Professor Sir John Tooke, talks about his vision for the UCL School of Life and Medical Sciences.

[youtube=http://www.youtube.com/watch?v=si1_O6sL_N4]

Professor Sir John Tooke, Vice-Provost (Health), Head
of UCL School of Life & Medical Sciences and Head of UCL Medical School

I had two papers accepted at the end of last year. Williams & Potts, “Group membership and staff turnover affect outcomes in group CBT for persistent pain” has been accepted by Pain. This is a retrospective analysis of service data with the intriguing finding that higher rates of staff turnover were associated with poorer patient outcomes. The first author is Dr Amanda Williams in the Dept. of Clinical, Educational and Health Psychology at UCL.

Colligan, Anderson, Potts & Berman, “Does the process map influence the outcome of quality improvement work? A comparison of a sequential flow diagrams and a hierarchical task analysis diagram” has been accepted by BMC Health Services Research and is now available online in provisional form. This study found that the style of process map used to describe a clinical task had a significant effect on what risks in the task individuals identified. First author Dr Lacey Colligan is now at the University of Virginia, but previously did her MSc with us at CHIME.

Henry

I’ve always been fascinated with the challenges of  modelling complex domains. Healthcare is one of those domains, without a doubt. Recent developments in finance however, revealed that even models and modelling practices accepted as well developed may turn into inadequate tools quite quickly. The danger of the missing variable in the model is huge.

I can still remember the first time I was introduced into concept of modelling a real life mechanism and one of my earliest questions was: how do you know you have all the relevant variables? The answer was “you don’t, but there are methods to represent this uncertainty in your model”. I guess what has happened (and is still happening) in economy is a bitter demonstration of the change in the magnitude of the uncertainty in the model. When panic is introduced into the models, the behaviour may change substantially. In the last financial crisis, expectations became a hugely effective factor, leading almost to halt of global supply and demand in many industries.

I’ve always believed in the value of variable hunting in large scale data, and these are the cases where having more granular models can help us manage processes better. For example, if the next wave of swine flu hits us with a mutation that makes it resistant to Tamiflu, how would the patients’ behaviour change? I have not given up on the hopes of building a large scale framework for healthcare data, that can help us perform this variable hunt, and there is no doubt that a significant amount of giants like Google, Yahoo, Microsoft and Facebook have theirs running already.

For the NYTimes article that discusses the situation in finance, go to this page.

As academics, we seek to disseminate our research findings, which generally means publication in a journal. However, we would like our work to be better known, so press releases are prepared and you hope some newspaper somewhere will take notice (and that they don’t garble the results too much).

I’m a co-author on a paper on habit formation that has just been published in the European Journal of Social Psychology. The study was by Phillippa Lally as part of her PhD supervised by Prof. Jane Wardle. (Jane supervised my PhD many years ago, which is how I got involved in assisting with the analysis.)

The study was covered by The Daily Telegraph here, with a write-up that explains the science well. But we’ve also gone international with coverage in The Times of India, two newspapers in Ukraine and in Komsomolskaya Pravda, the best-selling paper in Russia. This may seem like an odd spread, but it serves as a reminder of how newspapers work. As Nick Davies discusses in his seminal book on modern journalism, Flat Earth News, given the pressure to generate stories in a competitive market, you see stories being recycled from one paper to another. So, once one paper has covered the story, it becomes much more likely that others will as well. This presumably explains the concentration of coverage in Russia/Ukraine.

Komsomolskaya Pravda is a venerable institution, first published in 1925, but today, it’s a tabloid, and with a tabloid attitude to illustrating articles. So how do you illustrate an article about habit formation? Well, go here and look, and you can see the answer is with a Picasso and a lolcat. Quite what Picasso’s “Two women running on a beach” has to do with the subject, I’m not certain, but I suppose the drunk cat makes more sense…

Dr Henry Potts

PS: If you don’t know what a “lolcat” is, try here or here.

My kids can expend a fair amount of energy waving Wii handsets around. I have a colleague whose hi-spec laptop could be used for weight training. But, by and large, computers don’t help you exercise. Interesting, then, to learn about this project in which school children were given pedometers and encouraged to record the amount they walked on a website. Different classes in the school then competed to see which would first complete a virtual journey. At the same time the website allowed them to monitor their eating and advised them on healthy living.

These kinds of applications are sometimes called ehealth interventions. We used to talk about something called telemedicine, in which some form of telecommunications technology connected patients with doctors. The problem with that was that the difficulty in healthcare isn’t, for most of us in the West, that the doctor is too far away, it’s that he or she is too busy. Giving patients another way to get at the doctor doesn’t help with that problem. So people started to think about telemedicine as a way of connecting patients with some kind of service that might replace a doctor. NHS Direct is a scheme of this kind. But the assumption was that the transaction would be little more than an exchange of information. Patient sends data up the way, service provider sends advice down the way.

What is interesting about the way ehealth is developing, is that we are seeing applications where the ‘magic ingredient’ in the intervention isn’t advice. In one application patients use mobile phones to send blood glucose measurements to a company. The company then supplies them with feedback based on their measurements. This helps them in their self-management. There are also applications on computers that deal with depression, drinking problems and such conditions. A recent meta-analysis found the web or computer-based tools were effective in smoking cessation.

The interesting idea here is that we can consider ehealth applications where patients aren’t consulting a physician or looking up information, or necessarily receiving any kind of conventional healthcare advice. These applications work because they engage the user in a process which helps them learn to change their behaviour. It’s a kind of therapeutic relationship. But with a PC.