Sabrina Tulka1, Berit Geis2, Stephanie Knippschild2, Christine Baulig2, Frank Krummenauer2. 1. Institut für Medizinische Biometrie und Epidemiologie, Universität Witten/Herdecke, Alfred Herrhausen-Str. 50, 58448, Witten, Deutschland. Sabrina.Tulka@uni-wh.de. 2. Institut für Medizinische Biometrie und Epidemiologie, Universität Witten/Herdecke, Alfred Herrhausen-Str. 50, 58448, Witten, Deutschland.
Abstract
BACKGROUND: For scientific and ethical reasons randomized controlled clinical trials (RCTs) should be based on a sample size calculation. The CONSORT statement, an established publication guideline for transparent study reporting, requires a sample size calculation in every study publication. OBJECTIVE: The availability of sample size calculations in RCT publications on treatment of age-related macular degeneration (AMD) was investigated. The primary hypothesis of this investigation compared the prevalence of reported sample size calculations between journals with higher (≥5) versus lower (<5) impact factors (IF). MATERIAL AND METHODS: It was examined whether information on sample size calculation was available in a series of 97 publications of RTCs on AMD treatment published between 2004 and 2014. RESULTS: Only 46 out of 97 (47%) study publications provided information on the reason for the number of patients enrolled. The comparison of publications from journals with an IF ≥ 5 (63%, 30) and from journals with an IF < 5 (40%, 67) showed a statistically significant difference of 23% in the frequencies of available sample size calculations (95% confidence interval, CI 2%; 44%). Of the publications published before 2010, 43% reported a sample size calculation versus 51% of the publications afterwards. CONCLUSION: Publications in journals with higher IF more frequently reported a sample size calculation. More than 50% of the publications did not report any sample size calculation. Authors and reviewers of publications should pay more attention to the explicit reporting of sample size calculations.
BACKGROUND: For scientific and ethical reasons randomized controlled clinical trials (RCTs) should be based on a sample size calculation. The CONSORT statement, an established publication guideline for transparent study reporting, requires a sample size calculation in every study publication. OBJECTIVE: The availability of sample size calculations in RCT publications on treatment of age-related macular degeneration (AMD) was investigated. The primary hypothesis of this investigation compared the prevalence of reported sample size calculations between journals with higher (≥5) versus lower (<5) impact factors (IF). MATERIAL AND METHODS: It was examined whether information on sample size calculation was available in a series of 97 publications of RTCs on AMD treatment published between 2004 and 2014. RESULTS: Only 46 out of 97 (47%) study publications provided information on the reason for the number of patients enrolled. The comparison of publications from journals with an IF ≥ 5 (63%, 30) and from journals with an IF < 5 (40%, 67) showed a statistically significant difference of 23% in the frequencies of available sample size calculations (95% confidence interval, CI 2%; 44%). Of the publications published before 2010, 43% reported a sample size calculation versus 51% of the publications afterwards. CONCLUSION: Publications in journals with higher IF more frequently reported a sample size calculation. More than 50% of the publications did not report any sample size calculation. Authors and reviewers of publications should pay more attention to the explicit reporting of sample size calculations.
Entities:
Keywords:
CONSORT statement; Clinical trials; Quality of publications; Study planning; Transparent reporting
Authors: David Moher; Sally Hopewell; Kenneth F Schulz; Victor Montori; Peter C Gøtzsche; P J Devereaux; Diana Elbourne; Matthias Egger; Douglas G Altman Journal: BMJ Date: 2010-03-23