Are Top Journals Biased Against Eating Disorders Topics?

Brief Report
Are Top Journals Biased Against Eating Disorders Topics?
Susie Frost, M.Sc.
and/or bulimia nervosa and the number of articles published
about panic disorder and/or agoraphobia (i.e., disorders of
Rebecca Murphy, B.A.
comparable disease burden) in 29 high-impact journals over a
Peter Webster, M.A., M.R.C.Psych.
5-year period (1996–2001).
Ulrike Schmidt, Ph.D., M.R.C.Psych.
Results: There were almost twice as many published empirical
articles about panic disorder and/or agoraphobia (N=365) as
Objective: This study investigated whether there is a bias
there were about anorexia nervosa and/or bulimia nervosa (N=
against eating disorders research among the leading psychiat-
169).
ric, psychological, and medical journals.
Conclusions: The findings indicate a possible bias against eat-
Method: The authors performed a comparison between the
ing disorders research among some leading psychiatric jour-
number of empirical articles published about anorexia nervosa
nals. Alternative explanations and implications are discussed.
(Am J Psychiatry 2003; 160:363–365)
Anorexia nervosa and bulimia nervosa are severe and raphobia (4,600 disability-adjusted life-years), with a total of
often chronic eating disorders associated with high levels
10,192 disability-adjusted life-years.
of physical and psychological comorbidity and poor qual-
We compared the number of articles about anorexia nervosa
and/or bulimia nervosa with those about panic disorder and/or
ity of life. A unique feature of anorexia nervosa patients is
agoraphobia. A review was conducted of all articles between Feb-
their high mortality, which is twice that of other psychiat-
ruary 1996 and March 2001 that were published in the 10 highest-
ric inpatients. Eating disorders, particularly anorexia
impact psychiatric, psychological, and medical journals. These
nervosa, also pose a financial burden, with the cost of
were found from a search of the 1999 Journal Citation Reports,
treatment similar to that of the most severe psychiatric
Science Edition (Institute for Scientific Information) with the sub-
ject headings “psychiatry,” “psychology,” and “medicine, general
disorders (1).
and internal.” Table 1 shows the 29 journals reviewed and their
Despite the serious nature of these disorders, there is a
impact factors (one journal was listed twice as one of the 10 most-
prevailing view among medical professionals that eating
cited journals for psychiatry and psychology).
disorders are not genuine illnesses (2). Many health pro-
Articles about anorexia nervosa, bulimia nervosa, panic disor-
der, and agoraphobia were searched by using two databases,
fessionals hold stigmatizing attitudes toward eating disor-
MEDLINE and PsycINFO. Two search terms were used to select
ders, based on the belief that they are self-inflicted (3).
any article about either anorexia nervosa and/or bulimia nervosa
Compared to patients with diabetes, obesity, or schizo-
(“anorexia nervosa OR bulimia nervosa”) and panic disorder and/
phrenia, those with eating disorders are less well liked by
or agoraphobia (“panic disorder OR agoraphobia”). The “or” op-
medical and nursing staff—including psychiatrists (4, 5).
erator employed is an inclusive term and searches for articles that
contain any of the query terms (e.g., any articles referring to an-
The aim of this study was to investigate whether the
orexia nervosa or bulimia nervosa, including articles referring to
view of eating disorders as trivial and self-inflicted has
both). A total score of the number of articles about each topic was
contributed to a bias in publication rates of articles with
computed for each journal, excluding letters, comments, editori-
als, and duplicated articles across databases.
related topics in leading journals in recent years. We as-
To investigate whether there is an acceptance bias against arti-
sessed this by comparing the number of published articles
cles about eating disorders among high-impact journals, all 29
about anorexia nervosa and/or bulimia nervosa with
journals were contacted for information about their numbers of
those of disorders of equivalent disease burden.
articles about anorexia nervosa and/or bulimia nervosa and
panic disorder and/or agoraphobia submitted, accepted, and re-
jected between 1996 and 2001.
Method
We used disability-adjusted life-years, which measure the total
Results
impact of fatal and nonfatal outcomes, to identify disorders of
comparable disease burden to anorexia nervosa and/and bulimia
Table 1 shows the results of the journal review. Between
nervosa. Data were taken from the Australian Burden of Disease
February 1996 and March 2001, a total of 169 empirical
and Injury Study (6), since the leading causes of disease burden in
articles were published about anorexia nervosa and/or
Australia are similar to those for other established market econo-
bulimia nervosa in the highest-impact psychiatric, psy-
mies. The disability-adjusted life-years for eating disorders were
chological, and medical journals. This compares with 365
reported as 11,176 (5,835 disability-adjusted life-years for anor-
empirical articles published about panic disorder and/or
exia nervosa and 5,340 disability-adjusted life-years for bulimia
nervosa). Two anxiety disorders had a comparable disease bur-
agoraphobia. We excluded letters, comments, and edito-
den: panic disorder (5,592 disability-adjusted life-years) and ago-
rials, of which there were 64 about anorexia nervosa and/
Am J Psychiatry 160:2, February 2003
http://ajp.psychiatryonline.org
363

---

BRIEF REPORTS
TABLE 1. Numbers of Empirical Articles About Anorexia Nervosa and/or Bulimia Nervosa Versus Panic Disorder and/or
Agoraphobia in Leading Psychiatric, Psychological, and Medical Journals Between February 1996 and March 2001
Total Number of Articles About
Total Number of Articles About
Impact
Anorexia Nervosa and/or
Panic Disorder and/or
Journal Type and Titlea
Factor
Bulimia Nervosa (N=169)b
Agoraphobia (N=365)b
Psychiatric
Archives of General Psychiatry
10.9
17
39
Molecular Psychiatry
7.9
8
5
Schizophrenia Bulletin
6.6
0
1
American Journal of Psychiatry
6.3
33
91
Journal of Clinical Psychopharmacology
5.7
2
32
Neuropsychopharmacology
4.8
2
14
Journal of Clinical Psychiatry
4.2
10
108
British Journal of Psychiatry
4.1
30
24
Journal of the American Academy of Child and Adolescent
Psychiatry
3.6
24
10
Psychological Medicine
3.4
29
20
Psychological
Psychological Bulletin
7.8
0
0
Annual Review of Psychology
7.5
0
0
Psychological Review
6.8
0
1
Cognitive Psychology
4.2
0
0
Psychological Medicine
3.4
29
20
Journal of Experimental Psychology, Learning Memory
and Cognition
3.2
0
0
Neuropsychology
3.1
0
0
Psychophysiology
3.0
1
Journal of Memory and Language
2.9
0
0
Neurobiology of Learning and Memory
2.8
0
0
Medical
New England Journal of Medicine
28.8
4
2
Journal of the American Medical Association
11.4
1
7
Lancet
10.2
2
2
Annals of Internal Medicine
10.1
1
1
Annual Review of Medicine
7.2
2
0
Archives of Internal Medicine
6.7
0
2
British Medical Journal
5.1
4
2
American Journal of Medicine
4.9
0
3
Medicine
3.4
0
0
British Medical Bulletin
2.9
0
0
a From a search of the 1999 Journal Citation Reports, Science Edition (Institute for Scientific Information) with subject headings of “psychiatry,”
“psychology,” and “medicine, general and internal.”
b From a search of MEDLINE and PsycINFO. Two search terms were used to select any article about either anorexia nervosa and/or bulimia ner-
vosa or panic disorder and/or agoraphobia.
or bulimia nervosa and 86 about panic disorder and/or
journal received one article about panic disorder and
agoraphobia.
none about anorexia nervosa, bulimia nervosa, or agora-
The majority of the articles were published in the high-
phobia, and one journal received one article about an-
impact psychiatric journals: 131 of the 169 articles about
orexia nervosa and one article about panic disorder. The
anorexia nervosa and/or bulimia nervosa and 334 of the
one exception was the Journal of Clinical Psychopharma-
365 articles about panic disorder and/or agoraphobia.
cology, which received 60 articles about panic disorder
There were no articles published in the highest-impact
and/or agoraphobia (and accepted 45) and five articles
psychological journals about anorexia nervosa and/or bu-
about anorexia nervosa and/or bulimia nervosa (and ac-
limia nervosa between 1996 and 2001, and only one article
cepted three). Unfortunately, three of the journals with
was published about panic disorder and/or agoraphobia
the biggest differences in the numbers of eating disorders
during that time.
and anxiety disorders articles did not respond to our re-
quest, and the limited data meant we were unable to de-
Of the 29 journals approached for submission and ac-
termine whether there is an acceptance bias.
ceptance rates between 1996 and 2001, 21 responded. Ten
journals recorded this type of information, and 11 did
Discussion
not. The submission rate was generally low for articles re-
garding these disorders. Four journals received no articles
Our survey of leading journals found that, between 1996
about anorexia nervosa, bulimia nervosa, panic disorder,
and 2001, there were nearly twice as many articles pub-
or agoraphobia. Three journals received five or fewer arti-
lished about panic disorder and/or agoraphobia than
cles about anorexia nervosa and/or bulimia nervosa and
about anorexia nervosa and/or bulimia nervosa, despite
none about panic disorder and/or agoraphobia. One
the disorders’ comparable disease burden. Two of the
364
http://ajp.psychiatryonline.org
Am J Psychiatry 160:2, February 2003

---

BRIEF REPORTS
journals included were psychopharmacology journals.
research into eating disorders among some of the leading
Drug treatments are currently not commonly used in the
psychiatric journals. This may reflect editors’ and review-
treatment of anorexia nervosa; thus the likelihood of arti-
ers’ negative attitudes toward these disorders, possibly
cles about anorexia nervosa being submitted to these
viewing them as unimportant for the general psychiatrist
journals was low. However, even if these two journals were
compared to anxiety disorders, which are seen as “bread-
excluded from the count, the nearly 2:1 ratio between anx-
and-butter” psychiatry. The fact that anorexia nervosa and
iety and eating disorders articles published in leading
bulimia nervosa are predominantly female disorders com-
journals remains intact.
pared to the more equal sex distribution of panic disorder
Given the dearth of information about ratios of submit-
and agoraphobia may contribute to the view that eating
ted to accepted articles regarding the two types of disor-
disorders are of lesser importance.
ders, two basic possibilities need to be considered in try-
In summary, the question of whether top journals are
ing to explain this finding. One possibility is that the ratio
biased against studies of eating disorders cannot be con-
of submitted to published articles regarding both groups
clusively answered by our study but deserves further at-
of disorders is similar but that fewer eating disorders arti-
tention. We recommend that all generalist journals moni-
cles than anxiety disorders articles are being submitted to
tor the ratio of submitted to published articles concerning
top journals. This could be the result of fewer research
particular topics in order to avoid the possibility of dis-
studies being carried out about eating disorders than
criminating against particular disorders.
about anxiety disorders, perhaps because less funding is
available. However, when we searched the U.S. National
Received Aug. 13, 2001; revision received April 4, 2002; accepted
Institute of Health’s Computer Retrieval of Information on
July 25, 2002. From the Eating Disorders Research Unit, Institute of
Psychiatry, De Crespigny Park. Address reprint requests to Dr.
Scientific Projects database (http://crisp.cit.nih.gov) to
Schmidt, Eating Disorders Research Unit, Institute of Psychiatry, De
ascertain the numbers of projects funded as a crude indi-
Crespigny Park, Denmark Hill, London, SE5 8AF, U.K.; u.schmidt@
cator of potential funding discrepancies, we found a simi-
iop.kel.ac.uk (e-mail).
Susie Frost’s and Rebecca Murphy’s research is supported by a
lar number of grants funding agoraphobia and/or panic
grant from the PPP Foundation.
disorder research (N=238) as for anorexia nervosa and/or
bulimia nervosa research (N=200) for the 5-year period
covered by our study. Alternatively, eating disorder re-
References
searchers may believe that their chances of getting their
1. Nielsen S, Møller-Madsen S, Isager T, Jorgensen J, Pagsberg K,
articles published in high-quality journals are limited and
Theander S: Standardised mortality in eating disorders—a
therefore do not submit their articles to such journals. In
quantitative summary of previously published and new evi-
particular, it may be that these authors preferentially sub-
dence. J Psychosom Res 1998; 44:413–434
mit their articles to one of the specialty eating disorders
2. Agras WS: The consequences and costs of the eating disorders.
journals. However, there are also specialist anxiety disor-
Psychiatr Clin North Am 2001; 24:371–379
3. Vandereycken W: Naughty girls and angry doctors: eating dis-
ders journals, and on the face of it, the opportunities for
order patients and their therapists. Int Rev Psychiatry 1993; 5:
submitting articles to specialty journals seem comparable
13–18
across the two types of disorders.
4. Fleming J, Szmukler G: Attitudes of medical professionals to-
A second possibility is that the numbers of articles sub-
wards patients with eating disorders. Aust NZ J Psychiatry
mitted for the two groups of disorders is similar but a
1992; 26:436–443
higher proportion of the eating disorders articles are re-
5. Gowers SG, Shore A: The stigma of eating disorders. Int J Clin
Pract 1999; 53:386–388
jected. This could be because the eating disorders articles
6. Mathers C, Vos T, Stevenson C: The Burden of Disease and In-
are generally of poorer quality, which seems rather un-
jury in Australia: AIHW Catalogue Number PHE 17. Canberra,
likely. More likely, our findings may indicate a bias against
Australian Institute of Health and Welfare, 1999
Am J Psychiatry 160:2, February 2003
http://ajp.psychiatryonline.org
365

---