Intended for healthcare professionals

Rapid response to:

Editor's Choice

Pills, thrills, and bellyaches

BMJ 2005; 330 doi: https://doi.org/10.1136/bmj.330.7488.0-h (Published 17 February 2005) Cite this as: BMJ 2005;330:0-h

Rapid Response:

Decline of BMJ...more

Decline of BMJ

Mark Struthers claims my figures contain a distortion due to his
postings about the Middle East: and that this "will most certainly have
skewed the figures on rational responding," [1] and that this "must surely
account for the massive decline in BMJ activity since the beginning of
this year. I can’t see that charging for access is a good idea either."
[1] I agree that using the search engine figures provide an only crude
figure of BMJ activity. But it must show 'something'...a something that
has stayed steady in 1999-2003 and has then declined since last July. I
think many ups and downs in the year can have distorted the figures, but
these would tend to cancel each other out over time. The data may contain
many distortions and do not also forget the impact of the censorship
issue…more rapid responses must have been received since September that
were not posted due to them being defamatory or offensive in some way.
This policy must have contributed something to the observed ‘decline.’ It
is hard to quantify this aspect.

Herbert Nehrlich claims I "(and others) attacked Richard Smith when
he was editor. Now you attack him for saying that things are better since
he has left.” [2] That is not strictly true. I did not ‘attack’ him…how
could I? I would not even know who he was if he rode past me on his
bicycle, let alone attack him.

What I did was to criticise him when he published a disgraceful
obituary to David Horrobin and then compounded his error by refusing to
retract or amend it, and refusing to apologise for the upset he had caused
to the Horrobin family. Then he continued to pay the same person to write
further BMJ obituaries. I was not alone in thinking that such a gross
error of judgement was and is unforgivable.

Nor do the criticisms I have recently made are said because he has
gone, because he is a living being called ‘Richard Smith’ or because he
was once BMJ editor. They are made simply and solely because the comment
he made seemed woefully inappropriate, if not downright inaccurate. And
here below are the figures to support my point.

Here is a summary of the search engine figures:

Year

Mean items per month

% of 2002

1999

198.75

81.39932

2000

237.6667

97.33788

2001

239.5833

98.12287

2002

244.1667

100

2003

216.75

88.77133

2004

182

74.53925

Mean

219.8194

90.02842

What this data shows is that compared to the years 2000-2002, then
2003 and 2004 are below par and the gap seems to be widening.

Here is a second table showing further data from the search engine
info:

Month/year Search engine data 1999-2003

monthly mean
% downturn

Jun-04

205

231.8333

11.57441

Jul-04

187

215.1667

13.09063

Aug-04

146

177.5

17.74648

Sep-04

145

206

29.61165

Oct-04

156

203.6667

23.40426

Nov-04

81

179

54.7486

Dec-04

118

225.5

47.67184

Jan-05

82

224.3333

63.44725

Feb-05

30

188.6667

84.09894

       

What this data shows is that overall ‘BMJ activity’ has been
declining since June 2004 to the tune of first 10-20% [June to August] of
the 1999-2003 averages, then to 20-30% [Sept and Oct] and more recently by
50-84% [November to Feb].

Here is the corresponding data for recent rapid responses:

 Month/year

Mean rapid

responses


per article

% of

Nov 04
Downturn as

% of Nov 04

Oct-04

4.55

100%

0

Nov-04

4.26

93.62637

6.373626

Dec-04

4.117356253

90.49135

9.508654

Jan-05

1.698038022

37.31952

62.68048

Feb-05

1.406749079

30.91756

69.08244

This table shows that the number of rapid responses per article, in
each of the months shown, has declined from 4.55 in November to around 1.4
this month. This in turn represents about 30% of the November amount and
hence a 69% reduction in the BMJ rapid responses since that date. The
additional fact that for January 2005 the percentage reduction in rapid
responses [62.68%] matches almost exactly the percentage reduction shown
by the search engine data [63.44%] illustrates that something real is
being measured by these figures and that something is in steep decline.

I have provisionally suggested that the cause of this reduction is
the unilaterally introduced removal of email addresses from the site
[since October] and the ‘pay to view’ decision for articles from January
onwards. The data certainly supports such a view. The biggest drop has
been since December and hence it seems safe to conclude that it is the
‘pay to view’ policy that is the bigger of those two possible causes.
Whether it is the whole truth of the matter is, however, for others to
judge.

Several people have called for the email addresses to be put back on
the site and some also for the ‘pay to view’ policy to be reconsidered or
rescinded. If BMJ wishes to recover its steep collapse in popularity then
it might well be advised to seriously consider reviewing these recently
introduced changes to the site, which appear to be losing the journal the
support of the worldwide web-browsing public.

Sources

[1] Mark Struthers, Thrills, spills and happy pills 25 February 2005

[2] Dr. Herbert H. Nehrlich, Objection, Your Honour ! 25 February
2005

Competing interests:
None declared

Competing interests: No competing interests

02 March 2005
Peter Morrell
Hon Research Associate, History of Medicine
Staffordshire University, UK