Intended for healthcare professionals

Education And Debate

Injury surveillance programmes, ethics, and the Data Protection ActSharing data to prevent injuriesPotential problems for tenantsThe legal positionEthical viewpoint

BMJ 1999; 319 doi: https://doi.org/10.1136/bmj.319.7206.372 (Published 07 August 1999) Cite this as: BMJ 1999;319:372

Injury surveillance programmes, ethics, and the Data Protection Act

Public health is the arena in which clinical medicine, epidemiology, management, politics, and the law all meet—or perhaps more accurately, collide. Often there seem to be conflicting imperatives. Even within a department colleagues may disagree on policy. Such disagreement can be productive if as a result important policy issues are explored. The case described here raises important issues about the legal and ethical basis of the government's injunction to health authorities to work with other agencies, which is central to the current green papers on public health. A consultant, a director of public health, a health authority lawyer, and an ethicist present their views.

Sharing data to prevent injuries

  1. Ronan A Lyons, consultant in public health medicinea,
  2. Jo Sibert, professor of community child healthb,
  3. Michael McCabe, clinical director of accident and emergency medicinec
  1. a Iechyd Morgannwg Health, Swansea SA1 1LT
  2. b Department of Child Health, University of Wales College of Medicine, Cardiff CF64 2XX
  3. c Morriston Hospital, Swansea NHS Trust, Swansea SA6 6NL
  4. Iechyd Morgannwg Health, Swansea SA1 1LT
  5. Bevan Ashford Solicitors, Waterloo House, Fitzalan Court, Cardiff CF24 0BA
  6. Bioethics Centre, Dunedin School of Medicine, University of Otago, PO Box 913, Dunedin, New Zealand

    The purpose of the NHS is to improve the health of the population. This guiding principle was first laid down in the Strategic Intent and Direction for the NHS in Wales: “Working with others, the NHS should aim to take the people of Wales into the 21st century with a level of health on course to compare with the best in Europe.”1 This was subsequently taken up by the Health of the Nation and Our Healthier Nation initiatives in England and Better Health, Better Wales, the most recent green paper in Wales.14 In England, the prevention of injury is one of four areas for health gain in the population and in Wales it is one of 10 areas for health gain. Both old and new English and Welsh documents identify the need for injury surveillance systems and that the NHS needs to work with others to prevent injuries. Thus, the Welsh Protocol for Investment in Health Gain for Injuries sets targets for health authorities, among which are “working with others, each health authority purchasing team and local authority should develop a multidisciplinary, multiagency strategy for addressing injury prevention, based upon targeting those groups most at risk,” and “health authority purchasing teams should identify black spots for child injuries and, working with local authority partners, develop strategies for dealing with them by 1994.”5

    Injury surveillance system

    Fine sentiments, but what happens when it comes to putting the strategy into practice? In 1993, responding to this agenda the Department of Public Health at Morgannwg Health Authority (then West Glamorgan Health Authority), set up an injury surveillance system on the basis of collaboration with consultants in accident and emergency medicine in the three local hospitals that provide accident and emergency services for the local population. The database and its uses are described elsewhere, but it does collect full postcoded data for geographical mapping.6

    Community based programme

    As a result of the identification of injury black spots in certain electoral wards, a community based programme for preventing injuries in children was started in three high injury areas, with backing from the Welsh Office and local health authority. In 1997, we were invited to a meeting with a partner local authority to discuss injuries in houses in multiple occupation—that is, dwellings occupied by more than one household excluding purpose built blocks of flats. Injury rates are high in such buildings, and inspection of these buildings frequently shows remediable hazards. On discovery of a hazard, local authority officers can issue improvement notices that may make a landlord eligible to apply for a remediating grant. In more serious cases, the officers can apply for a closing order or control order using powers from the 1985 Housing Act. A problem for the local authority is that not all houses in multiple occupation are known to them, and they do not have the staff to visit all those they do know.

    The possibility of using the injury surveillance database to assist our colleagues from the local authority arose in the discussion. Briefly, the idea was to identify the most frequently occurring postcodes in the database and, by providing these to the local authority, narrow the focus to those areas with highest injury rates. Postcodes contain on average 14 contiguous addresses so that individual houses were not identified. By concentrating their efforts on the most likely areas, environmental health officers might become more proficient at preventing children from being injured in houses in multiple occupation.

    This idea was discussed by the local health authority, but the director of public health believed that passing such information to the local authority might contravene the Data Protection Act. The matter was referred to the health authority's solicitors and this viewpoint was upheld.

    Figure1

    Child injury rates are high in houses in multiple occupation

    (Credit: PHILIP WOLMUTH)

    We are now left in a position of knowing where childhood injuries occur but of not being allowed to pass the information on to public bodies It is permissible to identify comparatively large areas (for example, electoral wards) with high childhood injury rates but not smaller areas where intervention could be more focused and cost effective. It is almost as if we are asking our environmental health colleagues to play the blindfold game of pinning a tail on a donkey. Unfortunately, as injury is the leading cause of childhood deaths (600-700 per annum in England and Wales) and a frequent cause of attendance at accident and emergency departments (about 30% of children attend in any one year), it is a sad game to have to play.6

    If, as suggested by the other contributors to this debate, our suggested course of action does not have ethical or legal backing, then we are left in a quandary. Our Healthier Nation specifically tasks the NHS with improving health through partnership arrangements. In the summary it says that “health authorities will have a key role in leading local alliances to develop health improvement programmes, which will identify local needs and translate the national contract into local action.”3 How can these disparate views be reconciled? Is there a need to amend the public health legislation to allow us to protect children from their greatest cause of mortality and morbidity?

    Footnotes

    • Competing interests None declared.

    References

    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.

    Potential problems for tenants

    1. Peter D Donnelly, director of public health (peter.donnelly{at}morgannwg-ha.wales.nhs.uk)
    1. a Iechyd Morgannwg Health, Swansea SA1 1LT
    2. b Department of Child Health, University of Wales College of Medicine, Cardiff CF64 2XX
    3. c Morriston Hospital, Swansea NHS Trust, Swansea SA6 6NL
    4. Iechyd Morgannwg Health, Swansea SA1 1LT
    5. Bevan Ashford Solicitors, Waterloo House, Fitzalan Court, Cardiff CF24 0BA
    6. Bioethics Centre, Dunedin School of Medicine, University of Otago, PO Box 913, Dunedin, New Zealand

      Directors of public health have the difficult role of combining the functions of a public health advocate with those of an executive member of a statutory authority. The latter role requires that they do not expose fellow authority members to legal sanction as a result of breaking the law. My understanding of the Data Protection Act was that the action proposed by Lyons and colleagues could well contravene it. It seemed questionable that we would be utilising data collected by our agency for one purpose to supply without permission to a separate agency for a distinct purpose. I therefore sought legal advice from the health authority's retained solicitors.

      I derived no great pleasure from seeming to act as “corporate man” but I saw no legitimate way of doing otherwise. It also seemed that there were arguments against the proposed action beyond the needs of simply complying with the Data Protection Act. This concerned the need to protect people from persecution by landlords.

      Having spent 5 years as a city councillor serving on committees for environmental services and housing, I am aware of the pressures landlords can exert on tenants—for example, if they complain to environmental services about the condition of their rented property This pressure can deter tenants from complaining even when property is in a dangerous condition. We may not approve of this, but it would be paternalistic and naive to insist that people must complain and thereby expose themselves to persecution and possible homelessness If such unlawful pressures result from complaints about housing conditions, they could certainly occur if landlords were to find out that they had been targeted because of their tenants' attendance at an accident and emergency department.

      I am not defending such illegal practice by landlords, but part of public health is an appreciation of what occurs in the real world, and I fear that within a few months of the proposed policy being introduced, some tenants would refrain from attending their local accident and emergency department with minor injury because of fear of retribution.

      The figures quoted by Lyons and colleagues are clearly a cause for concern The vast majority of the deaths are, however, from road traffic accidents and accidental poisoning rather than factors remediable through housing enforcement action. I therefore conclude that even if the proposed practice was legal, it may still not be justifiable for the overall good of the public.

      Footnotes

      • Competing interests None declared.

      The legal position

      1. Tessa Shellens, partner (t.shellens{at}bevanashford.co.uk)
      1. a Iechyd Morgannwg Health, Swansea SA1 1LT
      2. b Department of Child Health, University of Wales College of Medicine, Cardiff CF64 2XX
      3. c Morriston Hospital, Swansea NHS Trust, Swansea SA6 6NL
      4. Iechyd Morgannwg Health, Swansea SA1 1LT
      5. Bevan Ashford Solicitors, Waterloo House, Fitzalan Court, Cardiff CF24 0BA
      6. Bioethics Centre, Dunedin School of Medicine, University of Otago, PO Box 913, Dunedin, New Zealand

        So that an injury can be dealt with patients attending accident and emergency departments provide confidential information about themselves to the medical and nursing staff. Assuming they have been given information about the use to which their records may be put, patients may have a vague idea that the information is also being kept for “NHS purposes” which are well defined in the Department of Health and Welsh Office Guidance on the Protection and Use of Patient Information.1 On page 2 of that document, fairly general guidance is given as to the type of NHS functions where information about patients may be utilised: paragraph 1.3 on page 3 states:

        “Patient information will be seen and used by a number of NHS professionals and administrative staff, as well as staff of other agencies contributing to a patient's care. Most patients would be unlikely to trust staff with detailed information about themselves and their clinical condition if they thought this might be passed onto others without proper control. It is therefore a central tenet of the NHS that, in the words of the Patient's Charter, you have the right to see your health records, which should be kept confidential within the NHS. In addition, the present guidance makes it clear that personal information should be anonymised where possible.”

        The report of the Caldicott committee emphasised the need to promote awareness of the guidance.2 When the patient gives information to doctors or nursing staff, the patient cannot possibly remotely expect that this guidance would be interpreted to justify the passing of information to another organisation not specifically concerned with that patient's health, but with the specific intention of enforcement of public health regulations against another body—namely, a landlord. This direct link is too remote.

        The proposed disclosure is, therefore, likely to constitute a breach of the common law duty of confidence and to be regarded as contrary to the eight principles of the Data Protection Act 1984. The 1998 Data Protection Act strengthens and extends those principles. To protect public health, there does exist a series of statutory requirements setting out exemptions to allow for the provision of public health information to other agencies. This type of disclosure is not covered by statutory exemption. There is also the general exemption of considering the wider “public interest.” It is, however, difficult to see such circumstances being sufficient to constitute a public interest argument.

        The purpose for disclosure in the present case would seem to directly conflict with the concept of anonymity and aggregation—that is, an anonymous compilation of information taken from raw case records. To avoid acting contrary to the Data Protection principles or the common law duty of confidentiality, then the information must be sufficiently and adequately anonymised so that there is not any possibility of individual patients being identified by reference to their specific housing. Choosing a small geographical area means this may be more than a possibility, and to give the information specific to postcodes would seem to be insufficiently anonymised because of the potential to identify individuals.

        The essential rule of thumb must be whether a patient could possibly envisage that information provided by him or her to an accident and emergency department could be used in this way. In my view, the answer to that must be “no.” The proposal could lead to potential criticism by a patient not only to the Data Protection Registrar but also to the Health Service Commissioner, and could expose the clinical staff to an allegation of breach of confidentiality.

        Footnotes

        • Competing interests None declared.

        References

        1. 1.
        2. 2.

        Ethical viewpoint

        1. Donald Evans, professor of medical ethics (donald.evans{at}stonebow.otago.ac.nz)
        1. a Iechyd Morgannwg Health, Swansea SA1 1LT
        2. b Department of Child Health, University of Wales College of Medicine, Cardiff CF64 2XX
        3. c Morriston Hospital, Swansea NHS Trust, Swansea SA6 6NL
        4. Iechyd Morgannwg Health, Swansea SA1 1LT
        5. Bevan Ashford Solicitors, Waterloo House, Fitzalan Court, Cardiff CF24 0BA
        6. Bioethics Centre, Dunedin School of Medicine, University of Otago, PO Box 913, Dunedin, New Zealand

          Threats to an individual's health include hazardous pursuits, environmental hazards, and other people's dangerous behaviour. It is not the responsibility of doctors to address all these threats. Unwarranted intervention would encroach on the privacy and freedoms of the public, making doctors agents of social control. Harm would thus be done both to civil liberties and to the trust that is essential in the doctor-patient relationship. Where in this context should we place the proposal to use, without consent, confidential information collected in the medical setting?

          Hazardous pursuits

          Health threats might be causal links between the work setting and disease incidence. Doctors are specially qualified to investigate and monitor these. It is less clear that threats to the health of workers prone to accidents such as falls are best handled by doctors. Although doctors are best qualified to diagnose the extent and nature of damage caused by such accidents, they are not trained to devise protections from such hazards. This responsibility is best borne by others.

          Domestic contexts often represent health hazards and make the task of living itself a hazardous pursuit. Where there is a link between the hazard and some disease state the law provides for cooperation between doctors and others to remedy the problem. But the proposal is more akin to the accident category of threat, and thus does not provide a case for doctor intervention.

          Hazardous voluntary pursuits such as boxing are even less problematic Enthusiasts choose such pastimes and do not welcome interference from doctors. By virtue of their specialist knowledge, doctors might form a powerful pressure group to curb such activities. But few believe it their business to control the sport any more than it is the business of doctors to report or apprehend motor cyclists who fail to wear crash helmets. Insofar as place of abode is largely a matter of choice, similar considerations attach to doctors' interventions in accidents due to remediable faults.

          Environmental hazards

          Environmental pollutants are possible causes of disease in the population Those responsible for such pollution are subject to legal controls Doctors are specially qualified to identify and measure such threats and are obliged to seek remedial action. Similar considerations apply to health hazards connected with sanitation and drainage that might be associated with particular dwelling places. There is a statutory obligation on the public health physician to identify these irrespective of the relations between the landlord and tenant.

          Other kinds of environmental hazards are not the doctor's responsibility. All cities have accident black spots where people are more likely to suffer injuries than in normal situations. Road safety measures are an important form of health protection, but it does not fall to doctors to identify the need for, nor implement, the provision of such measures. The threats involved in the proposal are more akin to the latter suggesting that doctors are not the proper people to address them.

          Dangerous people

          Aggressive, reckless, or negligent behaviours of others threaten citizens' health. In cases where a clinical condition explains the threat, doctors have an indisputable role to play in health protection. In other cases different bodies have this responsibility. The negligent builder, the reckless driver, and the violent criminal constitute identifiable dangers to everyone. But is not the business of doctors to police their activities to protect the public, for clinical expertise is not required to establish either the explanation or the remedy of the dangers.

          Although in the case of child injuries caused by criminal assault it is not the business of the doctor to engage in police work, nevertheless the concealment of grave suspicions is unethical. But the cases with which the proposal is concerned do not concern injuries of this sort Careless landlords are more akin to those categories of dangerous people identified as not being the responsibility of the doctor Comparison with other public health cases offers little support to the proposal, and suggests that it would be unethical for the information in question to be divulged without consent.

          Footnotes

          • Competing interests None declared.

          View Abstract