Evidence submitted by Computer Weekly
(EPR 64)
Computer Weekly supports the main objectives
of the NHS's National Programme for IT NPfIT], particularly the
aim of providing an accurate, useful and up to date electronic
health record that is always available to clinicians when they
need it.
Some of our readers have told us how the treatment
of their relatives was not what it should have been because the
paper records were not available.
However—and this is one of our biggest
concerns about the NPfIT—strong, well supported objectives
do not justify a flawed project.
For more than 15 years we have investigated
the common factors in IT-based projects and programmes that succeed
or fail to meet expectations. We hope the NPfIT succeeds and there
have indeed been some specific successes. On the whole, however,
the programme has already fallen into traps that have ensnared
some other large, high-risk IT-based projects and programmes.
There are doubts it will work as originally intended.
For these reasons we believe that there should
be a published and genuinely independent review of the NPfIT.
If all is well the results of the review may placate those who
are concerned about the £12.4 billion programme. It may also
provide evidence with which to market the programme's benefits
to doctors and nurses.
If all is not well, a review would provide evidence
of the need for a rethink; and the sooner the better, before billions
are spent without commensurate benefit to the NHS and patients.
In 1998 Computer Weekly urged the Transport
subcommittee to ask the government to commission an independent
review of a project to build air traffic control systems at Swanwick
in Hampshire. That review's recommendations arguably made the
difference between success and failure.
The Swanwick project had then been running about
four years behind schedule. Our call for an independent review
was opposed strongly by National Air Traffic Services. Its directors
told the Transport subcommittee that a review would be an unnecessary
distraction—the argument put forward by the Department of
Health in opposing a published independent review of the NPfIT.
In the case of the Swanwick systems, the government decided in
1999 to commission independent reviews in opposition to National
Air Traffic Services.
The Department of Health's rejection of a published
independent review of the NPfIT is one of our main concerns about
the programme. We are advised by senior members of the Association
for Project Management that a high-level review of the NPfIT could
be carried out in less than a month and by no more than six people
if they are given access to the right documents and senior personnel.
It would establish the health or otherwise of the programme.
Below we list some of those traps and some of
chief concerns expressed to us about the NPfIT:
1. The delay in the delivery of core software,
which in the case of the NPfIT includes a Care Records Service,
has proved in some past projects to be a symptom not of teething
difficulties but deep-rooted defects in design, scope and ambition
that could affect the successful outcome of the programme.
2. Delays in the delivery of core software
were symptoms preceding the failure of the Libra project, the
main aim of which was to install a unified case management system
for magistrates' courts across the country. Today, more than 15
years after the initial project attempt failed, no single system
has been rolled out across the country, though there have been
several attempts. One problem is complexity: magistrates' courts
have different ways of doing things—as have hospitals. And
despite Libra's complexity, it is a tiny project compared with
the NPfIT. Libra's case management system is a fortieth the size
of the NPfIT.
3. Long delays in the successful delivery
of core technology were also symptoms in the failure of a system
for the Performing Right Society. The Society's overly ambitious
PROMS system—Performing Right On-line Membership System—had
to be scrapped, largely because the new business processes that
accompanied the IT had not been thought through adequately. We
can provide other similar examples if requested.
4. Two major suppliers have withdrawn from
the NPfIT programme for reasons that have never been fully ascertained.
We believe that Accenture and IDX should be asked why they withdrew.
Over-optimism among suppliers and government agencies is a common
factor in IT-related failures. For example, after a Single Payment
System failed to meet its original objectives at the Rural Payments
Agency, two ministers told Parliament they were given over-optimistic
reports from their civil servants on the state of the IT project.
We believe that directors of suppliers who decide to pull out
of projects should have their concerns aired and taken into consideration.
Instead, it seems to us, the views of executives at Accenture
and IDX have been sidelined or not even sought.
5. The complexity of the NPfIT, and whether
its objectives are achievable as originally configured, may have
been factors in the withdrawal of Accenture and IDX. There are,
for example, concerns about whether it is possible for software
that has been designed for a small number of individual sites
to work at many health sites with diverse needs.
6. We and the National Audit Office have
studied common factors in project successes and some things stand
out: a clear objective (which congestion charge has for example)
and simplicity of design and ambition. The NPfIT, it can be argued,
does not have a clear objective and it has labyrinthine complexity.
Its ambition, complexity and cost make it the largest civil IT-based
programme in Europe and perhaps the world.
7. Truth and openness become the first casualties
in projects that are in serious trouble. Defensiveness and a resentment
of criticism—and the critics themselves—have characterised
the NPfIT.
8. The Department of Health and Connecting
for Health have put much information in the public domain, but
not the key facts we have requested. Our NPfIT-related requests
under the Freedom of Information Act have been rejected. And though
Connecting for Health publishes statistics it does not publish
any details of the specific high severity incidents that affect
the hospitals that install new national systems. These incidents
can affect the care and treatment of patients.
The Department of Health has declined to publish
the results of its reviews on the NPfIT, including those undertaken
by consultants and, separately, by the Office of Government Commerce.
If any of these highlight fundamental flaws in the programme,
it is in the public interest that they are published.
IN SUMMARY:
We would expect Connecting for Health to argue
that our concerns are misplaced and that the programme is headed
for success. To this we would suggest that it publishes evidence
of success in the form of the published results of a genuinely
independent review. We believe that the review carried out by
consultants Arthur D Little report on the Swanwick air traffic
control systems was a model exercise. As we said earlier, we believe
that this report made the difference between success and failure.
We would like the NPfIT to succeed, and so we
would urge the Health Committee to ask the government to commission
a published, genuinely independent of the NPfIT.
Tony Collins
Executive Editor
Computer Weekly
19 March 2007
| 
