The Government's plans for NHS ITmay be well-meaning, but the basis
for change is not supported by the available evidence, even in the
Wanless Report, and it will harm the service and patients
There is no more pressing priority for the Government than
improving the NHS - if possible before the next election. It has
less than three years. The money is available, although increases
in salaries may absorb more than had been expected. How best to
spend what is left? Surely to improve the lot of the patient?
Apparently not. The Government has chosen a course that is likely
to make it worse: sweeping and massively expensive changes to NHS
computer systems. We are told it is "the IT challenge of the
decade" and "a Herculean task".
Why don't people learn? Why are big IT projects seen as a badge of
virility or a sign that we really mean business? They nearly always
cause trouble: the bigger the change, the bigger the trouble.
Difficulties with the Government's earlier IT plans (this is the
third) demonstrate that the risk is especially great for such a
uniquely complex organisation - employing 1.3 million people with
50 million potential patients.
Ambitious IT changes rarely deliver what is promised and commonly
cause serious inconvenience for those they are intended to benefit:
in this case, the patients. Surely anyone who wishes the NHS well
would be striving to introduce the minimum necessary IT change, the
smallest possible challenge?
This is not a Luddite rant. Computing systems are an essential part
of healthcare delivery. There is undoubtedly a case for extension,
innovation and improvement and extra funding is plainly needed.
But, particularly for the NHS, plans for change, however desirable,
must be balanced against risk - and, where there is serious
uncertainty, doing the minimum necessary must be the best
course.
In contrast the recently published Department of Health plan,
Delivering 21st Century IT Support for the NHS, sets out a huge
programme involving massive risk. Yet the case for the programme is
not, to use current medical jargon, evidence-based.
It starts with a "vision". This word, along with "integration" and
"centralisation", is one of the most dangerous words in computing.
Central control and "ruthless standardisation" will bring about a
wonderful new world where health professionals and managers will
have instant and simple access to a wealth of information (case
histories, test results, research data, resource services, etc)
designed to support the patient "quickly, conveniently and
seamlessly".
This dream requires a major new NHS-wide IT infrastructure, a new
procurement strategy and centrally-defined data and system
standards, focusing initially on national health records, booking
systems and prescriptions. It sounds splendid. But such plans
always do, particularly when technically naive senior civil
servants, in alliance with enthusiastic industry representatives,
are painting an idealised picture for ministers. That is before the
dull practicality of the real world intervenes. Here are two
examples:
- Ruthless standardisation means that perfectly good but
non-standard local systems - often introduced after much trial and
agony, that are at last working and serving staff and patients -
will have to go. There are many such systems. Is dismantling them
really a good idea? Is it desirable to pile new problems and
"challenges" on health professionals and management, let alone the
patient?
- Electronic patient records are a critical component of the
programme. The concept involves huge problems: health information
is far more complex in nature and detail than, for example,
financial information. The Government has already experienced
difficulties - although 35% of NHS trusts were supposed to have
implemented these records this year, so far only a handful have
done so and the target of 100% by 2005 looks increasingly
difficult. And concerns about data privacy and human rights are a
growing worry, particularly regarding such a personal matter as
health. Recent ID card worries suggest that a centralised system
for health records would exacerbate these concerns.
So an exciting vision risks damage and disruption for an already
vulnerable healthcare service. The Government even recognises this:
the NHSITplan notes that "significant risk will be involved". And a
senior health department official recently described it all as
"incredibly ambitious, we're betting the farm on this". But why?
Where is the evidence that such a risk is justified?
What is envisioned would clearly be desirable. But, to justify a
huge gamble with the nation's healthcare, the potential outcome
must be more than desirable - there must be plain evidence of major
and achievable benefit. No other test will do. The NHS IT plan
provides no such evidence. Perhaps that was not its function: it is
a plan for action. For the strategy we must go elsewhere.
The Wanless Report was commissioned by the Treasury to examine
healthcare funding and gave prominence to the need for much greater
investment in IT. Delivering 21st Century IT Support is the
response to that. Key Wanless recommendations are that IT spending
should be doubled (and protected to ensure it was not diverted
elsewhere), that national standards for data and IT should be set
centrally "and vigorously applied" and that investment should be
aimed at "better integrated and more flexible" IT.
So far as funding is concerned, the principal justification is that
spending per employee is lower than in other sectors of the economy
and is less than is spent in overseas healthcare services.
Doubtless true - but not of itself an argument for spending more.
Clear evidence demonstrating the likelihood of major benefits
coming from greater funding and supporting the centralise/integrate
theory is needed. But there is no such evidence.
Instead there is assertion: "The benefits of ICT [ie IT] will not
come through significantly until the necessary infrastructure is
built." That is despite a statement towards the end of the report
that "decisions to invest in ICT need to be accompanied by firm
evidence of the costs and benefits". Exactly.
Unfortunately, although it notes the "clear risk given the scale of
such an undertaking", the Wanless Report fails to provide that firm
evidence. The closest it gets is its comment that evidence (coming
from Kaiser Permanente, a US healthcare provider and currently
controversial government favourite) "suggests that significant
benefits are achievable".
In the light of the potentially damaging outcome of what is now
planned, a mere suggestion is quite inadequate.
To gamble with the future of the nation's healthcare is arguably
unwise in any circumstances. To do so when the chances of success
are low is irresponsible. To do so when the costs of even a
successful outcome are high and its value uncertain must be
foolish. We seem to be embarking on a course that is both
irresponsible and foolish.
Wanless may be right about the inadequacy of NHS investment in IT.
Probably greater expenditure is needed. If so, where would it be
most beneficial in a reasonable timescale? My experience is that it
is usually best to start from the bottom and work up - the
antithesis of what is proposed. Identify the best local examples of
effective IT-enabled healthcare delivery in the NHS (not in
California) and build carefully on those. I am no expert on NHS IT
but there are many who are - including some clinicians - they
should be heard. There may be some who believe that additional IT
expenditure is not the best way of delivering a better service to
the patient. They also should be heard. In other words, we need a
debate.
Some months ago, the chancellor spoke of his wish for a great
debate about the future of healthcare in Britain. It has not
happened yet. But, as the programme defined by the NHSITplan does
not get fully started until April 2003, there is time for a
widely-based and informed debate about whether these proposals are
"a risk too far". It would be widely welcomed by NHS staff,
healthcare professionals and the public.
Robin Guenier is chairman of iX Group, a company that uses the
Internet to provide services to the medical sector. In 1996, he was
chief executive of the government Central Computing and
Telecommunications Agency and was appointed executive director of
Taskforce 2000 by the Department of Trade & Industry