← SASnRDefine and Call a Macro with Parameters

Introduction

Clinical programs repeat the same logic for different treatment groups, variables, or datasets — a reusable macro (SAS) or function (R) avoids copy-paste and keeps code maintainable.
We define a block of logic once, give it parameters for the parts that change, and call it with different arguments each time.
SAS uses %MACRO / %MEND with keyword parameters and optional defaults. R uses function() with named arguments and optional defaults.
This lesson builds a small utility that subsets an ADSL-like dataset by treatment group and calculates mean age — the kind of reusable helper every clinical program benefits from.
After running, we check that each call produces the correct subset and summary for the requested treatment arm.

Define and Call a Macro with Parameters

SAS (Base SAS)+

*==============================================================================;
* Create sample ADSL data;
*==============================================================================;

data adsl;
    input usubjid $ trt01p $ age sex $;
    datalines;
SUBJ001 Placebo 67 M
SUBJ002 DrugA   59 F
SUBJ003 Placebo 72 F
SUBJ004 DrugA   65 M
SUBJ005 Placebo 60 F
SUBJ006 DrugA   55 M
SUBJ007 Placebo 70 M
SUBJ008 DrugA   63 F
;
run;

*==============================================================================;
* Define a macro with keyword parameters and defaults;
*==============================================================================;

%macro trt_summary(ds=adsl, var=age, trt=);

    *--- Subset to the requested treatment arm ---;
    data _subset;
        set &ds.;
        where trt01p = "&trt.";
    run;

    *--- Calculate summary statistics ---;
    proc means data=_subset n mean std min max;
        var &var.;
        title "Summary of %upcase(&var.) for &trt.";
    run;

%mend trt_summary;

*==============================================================================;
* Call the macro for each treatment group;
*==============================================================================;

%trt_summary(trt=Placebo);
%trt_summary(trt=DrugA);
%trt_summary(ds=adsl, var=age, trt=Placebo);

To define a reusable block in SAS, we wrap our logic in %MACRO name(params) ... %MEND;.
Keyword parameters are declared with name=default syntax — here ds=adsl and var=age have defaults, while trt= is required (no default).
Inside the macro, we reference each parameter with the &name. syntax, just like any macro variable.
The trailing dot on &ds., &var., and &trt. tells SAS where the variable name ends — essential when the next character could be mistaken as part of the name.
We use %upcase(&var.) in the TITLE statement to display the variable name in uppercase — showing that macro functions work inside macro code.
To call the macro, we write %trt_summary(trt=Placebo); — parameters we omit pick up their defaults.
The third call spells out every parameter explicitly, which is good practice for readability in production code.
After running, we confirm two PROC MEANS outputs: one for the Placebo arm, one for DrugA, each showing N, mean, std, min, and max of AGE.

R (tidyverse)+

library(tidyverse)

#==============================================================================
# Create sample ADSL data
#==============================================================================

adsl <- tribble(
  ~usubjid,  ~trt01p,    ~age, ~sex,
  "SUBJ001", "Placebo",  67,   "M",
  "SUBJ002", "DrugA",    59,   "F",
  "SUBJ003", "Placebo",  72,   "F",
  "SUBJ004", "DrugA",    65,   "M",
  "SUBJ005", "Placebo",  60,   "F",
  "SUBJ006", "DrugA",    55,   "M",
  "SUBJ007", "Placebo",  70,   "M",
  "SUBJ008", "DrugA",    63,   "F"
)

#==============================================================================
# Define a function with default arguments
#==============================================================================

trt_summary <- function(ds = adsl, var = age, trt) {
  ds %>%
    filter(trt01p == trt) %>%
    summarise(
      n    = n(),
      mean = mean({{ var }}),
      sd   = sd({{ var }}),
      min  = min({{ var }}),
      max  = max({{ var }})
    )
}

#==============================================================================
# Call the function for each treatment group
#==============================================================================

trt_summary(trt = "Placebo")
trt_summary(trt = "DrugA")
trt_summary(ds = adsl, var = age, trt = "Placebo")

In R, the equivalent of a SAS macro is a plain function created with function().
We declare parameters with defaults: ds = adsl and var = age mirror the SAS ds=adsl and var=age.
The trt parameter has no default, so it must be supplied on every call — just like the SAS trt= with no default value.
To pass a column name as a parameter, we use tidy evaluation with {{ var }} (curly-curly) inside dplyr verbs — this tells dplyr to treat var as a column reference, not a literal value.
The trt parameter is a plain string, so we compare it with == directly — no curly-curly needed for string comparisons.
We call the function with trt_summary(trt = "Placebo"), and R prints the resulting tibble with n, mean, sd, min, and max.
The third call explicitly names every argument, matching the SAS convention of spelling out all keyword parameters for clarity.
After running, we confirm each call returns a one-row tibble with the correct summary statistics for the requested arm.

Happy learning!

Search lessons

Introduction

Define and Call a Macro with Parameters