Reporting
conflicts of interest may limit
credibility of research
BMJ
is hard pill to swallow, even for
graduates
Abstract
reasoning makes faster work for
editors
Quarter
of NHS nonconsultant posts are for
non-standard grades
 Reporting
conflicts of interest may limit
credibility of researchReaders
may be more likely to consider research
where a conflict of financial and
academic interests has been declared
as less credible, according to an
RCT by Samena Chaudhry and colleagues.
170 readers of the BMJ read a manuscript
detailing the effects on daily functioning
of pain caused by skin disorder
herpes zoster. Those who were led
to believe that the authors were
from an ambulatory care centre with
no competing interests rated the
paper significantly more favourably
than those who believed the authors
to be employees and potential shareholders
in a fictitious pharmaceutical company.
Believability, interest, importance,
relevance and the validity of the
paper were all rated significantly
lower when the authors were believed
to have competing interests.
[Full
Text Paper PDF]
Back
to top >
 Hard
pill to swallow, even for graduates
A
study by Anna Ellis, former Student
BMJ editor, suggests that the level
of English used in the BMJ may be
beyond the grasp of all but the
brightest university students. Investigators
assessed 20 randomly selected Education
& Debate articles published
in the journal between April 2001
and March 2003 against FOG and FLESCH
classifications to determine the
average ‘reading level’
of the articles. A number of criteria
were used including average length
of sentence, number of difficult
words and total number of syllables
in the text. The findings show that
when compared against normative
reading levels, the BMJ articles
scored on average ‘very difficult’
- the top category for reading ease
purportedly beyond the reading level
of a college graduate. Readability
of the articles, which was found
to get more difficult as the article
progresses, was not significantly
improved by editing. The authors
of the paper stress that the findings
should not be taken too literally.
“Readability is a complex
interplay of factors including interest,
layout and technicalities of the
text”.
Back
to top >
 Abstract
reasoning makes faster work for
editors A
recent experimental study in BMJ
editorial has shown that editors
can make a reliable decision to
reject or send to referee 60% of
submitted manuscripts based on the
abstract alone. The study showed
that in virtually all cases, the
decision to reject or referee manuscripts
based on reading the abstract was
not changed following full reading
of the manuscript which increased
the average time required to make
a decision by 200%. Decisions based
on abstracts took between 1 and
nine minutes compared with between
1 and 25 minutes for full manuscript
reading. The results of the study
suggest that the requirement for
editors to be increasingly ‘quick
and brutal’ in rejecting manuscripts
does not compromise the quality
of editorial decision-making. Paper
published by ESE.
[Full
Text Paper PDF]
Back
to top >
 Quarter
of NHS nonconsultant posts are for
non-standard grades
Nearly
a quarter of all nonconsultant medical
posts advertised in BMJ Careers
are for non-standard grades, according
to a recent study by Sabina Dosani
and colleagues. The study found
that an average of 22% of posts
in BMJ Careers, which advertises
nearly all of the available hospital
posts in the UK, were for non-standard
grades, often contravening the European
Working Time Directive. The survey
found that the posts are created
to meet NHS service requirements
when there is insufficient deanery
funding for recognised training
posts. Advertisers expect the posts
to be filled by overseas graduates
from non-EEA countries who may not
recognise that the positions are
for non-standard grades. The BMJ
Research study is the first to document
the number of non-standard training
grades available in the NHS which
are not included in the Department
of Health’s annual census.
[Full
Text Paper PDF]
Back
to top >
Leanne Tite,
BMJ
Research
in progress >>
|
