Subspecialty coverage in articles in ACP Journal Club
ACP J Club. 1992 Nov-Dec;117:A17. doi:10.7326/ACPJC-1992-117-3-A17
Comments from readers have almost universally praised the concept, structure, and policies of ACP Journal Club. Some subspecialists, however, have complained that their interests have been under-represented among the articles selected for review. Because I wrote one such letter as a member of the editorial board and at least one other member of the medical oncology community has raised the same issue, I volunteered to write an editorial exploring the coverage of oncology topics by ACP Journal Club.
The first question is whether my colleagues and I are correct in our impression that the number of articles included in ACP Journal Club on oncology is not proportionate to the importance of cancer as a disease. As a first stab at this question, I reviewed the articles published in ACP Journal Club since its inception and classified them according to area of subspecialty. The results are displayed in the left-hand column of the table below. My experimental methods would certainly not meet ACP Journal Club standards, and others might classify the articles somewhat differently (for example, I assigned cerebrovascular disease to neurology and venous thromboembolic disease to hematology). However, I think the findings are striking: The likelihood of an article appearing in the ACP Journal Club is not closely related to the frequency of the disease it discusses in the population. Some sort of selection is clearly going on.
Table 1. Number (and Percentage) of Articles by Specialty Published in ACP Journal Club and Annals of Internal Medicine*
|Discipline||ACP Journal Club||Annals of Internal Medicine|
|Cardiology||78 (28)||32 (16)|
|Lipids||12 (4)||6 (3)|
|Neurology||24 (9)||6 (3)|
|Allergy||2 (1)||3 (1)|
|Infectious disease||21 (8)||24 (12)|
|AIDS||13 (5)||18 (9)|
|Cancer||12 (4)||20 (10)|
|Rheumatology and physical medicine||10 (4)||16 (8)|
|Pulmonary medicine||16 (6)||13 (6)|
|Nephrology||7 (3)||10 (5)|
|Gastroenterology||24 (9)||25 (12)|
|Endocrinology||24 (9)||8 (4)|
|Geriatrics||11 (4)||6 (3)|
|Hematology||8 (3)||9 (4)|
|Other||13 (5)||7 (3)|
*Totals for the percentage columns do not equal 100 because of round-off error.
ACP Journal Club is directed at internists, however, so the issue is not the frequency and importance of diseases in the population but the frequency of patients with these diseases in internists' offices. Unfortunately, I did not have access to such data. As a substitute I decided to carry out a similar exercise with articles published in Annals of Internal Medicine—assuming that its editors would know what issues were important to internists and would select articles accordingly. The results of a survey of Annals articles appearing concurrently with the publication period of ACP Journal Club are listed in the right-hand column of the table. The difference in the distributions of articles by specialty for the two publications is highly statistically significant. Again it appears that, when compared to this informal measure of expectations, the likelihood of an article's appearing in ACP Journal Club is influenced by subspecialty topic. The editorial “distillate” is indeed “select” (1).
At issue, of course, is whether the selection process is appropriate. The operating principle of ACP Journal Club is that the articles abstracted will have “direct relevance to the management of internal medicine problems” (1). Relevance is unfortunately a subjective criterion, and it is possible that one's perceptions of relevance are influenced by training and environment. For this reason, the editors encourage comments from readers on this issue and will happily take steps to better serve the needs of this audience. In addition, however, oncologists like me will have to consider the possibility that much of their published research, although passing editorial standards for methodology (2, 3), is irrelevant to the day-to-day practice of their parent specialty. Astute readers will have noticed that the Journal of Clinical Oncology, the journal of the medical oncology subspecialty association, is no longer reviewed regularly for ACP Journal Club. It was dropped because none of the articles published in a 1-year period met the criteria for inclusion in ACP Journal Club. In preparing to write this editorial, I reviewed the issues in question, and, with the possible exception of a description of the thromboembolic complications of adjuvant chemotherapy for breast cancer (4), I had to agree that none of the articles provided findings likely to influence decision making in internists' offices. This situation may simply reflect the nature of oncology practice and research, but it probably deserves attention and discussion.
Joseph L. Pater, MD
1. Haynes RB.ACP Journal Club's modus operandi. ACP J Club. 1991 Nov-Dec;115:A14.