Adapting and Adaptive Research

Abstract

Research conducted during epidemics may warrant adaptations or adaptive designs owing to practical constraints, time pressures, uncertainty, the importance of flexibility, and the potential for research to detract from epidemic response. Adapting research entails choosing different research designs or methods if research goals, contexts or constraints justify or require a different approach. Adaptive research, by contrast, is a type of research that prospectively plans for modifications after research has been initiated, while maintaining the validity and integrity of the research. While adaptation and adaptive designs introduce an important degree of flexibility to research conducted during epidemics and help to address research objectives and constraints, adaptation and adaptive designs require close ethical scrutiny and are no different from other research in that they are expected to align with universally accepted ethical standards. Important ethical questions exist regarding the conditions that justify adaptations to research, the kinds of adaptive research designs that can be ethically justified, and how ethics review bodies ought to evaluate such novel approaches to research in epidemic contexts. The five cases included in this chapter prompt reflection on the ethical considerations and implications of adapting research in response to epidemic-related risks and the public health measures deployed in response to those risks, as well as the ethical implications of not adapting research in such contexts. These cases also highlight ethical questions and issues arising during the conduct of adaptive trials, including when treatments under study, treatment doses, sample size, and other study features are reviewed in response to evolving evidence. This chapter invites reflection on these key ethical dimensions when considering adaptive designs and adaptations to standard research procedures during epidemics. What these cases make clear is that adaptive designs and adaptations to research do not reduce the need for rigorous scientific evaluation and adherence to universal ethical standards, and must be explicitly ethically justified and reviewed through transparent and inclusive processes.

5.1 Introduction

Conducting research during epidemics is of critical importance. Yet, owing to practical constraints, time pressures, uncertainty, the importance of flexibility, and the potential for research to detract from epidemic response, the ways in which research is conducted in this context may warrant adaptations or adaptive designs.

Adapting research entails choosing different research designs or methods if research goals, contexts or constraints warrant or necessitate a different approach. For example, as Case 5.1 illustrates, researchers who initially planned to conduct research involving face-to-face interaction to generate data may have to instead adopt remote forms of data generation, owing to epidemic-related risks and measures that restrict mobility. And as Cases 5.2 and 5.3 highlight, failing to adapt and instead pausing research, given the challenges presented by an epidemic, may have negative impacts on research participants.

Adaptive research, by contrast, is a type of research that prospectively plans for modifications after research has been initiated, while maintaining the validity and integrity of the research (Mahajan and Gupta 2010). For instance, research may be designed with a plan to revisit the treatments under study, the treatment doses, the sample size, and so forth (Pallmann et al. 2018). Case 5.4 provides a nice example of adaptive research via the RECOVERY Trial, wherein study arms were added when there was reason to believe an intervention offered a benefit or removed when sufficient data had been collected to establish that an intervention was associated with a lack of benefit. And Case 5.5 highlights the possible implications of not using an adaptive research design in the context of evolving scientific evidence.

While adaptation and adaptive designs introduce an important degree of flexibility to research conducted during epidemics and help to address research objectives and constraints, this does not mean that “anything goes”. Scientific rigour and validity, in addition to adherence to universally accepted ethical standards, remain essential. Consequently, important ethical questions exist regarding the conditions that justify adaptations to research, the kinds of adaptive research designs that can be ethically justified, and how ethics review bodies ought to evaluate such novel approaches to research in epidemic contexts.

5.2 Adapting – Not Deviating From – Scientific and Ethical Standards for Research

If epidemic contexts sometimes warrant or necessitate that research be adapted or adaptive, does this mean that exceptions should be made to the scientific and ethical standards that otherwise govern research? Such standards include both scientific standards, such as those commonly used for participant selection, sample size estimation and sample size allocation, as well as ethical standards, such as those governing the ethics review and informed consent processes (see Chap. 6).

London and Kimmelman argue that the challenges that rigorous scientific methods are designed to address do not disappear during public health emergencies like epidemics, nor do researchers’ obligations to align the conduct of their research with the public interest or to protect the interests of research participants, both of which are advanced by research ethics standards and regulations (London and Kimmelman 2020). In other words, they argue that “the moral mission of research remains the same: to reduce uncertainty and enable caregivers, health systems, and policy-makers to better address individual and public health” (p. 476). Consequently, while accepted ethical and scientific standards should be interpreted in light of, and adapted in response to, particular circumstances and contexts in epidemics, the aim must still be to generate the best possible evidence about important questions. Adaptive research designs and adaptations to research therefore do not sidestep the need for rigorous scientific evaluation and adherence to universal ethical standards and must be explicitly ethically justified and reviewed through transparent and inclusive processes.

The evidence base can evolve rapidly during an epidemic. Researchers and those charged with reviewing ongoing studies (e.g. research ethics committees) therefore have a responsibility to monitor emerging evidence from other research initiatives, review the implications for the studies they are leading or overseeing and decide whether those studies should be continued, modified, suspended or cancelled, in order that they continue only if they have scientific and social value and so that people are not asked to participate in research that is no longer likely to produce meaningful results or which poses risks without the prospect of benefit (PAHO 2020). In other words, the justification and ethical acceptability of research can vary throughout its duration as a result of rapidly evolving evidence. Decision-making in this context can be particularly challenging as evidence may be uncertain or contested. Consequently, the intervals at which studies are reviewed and report to research oversight bodies ought to be shorter and more frequent during an epidemic. Researchers should also develop plans that account for how their study might be affected by new evidence or adapted in response to it.

The Nuffield Council on Bioethics (Nuffield Council on Bioethics 2020, pxxii) raises two key questions that they suggest ought to be asked when considering adaptations to standard research procedures during public health emergencies, like epidemics:

Is this the right study for this location and this population/subpopulation? Who has been involved in identifying and characterising the problem that the research seeks to answer? Will local populations benefit from any positive findings?

Is this the right design for this location and this population? How have local needs, concerns or preferences been taken into account?

Following on from these questions, the Council offers two recommendations:

Study protocols should be developed with the input of local communities before being finalised, in order to ensure that proposed procedures are acceptable to communities, as well as meeting ethical requirements. Even in multi-site trials, there will be elements that can and should be operationalised differently in different sites in response to engagement and feedback.

Any exclusion criteria from studies should be clearly justified with reference to the risks and benefits for the group in question, in this context, rather than an automatic exclusion of ‘vulnerable groups’.

5.3 Adaptive Clinical Trials

Adaptive clinical trials (ACTs) are a particularly salient type of adaptive research that may be considered during epidemics. For instance, the magnitude and high case fatality rate of the 2014–2016 Ebola virus disease (EVD) epidemic in West Africa prompted calls for the accelerated evaluation and development of investigational therapeutic interventions that had shown promising results in the laboratory and in animal models. In response, a World Health Organization ethics advisory panel concluded that it was ethical to offer investigational agents with the intent to treat those suffering from EVD, and that a moral duty existed to evaluate these interventions in the best possible clinical studies (WHO 2014); however, it was unclear what the ethical requirements were for the appropriate design of such investigations. Proponents of placebo-controlled randomized controlled trials (RCTs), for instance, argued that these designs ought to be used as they were best able to generate robust, statistically valid evidence about safety and efficacy, which could be used to ensure all patients receiving treatment after the trial received the safest and most effective intervention (Joffe 2014). On the other hand, proponents of ACTs argued that ACTs would be preferable as they better allow for emerging, accumulated data to be used to rapidly identify and deploy beneficial new therapies to improve outcomes among trial participants (Adebamowo et al. 2014).

The principal argument favouring the conduct of placebo-controlled RCTs in the context of epidemics and other public health emergencies is that one ought to collect the best possible evidence in order to develop the safest and most effective intervention, and that a placebo-controlled RCT is the most appropriate, and perhaps morally obligatory, method of achieving this goal. The principal argument favouring the conduct of ACTs in the context of epidemics and other public health emergencies is that, owing to the severity and urgency present during epidemics, in addition to the higher fatality rates associated with conventional, supportive care in the absence of effective therapies, one should give greater weight to the well-being of the patients affected and therefore favour ACTs, given their ability to adapt to emerging evidence of treatment safety and efficacy (Singh 2023). Table 5.1 outlines the key relative merits of RCTs and ACTs, as well as the ethical considerations regarding each, in order to elucidate the potential value, as well as the potential pitfalls, of conducting ACTs during an epidemic.

Table 5.1 Relative merits of RCTs and ACTs and ethical considerations regarding each type of trial

5.4 Adapting Research to Epidemic Contexts

The ethical appropriateness of any research design should to some extent be informed by the context in which the research is to be conducted (Pullman and Wang 2001). That is, it has been argued that methodological orthodoxy ought to be eschewed in order to critically consider the research context, background information, risks of the research and the most appropriate means of answering specific research questions and achieving stated goals (Pullman and Wang 2001; Cartwright 2007; Ezeome and Simon 2010). Appreciating the motivations for and principal objectives of conducting clinical trials in the context of an epidemic may, at least in part, be instructive of which trial design ought to be favoured, and whether adaptations are ethically justifiable (if not ethically obligatory). For example, while not necessarily mutually exclusive, there were at least two central objectives that were advanced in relation to conducting trials in the midst of the EVD epidemic: (1) to aid the current humanitarian response and to make potential therapies rapidly available in order to save as many lives as soon as possible; and (2) to generate the most robust, scientifically valid data that would lead to the development of a licensed product that could, in turn, be used to ensure the safest and most efficacious intervention was available for patients receiving treatment following the conclusion of the trial. Preference for either trial design in the context of an epidemic may therefore be dependent, at least in part, on which objective is considered the priority.

A World Health Organization ethics advisory panel argued that investigational therapeutic options should not divert resources or attention from the public health measures, which they claim ought to remain the priority in an epidemic response (WHO 2014). In 2022, WHO argued that “emergency use of interventions for which there is insufficient evidence of safety or efficacy for regular use in health systems is ethically permissible outside clinical trials or other research contexts, if the primary aim is clinical benefit for individual people or groups or benefit for populations, and if such use during public health emergencies complies with a sound ethical framework that ensures adequate justification, ethical and regulatory oversight, consent process and contribution to evidence” (WHO 2022). Others have warned that research conducted during an epidemic or other public health emergency could have the effect of encouraging the modification of public health priorities, perhaps from providing a humanitarian response to the rigorous collection of data (Ezeome and Simon 2010). As such, if substantial resources are to be invested to conduct a trial during an epidemic, then there is a strong argument to be made that a moral responsibility exists to use those resources in such a way that they benefit those affected by the epidemic and curb the further spread of the epidemic. While any trial design has the potential to direct attention away from the immediate epidemic response, it appears that ACTs may be more congruent with the immediate epidemic response, although, placebo-controlled RCTs could be designed in a way that makes them align better with the advantages of ACTs. This could be accomplished, for example, by utilizing stepped-wedge RCTs, which involve random and sequential crossover of clusters of participants from a control arm (or arms) to the experimental arm (or arms) until all clusters have been exposed to the experimental intervention (Hemming et al. 2014). Or, placebo-controlled RCTs could utilize data safety and monitoring boards, who are charged with reviewing interim data and implementing early stopping rules based on safety and/or efficacy thresholds.

It is important to acknowledge that, for any research conducted in the context of an epidemic, the ability of participants to provide informed consent may become compromised or the consent process may become less feasible (see Chap. 9). This may be due to participants’ lack of mental or physical capacity in such dire circumstances, a lack of local health-care workers available to recruit participants, and/or a strong therapeutic misconception undermining participants’ abilities to appreciate the clinician’s dual role as researcher and health-care worker (Pullman and Wang 2001; Ezeome and Simon 2010; Adebamowo et al. 2014; Kass 2014; Tangwa 2014). As such, some argue that every effort must be made to provide the most effective treatment to every trial participant, given current information, and that ACTs attempt to accomplish this very task while still ensuring that research objectives can be pursued (Pullman and Wang 2001). The dire circumstances and the prospect of inevitable therapeutic misconception during the EVD epidemic led some to argue that entering West Africa with the aim of doing anything other than saving the lives of those affected by EVD and curbing the spread of the epidemic would be morally irresponsible (Tangwa 2014). This sentiment, if it is to be balanced with the motivation and need to collect crucial evidence about the safety and efficacy of investigational therapies, may be supportive of adopting an adaptive design.

5.5 Conclusion

Epidemics should prompt researchers to evaluate whether their research ought to be adapted or whether adaptive designs might be appropriate and ethically justified. What is clear is that adaptive designs and adaptations to research do not obviate the need for rigorous scientific evaluation and adherence to universal ethical standards, and must be explicitly ethically justified and reviewed through transparent and inclusive processes. Involving the voices of local, affected communities in research planning, design and oversight remains crucial. Engaging local communities in such aspects of the research may foster trust in the research and epidemic response and better ensure local values and customs are both respected and represented (Modlin et al. 2023). Consequently, the input of those affected by an epidemic and who may be impacted by any research conducted ought to be considered of the utmost importance in responding to the question of whether and how research might be adapted.

Appendices

Case 5.1: Adapting Face-to-Face Interviews to Respect Infection Control Measures

This case study was written by members of the case study author group.

Keywords

Research design and adaptation; Privacy and confidentiality; Data protection, access and sharing; Vulnerability and inclusion; Qualitative research; Digital and remote healthcare and research; Researcher safety

In April 2020, a non-governmental research group decided to conduct a prospective study in a Caribbean country to inform and orient governmental strategies for controlling the spread of COVID-19. The initial study design incorporated a population-based survey that would seek to recruit participants with a variety of demographic characteristics. The research team would evaluate their knowledge of COVID-19 and try to understand the reasons behind their non-compliance with the main infection-control measures proposed by the health authorities, including wearing masks and practising hand hygiene and physical distancing.

The researchers planned to undertake face-to-face interviews with study participants after obtaining informed consent. This strategy involved visiting crowded places like markets and talking to several people for about 30–45 min each. To meet the study objectives, it would be necessary to interview people who did not follow the infection-control measures. However, this mode of data collection might also prevent researchers from complying with physical distancing requirements themselves, raising concerns about the safety of both the researchers and the participants. The study schedule might also be affected by lockdown periods, and if it was, it would be unclear when research could restart.

To address the research team’s concerns about exposing researchers and participants to infection, the principal investigator proposed to revise the protocol and conduct an online survey instead, using a structured four-page questionnaire. Before going through the questionnaire, participants would be shown a page presenting the research team, the study objectives, and the main ethical obligations, including data confidentiality. At the end of this page, the respondent would be invited to complete the questionnaire. Only two team members, a data officer responsible for the quality control and data analysis and the principal investigator, would access the data collected, which would be anonymized. The research team planned to advertise the questionnaire to the study population via organized groups: church groups, scientific communities and neighbourhood committees.

However, the choice of an online survey, while reducing risks to participants and researchers, has some limitations. For example, the study sample would not be representative of the population, because some people did not have access to the internet or did not have enough knowledge of online platforms to take part in an internet-based study. Translating the questionnaire into a local language at the request of the research ethics committee would make it accessible to members of the study population who did not read an official European language. However, the exclusion of the part of the population who could not read at all would still stand. The face-to-face survey would have given the research team the opportunity to include this population by reading the questionnaire to them. Additionally, an online survey would provide less opportunity to investigate the causes of non-compliance in depth, with implications for the interpretation of results. During the administration of the pilot questionnaire, concerns also arose about the reluctance of the respondents to reveal their economic status in an online survey.

Questions

• 1. Should national research ethics committees provide guidance about ethical issues and considerations arising during research in pandemics in order to help researchers develop appropriate study designs? Why? If such guidance is produced, what issues should it address?

• 2. In a pandemic, how should researchers and public health officials address tensions between the need to conduct relevant research rapidly and the need to respect participants’ safety and integrity?

• 3. The COVID-19 pandemic has exacerbated existing social disparities and created new vulnerabilities. How should these be addressed during the design and implementation of a research project during the pandemic?

Case 5.2: A Community-Based Intervention for Indigenous Older Persons with Mild to Moderate Dementia

This case study was written by members of the case study author group.

Keywords

Research design and adaptation; Risk/benefit analysis; Pausing and halting research; Vulnerability and inclusion; Community engagement and participatory processes; Resource allocation; Non COVID-19 research

Cognitive stimulation therapy (CST) has been shown to improve cognition, mood and quality of life in adults with mild to moderate dementia. Delivered in twice-weekly group sessions over a 7-week period, it is considered safe and is used internationally, including in some regions of a country in the Western Pacific.

Within this country, gerontology researchers partnered with an indigenous community in a relatively remote area, where CST was not available. Working together over time to develop trust and mutual understanding, the researchers and the community adapted the programme for culturally appropriate delivery, by using the indigenous language. The researchers and community commenced a trial to determine its effectiveness in this population, with several rounds of participant recruitment planned. The indigenous community was instrumental in recruiting participants and facilitating the trial.

The first round of the intervention study, involving ten participants, was completed before COVID-19 emerged. It demonstrated significant improvements over baseline measurements for both cognitive function and mood and was well received by the participants and their wider family networks. The second group of participants had been recruited, and baseline measurements completed, when the country was placed under long lockdown restrictions. In line with public health directives, the trial was put on hold.

After several months, lockdown restrictions were eased, and some normal activities were able to resume. The dementia of some recruited participants had progressed from moderate to severe in the intervening period, so they were no longer eligible to participate in the study. Baseline measurements of dementia, cognition, mood and quality of life would have to be completed again for the remaining participants, and new potential participants identified. Researchers and their indigenous partners considered whether to proceed with the recruited participants and decided against doing so. Considerations included the burden of running the programme upon local community workers, who were stretched by COVID-19-response work, and concern that the pandemic increased the risks of participating and that this would cause anxiety among participants and their families. A decision was made to re-attempt recruitment at a later date when the outlook regarding COVID-19 rates in the community and the responses of public health were more certain.

During the months following that decision, localized lockdowns occurred outside the study area, highlighting the on-going risk of transmission of the SARS-CoV-2 virus. Community workers who would be involved in recruitment and in delivering the intervention were managing increased demands for their services. The researchers had time-limited funding to develop relationships with indigenous health providers in five more sites, with a view to rolling out the adapted programme more widely. However, those efforts depended upon the trial being completed.

Once COVID-19 was well contained nationally, there were no government restrictions preventing the trial, and there was no evidence of the virus circulating in the community. However, in the absence of nationally licensed COVID-19 vaccines and curative treatments, researchers remained reluctant to recommence recruitment, owing to the on-going risk of a possible COVID-19 outbreak, with attendant risks to older people. They, and their community partners, were unclear about how to determine when recommencement of the trial would be justified.

Questions

• 1. What ethical considerations and other factors should determine when, and whether, to recommence this trial?

• 2. Should a higher level of caution regarding the risks of restarting CST apply if it’s being provided as part of a research study rather than as routine care? Why? Are characteristics of the participant group (such as a dementia diagnosis) relevant to this question?

• 3. How should the potential benefit of participation in this trial figure in reasoning about when recommencement is justified?

• 4. Should funding considerations influence decisions about whether and when to recommence this study? Why?

Case 5.3: Suspending Participation in Research

This case study was written by members of the case study author group.

Keywords

Research design and adaptation; Safety and participant protection; Risk/benefit analysis; Researcher roles and responsibilities; Access to experimental treatments; Vulnerability and inclusion; Digital and remote healthcare and research; Non COVID-19 research

During the COVID-19 pandemic the population of a Latin American country was vulnerable, owing to a very high prevalence of cardio-metabolic illness. Such illnesses are a major risk factor for a bad prognosis following infection with the SARS-CoV-2 virus because of their effect on the immune system and the chronic inflammatory state they promote (Carter et al. 2020; Jayawardena et al. 2020).

Research centres in the country are, in general, responsible for the health of their research participants. Some research centres also provide clinical care, and thus have responsibilities to their patients as well. It is important for them to uphold institutional directives and guidelines for infection control in order to minimize risks to patients and research participants. In the pandemic, novel ways of conducting research have had to be developed to decrease potential exposure to COVID-19. Some studies introduced home delivery systems for medicines and equipment so that participants were not unnecessarily exposed to infection at a research centre or clinic. However, the storage and handling of research products is a delicate process, which must follow adequate procedures and be appropriately documented.

During the pandemic a Phase III pharmacological study of moderate rheumatoid arthritis was conducted. To reduce the risk of infection, scheduled visits to the research centre by participants who reported that they had been infected with COVID-19 were replaced by telephone calls. Participants who did not report COVID-19 infection continued with routine face-to-face visits. A participant in the study experienced a 20% improvement in swelling and pain in his joints (when compared to baseline measurements at his first visit). At approximately week 84 of the study the participant was diagnosed with COVID-19. At this point his participation in the study was halted, as the medication being trialled had immunosuppressive activity.

In the context of evolving scientific understanding of COVID-19, the researchers needed to consider preventive measures that might need to be implemented to avoid furthering risks to participants. In this case questions arose about whether participants in a study of a treatment with immunosuppressive effects should resume research participation after their COVID-19 infection had resolved, and if so, when. The potential effects of the medication being trialled on both the severity of the rheumatoid arthritis, and the length and severity of COVID-19 symptoms were considered.

Questions

• 1. What ethical considerations should be taken into account when making decisions about the potential suspension and recommencement of research participation following a participant’s COVID-19 diagnosis? What role, if any, should participants’ views and preferences have in such decisions?

• 2. How should research centres prioritize research addressing chronic health needs in a pandemic? What specific considerations may arise when the proposed research involves a medication which has the potential to address chronic health problems but has immunosuppressive activity?

• 3. What responsibilities should researchers have to participants who are unable to continue receiving study interventions following a diagnosis of COVID-19?

• 4. What ethical issues should be considered when deciding if research studies should implement home delivery systems for research equipment and medication to minimize the number of site visits needed during a pandemic?

References

• Carter, S.J., M.N. Baranauskas, and A.D. Fly. 2020. Considerations for obesity, vitamin D, and physical activity amid the COVID-19 pandemic. Obesity 28: 1176–1177. https://doi.org/10.1002/oby.22838.

• Jayawardena, R., P. Sooriyaarachchi, M. Chourdakis, C. Jeewandara, and P. Ranasinghe. 2020. Enhancing immunity in viral infections, with special emphasis on COVID-19: A review. Diabetes & Metabolic Syndrome: Clinical Research & Reviews 14: 367–382. https://doi.org/10.1016/j.dsx.2020.04.015.

Case 5.4: Ethics and Adaptive Trials in the COVID-19 Pandemic

This case study was written by members of the case study author group.

Keywords

Research design and adaptation; Vulnerability and inclusion; Consent; Ethical review; Researcher roles and responsibilities; Research priority setting; Risk/benefit analysis; Research publication ethics; Treatment repurposing; Multi-centre research; Pre-prints

A broad range of potential treatments have been proposed for COVID-19. Conducting timely and rigorous research to evaluate the effectiveness and safety of possible treatments is key to informing effective public health responses to COVID-19. The Randomised Evaluation of COVID-19 Therapy (RECOVERY) study is an adaptive randomized controlled trial, initially established in the UK to evaluate which treatments may be more effective than the usual standard of care patients receive when admitted to National Health Service (NHS) hospitals with COVID-19 (University of Oxford 2021a).

Adaptive trials are so called because the protocol pre-specifies that certain elements of the trial may be adapted during the trial. In the RECOVERY trial, the key adaptive element is that new study arms are added when there is sufficient reason to believe an intervention may offer a benefit but when there is also uncertainty. Study arms are closed once sufficient data have been collected to establish whether the intervention is associated with a benefit, or lack of benefit or even harm, for participants. Large-scale adaptive trials can be powerful and give results to inform policy more efficiently than traditional clinical research designs (Pallmann et al. 2018). During the first 11 months of RECOVERY, for example, over 38,000 participants were recruited at over 170 sites. By February 2021, the study had published preliminary or complete findings about the benefit or lack of benefit associated with dexamethasone, tocilizumab, hydroxychloroquine, lopinavir-ritonavir, convalescent plasma and azithromycin in the treatment of hospitalized COVID-19 patients (University of Oxford 2021b). On-going study arms were assessing the benefits of monoclonal antibodies, aspirin, colchicine, baricitinib and dimethyl fumarate.

Review

Adaptive trials can be challenging to review for ethics committees and regulatory bodies, owing to their complexity and on-going evolution; by January 2021, RECOVERY was on version 13 of the protocol. During public health emergencies, such challenges can be exacerbated by the need for effective multi-site review within an expedited timeframe. In the UK, the RECOVERY protocol and regular protocol amendments have received expedited review by a single ethics committee, which provides national approval for the research to be conducted at all participating NHS sites. National regulatory review of RECOVERY is undertaken using routine review processes within an accelerated timeframe. To support these expedited regulatory and ethical review processes, RECOVERY’s principal investigators liaise with reviewers to provide updates about proposed protocol amendments and the timelines within which they will be submitted.

Public Health Responses

Within public health emergencies, the need to conduct research to address public health priorities must be evaluated in conjunction with the importance of ensuring that such research does not adversely impact pandemic response efforts (Nuffield Council on Bioethics 2020; WHO 2020). A key priority in the design and implementation of RECOVERY has been to minimize the impact of such a large-scale study on the provision of clinical care within NHS hospitals. Practical measures to achieve this include the development of a short and simple online case report form for data collection, substantial resources to support site staff, and the use of linkage to routine health data to collect information on patient outcomes (University of Oxford 2021c).

Inclusion

Current standards in research ethics highlight the need for fair and inclusive approaches to the selection of research participants, recognizing the importance of generating relevant evidence to inform approaches to addressing their health needs, as well as ensuring appropriate protections are in place (CIOMS 2016). Within the context of public health emergencies, the importance of inclusive approaches has been highlighted (Nuffield Council on Bioethics 2020), including, for example, calls to ensure that pregnant women are appropriately included in COVID-19 treatment trials (Taylor et al. 2021). All patients admitted to NHS hospitals in the UK with suspected or confirmed COVID-19 are potentially eligible to participate in RECOVERY. Potential participants will be excluded from the trial if their attending clinician believes that their medical history puts them at significant risk if they participate, or if one of the active treatment arms in the trial is considered specifically indicated or contra-indicated for the patient. Age-related exclusion criteria are associated with some treatment arms, and in trial sites outside the UK, but across the study as a whole, participant ages range from less than 1 year old to 103. Pregnant and breastfeeding women are excluded from treatment arms incorporating contra-indicated interventions but are eligible for other arms (https://www.recoverytrial.net/for-site-staff/site-teams).

Consent

Seeking consent to adaptive trials raises a number of practical ethical issues associated with the complexity of the research, on-going amendments to study arms, and evolving evidence about the risks and benefits of specific interventions (Global Forum on Bioethics in Research 2017). Within the RECOVERY trial, recruitment processes additionally need to be responsive to potential limitations on participants’ capacity to consent to research, including in children, very elderly patients and patients with severe disease. Participant information sheets for adults, children, parents and guardians have been designed to be clear and concise – no more than three pages long. Information for patients is also available on the RECOVERY website, including videos describing the interventions being evaluated (https://www.recoverytrial.net/study-faq). Resources for trial sites include training and standard operating procedures for the recruitment of competent and incompetent patients (https://www.recoverytrial.net/for-site-staff/training/background-and-informed-consent).

Selecting Study Interventions

During the pandemic, a wide range of therapies have been proposed to treat COVID-19, with varying quality of supporting evidence. Given the scale and pace of global COVID-19 research, the evidence base about specific treatments has frequently evolved rapidly. Within this complex context, RECOVERY investigators make decisions about which treatments to trial, and review recommendations from the study’s independent data-monitoring committee about appropriate points to halt recruitment into specific trial arms (https://www.recoverytrial.net/for-site-staff/site-set-up-1/data-monitoring-committee-correspondence). Choices of which treatments to trial are taken by RECOVERY’s principal investigators and the UK Chief Medical Officer, and have been informed by WHO priorities, reviews from the UK’s New and Emerging Respiratory Virus Threats Advisory Group and the UK COVID-19 Therapeutics Advisory Panel. Key considerations include existing evidence about the safety and efficacy of the intervention, whether it is available in sufficient quantities to evaluate in the trial, and whether, if shown to be successful, treatment could be rapidly scaled up (Wise and Coombes 2020).

Reporting Research Results

The pre-publication and publication of research results can have a rapid, substantial and multinational effect on research and public health responses in an epidemic (Hofmann 2020). Adaptive trials seek to produce findings more rapidly than traditional clinical trials, and when there is sufficient evidence to justify halting recruitment into a specific treatment arm, questions arise about when and how preliminary findings should be disseminated. Preliminary findings from RECOVERY are typically reported in press releases (https://www.recoverytrial.net/news/) and simultaneous pre-prints, with study results subsequently being published in peer-reviewed journals (https://www.recoverytrial.net/results/).

Questions

• 1. What specific ethical issues arise when conducting consent processes for an adaptive trial in a pandemic, and how should they be addressed?

• 2. Whose perspectives should inform decisions about which potential treatments should be assessed in the RECOVERY trial, and what ethical considerations should inform that decision-making?

• 3. How should researchers and health-care providers respond ethically to press releases about preliminary research results during a pandemic? Why?

• 4. What ethical arguments could be made for and against conducting expedited national ethical reviews of RECOVERY protocols and amendments (rather than review at each trial site)?

References

• CIOMS. 2016. International ethical guidelines for health-related research involving humans. Geneva: Council for International Organizations of Medical Sciences.

• Global Forum on Bioethics in Research. 2017. Ethics of alternative clinical trial designs and methods in LMIC research. https://www.gfbr.global/wp-content/uploads/2018/12/GFBR-2017-meeting-report-FINAL.pdf.

• Hofmann, B. 2020. The first casualty of an epidemic is evidence. Journal of Evaluation in Clinical Practice 26(5): 1344–1346.

• Nuffield Council on Bioethics. 2020. Research in global health emergencies. London: Nuffield Council on Bioethics.

• Pallmann, P., et al. 2018. Adaptive designs in clinical trials: Why use them, and how to run and report them. BMC Medicine 16(1): 29.

• Taylor, M.M., et al. 2021. Inclusion of pregnant women in COVID-19 treatment trials: A review and global call to action. The Lancet Global Health 9(3): e366–e371.

• University of Oxford. 2021a. RECOVERY: Randomised evaluation of COVID-19 therapy. https://www.recoverytrial.net/.

• University of Oxford. 2021b. RECOVERY: Randomised evaluation of COVID-19 therapy. https://www.recoverytrial.net/results.

• University of Oxford. 2021c. RECOVERY: Randomised evaluation of COVID-19 therapy. https://www.recoverytrial.net/for-site-staff.

• WHO. 2020. Ethical standards for research during public health emergencies: Distilling existing guidance to support COVID-19 R&D. Geneva: World Health Organization.

• Wise, J., and R. Coombes. 2020. Covid-19: The inside story of the RECOVERY trial. British Medical Journal 370: m2670.

Case 5.5: The Impact of New Scientific Evidence on On-going COVID-19 Studies

This case study was written by members of the case study author group.

Keywords

Research design and adaptation; Pausing and halting research; Ethics committee remits and responsibilities; Ethical review; Researcher roles and responsibilities; Social and scientific value; Treatment repurposing

A national research ethics committee (NREC) reviewed and approved three local research initiatives testing the efficacy of hydroxychloroquine for COVID-19: one for patients with mild to moderate COVID-19 (Trial A), another for patients with critical and severe COVID-19 (Trial B), and a third one that tested hydroxychloroquine as prophylaxis (Trial C).

After the approval of these trials, the preliminary results of a randomized, controlled, open-label adaptive trial demonstrated that there was no clinical benefit from administering hydroxychloroquine to hospitalized people with COVID-19 (University of Oxford 2021) (see also Case 5.4). The enrolment of participants to this arm of the trial was subsequently stopped. After the release of these data, another global clinical trial also suspended its hydroxychloroquine arm in order to analyse its interim results, and finally cancelled the arm because continuation was considered futile (WHO n.d.).

Some members of the NREC heard through news media about the measures adopted in these trials in response to the new scientific evidence and shared this information immediately with the committee. The NREC scheduled an extraordinary meeting on the following day to discuss the implications of this information. The committee considered that these events might have an impact on the ethical acceptability of the three trials it had approved (PAHO 2020). However, some members were not sure how the NREC should respond, given that it had not had access to the details of the new evidence. The committee finally decided that as it is a researcher’s responsibility to report such information to the NREC, they should wait to receive the researchers’ reports. They also agreed that in the midst of the pandemic, their priority was to review COVID-19 protocols recently submitted for ethics approval rather than to ask for more reports from previously approved studies.

The updates that the NREC received from the researchers of the trials testing hydroxychloroquine varied, as indicated below:

• Trial A: After 3 days, the principal investigator informed the NREC that they had suspended the trial in order to evaluate the impact of the new data on the efficacy and safety of hydroxychloroquine for their participants, who had mild to moderate COVID-19. Given that the new evidence was specifically for hospitalized COVID-19 patients, the research team was not sure if they were justified in extrapolating these results to the participants of their trial.

• Trial B: After 2 weeks, the principal investigator sent the periodic progress report according to the deadline that had been previously established by the NREC. The report stated that 23 participants were enrolled, and that the trial was being conducted in accordance with the protocol. No mention was made of the recent reports from other trials.

• Trial C: Three weeks after reports emerged about the suspension of the other hydroxychloroquine trials, the NREC had not received any information from the research team.

In response, the NREC took the following action. For Trial A, the committee asked the principal investigator to communicate to the participants the reasons for the suspension of the trial, and to inform the NREC of their final decision regarding the continuation of the trial as soon as they had finished analysing the evidence. For Trials B and C, the NREC asked the principal investigators to justify the continuation of the study on the basis of the newly available evidence. In the case of Trial C, the principal investigator replied that he had not provided any report because the new evidence did not affect the rationale for testing hydroxychloroquine as prophylaxis for COVID-19.

Questions

• 1. What are the implications of the rapid production of new evidence during a pandemic for the ethics oversight of COVID-19 research?

• 2. What are the responsibilities of COVID-19 researchers with respect to emerging scientific evidence that could affect the justification for conducting their research and/or how it is conducted?

• 3. Should research ethics committees adopt special operating procedures for the oversight of ongoing COVID-19 research? If so, what should these procedures include and why?

• 4. How should the ethical analysis of on-going protocols be conducted in light of new evidence, in order to ensure their continuing ethical acceptability? What questions should guide this assessment?

References

• PAHO. 2020. Guidance for ethics oversight of COVID-19 research in response to emerging evidence. Washington: Pan American Health Organization. https://iris.paho.org/handle/10665.2/53021.

• University of Oxford. 2021. RECOVERY: Randomised evaluation of COVID-19 therapy. https://www.recoverytrial.net/.

• WHO. n.d. “Solidarity” clinical trial for COVID-19 treatments. World Health Organization. https://www.who.int/emergencies/diseases/novel-coronavirus-2019/global-research-on-novel-coronavirus-2019-ncov/solidarity-clinical-trial-for-covid-19-treatments.