Eric Manheimer1, Jeanette Ezzo, Victoria Hadhazy, Brian Berman. 1. University of Maryland School of Medicine, Center for Integrative Medicine, Kernan Hospital Mansion, Baltimore, MD 21207, USA. emanheimer@compmed.umm.edu
Abstract
BACKGROUND: Systematic reviewers generally evaluate randomized controlled trials (RCTs) based on the published reports. We evaluated whether the description of methods in the published reports is an accurate and complete reflection of study procedures used. METHODS: The authors of 51 RCTs included in a systematic review of acupuncture for chronic pain were sent a brief survey that included questions related to the following three important study quality dimensions: (1) generation of allocation sequence, (2) allocation concealment, and (3) blinding of outcomes assessor. RESULTS: We received 35 of 51 responses for an overall response rate of 68.6%. Of 35 studies described as randomized in published reports, associated survey responses indicated that four actually used quasi-randomized methods. Among published reports with missing information on these quality dimensions, 27 of 32 studies used adequate methods for the generation of allocation sequence, 13 of 34 used adequate allocation concealment and 2 of 10 were blinded, according to survey responses. Survey responses generally confirmed information about randomization and blinding already described in investigators' RCT publications. CONCLUSION: Surveying RCT investigators uncovered some information about study quality dimensions not described in published reports.
BACKGROUND: Systematic reviewers generally evaluate randomized controlled trials (RCTs) based on the published reports. We evaluated whether the description of methods in the published reports is an accurate and complete reflection of study procedures used. METHODS: The authors of 51 RCTs included in a systematic review of acupuncture for chronic pain were sent a brief survey that included questions related to the following three important study quality dimensions: (1) generation of allocation sequence, (2) allocation concealment, and (3) blinding of outcomes assessor. RESULTS: We received 35 of 51 responses for an overall response rate of 68.6%. Of 35 studies described as randomized in published reports, associated survey responses indicated that four actually used quasi-randomized methods. Among published reports with missing information on these quality dimensions, 27 of 32 studies used adequate methods for the generation of allocation sequence, 13 of 34 used adequate allocation concealment and 2 of 10 were blinded, according to survey responses. Survey responses generally confirmed information about randomization and blinding already described in investigators' RCT publications. CONCLUSION: Surveying RCT investigators uncovered some information about study quality dimensions not described in published reports.
Authors: Maria E Suarez-Almazor; Carol Looney; Yanfang Liu; Vanessa Cox; Kenneth Pietz; Donald M Marcus; Richard L Street Journal: Arthritis Care Res (Hoboken) Date: 2010-09 Impact factor: 4.794
Authors: Anna Enblom; Anna Johnsson; Mats Hammar; Gunnar Steineck; Sussanne Börjeson Journal: Evid Based Complement Alternat Med Date: 2011-06-20 Impact factor: 2.629