The bedrock requirement to obtain informed consent before patients may be enrolled in research has been eroding. I’ve documented the different ways and different reasons for this several times here over the years (Informed Consent for Babies: When Experts Disagree, Informed Consent in Infant Research: Ethical Problems Remain, Informed Consent in Comparative Effectiveness Research and The Erosion of Informed Consent in Medical Research). The latest example is a clinical trial of severely agitated patients conducted by Hennepin Healthcare in Minnesota. Hennepin halted its research after information about lack of consent was published in the Minneapolis Star Tribune, resulting in a public uproar and criticism from local politicians.
The study’s subjects were individuals with “prehospital agitation” whom emergency medical service (EMS) workers injected with either the sedative drug ketamine or another commonly used drug, midazolam. Which drug the patients received was determined by the research protocol. The drug was administered during incidents in which individuals were perceived as severely agitated or aggressive. Although patients were not asked to consent to having the drug administered and being enrolled in the study, they were given the choice to opt out afterward. Opting out, of course, simply meant that their data would not be used in reports of the research outcomes. It was not consent for the research itself.
Police Involved in Medical Research?
The Star Tribune reported that the Minneapolis police urged EMS workers from Hennepin Healthcare to inject individuals with the sedative, a powerful tranquilizer, sometimes even when the individuals were protesting its use. Why were the police involved in medical research? In some of these episodes, the individuals who were injected were suspected of crimes, but in other cases they were not. Some individuals were restrained prior to injection. There were clearly consequences, as some of those injected suffered heart or breathing failure and had to be revived. Intubation was required in some cases.
Not surprisingly, officials from Hennepin Healthcare objected to the report about these incidents, claiming that it was a “reckless use of anecdotes.” A subsequent Star Tribune article reported that the sponsoring institution defended the use of the drug as “medically necessary,” with the chief medical officer claiming that “ketamine can be a lifesaving tool when paramedics encounter people showing signs of ‘excited delirium.’” The hospital later added, “Hennepin Healthcare would never conduct research without appropriate consent from patients involved.” However, as was reported in the newspaper, patients were told after their involvement in the study that they could choose to have their data removed retroactively. That was one way the researchers obtained “appropriate consent” for this study.
“Waiver of Consent” Studies
The investigators described the research as a “prospective observational study,” and Hennepin’s Institutional Review Board treated it as a “waiver of consent” study. To be eligible for such a waiver, a study must be classified as “minimal risk.” The consent form for using the data informed the patients after the fact that they had been in the study. Here is what the consent form said about the risks: “Because this study involves collection of data in a setting where usual care was conducted, you were not consented prior to enrollment. This is permitted under federal regulations for Waiver of Consent Research.” In other words, the rationale for treating the study as “minimal risk” is that the research compared two drugs considered to be “usual care” for the patients’ condition. In this study, the two drugs were compared head-to-head in what is known as comparative-effectiveness research. However, the federal regulations make no mention of usual-care studies.
This episode is one more instance of a dispute that has been raging for at least the past five years. One side in this debate argues that when comparative-effectiveness research compares two interventions in common use among doctors, it counts as “minimal risk.” There is no greater risk to patients than they would undergo if they were not in the study. The opposing view contends that this is not always true, since, in usual care, patients might receive the treatment they were not assigned to in the study. An earlier study similar to the current one but using haloperidol—another drug used to calm severely agitated patients—had shown that “ketamine…is associated with significantly higher complication and intubation rates” than haloperidol. However, “ketamine was superior to haloperidol in terms of time to adequate sedation for severe prehospital acute undifferentiated agitation.” The authors concluded that “using ketamine for prehospital agitation needs to be balanced against the context of risk vs. benefit.” The earlier study makes it clear that the risks and benefits of the two “usual care” treatments differed. So the patients in the current study might have received a different drug with a different risk-benefit profile if they had not been enrolled in the study.
I am among those who maintain that a study whose risks include the need for intubation and other complications, or alternatively, a longer time for the drug to achieve the goal of sedation, should not be considered “minimal risk.” Although it is true that severely agitated patients would be likely to receive one of these two drugs were they not in the study, the risks of either drug are surely greater than minimal risk. I cannot help but suspect that classifying usual-care research as “minimal risk” is a convenient way of avoiding the need to obtain informed consent, viewed by many researchers as a burdensome bureaucratic requirement. As this current episode and my earlier posts on this blog demonstrate, the erosion of informed consent in research is an ongoing ethical issue for patients and everyone concerned about the future of medical research.
Comments on this entry are closed.
Regardless of what the risk was, there is no way to gain informed consent beforehand in this study. Minimal risk vs. not is a distinction without a difference. If a patient can consent, they would not need emergency sedation. Getting consent from family on scene is also ludicrous- they are going to read, comprehend a 15 page consent, and make an intelligent decision at 2am in the morning while their relative is going crazy and about to seriously injure themself? So, unless we want to practice blindly and never find out which is actually better, we ought to do a study like this, which is only really possible under a waiver of informed consent.
You read one study, but anyone who knows and uses the drugs knows that Haloperidal has significant risks too- they are rarer, but can be far more dangerous. It lengthens the QT interval and predispose to fatal arrythmias. Perhaps ketamine results in more intubations. But i suspect with enough patients, they will find that haloperidol causes more deaths. Indeed, it is on the Beer’s list of drugs to avoid in the elderly due to risk of increased mortality.
Ultimately, studies like these are ethical and critical if we wish to take better care of our patients.