The trusted source for
healthcare information and
It’s no secret that the process of obtaining informed consent for research involving human subjects is never easy.
Investigators must try to present all of the potential risks and benefits of the intervention in a way that patients can understand; potential subjects must understand that they still may be randomized to the standard therapy, despite enrollment. And the investigators must make sure that they do not bias the participant’s decision so that the consent is truly voluntary.
Obtaining true "informed" consent is next to impossible in the best circumstances. The process is even more complicated, however, when the proposed study participant is a newborn.
"In the neonatal period, it is particularly difficult," says Jon Tyson, MD, a neonatologist and Michelle Bain distinguished professor of medicine and public health at the University of Texas-Houston. Tyson also is a member of the Neonatal Research Network, a federally funded effort to design multicenter trials in neonatal research.
"The fathers are often not there. Even if they are there, if the parents are not married, the father cannot legally consent. The mothers have experienced all of the changes that go on with pregnancy, labor, and delivery. Half of the mothers of our high-risk babies had cesareans, so they have had major surgery with anesthesia. It is an extremely difficult time to get consent in a way in which you feel that there is a meaningful exchange of information."
A recent study by British researchers1 of 200 parents and 107 neonatologists involved in clinical research in nine European countries found problems with the informed consent process in 70% of cases. In addition, even though most of the parents reported that they felt the informed consent process to be vital, a lower percentage of the doctors placed as high a value on the process.
"Our findings suggest that the quality of consent given by parents to neonatal research is commonly compromised," the authors wrote. "Our findings show that consent can be flawed, especially for research into emergency treatment or that with more risk or discomfort than standard treatment."
But the problem isn’t limited to one part of the world, experts say. It is a problem very familiar to neonatologists and perinatalogists on this side of the Atlantic as well.
"Most people think of informed consent as a process that goes on between rational, thinking adults, but that is not always the case," continues Tyson. "Anybody who has a sick child is going to be very distressed. It is a very challenging circumstance."
In addition, it is quite difficult to help the parents distinguish between consenting to participate in a study and consenting to allow a new treatment for their baby, adds Michael Cotton, MD, neonatologist at Duke University in Durham, NC.
"Often they think that giving consent is necessary for their baby continuing to receive care, and that consent will buy them new care," explains Cotton. "You have to be very judicious and very clear in explaining the randomization, that their baby may not get the new treatment, even though their baby may be very, very sick."
The ethical dilemma becomes more apparent to the investigator as well. The parents may want the unproven treatment, and the investigator, who may actually believe in the treatment, must be extremely careful not to unduly influence the decision.
"You [the clinician] may know the lab literature and other studies of the intervention you are now researching; and you know it is probably going to be OK, and you want to go ahead and give it to the baby but you can’t because you have to get consent," he says. "You can’t get their consent by being biased, by saying, If it were my baby, I’d do this.’ If you were the baby’s doctor you could; but if you are the person obtaining consent, you really shouldn’t, because you want it to be their decision and be up to them based on the merits of the study, not on your experience as a parent."
Therefore, on any neonatal consent form, it should specifically state that the baby will continue to get the standard treatment whether or not the parents agree to enroll in the study, notes Cotton.
As the researchers noted, consent is particularly hard to obtain when the research is on an emergency treatment or the baby is extremely ill. "We are currently involved in a big clinical trial on head cooling for babies with hypoxic ischemic encephalopathy," explains Cotton. "There’s about a five-hour time limit for inclusion in the study. To read a protocol with the parents and help them make a decision, it takes about an hour; if you have a five-hour window to get a kid enrolled, it makes things kind of tight."
Considering the impact of the conversation, it would be best if the person getting consent is not someone already involved in the mother’s or infant’s care, Cotton says. But that is not often the case.
"The person obtaining consent is in a tough situation in the [neonatal intensive care unit] because a lot of times the person obtaining consent is the caregiver," he states. "The parent can get confused and think that the research is part of the treatment and one and the same. That goes to biasing them and making them think that their child’s treatment is somehow going to suffer if they do not enroll in the study."
At Duke, researchers try to ensure that the neonatologist treating a baby is not the same investigator getting consent from the parents. In fact, in most publicly funded studies and many new drug trials, the study protocol specifically delineates who should obtain consent from parents, he adds.
Even when a completely objective third party presents the information and there’s enough time to thoroughly explain the study protocol, research shows that parents — even when they think they understand the information given — often have serious misconceptions about what they have agreed to do, Cotton continues. 2,3
"Studies looking at parents and their understanding, the testing done of people after they have had consent obtained from them as to what they actually consented to are really quite striking," he says. "Only about 10% to 20% get it, and the others really have no idea what they just did. Some people think they signed up for the treatment. Then, you have the parent who says, Whatever you say, doc.’ And, what do you do with that?"
The question of how to handle this remains largely unanswered, says Cotton, and really needs to be studied.
"In the large research study that we are involved in here, one of the things we eventually want to look at is the consent process, but just how we are going to go about it, what questions we should ask, we aren’t sure," he says.
Another complication, Cotton adds, is that researchers must obtain a separate informed consent from the parents to be included in the study, of the informed consent process itself, in addition to informed consent for the overall trial. "We know that [many parents] don’t get it, and we don’t know why and we don’t know what we should do to help it out and make it better."
Considering that many neonatal trials are conducted to determine the effectiveness and possible side effects of the intervention being studied, some experts question whether the term "informed consent" is even appropriate.
"The very reason you do a clinical trial is because you do not have the information that you need to make the decision that is informed enough to know whether the treatment you receive is beneficial, harmful, or of no effect," explains Tyson, who, with Paula Knudson, executive coordinator of the institutional review board (IRB) at the University of Texas Health Sciences Center, wrote a commentary accompanying a recent Lancet article.4
"In that situation, it is not like you are coming in for treatment for cancer. There is [a choice between] surgical treatment or radiation, and the side effects and risks and benefits of both therapies are well-studied; it is a matter of individual patients applying their particular values to those well-known, well-described lists of treatment benefits and treatment hazards," adds Tyson.
The current "default" situation in many NICUs is that, because it is so difficult to perform trials in neonates, and the knowledge base about how to treat newborns with certain illnesses is relatively small, many unproven treatments become part of standard clinical practice in different areas without being studied in well-designed clinical trials, say both Cotton and Tyson.
"That is really a major issue in all of medicine," notes Tyson. "Most therapies are used according to the judgment of physicians, based on the available evidence. But, the evidence is frequently not very good. It has been estimated by a number of different sources that 85% of all therapies in clinical practice have not been rigorously tested. I was involved in a review of 24 treatments commonly used in the first two hours after birth. Of those, only one or two had really been tested in good clinical trials."
Many clinicians may feel a frustration that informed consent is hard to obtain, and that, even when obtained, parents may not truly understand the consent they have given anyway, Cotton believes, noting the percentage of physicians in the Lancet study not in favor of obtaining full, informed, consent from parents.
"There is so much that is individualized to the NICU," he says. "It becomes the standard of care, and that is difficult to randomize. There are just not a lot of big, giant studies in this area."
Currently, there are two efforts, the Neonatal Research Network funded through the National Institutes of Health, and the Vermont-Oxford Network, trying to build databases of clinical information and design study protocols in multicenter, multinational trials to answer questions about certain neonatal therapies. These efforts will help provide more information, but this is still a difficult area.
"Right now, you have this problem that if you require explicit and full disclosure of a treatment in a randomized trial. You don’t require the same level of disclosure in clinical practice; you make it much more difficult for the physician to administer the treatment in a trial than in practice," says Tyson. "Then, well-meaning regulations can discourage proper testing and encourage routine clinical use of a therapy of uncertain value."
A key effort in solving this dilemma will be to have ethics committees and IRBs that are better informed about the real situations facing neonatal researchers and will develop regulations that address those issues, says Knudson. "I think often they do not understand what the clinician/investigator is facing at the time. And, we are forced to blindly follow the regulations we are given to follow instead of reaching out to determine what the real process actually is."
IRB and ethics committee members are dedicated people who most often "volunteer" their time to oversee clinical care and research trials, says Tyson, but they need more resources. "Better funding for the people responsible for deciding what studies can be done and how they can be done is very important."
1. Mason SA, Allmark PJ. Obtaining informed consent to neonatal randomized controlled trials: Interviews with parents and clinicians in the Euricon study. Lancet 2001; 356:2,045-2,051.
2. Editorial. Your baby in a trial. Lancet 1995; 345:805-806.
3. Modi N. Neonatal research. Lancet 1998; 351:530.
4. Tyson JE, Knudson P. Views of neonatologists and parents on consent for clinical trials. Lancet 2001; 356:2,026-2,027.
• Jon E. Tyson, MD, Center for Population Health and Evidence-Based Medicine, University of Texas-Houston Medical School, P.O. Box 20709, Houston, TX 77205-0708. Web: www.uthouston.edu.
• Paula Knudson, University of Texas-Houston Health Science Center, P.O. Box 20709, Houston, TX 77205-0708. Web: www.uthouston.edu.