Trip Database Blog

It’s been a busy day. We’re going to be ‘staff light’ next week so I’ve put in a fair few hours today. Of the 7 questions attempted few could be answered with robust evidence; a few examples:

What is the difference between soluble and insoluble fibre? How does the soluble form reduce cholesterol.
The first part was fairly straightforward, while the second part was conjecture that doesn’t appear to have moved on in a number of years. I suppose that it doesn’t matter greatly as long as there is an effect, but even so…

What current evidence for best practice is there in the conservative podiatric management of paediatric pes cavus?
Cochrane reviewed this topic in 2007 (not restricted to children) and found no real evidence, apart from one area. We did find a 2008 retrospective review of cases in children, but that’s hardly the most robust evidence!

In minor surgery following abscess incision and drainage, is packing necessary and if so how long and how frequently should the abscess be packed?
This was a good one! Why? Because the published evidence appeared to clash with our medical directors experience!!! His view was that the research was of low quality. But then we get down to the ‘nitty gritty’ of evidence-based practice – what is the role of experience?

In the absence of robust evidence, when there is only poor-quality research, where does a doctor’s (or nurse’s) experience fit into things? I often feel that ‘evidence-based’ practice judges experience too harshly. Perhaps not judges it; sneers might be a better word.

I still feel that DUETs has some promise in highlighting gaps in the research evidence. DUETs is attempting to record gaps in the evidence (relating to therapeutics) highlighted from research recommendations and clinical Q&As. The idea being to inform/enhance the research procurement process. In other words, if you know that lots of people are interested in a particular clinical question and the evidence-base is poor – procure some research!

As part of our TRIPanswers (to be launched July/August 2008) we’re taking a different angle at highlighting the research gaps – with our ‘tag cloud of clinical uncertainty‘. Ours is a more pragmatic approach. Perhaps the biggest differences will be what constitutes uncertainty (we’ll take a less academic approach) and what can be covered (we’ll cover everything, not just therapeutics). Our pragmatic approach will allow us to ‘cover’ more ground and we should launch with between 500-1,500 uncertainties.

If either (or both) approaches makes an impact it’ll have been worthwhile.

We’re currently exceptionally busy working on new projects and will highlight them below.

• TRIP Spain. A Spanish version of TRIP is currently being beta-tested and should be available soon.
• TRIPanswers. A very special Q&A project. No public details yet, but we’ve just signed off designs and the build starts next week. This should launch early summer 2008.
• TRIPcpd. Again no public details here, other than the obvious clue in the name. Due late summer 2008.
• Rolling cache. Due to the huge numbers of users the system is starting to creak. This will be a two stage change. Firstly, the results count displayed in the results categorisation section will disappear. This will speed things up to sub one second searches. Secondly, the results will re-appear (4-6 weeks later) following optimisation and a new method of displaying the results.
• Advertising system. Due to the poor performance of Google ads we’re going to try a new system, should be available in 4-6 weeks. Although no real enhancement to the site I’m certainly expecting a much better return and possibly some healthy income!

If the advertising system starts bringing in the money we’ll start on two other projects TRIPmed and TRIPpatient for the end of 2008!

This may well be a bore for everyone but the TRIP Database has passed a significant milestone. In March we were searched 1,001,360 times.

When TRIP was available via a subscription only we were searched around 25,-30,000 times per month. We switched to free access towards the end of 2006 and have grown dramatically since then. As a result of all this we have had to introduce new hardware and new software solutions to keep up with the demand. A new system should go online in 4-6 weeks which will see searches becoming over 5 times quicker.

Although there has been a good chance of reaching the million mark for a few days (and we only just made it), we at TRIP are delighted with March’s results.

I think it’s a bit early for champagne!

21,092 to go…..

A particularly slow weekend has got me nervous. But, as of now we have been searched 971,444 times this month. Therefore, we have just over 17 hours to be searched 28,556 times.

Well, two days to go and things are looking good.

As of yesterday we had been searched 942,964 times. So, for the last two days of the month we need to be searched 57,036 times.

I think we might just make it…

“Their TRIP database is a very useful resource for the practitioner and one which should be included as one of your favourite sites or bookmarks.”

Click here for the full review.

In my struggle to understand web 3.0/semantic web I’ve been exchanging a fair few e-mails with people, notably Dean at the Google Scholar Blog. I feel as though I’m starting to understand the key concepts – but I still need to get a bit more guidance before I try and explain it here. But linked into this is a great post Semantic Web Patterns: A Guide to Semantic Technologies on the ReadWriteWeb blog.

Two other interesting finds:

• Free meta-analysis software, MIX – Meta-analysis with Interactive eXplanations. Alas, my version of Excel (2007) is not compatible with the programme, so I can’t check it out. For those ‘lucky’ enough to have an earlier version of Excel – I hope you enjoy it. You can read more about the software via this BioMedCentral article.
• Searchme. I manged to get an invite to be a beta-tester. I love it! I’m not sure how helpful it would be in the clinical world. But it does allow you – very quickly – to scan the search results. I’ve embedded a demo below.

This may be a bore for others, but I had hoped to break the 1 million searches per month by the end of 2008. Well, it might happen much sooner.

As of the end of the 24th March we had been searched 769,334 times per month, so 230,666 searches off the 1 million, with a week to go. That averages 32,952 searches per day. We’re beating that for weekdays, but dipping under it for weekends.

One thing I do predict is that I’ll be the only tense watcher of these statistics!

If we don’t do it in March we’ll do it pretty soon as two new guarenteed projects should deliver significant new traffic while a 3rd (in negotiation) will add even more.