Article Figures & Data
Tables
Characteristics No. Note: Randomized controlled trials reported in Annals of Internal Medicine, British Medical Journal, The Journal of the American Medical Association, The New England Journal of Medicine, and The Lancet between September 2006 and February 2007. a Results total to more than 48, as several trials were excluded for more than 1 reason. Excluded trialsa 48 No, incorrect, or not publicly accessible trial registration database recorded in the main publication 11 Registry did not record a primary outcome, or the outcome recorded was too vague to use 20 Publication was not the main report of the trial results 23 Included trials Information extracted: study reference and copy of study, trials registration reference and copy of registration, primary and secondary outcomes in the final report, primary and secondary outcomes in the trials registration database 110 Journal Included Trials Excluded Trials Total Annals of Internal Medicine 12 3 15 British Medical Journal 20 11 31 The Journal of the American Medical Association 15 7 22 The Lancet 19 12 31 The New England Journal of Medicine 44 15 59 Total 110 48 158 Database or Registry No. of Reports Comments Note: Eleven reports did not list a registration database. Australian Clinical Trials Registry 2 No longer active. Merged into Australian New Zealand Clinical Trials Registry (ANZCTR: http://www.anzctr.org.au) Managed by NHMRC Clinical Trials Centre, University of Sydney Clinicaltrials.gov (http://www.clinicaltrials.gov) 112 Managed by US National Institutes of Health European Clinical Trials Database (http://eudract.emea.europa.eu) 1 Not publicly searchable. Managed on behalf of the European Commission International Standard Randomised Controlled Trial Number Register (http://www.isrctn.org) 30 Managed by Current Controlled Trials, Ltd (http://www.controlled-trials.com). Also contains a searchable meta-register of other trials databases National Research Register 1 No longer active. Archive available at: http://portal.nihr.ac.uk/Pages/NRRArchive.aspx. Managed by UK National Health Service – National Institute for Health Research Study Characteristic No. % (95% CI) Note: Studies could have multiple changes. Primary outcomes No change 76 69 (60–77) Change 34 31 (23–40) Change by addition 10 – Change by promotion from secondary 3 – Change by deletion 20 – Change by demotion to secondary 6 – Secondary outcomes No change 33 30 (22–39) Change 77 70 (61–78) Change by addition 54 – Change by demotion from primary 5 – Change by deletion 48 – Change by promotion to primary 3 – Primary Outcomes Secondary Outcomes Study Description Registry Publication Registry Publication Notes ER=extended release; 5-FU=5-fluorouracil; NICU = neonatal intensive care unit. Kaplan et al9,10 1. Patient perception of treatment benefit at week 12 1. Reported 24 outcomes grouped in several domains 7 outcomes, 2 not mentioned as out- comes in registry Several clinically important secondary outcome measures listed in the registry (continence, overactive bladder, erectile function) are not mentioned in the publication Safety and efficacy of tolterodine ER, tamsulosin, or the combination, against placebo in men with lower urinary symptoms 2. Treatment benefit, treatment satisfaction, and willingness to continue treatment questions at week 12 2. Not reported as a primary outcome but some measures reported as secondary outcomes Bosset et al11,12 1. Disease-free survival 1. Not reported None listed None listed Disease-free and progression-free survival both discussed in the publication, but not as outcomes Treatment of rectal cancer with radio- therapy and 5-FU 2. Overall survival 2. Reported Michalowicz et al13,14 1. Gestational age at birth 1. Reported None listed 1. Birth weight Demotion of 1 primary outcome and addition of multiple secondary outcomes Effect of treatment of periodontal disease on the risk of pregnancy complications 2. Birth weight 2. Reported as a secondary outcome 2. Infants small for gestational age 3. Apgar scores 4. Admissions to NICU
Additional Files
The Article in Brief
Undisclosed Changes in Outcomes in Randomized Controlled Trials: An Observational Study
Robert Ewart , and colleagues
Background Changing the outcomes of a clinical trial, without explanation, calls its validity into question. This study examines how often outcomes are changed between trial registration and final publication.
What This Study Found Authors frequently change the outcomes of clinical trials between trial registration and publication without explanation or disclosure. Analyzing 110 randomized controlled trials published in consecutive issues of five major medical journals in a six-month period and their associated trial registry entries, researchers found that a primary outcome had been changed in 34 (31 percent) of the articles and a secondary outcome was changed in 77 (70 percent) without disclosure.
Implications
- The authors assert that allowing researchers to change their outcomes without reason or disclosure threatens the underpinnings of the scientific enterprise and the ability of practicing physicians to apply published research to patient care.
Jump to section
Related Articles
Cited By...
- Estimating the prevalence of discrepancies between study registrations and publications: a systematic review and meta-analyses
- Hidden changes to prespecified primary outcomes of clinical trials completed between 2009 and 2017 in German University Medical Centres: A meta-research study
- Peer reviewed evaluation of registered end-points of randomised trials (the PRE-REPORT study): protocol for a stepped-wedge, cluster-randomised trial
- Systematic review protocol assessing the processes for linking clinical trial registries and their published results
- Many Randomized Trials of Physical Therapy Interventions Are Not Adequately Registered: A Survey of 200 Published Trials
- Despite Law, Fewer Than One In Eight Completed Studies Of Drugs And Biologics Are Reported On Time On ClinicalTrials.gov
- Frequency and reasons for outcome reporting bias in clinical trials: interviews with trialists
- Discrepancies between research advertisements and disclosure of study locations in trial registrations for USA-sponsored research in Russia
- In This Issue: Multifaceted Research