.-
help for ^badiag^                                                 (STB-55: sbe33)
.-

Bland-Altman plots
------------------

        ^badiag^ varname1 varname2 [^if^ exp] [^in^ range] [^, format(^str^)^ 
              ^avlab(^str^) difflab(^str^) err(^real 2^) ^graph_options]


Description
-----------
This is a variation on the well-known Bland-Altman (BA) plot.
BA plots are used in the comparison of paired data from two methods of 
measuring the same quantity (e.g. two blood pressure devices), 
where the relationship between the measurements is not in doubt, and 
emphasis is on the absolute size of the differences. Only if it is sufficiently 
small will ther mew machine be regarded as an adequate substitute for the standard.

The standard BA plot shows the average and difference between the measurements, 
(varname1 + varname2)/2 and varname2 - varname1
with 95% reference lines at mean +/- 2 SD (or 1.96 SD with the ^err^ option.)

^badiag^ is a variation of the command ^baplot^.  In both cases, 
the standard reference interval for the difference is calculated).  However, 
^badiag^ adds these to a plot of 


^badiag^ plots actual values of varname1 against varname2, with the Bland-Altman 
reference lines added.  The result is to rotate the well-known Bland-Altman plots 

Options
-------

^format(^str^)^ sets the format for the results given.

^avlab(^str^)^ gives a variable label to the average before plotting the graph.

^difflab(^str^)^  gives a variable label to the difference before plotting the
    graph.

graph_options are any of the options allowed with ^graph, twoway^.


Examples
--------

 . ^use col_icp^
 . ^badiag icp colorime, avlab("ICPOES vs Colorimetry)^


Author
------
        Paul Seed
        School of Medicine
        King's College London, UK
        email: paul.seed@@kcl.ac.uk

References
----------
Bland, J. Martin and Altman, Douglas G.  Statistical methods for assessing agreement 
between two methods of clinical measurement. Lancet, 1986; i: 307-310. 

Bland, J. Martin and Altman, Douglas G.  Measuring agreement in method comparison s
tudies. Statistical methods in medical research, 1999; 8: 135-160



Also see
--------

    STB: STB-55 sbe33