Article type
Year
Abstract
Background: Bias in controlled trials research has been examined in a number of meta-epidemiological studies.(1-6), and findings from these studies have contributed to improving the quality of RCTs and meta-analyses. We have combined these studies to create a database that will be used to examine associations between trial characteristics and treatment effects, and to provide a resource for future work. Since the original studies were done independently, the extent of overlap of included systematic reviews and RCTs is unknown.
Objectives: By assigning a unique identity for each unique review and RCT we aimed to identify duplicates, and to describe characteristics of reviews and their included trials.
Methods: Data were combined into an Access database. Unique identifiers were assigned for each systematic review and each trial using the identification number assigned by cataloguing databases. For this, the hierarchy of searching was MEDLINE followed by EMBASE. References indexed in neither MEDLINE nor EMBASE were hand-checked for duplicates and unique studies were assigned a new identification number. Duplicate reviews and trials in the combined database were identified. Results: The combined data-set comprised 296 meta-analyses, derived from 286 systematic reviews. Separate meta-analyses from the same systematic review were not considered to be duplicates. Two hundred and sixty two meta-analyses occurred only once, while 17 were duplicated at least once. The majority of the duplicates were from the Cochrane Database of Systematic reviews (CDSR). There were 3468 references for included RCTs, among which there were 2948 unique references (Table 1). Of these, 2585 occurred only once while 363 were duplicated at least once. The majority of these were in duplicated systematic reviews. Seventy-six per cent of the included RCTs were indexed in MEDLINE, 5% in EMBASE only, while the remaining 19% were not indexed in either.
Conclusions: There was surprisingly little duplication across the included meta-epidemiological studies. This large database of systematic reviews, meta-analyses and RCTs will be used to examine the epidemiology of bias in clinical trials.
References: 1. Schulz KF, Chalmers I, Hayes RJ, Altman DG. Empirical evidence of bias - Dimensions of methodological quality associated with estimates of treatment effects in controlled trials. J Am Med Assoc. Feb 1995;273 (5):408-412. 2. McAuley L, Pham B, Tugwell P, Moher D. Does the inclusion on grey literature influence estimates of intervention effectiveness reported in meta-analyses? Lancet. Oct 2000;356(9237):1228-31 3. Kjaergard LL, Villumsen J, Gluud C. Reported methodological quality and discrepancies between large and small randomized trials in meta-analyses. Ann Intern Med. Dec 2001; 135:982-989. 4. Jüni P, Holenstein F, Sterne J, Bartlett C, Egger M. Direction and impact of language bias in meta-analyses of controlled trials: empirical study . Int J Epidemiol. In press 2002. 5. Sampson M, Barrowman NJ, Moher D, Klassen TP, Pham B, Platt R et al. Should meta-analysts search Embase in addition to Medline? J Clin Epidemiol. Oct 2003;56(10):943-55. 6. Royle P, Milne R. Literature searching for randomised controlled trials used in Cochrane reviews: rapid versus exhaustive searches. Int J Technol Assess Health Care. Fall 2003;19(4):591-603.
Objectives: By assigning a unique identity for each unique review and RCT we aimed to identify duplicates, and to describe characteristics of reviews and their included trials.
Methods: Data were combined into an Access database. Unique identifiers were assigned for each systematic review and each trial using the identification number assigned by cataloguing databases. For this, the hierarchy of searching was MEDLINE followed by EMBASE. References indexed in neither MEDLINE nor EMBASE were hand-checked for duplicates and unique studies were assigned a new identification number. Duplicate reviews and trials in the combined database were identified. Results: The combined data-set comprised 296 meta-analyses, derived from 286 systematic reviews. Separate meta-analyses from the same systematic review were not considered to be duplicates. Two hundred and sixty two meta-analyses occurred only once, while 17 were duplicated at least once. The majority of the duplicates were from the Cochrane Database of Systematic reviews (CDSR). There were 3468 references for included RCTs, among which there were 2948 unique references (Table 1). Of these, 2585 occurred only once while 363 were duplicated at least once. The majority of these were in duplicated systematic reviews. Seventy-six per cent of the included RCTs were indexed in MEDLINE, 5% in EMBASE only, while the remaining 19% were not indexed in either.
Conclusions: There was surprisingly little duplication across the included meta-epidemiological studies. This large database of systematic reviews, meta-analyses and RCTs will be used to examine the epidemiology of bias in clinical trials.
References: 1. Schulz KF, Chalmers I, Hayes RJ, Altman DG. Empirical evidence of bias - Dimensions of methodological quality associated with estimates of treatment effects in controlled trials. J Am Med Assoc. Feb 1995;273 (5):408-412. 2. McAuley L, Pham B, Tugwell P, Moher D. Does the inclusion on grey literature influence estimates of intervention effectiveness reported in meta-analyses? Lancet. Oct 2000;356(9237):1228-31 3. Kjaergard LL, Villumsen J, Gluud C. Reported methodological quality and discrepancies between large and small randomized trials in meta-analyses. Ann Intern Med. Dec 2001; 135:982-989. 4. Jüni P, Holenstein F, Sterne J, Bartlett C, Egger M. Direction and impact of language bias in meta-analyses of controlled trials: empirical study . Int J Epidemiol. In press 2002. 5. Sampson M, Barrowman NJ, Moher D, Klassen TP, Pham B, Platt R et al. Should meta-analysts search Embase in addition to Medline? J Clin Epidemiol. Oct 2003;56(10):943-55. 6. Royle P, Milne R. Literature searching for randomised controlled trials used in Cochrane reviews: rapid versus exhaustive searches. Int J Technol Assess Health Care. Fall 2003;19(4):591-603.