This article first appeared on 19.3.13 on digitalhealth.net.
The NHS Commissioning Boards new director of patients and information, Tim Kelsey, wants everyone to write apps.
Recent news items on eHealth Insider floating his idea that clinicians – or even patients – should do some programming drew a lot of negative comments.
Posters told stories about naughty clinicians creating their own software and forgetting about any information governance implications; creating systems that required huge amounts of work to fix; and generally disturbing the calm tranquillity of the IT department.
I think it’s not that bad an idea. I also worry about complacent IT departments and suspect that the reason these naughty users started to design their own systems was that they weren’t being listened to by them.
Understanding the possible
Having said that I agree with the idea of teaching people to program, I think this is more about skilling people up to understand programming – what is possible and easy, what is difficult or time consuming – than with getting everyone to write thousands of lines of code.
I’ve been amazed by how many patients have told me they are being sent on courses about app writing. The last two to do this were teachers, but others have been businessmen.
I came across a newspaper story about a reporter who went on one of these courses.
To her surprise, she didn’t find a load of people wanting to retrain or change jobs. Instead, she found the course full of people in middle to senior management positions who were feeling left behind by the tech.
They just didn’t get the technology; they didn’t know its possibilities, the difference between online and offline apps, how some apps are platform specific while others can work on most devices.
They knew even less about app stores, marketing strategies and business models. They wanted to understand the devices better; some to keep their positions in their company and some to actually deliver more.
A skilled user might actually make the job of IT easier. If a person understands that X is a nightmare to do, they are less likely to ask for it.
Or, if they do, they are going to understand the resource implications. The converse is, of course, true. If you know that a job is easy, you are less likely to take any crap over it.
Taking no nonsense
We recently were told by our ‘moving to Emis Web’ trainer that one of our administrators would have to sit down and type 01270 into all of our 22,000 patient records because we hadn’t recorded the STD code of the patients’ telephone numbers.
Emis Web apparently needs this for Spine compliance. My practice manager was initially taken in by this demand; and I wonder how many others have been.
However, given my limited programming background, I knew that this should be a fairly easy piece of SQL; the sort of thing you might give to a job applicant at interview to test them.
Luckily, our Emis regional manager understands this kind of job should be doable by computer and is trying to get it sorted out. But it’s a classic example of following a script and not knowing what can and what can’t be done.
Opening up APIs: let’s go further
I notice that the opening up of application programming interfaces is in the news, with reports that the NHS Commissioning Board is going to require GP suppliers to open up their APIs as part of a new GP Systems of Choice deal.
This is something that I’ve been after for ages. But I’ve decided that what we need is more than that.
I’ve often been amazed by Microsoft Office. For most things that you do with Office, you only need a really simple word processor. Yet Office is amazingly complex, so hardly anyone ever uses it to its full potential.
It has a macro language built in and, if that wasn’t enough, there is VBA for applications – a type of programming language that allows you to customise the entire program.
Apple has something similar built into its operating systems. It is called Automator and Applescript. Automator is perhaps a vision of the future. It is very graphical; you drag and drop actions and build them into quite powerful tools.
I wonder if what we need is not just open APIs for IT professionals to use, but some form of macro/Automator system built into our clinical systems that users can play with.
New prospects for support
Getting back to Kelsey’s big idea, I think we should be honest. Even if getting more people to understand the basics of programming is a good idea, I doubt that many of the people going on courses will actually end up coding themselves.
So is there a role for a middle person? Somebody who can turn your ideas into a tender; get financing; find the programmers and run the project. Somebody who can deliver the idea, but this time do it right – make sure it meets IG demands and so on and so forth.
Is this something an IT department should be doing? I’ve not heard of any having a go.
I followed the most recent NHS Hack Day in London. It was interesting – people came up with loads of ideas.
I was also involved in the NHS North West web 2.0 accelerator program. It too came up with loads of ideas, including RallyRound which was one of the apps shown nationally the other day at the NHS Innovation Expo.
There clearly isn’t a shortage of ideas to make things better, so why is there so little innovation?
My father, before he retired, was an academic. During his career, universities went through a period of setting up private development companies.
They recognised that their academics were coming up with great ideas, but that they were either leaving to set up companies to try and develop them or failing. They decided to help them; for a share of any success.
Does this happen in the NHS? I’ve certainly never come across it yet. I’ve come across loads of people with brilliant ideas, some of which make it, most of which don’t.
Ok, a lot probably were never going to make it, but perhaps apps and computer programs are like drugs in the pharma world?
Fund the R&D, most projects fail, a few break even, and a couple make a fortune. Only for the NHS, the payback doesn’t have to be monetary – it could just be better clinical care.