Intended for healthcare professionals

Education And Debate

Is there a moral obligation not to infect others?

BMJ 1995; 311 doi: (Published 04 November 1995) Cite this as: BMJ 1995;311:1215
  1. John Harris, professor of bioethics and applied philosophya,
  2. Soren Holm, senior research fellowb
  1. aCentre for Social Ethics and Policy, University of Manchester, Manchester M13 9PL
  2. bDepartment of Medical Philosophy and Clinical Theory, University of Copenhagen, DK-2200 Copenhagen, Denmark
  1. Correspondence to: Professor Harris.
  • Accepted 3 August 1995

The emergence of HIV infection and AIDS has refocused concern on the obligations surrounding the carrying and transmission of communicable diseases. This article asks three related questions: Is there a general duty not to spread contagion? Are there special obligations not to communicate disease in the workplace? And does the mode of transmission of the disease affect the ethics of transmission and, if so, how and to what extent? There seems to be a strong prima facie obligation not to harm others by making them ill where this is avoidable, and this obligation not to communicate disease applies as much to relatively trivial diseases like the common cold as it does to HIV disease. The reasonableness of expecting people to live up to this obligation, however, depends on society reciprocating the obligation in the form of providing protection and compensation.

The AIDS epidemic has revitalised many questions about communicable diseases. These questions arose at a time when most communicable diseases were untreatable and controlling the spread of contagion was a mainstay in the battle against disease, illness, and death. The responsibilities of disease carriers and the response of society towards communicable diseases have again come into focus, but many of the new answers given to these old questions have been influenced by the special features of HIV disease.1

We believe that these questions are pertinent to a far wider range of diseases than just HIV infection and that they can best be analysed within a framework of reciprocity. Such a framework means that the burdens put on people carrying a disease must be reciprocated by society's commitment to treatment, care, non-discrimination, and, in certain circumstances, compensation.2

In this paper we analyse three related questions about the responsibilities of carriers of communicable diseases: Is there a general duty not to spread contagion? Are there special obligations not to communicate disease in the workplace? And does the mode of transmission affect the ethics of transmission and, if so, how? We start in the workplace.

Communicating disease in the workplace

We know from public opinion surveys that people are hesitant about working with people with HIV infection or AIDS.3 4 This attitude is paradoxical since the risk of contracting HIV through normal social contacts is extremely low and probably zero.5 Other communicable diseases are much more contagious and threatening. Many such diseases could be mentioned, but we will use the most common--the common cold and influenza, both airborne infections with a high rate of infectivity.

The common cold and flu are not usally lethal diseases in healthy human beings, but they do debilitate people for a short while, and because they are common they cause a great loss of working days. While not life threatening to healthy individuals, the annual death rate from upper respiratory tract infections is not negligible. On average 2000 people die in the Netherlands each year as a direct result of influenza infections, and mortality increases rapidly with age.6 In 1985 an American study estimated that the costs of ambulatory treatment for respiratory tract infections in the United States was about $10 billion and the cost of lost working days more than $9 billion a year.7 In both conditions immunisation gives only partial protection and there is no effective treatment. The question we therefore want to put is: What moral responsibilities does a person have towards her colleagues if she believes that she has contracted a cold or flu?


Given the high infectivity of these conditions she can be almost certain that if she goes to work one or more of her colleagues will be infected. Her going to work will therefore result in some other people becoming ill possibly needing time off work. She may also, of course, infect others on the way to work, on public transport.

If she stayed at home she would not infect her colleagues but might lose wages or opportunities or let down colleagues and impose an increased workload on them. How important any loss of wages would be would obviously depend on the degree to which she needed the income and on the degree of compensation that might be available. This calculation is one she can make in her own case but not for others whose circumstances may be different and whose autonomous choices should not be pre-empted.

A strict duty to stay home from work when one is infected with the common cold or flu would create financial hardship for many people. It therefore follows from the reciprocity thesis that such a duty, if it were to be imposed by legislation or by the informal rules of a particular workplace, would have to be balanced by compensation for any loss. The alternative is, of course, not to require any self imposed isolation but to compensate those who become ill where this involves any financial loss. This might be less costly overall since it would not require policing.

An alternate model would be one dictated by the threat of litigation. If those who transmitted disease to their fellow workers were liable to compensate them then the threat of litigation or of compensatory settlements might have the same effect. The problem, of course, would be establishing from whom a particular disease was acquired. Although an acute problem now, this might be changed by scientific advances in virus identification. There are, however, many diseases that are infectious before people feel ill, and a comprehensive liability scheme would have to penalise transmission from these unknowing carriers. This obviously raises serious moral problems, too.

Communicating disease in social settings If communicating a disease is to inflict a harm proportionate to the severity of the disease and its consequences then the same moral obligation not to inflict such harm on others applies as much to disease as it does to other knowingly inflicted harms and applies as much in the social context as it does at work. Here we need to distinguish casual contact in the street or other public places from private environment. This is not because there is any difference in the moral obligation owed between casual and other contacts but stems from two other considerations. One is the consent, real or presumed, to risky social contacts that applies where people choose to share social space with others. The second might be termed “the last refuge” consideration.


It is not just people infected with HIV virus …


The responsible individual will stay at home when she is ill, both to help speed recovery and avoid infecting others. So those whom she shares her home with will be at increased risk. Two factors must be considered here. Firstly, by the time the individual is aware that she is ill she may already have infected her intimate contacts. Secondly, her home is her last refuge: she cannot reasonably be expected to have a home which is conditional on the episodic consent of those with whom she shares it--unless the danger is so great that isolation is required for everyone's safety.1


Other, more restricted, social contacts--with neighbours, at social venues, clubs, etc--seem to require the same rules as should obtain at the workplace. That is, an infected person should temporarily isolate herself or seek the specific consent of those with whom she comes into contact. This restriction may in some cases conflict with the basic needs of the person with disease. If somebody is living alone it may be impossible to avoid, for example, going shopping for food. In the case of the common cold it seems excessive to require self imposed starvationin the discharge of the duty not to put others at risk, just as it seems excessive to require great economic sacrifice.

Does mode of transmission matter?

The second question is whether the mode of transmission should affect the way we analyse the situation from the ethical point of view. In the debate about HIV infection the mass media has shown a tendency to distinguish between so called “innocent” victims, such as those with haemophilia and children, and allegedly less innocent homosexuals and drug addicts.

To make such a distinction morally relevant one would have to show either that the mode of transmission itself or that personal involvement in the causation of transmission was morally relevant. The residual question is, of course, the degree of voluntariness of the transmission and hence that of personal responsibility for consequences.

An interesting pair of conditions to consider is those caused by herpes simplex virus I and II--respectively, the common cold sore (herpes labialis) and genital herpes (herpes genitalis), though both viruses can cause either condition if introduced in the corresponding part of the body. The herpes family also contains many other species, including herpes zoster, giving rise to chicken pox and shingles. The sentence, “I have got herpes” therefore has different connotations according to the localisation of the herpes lesion it refers to.

We may well pause before kissing a person with a herpes lesion on his or her lips, but we would not normally hold parents morally culpable if they kissed their child with such a lesion or a lover who kissed his or her partner. Our attitude to genital herpes is quite different. Genital herpes is a venereal disease, and people who transmit it to “unsuspecting and innocent victims” are often considered morally culpable. Admittedly, genital herpes may be more painful than a cold sore, but the difference is not sufficiently great to explain the profound difference in social attitude. Also infection with genitalherpes is a risk factor for the development of carcinoma of the cervix, but this discovery is fairly recent and not widely known outside medical circles. It cannot therefore be the basis of any difference in general social attitudes. Is there a morally relevant difference at play, or is it just prejudice or the vestiges of the old, and dubious, connection between sexuality and “morality”?8


To argue that it matters morally that some disease is transmitted through sexual intercourse we would have to show that this mode of transmission is not only materially but also morally different. It is difficult to see how this could be so, and even a superficial analysis of common moral attitudes shows them to be inconsistent. To compare the two herpes conditions, both are transmitted through intimate contact of the mucous membranes of a sort which is usually prompted by positive feelings between the two people involved (although in many societies kissing is a formal greeting); even the transmission of herpes labialis may be part of a sexual act. What difference can we point to between the two cases? That they affect different parts of the body is hardly enough to create a significant moral difference.

Initially there seems to be more mileage in the idea that what matters morally is not whether some disease is transmitted through sexual intercourse but whether the person infected was in some way personally involved in the act causing transmission. One could claim that if people participate actively in the act which causes transmission, in a situation where they know that there is a risk of transmission, they become morally responsible. We would need provisos exempting paramedics, nurses, and doctors from this judgment, since they often perform acts with a potential to cause transmission of disease through their need to treat sick people, but we do not usually claim that the doctor with hepatitis B contracted at work is morally to blame. And we generally find it lamentable that the AIDS epidemic has made people hesitant toperform mouth to mouth resuscitation.9


… but also those infected with herpes virus …


… and the common cold who should take action not to transmit their infections


But how could such a proviso work, without introducing a distinction between morally correct and morally blameworthy acts not based on participation in the act, but on an independent evaluation of the kind of act it is? What makes mouth to mouth resuscitation different from kissing? Could we formulate a general rule that would exempt medical staff from blame when they engage in potentially infecting behaviours in the line of duty, while still blaming people transmitting infection during sexual intercourse without implying that sexual intercourse is itself morally suspect? Are we really dealing with the last remains of the puritan belief that pleasure is sin?

We could try to distinguish between acts that are necessary--that is, medical acts--and acts that are unnecessary--that is, sexual acts. But this distinction holds only if the concept of necessity excludes any trace of psychological or biological necessity. Even then there would still be problems with a whole range of medical acts that are not necessary in any strict sense--but are done, for example, to make the patient feel more at ease.

A distinction between socially useful and socially non-useful acts would not work either, since many sexual acts are extremely socially useful. Without sexual acts society would soon cease to exist. It is sometimes claimed that health care workers have a professional duty to expose themselves to potentially infective contacts and so fall into a different moral category from others who expose themselves on the basis of personal choice. This is an extraordinary claim. Certainly doctors, for example, have no legal obligation to expose themselves to significant personal risk. It is far from clear that doctors' duty ofcare extends to running significant risks not only of infecting themselves but, via such infection, of communicating disease to others of their patients. As the recent case of R v Gaud has shown, a surgeon infected with hepatitis B who continued to practise was convicted of causing a public nuisance and gaoled for a year.


The main claim we are making is not, however, about whether those who are healthy have or have not a duty to care for those with communicable diseases, but whether or not those with such diseases have a duty to stay away from potentially infecting contacts and not transmit them further. This must, of course, also apply to health care workers who expose themselves to infection and then see other patients. Now, of course, the risk to those other patients of not receiving the health care they need may be greater than the risk of being infected by their health care worker. When there are no non-infected health carers available those who carry infection can at least obtain consent to exposure as they would and should obtain consent to treatment.

Therefore the mode of transmission itself cannot form any basis for a sound moral judgment. We could move to a convention that all risky contact becomes culpable (as we might were HIV transmitted in the same way as the common cold), but that would require a comprehensive change in social behaviour of a kind difficult to contemplate in the absence of dire emergency.1 We may well want to identify a class of contacts as involving reckless endangerment, but an extension of this class to all contacts involving risk of transmission would be excessive.


The moral duty to behave responsibly and not knowingly put other people at risk is not a duty that is confined to HIV infection or to other life threatening diseases. It is a duty which all people with communicable diseases have. It is, however, also a duty which we can expect people to discharge only if they live in a community that does not leave them withall the burdens involved in discharging this duty. The diseases we have discussed--the common cold and flu--are usually regarded as fairly trivial, but that is why we have chosen them as examples. If a duty not to communicate disease can be established for diseases of this kind it will be a general duty, and not a duty limited to serious or life threatening diseases. If we had used tuberculosis, syphilis, or smallpox as our examples our argument could have been influenced by the common perception of these diseases as serious and dangerous.

We believe we have shown a strong prima facie obligation not to communicate disease when this is avoidable. To fail to act on this obligation is deliberately to harm others. Expecting people to accept such an obligation in the absence of a system of compensation is, however, unreasonable. In many cases the loss would be relatively trivial and cost less than the mechanism of compensation. In other cases this would not be so.


This paper was written as part of a project for the Commission of the European Communities biomedical and health research programme on “AIDS: ethics, justice, and European policy.” We gratefully acknowledge the stimulus and economic support provided by the commission.