Who wants to be the flu doctor?

In 1841 the William Brown struck an iceberg in the North Atlantic. The doomed ship sank with 31 passengers. The remaining passengers scrambled onto lifeboats. In one of the lifeboats, the 41 escapees were so heavy that the first mate feared that the boat would sink. He instructed his eight crewmen to “go to work, or we shall all perish.” Sixteen passengers were thrown overboard.1

A couple of weeks ago, some GPs contacted a clinical ethics committee. They cited guidance issued by the Royal College of General Practitioners, the BMA, and the Department of Health: “Every GP practice must identify the person in the practice who will act as the practice lead on flu pandemic issues.”2 Once in the front line, this flu doctor may be more exposed to flu patients than his or her colleagues. Perhaps unsurprisingly, none of the GPs in their practice wanted the job. The question, admittedly oversimplified, was who should they throw out of the boat to be the flu lead? The guidance did not specify how the lead should be selected.

The optimal way to resolve the problem is to seek a volunteer. By reassuring the staff of the relatively small risk and reminding them of the training offered, their duty of care, and the benefits to patients, it may be possible to persuade one member to assume the role. To entice volunteers, the practice may want to offer incentives, such as “risk pay.” Let us assume, however, that in spite of the reassurances and benefits, none of the GPs step forward. What next? There are two main options, each unpalatable—(1) pressure someone to be the flu lead, or (2) ignore the guidance, which would disadvantage patients in the area, increase the risk and workload of colleagues elsewhere, and put the GPs at administrative risk.

In 1841 the first mate ordered the crew on the lifeboat “not to part man and wife, and not to throw over any women.” It is doubtful that the second part of this gentlemanly principle would hold water in 2009. The first part raises the issue of competing duties.3 If a GP is married, he or she has responsibilities as a spouse. If the GP has children, he or she has responsibilities as a parent. These responsibilities clash with the medical duty of care. If such GPs are flu leads, they run a risk of failing in their duties as spouse and parent. They may be quarantined or decide willingly to stay away from their loved ones when treating flu patients. Single, childless GPs, unburdened by such conflict of duties, may have a stronger moral obligation to volunteer.

In the real world, however, this is unhelpful. A GP might protest: “I’ve got just as strong a commitment to my long term partner” or “I’ve got a demented mother who needs me.” How would we decide which competing duties are strongest? And, of course, people can invent bogus reasons that would be impractical or impossible to verify. Even if it were possible, there is no guarantee that the reluctant recruit would turn up to work when called upon. If one candidate is clearly better suited to the task than the others (“clearly” because in such situations it would be tempting to find spurious reasons why one’s colleagues are better qualified for the job)—for instance, he or she was an infectious disease consultant before becoming a GP, then this would strengthen the individual’s obligation to volunteer, although again this may be offset by other non-professional obligations.

If no substantive criteria for selection can be chosen, either because of intractable disagreement or impracticality, a procedural system should be adopted. The first mate’s initial suggestion for choosing the victims was to cast lots, and this system, if a little crude in appearance, has the benefit of objectivity, equality, and practicality. In a critical care situation with insufficient beds or in a sinking, overpopulated boat, when time is of the essence and sentimentality perilous, a lottery system may be the best available method. It is also unlikely to create as much unrest in the practice as other methods based on matrimonial status, number and age of children, clinical experience, age of practitioner, and so on. The coercive element is reduced if the candidates consent to the selection process. To sweeten the pill, the unwilling appointee could receive the benefits originally offered to prospective volunteers.

I have outlined some thoughts on this emerging issue, but GPs should discuss between themselves the fairest way to select the flu lead and the matter of risk distribution. The team may agree to exonerate some members from consideration, appoint more than one flu lead, or distribute the flu patients equally among members. If the volunteer option fails, some GPs might reject the lottery system, preferring another method. Reasoned deliberation should then determine which method is best in the circumstances.

It is doubtful that any system will be unproblematic. The volunteer may not be the best person for the job. The GP who draws the short straw may have a pregnant wife, disabled children, or frail parents and be more immunologically susceptible to the disease than his or her colleagues. Each alternative will have its advantages and disadvantages, respecting some moral values and ignoring others. In such difficult dilemmas, justice asks for no more than a morally defensible and transparent process of decision making. As the maxim goes, justice must not only be done but must be seen to be done.