Top down never works – it is time to let patients succeed with bottom up

Earlier this week, the UK’s National Audit Office issued its report The National Programme for IT in the NHS: an update on the delivery of detailed care records systems. The country finally had a formal admission that the top down strategy had failed, and would never work:

The rate at which electronic care records systems are being put in place across the NHS under the National Programme for IT is falling far below expectations and the core aim that every patient should have an electronic care record under the Programme will not now be achieved.

For years, everyone knew the failure had happened, and for years before that, many warned that it would happen. I was one of the latter.

My worry is the government will think the problem was with the decisions that were made by central planners, or with the central planners making the decisions. Neither is not true. It is that the decisions were made by central planners. Top down programs simply do not work.

The only solution that will work is a bottom up one. And the only bottom up solution is a patient-controlled one. That means the opportunity is there for patients to help themselves and help the health care industry. By taking control of the records, they can fix the system. That is why we built Patients Know Best.

For me personally, the new announcement brought back memories.

I dug up an article I wrote for the back in 2001, about NHSNet, the forerunner to 2002 National Program for IT, that the NAO admitted had failed. The article was titled Why NHSNet can never be good, and getting it published was thanks to the kindness of Dr Jeannette Murphy from UCL because I was young and new to the field of health informatics. She supported my writing even though, as a first year junior doctor, I did not know much. But the principle was still clear: top down programs cannot succeed. By the time the National Program for IT was announced, I knew that all the innovation that had made the UK a world leader in health care IT would freeze as the centre took over. In frustration, I left the country to continue to improve my skills.

But by 2008, enough of the lessons were becoming clear enough even for those at the centre that they began asking for innovations from start-ups. And so I returned to the UK to found Patients Know Best. We had an opportunity before us to make bottom up work, and to do so by putting patients in control.

I leave you with the text of the article for historical entertainment.

In 1991, I lived with my grandmother in Bahrain. The room she gave me was originally my uncle’s, which meant I had access to all his books. This is a true education.

One book I read was about the Fifth Generation effort in Japan. Written in the 80’s, it described the Japanese government’s efforts to catch up and overtake the USA in IT. The book spoke of billions f dollars and hundreds of bright minds, all devoted to this giant governmental vision.

I was gripped.

All through the 90’s, I followed the Japanese computer industry for signs of their inevitable success. Nothing materialised. The billions spent have done nothing to change the nation’s fortunes. Indeed Japan’s gap widened as it was late to catch on to the internet revolution.

It is remembering this that I think of NHSNet.

What is NHSNet?

NHSNet is network designed with the NHS in mind. Its guiding principles were clinicians’ needs for a system to cut through the paperwork of healthcare. This system would be fast, delivering lab results instantly to the doctor that ordered them. It would be secure, insuring the confidentiality of clinical information. And it would provide timely trusted information that doctors would use to make clinical decisions.

What went wrong?

From the start, there were problems. This led to several rounds of delays and extensions. Now, several years on, the governments targets of having all GP’s on the system have still not been met. Let’s start with the governments’ plan. It was, august. A bold vision ideal for a politician’s speech. Consider X400, the standard proposed for transferring messages such as lab results. The requirements were so large and complex, that to date, only one manufacturer has implemented them – Microsoft.

Furthermore the standards set out were only really accepted in the UK. The market available is thus immediately limited and unattractive in size considering the investment required.

Contrast this with Health Level 7 (HL7). This is an international set of standards for medical software. As you can guess by the name, there are 7 levels of compliance. From the outset, it thus provided software developers with a gentler slope to climb, rather than the all-or-nothing mentality of NHSNet. In addition, this is an open international body. Developers that stick to it know that their market is worldwide. It is with this understanding that so many solutions have been developed and deployed worldwide.

Secondly, this effort has created monopolies. The knowledge that X400 compliance will be forced upon them has caused most health authorities to switch to one provider – Microsoft. Furthermore all spending on IT solutions other than NHSNet’s has been frozen. Innovators in the UK cannot find a market at home for their products without the government’s blessing.

By contrast in the US, the kinds of solutions that NHSNet was meant to provide are being met by several small innovative companies. With time, the more successful will acquire the less successful, or consolidate with bigger companies. HL7 compliance means their software may work with others’ without a monopoly arising.

Thirdly, the funding priorities were wrong. GP’s must invest thousands of pounds to ensure their surgeries are compliant, and wait several years before they can see the investment recouped in savings. Understandably, health care trusts have seen the scheme as a burden rather than an opportunity, and have only taken it up due to legal requirements.

More controversial is whether or not there will be any saving at all. There are some areas where large cost savings can be made quickly, and with a low initial investment. Further, the investments must be made in the right sequence, to affect the limiting steps in order of importance. Rather than identifying and targeting these, NHSNet’s creators pursued complete compliance with the grand vision from the start.

It’s the government , stupid

The main problem behind all this is that big government efforts do not work. It’s not about the Conservative or the Labour government. It’s about government. So this time round, I shall not wait eagerly for big results to flow from big efforts. I know they never will.

Published April 2001 as editorial in the newsletter
of the British Medical Informatics Society


  1. One problem is that some surgeries cannot even run an efficient administration which means test results get mixed up, different GPs do not read notes properly, receptionists are left to deal with frustration and upsets.but also it means people lack confidence in medics being able to run sites such as PKB
    Also it does seem so far that the service benefits those who are well enough off to own IT equipment, even a smart phone is out of reach of many, plus there is a £100 fee which is an impossible charge to many individuals. Having said I still think PKB has the right attitude!

  2. Thanks for your comments Susanne,
    a few answers to your points, and then I look forward to the next round of questions!

    1. Usually it is receptionists and secretaries who provide copies of the data to patients at our lead sites. This is not just because they are quite organised(!) but also because it frees up clinicians’ time to focus on other clinical tasks.

    2. The service is available to anyone who can reach a web browser. There are more mobile phones than people in the UK, and “smartphones” does not mean iPhone, it means most phones on the market, including ones which are available free of charge with a contract. Having said that, patients can use PKB in a relative’s home, on their library computer, or even a Nintendo Wii. Anything with a web browser is enough to get going.

    3. £100 fee is for patients who want to pay directly, but many have asked to be able to start without having to wait for their clinical teams. But clinicians who pay for the software on behalf of the patients tend to get institutional subscriptions with much lower prices per patient, and they save money – and improve care – by working in this way with their patients.

Leave a Reply