Journal of Orthodontics, Vol. 31, No. 4, 277-, December 2004 doi:10.1179/146531204225020614
© 2004 British Orthodontic Society
Quality Evidence
Friedy Luther
There would be little point in a journal existing if it did not aim to improve its standards all the time. There are various ways one could try and achieve this, but trying to attract the best papers in the first place is one obvious mechanism. How can we attract the ‘best’ papers? What do we even mean by ‘best’? Most readable? Most scientifically well constructed? Most clinically relevant? Most clinically relevant for the future? Perhaps all of these things are true, as well as no doubt other factors, but if we look at it from an author’s point of view, we are probably talking about some opposites! Authors usually want quick responses and all their papers to be accepted with least hassle for them, regardless of how bad the paper is, or how irrelevant or lacking in originality. Unfortunately, we end up needing to balance these things—in the interests of maximizing quality and reducing wasted time for a busy readership.
Besides these external factors, we aim to have good internal (i.e. journal) workings to get the best out of what has been submitted and, of course, we rely on peer review for this. Whilst peer review can be criticized (yes, it can be slow, it can be biased and anyway, is the most appropriate reviewer always available?), it is probably still the ‘least bad’ option we have, even though imperfect. However, we do ask a lot of our referees—we rely on them putting aside a lot of time; knowing a particular area very well and keeping right up to date with it; knowing all about study design and, yes, even statistics.
We all need to know some statistics, but I wonder, would we get a statistician to do orthodontics on our patients? Are we necessarily always the best people to know what statistics are good for our studies? Can we be sure we are up to date with current best practice?
In future, the Journal of Orthodontics will aim to send all those relevant scientific papers being considered for publication for formal statistical review. Time will tell whether standards will improve, but I can see no harm in trying.
