APA_BrCa_RAM: (A)sian-(P)acific-(A)merican (BrCa) (R)isk (A)ssessment (M)acro
Installing the Asian-Pacific-American BrCa Risk Assessment Macro Ver 1.0 05/12/11
"Projecting Individualized Absolute Invasive Breast Cancer Risk in Asian and
Pacific Islander American Women" Matsuno, Costantino, Ziegler et al. JNCI 2011
This module pertains to Windows 2000, Windows XP, Windows Vista and Windows 7
operating systems for desktop/laptop PCs with installed SAS 9 software.
Contents of the Zip file "APA_BrCa_RAM.zip":
(a) APA_BrCa_RAM sas macro which performs abs risk and 95% CI calculations
for Asian-Pacific-American women based on the Asian-American
Breast Cancer Study (AABCS) Relative Risk (RR) model.
the first 350 lines of the sas macro "APA_BrCa_RAM" contains
extensive documentaion/clarification on the use and operation
of this macro.
(b) APA_BrCa_example.sas a simple demonstration sas program which reads in raw data,
involkes the sas macro "APA_ABrCa_RAM" and obtains abs risk
and 95% CI.
the first 30 lines of this sas program contains a brief
description of what this program accomplishes. additional
clarification is scattered throughout the program.
(c) APA_RR.fil raw data file for input to "APA_BrCa_example.sas"
(d) APA_BrCa_example.out sample output from "APA_BrCa_example.sas"
(e) ReadMe.fil this file you are reading
After unzipping "APA_BrCa_AbsRsk.zip", check to see that all files listed above are
included. At this point, a cursory examination of "APA_BrCa_RAM", "APA_BrCa_example.sas"
and "APA_RR.fil" will prove helpful. These files can be veiwed using your favorite text
editor or Notepad which is included with Windows.
Installation and operation of the sas macro "APA_BrCa_RAM":
(a) unzip the entire contentes of the zip file "APA_BrCa_RAM.zip" into a folder on
your hard drive.
(b) execute the provided demonstration sas program "APA_BrCa_example.sas".
(c) compare the output "APA_BrCa_example.lst" with the provided file
"APA_BrCa_example.out". The two files should be identical. If so, congratulations,
you have successfully installed the software.
Characteristics of the input sas file for the Asian-Pacific-American Risk Assessment SAS Macro:
(a) the input sas file name and output sas file name are provided by the user upon the
invocation of the sas macro "APA_BrCa_RAM".
(b) the input sas file must contain the following 9 quantities. any additional variables
on the input sas file are ignored. No particular ordering is required.
(1) ID unique ID (non-negative integer 1,2,3....9999999)
(2) current age in years in the age interval [20, 90) e.g. 35.8
(3) projctn age in years in the age interval (current age, 90] e.g. 58.9
constraint of: 20 =< current age < projctn age <= 90
(4) # of breast biopsies (non-negative integer counts 1,2,3...)
(5) age at menarche in years (non-negative integer years)
(6) age of first live birth in years (non-negative integer years)
constraint of: age menarche <= age at first live birth
(7) # of first degree relatives with breast cancer
(non-negative integer counts 1,2,3...)
(8) biopsy displays atypical hyperplasia?
0 = no
1 = yes
99 = unk or not applicable response
IF # breast biopsies = 0, hyperplasia MUST BE 99 (not applicable)
IF # breast biopsies > 0, hyperplasia can be 0,1 or 99 (unknown)
(9) race of women
1=Chinese 2=Japanese 3=Filipino 4=Hawaiian 5=Other Pacific Islander 6=Other Asian
for records where race is not 1 or 2 or 3 or 4 or 5 or 6
projected abs risk defaults to the sas missing value "." .
more detailed documentation is provided in the first 350 lines of the SAS macro
in the above [) represents an interval closed on the left and open on the right
(] represents an interval open on the left and closed on the right
this program will categorize the raw values of # breast biopsies, age menarche,
age first live birth and # 1st degree relatives with brca into risk categories
according to the rules outlined in the JNCI 2011 manuscript listed at the beginning
of this document.
note that the program, being an analytical tool, will make the most rudimentary
checks for logical consistency between the input variables. It is the user's
responsibility to make sure that all input covariate values are logically
consistent and correct within each person. As a final step, the program
will generate a missing projected Absolute Risk for records which it deems to
be in error. The user should examine the output file for any records with a
missing Absolute Risk and rectify the error(s) for these records.
e.g. if NBiops equals zero than atypical hyperplasia must be 99 (not applicable)
or Projection Age in years must be greater than Current Age
Dr. Mitchell Gail firstname.lastname@example.org Scientific/Statistical issues
Mr. David Pee email@example.com Technical/software problems
Terms & Conditions
Source Code for APA model: Asian-Pacific American Women Breast Cancer Risk Calculation Module
Non-Proprietary Software Transfer Agreement
Provider: National Institutes of Health (NIH), National Cancer Institute (hereinafter "NCI")
By accepting Software, Recipient agrees to the terms of this Agreement.
The Software transferred under this agreement was created by Federal Government employees in
the course of official duties. Transfer of Software to Recipient does not constitute
endorsement by the NCI of the Recipient or any product, service or company and no endorsement
should be inferred. SOFTWARE IS NOT INTENDED FOR TREATING OR DIAGNOSING HUMAN SUBJECTS.
The APA model risk calculator may be updated periodically as new data or research becomes
available. Software is supplied AS IS, without any accompanying services or improvements from
NIH. SOFTWARE IS SUPPLIED TO RECIPIENT WITH NO WARRANTIES, EXPRESS OR IMPLIED, INCLUDING ANY
WARRANTY OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. NCI makes no representations
that the use of Software will not infringe any patent or proprietary rights of third parties.
All risk as to quality and performance of Software is with Recipient. In no event will the
United States Government or NCI be liable to Recipient for damages arising out of the use or
inability to use Software, including but not limited to loss of data or data being rendered
inaccurate or losses sustained by Recipient or third parties