Anecdotal patient descriptions have been of great importance for the development of clinical medicine and previously had a central place in medical publishing. This tradition is also true for Norway (1 , 2 ). But does the anecdotal patient history still have a place in today’s evidence-based medicine?
Randomised studies and meta-analyses provide the best basis for clinical treatment guidelines. This means that case reports now play a smaller role in medical publishing. But there are several good reasons why case reports should continue to be published. The thalidomide story showed how individual observations could change clinical practice (3 ). Among the drugs withdrawn from the market in Spain from 1990 to 1999 in the interests of patient safety, case reports were the most important information source in 18 of 22 instances (4 ).
Case reports have an intuitive appeal and capture the reader’s interest in a different way than a meta-analysis. Case reports also have an important function for doctors in training (5 ). For these reasons, since 1999 the Journal has published case reports as educational patient stories, accompanied by expert commentaries, in a dedicated column (6 ). The main point of these case reports is not to present rare conditions, but to promote excellence in clinical medical thinking (2 ). In a reader survey for the Journal in 2007, many readers stated that they would like to see more case reports (7 ).
For a general medical journal such as the Journal, case reports should ideally represent a diversity of patients and subjects. Does this happen? Which patients are described, and to what extent are the various medical specialties represented? When the column was introduced in 1999, the vision was that it would represent a relay race from hospital to hospital across the country, also outside the university hospitals (6 ). To what extent has this goal been achieved? This study has surveyed patients and authors for the 157 case reports and expert commentaries published in the period 1999–2011.
Materials and methods
Relevant articles were identified on www.tidsskriftet.no
For the patients described in the articles, data were recorded on primary diagnosis, gender, pregnancy and age category (0–1 years, 2–10 years, 11–20 years, 21–30 years etc.). For articles that described several patients with the same primary diagnosis (n = 2), only data on the first patient were used.
For the authors, all workplace related affiliations were recorded. Each author was then classified by geographical location (1 of 19 counties) and medical specialty (1 of 44 specialties). Classification of specialty was made based on the author’s primary place of work. When workplace indicated a sub-specialty (e.g., «Cardiac Department»), the sub-specialty was recorded and not the main specialty. If in doubt when assigning specialty, information about secondary work place was used. If still in doubt, a preliminary search on the author’s name in Google was done, followed by targeted searches among accredited specialists on the Norwegian Medical Association’s website www. legeforeningen.no. If still in doubt, the author was classified under «unknown specialty» (n = 4). For workplaces with the words «emergency medicine» in the name, searches were made on the author’s name among accredited specialists in anaesthesiology, cardiology, internal medicine and psychiatry; writers who were not listed in any of these were classified under anaesthetics. Two authors were classified as psychologists, two as clinical scientists, one as a dentist and one as a pharmacist.
Results
Literature searches identified a total of 157 case reports with 157 accompanying expert commentaries published in the study period. After the introduction of the column in April 1999, 12 case reports were published in the same year. The number then fell to 0 case reports in 2003, followed by an increase to a peak of 23 case reports in the years 2007 and 2008 (e-fig. 1).
e-figure 1 Case reports and commentaries in the Journal of the Norwegian Medical Association from 20 April 1999 to 31 December 2011
Of the 157 patients, 54% were women. The most common age category was 51–60 years. Among the women, 13 pregnancies were noted. Most of the main diagnoses recorded were subject to only one case report, with the exception of three articles about endocarditis and two papers each for amyloidosis, dermatomyositis, Kawasaki disease, leishmaniasis, epidemic nephropathy, and tuberculosis.
The articles were written by a total of 676 authors, representing 157 first authors and 359 co-authors of case reports, and 160 authors of the accompanying commentaries. Case reports had a median of 3 authors (range 1–7). All commentaries, except for three with two authors, had a single author.
Medical specialties
The 157 first authors most often worked at departments of internal medicine and its sub-specialties (n = 54). This was followed by neurology (n = 21), general surgery with sub-specialties (n = 18), paediatrics (n = 15) and anaesthetics (n = 10). The non-clinical specialties were less often represented, and mainly by co-authorships. Most frequently represented were pathology (n = 27) and radiology (n = 22).
When related to the number of accredited specialists (8 ), neurology had most authorships (Fig. 2). The large specialties general medicine and psychiatry were poorly represented and contributed 5 and 7 of 676 authorships, respectively.
Figure 2 Medical specialty of 676 authors in the column «Case reports» with accompanying commentaries shown with the total number of approved specialists in Norway (8 )
Counties
Most first authorships originated from Oslo (n = 47), Nordland (n = 17) and Hordaland (n = 15). When related to the number of physicians in each county (9 ), Nordland, Sogn og Fjordane and Troms had most first authorships (Fig. 3). Accompanying commentaries (invited by the editors) most often originated from Bergen, Oslo and Sør-Trøndelag, both with and without adjustment for the number of doctors in each county.
Figure 3 County affiliation of 676 authors of case reports and commentaries, shown with the total number of doctors in each county (9 )
Discussion
This study shows that case reports and accompanying expert commentaries during a period of 12 years have presented a broad range of patients and diagnoses, whilst the medical specialties have been unevenly represented. Domination of neurological cases is consistent with the pattern for case reports published in The Lancet (10 ). A majority of first authors from clinical specialties is not unexpected for case reports, but remarkably few case reports originated from psychiatry and general practice. Primary care has previously been encouraged to contribute to this column (2 ), but the presented findings provide a rationale for also extending this invitation to other specialties.
Author lists that include five or six different specialties (11,12) suggest impressive multi-disciplinary collaboration. Probably all clinicians who have been actively involved in the diagnostic work-up and treatment of a patient should be informed before a case report is submitted for publication. It has been pointed out that radiologists and other specialties are not always represented in the author list when a case report is published, even if they have contributed substantially in the diagnostic process and treatment of the patient (13 ).
The number of authorships in the counties varied, but all counties were represented. This is particularly interesting, considering that the original scheme of inviting contributions to the column (6 ) rapidly collapsed (2 ). It is worth noting that individual efforts impacted on the specialty and county level: Erik Waage Nielsen (anaesthesist in Bodø) and Odd Kildahl Andersen (internal medicine specialist in Harstad) have authored 11 and seven case reports in the column, respectively. These efforts are impressive at a time when case reports may yield less academic merit than original articles.
The study has several limitations. Authors with affiliation to a given specialty, e.g. general practice, may have written a case report while working in another specialty, e.g. internal medicine, and if so, this would not have been identified. The great majority of manuscripts were submitted spontaneously throughout most of the study period, while expert commentaries were invited by the editors. Information about rejected manuscripts was not available. It is therefore not known whether the published case reports are representative of all case report manuscripts submitted for consideration for publication.
