BACKGROUND: Careful consideration of site effects is important in the analysis of multi-site clinical trials for drug abuse treatment. The statistical choices for modeling these effects have implications for both trial planning and interpretation of findings. OBJECTIVES: Three broad approaches for modeling site effects are presented: omitting site from the analysis; modeling site as a fixed effect; and modeling site as a random effect. Both the direct effect of site and the interaction of site and treatment are considered. METHODS: The statistical model, and consequences, for each approach are presented along with examples from existing clinical trials. Power analysis calculations provide sample size requirements for adequate statistical power for studies utilizing 6, 8, 10, 12, 14, and 16 treatment sites. RESULTS: Results of the power analyses showed that the total sample required falls rapidly as the number of sites increases in the random effect approach. In the fixed effect approach in which the interaction of site and treatment is considered, the required number of participants per site decreases as the number of sites increases. CONCLUSIONS: Ignoring site effects is not a viable option in multi-site clinical trials. There are advantages and disadvantages to the fixed effect and random effect approaches to modeling site effects. SCIENTIFIC SIGNIFICANCE: The distinction between efficacy trials and effectiveness trials is rarely sharp. The choice between random effect and fixed effect statistical modeling can provide different benefits depending on the goals of the study.
BACKGROUND: Careful consideration of site effects is important in the analysis of multi-site clinical trials for drug abuse treatment. The statistical choices for modeling these effects have implications for both trial planning and interpretation of findings. OBJECTIVES: Three broad approaches for modeling site effects are presented: omitting site from the analysis; modeling site as a fixed effect; and modeling site as a random effect. Both the direct effect of site and the interaction of site and treatment are considered. METHODS: The statistical model, and consequences, for each approach are presented along with examples from existing clinical trials. Power analysis calculations provide sample size requirements for adequate statistical power for studies utilizing 6, 8, 10, 12, 14, and 16 treatment sites. RESULTS: Results of the power analyses showed that the total sample required falls rapidly as the number of sites increases in the random effect approach. In the fixed effect approach in which the interaction of site and treatment is considered, the required number of participants per site decreases as the number of sites increases. CONCLUSIONS: Ignoring site effects is not a viable option in multi-site clinical trials. There are advantages and disadvantages to the fixed effect and random effect approaches to modeling site effects. SCIENTIFIC SIGNIFICANCE: The distinction between efficacy trials and effectiveness trials is rarely sharp. The choice between random effect and fixed effect statistical modeling can provide different benefits depending on the goals of the study.
Authors: Kathleen M Carroll; Chris Farentinos; Samuel A Ball; Paul Crits-Christoph; Bryce Libby; Jon Morgenstern; Jeanne L Obert; Doug Polcin; George E Woody Journal: J Subst Abuse Treat Date: 2002-09
Authors: Robert E Booth; Barbara K Campbell; Susan K Mikulich-Gilbertson; Carrie J Tillotson; Dongseok Choi; James Robinson; Donald A Calsyn; Raul N Mandler; Lindsay M Jenkins; Laetitia L Thompson; Catherine L Dempsey; Michael R Liepman; Dennis McCarty Journal: AIDS Behav Date: 2011-01
Authors: John F Kelly; Robert L Stout; Leonard A Jason; Nilofar Fallah-Sohy; Lauren A Hoffman; Bettina B Hoeppner Journal: Alcohol Clin Exp Res Date: 2020-02-03 Impact factor: 3.455
Authors: Joanna Jacobus; Charles T Taylor; Kevin M Gray; Lindsay R Meredith; Anna M Porter; Irene Li; Norma Castro; Lindsay M Squeglia Journal: Drug Alcohol Depend Date: 2018-04-11 Impact factor: 4.492
Authors: Joshua L Roffman; Eren D Sipahi; Kevin F Dowling; Dylan E Hughes; Casey E Hopkinson; Hang Lee; Hamdi Eryilmaz; Lee S Cohen; Jodi Gilman; Alysa E Doyle; Erin C Dunn Journal: PLoS One Date: 2021-04-28 Impact factor: 3.240
Authors: Joshua D Lee; Edward V Nunes; Patricia Novo Mpa; Genie L Bailey; Gregory S Brigham; Allan J Cohen; Marc Fishman; Walter Ling; Robert Lindblad; Dikla Shmueli-Blumberg; Don Stablein; Jeanine May; Dagmar Salazar; David Liu; John Rotrosen Journal: Contemp Clin Trials Date: 2016-08-10 Impact factor: 2.226
Authors: Katie Witkiewitz; John W Finney; Alex H S Harris; Daniel R Kivlahan; Henry R Kranzler Journal: Alcohol Clin Exp Res Date: 2015-08-11 Impact factor: 3.455
Authors: Katie Witkiewitz; John W Finney; Alex H S Harris; Daniel R Kivlahan; Henry R Kranzler Journal: Alcohol Clin Exp Res Date: 2015-08-06 Impact factor: 3.455
Authors: Benedetto Vitiello; Mark A Riddle; Gayane Yenokyan; David A Axelson; Karen D Wagner; Paramjit Joshi; John T Walkup; Joan Luby; Boris Birmaher; Neal D Ryan; Graham Emslie; Adelaide Robb; Rebecca Tillman Journal: J Am Acad Child Adolesc Psychiatry Date: 2012-07-31 Impact factor: 8.829
Authors: Benjamin G Druss; Manasvini Singh; Silke A von Esenwein; Gretl E Glick; Stephanie Tapscott; Sherry Jenkins Tucker; Cathy A Lally; Evelina W Sterling Journal: Psychiatr Serv Date: 2018-02-01 Impact factor: 3.084