Sidelines 15 (3)
Steve Duffy,
Kate Light, Lisa Mather and Vickie Orton
Centre for Reviews and Dissemination
University of York
Web:
http://www.york.ac.uk/inst/crd
The Sidelines abstracts in this
issue of Inform focus on developments in clinical trial reporting,
the potential additional morbidity data that may be generated from
NHS Direct, a survey into the use of complementary medicine in the
UK, and research into the effectiveness of the internet as a
resource for answering consumers’ health questions.
Cooper, D., Chinemana, F..
(2004) “NHS Direct derived data: an exciting new opportunity or an
epidemiological headache?”, Journal of Public Health, vol. 26,
no. 2., pp. 158-60.
NHS Direct, a national telephone
helpline for health advice, was established in 1998 to provide health
information and advice to callers and refer them to an appropriate
service. This article discusses the NHS Direct call record, and how
the data collected can be used for disease surveillance and
epidemiological purposes.
NHS Direct nurses use clinical
decision support software to handle calls to NHS Direct. Callers are
classified (triaged) on the basis of their symptoms using this
software. Computerized call records contain information on the
caller's name, age, address, sex, time of call, symptoms and
disposition (the outcome of the call), as well as a more detailed
clinical description. Ethnicity has been included in this list since
Summer 2003.
The call reporting system was
originally designed for performance management purposes, however the
data collected can be used to provide an overview of symptoms in the
community. Calls can also provide information on the most common age
of the caller, peak times of calls, geographical distribution of
callers, and other factors.
There are, however, some
limitations to the data, including the fact that NHS Direct calls are
a fraction of GP contact rates, so may not yet reflect the health of
all sections of the population. Other issues are that NHS Direct was
implemented in four waves between 1998-2000, so there will be some
geographical variation in data collected, that calls are re-routed to
other call centres at busy times, that commuters may call from their
place of work, thus elevating calls in urban areas, and that the age
distribution of calls is heavily skewed towards children.
Despite the issues outlined in
this article, the authors believe that the value of NHS Direct derived
data for public health purposes is likely to increase as NHS Direct
expands, as staff become more experienced and as the technical
infrastructure of NHS Direct becomes more advanced.
Steinbrook, R. (2004) “Registration of clinical trials – voluntary or
mandatory?”, New England Journal of Medicine, vol. 351, no. 18,
pp. 1820-1822.
As the US government moves towards
making the public registration of clinical trials mandatory this
article discusses whether this is helpful or not. It raises issues
about the disclosure of sensitive company information. The article
suggests that public trials registers are already in place and that
pharmaceutical companies may be reluctant to conduct trials if the
information will become publicly available to their competitors before
completion of the trial.
Since the New York Attorney
General successfully sued GlaxoSmithKline for withholding negative
information about the antidepressant paroxetine, the debate has gained
impetus. In response to growing concerns the Pharmaceutical Research
and Manufacturers of America (PhRMA) has set up a new trials database
http://www.clinicalstudyresults.org.
In September 2004 the
International Committee of Medical Journal Editors (ICMJE) announced
that member journals will require companies to register trials at or
before the onset of patient recruitment as a condition of
publication. No particular database has been recommended but trials
must provide the minimum data as required on clinicaltrials.gov
The Fair Access to Trials Act,
proposed in the US, would require mandatory registration of trials.
However, it has no co-sponsors in the House or Senate, so legislation
may depend on the outcome of the election.
Moher, D., Altman, D.G.,
Schulz, K.F., Elbourne, D.R. (2004) “Opportunities
and challenges for improving the quality of reporting clinical
research: CONSORT and beyond”, Canadian Medical Association
Journal, vol. 171, no. 4, pp. 349-350.
With the announcement by the ICMJE
(International Committee of Medical Journal Editors) that it will
require all trials to be registered at or before the recruiting stage
for submissions to be accepted for publication, this paper looks at
the role of CONSORT (Consolidated Standards of Reporting Trials).
In 1995 a series of studies found
that trials may be biased when trials use inferior methods or are
reported without adequate descriptions of methods. The studies also
concluded that failure to conceal the allocation of participants to
treatment groups is associated with an exaggeration of the
effectiveness of an intervention of 30% or more.
This led to the drafting of the
CONSORT recommendations. Three studies have reported that the quality
of reports of RCTs appearing in journals that have backed the CONSORT
statement are of superior quality than those in journals that have not
adopted the statement. But the studies indicated that all journals
were still publishing inadequate reports of RCTs of and that more
effort needs to be made to improve standards.
Since the publication of CONSORT,
several related initiatives have been developed including QUOROM for
systematic reviews of randomized trials, MOOSE for systematic reviews
of observational studies, STARD for diagnostic tests and guidelines
for reports of controlled trials of botanical medicines and STROBE for
observational studies.
The paper concludes that resources
must be spent on improving reporting so that the results of RCTs can
become the foundation of evidence-based health care.
Abbasi, K. (2004) “Compulsory registration
of clinical trials: will be a requirement before submission to the BMJ
from July 2005”, BMJ, vol. 329, pp. 637-639.
In 1999 the BMJ and the Lancet
stated that “the case for registering all clinical trials – first
advanced a decade ago- is now unanswerable”. To this end the BMJ
experimented with an “amnesty” for unpublished trials, but with little
success. It also considered compulsory registration but felt that the
trials registries available were too diverse and lacked the necessary
organisation.
Since Eliot Spitzer, a New York
attorney general, won a landmark case against GlaxoSmithKline
demanding disclosure of previously undisclosed trial data on
paroxetine in adolescents, the move for reform has gathered pace. Many
international medical journals are now demanding that researchers
organising trials recruiting patients after 1st
July 2005 must register the trial in a public trials register, before
their work will be considered for publication. Any researchers
recruiting participants to trials before 1st
July 2005 must register the trial before 13th
September 2005 to be considered for publication.
The article notes that decisions
by journal editors may have a slight impact, and the reaction of
industry will be watched with interest. How many companies will risk
missing the opportunity to publish favourable results in a prestigious
journal?
Thomas, K., Coleman, P. (2004) “Use of complementary
or alternative medicine in a general population in Great Britain.
Results from the National Omnibus survey”, Journal of Public
Health, vol. 26, no. 2, pp. 152-157.
This paper presents results from a
survey of a representative sample of adults in Great Britain
interviewed about the use of complementary or alternative medicine
(CAM).
The sample was derived from the
National Omnibus Survey carried out in 2 out of every 3 months each
quarter, by the Office for National Statistics (ONS). In 2001 8
questions relating to CAM use were added to the survey interview
schedule.
On analysis 10% of the population
reported using CAM in the previous 12 months. Most users reported
using more than one CAM therapy. CAM use is similar across England,
Scotland and Wales, and men and women reported using CAM in equal
proportions. CAM use is lowest in the youngest and oldest age groups,
and over half of CAM users had not told their GP or health care
provider about visiting CAM practitioners. The study also looked at
reasons for receiving CAM therapy and reasons for choosing particular
CAM practitioners.
The authors recognise that the
small sample size has limited their results. They suggest the need
for a population sample 4 times greater than this sample to enable
better sub-group analysis, and recommend that the CAM question module
should be repeated over at least 4 consecutive ONS Omnibus surveys.
Spink, A., Yang, Y.,
Jansent, J., et al.
(2004) “A study of medical and health
queries to web search engines”, Health Information and Libraries
Journal, vol. 21, pp. 44-51.
The authors of this study
investigated the characteristics and topics of medical and
health-related web searching by examining queries made to three
commercial search engines: Excite, AlltheWeb.com and Ask Jeeves. In
addition, a more detailed study of the nature of advice-seeking was
carried out using the Ask Jeeves search engine.
A large part of the paper,
however, reports trends in the use of the Internet for health-related
information. For example, while there has been an increase in the use
of health information web sites, there has been a corresponding
decline in the use of commercial search engines to identify
health-related information. Analysis of the web logs for both Excite
and AlltheWeb.com confirmed a decrease in medical and health-related
queries when compared with an earlier survey.
Web searching for health topics is
shown to be comparable to other web searching in terms of length,
complexity and lack of query re-formulation. This suggests that users
are heavily reliant upon the efficacy of the search engine to retrieve
relevant information.
The subjects of web queries
submitted to Ask Jeeves and Excite are briefly discussed and
categorised into very broad headings such as pregnancy, general health
and human relationships. Although there are several noticeable
differences between the topics submitted to the two search engines,
the authors do not discuss reasons for these differences.
The examination of Ask Jeeves, in order to study health
advice-seeking, revealed that queries were often submitted as though
to a person, so demonstrating that some users failed to understand web
search processes.
The authors conclude that
commercial search engines’ use of natural language may be problematic
and that users would retrieve information more effectively by using
precise clinical terms and specialised vocabulary.
Back to Inform 15 (3)
| 
