The Ottawa Statement on the Ethical Design and Conduct of Cluster Randomized Trials

The Ottawa Ethics of Cluster Trials Consensus Group sets out 15 recommendations for the ethical design and conduct of cluster randomized trials.

. A detailed description of the consensus process is provided elsewhere .

Ethical guidance for researchers and research ethics committees
The Ottawa statement sets out 15 recommendations for the ethical design and conduct of CRTs, including guidance on the justification of cluster randomization, the need for research ethics committee review, the identification of research participants, obtaining informed consent, the role of gatekeepers in protecting group interests, the assessment of benefits and harms, and the protection of vulnerable participants. The recommendations are listed in Table 1.
Here we summarize the major conclusions of the Ottawa statement.

Justifying the cluster randomized design
Compared to individually randomized trials with the same number of individuals, CRTs are statistically inefficient (Campbell et al., 2000). Owing to inherent features of their design, they are also more prone to bias (Hahn et al., 2005). Accordingly, researchers must justify the choice of cluster randomization (recommendation 1). Acceptable reasons include a study intervention that can only be administered at the group level, avoiding experimental contamination arising from intervention and control participants interacting with one another, enhancing participant compliance, or increasing administrative efficiency. Researchers should not adopt the design in a veiled attempt to avoid obtaining informed consent from research participants.

The need for research ethics committee review
There is broad agreement in national and international guidelines that research involving human participants must be prospectively reviewed by a research ethics committee. Research may be defined as a systematic investigation designed to produce generalizable knowledge. CRTs, including those evaluating quality improvement and knowledge translation interventions, fulfil the definition of research and, accordingly, must be reviewed by a research ethics committee (recommendation 2). Research ethics committees ought to undertake a proportionate approach to research review. CRTs that pose low risk and do not involve vulnerable participants may be eligible for an expedited or delegated review.

Identifying research participants
The clear identification of research participants is central to the implementation of research protections (McRae et al., 2011b). In individually randomized trials, the units of allocation, intervention, and outcome measurement are usually the same 3 Identifying research participants Researchers should clearly identify the research participants in CRTs. A research participant can be identified as an individual whose interests may be affected as a result of study interventions or data collection procedures, that is, an individual: (1) who is the intended recipient of an experimental (or control) intervention; or (2) who is the direct target of an experimental (or control) manipulation of his/ her environment; or (3) with whom an investigator interacts for the purpose of collecting data about that individual; or (4) about whom an investigator obtains identifiable private information for the purpose of collecting data about that individual.
Unless one or more of these criteria is met, an individual is not a research participant.

Obtaining informed consent
Researchers must obtain informed consent from human research participants in a CRT, unless a waiver of consent is granted by a REC under specific circumstances.

5
When participants' informed consent is required, but recruitment of participants is not possible before randomisation of clusters, researchers must seek participants' consent for trial enrollment as soon as possible after cluster randomisationthat is, as soon as the potential participant has been identified, but before the participant has undergone any study interventions or data collection procedures.

6
A REC may approve a waiver or alteration of consent requirements when (1) the research is not feasible without a waiver or alteration of consent, and (2) the study interventions and data collection procedures pose no more than minimal risk.

7
Researchers must obtain informed consent from professionals or other service providers who are research participants unless conditions for a waiver or alteration of consent are met.

Number Ethical Issue
Recommendation 8 Gatekeepers Gatekeepers should not provide proxy consent on behalf of individuals in their cluster. 9 When a CRT may substantially affect cluster or organisational interests, and a gatekeeper possesses the legitimate authority to make decisions on its behalf, the researcher should obtain the gatekeeper's permission to enrol the cluster or organisation in the trial. Such permission does not replace the need for the informed consent of research participants.
10 When CRT interventions may substantially affect cluster interests, researchers should seek to protect cluster interests through cluster consultation to inform study design, conduct, and reporting. Where relevant, gatekeepers can often facilitate such a consultation.

11
Assessing benefits and harms The researcher must ensure that the study intervention is adequately justified. The benefits and harms of the study intervention must be consistent with competent practice in the field of study relevant to the CRT.

12
Researchers must adequately justify the choice of the control condition. When the control arm is usual practice or no treatment, individuals in the control arm must not be deprived of effective care or programmes to which they would have access, were there no trial.

13
Researchers must ensure that data collection procedures are adequately justified.
The risks of data collection procedures must (1) be minimised consistent with sound design and (2) stand in reasonable relation to the knowledge to be gained.
14 Protecting vulnerable participants Clusters may contain some vulnerable participants. In these circumstances, researchers and RECs must consider whether additional protections are needed.

15
When individual informed consent is required, and there are individuals who may be less able to choose participation freely because of their position in a cluster or organisational hierarchy, RECs should pay special attention to recruitment, privacy, and consent procedures for those participants.
Source . (e.g. patients with a defined illness) and, as a result, identifying research participants is straightforward. In CRTs, these units may differ within a single study, complicating the identification of study participants. Research participants are those individuals who are most directly affected by the conduct of research, and researchers and research ethics committees have an obligation to identify them clearly and protect their interests. The Ottawa statement defines a research participant as any cluster member who is the recipient or the direct target of a study intervention (including the control condition), with whom researchers interact for study purposes, or about whom identifiable private information is collected (recommendation 3). In the case of CRTs involving a group-level intervention, all cluster members directly targeted by the intervention ought to be considered research participants. In the case of CRTs evaluating a knowledge translation intervention directed at health providers, the health providers are research participants; however, their patients ought to be considered research participants only if they are intervened upon, interacted with, or their identifiable private information is collected.

Obtaining informed consent
There is a general presumption that the informed consent of research participants must be obtained. In CRTs, informed consent procedures may be complicated by randomization of clusters before the identification of cluster members, clusterlevel interventions, and large cluster sizes (McRae et al., 2011a). The Ottawa statement requires researchers to obtain the informed consent of research participants, including health providers who are research participants, unless a research ethics committee grants a waiver of consent (recommendations 4 and 7). Researchers should seek the informed consent of research participants prior to randomization of clusters, where possible. When cluster members cannot be identified prior to randomization of clusters, their enrollment is permissible provided that informed consent is obtained as soon as possible and prior to any study interventions or data collection procedures (recommendation 5). Because of the challenges of seeking informed consent in CRTs, waivers of consent are especially important in this design. A research ethics committee may waive the requirement for informed consent when the study would not otherwise be feasible and the risks of study participation are no more than minimal (recommendation 6). Minimal risk refers to the risks of daily life, and includes the risks associated with routine physical examination and review of medical records. Feasibility will depend on a variety of factors, such as cluster-level intervention, cluster size, ease of contacting cluster members, complexity of the consent process, and research infrastructure. The burden of demonstrating that conditions for a waiver of consent obtain falls to the researcher.

The role of gatekeepers
Gatekeepers are individuals or bodies, such as hospital directors or municipal councils, who may be called upon to protect the interests of organizations or communities that are the setting for a CRT (Gallo et al., 2012). Historically, gatekeepers have fulfilled a variety of roles in CRTs, including providing informed consent on behalf of cluster members when it cannot otherwise be obtained. The Ottawa statement concludes that gatekeepers may not legitimately provide proxy consent on behalf of cluster members (recommendation 8). When cluster members are research participants, their informed consent is required unless a research ethics committee grants a waiver of consent. Gatekeepers may nonetheless play an important role in protecting organizational and community interests in CRTs. When a CRT has important implications for group interests, the permission of the gatekeeper to enroll the cluster ought to be obtained when the gatekeeper possesses the legitimate authority to provide such permission (recommendation 9). Gatekeepers may also help protect group interests by facilitating consultation between researchers and cluster members about the study (recommendation 10). Mechanisms for cluster consultation include open public fora, community advisory boards, and meetings with opinion leaders.

Assessing study benefits and harms
In individually randomized trials, the analysis of benefits and harms focuses on the consequences of the study for individuals. In CRTs, however, study interventions and data collection procedures may have implications for groups as well as individuals, and this complicates the assessment of benefits and harms (Binik et al., 2011). The Ottawa statement concludes that researchers must ensure that study interventions are consistent with competent practice in the relevant specialty of practice, e.g. medical practice, public health, or health policy. Random assignment to study interventions is justified when the relevant community of experts is uncertain as to the better performing practice (recommendation 11). The control condition must be adequately justified and, at a minimum, must provide usual care within the study context (recommendation 12). Finally, risks associated with data collection procedures must be minimized and deemed reasonable in relation to the importance of the knowledge to be gained (recommendation 13).

Protecting vulnerable research participants
Vulnerable research participants may be viewed as falling into four groups: children; incapable adults; people at undue risk of harm as a result of study participation; and people in subordinate positions within social or organizational structures.
Standard protections for these groups are outlined in national and international research regulations. Including vulnerable participants in CRTs poses the special challenge that their presence within cluster may be hidden, and researchers may fail to employ standard protections. In some cases, the study intervention may run the risk of exacerbating pre-existing inequalities within clusters (Conrad and Edwards, 2011). Accordingly, the Ottawa statement requires that special care must be taken to identify and protect vulnerable participants (recommendation 14). Additionally, CRTs are commonly conducted within healthcare organizations and research participants may include health professionals, service staff, or employees whose position within the organization makes them less able to express a free choice regarding study participation. In these circumstances, researchers must ensure that discussions about trial participation are carried out in a manner that limits the potential for coercive influence from employers (recommendation 15). The disclosure process should discuss career-related risks, including risks from detection of negligence and incompetence.

Next steps
The Ottawa statement on the ethical design and conduct of cluster randomized trials aims to provide guidance for researchers and research ethics committees on the design, conduct, and review of CRTs in health research. It supplements national and international ethics guidelines for randomized controlled trials and other human research, with special consideration for the unique characteristics of CRTs. Further work is required, however, to encourage use of the Ottawa statement and maximize its utility for researchers and research ethics committees. Research ethics committees operate within national regulatory frameworks, and countryspecific analyses of the use of the Ottawa statement within national laws and guidelines are required. Given the considerable variability of CRTs and the broad scope of their application, case studies highlighting ethical issues and their resolution within specific research domains are needed.