House of Commons - Public Accounts

Select Committee on Public Accounts Sixty-Second Report

THE PURCHASE OF THE READ CODES AND THE MANAGEMENT OF THE NHS CENTRE FOR CODING AND CLASSIFICATION

THE PURCHASE OF THE READ CODES

9. Since the 1980s, NHS Hospitals have used clinical coding systems to provide central returns of their activity to the NHS Executive. These developments were largely driven by the requirements of resource management and central returns, but computerised clinical information systems had wider potential. Data would be entered into the system once only, at the level of direct patient care, and would then be available to meet the information requirements of direct and indirect patient care, and of management at local and national level, improving quality and reducing duplication and scope for error. To enable clinical data to be readily retrieved, analysed and aggregated there was first a need for an agreed thesaurus of clinical terms. The next step was translation of the terms into codes for incorporation into clinical computer systems.[8]

10. In 1987, the Joint Computing Group of the Royal College of General Practitioners and the General Medical Services Committee of the British Medical Association established a technical working party to consider whether a coding system could be adopted for use in general practice. The Group considered the coding systems which were available at the time, including the Read Codes, which had been developed by Dr James Read, a general practitioner who practised in Loughborough, Leicestershire.[9]

11. The working party recommended that provided the Read Codes were developed to comply with its requirements, they should be adopted as the basis of a United Kingdom standard classification of general practice data. They considered that the full benefits of a standard clinical classification would only become apparent if adopted throughout the NHS. They therefore recommended that the implications of their report on the use of coding systems should be considered across the NHS. The NHS Executive accepted these recommendations.[10]

12. With the Joint Computing Group, the Executive concluded that they would not be in a position to work with the clinical professions to ensure the development of the Codes with their content under professional control while ownership effectively remained with a private individual. They therefore opened negotiations in 1989 with Dr Read with a view to purchasing the intellectual property rights to the Codes.[11]

13. In December 1989, the NHS Executive reached a provisional agreement with Dr Read for the purchase of the Codes, together with all related hardware and software, for a total of £600,000. The arrangements for the purchase included the establishment of a separate selffinancing body to control, develop and maintain the Read Codes and to ensure their commercial exploitation.[12] This proposal was subsequently withdrawn. Following further negotiations, the Executive agreed in March 1990 to purchase the worldwide rights to the Codes from Dr Read for a revised price of £1.25 million.[13]

14. The National Audit Office could find no documentary evidence of the reasons for the withdrawal of the original price of £600,000. But the NHS Executive told them that, following informal discussions with the Treasury, the Executive preferred to seek an arrangement whereby charges to NHS bodies using the Codes would not have been set so as to recover the costs both of CAMS Limited and of the body developing the Codes.[14]

15. The Committee asked why there seemed to be such a leap in price. The Executive told us that the first price did not involve the NHS taking responsibility for the Codes. A freestanding, selffinancing body with Dr Read as its Director, would have developed the Codes, and this would have meant that NHS users would have paid a price that included the cost of developing the Codes. The price paid, £1.25 million, reflected the policy intention of ministers to reduce the cost to NHS users. Under the agreement, the NHS Executive developed the Codes and funded that development centrally and NHS users took financial responsibility only for the installation and maintenance of the Codes.[15] We asked if these arrangements were an attempt to make the costs of the Read Codes less visible. The Executive told us that from the outset, there had been a policy decision that the Codes should be developed within the NHS and centrally funded.[16]

16. We asked whether the purchase represented value for money. The Executive told us that the people who made the decision judged that it represented value for money, and that in retrospect they had no basis on which to demur from that judgement.[17]

17. The NHS Executive did not prepare their own business case for the purchase of Read Codes and the Joint Computing Group's proposals had not fully evaluated the likely costs of the Codes.[18] The Executive told us that they had not undertaken a full cost benefit analysis, but had accepted the recommendations of the Joint Computing Group. The Group had looked at eight different systems and had chosen the Read Codes as being the most suitable for development; and had identified the benefits in some detail. The Read Codes were the only system that was generally acceptable to the profession. [19]

18. We asked how much the NHS Executive had spent on the Read Codes. They told us that in total expenditure was some £32 million to the end of 199798 (Table 1). [20]

Table 1

Expenditure on Read Codes to 31 March 1998

£m

Purchase
£1.25

Development
£22.20

Sub Total (Spend by the NHSE)
£23.45

Licence Fees paid to CAMs Ltd
£8.85

Total
£32.30

19. We asked about the cost of clinicians' time spent on developing the Codes. The Executive told us that the £32 million included £3.8 million to develop the Version 3 thesaurus. The bulk of this cost was in servicing 55 specialty working groups. But while the Chairmen of these groups received honoraria, the other clinicians involved only received reimbursement of expenses. The Executive had no estimate of how much time these clinicians had committed.[21]

20. We asked about the implications for the future of the Read Codes of a report commissioned by the NHS Centre for Coding and Classification, in which a consultant estimated the cost of installing Read Codes Version 3 in general practices at £100 million. The Executive agreed that it was now time to undertake an independent evaluation of the project, and any decision to abort the project was one for ministers, especially given the strength of clinical support.[22]

21. At the time they purchased the rights to the Read Codes, the NHS Executive's delegated authority for expenditure on information technology projects was £1 million. While the purchase was not an IT project in the normal sense, it was an investment intended to add value to NHS clinical systems. There is evidence that the Executive consulted the Treasury about the purchase of the rights to the Read Codes. But neither the Executive nor the Treasury could produce documentary evidence that the Treasury approved the final agreement.[23] We asked the Executive whether the Treasury approved the purchase. They confirmed that there was no documentary evidence of Treasury approval, and that the only evidence of contact with the Treasury was in internal Executive papers that made references to discussions. They accepted that this was wrong.[24]

Conclusions

22. The NHS Executive withdrew their original offer of £600,000 for the Read Codes and, shortly afterwards, paid more than twice as much, £1.25 million. We note that the intention of the revised deal was to ensure that the costs of developing the Codes would be borne centrally and not by users. But we are concerned that the Executive were unable to provide documentary information to confirm the assessments made at the time, nor a convincing analysis of whether the extra cost of purchase represented value for money to the taxpayer.

23. Total expenditure on Read Codes to March 1998 was some £32 million and implementation of the Codes may cost many times more than that. It is a serious failing that the NHS Executive did not prepare a properly structured business case, including a cost benefit analysis or investment appraisal, to justify their decisions to purchase the Read Codes, to set up the Centre for Coding and Classification and to give CAMS exclusive rights to distribute the Codes.

24. We have commented before on the failure of the NHS Executive to adopt sound business planning for investment in information technology, in our reports on the Hospital Information Support Systems Initiative[25] and the Wessex Regional Health Authority Regional Information Systems Plan.[26] It is a serious failure of management by the NHS Executive that again it did not adopt sound project management arrangements. We look to them to ensure that full appraisals underpin all future NHS investment in information technology.

25. We are surprised that the Executive had no documentary evidence to show that they had received Treasury approval for the purchase and development of Read Codes. This lack of clarity reflects badly on the Executive but we note that any uncertainty about the NHS Executive's delegated powers has now been resolved.

8 C&AG's report, (HC 607 of Session 1997-98) paras1.4-1.6 Back

9 ibid, paras 1.7-1.8 Back

10 C&AG's report, (HC 607 of Session 1997-98) para 1.12 Back

11 ibid, paras 1.14-1.15 Back

12 ibid, para 1.17 Back

13 ibid, paras 1.17, 1.20-1.21 Back

14 ibid, para 1.20 Back

15 Q36, 168, 215 Back

16 Qs 14-20, 24 Back

17 Qs 36-37 Back

18 C&AG's Report (HC 607 of Session 1997-98), para 1.31 Back

19 Qs 152-157, 169, 198 Back