Issues in use of SAS PROC.MIXED to test the significance of treatment effects in controlled clinical trials

Journal of Biopharmaceutical Statistics

Volumn 10, Issue 2, 2000, Pages 265-286

  Ahn, Chul ^a,b   Tonidandel, Scott ^a   Overall, John E ^a  

^a UNIVERSITY OF TEXAS   (United States)

^b UNIVERSITY OF TEXAS MEDICAL SCHOOL   (United States)

Author keywords

Clinical trials; Covariate controls; Error structure; Mixed models; Modeling dropouts; Power; Repeated measures; Tests of significance

Indexed keywords

CLINICAL RESEARCH; MEASUREMENT; PRIORITY JOURNAL; REVIEW; SCORING SYSTEM; STATISTICAL ANALYSIS; STATISTICAL MODEL; STATISTICAL PARAMETERS;

ALGORITHMS; COMPUTER SIMULATION; CONTROLLED CLINICAL TRIALS; MODELS, STATISTICAL; NONLINEAR DYNAMICS; PATIENT DROPOUTS; RESEARCH DESIGN; SOFTWARE; TREATMENT OUTCOME;

EID: 0033819067     PISSN: 10543406     EISSN: None     Source Type: Journal    
DOI: 10.1081/BIP-100101026     Document Type: Review

References (6)

1. Problematic formulations of SAS PROC.MIXED models for repeated measurements
2. Estimating sample sizes for repeated measurements designs
3. How many repeated measurements are useful?
4. Power of univariate and multivariate analyses of repeated measurements in controlled clinical trials