Poisson Regression Analysis of Illness and Injury Surveillance Data
E. L. Frome J. P. Watkins E. D. Ellis
Center for Epidemiologic Research
Oak Ridge Institute for Science and Education, Oak Ridge, TN, USA
C. H. Strader
U. S. Department of Energy
Date Published: December 2012
The Department of Energy's Illness and Injury Surveillance
Program(IISP) has the goal of identifying groups of workers who may be
at increased risk for occupationally related injury and illness. The
data used here are from the Worker Health Summary,
1995-2004,(WHS_IISP) Report 07-OEWH-1073 available at
DOE Office of Health, Safety and Security, or
from the DOE Information Bridge
www.osti.gov/bridge/ search term ORISE-07-OEWH-1073.
The purpose of this website is to describe the
computational details required to obtain some of the results in
the technical report PRA_IISP (Item 1 below), and additional results not in the
report. The R language and environment for statistical computing and
graphics is used for all data analysis--- see Item 3 below.
- Poisson Regression Analysis of Illness and Injury
Surveillance Data,
Technical Report ORISE-09-OEWH-0176 (2012). The
purpose of this report is to review the use of regression methods for
the analysis of count data and to describe how theses methods can be used
for the analysis of the IISP data.
- This folder [RTFMEM] contains
the results of fitting the MEM to each of the six DOE sites in the
WHS_IISP report that have contributed data to the IISP for all ten
years considered in the report. A two page summary of age-adjusted
5 day absence rates is given for each of five health events for each DOE
site, and for all sites combined.
- Information about and how to obtain the latest version of R
with links to R documentation and at least 110 books describing R and
how to use R is given [HERE---GET R]
- This link contains several R utility functions and a data
frame (drtw.rda) with the IISP RTW data that were used to obtain
numerical results in Item 1 and Item 2. The R functions and data
frame are in the folder RFIISP. Download v5iisp.zip by right clicking on the link, and
selecting "save target as"
- Unzip v5iisp.zip and move the folder
RFIISP and the file readdata.txt into a folder where you will run R,
e.g. "..\demopriisp".
-
Run R in the folder "..\demopriisp".
- This step assumes that you have R installed and have some
experince with R. If this is not the case refer to the information in
Step 3 above. Open the file readdata.txt
with a text editor and follow instruction in A or B
- Follow the directions on Lines 7 - 10 , OR
- copy
the uncommented lines into the R window.
A good text editor to use with R is
NppToR with Notepad++ [
see NppToR_Rjournal_June_2010]
- If you follow 7-A you will see the text file
PRA_IISPdemo.txt in the working directory "..\demopriisp".
This is the same as Appendix 2 in the Tech Report in Item 1
Revised 17 Dec 2011 by E.L. Frome
Revised 20 July 2012 by E.L. Frome
Revised 3 Dec 2012 by E.L. Frome
Last Revised 29 Dec 2012 by E.L. Frome