Abstract
In March 2013, four years after trial closure, the U.S. Office for Human Research Protection (OHRP) wrote to the principal investigator of the SUPPORT trial (The Surfactant, Positive Pressure, and Oxygenation Randomized Trial) 1 informing him that the researchers were in violation of the regulatory requirements for informed consent, stemming from their failure to describe the reasonable foreseeable risks of blindness, neurological damage and death in the consent documents. The SUPPORT trial was conducted at 20 sites between 2004 and 2009, enrolling 1300 premature babies. 1 The trial was undertaken because neonatologists lacked reliable scientific evidence regarding the blood oxygen saturation levels that were the safest and most effective for these extremely premature (24–27 weeks) babies. 2 The trial sought to refine the range to determine the optimal oxygen saturation from within the range of 85–95% being routinely used in standard clinical care, to reduce the incidence of retinopathy of prematurity.
This untimely challenge and the ensuing controversy set back progress, confused the biomedical research community, alarmed the public and the parents of participating neonates, and befuddled Institutional Review Boards. 3 But it also stimulated the debate about the best way to seek consent, 4 with numerous contributions in the New England Journal of Medicine2,3,5 and the British Medical Journal. 6
In this case, we might begin by considering if is it beneficent, at a time of extreme trauma, to ask the parents of a very premature baby whose life hangs in the balance, whether they agree to the clinician choosing from the range of standard treatment options. On the one hand, this choice from the range of possible oxygen administration could lead to the likelihood of ultimately taking home a blind child (who will be prone through its developmental immaturity, to suffer from various neurological defects that will affect its ‘normal’ development). Or, on the other hand, perhaps minimising that possibility, a different choice from the standard range could potentially be at the expense of its life? Importantly, how is the clinician best able to communicate all the uncertainties that surround this proposed intervention; how might he best explain these uncertain realities, retaining trust without destroying hope? In the face of these terrible but uncertain potential consequences, where the outcomes on offer within the standard treatment options are in doubt, might not randomisation be not only the most sensible option, but also the most compassionate for both parties? And, if we believe in that proposition, what is the best way to formulate concordant action to best deal with this approach, and when? More importantly, who should be involved in deciding how we should proceed? If we are agreed that we should not continue to proceed in unresolved ignorance, how can agreement be reached to define a common goal and the means of achieving it speedily, beneficently and properly? How should public consent be achieved?
Compassionate clinicians, faced with that appalling dilemma of having to communicate their uncertainty to the parents about the unknown optimal oxygen concentration to administer, feel an obligation and a need to both reduce the degree of uncertainty and perhaps also to lift a measure of the awful parental responsibility of making a stark, under-informed choice between a potentially blind baby or a potentially dead baby. Or, in our Society, where averting and delaying the time of death is viewed by many as the over-riding goal, should the communicative approach be based on this premise, or work to a different paradigm? Aside from iatrogenically induced blindness, the neonate is clearly likely to have to overcome – through the good offices of all those who will ultimately care for the child – many developmental problems arising from our new-found technical abilities to prolong life. Debate itself exploring how we should balance the value of quality of life with the value of quantity of life in an over-regulated, risk-averse, shame-and-blame, bureaucratically driven provision of healthcare, is fraught with difficulty. Taking and shouldering personal responsibility within large organisations, challenging norms, stepping outside safe conformity, all require courage and stamina.
When considering ‘the progress’ that medicine has made, any judgement of an event needs to be set against the values and norms of society pertaining at that time. In living memory, midwives and doctors, faced with delivering a ‘life sentence’ at the moment of birth, not just on the baby, but on the carers, could, without too much fear of retrospective challenges from authority, take matters into their hands by ‘letting nature take its course’ humanly aided by their refraining from heroic actions to preserve life at all costs. We are seeing more and more retrospective challenges to perceived wrongdoing: in child abuse and in warfare, for example. Where should the limits be set?
A current inability to be able to accept the kindness of death, a belief that, at all costs, if we are technically able to save a life it should be saved, drives not only the practice of medicine, but sometimes the application of draconian, inflexible regulation and governance. This can lead to retrospective demands for justice that show precious little mercy or understanding of what this means in reality for practitioners now, and particularly for citizens having to make impossible choices caught up in life and death situations. Regulators’ measure of satisfaction is to see that everything was done ‘correctly’ according to the rules. But rules change over time.
How far back should regulators reach? Delayed challenges can draw on hindsight to justify current criticisms of past endeavours to address uncertainties about degree of benefits and harms associated with interventions in use at that earlier time, administered then with the best of intentions. Would those night nurses who turned up the oxygen in decades past want now to be accused of causing blindness to their charges? 7 Where’s the beneficence in that approach, to anyone? Regulating bodies must learn that their ill-timed interventions too can have consequences that do not benefit the populations they should be serving. The longer the delay in making the challenge, the greater is the potential for all-round harm. In pursuit of correctness, it is easier to condemn and destroy than to retrieve, renovate and rebuild. The shortcomings of the past should be used to shape the values of the future, by remembering the virtues of compassion, trust and hope that motivated earlier pioneers.
We can see this progress in action in the UK trials that followed close on the heels of the SUPPORT trial. 8 Clinicians and researchers learned collaboratively through their health-professional networks, exchanges and publications, not only about technical findings and advancements but also by working with those for whom they care: the patient-families, who can throw light on what recipients of that care value most at that crucial time of becoming parents. This way of working together – clinicians with affected families – will illumine the best concordant approach to improving every aspect of providing the most suitable care with compassion and improved understanding, both within trials and in usual care.