David J Biau1, Raphaël Porcher, Isabelle Boutron. 1. INSERM U717; Departement de Biostatistique et Informatique Médicale, Hôpital Saint-Louis-Université Paris 7, Paris, France. djmbiau@yahoo.fr
Abstract
OBJECTIVE: To systematically review the account of center and provider effects in large surgical and interventional randomized controlled trials. STUDY DESIGN: A systematic review of multicenter interventional randomized trials of 200+ patients. The search included Medline from 1 January 2000 through 11 October 2005 and a hand search of the bibliographies of retrieved articles. One author abstracted all data using standardized abstraction forms; a second reviewer assessed a random sample of reports as quality-assurance procedure. RESULTS: Sixty-eight reports met inclusion criteria. The trials predominantly reported on cardiology (n=23, 34%). The number of participating providers was reported in 11 trials (16%). Both the performed control and performed interventional procedures were described in 43 trials (63%). The use of stratified random allocation on center was reported in 26 trials (38%) and on provider in 6 trials (9%). The analysis was adjusted for center in four trials (6%) and for provider in three trials (4%). CONCLUSIONS: Only few trials account for center or provider effect in the design and analysis. Authors and journal editors could play an important role in improving the reporting of trials.
OBJECTIVE: To systematically review the account of center and provider effects in large surgical and interventional randomized controlled trials. STUDY DESIGN: A systematic review of multicenter interventional randomized trials of 200+ patients. The search included Medline from 1 January 2000 through 11 October 2005 and a hand search of the bibliographies of retrieved articles. One author abstracted all data using standardized abstraction forms; a second reviewer assessed a random sample of reports as quality-assurance procedure. RESULTS: Sixty-eight reports met inclusion criteria. The trials predominantly reported on cardiology (n=23, 34%). The number of participating providers was reported in 11 trials (16%). Both the performed control and performed interventional procedures were described in 43 trials (63%). The use of stratified random allocation on center was reported in 26 trials (38%) and on provider in 6 trials (9%). The analysis was adjusted for center in four trials (6%) and for provider in three trials (4%). CONCLUSIONS: Only few trials account for center or provider effect in the design and analysis. Authors and journal editors could play an important role in improving the reporting of trials.