Background The internal and external validity of randomised controlled trials require detailed analyses of modes of patient recruitment, description of the settings and locations of the study and the individual center effects as described in the CONSORT statement (1,2). However, most of the trials published in major rheumatology journals lack this information.
Objectives We surveyed the reporting of the eligibility criteria for participants, the settings and locations where the data were collected according to CONSORT and the center effects on outcomes in randomised clinical trials in major rheumatology journals.
Methods A hand search was made for randomised controlled trials published in 2011 and 2012 in Ann Rheum Dis, Arthritis Rheum, Rheumatology, J Rheumatol, and Clin Exp Rheumatol. The items were separately tabulated for eligibility criteria for participants such as referral or self selection, the settings and locations of the study such as country, city; the immediate environment the study was performed in such as community, office practice, hospital clinic, inpatient unit, primary, secondary and tertiary health care and other factors related to the setting exemplified by climate, physical factors, economics, geography, social and cultural milieu (race and education), and problems with transportation. The individual center effects on outcomes in the main text were also explored. Two observers separately evaluated the papers and the disagreements were resolved by consensus.
Results We evaluated 102 studies including four with childhood diseases. The following data were available for referral patterns: physician referrals (n=5), self selection (n=6) or both (n=2); locations: country (n=54), continent (n=12), and local city (n=31); settings: community (n=2), hospital clinic (n=27), inpatient unit (n=2); secondary (n=1) or tertiary health care centers (n=13); socio-economics (n=6), educational levels (n= 5), and race (n=42). No data were given for the remaining items. Multicenter differences were only reported in 1 of 24 papers in Ann Rheum Dis, 8 of 28 in Arthritis Rheum, 1 of 6 in Rheumatology, 2 of 11 in J Rheumatol, and 1of 3 in Clin Exp Rheumatol.
Conclusions A few randomised controlled trials give enough information on eligibility criteria for participants and the settings and locations as well as the center effects. These cast doubt on the robustness of the results. More data are needed to apply these outcomes to other patient populations.
Schulz et al. CONSORT 2010 Statement: updated guidelines for reporting paralel group randomised trials. BMJ 2010;340:c332.
Moher et al. CONSORT 2010 Explanation and Elaboration: updated guidelines for reporting parallel group randomised trials. BMJ 2010;340:c869
Disclosure of Interest None Declared