Protecting research subjects under the waiver of informed consent for emergency research: Experiences with efforts to inform the community☆☆☆

Article Outline

• Abstract
• Introduction
• Materials and methods
• Results
• Discussion
• Acknowledgements
• Appendix
• References
• Copyright

Abstract 

Study objectives: The US Food and Drug Administration (FDA) and Department of Health and Human Services created a method for obtaining a waiver of informed consent to permit research in emergency settings when patients are not competent. One of the requirements to obtain a waiver is public disclosure about the research. We describe approaches used to meet the public disclosure requirement. Methods: We performed a qualitative content analysis of the FDA's repository for mandatory documentation of public disclosure efforts as of November 1999. Data from 4 studies were analyzed: 2 multi-institutional trials involving trauma and 2 smaller trials involving cardiac arrest. These efforts included both 1-way communications, such as announcements, and 2-way communications, such as focus groups and public meetings. Participation in and common concerns expressed during 2-way communications were noted, as was evidence of mechanisms of advance refusal and the collection of quantitative data on public opinion. Results: One-way communications were the predominant method of disclosure. Many 2-way communications were not directed toward lay persons. Those that were often involved fewer than 15 participants. Concerns regarding inability to refuse study participation, racial biases affecting study design and execution, and ambiguity with regard to how community input would be used in study decisionmaking were raised in several 2-way communications. A minority of participating hospitals announced mechanisms of advance refusal or collected quantitative data on public opinion concerning proposed research. Conclusion: Initial experience with obtaining a waiver of informed consent for emergency research suggests that community involvement and specific public concerns should be carefully considered in future applications for such a waiver. [Ann Emerg Med. 2003;41:72-78.]

See editorial, p. 79 .

Back to Article Outline

Introduction 

Informed consent offers a means of protecting human participants in biomedical research. However, obtaining informed consent might not be feasible in certain clinical settings because of altered mental status and the urgency of intervention (eg, cardiopulmonary resuscitation [CPR]). In an effort to permit clinical research in emergency settings when informed consent is not feasible while simultaneously protecting the rights and interests of the patients who are the proposed participants in such research, a federal rule permitting a waiver of informed consent under restricted conditions was promulgated in 1996. For such a waiver to be obtained, the institutional review board (IRB) responsible for the review, approval, and continuing review of the study must document that numerous criteria have been satisfied.¹ Among these criteria are community consultation and public disclosure before initiating proposed research.

Under the community consultation requirement, the input of communities in which the research will be conducted and from which patients will be drawn must be considered by the IRB before approving the study. As noted in draft guidelines published by the US Food and Drug Administration (FDA), community input is to be gathered during consultation sessions that (1) promote comprehension of both the proposed study and the waiver of informed consent through discussion with community members and (2) elicit community opinions and input.² Specific issues to be considered when designing such community consultation efforts are outlined elsewhere.², ³, ⁴

Under the public disclosure requirement to obtain a waiver, communities in which the study will be conducted and from which participants will be drawn must be notified of the intention to conduct the research and of the nature of the research.¹ We sought to describe and analyze efforts aimed at meeting the public disclosure requirement.

Back to Article Outline

Materials and methods 

We performed a qualitative content analysis of docket 95S-0158, the FDA's repository for mandatory documentation (eg, photocopies of announcements, transcripts of community meetings) of efforts made to fulfill the public disclosure requirement for obtaining a waiver of informed consent for emergency research. This report represents a review of docket contents as of November 1999, which includes information about 4 separate studies (Table 1).

Table 1. Trials included in docket 95S-0158.

IND, Investigational new drug; IDE, investigational device exemption.

IRB approval for our analysis was not required because docket contents are in the public domain.⁵

The information in the docket was obtained on compact disc (90.5 MB) from the FDA. One author (ANS) initially read through the docket contents and developed a framework to organize the data. Both authors then met to discuss the framework for data organization and the categorization of recurring themes found in the initial review. One author (ANS) returned to the docket to categorize and code the data. Summary tables were then prepared and interpreted by both authors.

Documentation of disclosure activities was organized by the study and hospital for which it had been submitted. Next, activities were categorized by type of communication. One-way communications generally consisted of announcing the study and involved minimal interaction between research personnel and audience members. One-way communications also announced 2-way communication events. Two-way communications involved interaction with lay and nonlay audiences (eg, medical personnel and administrators).

One-way communications and 2-way communications were grouped by type of communication (eg, meeting with a community group). This grouping was informed by the title of the activity identified in the docket, as well as any related description or documentation. Use of a type of 1-way communication or 2-way communication by a hospital participating in 1 of the 4 studies was recorded, and total 2-way communication efforts in each grouping were tallied for each study. Counts of 1-way communication efforts were not performed because documentation of the actual number of 1-way communication efforts undertaken by each participating hospital was typically unavailable or unclear.

Two-way communications directed at lay persons were further categorized on the basis of the availability of (1) data on number of participants and (2) transcripts or minutes. Themes regarding public concerns were identified inductively during preliminary review of those 2-way communications with transcripts or minutes in the docket. These themes were subsequently grouped into 5 categories: (1) interest in a mechanism to allow prospective refusal or identification of uninterested patients; (2) interest in how community input would be used in study decisionmaking (eg, to continue, alter, or cancel the trial); (3) concern regarding the sufficiency of disclosure and consultation activities; (4) concern that racial bias had or would play a role in study design or execution; and (5) other suspicions (eg, focus on hospital and/or sponsor profits outweighing the best interest of patients). Codes were developed for these 5 index categories. In subsequent reviews of the docket, documented public input was then coded and reorganized according to the 5 categories. Similarly coded 2-way communications were regularly compared to ensure consistency.

Given the public interest in mechanisms to prospectively refuse participation and identify uninterested patients, which was apparent during the initial review of the docket, announcements of the existence of such mechanisms in 1-way communications or 2-way communications were recorded. Additionally, because of the interest expressed in how community input would be used when making decisions regarding study progress, documentation indicating efforts to systematically gather data on community opinions (eg, by polling 2-way communication participants) was noted.

There was evidence in the docket (eg, statements made in cover letters accompanying required documentation of efforts) that sponsors considered some (93%) of these 2-way communications to be part of their community consultation process. However, because submission of documentation regarding community consultation is not required, the material we identified would likely provide an incomplete picture of community consultation efforts and thus be insufficient for a proper analysis of the community consultation process.

Back to Article Outline

Results 

Information about 4 different studies was included in the docket. Study A had 22 participating hospitals, study B had 9 hospitals, study D had 4 hospitals, and study C had 1 hospital.

One-way communications were the predominant type of public disclosure activity, with local newspapers and institutional publications the most commonly used media (Table 2).

Table 2. One-way communication efforts.

Ninety percent or more of the hospitals in each study advertised telephone numbers or addresses in 1-way communications for audience comments and questions. However, the majority did not report whether telephone calls or letters were received or how many were received.

Five study A hospitals and 1 study B hospital did not make use of 2-way communications. Of 2-way communications directed at nonlay audiences, meetings with specific medical or administrative groups were most commonly used (Table 3).

Table 3. Two-way communications involving nonlay audiences.

The percentage of total 2-way communications targeted at lay audiences varied by study from 60% to 100%. Meetings with specific community groups and open public meetings were most commonly conducted (Table 4).

Table 4. Two-way communications involving lay audiences and participation.

With regard to 2-way communications used in study A, the number of participants was documented for 10 meetings with community groups, 8 open public meetings, and 2 talk radio programs (number of callers), for a total of 264 participants. For study B, participation data were available for 7 meetings with community groups, 2 open public meetings, and 2 telephone polls, for a total of 182 participants. For study C, participation data were available for 1 meeting with a community group and 1 telephone poll, for a total of 48 participants. For study D, these data were available for 2 open public meetings with a total of 31 participants. Although participation varied widely, the average number of participants was in the teens for all studies when the 3 total telephone polls were excluded. The use of telephone polls increased participation substantially.

Documentation of community concerns through minutes or transcripts varied by study. In study A, there was such documentation for 11 meetings with community groups, 7 open public meetings, and 2 talk radio programs. In study B, there was documentation for 1 meeting with a community group and 2 telephone polls. In study C, there was documentation for a meeting with a community group. In study D, there was documentation for an open public meeting. Data regarding the aforementioned 5 themes are displayed in Table 5.

Table 5. Concerns expressed by community members during 2-way communications.*

TWCs, Two-way communications.

One study A hospital and 3 study D hospitals reported mechanisms to prospectively refuse participation. The study A hospital announced the availability of a notice to be carried by those not interested in participating in the trial. Individuals in study D's research population, limited to inpatients, were able to refuse participation in advance by speaking to a nurse about such a preference during an inpatient stay.

Of 2-way communications directed at lay persons, 2 in study A, 2 in study B, 2 in study C, and 1 in study D reported the systematic collection of data (eg, formal polling) on participants' support of the study or their willingness to participate. Data collected on study A 2-way communication participants (total of 17 participants for 2 meetings with community groups) demonstrated support for the study. Data from the 2 study B telephone polls, each with more than 500 participants, found that 62% and 71% would want the intervention without consent, and 64% and 76% believed the waiver was justified. The study C telephone poll included 35 community members, finding that 77% supported the study. The 13 out-of-hospital cardiac arrest survivors and family members that comprised the study C 2-way communication meeting all supported the study. Finally, all 25 attendees at the study D public forum approved of the study.

Back to Article Outline

Discussion 

Our review highlights some important concerns for future efforts at meeting the public disclosure requirement for obtaining a waiver of informed consent for emergency research, specifically the need for broad community involvement, providing and announcing mechanisms of advance refusal, and justice with respect to race and ethnicity.

If 2-way communications are used as part of the public disclosure process, 3 issues should be considered. First, participating institutions might have catchment areas spanning multiple counties, states, or both. Second, community members who respond to announcements regarding 2-way communications and participate in such events are not necessarily representative of the population from which patients will be drawn. Third, opinions voiced during 2-way communications should not be taken to represent the sentiments of all of those in attendance, unless participants are individually queried.

A well-designed effort to poll individuals who are representative of the community in which the research is to take place and from which patients are to be drawn can provide quantitative data on community opinions regarding the study's acceptability. Such a structured approach to the tabulation, analysis, and incorporation of community opinions into decisionmaking serves as a powerful response to the question raised in 6 of the 25 2-way communications with appropriate documentation available for analysis: How does community input affect decisions on study review and approval? Specific issues to be considered in poll design are described in detail elsewhere.⁶

In nearly one half of all 2-way communications with minutes or transcripts submitted to the docket, participants asked how someone might exclude himself or herself from automatic study enrollment in advance, how uninterested patients might be identified at the time of enrollment, or both. These concerns echo comments submitted during the 45-day comment period that followed the publication of the proposed rule in September 1995.¹ It appears, however, that a mechanism for advance refusal of participation, suggested but not required by the final rule, was explicitly announced by only 4 of the 36 hospitals in our review. This is problematic because, arguably, a critical component of protecting the autonomous wishes of cognitively impaired individuals under a waiver of consent is the existence and announcement of a mechanism to prospectively (before a potential patient becomes cognitively impaired) refuse study participation.

In conveying information about the proposed study to community members, public disclosure efforts might help inform decisions regarding advance refusal of participation. Moreover, mechanisms for communicating advance decisions regarding participation in a specific study might be announced in public disclosure efforts. The March 2000 draft guidance issued by the US Department of Health and Human Services and the FDA provides 2 examples of mechanisms for conveying advance decisions regarding study participation: Medic alert bracelet indications and driver's license statements.² Credit card–sized notices of refusal to participate (similar to cards refusing the administration of blood products) and comments addressing emergency research on advance directives might also be considered.

It is noteworthy that the sole study in our review in which most hospitals advertised a mechanism of advance refusal differed from the other studies in that patients were drawn from an inpatient population. Advance decisionmaking was possible in study D through communication with nursing staff. Understandably, an emphasis on simplicity has been viewed as key in the design of workable mechanisms of advance decisionmaking in noninpatient populations.

In nearly one third of documented 2-way communications, participants voiced concerns about minority patients shouldering an unfair burden with regard to research participation. This is by no means a new concern. Dula⁷ has warned that the waiver of informed consent might take an unfair toll on minority populations for several reasons. First, trauma centers participating in emergency research are likely to be located in inner cities with high numbers of resident ethnic minorities. Second, ethnic minorities are disproportionately affected by violent crimes, a chief cause of many of the medical conditions that are likely to be studied under a waiver of informed consent. Third, the history of research abuses of ethnic minorities might impede investigator efforts to meaningfully fulfill public disclosure and community consultation requirements simply because community members might be hesitant to allow investigators into their communities to discuss proposed research. Fourth, “The infrastructures in many large urban cities are so chaotic, neglected, and poverty-stricken that researchers will encounter problems in finding a safe place to go [for community discussions].”⁷

The distrust of medicine and science held by some members of ethnic minority communities must be recognized. Mechanisms to ensure the trustworthiness of proposed research activities should be put in place in an effort to build trust in medical research. Frank community-based discussions with sincere efforts to gauge opinions of community members might serve to address suspicion of medicine and medical research in certain communities.

Much can be learned from the array of creative approaches that were used to meet the public disclosure requirement for a waiver of informed consent for emergency research. However, in interpreting the results of this study, it is important to bear in mind that only 4 studies were examined and that the only data analyzed were those found in documentation submitted to the docket. Because regulations 21 CFR 312.54(a) and 812.47(a) require records of public disclosure activities to be submitted to the docket, with no similar requirement for community consultation activities, it is likely that our review does not provide a complete picture of all communication efforts.⁸, ⁹ No efforts were made to contact investigators, sponsors, IRBs, or institutions to obtain additional information. Furthermore, one author abstracted data from the docket, which might have introduced bias. In addition, errors might have been made in data grouping and coding because of incomplete documentation submitted to the docket. Finally, our analysis of concerns about emergency research was limited to 2-way communications involving lay persons with minutes or transcripts that were submitted to the docket.

Conducting research in the emergency setting is essential for the development of new and effective treatments for life-threatening conditions. Unfortunately, these situations make obtaining meaningful informed consent, a cornerstone of protection afforded to participants in research, unfeasible. As additional investigators begin to conduct research using a waiver of informed consent, it is crucial to examine the substitute protections being afforded to those involved. Thus, further investigation with additional data sources is imperative to continue to monitor the suitability and appropriateness of different measures that are used in obtaining waivers. Future prospective work will help in meeting this task. This work could include use of multiple focus groups to assess the relevance of the themes identified here, as well as others that might emerge. With such qualitative data in hand, it would be possible to develop rigorous methods to systematically sample those who would be most affected by planned research in the emergency setting to evaluate the mechanisms designed to protect their rights and interests.

Back to Article Outline

Acknowledgements 

We thank Bonnie M. Lee (Health Issues Analyst, FDA Office of Regulatory Affairs and co-author, CFR 50.24: Exception from informed consent requirements for emergency research) and Robert L. Wears, MD, MS (University of Florida Health Sciences Center, Jacksonville, Department of Emergency Medicine) for their comments and guidance. This study began as course work for a Masters of Health Sciences in Clinical Research at Duke University School of Medicine.

Back to Article Outline

Appendix 

Author contributions

. 

Both authors were involved in conceiving and designing the study. ANS was primarily responsible for data extraction. Both authors took part in data analysis and manuscript drafting. Both authors take responsibility for the paper as a whole.

 

Back to Article Outline

References 

1. Department of Health and Human Services/Food and Drug Administration/Office of the Secretary . Protection of human subjects: Informed consent and waiver of informed consent requirements in certain emergency research. Fed Reg. 1996;61(192):51497–51533
   • View In Article
2. Department of Health and Human Services/Food and Drug Administration . Guidance for institutional review boards, clinical investigators, and sponsors: Exception from informed consent requirements for emergency research. Draft Guidance. Available at http://www.fda.gov/ora/compliance_ref/bimo/err_guide.htmApril 2000; Accessed
   • View In Article
3. Sharp RR, Foster MW. Involving study populations in the review of genetic research. J Law Med Ethics. 2000;28:41–51
   • View In Article
   • MEDLINE
   • CrossRef
4. Melton GB, Levine RJ, Koocher GP, et al.  Community consultation in socially sensitive research: Lessons from clinical trials of treatments for AIDS. Am Psychol. 1988;43:573–581
   • View In Article
   • MEDLINE
   • CrossRef
5. 45 CFR 46.101.b.4.
   • View In Article
6. Sutherland HJ, Till JE. Opinion polling and decision making: A critical appraisal of quality of life assessment. Qual Life Res. 1994;3:155–162
   • View In Article
   • MEDLINE
   • CrossRef
7. Dula A. Bearing the brunt of the new regulations: Minority populations. Hasting Center Rep. 1997;27:11–12
   • View In Article
8. 21 CFR 312.54(a).
   • View In Article
9. 21 CFR 812.47(a).
   • View In Article
10. Sloan EP, Koenigsberg M, Gens D, et al.  Diaspirin cross-linked hemoglobin (DCLHb) in the treatment of severe traumatic hemorrhagic shock: A randomized controlled efficacy trial. JAMA. 1999;282:1857–1864
    • View In Article
    • MEDLINE
    • CrossRef
11. ICOS Corporation Web site . ICOS announces clinical results of hemorrhagic shock trial. Available at http://www.corporate-ir.net/ireye/ir_site.zhtml?ticker=ICOS&script=460&layout=0&item_id=178159October 19, 2002; Accessed
    • View In Article
12. Longstreth WT, Fahrenbruch CE, Olsufka M, et al.  Randomized clinical trial of magnesium, diazepam, or both after out-of-hospital cardiac arrest. Neurology. 2002;59:506–514
    • View In Article
    • MEDLINE
13. Kremers MS, Whisnant DR, Lowder LS, et al.  Initial experience using the Food and Drug Administration guidelines for emergency research without consent. Ann Emerg Med. 1999;33:224–229
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (60 KB)
    • CrossRef