Editorial

Clinical Trial Registration

in Physiotherapy
Journals:
Recommendations from
the International Society
of Physiotherapy Journal
Editors
Leonardo O.P. Costa, Chung-Wei Christine Lin, Debora Bevilaqua Grossi, Marisa
Cota Mancini, Anne K. Swisher, Chad Cook, Dan Vaughn, Mark R. Elkins, Umer
Sheikh*, Ann Moore, Gwendolen Jull, Rebecca L. Craik, Christopher G. Maher, Rinaldo
Roberto de Jesus Guirro, Amlia Pasqual Marques, Michele Harms, Dina Brooks, Guy
G. Simoneau, John Henry Strupstad

Clinical trial registration involves placing importantly, however, it tackles two big
the protocol for a clinical trial on a free, problems in clinical research: selective
publicly available and electronically reporting and publication bias.
searchable register. Registration is Selective reporting involves investigators
considered to be prospective if the protocol only reporting the most favourable results
is registered before the trial commences when they publish a trial, instead of
(i.e., before the first participant is reporting the results for all the outcomes
enrolled). Prospective registration has that were measured. Reporting only
several potential advantages. It could help favourable outcomes can create a
avoid trials being duplicated unnecessarily misleading appearance of the effect of a
and it could allow people with health therapy in the published literature. For
problems to identify trials in which they example, imagine that a completely
might participate. Perhaps more ineffective intervention is tested across

[Journal of Physiotherapy & Sports Medicine][Volume 1][Issue 2][December 2012] Page 60

several trials and each trial measures the apparent effect of an intervention
multiple outcomes. Most outcomes will across the published data. For example, a
show no significant effect of the trial in which the intervention appeared to
intervention. However, occasionally an be effective may be published, while the
outcome will show significant benefit or three other trials in which the intervention
harm simply by chance. If the researchers appeared ineffective or harmful languish in
publish the positive outcomes but not all of the filing cabinets of the investigators. If a
the non-significant and negative outcomes, trial is registered but never published,
readers could interpret falsely that the authors of a systematic review can still
intervention is beneficial. A similar find the trial on the register and contact the
problem could occur when outcomes are authors to request the unpublished data for
analysed at multiple time points. inclusion in the review. Therefore,
Researchers may report that an prospective registration of clinical trials
intervention improves walking speed at 6 could further limit bias affecting the body
months, but fail to mention that it does not of evidence that is available in published
improve walking speed at 1, 2, 3, 9, 12 and physiotherapy trials.
24 months. Prospective registration of Prospective clinical trial registration
1
clinical trials combats this problem in encourages transparency and may also
several ways. Journal editors and make it more difficult for fraudulent
reviewers can compare the range of authors to fabricate data. For example,
outcomes reported in a manuscript against some journals now ask for individual
those listed in the registered protocol, patient data to be provided routinely for
requesting that any discrepancies be checking2 or audit data when fraud is
resolved by following the protocol. suspected3. Data collection should have
Readers can also compare the outcomes in occurred during the dates of data collection
the registered protocol against those in the defined on the registry. Because many
published report, taking greater outcomes are measured and stored
reassurance when they are consistent. electronically with date stamps, this would
Publication bias arises when trials with increase the planning and complexity
positive results are more likely to be involved in fabricating data, especially if
published than trials with non-significant the fabricated data are to withstand the
or negative results. Like selective scrutiny of an audit. Also, researchers who
reporting, this can also spuriously inflate obtain unwelcome data from a particular

[Journal of Physiotherapy & Sports Medicine][Volume 1][Issue 2][December 2012] Page 61

subgroup of patients may be tempted to subject (p.3)6. Some ethics committees
eliminate it by retrospectively introducing have made trial registration a condition of
an additional exclusion criterion. If their ethical approval.
protocol has been prospectively registered, Although some physiotherapy journals
however, this would be publicly evident to have also encouraged clinical trial
anyone who compared the registered registration7,8,9, only about 6% of the
protocol and the report of the trial. randomised trials investigating the effects
of physiotherapy interventions published
1. How common is clinical trial in 2009 had been registered
registration? prospectively10. In an attempt to rectify

The first major register for healthcare trials this situation, this editorial recommending

was established in 19984. Although prospective registration has been co-

thousands of trials were soon registered, authored by several members of the

the majority of trials remained International Society of Physiotherapy

unregistered. In 2004, clinical trial Journal Editors (ISPJE). The remainder of

registration was endorsed by the the editorial will: define which trials

International Committee of Medical should be registered; explain how

Journal Editors (ICMJE)4. In addition to researchers can register their trials;

endorsing clinical trial registration, announce tougher policies about clinical

member journals of the ICMJE made trial registration that are being adopted by

prospective registration compulsory for all some member journals of the ISPJE; and

clinical trials that commenced participant identify who can contribute to ensuring

recruitment after 1 July 20054. Many other that clinical trial registration achieves its

journals also endorsed clinical trial potential benefits.

registration and the number of registered

trials increased rapidly5. Since then, many
organisations have added their support for
clinical trial registration. For example, in
2008 the World Medical Association
included a new item on the Declaration of
Helsinki stating that Every clinical trial
must be registered in a publicly accessible
database before recruitment of the first
2. Which trials should be
registered?
Any clinical trial should be prospectively
registered before the first participant is
recruited into the study. The World Health
Organization defines clinical trials as any
research study that prospectively assigns
human participants or groups of humans to

[Journal of Physiotherapy & Sports Medicine][Volume 1][Issue 2][December 2012] Page 62

one or more health-related interventions to comparison(s) studied, study hypotheses,
evaluate the effects on health outcomes11. primary and secondary outcomes,
eligibility criteria, sample size, blinding,
3. How can I register my trial? funding, principal investigators and dates
Clinical trial registration should be quick, of commencement and anticipated
easy and free of charge. Many clinical trial completion of the study. It is common for
registries have been established, including trial registries to review the information
some that focus on a particular disease (eg, for completeness and clarity, so some
Internet Stroke Center Trials Registry, editing might be needed. The registry will
www.strokecenter.org/trials) or then provide a unique trial registration
geographical region (eg, Pan African number to the researchers. This number
Clinical Trials Registry, www.pactr.org). should be included in all reports of the
Researchers often choose to register their trials results as a link to the registered
trials in their countrys national register, protocol for editors, reviewers and readers.
although this is not compulsory. It is more Prospective registration can be done any
important that researchers choose a time before the first participant is
registry that elicits and documents all the recruited. Many researchers wait until
relevant content from the original protocol immediately before recruitment starts, so
(outlined below) and that has satisfactory that any late changes to the protocol (such
quality, validity, accessibility, unique as alterations requested by an ethics
identification, technical capacity and committee) do not necessitate an
administration. To assist researchers, the amendment to the registry entry. Although
World Health Organization maintains a list not ideal, protocol amendments are
of registries that meet these criteria sometimes made after recruitment starts.
(http://www.who.int/ictrp/network/primary These should be updated on the registered
/en/index.html). Currently 16 registries are protocol as well. The trial registry will
listed. Among these, researchers could publicly document what changed and on
choose one that processes applications what date.
swiftly or that allows communication
using their native language. When 4. ISPJE member journals
registering their protocol, researchers will introducing mandatory
be asked to provide information such as prospective registration
descriptions of the intervention(s) and policies

[Journal of Physiotherapy & Sports Medicine][Volume 1][Issue 2][December 2012] Page 63

The executive of the ISPJE strongly participant recruitment after 1 January
recommends that member journals adopt a 2009. The following table lists other
policy of mandatory prospective member journals and their nominated dates
registration for all clinical trials. Several to implement mandatory prospective
member journals are implementing such clinical trial registration, as well as the
policies. Physical Therapy has already trials that this policy applies to (based on
implemented a policy of mandatory the commencement date of participant
prospective clinical trial registration, recruitment).
which applies to trials that commenced
ISPJE Journal Dates to Implement Prospective Clinical Trial
Registration

Name of the Journal Mandatory Registration Start of Recruitment Date
Date*
Brazilian Journal of Physical 1 January 2014 31 December 2013
Therapy/Revista Brasileira de
Fisioterapia
Cardiopulmonary Physical 1 January 2015 1 January 2014
Therapy Journal
Journal of Manual and 1 January 2014 6 June 2013
Manipulative Therapy
Journal of Physiotherapy 1 January 2013 1 January 2006
Journal of Physiotherapy & 1 January 2014 1 June 2013
Sports Medicine
Manual Therapy 1 January 2014 1 June 2013
Physical Therapy & 1 January 2014 31 December 2013
Research/Fisioterapia e
Pesquisa
Physiotherapy 1 January 2013 1 January 2013
Physiotherapy Canada 1 January 2013 1 January 2013
The Journal of Orthopaedic & 1 January 2013 1 January 2013
Sports Physical Therapy
Tidsskriftet Fysioterapeuten/ 1 January 2014 1 July 2013
Norwegian Journal of
Physiotherapy
*The date after which prospective clinical trial registration becomes mandatory.

This policy applies to trials that commence participant recruitment after this date.

5. Who else can help ensure that clinical trial registration achieves its
clinical trial registration potential benefits. Everyone in the

achieves its potential profession can ensure that their colleagues

are aware of clinical trial registration and
benefits?
its importance. Educators should ensure
In addition to the recommendations for
that the research component of
researchers and editorial boards outlined
physiotherapy training programs explains
above, others can contribute to ensuring

[Journal of Physiotherapy & Sports Medicine][Volume 1][Issue 2][December 2012] Page 64

the importance of trial registration. International Society of Physiotherapy
Journal Editors
Clinicians can also advise or help patients Leonardo O P Costa and Chung-Wei Christine
Lin
to search trial registers to identify relevant
trials for which the patient might Brazilian Journal of Physical
Therapy/Revista Brasileira de
volunteer. Administrators of clinical trial Fisioterapia
Debora Bevilaqua Grossi and Marisa Cota
registries that do not meet the WHO Mancini
criteria can strive to attain this status.
Cardiopulmonary Physical Therapy
Grant review panels can make funding Journal
Anne K Swisher
contingent upon prospective registration
for proposed clinical trials. More ethics Journal of Manual and Manipulative
Therapy
review committees can make their Chad Cook and Dan Vaughn
approval of trials contingent upon
Journal of Physiotherapy
prospective registration as well. However, Mark R Elkins

even universal prospective registration Journal of Physiotherapy & Sports

may make no difference to selective Medicine
Umer Sheikh
reporting and publication bias unless there
Manual Therapy
is an expectation that protocols will be Ann Moore and Gwendolen Jull
compared to published reports before
Physical Therapy
publication. Therefore, journal editors and Rebecca L Craik and Christopher G Maher
peer reviewers must remember to check
Physical Therapy &
for discrepancies between submitted Research/Fisioterapia e Pesquisa
manuscripts and registry entries. Rinaldo Roberto de Jesus Guirro and Amlia
Pasqual Marques
Physiotherapy clinical trials that are
conducted and reported according to a pre- Physiotherapy
Michele Harms
specified protocol are more likely to
provide credible information than those Physiotherapy Canada
Dina Brooks
that do not. Prospective clinical trial
The Journal of Orthopaedic & Sports
registration is therefore of great potential Physical Therapy
value to the clinicians, consumers and Guy G Simoneau

researchers who rely upon clinical trial Tidsskriftet Fysioterapeuten/Norwegian

Journal of Physiotherapy
data and that is why ISPJE is John Henry Strupstad
recommending that members enact a
policy for prospective trial registration.

[Journal of Physiotherapy & Sports Medicine][Volume 1][Issue 2][December 2012] Page 65

References 7. Askie L, Ghersi D, Simes J. Prospective
registration of clinical trials. Aust J Physiother.
1. Sim I, Chan AW, Gulmezoglu AM, Evans T,
2006;52:237-239.
Pang T. Clinical trial registration: transparency
8. Costa LO, Maher CG, Moseley AM, Sherrington
is the watchword. Lancet. 2006;367:1631-1633.
C, Herbert RD, Elkins MR. Editorial:
2. Herbert RD. Researchers should make data
endorsement of trial registration and the
freely available. Aust J Physiother. 2008;54:3.
CONSORT statement by the Revista Brasileira
3. Smith J, Godlee F. Investigating allegations of
de Fisioterapia. Rev Bras Fisioter. 2010;14:V-
scientific misconduct - Journals can do only so
VI.
much; institutions need to be willing to
9. Harms M. Clinical trial registration. Physiother.
investigate. BMJ. 2005;331:245-246.
2011;97:181.
4. de Angelis C, Drazen JM, Frizelle FA, et al.
10. Pinto RZ, Elkins MR, Moseley AM, et al.
Clinical trial registration: a statement from the
Comparison of registry entries and published
International Committee of Medical Journal
reports of randomised trials: an audit of 200
Editors. N. Engl. J. Med. 2004;351:1250-1251.
published trials. Phys Ther. Accepted
5. Laine C, Horton R, DeAngelis CD, et al. Clinical
12/9/2012.
trial registration - Looking back and moving
11. World Health Organization. International
ahead. N. Engl. J. Med. 2007;356:2734-2736.
Clinical Trials Registry Platform (ICTRP) 2012.
6. World Medical Association Declaration of
Available from http://www.who.int/ictrp/en/
Helsinki. Ethical principles for medical research
Accessed 09/07/2012.
involving human subjects: World Medical
Association Declaration of Helsinki. 2008:1-5.
Available online at www.pgip.co.uk/jopsm
POSTGRADUATE INSTITUTE OF
PHYSIOTHERAPY

* Corresponding author. E-mail address: u.sheikh@pgip.co.uk (Umer Sheikh)

2012 PGIP. All rights reserved.

[Journal of Physiotherapy & Sports Medicine][Volume 1][Issue 2][December 2012] Page 66