A study of warning letters issued to institutional review boards by the United States Food and Drug Administration

PURPOSE: This study explores the ethical issues contained in warning letters (WLs) issued to institutional review boards (IRBs) by the United States Food and Drug Administration (FDA). METHODS: The online FDA Warning Letter Index was reviewed for letters issued to IRBs in the United States under the violation categories "Institutional Review Board" and "IRB" for the period January 1997 through July 2004. The resultant letters were evaluated for violations in 4 regulatory themes: having and following written procedures for research review; documentation of research review; IRB membership and conflict of interest; and informed consent. RESULTS: Fifty-two (52) FDA WLs were issued to IRBs during this period. Hospital/medical centre IRBs received the most letters (n = 34), followed by university IRBs (n = 9) and private IRBs (n = 9). The most common regulatory violations were failure to have and follow adequate written procedures about how the review of research is conducted (50 WLs); failure to prepare and maintain adequate documentation of IRB activities (47 WLs); and failure to provide adequate continuing review of approved studies (36 WLs). Nineteen WLs were issued for consent form issues. CONCLUSIONS: Warning letters are informative with regard to clinical research regulations and research subject protection. The content of these letters consistently indicates weaknesses in review and documentation activities of audited IRBs, potentially signalling similar issues among IRBs across the United States. Our findings, in a setting of overburdened IRBs who, in general, passively monitor studies, raise concerns about study oversight and optimal protection of research subjects.     

Structure and Practice of Institutional Review Boards in the United States

Objective: The institutional review board (IRB) is a critical element in the protection of patients' and subjects' rights with regard to their participation in research protocols. The purpose of this study was to describe the structure and current practices of IRBs in the United States.
Methods: A self-administered questionnaire was mailed to the IRB chair of each U.S. hospital with a capacity of at least 400 beds (n = 907). The survey contained 21 questions outlining committee size and structure, review of research proposals, and policies concerning scientific misconduct. Chairs also were asked what advice they would offer a young investigator preparing a proposal for submission.
Results: A total of 488 surveys (54%) were returned; 447 of the responding institutions had an IRB committee. Committees had an average of 14 members, representing 27 medical specialties. Orthopedics had the least IRB representation (10% of committees), followed by emergency medicine (12%) and ophthalmology (15%). The majority of research proposals go through 5 specific steps once submitted for review. Common reasons for proposal rejection were improperly designed consent form (54%), poor study design (44%), unacceptable risk to subjects (34%), ethical or legal reasons (24%). and scientific merit (14%). When a research proposal is rejected, 86% of the responding IRBs assist the investigator in making appropriate revisions. Although a number of IRBs (17%) have dealt with scientific misconduct allegations, only 58% have a written policy regarding research integrity.
Conclusion: Despite variations in committee structure and representation, IRBs have similar procedures for governing research. Investigators should be familiar with these procedures and are encouraged to discuss their proposal with an IRB representative prior to formal review.     

Global requirements for DNA sample collections: results of a survey of 204 ethics committees in 40 countries

The Industry Pharmacogenomics Working Group has an interest in attaining a better understanding of global requirements for sample collections intended for pharmacogenetics research. To have adequately powered pharmacogenetics studies representative of the clinical trial population, it is important to collect DNA samples from a majority of consenting study participants under many institutional review board/ethics committee (IRB/EC) jurisdictions. A survey was distributed to gather information from local and central IRBs/ECs. The survey included questions related to the approval of pharmacogenetics studies, collection and banking of samples, and return of data to subjects. A total of 204 responses were received from global IRBs/ECs with pharmacogenetic experience. The data show that requirements for approval of pharmacogenetic research differ between IRBs/ECs within and between countries but not between regions of the United States. A better understanding of differing requirements should facilitate global sample collection of DNA for pharmacogenetics research and may provide the basis for harmonized regulations for collection of genetic samples in the future.     

Variation in Institutional Review Board Responses to a Standard Protocol for a Multicenter Clinical Trial

Multicenter clinical trials require approval by multiple local institutional review boards (IRBs). The Multicenter Airway Research Collaboration mailed a clinical trial protocol to its U.S. investigators and 44 IRBs ultimately reviewed it. OBJECTIVE: To describe IRB responses to one standard protocol and thereby gain insight into the advantages and disadvantages of local IRB review. METHODS: Two surveys were mailed to participants, with telephone follow-up of nonrespondents. Survey 1 was mailed to 82 investigators across North AMERICA: Survey 2 was mailed to investigators from 44 medical centers in 17 U.S. states. Survey 1 asked about each investigator's local IRB (e.g., frequency of meetings, membership), whereas survey 2 asked about IRB queries and concerns related to the submitted clinical trial. RESULTS: Both surveys had 100% response rate. Investigators submitted applications a median of 58 days (interquartile range [IQR], 40--83) after receipt of the protocol, and IRB approval took an additional 38 days (IQR, 26--62). Although eight applications were approved with little or no changes, IRBs requested an average of 3.5 changes per site. Changes involved study logistics and supervision for 45%, the research process for 43%, and the consent form for 91%. Despite these numerous requests, all eventually approved the basic protocol, including inclusion criteria, intervention, and data collection. CONCLUSIONS: The IRBs showed extreme variability in their initial responses to a standard protocol, but ultimately all gave approval. Almost all IRBs changed the consent form. A national, multicenter IRB process might streamline ethical review and warrants further consideration.     

Conducting research using the emergency exception from informed consent: the Public Access Defibrillation (PAD) Trial experience

BACKGROUND: The Public Access Defibrillation (PAD) Trial, a prospective, multicenter, randomized clinical trial comparing two prehospital resuscitation strategies, was conducted under the regulations for exception from informed consent (21CFR50.24) in 24 communities in North America. These regulations place additional requirements for human subject protection on investigators and Institutional Review Boards (IRBs), including conducting community consultation (CC) and public disclosure (PD). OBJECTIVE: To describe the IRB approval process at study sites and the number and types of community consultation and public disclosure activities conducted. METHODS: The 24 study sites in the United States and Canada submitted IRB applications, CC and PD plans, and a structured report on IRB process and investigator perceptions to the Clinical Trial Center at the University of Washington. RESULTS: The primary IRBs for all 24 trial sites and a total of 101 IRBs approved the study. The median interval from submission to approval was 108 days (IQR 43-196), and the mean number of revisions was two (range 0-7). Investigators conducted nearly 12,000 activities to achieve CC and PD; activities varied greatly from site to site in both type and quantity. CONCLUSION: The length of time to obtain IRB approval and the extent of community consultation and public disclosure varied greatly among trial sites in meeting the current regulations for conducting emergency research with exception from informed consent. This suggests that more specific guidance may be useful and that determination of effective strategies for community consultation and public disclosure is needed.     

Regulatory challenges: lessons from recent West Nile virus trials in the United States

Delays in research on emerging infections could deprive the public of appropriate therapies. This report describes challenges encountered in implementing two multicenter protocols of West Nile virus (WNV) infections in the United States during 2003. Protocol development times, federal regulatory approvals, and local Institutional Review Boards (IRB) approvals were compiled. Twenty eight institutions participated in a natural history study and 27 in a therapeutic trial of WNV developed through the National Institute of Allergy and Infectious Disease Collaborative Antiviral Study Group (CASG). The CASG compiled protocol development times, federal regulatory approvals, and local IRB approvals. Additional information on the local IRB process was obtained by survey of the investigators. Because of the lengthy development and approval process, protocols were distributed after the start of the epidemic season, most sites were unable to enroll subjects at the peak of the season, and a number of sites lacked IRB approval at the end of the season.     

Ethical Attitudes of Nurse, Physician, and Unaffiliated Members of Institutional Review Boards

PURPOSE: To evaluate the recommendation for adding unaffiliated members to institutional review boards (IRB) by comparing the attitudes and influence of IRB members of different backgrounds, primarily nurses, physicians, and unaffiliated members. DESIGN: Survey. METHOD: A closed-ended self-administered questionnaire concerning ethical issues in human subjects research was completed by 284 IRB members in a nonprobability sample of 27 IRBs in 12 U.S. states. The attitudes and influence of IRB members with different backgrounds were compared. FINDINGS: Nurses rated most of the issues as more important than did all other members; physicians rated most of the issues as less important than did all other members; and unaffiliated IRB members rated the issues similar to the whole. Nurses and unaffiliated members were ranked the least influential IRB members, and IRB chairs and physicians were ranked the most influential. CONCLUSIONS: The responses of the IRB members in this study indicate that adding more unaffiliated members to IRBs is unlikely to increase IRB concerns with ethical issues. Adding more nurse members to IRBs might increase IRB concerns with ethical issues if the level of influence of the nurses is increased.     

The challenges of IRB review and human subjects protections in practice-based research

Practice-based research raises unique challenges with respect to human subjects protections and institutional review board (IRB) review. In this paper, three challenges posed by practice-based research are analyzed: (1) IRB review for clinician investigators who are not affiliated with an institution that has an IRB; (2) multiple IRB review; and (3) required human subjects protection training of key personnel. Investigators should be proactive in addressing required IRB review in practice-based research. In particular, they need to ensure that appropriate IRB review is obtained for all performance sites and plan for review for unaffiliated investigators. Practice-based research investigators and professional societies should educate IRB members and policy makers and publish articles regarding how IRBs might best address human subjects concerns in practice-based research. Furthermore, practice-based research investigators and professional societies should work with IRBs and the Office for Human Research Protections to facilitate centralized or cooperative review of practice-based research. Finally, practice-based research investigators need to ensure that their clinician investigators receive appropriate training in human subjects protection. Practice-based research investigators and professional societies can facilitate this process by helping to define and implement the appropriate training for busy practice-based research clinicians and office staff.     

Do patient consent procedures affect participation rates in health services research?

BACKGROUND: Few studies have examined the effects of Institutional Review Board (IRB) requirements to contact potential research participants. OBJECTIVE: To examine the association between requirements to contact potential research subjects and participation rates in a multisite health services research study. RESEARCH DESIGN, SUBJECTS: Prospective observational study of survey participation by 2673 individuals with diabetes and 1974 individuals with congestive heart failure treated at 15 clinical sites in the United States that had implemented a quality improvement intervention. MAIN OUTCOME MEASURES: Telephone survey response rates. RESULTS: Of 15 IRBs, seven required sites to obtain authorization from participants to release contact information to the study team. Five required oral and two required written advance permission. The response rate was 58% (913/1571) at sites where no advance permission was required, 39% (989/2530) from sites that required oral advance permission and 27% (145/546, P <0.001) at sites requiring written advance permission. Although 85% of eligible participants contacted directly by the study team consented to complete the survey, only 43% of individuals at sites requiring written advance permission allowed the release of contact information to the study team. CONCLUSIONS: Many potential participants did not grant advance permission to be contacted by the study team. Requiring advance permission reduced participation rates, especially at sites requiring written authorization.     

Computerized information management for institutional review boards

The use of human subjects for medical research in most industrialized nations requires the scientific and ethical scrutiny of research proposals by a governing institutional review board (IRB) or its equivalent. As part of their primary charge to protect human subjects, IRBs are responsible for the regulatory oversight of not only the research protocol itself but also the research conduct of the investigators and, if applicable, the funding sponsor. This article will discuss the regulatory requirements for an accurate account of IRB protocols and investigators and present an overview of the general flow of information for an IRB protocol. The current and potential uses of information management systems by IRBs will also be reviewed and accompanied by a discussion of the potential advantages and disadvantages of various computerized information systems for management of clinical research.     

A Survey of IRB Process in 68 U.S. Hospitals

PURPOSE: To compare IRB processes in 68 U.S. hospitals for the same multicenter study. DESIGN: Survey of IRB processes in 68 U.S. hospitals during 2001-2002. METHODS: Requirements of IRB submission including type and duration of review and qualifications of principal investigator were compared by hospital bed size, region, and academic affiliation. FINDINGS: The majority of hospitals (63.2%) were on the East coast, and mean bed size was 465 (range: 77-2,112). About one-third (33.8%) required that the principal investigator listed on the application be from within the institution, 26.5% required evidence of human subjects research training, 10.3% required a conflict of interest statement. Mean number of pages for the application was 5.24 (1-31) and up to eight copies were requested. Time from submission of the IRB application to approval averaged 45.4 days (range, 1-303 days), and the majority of reviews were "expedited" (61.8%). Expedited reviews required more time (mean, 54.8 days) than did either exempt (mean, 10.8 days) or full (mean, 47.1 days) reviews. CONCLUSIONS: Current IRB review processes are cumbersome and nonstandardized, and review time varies widely. The absence of efficient and streamlined review might unnecessarily impede national clinical research projects without improving participant safety.     

Variability in Institutional Review Board Assessment of Minimal-risk Research

OBJECTIVES: To examine variability in responses from institutional review boards (IRBs) to submission of a proposed minimal-risk survey. METHODS: Identical research proposals to obtain information concerning beliefs about the needs of victims of intimate partner violence via surveys were submitted for IRB approval to three institutions in the Baltimore metropolitan area. One institution is an academic center, one is an inner-city hospital affiliated with the academic center, and the third is a suburban community hospital. The study population consisted of emergency department health care providers and individuals in emergency department waiting areas. RESULTS: Inconsistencies emerged among the three IRBs in the review process itself, the need for participant consent, and the need for revision of the consent form and study protocol. One institution approved the proposal in 15 business days after expedited review. The second institution approved the proposal in 12 business days and waived the requirement for informed consent. The third institution approved the research in 77 business days after three revisions. Questions raised included: methodology for selecting participants; appropriateness of surveying individuals in emergency department waiting areas; a request for background literature to assure that the research questions had not already been answered; and concerns about study methodology and sample size justification. CONCLUSIONS: In this sample, there is considerable variability in IRB processes even for minimal-risk studies.     

Implementation of community consultation for waiver of informed consent in emergency research: one Institutional Review Board's experience.

BACKGROUND: This article describes how one Institutional Review Board (IRB) chose to implement the issue of waiver of consent for a research study involving brain trauma victims brought to an emergency department. METHODS: Presentations were conducted in the state of Mississippi among cultural and ethnic groups representative of Mississippi's demographic composition. Individuals from the neurotrauma research team, including neurosurgeons and nurse study coordinators, conducted all of the presentations. One IRB member served as an objective "community liaison" and attended all presentations. This individual administered evaluation forms to attendees that measured their levels of comprehension and acceptance for the use of waiver of consent in the brain trauma study. RESULTS: All of the 137 attendees in 7 community consultation meetings gave their approval for the use of "waiver of consent." Continued community consultations are planned for the duration of the brain trauma study. CONCLUSION: Based on our experience, we conclude that in collaborating with local IRBs, research teams can successfully develop strategies for obtaining "acceptable community consultations" as required by regulatory mandates. We suggest that standardized community consultation guidelines be developed for obtaining waivers of informed consent in emergency research. Such criteria should form the basis for local IRBs to obtain their respective community consultations.     

Exception from Informed Consent: Viewpoint of Institutional Review Boards—Balancing Risks to Subjects, Community Consultation, and Future Directions

Differences in interpretation of the Final Rule for exception from informed consent (EFIC) requirements for emergency research result in inconsistencies in implementation and difficulties for some institutional review boards (IRBs) to approve such research. During a consensus workshop organized by the editorial board of Academic Emergency Medicine, participants discussed how IRBs balance the risks to human subjects in EFIC research, the conduct of community consultation and its role in IRB decision making, and future directions to improve and research EFIC effects. Areas of consensus and diversity of opinion were identified. During the workshop, the National Institutes of Health model of consensus building was used to develop statements pertaining to specific questions of the effects, directions, implementation, and ultimate goals for emergency research using EFIC. The program was composed of an overview of the history and issues related to EFIC or Final Rule research and presentations of viewpoints of experts in this area of research. A final consensus was developed regarding the major topics, including IRB perspective, effective community consultation (often considered the main difficulty in implementing EFIC research), and goals for future directions and research on the topic. Roundtable discussions and breakout sessions involving interested parties were used as a format. In regard to how IRBs balance risks, by consensus it was agreed the regulations stipulate that EFIC studies must involve treatment that is unproven or unsatisfactory. The committee agreed that resuscitation rates are currently unsatisfactory, and thus current treatments are unsatisfactory. Many treatments currently used as standard care have never been proven to be effective. IRBs and the public need education that resuscitation research is needed. The same can be said for other conditions to which this rule applies. Because IRB expertise differs across the country, a group of peer reviewers to act as consultants should be available to help IRBs determine if current treatment for a condition is unproven or unsatisfactory. In regard to community consultation, the experiences of others are important and helpful as guidance. The amount and formats of community consultation should correspond to the amount of risk involved in the study proposed. In regard to future directions, communities should be asked how they define “success” of community consultation and public disclosure. Research on community attitudes is critical. Ways to continue/add to research include the following: research including major National Institutes of Health/Centers for Disease Control and Prevention funding acquisition for evaluation of the clinical impact of EFIC research; education for research funding agencies about emergency research, including current outcomes (e.g., survival rates); participation of emergency medicine researchers in meetings of research ethicists/IRB members (Public Responsibility in Medicine and Research/Applied Research Ethics National Association); publication of experiences and of the effects of EFIC research; future update meetings such as this one at the Society for Academic Emergency Medicine meeting; and more membership on IRBs of emergency physicians. While IRBs must approve EFIC research based on their own local environment, additional guidelines from regulatory agencies may be helpful. In general, current treatments for EFIC conditions are unsatisfactory and many are unproven. A group of peer reviewers can act as consultants to IRBs that do not have this expertise.     

The historical, ethical, and legal background of human-subjects research

The current system of human-subject-research oversight and protections has developed over the last 5 decades. The principles of conducting human research were first developed as the Nuremberg code to try Nazi war criminals. The 3 basic elements of the Nuremberg Code (voluntary informed consent, favorable risk/benefit analysis, and right to withdraw without repercussions) became the foundation for subsequent ethical codes and research regulations. In 1964 the World Medical Association released the Declaration of Helsinki, which built on the principles of the Nuremberg Code. Numerous research improprieties between 1950 and 1974 in the United States prompted Congressional deliberations about human-subject-research oversight. Congress's first legislation to protect the rights and welfare of human subjects was the National Research Act of 1974, which created the National Commission for Protection of Human Subjects of Biomedical and Behavioral Research, which issued the Belmont Report. The Belmont Report stated 3 fundamental principles for conducting human-subjects research: respect for persons, beneficence, and justice. The Office of Human Research Protections oversees Title 45, Part 46 of the Code for Federal Regulations, which pertains to human-subjects research. That office indirectly oversees human-subjects research through local institutional review boards (IRB). Since their inception, the principles of conducting human research, IRBs, and the Code for Federal Regulations have all advanced substantially. This paper describes the history and current status of human-subjects-research regulations.     

Neuroscience nursing research: human subjects considerations.

Permission to conduct research must be obtained from all institutions which will be involved. The special nature of our target population may result in a particularly detailed review. Specific suggestions to facilitate approval to conduct research address utilization of institutional personnel, informed consent and concerns of the IRB. Seek guidance from knowledgeable employees. You can obtain a great deal of useful information from IRB secretaries and nurses who are engaged in research in the target institution. Informing potential subjects about the research is critical, and is best done using strategies such as assuring the language used enhances comprehension. Finally, be prepared to discuss concerns of the IRB. The ability of impaired subjects to communicate with you, the researcher, seems to be of particular concern to many IRBs.     

The cost of running American city hospitals: the Gorgas 1910 survey

BACKGROUND: COL William C. Gorgas was appointed Chief Sanitary Officer of the Isthmian Canal Commission during construction of the Panama Canal (1904-1914). In 1910, Gorgas sought to determine the administrative and operating costs of major metropolitan hospitals in the United States and compare these with similar costs in the Canal Zone hospitals. METHODS: Gorgas sent a questionnaire to hospitals in Atlanta, Baltimore, Boston, Chicago, Cleveland, New York, Philadelphia, Pittsburgh, San Francisco, and Washington, DC. The information requested included number of beds, daily census, details about resident and nursing staff, salaries, length of stay, and hospital cost per patient per day. RESULTS: The survey results provide information about metropolitan hospitals in the United States at the turn of the century. Hospital costs varied from $.22 to $2.76 per patient per day. CONCLUSION: Gorgas concluded that the costs of operating hospitals in the Canal Zone compared favorably with those in the United States.