Research bioethics in the Ugandan context II: Procedural and substantive reform

In July 1997, the voting representatives at the National Consensus Conference (NCC) on Bioethics in Health Research voted unanimously to adopt the proposed Guidelines for the Conduct of Health Research Involving Human Subjects in Uganda (Guidelines). This vote represented the culmination of a three-year

Attendees at the NCC included both voting representatives and non-voting participants. Voting representatives had been designated by the institutional entity with which they were affiliated to represent that entity at the NCC and to voice the concerns and opinions of the entity and its constituents. Voting representatives included individuals from various governmental organizations such as the Ministry of Health, the Ministry of Defense, the Attorney General's Office, the Uganda National Council of Science and Technology (UNCST), the National Drug Authority, and the National Cancer Institute; Makerere University; various medical associations, such as the Protestant Medical Association; nursing and pharmacists' professional organizations; various churches, legal service agencies, human rights organizations, and media personnel. The NCC had been widely advertised to encourage attendance and participation of non-affiliated persons. These included, for instance, university students, participants in ongoing research, and freelance media personnel. Both voting representatives and non-voting participants participated actively in the discussion and debate of the proposed Guidelines and revisions to the proposal. Proposed revisions to the proposed Guidelines could be made by any attendee from the floor. (See Appendix for listing of voting representatives.)

The newly adopted Guidelines were formulated following an examination of then-existing policies and procedures in Uganda, of numerous international documents governing research involving human subjects, and of regulations and policies governing human research in Australia, the United Kingdom, and the United States. These Guidelines make significant procedural and substantive changes in Uganda's process of bioethical review. Those that are most significant are discussed below in the context of Ugandan culture and history and currently accepted international principles, and are compared with United States regulations relating to research involving human subjects.

The Process and Scope of Review

The Guidelines establish multiple levels of review, beginning at the institutional level with institutional review committees (IRCs) and extending to the AIDS Commission for HIV-related research and to the NCST for all research, including that which is HIV-related. The regulations are intended to apply not only to universities and governmental entities, but also to nongovernmental organizations that may conduct health-related research, although not labeled as such, in conjunction with their program development or evaluation. Pursuant to the Guidelines, IRCs must now consist of at least 5 members of diverse tribes, religions, professions, and socioeconomic status. At least one member of the IRC may not be otherwise affiliated with the research-sponsoring institution in any way. Membership will rotate periodically and may be involuntarily terminated where it is established that an individual has committed misconduct or has refused to excuse him- or herself from discussion where there exists a conflict of interest. This attention to conflicts of interest and the provision for involuntary termination of membership represent significant departures from previously existing procedures and practices.1

These criteria for IRCs were formulated based on an examination of Guideline 14 of the International Ethical Guidelines for Biomedical Research Involving Human Subjects [CIOMS Human Subjects]2 and paragraph 38 of the International Guidelines for Ethical Review of Epidemiological Studies [CIOMS Epidemiology].3 Together, these provisions require the review and approval of research by independent ethical committees and provide guidance on the composition of such committees. Additionally, the Uganda Guidelines draw on the new Ugandan constitution, which prohibits discrimination on the basis of sex, race, color, ethnic origin, tribe, birth, creed or religion, social or economic standing, political opinion, or disability.4 The Guidelines further establish procedures for expedited review and standards for the submission of protocols for review.

In addition to requiring that investigators provide details relating to research methodology, the Guidelines require that the protocol include complete details of the informed consent procedure and of mechanisms designed to ensure the confidentiality of the data. IRC members are charged with the responsibility to ensure that any approved protocol (1) minimizes risks to the study participants, (2) adequately reflects and balances the risks and anticipated benefits to the participants, (3) provides for the equitable selection of research participants, (4) requires both the receipt and documentation of informed consent from each individual participant, (5) establishes adequate mechanisms to ensure the confidentiality of the data and the privacy of the participants, (6) establishes procedures to assure the safety of the participants, and (7) establishes additional safeguards for the protection of participants vulnerable to coercion or undue influence.5 These requirements implement in the Ugandan context various provisions of the CIOMS Human Subjects, including Guideline 1, which pertains to informed consent; Guideline 2, which relates to information to be provided to prospective participants; and Guideline 3, which addresses the obligations of investigators with respect to informed consent.6

IRCs will now have authority to temporarily and provisionally suspend or terminate research that is in violation of the IRC's original approval or that has been associated with unexpected serious harm to the research participants. Each institution maintaining an IRC must provide written assurances to the NCST regarding the principles to be followed by the institution to protect the rights and welfare of the participants, the procedures to be followed by the IRC, and the composition of the IRC.7

The impetus for the establishment of these IRCs and governing procedures derives primarily from four sources: (1) a desire to function at a level of sophistication consonant with the ethical principles that have been enunciated and accepted by the international scientific community; (2) an inability of institutions sponsoring research to control researchers and their research at a local level; (3) widespread social and economic inequality within Uganda, which can potentially be exploited by unscrupulous investigators; and (4) an apparent distrust of an essentially centralized system, with little or no input from those to be affected by the research. Each of these conditions merits further consideration.

The HIV epidemic in Uganda has forced Uganda to consider the acceptable parameters of health research conducted within its borders. As of the end of 1997, Uganda had estimated that 930,000 of its adults and children currently alive were infected with HIV, including approximately 9 percent of its adult population. A total of 1,700,000 children had been orphaned through the deaths of their parents due to HIV; 1,800,000 adults and children had already died from HIV; and by June 1998, Uganda had reported a total of 1,900,000 cases of AIDS among its women and children.8 A significant amount of foreign-sponsored research monies have flowed into Uganda in an attempt to prevent further transmission of HIV and to identify acceptable and feasible mechanisms to provide health care to those already infected. The provision of such funds by Western sources often carries with it a demand for compliance with internationally accepted principles of bioethics.9 The adaptation of the international principles by Ugandans for Ugandans serves two purposes. First, it alleviates fears associated with adopting a foreign-mandated code.10 Second, establishing local, institutionally-based IRCs may help to alleviate verbalized fears of foreign manipulation and exploitation of the HIV epidemic within Uganda.11

Prior to the adoption of the Guidelines, institutions sponsoring research in Uganda had no authority to review, approve, suspend, or terminate research, even where there existed blatant and significant departures from a protocol approved by the NCST and/or blatant disregard of the rights of the study participants. Widespread economic and social inequality, reflected in high rates of poverty, illiteracy, unemployment, and lack of access to medical care12 rendered it extremely unlikely that study participants would either be aware of any abridgment of their rights or would be willing or able to protest in response. A protest could, on a practical level, result in the withdrawal of the only care to which they had access-that provided through the study at issue. Although the NCST had authority to suspend approval of research, it rarely had sufficient knowledge of a study's impact or conduct on a local level to make such a determination.

Similar concerns fueled the requirements for the IRC composition. NCST and AIDS Commission review generally lacked input from the study and target populations to be affected by the research. The imposition of research on a community was perceived by some Ugandans as a demonstration of foreign imperialism, reflective of British colonialism, encouraging control of the less sophisticated by the privileged.13 Uganda's history of tyranny and torture imposed during the successive regimes of Idi Amin and Milton Obote14 may also contribute to discomfort with scientific research in general and "foreign" research involving human subjects in particular. The incorporation of the international and Ugandan constitutional principles into the new Guidelines reflects Uganda's recognition of its own diversity,15 the past persecution of specific targeted groups by Ugandans,16 and the need and desire to prevent such occurrences from happening once again.

Subsequent to local IRC review, a protocol must be submitted for review and receive the approval of the NCST. 17 NCST review is consistent with pre-Guidelines practice, whereby the NCST was responsible for the review and registration of all research protocols involving research in Uganda, regardless of the discipline from which the research arose or the nature of the research question. Accordingly, the Guidelines do not extend the authority of the NCST to new disciplines or types of applications, but they do broaden the scope of the NCST review to include ethical concerns in addition to methodological issues and the ministerial registration (recording) of the research. Although the scope of the NCST review pursuant to the Guidelines is similar to the review provided by the local IRC, it will accomplish several functions not attainable through local level review. Ethical review of the proposed research at a national level will permit the consolidation of experts in a particular field to review the proposal and will also allow the NCST, equivalent in some respects to the National Institutes of Health in the United States, to approve those protocols that are consistent with and most likely to achieve the nation's health research objectives.

General Requirements for Informed Consent

Previous ethical review of research proposals required the provision of informed consent, but failed to enunciate the basic elements by which to judge its adequacy. The Guidelines mirror, to a significant degree, the provisions enunciated in then-existing U.S. regulations,18 in Guidelines 1 through 4 of CIOMS Human Subjects,19 in paragraphs 1, 10 through 13, 26, and 47 of CIOMS Epidemiology,20 and the Nuremberg Code.21 The Guidelines include, for example, a prohibition against exculpatory language and mandate a description of the risks and benefits of the research, and statements that research is to be conducted, that participation is voluntary, and that the participant may withdraw at any time without loss of any benefits to which he or she is otherwise entitled.22 Notably, however, the Guidelines depart from paragraph 5 of CIOMS Epidemiology by specifically prohibiting an investigator from relying on the permission of a community leader for the participation of community members in research. In all situations other than those specifically excepted, such as minor children unable to give consent, the investigator must obtain the individual's consent to participate in the research.23

The development and adoption of this requirement of individual consent essentially necessitated a re-examination of various aspects of Ugandan customary law. Unlike many Western cultures, Ugandan traditional practice demands the subordination of an individual's wishes, such as an adult son or wife, to those of a specified family leader, such as the father or husband.24 Subordination of one's wishes may further extend to the interests of the community or the tribe as enunciated by the recognized leader. Attendees at the NCC recognized that the extent to which individuals conform their behavior to what has been traditional practice varies across individuals and groups and may depend upon many factors, including the extent of an individual's exposure to Western conceptualizations of consent and liberty; individuals' level of education and literacy; the extent to which family, tribal, and religious ties have been maintained; the extent to which a family or group has maintained a more traditional approach to the resolution of conflict; and the extent of an individual's or group's economic dependency on others. The Guidelines' rejection of a leader's permission as an adequate basis for an individual's participation in research reflects Uganda's experience with tyranny, torture, and the elimination of targeted groups at the hands of its leaders.25 The Guidelines do, however, consider the needs of individuals who may continue to subscribe to Uganda's tradition through the provision to participants of "sufficient opportunity to consider whether or not to participate" in order to minimize the possibility of undue influence or coercion.26 This provision allows potential participants to confer with others as they deem necessary or appropriate, while still requiring that specific individual's consent.

These conclusions and provisions are consistent with a number of the U.S. National Bioethics Advisory Commission's (NBAC) draft findings and recommendations. The NBAC specifically found that "[I]n some cultures, permission of a woman's husband, if she is married, or her father, is she is unmarried, is a customary requirement before the woman may be enrolled in research."27 The NBAC further noted that, in some cultures, individuals are not free to refuse to participate in research once the community leader has granted permission to conduct the research.21 The NBAC recommended in the first instance that, where "the potential subject wishes to involve the spouse in the consent procedures, the researcher should take appropriate steps to that end. 1121 In the latter instance, the NBAC cautions researchers to "ensure that individual subjects agree to participate without coercion from community leaders or heads of households."30

The Guidelines also reject a requirement of written informed consent. This reflects an acknowledgment of the past persecution of individuals based on their association with particular groups or entities, and a recognition of the high rate of illiteracy in Uganda. Although as of the time of the NCC there had been no empirical studies examining the willingness of individuals to provide written informed consent to study participation, many of the attendees at the NCC observed that, in their experience, it was not unusual for potential research participants to express discomfort or fear at the prospect of having to sign an instrument that would establish a formal record of their association with the research. This discomfort was most evident among women who wished to participate without their fathers' or their husbands' knowledge or consent. The rejection of written consent reflects an awareness of and sensitivity to such circumstances.

Additional Protections for Vulnerable Participants The Guidelines set forth additional protections for six classes of persons: pregnant women, children, prisoners, mentally ill and behaviorally disordered persons, soldiers, and refugees. In general, these provisions are consistent with the Nuremberg Code, the various Helsinki Declarations,31 Guidelines 5, 6, and 7 of CIOMS Human Subjects,32 pertaining to research involving children, mentally ill and behaviorally disordered persons, and prisoners, and regulations adopted by the U.S. Department of Health and Human Services.33 Several departures from the U.S. provisions are noteworthy.

Women

Unlike previous U.S. regulations, which until quite recently restricted the ability of pregnant women to participate in clinical research,34 the Guidelines provide that a pregnant woman may participate in clinical research that presents minimal risk to the fetus, or in research presenting greater than minimal risk to the fetus if the purpose of the research is to meet the health needs of the mother.35 The fetus is to be placed at risk to the minimum extent necessary to meet those needs. This provision potentially permits the health needs of the mother to override the potential risks to the fetus. The requirement for the father's consent to the woman's participation, which is premised on a recognition of joint parental consent for the health of the fetus, is eliminated in situations where the clinical research is intended to meet the needs of the mother.36 This provision, which is in contravention to Ugandan tradition recognizing the authority of the husband and father over his wife and children,37 reflects an awareness of women's social vulnerability and their vulnerability to disease transmission as the result of specific cultural practices and mores such as polygamy, wife inheritance,38 and male infidelity.39

Children

Drawing from then-current policies and regulations in the United Kingdom, Australia, and the United States,40 the Guidelines distinguish between consent and assent in the context of children's participation in research. Assent, requiring a lower level of understanding, must be obtained from the child in all cases as a condition of his or her participation in the research, in addition to the permission of the parents) or guardian(s), where the parents) or guardians) can be identified and located and have not abandoned the child.41 The requirement of assent constitutes, again, a significant departure from Uganda's past practice, which traditionally recognized and accepted the complete authority of the male parent or family leader over his children.

Like the United States, Uganda provides for the participation in research of children who are wards. Unlike the United States, the Uganda Guidelines permit research involving such children to be conducted only where it is related to the children's status as wards.42 Additional significant differences exist between the two countries' provisions.

First, the United States provision is limited to children "who are wards of the state or any other agency, institution, or entity."43 In contrast, Uganda's provision encompasses, as well, children who "have no identifiable parent or guardian or have been abandoned by their parents) or guardian."44 As currently written, the Ugandan provision would permit a child to participate in research prior to the assumption of responsibility for the child by a guardian, institution, agency, or governmental entity. It is not clear, for instance, that United States regulations would permit the inclusion in research of children who have been abandoned by their parents, absent the establishment of legal status as a ward vis-a-vis the state or an institution. The Ugandan provision, however, notably fails to provide guidance as to which entities or individuals are responsible for deciding whether the child may participate prior to the establishment of the child's status as a ward, in view of the child's inability to provide consent.

Additional differences exist. For instance, United States regulations permit the participation of children who are wards in research involving greater than minimal risk with no prospect of benefit to the individual participants, as well as in research that would not otherwise be approvable but is expected to yield findings critical to the understanding of the disease or its prevention.45 However, additional protections are then required, including the appointment of an advocate for each child who is a ward, in order to provide a perspective in addition to that of any parent or guardian. That advocate is-required to act in the "best interests" of the child.46 In contrast, Uganda's Guidelines explicitly prohibit the participation of orphans and street children in research involving greater than minimal risk, regardless of the benefit that may be derived from such research.47 This prohibition stemmed from a concern for the growing number of children who have been orphaned or abandoned due to HIV and their potential vulnerability to exploitation.48

Prisoners

The Guidelines' additional safeguards for the protection of prisoners, which derive from the provisions of the Nuremberg Code49 and Guideline 7 of CIOMS Human Subjects,50 are modeled after those found in then-current U.S. regulations.51 The Guidelines specifically recognize that the constraints attendant to involuntary confinement may affect the voluntary nature of an agreement to participate in research. They further recognize that the issues surrounding confinement and consent are not limited to penal institutions and criminal rulings, but also extend to detention in other facilities, detention pending a disposition of the matter at issue, and detention in connection with civil rulings.52 Awareness of these various permutations flows directly from vivid memories of the detention and torture endured by many Ugandans under Amin and Obote and the lack of recourse available.

Accordingly, the Guidelines specify that (1) a majority of the IRC members must not be associated with the prisons) to be involved; (2) at least one member of the IRC must be a prisoner or an ex-prisoner for review of research protocols involving prisoners; (3) the risks of the research are commensurate with those that would be accepted by nonprisoners; (4) recruitment and selection procedures for participation are fair and not subject to intervention by prison authorities; (5) a prisoner's decision to participate or not to participate will not be considered in decisions regarding his or her release and he or she is advised of this prior to deciding whether or not to participate; and (6) any possible advantages of participation are not so great as to impair the prisoner's ability to weigh the risks against the benefits. In addition, research involving prisoners may not be approved unless it has "the intent and a reasonable probability of improving [the participant-- prisoner's] health and well-being" and appropriate experts have been consulted in developing the protocol.53

Mentally Ill and Behaviorally Disordered Persons

An awareness of the need for provisions for the protection of persons with mental illness arose from the observation of increasing numbers of individuals who appear to be suffering from HIV-related dementia. Pursuant to Ugandan law and tradition, these individuals would otherwise be deemed able and required to make decisions regarding participation, by virtue of their age and position in their family. These protections also encompass persons who are behaviorally disordered, due to the difficulties inherent in diagnosing what may be mental illness. Research involving mentally ill or behaviorally disordered persons is consequently prohibited absent (1) the informed consent of the prospective participant to the extent of his or her capabilities and (2) the permission of an incompetent individual's guardian, conservator, or other authorized individual.54 This second requirement, then, prohibits the participation of incompetent individuals who do not have a guardian or conservator. Additionally, the consent of a guardian, conservator, or other authorized person on behalf of the incompetent individual requires the presentation of evidence establishing their legal authority to make such a decision. Research involving mentally ill or behaviorally disordered individuals is prohibited if the research (1) can be carried out with individuals who are in possession of their full mental faculties, (2) is not relevant to the health needs of those with mental or behavioral disorders, (3) involves more than minimal risk, or (4) is potentially no more advantageous to the individual than currently existing interventions.55

Soldiers

The newly developed Guidelines provide additional protections for two classes of persons not encompassed by U.S. regulations as requiring additional protections: soldiers and refugees. The desire to provide these protections for soldiers stemmed, in part, from concerns for the potential for abuse by Ugandan leaders. These concerns reflect continuing fears imprinted by the long years of instability and tyranny of Amin and Obote. In part, as well, concerns for soldiers reflected Conference participants' knowledge of the involuntary participation of American soldiers in research conducted by the U.S. government, most notably the LSD experiments of the 1960s and the radiation experiments of the Cold War era.56

Accordingly, the Guidelines pertain to all military personnel, regardless of rank. The requirements for approval of the protocol are similar to those for research involving prisoners, described above. The IRC reviewing research involving soldiers must include at least one enlisted soldier and may not otherwise include individuals currently associated with soldiers or the military.57

Refugees

Unlike the United States regulations, Uganda's Guidelines specifically enumerate refugees as a class of individuals meriting special protection due to their social and political vulnerability. Uganda has had significant direct experience with refugee populations in recent years due to the large influx of individuals seeking refuge from political turmoil and genocide in Rwanda and what was Zaire.58 Research involving refugees may not be approved (1) unless the research question is answerable only with the participation of refugees, (2) the research is relevant to the health needs of refugees and will benefit refugees as a class, and (3) no more than minimal risk is involved. At least one member of the IRC must be a representative of a human rights organization that has as a primary focus the protection of refugees and refugee populations.59

Clinical Trials

Various provisions of the Guidelines relating to clinical trials are noteworthy. First, the NCC representatives concluded after significant discussion that placebo-controlled trials are to be permitted where (1) based on "knowledge that is available at the commencement of the trial, the new drug or device to be tested does not confer any significant benefit compared to the placebo," and (2) "the placebo arm is provided with the treatment or diagnosis product or device that is considered to be the normal standard of care in the community in which the trial is being conducted."60 We do not attempt here to either address or resolve the ongoing controversy related to the use of placebo controls in research designed to address the prevention of maternal-- child transmission of HIV.61 Rather, we focus on the circumstances and arguments considered by the NCC in reaching this determination regarding the general acceptability of placebo-controlled trials.

In reaching this conclusion, NCC representatives considered the Helsinki Declaration's provision which requires that "[fln any medical study, every patient-including those of a control, if any-should be assured of the best proven diagnostic and therapeutic method"62 and which recognizes that the "best therapeutic method" may include an inert placebo "where no proven diagnostic or therapeutic method exists."63 The representatives were also cognizant that (1) the Helsinki Declaration does not specifically address clinical trials but, instead, was intended to serve as professional guidelines for physicians as enunciated by physicians64 and (2) dose response designs, often touted as being "ethically superior" to placebo-controlled trials, may present similar ethical difficulties.65 The NCC representatives further noted that potential trial participants may feel coerced to "volunteer" due to Uganda's current economic situation, the relative lack of medical care, and the widespread perception that the quality of medical care received in conjunction with participation in a clinical trial is significantly higher than that which would otherwise be obtainable. This perception is not without a strong basis in reality. Uganda has a population of approximately 16 million, a gross national product per capita of $170, one physician for every 100,000 people, and an annual per capita expenditure on health of $6.66 Ultimately, the representatives concluded that the application of the "best therapeutic method" in the Ugandan context refers to the standard of care that prevails in Uganda with respect to the particular disease or condition under study. In some cases, this treatment might be no treatment.67 This interpretation was believed to minimize the potential for coercion to participate in a trial that might be occasioned by the provision of medical care that would be deemed to be extraordinary in the local context.

However, representatives were also aware of the concomitant potential for the exploitation of potential participants. Three provisions were developed in an attempt to minimize this possibility. First, investigators must provide adequate and safe dental or medical care, as appropriate to the condition under study, to the participants during the course of the trial. Appropriate medical care and followup procedures are to be maintained following the conclusion of the trial for a period of time that is dependent upon the nature of the disease under study, the trial, and the interventions.68 Second, the investigator must assure that, if the intervention is found to be successful, he or she will "make every effort" to ensure its provision to trial participants without charge following the conclusion of the trial and will make "a reasonable effort" to ensure its provision to the local community.69 NCC participants refrained from mandating the provision of the drug or device following the trial's conclusion in recognition of the CIOMS Human Subjects provision that "any product developed... be made reasonably available to inhabitants of the underdeveloped community in which the research was carried out. . . " the permissibility of exceptions to this general requirement,70 and unavailability of such a guarantee even to trial participants in developed nations.71 Third, prior to the approval of a protocol, the investigator must present evidence of insurance to cover any participant claims for injury, disability, or death that may arise in connection with participation in a clinical trial.72 This provision for compensation is consistent with Guideline 13 of CIOMS Human Subjects, but goes beyond current United States regulations, which require only that participants be provided with "an explanation as to whether any compensation and an explanation as to whether any medical treatments are available if injury occurs and, if so, what they consist of..."73

The prohibition against the participation of street children and orphans in research involving greater than minimal risk may appear, at first examination, to be inconsistent with the acknowledgment that medical care provided in the context of a clinical trial may be of higher quality than that which would ordinarily be available. As perhaps one of the most vulnerable segments of the population, it could be argued that orphans and street children should be permitted to participate in such trials, which could constitute their only means of access to any health care. The NCC representatives considered this position but ultimately rejected it, despite the existence of the three protective clauses described above, based on their knowledge of past historical events, such as the Willowbrook experiments in the United States,74 and a strong desire to protect such vulnerable children from the harm that could be occasioned by their participation in a study. It was noted that, absent an effective enforcement mechanism, the allowance of studies conducted with street children and orphans involving greater than minimal risk could leave these children open to harm without an adequate voice to make that harm known. In essence, the NCC representatives determined to leave the proverbial Pandora's box unopened.

Conclusion

These Guidelines were formulated following careful review and consideration of the foundational principles derived from the Nuremberg Code and Helsinki Declarations and the factual situations presently confronting Uganda. It is anticipated that an annual or bi-annual national review of these Guidelines will occur in order to re-evaluate their soundness in a continuously changing context and to further develop and elucidate the ethical principles to be applied. Various aspects of these newly created Guidelines have already been identified as being problematic and in need of further attention, including the inability to assure diversification of the IRCs across all socioeconomic statuses, religions, tribes, and professions due to Uganda's diversity; the potential absence of non-institution-based scientists on the IRCs; the noticeable absence of a requirement for lay participation on the IRCs; and the lack of a viable enforcement mechanism to ensure compliance with the Guidelines.

It is critical that this process be sufficiently comprehensive to include and address varying perspectives. This represents a considerable departure from the almost exclusive reliance on medical professionals in the past. It will require the establishment and nurturing of linkages across disciplines and within and across communities. This endeavor is rendered all the more difficult due to the limitations of the currently existing communications and transport infrastructures in Uganda.

The media can potentially play a critical role in supporting and moving the process forward. First, the media can bring to the public's attention research situations that may merit further scrutiny. Research participants are unlikely to complain to authorities on their own and, until an effective monitoring system is developed and implemented, the press may represent the only effective conduit of information between local communities and members of institutional review committees. Similarly, the press and the radio, in particular, can help both to inform the nonprofessional community members about their rights as research participants and to recruit nonprofessionals to participate in the review of research in their communities.

Clearly, the implementation and continual revision of the Guidelines will be an ongoing process requiring attention to multiple tasks. These include the education on a nationwide basis of all researchers and potential researchers regarding the Guidelines and the basis for their adoption; the formation and maintenance of institutional review committees; the training of individuals, at both the professional and community levels, to assume positions on the local and national review committees; and the education of Uganda's residents, including the media, regarding their rights as research participants.

Uganda's efforts to develop guidelines for bioethical review of health research continue to provide a model to those countries currently in the process of developing review procedures. These efforts and the resulting Guidelines also challenge developed countries to re-examine their own choices and values, particularly as they are interpreted and applied in diverse cultural and social settings.

Funding

This research was funded in part by the AIDS International Training and Research Program at Case Western Reserve University, sponsored by the Fogarty International Center.

Acknowledgments

We gratefully acknowledge the critical review by Drs. Jerrold Ellner, Karen Olness, and Christopher Whalen of earlier drafts of this manuscript.

Also, each of the authors has served as an investigator or principal investigator of health research studies in Uganda. At the time of the NCC, neither of the authors served as an investigator or staff on studies involving populations identified by the NCC as vulnerable or involving clinical trials. Case Western Reserve University has had a longstanding collaborative relationship with Makerere University, which has included, among others, the conduct of clinical trials related to tuberculosis prophylaxis, HIV vaccine preparedness trials, and HIV prevention trials. The involvement by the authors on such studies has been peripheral in nature, e.g., reviewing informed consent forms.