25

national consensus about social preferences for neonatal treatment under conditions of highly uncertainty mortality in extremely preterm infants? John Robertson has characterized the Baby Doe rules as the product of a societal “consensus of sorts” [3]. He argues that the Baby Does rules are an appropriate compromise between the general (and quite legitimate) commitment to “respecting human life regardless of disability,” and the fact that the rules do not directly impose legal penalties on physicians but merely require states to set up protective procedures [3]. Yet, as Frader has noted, background behind development of the current AAP policy on Baby Doe suggests that the rules were the product of a forced political compromise to avoid substantially more punitive laws [2]. This tension evokes a phenomenon to which I have devoted a considerable amount of study – the tendency of health care providers and administrators to regard perceived legal constraints asreal constraints [35]. Stephen Toulmin has described this propensity in health care institutions as follows: “the most efficacious social facts in the actual hospital situation are, in real life, those perceptions themselves, not the objective risks and needs as these might be assessed by some impartial, outside observer” [36]. Even Robertson has recognized this point by noting that many physicians, hospital administrators, and lawyers still perceive the rules as creating an absolute legal presumption in favor of treating children likely to have disabilities, and that “technically this was inaccurate, but it was not an unreasonable conclusion” [3]. The implications of this are more striking now, however, because there is much more social controversy and litigation regarding extreme prematurity than there ever was about application of the rules to genetic anomalies. This strongly suggests that it is time to revisit the Baby Doe rules at the public policy level. My view is that the first choice would be to repeal the Baby Doe rules entirely and rely on the traditional approach that parents, with guidance from physicians, are the best arbiters of their children’s best interests. But if that proves not politically feasible, Robertson has suggested another option – to change the burden of proof that must be met in order to treat a child over the parents’ wishes [3]. From a practical perspective, in NICUs where the culture is dominated (de facto) by fear of the Baby Doe rules, parents currently have little or no input in most treatment decisions if they think such treatment would not promote the child’s interest or would impose a disproportionate burden of suffering. Under such an approach, parents would have a presumptive legal right to have their decisions about the child’s case respected unless the medical team could demonstrate by clear evidence that the child is likely to have a specified minimum level of cognitive functioning.