Statistical models appropriate for designs often used in group‐randomized trials

Group-randomized trials are characterised by the allocation of identifiable groups rather than individuals to study conditions; members within those groups are then observed to assess the effect of the intervention. It is convenient to categorize the designs employed in group-randomized trials along two dimensions, each with two levels. The first distinguishes between designs having just one or two time intervals and those having three or more intervals. The second distinguishes between nested cohort and nested cross-sectional designs. Following a brief review of the design and analytic issues common to group-randomized trials, and their general solutions, this paper presents the adaptations of the mixed-model analysis of covariance and random coefficients models that are required for the four combinations that result from this categorization scheme. The assumptions, strengths and weaknesses of each model are discussed.

[1]  R J Carroll,et al.  On design considerations and randomization-based inference for community intervention trials. , 1996, Statistics in medicine.

[2]  D M Murray,et al.  A Monte Carlo Study of Alternative Responses To Intraclass Correlation in Community Trials , 1996, Evaluation review.

[3]  A. Kristal,et al.  Selected methodological issues in evaluating community-based health promotion and disease prevention programs. , 1992, Annual review of public health.

[4]  Russell D. Wolfinger,et al.  Analysis Issues in the Evaluation of Community Trials - Progress Toward Solutions in Sas/Stat Mixed , 1994 .

[5]  D M Murray,et al.  Gauss or Bernoulli? , 1996, Evaluation review.

[6]  J. Ware,et al.  Random-effects models for serial observations with binary response. , 1984, Biometrics.

[7]  P J Hannan,et al.  Assessing intervention effects in the Minnesota Heart Health Program. , 1994, American journal of epidemiology.

[8]  David M. Murray,et al.  Design and Analysis of Group- Randomized Trials , 1998 .

[9]  R M Elashoff,et al.  Analysis of longitudinal data: random coefficient regression modelling. , 1994, Statistics in medicine.

[10]  B. Short,et al.  Intraclass correlation among measures related to tobacco use by adolescents: estimates, correlates, and applications in intervention studies. , 1997, Addictive behaviors.

[11]  H A Feldman,et al.  Cohort versus cross-sectional design in large field trials: precision, sample size, and a unifying model. , 1994, Statistics in medicine.

[12]  Anthony S. Bryk,et al.  Hierarchical Linear Models: Applications and Data Analysis Methods , 1992 .

[13]  R. Wolfinger,et al.  Generalized linear mixed models a pseudo-likelihood approach , 1993 .

[14]  A Donner,et al.  Linear regression analysis with repeated measurements. , 1984, Journal of chronic diseases.

[15]  D. Harville.  Maximum Likelihood Approaches to Variance Component Estimation and to Related Problems , 1977 .

[16]  David M. Zucker,et al.  An Analysis of Variance Pitfall: The Fixed Effects Analysis in a Nested Design , 1990 .

[17]  D M Murray,et al.  Planning for the appropriate analysis in school-based drug-use prevention studies. , 1990, Journal of consulting and clinical psychology.

[18]  Jerome Cornfield,et al.  SYMPOSIUM ON CHD PREVENTION TRIALS: DESIGN ISSUES IN TESTING LIFE STYLE INTERVENTIONRANDOMIZATION BY GROUP: A FORMAL ANALYSIS , 1978 .

[19]  E H Wagner,et al.  Data analysis and sample size issues in evaluations of community-based health promotion and disease prevention programs: a mixed-model analysis of variance approach. , 1991, Journal of clinical epidemiology.

[20]  D. Murray,et al.  Intraclass Correlation among Measures Related to Alcohol Use by School Aged Adolescents: Estimates, Correlates and Applications in Intervention Studies , 1996, Journal of drug education.

[21]  N. Breslow,et al.  Approximate inference in generalized linear mixed models , 1993 .

[22]  R. Bosker.  Boekbespreking van "A.S. Bryk & S.W. Raudenbusch - Hierarchical linear models: Applications and data analysis methods" : Sage Publications, Newbury Parki, London/New Delhi 1992 , 1995 .

[23]  Mitchell H. Gail,et al.  Community Intervention Trial for Smoking Cessation (COMMIT): I. cohort results from a four-year community intervention. , 1995, American journal of public health.

[24]  J Cornfield,et al.  Randomization by group: a formal analysis. , 1978, American journal of epidemiology.

[25]  Chris J. Skinner,et al.  Analysis of complex surveys , 1991 .

[26]  B. Short,et al.  Intraclass correlation among measures related to alcohol use by young adults: estimates, correlates and applications in intervention studies. , 1995, Journal of studies on alcohol.

[27]  N. Laird,et al.  Maximum likelihood computations with repeated measures: application of the EM algorithm , 1987 .

[28]  M. Gail,et al.  Community intervention trial for smoking cessation (COMMIT): II. Changes in adult cigarette smoking prevalence. , 1995, American journal of public health.

[29]  R. Wolfinger.  Covariance structure selection in general mixed models , 1993 .

[30]  D M Murray,et al.  Analysis of data from group-randomized trials with repeat observations on the same groups. , 1998, Statistics in medicine.

[31]  Ita G. G. Kreft,et al.  Multilevel Analysis Methods , 1994 .

[32]  P. McCullagh,et al.  Generalized Linear Models, 2nd Edn. , 1990 .

[33]  S. Kelder,et al.  Statistical design of the Child and Adolescent Trial for Cardiovascular Health (CATCH): implications of cluster randomization. , 1995, Controlled clinical trials.

[34]  M. Feinleib,et al.  Statistical Models for Longitudinal Studies of Health , 1992 .

[35]  J. Ware,et al.  Random-effects models for longitudinal data. , 1982, Biometrics.

[36]  H A Feldman,et al.  Statistical design of REACT (Rapid Early Action for Coronary Treatment), a multisite community trial with continual data collection. , 1998, Controlled clinical trials.

[37]  J. Ware.  Linear Models for the Analysis of Longitudinal Studies , 1985 .