|Date: ||Sat, 16 Apr 2005 22:56:39 -0400|
|Reply-To: ||"Boye, Mark" <mark.boye@PFIZER.COM>|
|Sender: ||"SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>|
|From: ||"Boye, Mark" <mark.boye@PFIZER.COM>|
|Subject: ||Estimators for 2-equation Random Effects Panel Data|
|Content-Type: ||text/plain; charset="iso-8859-1"|
I am evaluating a clinical trial with 9 sequential days of data and am
attempting to model the following: dichotomous treatment group
assignment->variation in opioid consumption -> variation in patient-reported
opioid symptoms (e.g., drowsiness, nausea, difficulty with urination, etc).
The dependent patient-reported variables are dichotomous (symptom yes/no),
count (how many daily episodes), and continuous(patient-rated symptom
Including group assignment and opioid consumption on the RHS of a single
equation would possibly introduce endogeneity issues and would also fail to
completely model this causal sequence; I therefore envision a 2-equation
solution first regressing the continuous variable opioid consumption onto
group assignment. Thereafter the second equation would incorporate the
residual to estimate the effect on my various patient-reported outcome
measures (e.g., 0/1, count, and continuous measures).
Research questions I am attempting to answer include the treatment effect on
both opioid consumption and opioid symptoms, and the dose-response effect of
opioid consumption on opioid-related symptoms.
Are there 2-equation random effect panel estimators available in SAS to fit
such models to these data? If not, are you aware of any other routines in
other software that I could use to conduct these analyses.
Alternatively, are there routines for comparing estimates of separate models
for each treatment group and then conducting cross-group comparisons on
parameters of interest?
Mark E. Boye MBA, MPH, Ph.D.
Senior Outcomes Research Scientist
WW Outcomes Research
2800 Plymouth Road
Ann Arbor, MI 48105
Unless expressly stated otherwise, this message is confidential and may be privileged. It is intended for the addressee(s) only. Access to this E-mail by anyone else is unauthorized. If you are not an addressee, any disclosure or copying of the contents of this E-mail or any action taken (or not taken) in reliance on it is unauthorized and may be unlawful. If you are not an addressee, please inform the sender immediately.