Justin Clark recently took up the role of Trials Search Coordinator with the Acute Respiratory Infections Group at Bond University. Here, he shareshis thoughts as a newcomer to Cochrane and a seasoned expert in the world of searching. Over to you Justin...
As a health librarian who considers himself reasonably accomplished and quite good at the job, I admit I was overjoyed to be given the opportunity to take on what, in my opinion, is the most interesting and challenging role a health librarian can attempt: being a Trials Search Coordinator (TSC) for Cochrane.
Since starting the job in May, I've come to the conclusion that I was correct. It is really challenging. Far more challenging than I could have imagined, but also far more interesting. I can think of few other roles where I would get to work with researchers from such wide and varied locations and with such a diverse range of interests. So there's no looking back and definitely no regrets on my decision to take the job - it's one of the best career moves I have ever made.
There have already been numerous highlights to date. Obviously the work itself is very rewarding, but there's more to it than that. There's a great sense of camaraderie amongst the TSC community, with everyone willing to help each other, provide advice or pointers or even just say hello. The sense of community at Cochrane is very welcoming and it feels great to be a part of it. The support structures in place have made taking on the role a lot easier, with the great help files and the even greater TSC support crew . They have helped out with such speed and enthusiasm and are also very easy to talk to. There were a lot of laughs as well as information in my induction sessions!
The last highlight is to finally sit down at my desk and look at the search strategies from reviews that I need to update. It is then that I breathe a heavy sigh of relief as I realise that all I really have to do is cut and paste them into a database and hit the search button. Speaking as someone who has been involved in a lot of systematic reviews, both writing them and teaching people to write them, this is quite a relief. Although I'm sure I'm not telling any of my TSC colleagues anything they don’t know, the standard of systematic searching outside of Cochrane is, well, woeful would be a polite way of putting it.
It's been quite a pleasure to look at the searches that have been done in the past without developing a migraine trying to work out what on earth these people were actually trying to find. By extension, it's been interesting to look at the different ways people who know what they are doing search. You can see what you have in common and learn a few cool new strategies. I am happy to admit that I may have borrowed the odd search idea from other people from time to time.
Of course coming in as an outsider to a new role, organisation and a different way of doing things doesn’t just highlight the things that are done well, of which there are many. You also get the chance to have a look at the things that you think could be done differently. I did notice two things when I started that I felt could do with a bit of attention.
The first was the fact that I was cutting and pasting search strategies into databases and hitting the search button. I'm unsure of when saving search histories became a common feature of databases, but I think it is a tool that is underutilised. I was surprised that there seems to be no central repository of saved searches for Cochrane Reviews so with a simple click of a button you could run an update of a review.
The second thing is something not quite so obvious. It's the way search strategies are reported. Although the searches themselves are of a very high quality, the way they're reported means that you need to know quite a bit about searching in order to replicate them. This is fine for TSCs but my own experiences with clinicians and researchers would suggest that most of them would not be able to re-do these searches and replicate the results. In fact, most of them would simply go cross-eyed and stare blankly at them for a while before moving on to other parts of the review. This has made me start to think of ways to write out search strategies so that even the most inept of database users can understand how the articles that end up informing all of the Cochrane reviews actually got there in the first place. Not an easy task when you think of just how bad most people are at systematic searching!
But the good news is that at the Centre for Research in Evidence-Based Practice here at Bond University we're working on a tool to do some of these things. It's currently set up to convert PubMed or Ovid searches to the other databases and also auto-run those searches. You can also insert comment lines into your searches, much like computer programmers do in their code to help people understand what's going on. It's not perfect but already saves me a lot of time. Hopefully tools like this and an increase in the use of existing tools will make the searching that underpins Cochrane reviews even better into the future.