A new journal for case reports only, The Journal of Medical Case Reports, has spawned an discussion at The Scientist about whether we should even have case reports in journals:

Does the medical literature need more case studies? A new journal is betting it does, even as editors at other journals say the answer is no.

Historically, case reports have proven extremely valuable to clinicians faced with diseases they knew little about. But in an age where countries spend more on research than ever before investigating both rare and common diseases, some experts argue that the obscure nature of many case reports makes them of little value to the average practitioner.

Case reports typically receive fewer citations than research articles, putting them in danger of being phased out at journals where citation data rule decisions, said Matthew Cockerill of BioMedCentral (BMC, sister company to The Scientist), which publishes the new online open access journal, The Journal of Medical Case Reports.

Indeed, a 2005 study in the Journal of the American Medical Association showed that case reports receive the fewest citations of all other study designs. In a group of 416 case reports published between 1991 and 2001, less than two percent received 10 or more citations in the first two years of publication.

Dedicating an entire journal to case reports ensures that these valuable stories will be told, and grouping them together helps doctors establish timelines or patterns for rare pathologies, Cockerill noted.

However, Drummond Rennie, deputy editor of the Journal of the American Medical Association -- a journal that does not publish case reports -- said he is skeptical about the value of the new journal. Case reports that merely add another patient to a roster of those who have rare diseases accomplish little in the medical literature, he told The Scientist. The role of journals is "to advance knowledge," he added. "The first few cases [of a rare condition] are great, someone had to describe it. After that they have to say why they got [the condition] and what are we going to do about it."

As a general rule, I don't tend to cite case reports. From a scientific point of view, reviews and controlled studies are much more useful to me.

However, I can think of two reasons why we shouldn't get rid of them entirely:

1) Rare neurological cases -- Sometimes cases in neurology present in truly rare and bizarre ways. In these cases, the issue isn't so much how you are going to treat the patient -- there usually isn't much you can do -- so much as where the lesion is and what the details of the presentation are. Understanding lesions of this type is an important pillar of how we localize functions to different parts of the brain -- because you can't very well go about removing parts of peoples brains just to see what they do. They allow us to know the results of experiments we cannot perform ethically.

I can give you an example of this. I published an article a while back about a lesion where the patient lost his addiction to drugs as a result. This is a really interesting finding that informs our understanding of addiction pathways in humans. I am glad they published it even if it may not be cited regularly.

2) Teaching tools -- I am still learning to be a doctor, and one of my favorite ways to learn is using case reports. I am particularly fond of the ones in the New England Journal of Medicine. They help me put the presentation of a patient into a broader context. Furthermore, even if it is an obscure presentation, medicine is all about approaching a problem, and I can apply the approach used in the case report to more common cases.

Now I don't know if either of those two examples imply that there should be a whole journal for case reports, but I do think that they are good enough reasons to keep them around a while longer.

Any other arguments for or against case reports...