[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: New version of -bmjcip- on SSC

From   "Newson, Roger B" <>
To   "''" <>
Subject   st: New version of -bmjcip- on SSC
Date   Thu, 8 Oct 2009 14:38:44 +0100

Thanks to Kit Baum, a new version of the -bmjcip- package is now available for download from SSC. In Stata, use the -ssc- command to do this, or -adoupdate- if you already have an old version.

The -bmjcip- package is described as below on my website. The new version adds a new possible value -htmlsuper- to the -esub()- option, allowing the user to recode numbers with "e+" or "e-" substrings to numbers with HTML superscripts. (The existing -esub()- possibilities included TeX and RTF superscripts.) And it also adds the ability to use input varlists of 1, 2, 3 or 4 variables. These are interpreted, respectively, as a P-value, an estimate with a P-value, an estimate with lower and upper confidence limits, or an estimate with lower and upper confidence limits and a P-value. This will be useful if the user wants to report a table of chi-squared test with P-values.

I have also added to the on-line help an example showing users how to implement the standard 20th century policy, in medical journals, of reporting P-values below 0.0005 as "<0.0005". This traditionally saved medical scientists the trouble of reporting P-values with exponents (superscripted or otherwise). However, it is probably not a good idea if you are reporting the highlights of a whole genome scan in which a million gene polymorphisms have been assessed for association with a disease, because, on average, 500 of these associations will have sample P-values less than 0.0005 by chance, even if all associations are null in the population at large. Under these standard 21st century conditions, a P-value less than 0.0005 should not be interpreted as being "as good as zero".

Best wishes


Roger B Newson BSc MSc DPhil
Lecturer in Medical Statistics
Respiratory Epidemiology and Public Health Group
National Heart and Lung Institute
Imperial College London
Royal Brompton Campus
Room 33, Emmanuel Kaye Building
1B Manresa Road
London SW3 6LR
Tel: +44 (0)20 7352 8121 ext 3381
Fax: +44 (0)20 7351 8322
Web page:
Departmental Web page:

Opinions expressed are those of the author, not of the institution.

package bmjcip from

      bmjcip: Format confidence intervals and P-values for medical journals

      bmjcip reformats a list of numeric variables, containing estimates
      and/or confidence limits and/or P-values, to string variables,
      containing the same values formatted for presentation in medical
      journals.  The new string variables replace the old numeric
      variables, and have the same names, variable labels, variable
      characteristics, and positions in the dataset.  bmjcip is usually
      used in output datasets (or resultssets) produced by the parmest
      package, and the reformatted string variables are usually later
      output using the listtex package, for input to Microsoft Word or
      other word processors.  It requires the sdecode package in order to
      work.  The sdecode, parmest and listtex packages can all be
      downloaded from SSC.
      Author: Roger Newson
      Distribution-Date: 07october2009
      Stata-Version: 11

INSTALLATION FILES                                  (click here to install)
(click here to return to the previous screen)

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index