I have watched most of the presentations at the OHRP meeting, so you don’t have to. Many of the critics of SUPPORT make the same mistaken assumption, that usual care in the NICU is to individualize oxygen saturation targets, I don’t understand why these people (in addition to the people I have criticized before, there are 2 adult ICU docs, and the founder and president of something called the Alliance for Human Research Protection, Vera Sharav) couldn’t be bothered to actually ask someone who knows?
I wouldn’t go to a public meeting to criticize a research project in adult hematology and state with confidence that there was no usual care arm in the trial without talking to someone who knew what usual care was! One of the adult ICU docs gives the example of a baby on high oxygen that has a saturation of 92% and states that the usual response would be to reduce the oxygen saturation goals in order to reduce the oxygen exposure. I don’t know how he practices, but I certainly don’t do that. I also didn’t do that for the 15 years that I worked in the PICU either.
Lois Shepherd, another JD who is a bioethics specialist, in her presentation states something like ‘some defenders of the SUPPORT consent forms think that standard of care research means that the study is comparing 2 interventions that are commonly used and within the bounds of good medical practice’ she goes on to state ‘the SUPPORT trial would not fit within that definition either’. As is usual for such comments, she gives no reference or data, but just makes the assertion. She is, of course, wrong.
Some of the presentations, such as the one by Professor Annas, really suppose that the patient is an independent agent who seeks a medical opinion to determine which treatment to have. In such a circumstance they have the right to examine the options available with the various risks and benefits of each alternative. It is based I think on the model of the adult getting a consultation in a doctors office, the one-on-one decision making that is familiar to anyone who has watched doctor movies from the 50’s. The team I lead in my NICU consists of around 500 people (nurses, auxiliaries, RTs, clerks, cleaners, doctors at all levels of training, professionals in other fields as well) who are responsible simultaneously for around 65 babies. A good proportion of those babies would be dead if it were not for the NICU.
In that situation, the model of informed consent, which is so crystal clear to Pr Annas that he doesn’t really understand the controversy, starts to become really murky. Each day on rounds, for a particular patient I might make 10 different decisions, about the rate of feed increase, and when feeds should be enriched; about whether we will wean the rate of the ventilator or the pressures; about whether I will stop the antibiotics, continue for another 3 days, or do a CRP to help me make the decision; about whether I will ask for an echocardiogram, and if the PDA is indeed patent will I start treatment; and so on and so on. I try to make those decisions based on the best available evidence, and to weigh up the risks and benefits of each alternative. Many of the decisions are guided by protocol, because there is some evidence on which to base a protocol, or if not, the protocol is in place because we just want to make an arbitrary agreement. So our feeding guidelines are a protocol for the commencement and advancement of feeds in preterm babies. Most often the protocol works fine, and I do not adjust it, on some occasions I will decide to change the rate of increase of the feeds for certain clinical reasons.
Maybe Annas and Lois Shepherd will consider me negligent, but I do not discuss most of those options with the parents! Most of those day to day decisions if there is no unit protocol are made based on best guesses, minimization of harm, etc. Each of them might indeed carry risks and benefits that are different to the alternatives. When it is likely that a particular decision changes the risks of important outcomes, then I discuss with the parents. As I wrote several years ago, for example, before giving steroids to a child with severe bronchopulmonary dysplasia we should discuss with parents the possible long term consequences, as well as the short term benefits. As both are to some extent predictable based on published evidence (despite its limitations).
Our patients are often at nearly 100% risk of death if we do not intervene with intensive care, so, in our initial discussion with the parents we will often say that the patient may die despite our efforts. But death is rarely directly a risk of the treatment they are receiving, we dramatically reduce the chance of dying by instituting intensive care. So in the daily decisions about treatment options, if we do not know which of 2 options has the better chance of survival, then how am I supposed to inform the parents about the ‘material risk’ of death (which Dr Annas uses as an example of a risk that must always be disclosed)? For most of my decisions the risk of death is probably not particularly different between the alternatives but other important outcomes might be, and there will be some risk of death whatever I do.
If I were to design a trial to compare different feeding protocols, because I am truly uncertain whether one might be better than the other (and there are many, many different protocols out there), then I would do that because an evaluation of the evidence shows that use of either protocol would be an entirely reasonable clinical decision, even in the absence of the trial. Also, it must be crystal clear, if a baby is not tolerating their particular protocol, and I think they should deviate from it, then that is what happens. Research or no research.
Some of the presentations at OHRP such as George Annas and the entirely over the top Vera Sharav (who is tearfully outraged that we study fragile preterm babies) seem to think that we become unthinking, uncaring automatons, blindly imposing research protocols on our helpless patients the moment the misleading consent form is signed. Well, no. The parents assent to NICU care for their baby, and place a great deal of trust in me to make many hundreds of small and large decisions for their babies. Even some of the larger decisions, such as an intubation for surfactant, are not discussed with the parent beforehand to obtain their consent if it is clearly in the best interest of the baby. If a research protocol imposes a course of action that is not in line with those interests then we deviate from it. Without worrying about it. Every large study has some protocol violations, many of those are because someone in the team thinks the study directed approach is not consistent with the best interest of the patient. Even after informed consent for the trial.