Current issues of ACP Journal Club are published in Annals of Internal Medicine


Where's the meat in clinical journals?

ACP J Club. 1993 Nov-Dec;119:A22. doi:10.7326/ACPJC-1993-119-3-A22

Related Content in the Archives
• Letter: Where's the meat in clinical journals?

For the clinician attempting to nourish lifelong learning, medical journals may seem to provide slim pickings: a few scraps embedded in a mass of materia nonmedica from preclinical test tubes that only scientists can digest and a malnourishing mishmash from preliminary clinical studies. ACP Journal Club picks through this concoction and extracts the juiciest morsels for clinical practice.

ACP Journal Club's selection process begins with review of 6 or more consecutive issues of any journal that has original articles of possible relevance for internal medicine. 33 journals were reviewed in 1991 and 58 in 1992. Journals that provide the highest number of articles that meet our criteria for scientific merit and direct relevance for clinical practice in internal medicine then become part of the core for our review process. We review each issue of these core journals and cite in ACP Journal Club all articles that meet our criteria. Thus, if our criteria for studies of prevention, therapy, diagnosis, prognosis, causation, quality improvement, and health economics suit your clinical journal reading needs, you can be assured that you will not miss any appropriate articles in these journals.

For the most important articles, we prepare an abstract and invite an expert, usually with both clinical and applied research experience, to provide a commentary. Just fewer than half the articles that met our criteria in 1992 appeared in the abstract/commentary section of ACP Journal Club.

The remaining articles that meet our criteria appear in the “Other Articles Noted” section. There are several reasons that these articles are not selected for abstracting. First, some have important information but we have already abstracted 1 or more articles with similar findings in previous issues. Second, some articles are more important to pure subspecialists than to internists with more general interests. If there is doubt about the interest for general internists, 4 associate editors rate the relevance of the article independently and the editor makes the final decision. Third, some articles report scientifically sound studies and have internal medicine content but that content is not particularly useful for practitioners. (For example, drug A does as good a job as drug B but has no advantages; or a clinical hypothesis that is not currently the basis for clinical practice is tested and found to be not valid.) Fourth, although an article meets our criteria, the study it reports has a serious scientific flaw. Our criteria are neither comprehensive nor watertight, and we relegate studies to the “Other Articles Noted” list if they meet our criteria but still have important methodologic problems. Finally, because of space limitations, we are not always able to abstract all the articles that we would like to.

Our selection process provides a way to determine which journals contain the most important evidence for the clinical practice of internal medicine. The table below gives statistics for the top 20 contributors in 1992, journals that provided at least 2 articles that we abstracted. The leading journal, by our criteria, was clearly The New England Journal of Medicine, whether by absolute numbers of articles abstracted or selected or by proportion of articles with abstracts published in that journal. This perhaps will not surprise our readers, but the performance of other general journals, including JAMA, Lancet, and the British Medical Journal, may. In fact, 4 of the top 6 journals are general journals. The 2 leading internal medicine specialty journals are Annals of Internal Medicine and Archives of Internal Medicine. Annals was the second most productive journal for ACP Journal Club in 1992, in terms of the proportion of its articles with abstracts that were reabstracted in ACP Journal Club.

Table 1. Top 20 Journals Contributing to ACP Journal Club in 1992*

Journal Number of Articles with Abstracts Number (%) of Articles Meeting ACP Journal Club Criteria Number (%) of Articles Abstracted in ACP Journal Club
N Engl J Med 254 43 (16.9) 34 (13.4)
JAMA 303 37 (12.2) 20 (6.6)
Ann Intern Med 246 33 (13.4) 18 (7.3)
Lancet 410 30 (7.3) 16 (3.9)
Arch Intern Med 262 27 (10.3) 13 (5.0)
BMJ 283 24 (8.5) 17 (6.0)
J Intern Med 157 17 (10.8) 3 (1.9)
Diabetes Care 232 17 (7.3) 3 (1.3)
Circulation 541 15 (2.8) 4 (0.7)
Am J Cardiol 649 15 (2.3) 4 (0.6)
Chest 780 13 (1.7) 2 (0.3)
Am Rev Respir Dis 520 12 (2.3) 4 (0.8)
J Am Coll Cardiol 407 11 (2.7) 0 (0.0)
Am J Med 298 10 (3.4) 5 (1.7)
Gastroenterology 494 9 (1.8) 3 (0.6)
J Clin Epidemiol 144 7 (4.9) 4 (2.8)
J Gen Intern Med 71 6 (8.5) 3 (4.2)
Neurology 445 6 (1.3) 2 (0.4)
Can Med Assoc J 70 5 (7.1) 3 (4.4)
Spine 271 2 (0.7) 2 (0.7)

*Other journals reviewed that had at least one article meeting ACP Journal Club criteria, in descending order of yield: Ann Emerg Med, J Am Geriatr Soc, J Fam Pract, J Infect Dis, Am Heart J, Ann Med, Gut, J Allergy Clin Immunol, J Clin Gastroenterol, J Internal Med Res, Age Ageing, Am J Prev Med, Aust N Z J Med, Arch Neurol, Arthritis Rheum, Blood, Br Heart J, Br J Addiction, Can J Cardiol, Clin Pharmacol Ther, Crit Care Med, Eur Heart J, Gastrointest Radiol, Hepatology, Hypertension, J Cardiovasc Pharmacol, J Hypertens, J Manipulative Physiol Ther,J Neurosurg, J Urol, Med Care, Ophthalmology, Physicial Therapy, Prev Med, Scand J Gastroenterol, Scand J Rehabil Med.

Of interest to clinical readers will be the generally low yield of clinically useful and strong evidence presented in these journals. For the leading journal, the yield in 1992 was less than 1 article per issue and, for journals lower than the top 6, the yield decreased to less than 5 articles per year. This highlights the fact that medical journals serve several purposes, the most important of which is probably scientist-to-scientist communication (1). Communications from scientists to which clinicians need to pay attention comprise a small part of the content of peer-reviewed medical journals. Secondary publications such as ACP Journal Club serve to concentrate the information that has been adequately validated for consideration by practitioners.

From the perspective of the clinical practice of internal medicine, where is the meat in clinical journals? First, the best cuts are represented in the abstract/commentary section of ACP Journal Club. Second, the “Other Articles Noted” section contains some additional useful material, particularly for internists with subspecialty interests. Readers who want direct access to the full text of articles should subscribe to as many top journals in the table as they can afford. Besides the articles in these journals that meet our criteria, internists will find other useful items-if there is time!

R. Brian Haynes, MD


1. Haynes RB. Loose connections between peer-reviewed clinical journals and clinical practice. Ann Intern Med. 1990;113:724-8.