Trip Database Blog

A truely bizarre/shocking story about a COI incident involving JAMA. This article reports it very nicely.

Twitter has been alive with chat about this – see here.

Internet Explorer is now available as version 8 (IE8). Unfortunately, as is so often the case, the new technology causes parts of the site to break. Fortunately, the breaks do not appear appear to be too severe and no massive rush to fix them – but they will get fixed!

Work will shortly be starting on the next upgrade to TRIP. There are currently 19 separate projects as part of this upgrade and work should be completed by the start of July. The main pieces of work include:

• Improved speed
• Significant enhancement of CPD/CME capabilities
• Overhaul of ‘My-TRIP’
• Incorporate a new design for the site using, for the first time, a specialist external design agency
• Creation of a ‘New/Newsworthy’ section
• Integrate extra content (moving content from specialist TRIP’s into the main TRIP search)
• Creating a background knowledge box
• Create an export facility (send via e-mail or export as a .txt file)
• Improvement to the search algorithm
• Improve medical images

As these pieces of work are completed and tested internally I’ll blog in more detail about each feature.

The latest update of new content on TRIP is now searchable – a total of 453 records have been manually added plus a similar amount automatically added.

I like to think that TRIP has developed a powerful brand name. Equally importantly the brand name appears to evoke a high level of fondness. So much so we decided to protect our brand and last year applied for a trademark. We have just been notified that this has been granted. In the future, should someone try to use the name TRIP in the medical sphere, we have a high level of protection.

Interesting quote from a recent interview with Marissa Mayer (Google) taken from this TechCrunch interview:

Question: Is it fair to say that search is in its infancy.

Answer: Very much so. It was interesting for our engineers to see that early index and see how far we’ve come in ten years. But when you think about what would be the perfect search engine, what is an answer as opposed to a result? Why are we handing you just links and URLs? You know, what does it mean to try and synthesize a video or an image or a diagram that better explains your answer or maybe even grabs facts from all the different pages and helps you do comparisons. There’s just a lot of different things we can do…..

I’ve highlighted the bit that grabbed my attention and it follows a theme I’ve blogged before (click here to see a post from 20th Nov 2008). Basically, search engines do not answer questions, they post results.

Would clinicians (or other search engine users) want answers or search results?

Until people realise the difference between the two and realise their users want answers not results I’m afraid the aspirations will be limited. Our, modest but expanding attempt, at supplying answers (see TRIP Answers) is doing really well, but it has less than 6,000 answers. Having 100,000 clinical Q&As quality marked and easily searchable is the challenge. The irony is the ‘answers’ are in all the documents that exisit. Take CKS guidelines, these are very long documents (too long for clinicians to easily navigate) that must contain over 100 clearly defined Q&As and with around 200 guidelines that’s around 20,000 Q&As (from experience CKS answers the questions GPs have, it’s by far and away the best resource). Another example is the Green Book (the Government’s document on immunisation practice), there are 25+ chapters and each has over 15 clearly defined Q&As within them.

Here’s hoping that people start to understand – soon – the difference between results and answers.

Another great find via twitter:

Guide to Biostatistics

“Here is a compilation of important epidemiologic concepts and common biostatistical terms used in medical research. You can use it as a reference guide when reading articles published on MedPage Today or download it to keep near the reading stand where you keep your print journals. For more detailed information on these topics, use the reference list at the end of this presentation.”

I’ve had a good relationship with ARIF for years and they’re kind enough to send me their links (saving me having to do it). As I was carrying out the upload I noticed that ARIF has the lowest publication ID. Each publication is given an ID and ARIF’s is 8 (Cochrane = 14 and Bandolier = 9) we’re currently up to around to publication IDs in excess of 1,100!

Ultimately, these IDs are meaningless to users but it does show the length of relationship!

I started this 3-4 years ago and kept it up for a few months, then workload pressures made me stop updating it – so I forgot about it – until now! It’s a great visualisation of the the contacts we have and I hope to be more diligent in updating it in the future! To view the map click here, alternatively, I’ve embedded a smaller version below.