12.4 a.m.
I am grateful for the opportunity to raise this subject in this Adjournment debate, and I am glad that the Minister of State is here to reply to it, in view of the number of Parliamentary Questions that I have put to him in recent months.
I believe that there is a growing fear among many people about what I call "computer power". Computer power is the ability to record, keep and collate a great deal of information about every one of us, and it is a power which could be misused or exploited against the individual. I want to examine this growing computer power within the National Health Service. There are about 2,000 people involved in all types of work dealing with computers in the NHS. More than 100 of them are systems designers—men and women who are specialised in preparing new systems and new types of record in order to make the present system, as they see it, more efficient—combined with some 150 computer operators. I wish to make it clear that I do not accuse any of those involved in this work of anything other than the highest motives. But, in the long term, the position which could be created, as I hope to show in my later comments, is more frightening than that in George Orwell's "1984". I want to examine the Annual Review of National Health Service Computing 1974, Parts 1 and 2. I find these documents both fascinating and frightening. They are fascinating because they are clearly well thought out, with the idea of building up a massive number of computer record systems within the NHS based, initially anyhow, upon the regional health authorities. They are frightening because they will mean that the medical records of children to come will go on to a computer from the moment that they are born to the moment that they die. I quote from Part 1, where the intentions are clearly stated. Section 13 on page 8 says:There we are told openly what the intentions are. A look at some of the computer records that the Department wishes to set up builds up an alarming picture. I take, for example, the Multipurpose Child Surveillance computer record on page 79 of Part 2. There we read:"Since the last Review in 1973 the work of Regional computer installations has continued to grow and the reorganisation of the NHS has brought with it not only a need for improved management information systems but also responsibility for work previously undertaken by local health authorities, which itself was growing. There is no doubt that the range of computing activities in the NHS is expanding and will continue to expand for many years ahead."
That is priority No. 1. Then I move to the Surveillance Registers computer record:"In general, a computer-held record is established with the birth of the child."
I submit that that could have a very wide interpretation. Then we come to a Pre-Registration Hospital Record. This would mean that in advance of patients attending at hospital, to this first record would be added all the patient's subsequent hospital records. We then run into the In-Patients Admissions Computer Records System. I quote from page 108:"These records give information on those members of the population at risk of developing various diseases or disabilities."
Then we come to the point when we are discharged from hospital. We have the Discharge Computer Record System, which involves the transfer of information to other health agencies. However, it is in relation to the next batch of systems that I find the greatest cause for worry and concern about the future. These are called the Personal Medical Record Systems, which include systems for a basic patient record, an abbreviated patient record, an extended clinical record—the final aim of this is to eliminate written hospital or GP case notes entirely—a personal health record, and then an integrated patient record, with information about both hospital and GP treatment. Any of those record systems taken in isolation may not appear very important, but when one looks at them together to see what the aim and the intention is, one appreciates that it will mean a very comprehensive record of the entire medical history, experience and life of every man, woman and child in the country. Many of these developments are in the future. As I said, they will be based on the regional health authorities. But if these systems are set up they will inevitably lead—perhaps the Minister will confirm this—to a national computer record. A number of questions arise from these proposals and intentions and the general thinking of the Department at present. The first relates to confidentiality. As the Minister will know, I am not a tremendous fan of the Lane Report in many of its aspects, but on page 99 at paragraph 316 these words are used:"Retrieval of previous information as appropriate is carried out … providing relevant data for a variety of purposes."
If systems of the type I have outlined are set up and are working as I am sure they are intended to work, confidentiality will go out of the window. It will mean that many people will not feel totally and completely free to talk to their doctors about any and every problem. Men or women who talk to their GP about their sexual problems or other difficulties do so at present in the firm and complete knowledge that anything they say to their doctor will never go outside the room. Once records are built up in the way proposed and thought of in the National Health Service, that will not be the case. As more information is stored, the chances of theft and misuse become more tempting and more likely. What about the type of information which will inevitably go on to the discs and tapes, which would record that a woman had had a child before marriage, or an abortion, or that a man had been an alcoholic in his teens or twenties? All this sort of information would be bound to end up on such a computerised system. This could gravely damage confidence between doctors and patients. Mistakes are inevitable. How can they ever be put right once they are recorded on this massive system? In a Written Answer on 28th January, the Minister told me:"There is another aspect of confidentiality to which we desire to draw attention. Confidentiality should be preserved not only in respect of communications between doctor and patient but also in relation to communications between doctors and to reports of all kinds, including social reports."
When this computerised system is more advanced, the Department will have to find ways and means by which records can be checked or disputed to ensure protection for patients. The problem needs a new approach. There are four great dangers in computer power. First, when conflicts arise between individuals and the authorities, like the National Health Service, the individual is at a disadvantage now. When faced with a computer which stores information, perhaps linked to other computers, he will be at an monumental disadvantage. Second, information obtained now for perfectly good reasons could be misused in future, intentionally or accidentally, and the risk will increase as the system becomes more sophisticated. Third, once established, computer systems are difficult to destroy. A computer can be dismantled, but that will happen rarely. Even then, a more efficient machine will be installed. To be effective, one would have to get hold of the records—tapes, discs or microfilm. So their power increases. Fourth, there could be errors of information which are difficult to correct and which could easily and gravely damage an individual. New laws are needed to establish a strong inspection body, to enforce stringent checks and safeguards, to control tapes, discs and microfilm. One day, someone will think of linking up all the computers in the great Departments of State. If that happens, the individual will lose his privacy and freedom and will be at the mercy of an all-powerful and inhuman system."A patient, whether an in-patient or an outpatient, has no right to examine the medical records relating to him"—[Official Report. 28th January 1975; Vol. 885, c. 90.]
12.19 a.m.
The hon. Member for Brighton, Kemptown (Mr. Bowden) has raised an important subject, which he discussed as "computer power". Discussion of this subject should be open. The documents he quoted were documents that I sent to him in the belief that the wider the discussion and the greater the information about these developments the better. So I do not take the view that this area should be swept aside. The more we consider the problems involved the better.
I do not therefore propose to describe in detail the wide range of activities applied within the National Health Service, described in the two volumes which I have sent to the hon. Member. In many areas, for example financial and administrative work, their use reflects accepted commercial and industrial practice. In those areas, as a major employer—the largest single employer in the country—we would be expected to be using the newest techniques. I propose to limit my remarks to those activities on which the hon. Gentleman concentrated, mainly where computers handle information about patients. In looking at these activities it is necessary to keep a clear distinction between those activities which are common within the NHS and those which are limited to a small number of experimental sites. There are only a few applications involving patients' data which extend throughout the NHS. The only computing facilities operating throughout the NHS are those provided by the installations managed by regional health authorities which provide bureau facilities for their areas and districts. All regions either have a computer which can provide facilities on the scale required or share one as a temporary measure. It is these computers which process the administrative and financial applications of health authorities. At the present time the main application which they process involving data on patients is the statistical work arising from "Hospital Activity Analysis". This system has been progressively introduced into the NHS over the past six years. Its main purposes are to give individual doctors a summary of data on his patients and to provide an aggregation of such data for general clinical and management purposes. Coded data on diagnosis and operations are held by the computer. Patients' names are not generally required for hospital activity analysis and are usually obliterated before data capture forms leave the hospital. Some authorities do however include this item in response to a request from clinicians for named clinical summaries. Addresses are not included, residence details being reduced to a code so that catchment areas can be analysed. Over 90 per cent of non-psychiatric hospitals participate in this scheme. In future these regional machines are likely to process the registration and other work of family practitioner committees. A pilot study in this area has been encouraging and a further trial is to be mounted. Under this system the identification details of patients on doctors' lists would be held on computers with the primary purpose of calculating the remuneration of general practitioners but not clinical information. Another widespread application is to be found in the field of child health. A large number of local health authorities had systems prior to the reorganisation of the NHS and many area health authorities have continued to operate these systems on local government computers. The most common use and one which has been most beneficial, is the administration of immunisation procedures which ensure that GPs and parents are informed when further immunisation is due. The main items of data held relate to identification details about attendances and immunisations, which are crucial to the system. I would add that it is quite widely believed that the use of computers for this purpose has contributed to the high immunisation rates achieved in many parts of this country. I believe it has been a successful use. These are the only widespread applications in use or envisaged at present. In addition there is a good deal of experimental activity but the number of sites involved is relatively small. Fourteen projects are being financed by my Department to explore the extent to which computers can be used to improve the management of hospital resources. These experiments inevitably involve the computer holding details about patients since hospitals revolve around the needs of patients. We cannot exclude that. The tasks for which these computers are used include making out-patient appointments, maintaining waiting lists for admission, and handling the administrative procedures involved in admitting and discharging patients. Identification details and some limited clinical information, for example, diagnosis and operation planned, are needed to carry out these tasks. Some systems go a little further and hold what amounts to an abbreviated version of the medical record. This is what has mainly concerned the hon. Gentleman. The intention is that data on patients should be readily available at all times, a situation not always possible with a manual system. To have such details as current diagnosis, blood group, drug therapy and known drug sensitivities available can be of particular value in an emergency. I should add that in no case does the computer contain the entire clinical record and in no case does it replace the existing case notes, which are still maintained manually. I do not want to exclude that situation from developing. In addition to these systems there are a larger number, financed both by my Department and by health authorities out of their normal budgets, which involve a small computer dedicated to a particular scientific or clinical use. Many are in pathology or radiology departments. These systems hold identification details of patients to enable requests for investigations and tests and their results to be handled speedily. There is a growing demand for tests and investigations of all sorts and it is hoped that computers will be able to solve the problems posed by the clerical processing of requests and reports. Only one of the experimental projects handles general practitioner records and there are currently no immediate plans to extend this work or to mount further experiments of this nature. The family practitioner committee system I have already mentioned will not hold medical data, merely identification details. I should like to mention the particular use of a computer for organ matching purposes. The South Western Regional Health Authority machine holds a file of over 900 potential recipients of kidney transplants in the United Kingdom. When a kidney becomes available, details of the matching characteristics are telephoned to the National Tissue Typing Reference Laboratory at Bristol and fed into the computer, which makes complex comparisons with the characteristics of potential recipients and produces results in the form of the best matches. The organ matching service and the doctors concerned agree the distribution of the kidneys and arrangements are made for their transportation. The possibility of introducing an international organ matching service is being explored by the EEC. I should at this point stress again that in relation to the number of hospitals in the NHS, the number who have any form of computer is very small. We are still experimenting with the use of computers in the administration of health care, and it is still too early to say whether the benefits of using a computer will compare sufficiently favourably with the costs to justify their wider use. It is, however, my belief that, given the demands for health care and the limited resources available, we should be looking at new methods of using computers to see whether we can make better use of our limited resources. The conduct of these experiments, however, must involve the use of data relating to individual patients and to that extent I share the hon. Gentleman's concern that adequate steps should be taken to safeguard the confidentiality of such data. The planning of an experimental computer project requires a real partnership between professionals in health care and in computing. Medical and nursing staff participate in all stages of the development of a project and bring to this work the traditions and ethics of their professions. They are dealing all the time with manual records of patients, and the standards of confidentiality used on them have to be applied to computer records. They are determined to maintain or even improve upon the traditional confidentiality of the doctor-patient and nurse-patient relationship. In the course of their involvement some have come to the view that, rather than pose a threat, computers can provide a higher standard of security and confidentiality of patient information. There are some doctors and nurses who are worried about the proposals. My Department has both encouraged the active participation of medical and nursing staff in the design of these systems, in itself an important safeguard, and issued advice to authorities managing experiments on methods for keeping data secure, and on a whole range of aspects. This advice covers such matters as the physical security of computer installations and equipment; the storage of computer records; built-in safeguards such as limiting access to certain data to particular classes of person and times of the day—for instance, while the out-patients' department is open; methods of identifying people seeking access to information, such as passwords; methods for logging access to information; and advice on the clauses to be written into staff contracts to bring home to the staff the serious nature of any breach of confidentiality. The hon. Gentleman suggested that patients ought to be able to see their medical record, particularly when it is held on a computer. In my view the fact that some part of the medical record is held on a computer is secondary to the general principles governing the disclosure to patients of the content of their medical record. The same issues arise whether it is a traditional manual record or a computer record. The hon. Gentleman has a point in what he says about the transfer of such information but, as I have explained, there are considerable safeguards in that respect. A patient—whether in a hospital or of a general practitioner—has no right as such to know the contents of the medical records relating to him. It is for his doctor to decide, in his professional judgment, what information should be given to the patient. I know that many people take the view that they are entitled to know as much about their own state of health as their doctor does. Others—and I believe that most doctors in this country agree—think that how much a patient should know depends on a large number of different factors, and especially on the patient's own circumstances. There will certainly always be some cases where a doctor is convinced that it would not be in the patient's well-being for him to have access to his full medical record, and, indeed, that this could be severely distressing. There cannot be any hard and fast rule about such matters. The long-standing policy of my Department has been, and remains, that it would not be right to attempt to intervene in professional medical opinion on matters of this kind. If we were to make a change in relation to computers, we would have to make a change for overall records. It is for the doctor to decide, in his professional opinion, which information should be given. I think that there is already a great deal more openness and frankness between doctors and patients in discussing medical conditions. So far as the accuracy of a patient's record is concerned, the prime difficulty is in deciding who can form a judgment. Apart from basic factual information, most of a case note involves professional judgments by doctors. Whether it is correct or not is therefore largely a question of medical opinion. I cannot see how we could make available what may be only an opinion or even a query put forward by a doctor to alert another doctor who may be giving treatment in future. In the event of a patient gaining access to his record and disagreeing with what is found there, the matter can be resolved only by a discussion between the patient and the doctor concerned. This is getting into the whole intimacy of the doctor-patient relationship, an area in which it would be difficult and unsuitable for the Department to lay down regulations. There has been some concern over the length of time records may be kept, and the possibility of their becoming a source of embarrassment in later years. Case notes are weeded at the end of each episode and some are retained for six to seven years in case they are required as evidence in legal proceedings. Certain records are retained in hospital beyond this time, for example because the patient's malady is of a continuing or recurrent nature or because there are features of the case which may assist with research. One would have to consider that aspect, too. I do not think that all these questions can be divorced from the whole consideration of information being given to patients. I am satisfied that the confidentiality of the information held on the computer systems which I have described is adequately safeguarded and I can assure the hon. Member that no proposal for linking health information on a national basis would proceed without full public discussion. We would not want to wake up and find that it had suddenly occurred. We should be prepared to discuss these issues. I think that the hon. Gentleman's main anxiety was if this information, taken in the context of medical treatment and for the Department of Health, were to be made available to other Departments and other areas of Government. I agree that the transferability of information is a much more serious issue and raises fundamental questions of privacy, which the Government are looking at. But if one confines the question within the medical system, there is wide access already to medical records in transfer from doctor to doctor and for medical research which goes beyond particular districts—The Question having been proposed after Ten o'clock on Wednesday evening and the debate having continued for half an hour, Mr. DEPUTY SPEAKER adjourned the House without Question put, pursuant to the Standing Order.
Adjourned at twenty-six minutes to One o'clock.