As a consultant radiologist at specialist cancer centre, Dr Rhidian Bramley gathers crucial information about patients.
It is the scans that he and his colleagues at The Christie NHS Foundation Trust perform that enable fellow clinicians to decide how best to treat a patient.
Yet, as Dr Bramley told delegates at the first CCIO Leaders Network Annual Conference, there used to be instances in which such information was not been seen or acted upon.
The consequences were serious: “There were incidents where this led to patient harm or death,” he told the event, which ran alongside EHI Live 2012 in Birmingham last week.
See a problem: solve it
As well as being a consultant radiologist, Dr Bramley is the trust’s chief clinical information officer. It was in this capacity that he sought to address the problem of results not being acknowledged by clinicians.
He felt certain it was a problem that could be easily addressed by an existing software solution. “Results acknowledgement should be simple to do, so we went to another site to see how someone else had done it,” Dr Bramley told delegates.
“We asked how long it took for clinicians to acknowledge results; but they weren’t actually doing it - the system didn’t support the workflow. Others sites didn’t know the percentage of clinicians that were acknowledging results.”
Such conversations continued for two years. Then, his organisation made the “revolutionary” decision to develop the software it needed in-house.
“Clinicians told us what they wanted and a colleague did the coding. He showed it to clinicians, they piloted it, word spread and more and more clinicians asked to use it. Within two months, we had 100% result acknowledgement across the trust.”
Staff at The Christie now use a number of information portals created in house - including a picture archiving and communications system portal built by Dr Bramley himself.
“We have had a big strategic shift and are now doing much more in house development,” he said.
“I've argued quite passionately within our organisation that there needs to be core capability to do some development. Setting up the development team is probably one of the most exciting things I've done as CCIO.”
It is also something that many other trusts are considering, and which was a source of much debate at the conference.
Should NHS organisations have in house development teams creating solutions that work for their own clinicians? And, at a more basic level, should those clinicians - and the CCIOs leading them - be able to code themselves?
The Christie is not the only trust to have decided that the answers to these questions are ‘yes’.
Dr Jonathan Richardson - clinical director of informatics at Northumberland, Tyne and Wear NHS Foundation Trust - told delegates that his organisation’s clinical dashboard app had been created by an in house team “probably at a fraction of the cost of it if we’d gone out to buy one.”
A similar story was told by Dr Jack Barker, CCIO at King’s College Hospital NHS Foundation Trust. The London trust has now created a number of applications, with Dr Barker leading development in some instances.
“I was hired to some extent to run the lung cancer service and I knew from the beginning we needed to collect data on quality and performance,” the consultant chest and general medicine physician told delegates.
“So I got into programming a bit - Access databases. Then we needed a web front end, so I worked with the IT team to learn how to do that and built a respiratory system that helps collect all the data needed.”
The trust’s clinical dashboard is now used every day as part of nursing handover, and most of it was built by Dr Barker with support from the trust’s IT team. Similarly, the organisation’s vital signs monitoring application was built by one of its clinicians.
Colin Sweeney, the ICT director at King’s, clearly explained the benefits of clinicians and his staff being practically involved in the development of informatics solutions. But he also told delegates of his worries about this approach.
“Everything that we develop we’ve got to support,” he pointed out. “So that does become a whole resource issue. That’s the sort of thing that keeps me awake at night - what happens if certain developers leave the trust?
“With the vital signs application, I was very concerned about it being used trust wide and its need to be supported by more than one man and his dog. So we have an external company backing us up as far as support is concerned.”
The world is changing
King’s, The Christie, and Northumberland, Tyne and Wear might all have found ways to embrace clinicians who code and in-house development teams who create applications. But others at the conference questioned whether this is something the NHS should be doing.
Dr Paul Altmann, CCIO at Oxford University Hospitals NHS Trust, told delegates the stated desire of Tim Kelsey, the new national director for patients and information, to create clinicians who can code is a misguided one.
“It’s completely irrelevant whether clinicians can or cannot code,” he argued. “What’s more relevant is that they ask not how - the coding bit - but why; start at the beginning and not get seduced by being able to code something.
“We need to know what it is we want to be able to achieve and then work with better qualified colleagues in industry or in house to develop it.”
It was a viewpoint shared by Dr Sajjad Yacoob, director of medical informatics at Children’s Hospital Los Angeles.
“I have no idea how to change a carburetor, but I drive a car,” he said. “You don’t have to be a programmer but you need to have some understanding of how to use those tools.”
Even those who did not feel that coding ability should be a requirement for clinicians and CCIOs acknowledged it may not remain that way for long, however.
“One of the real challenges is not just to do something about how it is now but to try to predict how it's going to be,” argued Anne Cooper, national clinical lead for nursing. “We need to be able to envisage what it’s going to look like in ten years’ time.
“Would I like to see nurses who can code in the next decade? Yes. Do I think all nurses should be able to code? No.
"It is helpful to have deep dive [clinical staff] who can do the coding part but [for now] I don’t think everyone needs to be able to do it. I think leadership is much more important.”
why wait - agile developmentehireader18to14to12 108 weeks ago
Developing a "computer system" is not that different from building a hospital - you start from the bottom up, get the foundations right, and work hard. Trusts do not have the apppropriate resource or structures to do this (the resource is too expensive and the structures are, quite rightly, clinically driven) - Trusts need to retain Teams to "move" information (reports etc) around, development of Trust computer systems (including those giving patients and service users access to their info.) needs to take place elsewhere (these organisations, unlike some others, need to be built from the bottom up, with a flattened hierarchy and little management).
The future is in the hands of children, they are the best teachers (the principles of GUI were based on the way children played with the pointer on a screen), they live in the present. IF the NHS is to have development teams it should build them now and do it quickly, with regard to patient service user information, waiting should not be an option, it is racing away and the patients service users are losing out.
Programmingmlysons 108 weeks ago
The more buy-in from clinicians the better, but leave the software development to those trained to do it. Communication is the key element here, not training what will be the end-users to build the products themselves.
I think this raises a wider point that Trust in-house development teams point towards a future NHS in-house development team.
The inbalance between technical and management skills needs fixingehireader18to14to12 108 weeks ago
Iam a software developer who has worked in the healthcare sector for more than 25 years, recently for the NHS, currently in an IM&T department on a salary of just over 30K, I have been told software development is not one of its primary functions, which means I don%u219t fit (why they took me on, I don't know, cheap probably) and need to move on, the numerous contractors they have taken on a short term basis on a reasonable rate will fulfil the coding function. Just for interest my first job was with Hoskyns, I was taught 40% design, 20% code, 40% test (languages are more powerful now, but I still feel those figures are about right (maybe 40/15/45)).
Without a shadow of a doubt, there is an acute shortage of coders (not just in the NHS), this will get worse in the foreseeable future. Clinicians and technicians need to work together %u213 anyone who puts up barriers to prevent this happening, needs to take a sideways step. Thank you Dr Bramley for being honest.
If the clinicians and nurses are keen to code let them, encourage them, train them. Let the technicians work directly with the clinicians, trust them, pay them a fair wage, not a clinicians but a fair wage (something like a band 7/8 manager). Remember generally speaking when a clinician spends time with patients service users it generally is on a 1:1 basis, a coder could potentially write a chunk of code which could indirectly help many patients.
By the way I do not believe the failure of NpfIT was down to the software engineers, but due to the huge numbers of hangers on, or to put IT more bluntly, talkers, not doers.
Coding or clinical termstimbenson 108 weeks ago
I agree I confused the terms! But it is a simple fact that the most successful clinical systems in UK and overseas were originally created largely by doctors. There are two reasons. The selection process for medical schools tends to select people with high ability and it is much easier to write useful software if you really understand the problem you are trying to solve. Enormous sums of money have been wasted due to the failure of software engineers to fully understand why each requested feature was needed. In his "Mythical Man Month" published 40 years ago Fred Brooks proposed that the division of effort on any computer project should be: one third to planning and requirements, half to testing and tweeking, but only one sixth for coding. Sounds about right to me.
Big teams do not give good resultstimbenson 108 weeks ago
Another of Fred Brooks' insights was that if you add resources to a late project you make it later. This is due to the volume of internal communication getting in the way of doing the real work. Most successful projects have only a couple of people responsible for the key effort. NPFIT certainly was a victim of the mythical man month tar pit.
Not wishing to argue but...Groundhog Day 108 weeks ago
I agree entirely with the division of effort on a programming project, indeed when I entered the industry 20 years ago I also heard the same, and the reality of developing software would indicate that this would be a good starting point.... BUT
Can you evidence your assertion that enormous sums of money have been wasted due to the failure of software engineers?!
You shouldn't confuse the debacle which was CfH/NPfIT with developing code that didn't work. The bigger issue I think you will find is that clinical staff do not have standardisation of business processes within secondary care which is a much bigger issue than developing badly designed code.
CCIO must understand what programming is all aboutRupertFawdry 108 weeks ago
Massive amounts of money is not wasted by software engineers but by clinicians, managers and politicians who believe computers are magic and who fail to understand that they will only get good clinical systems if they have a real grasp of what writing software is all about. Anyone involved in ICT including doctors, managers and politicians ought to have written some programs. Once they understand what is needed THEN they never need to write programs again but they will be able to make better ICT decisions
The world gone mad!Groundhog Day 109 weeks ago
Forgetting the fact that I work as an IT specialist in healthcare, as a tax payer and user of the NHS I find the idea of a clinician writing software absurd and totally wrong..
As a tax payer I have paid for the medical training of these people to become clinicians not programmers..
When after being on an expanding waiting list, I attend an outpatient clinic to be told that Dr. Jones is running two hours late, I expect that it is due to them being overworked treating patients, not because they are only working 2 days a week seeing patients whilst hacking code for the other 3!.. this is ludicrous!
I fully support Dr's being involved in the analytical design of software, but seriously leave the coding to the professionals. I have worked in enough hospitals to have seen the impact of these little islands of information which once Dr. Jones has retired become unsupportable for either a trust or external supplier.
Microsoft Access had a lot to answer for in the NHS...
Clinicians can codetimbenson 109 weeks ago
The proof is in general practice, but when GP computing started many people thought that they would never be willing to do it. The original Exeter GP system, which bequeathed us the FP10 (Comp) prescription form, did not support coded nots, but Dr John Perry of the Oxford Community Health Project, developed the OXMIS codes and showed it was practical. His work was then extended by James Read with the eponymous Read Codes, which every GP uses. The secret to coding is to make the term list fit for the task in hand. General practice is just one specialty among 60; we have to recognise that we need another 60 people like James Read or John Perry. Note this means 60 obsessive individuals, not 60 committees.
Need for 60 obsessive individualsRupertFawdry 108 weeks ago
As you know couldn't agree more obsessively
Without obsession - and using my own finances - could not have achieved so much with the http://www.eepd.info and http://www.wisdam.info initiatives. Main problem is that 1 in 5-10,000 GPs adds up to a reasonable number of doctors with ICT knowledge and skills - but 1 in 5-10,000 obstetricians adds up to about 1 (if you know any other obstetrician worldwide with a similar obsession do tell them to get in touch
use of the term coding is confusingIt is I, LeClerc 108 weeks ago
The author's use of the term coding is causing some confusion here, he means programming, but coding (poatient data) by climnicians is such a relevant topic that the headline is a little ambigious.
But re clinicians programme computer code, why would they. surely they(us) have spent a fortune training them to be doctors etc, arn't they best treating people. Sure let them specify, lead and own IT developments, but programmers are cheaper than doctors.