This thoughtful reflection on a valuable questionnaire survey of patients' attitudes regarding being told that their dentist had been infected with hepatitis B is of very direct relevance to HIV, as the authors show.¹ The measured tone and analytical approach are a welcome change from the stridency that has characterised some of the debate elsewhere. I am very conscious that more time and effort has gone into drafting and redrafting, amending, revising and refining policy in this area (in the UK at least) than in any other area of HIV control, with the probable exception of blood transfusion. Yet this is the setting that has the lowest risk among all established routes. Why has it been so hard to establish a satisfactory policy and practice to deal with this situation?

Many factors, of course, apply. However, much seems to stem from a prevailing view, fostered by recent political administrations, that health care can be delivered in a risk-free way. This has also been linked with a “blame culture”, where adverse outcomes are seen as someone's fault, and where the professions (and the “establishment”) are regarded with distrust. It has been fuelled by levels of concern about risk for HIV (and other blood-borne pathogens to a lesser extent) that have stemmed more from media “stories” about (perceived) risk than careful risk assessment.

Hence Blatchford et al comment that public anxiety about HIV relative to hepatitis B was a factor in assessing how the incident was handled, implying that HIV was of greater concern - even though the risk of transmission is some two orders of magnitude less. Interestingly, they also comment that, because the risk of death with HIV is higher than for hepatitis B, this balances the overall risk. I am not sure of the validity of this argument, though it is certainly a widely held view. It is notable that in the mid-1980s, the perceived risks of HIV outweighed those of hepatitis B in respect of the low uptake of plasma-derived hepatitis B vaccine among health care workers, even though there were over 50 deaths from occupationally acquired hepatitis B among health professionals in the USA in one year and none from HIV.

Another interesting factor in the public, and especially the media - and hence political - concern about the issue of HIV-infected health care workers, was the issue of the confidentiality of the health care worker. This is necessarily in hazard from any look back procedure, from either direct or deductive disclosure. Confidentiality is generally and properly regarded as a right among patients (though the media have been all too willing to breach it where HIV has been concerned). However, where the patient is a health care worker, it has been regarded as a right that is all too readily dispensed with. The terms “secrecy” and “cover-up” were all too often substituted to influence opinion in the media coverage of the issue. This was linked to the fact that health care professionals were seen as having drawn up the policies, and were therefore judged to be protecting their own interests.

This was very evident when sections of the press elected to “out” a number of health care workers who had been following the existing policies, because the press regarded the risk as unacceptable, despite the careful scientific deliberation in drawing up the current policies. This led the government of the day to draw up new guidance based on a very different level of risk assessment and, above all, to introduce procedures that would almost inevitably have breached the confidentiality of the health care worker. While these were in place (thankfully briefly), there was an almost complete cessation of individuals coming forward with risk for or known HIV, even to those clinicians who had publicly rejected the new guidance. Those who had previously come forward said they would not have done so if the new guidance had been in place, because it gave them, and those close to them, no protection against press intrusion into their privacy.

Thus measures that were designed to protect the public against a very low risk, had the perverse effect of reducing the willingness of health professionals to self-declare their risk, because their own rights had not been adequately safeguarded. Yet self-declaration of risk is a critical component in the effectiveness of these policies, so the policy would have, if anything, increased any risk. Subsequent guidance has largely rectified this balance. Blatchford et al recognise the need to balance these issues, though they do not articulate fully the rationale in terms of the benefits to public health and private rights. It is notable that their survey of those contacted about an incident shows that many respondents wished to know the identity of the health care worker.