Analysis of Repeated Measures Studies with Multiple Regression Methods for Sample Survey Data
Lisa M. LaVange
and Gary G. Koch
1994, Presented
at the Drug Information Association Annual Meeting
Abstract
This
presentation discusses how recently developed statistical procedures
for fitting multiple regression models to sample survey data enables
more effective analysis for repeated measures studies with
complicated data structures. Situations where such methods are of
interest include dermatology studies where treatment is applied to
two or more sites of each patient, multi-visit studies where
responses are observed at two or more points for each patient, dental
studies where two or more teeth or dental areas of each patient
receive treatment or are monitored over time for outcomes such as
caries or progression of periodontal disease, multi-period crossover
studies, and epidemiologic studies for repeated occurrences of
adverse events or illnesses. For these situations, one can specify a
primary sampling unit within which repeated measures have intraclass
correlation. This intraclass correlation is taken into account by
sample survey regression methods through robust estimates of the
standard errors of the regression coefficients. Regression estimates
are obtained from model fitting estimating equations which ignore the
correlation structure of the data (i.e., computing procedures which
assume that all observational units are independent or are from
simple random samples). The analytic approach is straightforward to
apply with logistic models for dichotomous data, proportional odds
models for ordinal data, and linear models for continuously scaled
data, and results are interpretable in terms of population average
parameters. Several examples are presented to illustrate the
capabilities of the methodology.