Consent,confidentiality and curing sexually transmitted infection: an ethical trilemma

Introduction

Most would agree that the primary goal of medicine is to benefit the patient (see Beauchamp and Childress ¹ ). However, if practised without concern for the wishes of the patient, such beneficence may be regarded as paternalistic. The appropriate mechanism for reflecting such concern is usually characterized as ‘respect for autonomy’. This may be seen as setting moral limits on the professionals’ actions in pursuit of the goal of beneficence. Benefit for patients may only be achieved in ways patients would endorse through their own informed consent. The exercise of autonomy is the freedom of individuals to make meaningful choices concerning their health care ‘free from…limitations, such as inadequate understanding’ that might prevent such a choice (see Beauchamp and Childress ¹ ). Concern for confidentiality also arises from the principle of respect for autonomy, allowing individuals to decide the manner and extent to which information about themselves is shared with others. ²

In sexual health clinics, we are commonly concerned with the potential exposure of our patients to infections that may be sexually transmitted. In assessing such patients, we are often aware of the infection status of any sexual partner(s). Frequently, there is a potential conflict between our beneficent desire to prevent or mitigate the effects of a sexually transmitted infection (STI) in the consulting patient together with our desire to respect his/her autonomy through enabling informed consent while also respecting the autonomy of sexual partners through protection of their confidentiality. We may, for example, respect a woman's right to make a fully informed consent and, in so doing, break the confidentiality of her sexual partner by disclosing more information than he would wish. We may respect her partner's confidentiality, but ensure that she is treated with appropriate antibiotics through coercing her consent by the limited information we provide (thus failing to respect her autonomy). Or we may respect her partner's confidentiality and be adequately respectful of her right to informed choice, but fail to produce a beneficent outcome, in that she does not receive treatment. We are in a trilemma. Can these three competing obligations be balanced – and if so how? What do doctors do in practice?

Methods

A short clinical scenario was mailed electronically to a convenience sample of mainly specialty and associate specialist doctors working in sexual health clinics in the UK.

Results

Fifty-nine doctors (out of 285 mailed) returned responses to the exercise. They chose options as follows (Table 1):

OPEN IN VIEWER

Table 1

Options chosen by respondents

Option 1	Option 2	Option 3	Option 4
15	31	5	15

Thus, each of the options was endorsed by some respondents, while seven respondents endorsed more than one option. The option endorsed most frequently was option two – ‘ say that her partner has an infection which could be sexually transmitted…’. Most respondents elaborated in free text in support of their choice.

Discussion

While the results cannot be taken as representative of any group of doctors, they do illustrate the breadth of clinical responses that such a scenario may provoke in practice. It may be tempting to launch into the consultation in the manner recommended by the respondent providing comment M – by simply telling the patient that C4H is, of course, NSU on the basis that this information is ‘in the public domain’. (The author was able to confirm this by inserting ‘C4H’ into an Internet search engine and being able to identify ‘non-specific urethritis’ as the diagnosis within a few minutes.) Nevertheless this drives a ‘coach and horses’ through the logic of coding contact slips. Although the KC60 coding system was not designed for this purpose, it is commonly used as a vehicle to maintain confidentiality on these forms.

Sexual Health and HIV Property Type (SHHAPT) coding will doubtless be used in the same way. By implication, the use of a code conveys that the index patient has not given permission for their diagnosis to be made explicit to contacts (although this may not, in fact, be true). However, many services now ask the permission of the index patient to insert the true diagnosis in order to circumvent this uncertainty. Where the patient agrees this saves a good deal of time in assessing the level of awareness of sexual contacts, but even if NSU is written on the form, one can still not assume complete insight. Certainly, if a code is used, disclosure cannot take place on the casual basis that someone else might tell them or they might find out for themselves. Patients have a right to expect that information held about them will be held in confidence by their doctors. ⁸ Under General Medical Council (GMC) guidance, disclosure may be justified in the public interest if there is a ‘risk of death or serious harm’. ⁹ While this may apply in the passage of HIV, it is unlikely to apply in relation to a less serious STL Additionally, the harm generated by a reluctance to present to a clinic with a possible STI or pass on a contact slip may be great if confidentiality is seen to be eroded in this way. Another approach (response L) is to move directly from ‘bladder infection’ to ‘NSU’ on the basis that they may have the same symptoms. Nevertheless, a key difference would seem to be that one is sexually transmissible whereas the other is not. One might assume that once a patient attends a sexual health clinic he/she may already have an awareness of the possibility of STI. It may be that it is this implicit acknowledgment of sexual exposure allied with evident medical concern that provides the actual guarantee in practice that we have not over-stepped the mark in terms of disclosure; but again we cannot assume this insight.

If a disease can be described as ‘sexually transmitted’ this implies two things – first that it may have been sexually acquired (‘picked up’) and second that it may be sexually shared (‘passed on’ or ‘passed back and forth’). In an apparently long-term monogamous relationship, the possibility of the former is emotionally charged whereas the latter is far less so. Many respondents adopted the sexual sharing vocabulary avoiding exploration of how the infection might have been initially acquired. This may indeed be acceptable to some patients allowing for a certain amount of denial in those who might choose to adopt this as a coping strategy. It may also feel less uncomfortable to the health professional drawn into emotional identification with one or other party. Some will, however, wish to ask direct questions about the origin of the infection. At least one respondent elaborated robustly on the possibilities of acquiring a urethritis without recourse to another sexual partner:

Lists of differential diagnoses, without discussion of the relative frequency of these diagnoses, may suggest that each explanation is as likely as the next.

This clearly depends on the circumstances, however, and an STI would seem to be the most likely diagnosis in this scenario. However, the dictum ‘first do no harm’ is often used to express the primacy of the moral obligation of non-maleficence for good reason. NCNGU may have a non-sexually transmitted cause and this must be clearly conveyed, regardless of our awareness of his other sexual partner. Were we to identify chlamydia or gonorrhoea the situation would be different, but with NCNGU we presently cannot be certain that we are dealing with an STI. In this scenario it would plainly be wrong to jeopardize the relationship by asserting otherwise.

What about sidestepping the issue by identifying an infection in her? As pointed out by many respondents, best evidence now suggests that the presence of polymorphs in the cervix has no predictive value in identifying genital tract infection in women. Any doctor aware of this fact but indicating otherwise would be perpetuating an untruth. GMC good practice guidance ¹⁰ states that good doctors are ‘honest and trustworthy’ while specific guidance to general practitioners identifies one of the roles of a doctor as a ‘health educator’. Bell, ¹¹ in her advice to health advisers on partner notification, explicitly considers whether telling lies can ever be justified. Commonly, the act of lying may be defended on the basis that any harm is slight and ‘the end justifies the means’ (a consequentialist approach). However, consequentialism may be used to justify the breach of any moral duty unless safeguards are employed to limit its potential excesses. These safeguards are rules that cannot be broken no matter how good the outcome. Most professionals would regard the obligation of truth-telling as one such rule, and hence view any lie as a wholly unacceptable breach of trust. Option 1 thus appears ethically flawed.

Certainly, a sensitive exploration of this woman's understanding of her partner's diagnosis and the meaning of the contact slip would seem to be the wisest way to initiate any such conversation. If she fully understands his condition then the dilemma dissolves. If she reveals that her partner did mention ‘NSU’ (or something like it, as in response B), the conversation can become easier. However, if the title of the infection alone was conveyed it is still not clear that this amounted to ‘disclosure’. Her partner may not have adequately understood the pathogenesis of the condition and its possible implications, or may have simply left out important details, such as the sexual transmission of the infection.

In order to enable someone to make a medical decision that is adequately informed, the health professional must provide patients with the information that they want or need in relation to the diagnosis and appropriate treatment options. ¹² In law, there are two standards of information required for a valid consent. The first is an explanation of the ‘broad nature and purpose’ ¹³ of the proposed treatment – something to the effect that ‘this is a prescription for an antibiotic and it is intended to treat or prevent any bacterial infection’. Such a level of disclosure would be an adequate defence against a charge of battery (unconsented touching), but a higher standard of disclosure is required to refute a charge of negligence. This standard requires that the patient should be given adequate information about the diagnosis, the treatment options and any material or significant risks that apply to each including the risks of doing nothing. ¹⁴ In order to understand the risk of doing nothing, she would have to know that her partner has a potentially sexually transmissible infection, that she may have picked it up and that medical tests might nevertheless fail to identify it in her. Such a chain of events, might eventuate from option 3. There has been no such case to date, but it is possible to imagine a scenario in which a woman in this situation sues her doctor for failing to inform her of the risk of developing pelvic inflammatory disease, even in the absence of a positive test for Chlamydia trachomatis or Neisseria gonorrhoeae. If, as a result, she did not take prophylactic antibiotics and the harm of tubal factor infertility ensued, then negligence, causation and harm would all be potentially in place for a successful action. The duty of confidentiality to her partner may be cited in defense and it may be that the public interest in maintaining his confidentiality would be judged to trump her right to informed consent. However, it serves to make the point that such an inadequately informed decision might in other circumstances be deemed unacceptable. For these reasons, option 3 also appears ethically flawed.

Conclusion

Duties of confidentiality and respect for autonomy may well conflict in trying to achieve a beneficial outcome in the management of someone exposed to an STI. Sexual health doctors recognize this as a genuine dilemma (actually trilemma) encountered not infrequently in practice. Respondents dealt with this hypothetical scenario in a number of ways: firstly trying to establish if disclosure has already occurred – if so, the dilemma dissolves. Otherwise: attempting to reduce the emotional threat of STI through subtle vocabulary change; avoiding awkward questions through presenting medical management as a routine; suggesting she should accept treatment for her partner's benefit; or dealing with STI as an abstract concept. More contentious approaches included: identifying an infection in her (on insubstantial grounds), suggesting she should accept treatment for her partner's benefit; and/or by breaking his confidence to a greater or lesser extent. Although the latter appears clearly unethical, it is not clear that any of the former approaches stands up to ethical analysis either; closing down any opportunity for exploration, providing information that is lacking in important detail, or frankly misleading, or being subtly coercive all undermine autonomy.

The best approach would seem to be a subtle exploration of insight concerning the partner's condition, along with an explanation of the way that certain infections may be sexually ‘shared’. If necessary, (on direct questioning) this might include discussion of how the infection might initially have been acquired – but, in relation to NSU, must encompass its uncertain aetiology in many circumstances. These are frequently difficult consultations, partly because we may be drawn into emotional identification. Competing obligations to absent third parties appear to have drawn many of our respondents into ethically problematic responses. We should aspire to be medically well-informed, to act as honest educators, to respect confidences and withhold judgment. The fact may be that even though there might not be a ‘perfect fit’ solution, this must not draw us into exploiting convenient falsehoods, casually breaching confidences or simply avoiding sticky conversations.