Joe Abramson has developed an augmented Windows version (WINPEPI) of the PEPI suite of programs for epidemiologists. You can read a description of WINPEPI (published in Epidemiologic Perspectives & Innovations) by clicking http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3041648/pdf/1742-5573-8-1.pdf A list of currently available functions and procedures is available here.
The latest update (version 11.50) was posted to this site on August 8, 2015..
The downloaded file will be WinPepiSetup.exe. Alternatively, you can download it in a ZIP archive :
which you will have to open to access WinPepiSetup.exe.
When you run WinPepiSetup.exe, it will install the latest WINPEPI programs and manuals, and an index to the statistical procedures and programs, in any folder you select. Older versions will be replaced. Follow the instructions, and your desktop will have a portal to the programs and manuals, the index, and relevant Internet sites. No changes are made to the Windows registry.
As a third alternative, if you do not have permission to use an installation program (but can download executable files in a zip file), download WinPepiFiles.zip, which contains all the components of WINPEPI, but no installation program. You will have to open this zip file and copy its contents to a directory of your choice (overwriting any previous versions), and manually put a shortcut to winpepi.exe (the portal program) on your desktop.
All the programs are 32-bit applications, and will run in any version of Microsoft Windows except Windows 3.
A report from the McAfee Site Advisor confirms that "This link is safe. We tested it and didn't find any significant security issues”..
Years of potential life years lost due to mortality
Years of potential productive years lost due to disability
Critical prior interval, for Bayesian appraisal of an association
[An error in module S6 of COMPARE2, which exaggerated the sample sizes required for a comparison of means, has been corrected.]
LESS RECENT ADDITIONS:
DALY (disability-adjusted life years).
Tests for conformity with Poissson (random) distribution.
Test for compatibility of counts.
Peto odds ratio, for use in meta-analyses
Tolerance intervals and tolerance bounds
Confidence interval for difference between Youden indices
Kaplan-Meier and risk-difference curves for comparison of survival data (e.g. in a clinical trial).
To avoid problems with administrator rights in recent versions of Windows, all results are now automatically saved in PEPI.TXT in the Winpepi folder, instead of in C:\PEPI.TXT, where older results may be found.
The currently available WINPEPI programs are :
COMPARE2.EXE - For comparisons of two independent groups or samples. The current version number is 3.60.
DESCRIBE.EXE - For use in descriptive epidemiology including the appraisal of separate samples in comparative studies. The current version number is 3.02.
ETCETERA.EXE - Miscellaneous (randomization, random sampling, adjustment of multiple-test p-values, appraisal of synergism, assessment of a scale, correlation-coefficient tools, large contingency tables, three-way tables, median polish and mean polish, appraisal of effect of unmeasured confounders). The current version number is 3.04.
LOGISTIC.EXE - Multiple logistic regression. The current version number is 1.54.
PAIRSetc.EXE - For appraisal of differences and agreement between matched samples or observations. The current version number is 3.47.
POISSON.EXE - Multiple Poisson regression. The current version number is 1.27.
WHATIS.EXE - An expression evaluator (with storage of constants, interim results, and formulae) and calculator for p values (and their inverse), confidence intervals, and time spans. The current version number is 4.60.
For a paper describing the programs, see Abramson,J.H. WINPEPI updated: computer programs for epidemiologists, and their teaching potential. Epidemiologic Perspectives & Innovations 2011, 8:1 (can be accessed via the WinPepi portal, and is available on the Internet at www.epi-perspectives.com/content/8/1/1). The programs and their uses are also described in Encyclopedia of Research Design (ed. N.J. Salkind), vol. 3, Sage Publications, 2010, 1633-1636. A description was also published in the Epidemiology Monitor 2007 28(Oct.): 1-7, with an update in the March 2009 issue. There is a brief and usually out-of-date description in Wikipedia.
Please report any problems you experience or any suggestions you may have to improve the package. E-mail your comments &c.