Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: ANCOVA for pre post designs

From   David Airey <>
Subject   st: ANCOVA for pre post designs
Date   Tue, 23 Dec 2003 17:12:39 -0600

This is a question for the biostatisticians on the list.

I'm thinking of formulating a commentary on accepted research procedures in my area that I think could be improved by observing basic statistical arguments presented to researchers by biostatisticians.

It has been suggested that in a randomized clinical trial design with baseline (B) and followup (F) test measures comparing a control and treatment group (G), performing an ANOVA on the ratio pre/post is the worst choice of the 4 ways to deal with baseline differences:

(1) post: analyze F by G
(2) difference: analyze F-B by G
(3) ratio: analyze F/B by G
(4) ancova: analyze F = constant + b1*B + b2*G, for G differences

In light of biostatisticians' suggestion (e.g., Vickers, BMC Medical Research Methodology (2001) 1:6, that method (4) above is preferred most and method (3) is least preferred, does it apply to "prepulse inhibition" literature?

In an area of schizophrenia research, subjects show a deficit in basic sensorimotor gating, as measured by prepulse inhibition of the acoustic startle response. The startle response is simply startle to a loud noise. Prepulse inhibition is simply inhibition of that startle response by preceding the loud noise with a soft noise. In both non-schizophrenics and schizophrenics, startle is comparable, but prepulse inhibition is _less_ in schizophrenics. That is, both groups startle comparably to a loud noise, but schizophrenics startle less when a startling noise is preceded by a soft noise. So, there are two brain circuits underlying this behavior and the prepulse inhibition circuit is compromised in schizophrenics.

Across the board, prepulse inhibition papers generally employ method (3). That is, each subject is measured with trials for startle, and also (at the same testing session) with trials for prepulse inhibition, and then a ratio is formed for that subject 100*(startle-prepulse)/(startle) to represent the percent prepulse inhibition relative to startle. That percent (%PPI) is then analyzed as the response measure. I'm thinking the baseline measure discuss at the top of this email is analogous to the startle trial response, and the followup is analogous to the prepulse trial response. A difference is that both trial types are given pre and post to any (e.g., drug) intervention. An additional complication is that the %PPI ratio is analyzed in the context of both between-subject designs and within-subject designs.

Is method (4) applicable to prepulse inhibition studies, and is it applicable to both between and within designs?

By way of example, let's suppose we have the following between-subjects data:

group, startle, prepulse, ppi

where group indicates a between-subjects design where one group had a placebo and another group had the drug. Then, method (4) says to analyze group difference by the regression (ANCOVA) model, prepulse = constant + b1*startle + b2*group, and in Stata, this would be "anova prepulse group startle, continuous(startle)". The usual ANOVA on %PPI by method (3) would be "anova ppi group".

Next we could have the same question but using within-subjects data:

subject, startle, prepulse, ppi, treatment

where treatment indicates that each subject is exposed to the placebo and drug at different times in random order. Is it also possible to use a within-subjects design with the advantages of method (4) incorporated? Usually, one would do "anova ppi subject treatment" on %PPI. Does the model "anova prepulse subject treatment startle, continuous(startle)" make sense?


* For searches and help try:

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