Annex A > Chapter 19 - Statistics Relating to the Clinical Performance of Paediatric Cardiac Surgical Services in Bristol Compared with Other Specialist Centres during the Period 1984 to 1995 > Other statistics relating to clinical performance > Patient Administration System
<< previous | next >>
Patient Administration System
57 The principal source of data kept by administrative staff was the Patient Administration System (PAS). PAS was a computerised system for storing, analysing and recording information that was introduced within UBH/T in 1988. Mr Andrew Hooper, formerly UBH/T PAS Manager, in his written evidence to the Inquiry stated:
`PAS is an administrative system only. It was implemented to replace the manual administrative systems, which had been supported by locally developed computer systems in the Health Authority, prior to 1988. PAS does not replace clinical data. It is solely administrative. For example, it provides an index of all patients who have attended the Trust, whether on an in-patient or out-patient basis. It is able to provide information concerning waiting lists and also to print documents, for example, labels and letters, so saving administrative time. ... PAS feeds PAD [Patient Administration Database] with patient based information that is used for statistical information and contract purposes. ... the in-patient module is an administrative tool to enable staff to record the date of admission, the fact and date of transfer to wards, the date of discharge, any change of consultant, whether the patient is being or has been treated at another hospital within the Trust, and referral to another hospital if within the Trust. Most of this information is recorded in "real time", in other words it is recorded as the event occurs.' [92]
58 PAS contained information on patients' diagnoses and procedures. Mr Hooper was asked by Leading Counsel to the Inquiry about how the diagnostic information got onto the PAS:
`Q. So how did the diagnostic information get on there?
`A. The way that it works is obviously a patient comes into hospital, they have their care, are discharged from the hospital, and then the notes one would normally expect to go back to the consultant's secretary to have a summary dictated. That summary should be dictated as quickly as possible after the discharge, although some clinicians are better than others at doing discharge summaries, so the discharge summary would normally be a side of A4, the top half would have the demographic information, the middle portion would probably have the diagnostic information written down, into a main and secondary diagnosis. If they had an operation, there would be a section for the operation details. At the bottom you would have the text the medical staff dictated about the patient's stay. A copy of that discharge summary would obviously go to the GP and any other interested parties. A copy would be retained in the medical records. As soon as that discharge summary has been dictated and typed, those notes would then go off to the clinical coders for that episode of care to be coded. That would be done as quickly as the discharge summary was dictated. Most of the coding clerks would code directly, all the coders would code from the diagnostic and operational information that had been put on to that discharge sheet.' [93]
59 In his written comment on the preliminary overview of data sources published by the Inquiry, Mr Hooper stated:
`... it is not correct to describe it [the PAS] as a "case-based information system". It is an administrative system. The distinction lies in that of the 5, 000 plus users of the system in UBHT 99% of those people using it are only interested in the accuracy of the demographic information and episodal administrative information (i.e. dates of admission and discharge) contained in it. PAS then feeds the statistical information through to PAD [Patient Administration Database]. It has the ability to produce standard reports and utilise an enquiry package. ... it was an administrative system which fed information systems.' [94]
60 Referring to PAS systems in general, Ms Ann Harding, then Acting Director of the NHS Information Authority, told the Inquiry:
`... I think this is one of the problems that we have, the data is collected for the purposes which clinicians believe is managerial, and therefore not relevant to them, and I have a great deal of sympathy for that, because the level of detail at which a clinician would want the information for the purposes of audit is not readily encompassed within the levels of diagnosis and operative coding that we currently have.
`I also think that we have been quite lax in not feeding back to clinicians the information that is being submitted to the central returns about what it is they are doing. When we did give information back to clinicians, they said "I do not do that operation, I do not have that many patients with that diagnosis", and I think therefore one of the things we must do is to find ways of ensuring that the information does go back to them so they in some ways validate it before it is used for other purposes.' [95]
61 Mr Hooper was asked by Mrs Howard, a Member of the Panel, to comment on the setting up of stand-alone systems within the UBHT to support the clinical staff in the information that they wished to collect, in the following exchange:
`Q. You made it clear that this was very much an administrative system, and we have also heard about what I would phrase the "lack of ownership" from clinical staff with regards to that. Do you have any comment about subsequent setting up of stand-alone systems within the Trust to support the clinical staff in the information that they wished to collect? That would be the first part of the question. I would like to explore that after your answer.
`A. Certainly, as far as I am concerned, I have only ever been involved in implementing the corporate systems. I think probably the Trust ... philosophy has always been, where possible, we would like to use those corporate systems, but that is not to say that if a clinician or a group of clinicians wanted to go and purchase perhaps an audit system - they should be able to do that. I think the problem with doing that is, if you are inputting the data into two separate systems, it is always going to be difficult reconciling the information on the two systems.' [96]
62 The data in the PAS was used in ad hoc audit reports, and in making returns to external organisations. Referring to his use of data in the PAS, in his written statement to the Inquiry, Dr Pryn stated:
`... This was a system used by the ward clerks and enabled me to check whether or not the children had been discharged home and seen in outpatient clinics following surgery. This was an indicator as to whether or not they were indeed alive at the time of discharge.' [97]
The CHKS Report
63 In the early 1990s, [98] UBH/T began to use the services of CHKS Limited. Mr Gary Tharme, Sales and Marketing Director of CHKS Limited, wrote to the Inquiry:
`CHKS were formed out of a joint venture in 1990 between the King's Fund and CASPE Research and have been building a normative database of UK NHS acute clinical activity since 1992. ...
`The National Comparative Database was launched in 1992 and now has over 100 Trusts subscribing from all parts of the UK.
`Our aim is to continually improve the quality and use of clinical information in the NHS.
`We compare hospitals'clinical activity with others that are locally relevant. Comparisons can be simple or extremely refined. Typical comparisons can be at Trust, Hospital, Specialty, HRG [Health Resource Group], and procedure or diagnosis level. Target levels of performance can be derived, for instance looking at high performing Trusts. ...
`Trusts regularly send CHKS information about their activity, which we compare with a range of peer group hospitals that they feel, are relevant to them. Standard performance monitoring reports are sent to the client by return.' [99]
64 In a letter to the Inquiry, Mr Rashid Joomun, the UBH/T's Trust Information Manager, stated:
`The Trust provided CHKS with data monthly, which they processed and sent back high level reports in the form of comparative tables. They also provided us with a monthly database on which we could do analysis.' [100]
65 CHKS produced a report, [101] dated 1992, which contained figures of diagnoses, lengths of stay and deaths in relation to the Cardiology and Cardiothoracic Surgery specialties in UBH/T and in a group of similar hospitals for comparison. Statistics relating to paediatric cardiology and cardiac surgery in particular, were not separately identified in the report. [102]
66 In a letter to the Inquiry, Mr Joomun further stated:
`Action was taken to improve data quality. This was mostly centred around diagnostic coding. General Managers would be informed individually of their directorates' problems and it was expected that they would take the necessary action to remedy these problems. ... As far as I know, none of the Directorates had direct access to the routine reports produced by CHKS.' [103]
Hospital Episode Statistics
67 The PAS was used, further, to provide summary data on episodes of care for patients, for the national returns to the NHS known as Hospital Episode Statistics (HES). In his written statement to the Inquiry, Mr Richard Willmer, a Branch Head in the Statistics Division of the DoH, in describing HES, stated:
`The HES system collects records for all (both NHS and private) in-patient consultant episodes of care, including day cases but excluding regular day or night attenders, in NHS hospitals in England.' [104]
`HES was introduced on the recommendation of the NHS/DHSS Steering Group on Health Services Information, which was appointed by the Secretary of State for Social Services in February 1980. The Chairman, Mrs E Körner, published the First Report on the collection and use of information about the clinical activity in the National Health Service in 1982. The report which is commonly known as the "Körner"report, states at section 8.4;
` "The DHSS needed information about bed use for:
` "a. Policy development. Detailed statistical analysis may be required when pursuing issues arising from a preliminary analysis of simple tabulations.
` "b. Resource procurement and allocation. To prepare and argue the case for adequate funding for the NHS, the DHSS requires ready access to detailed information about the pattern of care in hospitals. Information is also needed for the operation of the resource allocation system, for the monitoring of the system's effects and for the development and improvement of the existing system.
` "c. Accountability. Health authorities are accountable to the Secretary of State and the Secretary of State to Parliament for the setting of policies and priorities for the use of NHS resources, and the use of resources to achieve those objectives.
` "d. Research and development activities. Both DHSS and OPCS [Office for Population Censuses and Surveys] carry out a range of epidemiological, operational research and economic studies, which contribute to policy development and service planning.
`The Hospital Episode Statistics (HES) system, and miscellaneous Körner Aggregate returns resulted from these recommendations, and is largely still based on the principles established in the First Report.
`... The main additional purposes ... are: performance assessment of the NHS by DoH and the NHS themselves; identifying inequalities in health and healthcare and small area studies eg effects of local environmental factors. Even now, there are known deficiencies in the data which impose constraints on the uses but with knowledge and care expert users are seeking to exploit the data more fully than in the past."' [105]
`HES data were originally collected centrally from the NHS through the Regional Information System (RIS) based in each of the Regional Health Authority (RHA) areas. The regions varied as to what data they held on their own databases (eg whether or not names and addresses were held), how they compiled their HES submission, and how data were shared with NHS colleagues. What did not vary was the subset of data items the RHAs supplied to OPCS for HES.' [106]
70 In his first report to the Inquiry, Professor Stephen Evans, one of the Inquiry's experts on statistics, stated:
`Bristol, in common with other hospitals within the National Health Service (NHS), is required to provide summary data on episodes of care for patients. It does this using the local computer-based Patient Administration System (PAS) for providing national returns to the NHS, known as Hospital Episode Statistics (HES). Most other Health Care Trusts in England and Wales have similar systems for producing these returns. More details on the national picture derived from HES are given in a separate report (Aylin et al. 1999). The Bristol PAS is a fairly comprehensive computer system developed from a long tradition of using computers for patient administration in the Bristol area that was begun in the late 1960's. Neither in Bristol nor elsewhere have patient administration systems generally been used for looking at clinical outcomes in a rigorous way. In most instances the recording of death is limited to those deaths which occur prior to discharge from hospital. The Bristol system has included patient outcome, in terms of death, for at least some patients beyond the period when a patient was actually in hospital. This is unusual and allows for a more comprehensive picture of the vital status of patients than is usually the case for administration-based systems.' [107]
`The purpose of the system is not to store information that a doctor requires to care for the patient, nor is it intended for audit of the quality of care. Any such use must take account of the likelihood that important details may not be recorded.' [108]
`The administrative clerks (known as "coders") who classify diagnoses and operations for the PAS are not medically qualified, and they may misunderstand the medical information in the medical records of patients. The medical records themselves may not be clear in describing the diagnoses or operations for every patient. The ICD [International Classification of Diseases] coding system may also have inadequacies when used in a very specialised area. Individual coders vary in their experience, but there are some highly experienced coders who are very good at carrying out the classification of diagnosis and operation.' [109]
<< previous | next >> | back to top
Footnotes
[92] WIT 0211 0002 - 0003 Mr Hooper
[97] WIT 0341 0040 Dr Pryn
[98] UBHT 0343 0002 Mr Joomun
[99] SEM6 0003 0131 - 0132 Mr Tharme
[100] UBHT 0343 0003 Mr Joomun
[101] HOME 0011 0001; `National Comparative Database, United Bristol Healthcare NHS Trust, Reports for Cardiology and Cardiac Surgery', CHKS Limited, 1992
[102] UBHT 0343 0003 Mr Joomun
[103] UBHT 0343 0006 Mr Joomun
[104] WIT 0189 0009 Mr Willmer
[105] WIT 0189 0004 - 0005 Mr Willmer (emphasis in original)
[106] WIT 0189 0006 Mr Willmer
[107] INQ 0012 0007 - 0008 ; `A report on local data relating to children who received cardiac surgery under the terms of reference of the Bristol Royal Infirmary Inquiry', October 1999, Professor SJW Evans
[108] INQ 0012 0012; `A report on local data relating to children who received cardiac surgery under the terms of reference of the Bristol Royal Infirmary Inquiry', October 1999, Professor SJW Evans,
[109] INQ 0012 0013; `A report on local data relating to children who received cardiac surgery under the terms of reference of the Bristol Royal Infirmary Inquiry', October 1999, Professor SJW Evans,