To The Editor:
I read with interest the recent paper "Reference Accuracy in Peer-Reviewed Pediatric Orthopaedic Literature" (2010;92:1155-61), by Davids et al., on the accuracy of references in the pediatric orthopaedic literature. Inaccuracy is a serious matter that needs to be resolved urgently, particularly in terms of quotation errors, which can then permeate throughout the entire body of orthopaedic knowledge.
What stimulated me to write was an obvious minor error in reference 26, BJM instead of BMJ, but this made me ask the question "Who else has noticed this error?" which leads to the wider question "How many have actually read this paper?" There may now be a journal impact factor, but are there any figures collected for readership of a particular article?
As medicine and journals become ever more specialized and the number and frequency of journals increase, the readership of an individual article must inevitably decrease. As a general rule, with modern-day pressures on time, readers will only read papers that are related to their particular areas of interest. Perhaps then it could be left up to the readers to check the references as they are likely to be fairly familiar with the field. This will work only so long as corrections are published quickly and prominently.
The current common practice would seem to be that it is the role of the authors to check the references. But should it not be the role of the reviewer? Surely, peer review strongly implies that the reviewer has an intimate knowledge of the field and should have an intimate knowledge of the relevant literature and could note errors in the referencing. Again this is difficult because of pressures of time, but it would be helped by limiting the number of references allowed. This idea could be helped by the journals explaining clearly and frequently what purposes references actually serve. With modern technology, it is very easy to quickly generate dozens of references, without spending long hours in the library, but how many of these are ever read or indeed relevant? Perhaps it is now time for the journals to alter their methods to take into account modern-day electronic search facilities.