Objectives. When prospective informed consent is not feasible, clinical research that presents more than minimal risk can proceed only after a community consultation and public disclosure process and the granting of exception from informed consent from the federal government. The applicability of exception from informed consent to pediatric resuscitation research has not been described. The objectives of this study were 1) to perform a community consultation and public disclosure process specific to a trial of induced hypothermia immediately after pediatric cardiac arrest and 2) to determine the applicability of exception from informed consent to randomized, controlled trials of emergency interventions after resuscitation from inpatient pediatric cardiac arrest.
Methods. Focus groups, information sheets with options for written responses, posted notices, e-mails, and telephone conversations with parents of critically ill children and hospital staff were conducted at a tertiary care children's hospital. Data were stored, organized, and retrieved using NVivo qualitative analysis software (QSR International).
Results. In focus groups (n = 8), parents (n = 23) and hospital staff (n = 33) concluded that prospective informed consent is not feasible for a trial of induced hypothermia after inpatient pediatric cardiac arrest. Focus group participants endorsed exception from informed consent for a trial of induced hypothermia but only if study information is easily available prospectively and if all parents have an explicit opportunity to decline participation in a verbal conversation before study enrollment. Separate from and without knowledge of the focus group results, 7 (100%) of 7 parents of past or current patients and 21 (50%) of 42 hospital staff who provided written opinions endorsed exception from informed consent for this study. Five (12%) of 42 hospital staff opposed, and 16 (38%) of 42 were neutral. In telephone conversations, 14 (70%) of 20 parents of children who were previously resuscitated from cardiac arrest endorsed exception from informed consent for this study, 3 (15%) of 20 opposed, and 3 (15%) of 20 were unsure.
Conclusions. Community consultation for inpatient resuscitation research can be conducted in a children's hospital, with hospital staff and parents of patients as the relevant community. Exception from informed consent is necessary and appropriate for a randomized trial of induced hypothermia begun within 30 minutes after pediatric cardiac arrest. A process in which families are informed prospectively and have a pre-enrollment option to decline participation will likely be acceptable to families, health care providers, and the institution.
Clinical trials of therapies with potential to improve outcome from pediatric cardiac arrest are impeded by the difficulty of obtaining traditional informed consent during or immediately after arrest. Induced hypothermia has been demonstrated to confer neuroprotection in adults who are resuscitated from out-of-hospital ventricular fibrillation,1,2 but clinical equipoise exists regarding its use in children. A randomized, clinical trial is needed and is in development.
If research that presents greater-than-minimal risk is to take place in the context of sudden life-threatening conditions, then traditional informed consent may not be possible. The Food and Drug Administration (FDA) and the Department of Health and Human Services (DHHS) published guidelines in 1996 allowing emergency research to proceed with an exception from informed consent in specific situations (Table 1). 3–6 Research with an exception from informed consent requires a process of community consultation and public disclosure and the subsequent approval of the involved institutional review boards (IRBs) and the FDA or DHHS. Determination of the relevant community and the type of public disclosure and consultation necessary are left to the discretion of investigators and IRBs. Clinical investigations involving adults have taken place with an exception from informed consent,7–10 but children have been excluded from these studies in part because of their unique informed consent issues. Although at least 1 randomized neonatal resuscitation trial was conducted abroad without informed consent before the 1996 guidelines,11 no published pediatric studies have been conducted under an exception from informed consent using a community consultation and public disclosure process.
The authors performed a community consultation and public disclosure process specific to a randomized, controlled trial of induced hypothermia initiated within 30 minutes after inpatient pediatric cardiac arrest. We hypothesized that the hospital community (health care professionals and parents of current and past patients) would determine that such a trial cannot be performed with traditional informed consent and that they would endorse the application of exception from informed consent to such a trial.
The components of the community consultation and public disclosure process conducted are summarized in Tables 2 and 3. Eight focus groups were conducted. Each focus group involved peers selected purposively on the basis of the community from whom participants would be drawn and within which the research would be conducted.12 Parents of all pediatric intensive care unit (PICU) patients were invited by an investigator to participate on the day before and the day of each focus group. All parents of children who were resuscitated from cardiac arrest between January 2000 and June 2002 (survivors and nonsurvivors) were invited to participate by an investigator through a letter and a follow-up telephone call. Health care providers and hospital staff were recruited through e-mail and posted notices. Hospital administrators were invited by telephone, with care taken to ensure participation by representatives of the legal department, risk management, nursing management, public relations, and high-level administration.
Focus groups were led by an investigator (R.M.N.) who was experienced in the conduct of focus groups, with at least 1 other study investigator and 1 observer from the IRB present at each focus group. Written informed consent was obtained from focus group participants, and each session was audio-recorded in duplicate, then transcribed by a professional medical transcriptionist. Each transcript was reviewed for accuracy, and all statements identifying a participant were removed.
Letters describing the proposed study and containing a form for elective written feedback were mailed to the parents of 49 children who survived for at least 24 hours after being resuscitated from cardiac arrest while admitted to this institution (outside of the neonatal ICU) between January 2000 and June 2002. At least 2 attempts were then made to contact each family by telephone. All parents were asked consistent, scripted questions and given the opportunity to ask questions and to share their opinion of exception from informed consent for a trial of induced hypothermia. All of these responses were documented verbatim and categorized as 1) endorsing exception from informed consent for a study of induced hypothermia, 2) disagreeing with exception from informed consent for this study, or 3) neutral/unsure.
Notices describing a trial of induced hypothermia with an exception from informed consent and encouraging feedback were posted in areas of high PICU parent traffic and were available on a table at the family resource center. A handout describing the proposed trial and inviting feedback was distributed to the parents of all children who were admitted to the PICU on 2 separate occasions.
An information letter describing the proposed investigation with exception from informed consent was sent by in-hospital mail to the all of the attending physicians (n = 121) in each of the following pediatric specialties: general surgery, neurosurgery, otolaryngology, emergency medicine, pulmonology, cardiology, cardiothoracic surgery, and oncology. The physicians were encouraged to respond to the authors if they had questions, concerns, or comments. The proposed trial was presented at grand rounds in the department of anesthesia and critical care medicine, at the PICU research council, and at PICU charge nurse meetings. All opinions voiced at these meetings were recorded by an investigator. Notices describing the proposed project were posted in areas frequently visited by clinicians, with methods of contacting investigators clearly indicated. An informational e-mail describing the proposed study and inviting feedback for those with questions, concerns, or comments was then sent to the entire children's hospital e-mail community, consisting of 8136 mailboxes, including physicians, nurses, administration, support staff, and other hospital employees. Every response was documented, rendered anonymous, and saved; responses were classified as endorsing, disagreeing, or neutral.
In focus groups, written letters, and e-mail communications, the rationale for induced hypothermia after resuscitation from cardiac arrest was described. It was explained that induced hypothermia has been demonstrated to provide neuroprotection in animal cardiac arrest models and in adults resuscitated after cardiac arrest resulting from ventricular fibrillation.1,2 It was described that for this proposed trial, target temperature would be 32°C to 34°C maintained for a period of 24 hours and that hypothermia would be achieved using a cooling blanket and generous application of ice packs. Focus group participants were told that the potential risks of induced hypothermia include shivering requiring the use of neuromuscular blockade, thermal injury to the skin, excessive bleeding, and cardiac dysrhythmias.
Focus group transcripts were stored, organized, and retrieved using the NVivo qualitative analysis software package (QSR International Pty Ltd, Doncaster, Victoria, Australia). A coding scheme to categorize segments of the transcripts according to relevant concepts and themes was developed on the basis of the content analysis of the transcripts and the prospectively designed research questions. Two investigators (M.C.M., F.R.W.) of different professional backgrounds (medicine, nursing) worked on a subset of the data to develop a shared understanding of the codes. Transcripts were imported into the NVivo program, and the same 2 investigators independently coded the transcripts. After coding, all differences between the coders were resolved to the satisfaction of both coders. The NVivo software was used to compile and review all statements relevant to each specific code.12,13 Focus groups were conducted until no new concepts were emerging (ie, thematic or theoretical saturation was reached),12 with a total of 8 groups.
Focus group participants (Table 4) concluded that a trial of induced hypothermia after pediatric cardiac arrest should be conducted and that it is not feasible to obtain a meaningful informed consent within 30 minutes after resuscitation. Many parents stated that they would agree to anything that a doctor asked them at that time and that they would not understand what was being asked (eg, “At that point . . . you're so desperate for anything to be done for your child, you're probably going to say yes anyway”).
In every focus group, the possibility of attempting to obtain informed consent at the time of hospital or ICU admission was discussed. Focus group participants identified the following problems with this approach: 1) the mental state of a parent when a child is admitted to the hospital or ICU (“The last thing that we needed to hear, when we brought a 3-day-old into the hospital was, ‘If your child happens to turn blue or needs to be resuscitated, we might do a test study’”); 2) the lack of attention likely to be paid to the informed consent document when a parent does not believe that the study will apply to his or her child; and 3) the parent's emotional burden of contemplating that his or her child may have a cardiac arrest, even when the child is at no substantial risk of this event. Health care providers also discussed the labor necessary to obtain informed consent from all patients who are admitted to either the hospital or the ICU.
Despite the impracticality of obtaining prospective consent, there was widespread agreement that providing information to parents before study enrollment is necessary. Parents suggested distributing a brochure describing the study to all parents at hospital admission and having notices posted prominently in areas frequented by parents throughout the study period. Parents further emphasized the desire to have a brief verbal discussion of the study in the 30-minute interval between resuscitation and study initiation, at which point they would be given the opportunity to decline participation.
Parents of past and current patients expressed discomfort with the control arm of the proposed study (“Asking someone if they would like to participate in a study is one thing, agreeing and not seeing things being done for your child, that's another. . . . I'm going to know my child was not selected. I'm not going to be happy about it.”). Parents expressed that they may be angry if their child had a bad outcome and had been randomized to the “wrong” arm of the study (“If it turns out that it's beneficial for the child to be cooled but you didn't cool my child because he wasn't picked, I would be furious that my son could have brain damage because you just decided not to do it.”). In 1 focus group, parents of children who were admitted to the PICU suggested that researchers not tell parents of children who were randomized to standard therapy that they were involved in a study (“And you won't tell the parents that this was a coin toss, of course. That would kind of be cruel”). Only 1 parent expressed an apparent preference for the control arm of the study (“Better let them just go ahead the way he is. I'm looking for a sure-fire thing.”).
Despite extensive explanation in lay terms by the focus group leader, many parents did not understand the nature of a randomized, clinical trial. A common misunderstanding was that all children enrolled in the study would be treated with hypothermia. Understanding the need for a concurrent control population was limited (“If we could be part of this study, absolutely. I would have said: he's getting the ice. Give it to him. I understand that it's random.”).
Some focus group participants raised concerns about the public perception of an institution and about legal liability (“The headline the next day after there was a negative outcome will be, ‘Research Performed Without Consent at Institution X.’”). The administrative staff focus group concluded that conducting a trial as part of a multicenter, nationally funded investigation would make the study more acceptable to the institution and to the public.
Responses to Written Informational Communications
Forty-two hospital staff responded by e-mail to either a written letter or the hospital-wide e-mail announcement; responses to these 2 different communications were pooled. Twenty-one (50%) of 42 respondents supported the application of exception from informed consent to the proposed research, 2 with the qualification that the study must be low or no risk. Five (12%) of 42 opposed exception from informed consent for the proposed investigation, and 16 (38%) of 42 posed questions or comments without expressing a positive or negative opinion regarding the research proposal. A hospital administrator received 14 additional comments in response to the general e-mail announcement. Three (21%) of 14 supported the investigation with specific qualifications (IRB and divisional approval, demonstration that risks are minimal), 3 (21%) of 14 disagreed with the proposed research, 2 of 14 asked whether the e-mail was a joke, and 6 of 14 were neutral. The e-mail and written letters were sent before completion of the focus groups. These communications did not describe the possibility of providing information to families at hospital admission and providing all families with the opportunity to opt out of participation before study enrollment as these proposals developed during focus group discussions.
On 2 occasions separated by 8 weeks, handouts describing the proposed study were distributed to all parents of children who were admitted to the PICU, with a comment sheet to return if they had opinions that they wanted to express. Sixty-one handouts were distributed. Four of 4 parents who chose to return written comments believed that the study should proceed. The investigators received no communications that were attributable to the 14 notices posted throughout the parent and health care provider areas of the PICU.
Responses to Letters and Telephone Calls to Parents of Children Who Were Resuscitated From Cardiac Arrest While Admitted to This Institution
Letters were sent to parents of all children who were resuscitated from cardiac arrest between January 2000 and June 2002 (n = 49). Twenty-three of 49 children had died during the same hospitalization, and 26 of 49 had survived. Five letters were returned undelivered because the family no longer lived at the address identified in the medical record. Three of 3 families who elected to return a sheet with written comments indicated that the study should proceed. The parents of 23 children who were resuscitated from cardiac arrest were reached by telephone. A summary of the opinions expressed in Telephone conversations is shown in Table 5. The letters were mailed and the telephone conversations were conducted before completion of the focus groups. These communications did not describe the possibility of providing information to families at hospital admission and providing all families with the opportunity to opt out of participation before study enrollment as these ideas developed during focus group discussions.
Parents and health care providers uniformly agree that prospective informed consent is not feasible for a randomized trial of induced hypothermia beginning within 30 minutes after resuscitation from inpatient pediatric cardiac arrest. Parents and health care providers cite the emotional state of the parents and the volume of information that parents must absorb after a child experiences a cardiac arrest as precluding the possibility of a meaningful informed consent process in the immediate postresuscitation period.
An alternative to seeking informed consent in the immediate postresuscitation period is to obtain informed consent at the time of hospital admission from all parents or from the parents of all ICU patients. A substantial amount of labor would be necessary to perform this task. As only ∼1.8% of children who are admitted to a PICU experience a cardiac arrest14 and only ∼67% of these children survive the event, at least 100 families would need to complete an informed consent process for each potentially eligible subject. The second drawback to prospective informed consent at PICU admission is the questionable validity of informed consent obtained from a parent who does not anticipate that his or her child will ever be eligible for study participation. Last, at the time of ICU admission, families are emotionally stressed and have a formidable amount of paperwork to address. Parents of critically ill children stated that they would find it upsetting to be approached for consent at the time of PICU admission. Approaching each family at the time of PICU admission would generate unnecessary anxiety in the large number of parents whose children would not become participants.
Without a complete process of informed consent, a trial of induced hypothermia can take place in the United States only with an exception from informed consent. However, parents and health care providers believe strongly that it is inappropriate to enroll children in this trial without parental awareness. In focus groups, parents and health care providers concluded that exception from informed consent may be acceptable if all families of potentially eligible patients are aware of the ongoing study and if parents are given an opportunity to decline participation before study enrollment. Focus group participants suggested the use of posters and brochures to inform families prospectively of the study and a brief verbal discussion in the immediate postresuscitation period to allow parents the opportunity to decline participation. This is consistent with the FDA and DHHS stipulations for conduct of exception to informed consent for emergency medicine research (see Table 1).
Understanding Research and Randomization
Parents in focus group discussions revealed significant discomfort with randomization and a strong intervention bias. Parents and health care providers expressed concern for the families of the children who were “not chosen” for cooling. Our research reinforces published literature indicating that many parents poorly understand the method of controlled clinical trials.15,16 Investigators who use community consultation methods must be careful to educate the communities involved about clinical trials and to confirm understanding.
Relevant Community for Research Involving Inpatient Cardiac Arrest
The legitimacy of community consultation depends on the shared experiences and values in a given community. In a heterogeneous society, determining the appropriate community for consultation regarding clinical research is difficult. Families of children who experience an inpatient cardiac arrest come from a multitude of ethnic, socioeconomic, and religious communities. The largest commonality among these families is the experience of having a critically ill child in the hospital setting. We took the approach that a community consultation and public disclosure process for inpatient resuscitation research ought to focus on the inpatient hospital community, rather than on the geographic catchment area for a particular institution. The IRB at this institution endorsed this initial approach in developing with us this community consultation and public disclosure process.
A challenge in any community consultation process is ensuring that members of the community have ample opportunity to provide input. In this community consultation process, information was broadly disseminated with contact information clearly indicated. All notices encouraged responses by any interested party. This was different from a survey tool with the expectation of a high response rate; rather, the goal was to continue to collect opinions until thematic saturation had been achieved. Consistent with previous community consultation literature,7,8 only a small minority of the individuals who were contacted personally or those who were exposed to posted notices chose to provide opinions to the investigators. The use of focus groups as an integral part of the community consultation process provided an in-depth assessment of community attitudes that would not otherwise have been available. The critical factor in a community consultation process is not the number of community members that provide formal response to the investigators but rather that there is sufficient opportunity for community members to provide input if they choose. Investigators who use community consultation methods must ensure that information is available and easily understandable and that means to provide input are clearly indicated.
A concern for institutions considering participation in a clinical trial that uses an exception from informed consent is legal liability and public relations repercussions. Conducting extensive communication with the hospital community before, during, and after the study period and performing clinical trials as part of a multi-institutional network with federal funding were suggested by administrative staff focus groups to increase community understanding and acceptance of this type of research.
Importance of the Timing of an Intervention
The timing of an intervention is critical to the decision of whether an investigator should seek permission to bypass the traditional informed consent process. Our community consultation and public disclosure process emphasized that a trial of induced hypothermia should be initiated within 30 minutes after cardiac arrest. Clinical trials of hypothermia after birth asphyxia have been conducted in the United States after obtaining informed consent, with a delay between asphyxia and initiation of cooling.17,18 However, we believe that for hypothermia to be maximally effective after cardiac arrest, cooling must begin soon after resuscitation.19,20
Every effort must be made to allow a parent to decide whether a child will participate in clinical research, even in situations in which complete informed consent is not feasible. For inpatient resuscitation research, there is an opportunity to inform parents of ongoing research at the time of hospital admission. If an intervention can be delayed for 30 minutes or more after resuscitation, then there is also an opportunity to briefly remind parents of the investigation and allow them to decline participation. It is not clear from this study how soon after resuscitation families would be able to participate in a valid informed consent process, but we posit that for therapies that can be delayed for several hours after resuscitation, complete informed consent should be sought. For inpatient resuscitation research involving therapies that must be instituted during cardiac arrest (eg, comparison of high-dose with low-dose epinephrine), the opportunity remains to make families aware of the study at the time of hospital admission, but there is no opportunity to allow families to decline participation at the time of eligibility. Additional research is necessary to determine the extent to which the hospital community would endorse exception from informed consent for such a trial.
Limitations of the Study
Although the authors attempted to reach as wide a segment of the hospital community as possible, responses were voluntary and we cannot rule out response bias. We consulted with the inpatient hospital community, including patients, staff, families, and administrators, but did not specifically consult the geographic community.
The results of this investigation are pertinent to randomized, controlled trials that compare 2 accepted therapies implemented shortly after resuscitation from inpatient cardiac arrest. Additional studies are necessary to determine the applicability of exception from informed consent to out-of-hospital pediatric resuscitation research, to experimental therapies that carry higher risk, and to emergency interventions that are applied during cardiopulmonary resuscitation.
In the United States, approval must be obtained from the FDA or DHHS for any clinical trial to proceed with an exception from informed consent. We have not yet sought approval for a trial of induced hypothermia after pediatric cardiac arrest, and it cannot be assumed that the community consultation and public disclosure process that we performed would be deemed acceptable by these regulating agencies.
This study provides 1 potential blueprint for the conduct of community consultation for an inpatient clinical trial with an exception to informed consent, using the hospital and its patients as the relevant community. This community consultation confirmed that traditional informed consent is not feasible for a randomized trial of a low- to moderate-risk therapy (induced hypothermia) instituted within 30 minutes after pediatric cardiac arrest. A community consultation and public disclosure process identified the optimal process as one in which families of potentially eligible patients are informed prospectively that a study is taking place (eg, through the use of brochures and posters) and have an explicit opportunity to decline participation before study enrollment.
- ↵Informed Consent Requirements in Emergency Research. OPRR Reports, 1996. Available at: ohrp.osophs.dhhs.gov/humansubjects/guidance/hsdc97-01.htm. Accessed August 5, 2002
- Protection of Human Subjects. Office of Regulatory Affairs, 1996. Available at: www.access.gpo.gov/nara/cfr/waisidx_02/21cfr50_02.html. Accessed August 1, 2002
- ↵Sloan EP, Koenigsberg M, Houghton J, et al. The informed consent process and the use of the exception to informed consent in the clinical trial of diaspirin cross-linked hemoglobin (DCLHb) in severe traumatic hemorrhagic shock. DCLHb Traumatic Hemorrhagic Shock study group. Acad Emerg Med.1999;6 :1203– 1209
- ↵Lewis RJ, Berry DA, Cryer H 3rd, et al. Monitoring a clinical trial conducted under the Food and Drug Administration regulations allowing a waiver of prospective informed consent: the diaspirin cross-linked hemoglobin traumatic hemorrhagic shock efficacy trial. Ann Emerg Med.2001;38 :397– 404
- ↵Saugstad OD, Rootwelt T, Aalen O. Resuscitation of asphyxiated newborn infants with room air or oxygen: an international controlled trial: the Resair 2 study. Pediatrics.1998;102 (1). Available at: www.pediatrics.org/cgi/content/full/102/1/e1
- ↵Pope C, Ziebland S, Mays N. Qualitative research in health care. Analysing qualitative data. BMJ.2000;320 :114– 116
- ↵Kodish E, Eder M, Noll RB, et al. Communication of randomization in childhood leukemia trials. JAMA.2004;29 :470– 475
- ↵Battin MR, Dezoete JA, Gunn TR, Gluckman PD, Gunn AJ. Neurodevelopmental outcome of infants treated with head cooling and mild hypothermia after perinatal asphyxia. Pediatrics.2001;107 :480– 484
- ↵Azzopardi D, Robertson NJ, Cowan FM, Rutherford MA, Rampling M, Edwards AD. Pilot study of treatment with whole body hypothermia for neonatal encephalopathy. Pediatrics.2000;106 :684– 694
- Copyright © 2004 by the American Academy of Pediatrics