 2 D 1 1 Does anyone have any advice?

SarahW13 · ‎06-05-2018

I didn't know that (about the create table option). The code from Astounding doesn't seem to be working when I try it. I think I'm just not using it correctly. Astounding (or anyone else), could you clarify that code for me? For the code a.value, would "a" be referring to the dataset name and "value" referring to the variable name? In the example datasets that I gave, I would be trying to copy all columns (value, med, start_date, end_date), with the four values for each id getting added to all rows for that corresponding id from the other dataset.

SarahW13 · ‎06-04-2018

Astounding, thanks for the reply! Is there any way to actually do this as a data step without creating a table in proc sql? My understanding about proc sql tables is that I can't actually use that table as a dataset. I want to do this merge in a way that creates an actual dataset that I can manipulate further.

SarahW13 · ‎06-04-2018

Hello all! I have two datasets with a different number of rows per person_id, and I would like to merge the two datasets in a way that each row from dataset B gets added to all rows for that corresponding person in dataset A. Below is an example of what the original datasets would look like: DATASET A: id value 1 A 1 B 1 C 2 A 2 C DATASET B: id med start_date end_date 1 1 1/5/13 2/10/16 1 2 3/10/16 4/10/16 2 1 1/8/13 1/10/13 2 2 3/5/18 5/3/18 2 3 5/5/18 6/1/18 This is what I would like the merged dataset to look like: id value med start_date end_date 1 A 1 1/5/13 2/10/16 1 A 2 3/10/16 4/10/16 1 B 1 1/5/13 2/10/16 1 B 2 3/10/16 4/10/16 1 C 1 1/5/13 2/10/16 1 C 2 3/10/16 4/10/16 2 A 1 1/8/13 1/10/13 2 A 2 3/5/18 5/3/18 2 A 3 5/5/18 6/1/18 2 C 1 1/8/13 1/10/13 2 C 2 3/5/18 5/3/18 2 C 3 5/5/18 6/1/18 Can anyone advise me how I can merge the two datasets so that all rows for each person in dataset B gets added to each row for that same person in dataset A? Thank you for your help!

SarahW13 · ‎02-19-2018

Warren and Sanjay, thank you both! Warren, thanks for the extensive code. I'm still pretty new to SAS (this is my first time generating Kaplan Meier curves in SAS), and I've never used annotations in SAS before. I'm trying to finish these plots within the next few days, so I'm going to look for an easier (probably less elegant) solution for now, but I am definitely going to comb through your code later for my learning. Sanjay, would it be possible for you to share the code that produced the first of the two plots that you posted? Thank you both again!

SarahW13 · ‎02-19-2018

Hello SAS community! I'm using SAS University. I have run proc lifetest to generate Kaplan Meier curves, and I managed to use Proc template to change the format of my x-axis, y-axis, entrytitle, etc. I used viewmin, viewmax, offsetmin, and offsetmax to better control my axes. I want both the x-axis and y-axis to start at 0, so I set viewmin=0 and offsetmin=0 for both the x-axis and y-axis. However, I have realized that my at risk table is causing the x-axis to start a little before 0. I'm attaching an image that shows what my axis shows. Does anyone know what proc template language would allow me to either adjust my y-axis or at risk table so that the x-axis min is exactly set to 0? Thank you for your help!

Online Status	Offline
Date Last Visited	‎12-30-2019 04:39 AM

Identify overlapping date intervals

Matching data by date intervals

Creating variable that indicates order by two other variables

Matching observations by ID and date

Extracting specific information from rows

Re: Problem with formatting with data forest plot

Problem with formatting with data forest plot

Identifying ids with two values

Identify ids with consecutive values meeting threshold

finding time periods that overlap

Re: Creating variable that indicates order by two other variables

Re: Matching observations by ID and date

Re: Matching observations by ID and date

Re: Matching observations by ID and date

Re: Identifying ids with two values

question about selectively removing duplicates

Re: Proc phreg - time-varying predictor

Re: Assign sequential id by id and date

Re: Comparing survival between matched cases and controls - proc phreg...

Re: Proc phreg: specifying time-varying predictor as class

Comparing survival between matched cases and controls - proc phreg?

Re: Proc phreg - time-varying predictor

Proc phreg - time-varying predictor

Proc phreg: specifying time-varying predictor as class

Re: Assign sequential id by id and date

Assign sequential id by id and date

Re: question about selectively removing duplicates

question about selectively removing duplicates

Re: Merging two datasets with different number of rows per id

Re: Merging two datasets with different number of rows per id

Merging two datasets with different number of rows per id

Re: Proc lifetest - at risk table preventing x axis from starting at 0

Proc lifetest - at risk table preventing x axis from starting at 0