16 key points in Gateway Reviews on NHS IT scheme

Key points in the 31 Gateway Reviews released on the NPfIT last week are:

i)  the professionalism and independence of the Gateway reviewers who have been able to take the “big view” while those working on the programme have been confined to small components only.

2)  that ministers have carried on regardless of the foundations-level recommendations and concerns expressed in the reviews. [They probably wouldn’t have been able to carry on regardless if the Gateway Reviews had been made public at the time.] 

3) that the NPfIT was doomed from the start, in Spring 2002. As one Gateway Review put it, many dedicated people were working hard on building the components for a car that hadn’t been designed.  To some extent that’s still true today.

4) that people didn’t really know what they were doing in the first critical months in 2002.  

“The Review Team noted that most of the programme planning completed to date seemed to rely mainly on the listing of milestones and key dates. This is not a realistic basis for planning a programme of this magnitude and complexity.”

5) that the initial plan was for new IT – not for changes to the way people work. So the preoccupation was with IT and not patients. It was hoped that new IT would drive change. But that rarely if ever succeeds. A focus on IT instead of business change means that the IT may get delivered successfully, and the supplier paid- but use of the system by doctors and nurses is limited to administration only and then reluctantly. Said one Gateway review:

There is much talk of what the IT programme will achieve, but little recognition of the potential impact of this on current practices, procedures and systems, both technical and organisational.”

6) that the costs and complexity were initially underestimated – by about £7bn –  because nobody had an understanding of what was needed. The first Gateway Zero put the “whole programme cost” at £5bn. It’s now £12.7bn.

7) that speed was unduly important. One gateway review suggested that key staff didn’t have time to take action on recommendations or learn lessons.

8) that the plan to appoint a small number of local service providers was set in concrete as a result of the Downing Street seminar in February 2002. The plan wasn’t the result of a well-constructed argument in the business case. Said a Gateway Review in October 2002:

“We were told categorically that the OBC [outline business case] was specifically to gain approval for the procurement of the PSPs [primary service providers, now called local service providers].”

9) that there was a “Head Office” mind-set. “The prevailing sense is one of the programme being in an ivory tower” said one Gateway review. It’s now the strategic health authorities who hold the NPfIT power. But Whitehall still pulls the strings.

10) that the Department of Health likes to blame suppliers for the delays and failures. But as one of the Gateway reviews said:

“The days of the supplier being responsible for delivery are long gone and that approach has consistently proven unsuccessful in even small projects. The programme will stand or fall on its ability to act as an intelligent customer.” That was in June 2002.

11) that the perception in June 2002, according to the first NPfIT Gateway Review in that month, was that the NHS’s track record on centrally delivered, IT-related solutions was poor. Which could explain why the Department of Health launched the mother of centrally delivered IT-related programmes.

12) that the programme as a whole, according to one Gateway Review, was not assessed against a list of Common Causes of Failure, as published by the National Audit Office. Only individual projects were assessed against the list.

13)  that there have been extra costs for the NHS of delays which have not been calculated, and those costs are still rising, with abortive training when systems don’t go live when planned.

14) that many implementations are “like for like”.  And for what good? One Gateway review said:

“Progress to date has in some instances been achieved tactically rather than strategically through the replacement of obsolete and ageing systems and meeting contract closure deadlines with essentially like for life systems offering only limited additional functionality.”

15) When Richard Granger was running the IT part of the NPfIT there was firm management of suppliers. One Gateway review said that BT – despite having worked on the programme for more than a year “has not yet presented an invoice that is fit to be paid”. Has that firm grip on suppliers loosened now Granger has gone? BT has negotiated an increase in its NPfIT contracts worth more than £500m.

16) A couple of the sentences in the Gateway Reviews end abruptly. Have they been redacted discreetly without the use of blackout pens? NHS Connecting for Health denies there has been any covert editing. 


On twitter Peter Murray, a nurse and fellow of the BCS, said he was surprised that the NHS’s National Programme had “failed” nine Gateway reviews. “You mean it passed some?” he tweeted.

Murray’s comment was not as frivolous as it sounded. By the time ministers launched the NPfIT in early 2002, 187 pages of specifications for electronic health records had been written.

Several months later a Gateway Review compared the e-records part of the scheme to a poorly-designed car.

The Gateway review said that there had been a “Grand Vision” which a number of enthusiastic people had immediately set to work to turn into something practical, but “without the requirement being fully defined and agreed”. The review added:

“The analogy to a group of specialist car builders who are separately providing the wheels, the body, the engine and the transmission for a car that does not have an agreed design, is probably relevant.”

And yet in the first five years of the programme ministerial statements on the NPfIT were fullsome in their praise for the scheme and gave no hint of any serious difficulties. 

There again, it has been clear that ministers in putative control of the NPfIT have had two overriding priorities. To ensure:  

–  they don’t know the unpleasant truths 

–  they’re not holding the polonium parcel when the music stops


If you don’t like the results of independent reviews of a large and risky project or programme do you simply stop having them?

Ipsos Mori did two annual surveys on the National Programme for IT but produced consistently unfavourable results. It wasn’t asked to do any more.  

Now it transpires that although the Office of Government Commerce carried out a Gateway Review 31 times on subsets of the National Programme for IT in the NHS, only two Gateway Reviews were on the entire programme, and then only at the Gateway Zero strategic assessment stage. 

The OGC wasn’t asked to do any more – despite its recommendation that there should be. 

The first Gateway Review of the whole programme was in June 2002 and gave no traffic light status. The second – and last – was in November 2004 and was at “red”. 


How to fix Government IT – Ex-Government CIO Ian Watmore

The 31 Gateway Reviews released last week – CfH website  

Gov’t ordered to release 23 gateway Reviews on ID Cards and other projects – Computer Weekly  

CfH’s Martin Bellamy is promoted to Cabinet Office – E-Health Insider