BookmarkSubscribeRSS Feed
djbateman
Lapis Lazuli | Level 10

I am creating a swimmer plot using the HIGHLOW statement in PROC SGPLOT to show how long a subject has been on our study.  In addition, I want to add an arrow on the end of a bar to denote that a subject is still active on our study; otherwise, there is a flat end.  My issue is that if a subject has recently joined our study, their time on study will be quite short (e.g., see row 11 in the data below--only on study for 1 week).  A short study time will result in the arrow not showing up as the end cap.

I was reading through the SAS documentation for PROC SGPLOT, and I found the following statement:

"Restriction:Arrow heads are not displayed for very short bars. Bar height must be at least twice the size of the arrow heads in order for the arrow heads to appear. "

My question here is this: Is there a work around for this restriction, or do I need to manually watch for these and maybe add an asterisk using the highlabel option and footnote the bars that are influenced?

I have thought about adding a condition to automatically add the asterisk, but I think the length of the bar will change every time I increase the length of the x-axis, so it isn't very consistent.  I'm not sure how to accommodate that.


proc format;

      value schedule

            2.5='TIW'

            8.5='QW';

run;

data swimmer;

      input N SCHEDULE $ DOSE TIMEONSTUDY ACTIVE $;

      format N schedule. dose 5.2;

      if active='Y' then ACTIVECAP='FilledArrow';

      BASE=0;

      cards;

1     TIW         0.2         10.5714           N

2     TIW         0.3         13.5714           N

3     TIW         0.45  7.5714            N

4     TIW         0.6         6.5714            N

6     QW          0.3         7                 N

7     QW          0.45  22                N

8     QW          0.6         4                 N

9     QW          0.8         10                Y

10    QW          1.0         6.1429            Y

11    QW          1.25  1                 Y

;

run;

proc sgplot data=swimmer;

      title 'Time on Study';

      footnote1 j=l height=.9 'NOTE: An arrow shows that the patient is still active on study treatment while a solid stop shows that the patient has discontinued treatment.';

      highlow y=n low=base high=timeonstudy / type=bar highcap=activecap lowlabel=dose;

      refline 5;

      xaxis min=-2 max=24 grid label='Time on Study (weeks)' values=(0 to 24 by 6);

      yaxis reverse display=(noticks) min=1 max=11 valueshint values=(2.5 8.5) label='Dosing Schedule & Dose Level';

run;

2 REPLIES 2
Jay54
Meteorite | Level 14


This is only a work around.  Using an iterative process "SwimmerTest", I find the smallest value of TimeOnStudy (T) that will show the arrowhead.  I copy TimeOnStudy to TOS, setting values < T to T.  Then I draw the graph using HIGH=TOS.  I displayed the actual value of TimeOnStudy as HighLabel.

I am unable to add the resulting image due to some bug in the "add Image" tool.

ods html close;
ods listing gpath='C:\Work\SASUser' image_dpi=100;

proc format;
      value schedule
            2.5='TIW'
            8.5='QW';
run;
data swimmer;
      input N SCHEDULE $ DOSE TIMEONSTUDY ACTIVE $;
      format N schedule. dose 5.2;
      if active='Y' then ACTIVECAP='FilledArrow';
      BASE=0;
      cards;
1     TIW         0.2         10.5714           N
2     TIW         0.3         13.5714           N
3     TIW         0.45        7.5714            N
4     TIW         0.6         6.5714            N
6     QW          0.3         7                 N
7     QW          0.45        22                N
8     QW          0.6         4                 N
9     QW          0.8         10                Y
10    QW          1.0         6.1429            Y
11    QW          1.25        1.0               Y
;
run;
proc print;run;

data swimmer2;
      input N SCHEDULE $ DOSE TIMEONSTUDY ACTIVE $;
      format N schedule. dose 5.2;
      if active='Y' then ACTIVECAP='FilledArrow';
      BASE=0;
      cards;
1     TIW         0.2         10.5714           N
2     TIW         0.3         13.5714           N
3     TIW         0.45        7.5714            N
4     TIW         0.6         6.5714            N
6     QW          0.3         7                 N
7     QW          0.45        1.3               Y
8     QW          0.6         1.2               Y
9     QW          0.8         1.1               Y
10    QW          1.0         1.0               Y
11    QW          1.25        0.9               Y
;
run;
proc print;run;


ods graphics / reset width=5in imagename='SwimmerTest';
proc sgplot data=swimmer2;
      title 'Time on Study';
      footnote1 j=l height=.9 'NOTE: An arrow shows that the patient is still active on study treatment while a solid stop shows that the patient has discontinued treatment.';
      highlow y=n low=base high=timeonstudy / type=bar highcap=activecap
              lowlabel=dose highlabel=timeonstudy;
      refline 5;
      xaxis min=-2 max=24 grid label='Time on Study (weeks)' values=(0 to 24 by 6);
      yaxis reverse display=(noticks) min=1 max=11 valueshint values=(2.5 8.5) label='Dosing Schedule & Dose Level';
run;

data swimmer3;
  set swimmer;
  tos=timeonstudy;
  if timeonstudy < 1.1 then tos=1.1;
run;
proc print;run;

ods graphics / reset width=5in imagename='Swimmer';
proc sgplot data=swimmer3;
      title 'Time on Study';
      footnote1 j=l height=.9 'NOTE: An arrow shows that the patient is still active on study treatment while a solid stop shows that the patient has discontinued treatment.';
      highlow y=n low=base high=tos / type=bar highcap=activecap
              lowlabel=dose highlabel=timeonstudy;
      refline 5;
      xaxis min=-2 max=24 grid label='Time on Study (weeks)' values=(0 to 24 by 6);
      yaxis reverse display=(noticks) min=1 max=11 valueshint values=(2.5 8.5) label='Dosing Schedule & Dose Level';
run;

sas-innovate-2024.png

 

Time is running out to save with the early bird rate. Register by Friday, March 1 for just $695 - $100 off the standard rate.

 

Check out the agenda and get ready for a jam-packed event featuring workshops, super demos, breakout sessions, roundtables, inspiring keynotes and incredible networking events. 

 

Register now!

How to Concatenate Values

Learn how use the CAT functions in SAS to join values from multiple variables into a single value.

Find more tutorials on the SAS Users YouTube channel.

Click image to register for webinarClick image to register for webinar

Classroom Training Available!

Select SAS Training centers are offering in-person courses. View upcoming courses for:

View all other training opportunities.

Discussion stats
  • 2 replies
  • 2537 views
  • 5 likes
  • 2 in conversation