Quality of Reporting of Randomized Controlled Trials of Pharmacologic Treatment of Bipolar Disorders: A Systematic Review
J Clin Psychiatry 2011;72(9):1214-1221
© Copyright 2014 Physicians Postgraduate Press, Inc.
Purchase This PDF for $40.00
If you are not a paid subscriber, you may purchase the PDF.
(You'll need the free Adobe Acrobat Reader.)
Receive immediate full-text access to JCP. You can subscribe to JCP online-only ($86) or print + online ($156 individual).
With your subscription, receive a free PDF collection of the NCDEU Festschrift articles. Hurry! This offer ends December 31, 2011.
If you are a paid subscriber to JCP and do not yet have a username and password, activate your subscription now.
As a paid subscriber who has activated your subscription, you have access to the HTML and PDF versions of this item.
Click here to login.
Did you forget your password?
Still can't log in? Contact the Circulation Department at 1-800-489-1001 x4 or send email
Objective: This study aimed to assess (1) the quality of reporting of randomized controlled trials of pharmacologic treatment of bipolar disorder, (2) the potential improvement in quality of reporting over time, and (3) differences in quality of reporting between journals that endorse or do not endorse the Uniform Requirements for Manuscripts Submitted to Biomedical Journals developed by the International Committee of Medical Journal Editors.
Data Sources: A systematic literature search was done to identify all randomized controlled trials published between 2000 and 2008 relevant to the pharmacologic treatment of bipolar disorder. The search strategy of the published National Institute for Health and Clinical Excellence guideline for management of bipolar disorders was used and adapted. All included and excluded clinical trials mentioned in the guideline and published from 2000 onward were reviewed for eligibility. For an update search from July 2004 through December 2008, an adapted search strategy was used in MEDLINE, EMBASE, PsycINFO, CINAHL, Ovid, and Cochrane Central Register of Controlled Trials. Titles and abstracts were scanned for relevance, and full texts were ordered in case of uncertainty to maximize sensitivity. Reference lists of retrieved systematic reviews were checked.
Study Selection: All full texts were checked for eligibility. Only relevant randomized controlled trials published between 2000 and 2008 were included. Abstracts, randomized controlled trials published before 2000, nonrandomized clinical studies, pooled analyses, editorials, reviews, case reports, observational studies, and unpublished reports were excluded.
Data Extraction: A checklist based on the Consolidated Standards of Reporting Trials (CONSORT) statement was used to assess quality of reporting of all included studies.
Results: A total of 105 randomized controlled trials were included in the analysis. Of the 72 applicable checklist items, 42% were generally reported adequately and 25% inadequately. Reporting was especially poor for randomization procedures, with, for example, 16% of studies defining generation of random allocation sequence and 15% defining method of allocation concealment. Inadequate randomization increases the potential for bias to influence the final results. Authors of clinical guidelines or health technology assessments are forced to exclude or downgrade trials with inadequate reporting on randomization. Also, information with essential clinical relevance was generally reported inadequately, such as the effect size (in 18% of studies) and the number needed to treat (in 8% of studies). Both effect measures are more important for clinicians than individual point estimates that have been reported adequately. No consistent trend could be shown for improvement in quality of reporting over time or for reporting of essential methodological items differently in journals that endorse the Uniform Requirements for Manuscripts (URM). The reporting of information on clinical relevance and generalizability of results, however, showed a consistent trend toward better reporting in journals endorsing the URM, with significant differences for the reporting of secondary outcomes (100% vs 89.9%; P = .03) and adverse events (93.2% vs 73.8%; P = .011) and interpretation of results with regard to totality of data (30.2% vs 11.5%; P = .029).
Conclusions: Our findings suggest that, while some trial-related information is well reported, a good part of the reporting quality of randomized controlled trials in bipolar disorder falls well below the required and also practically feasible level for many aspects essential for adequate interpretation of methodological quality and clinical relevance. Authors should be further encouraged to follow the CONSORT criteria.
J Clin Psychiatry
Submitted: April 7, 2010; accepted June 18, 2010.
Online ahead of print: January 25, 2011 (doi:10.4088/JCP.10r06166yel).
Corresponding author: Daniel Strech, MD, PhD, Hannover Medical School, CELLS–Centre for Ethics and Law in the Life Sciences, Institute for History, Ethics and Philosophy of Medicine, Carl-Neuberg-Strasse 1, 30625 Hannover, Germany (firstname.lastname@example.org).