Background

Late-phase randomised controlled trials often require hundreds or thousands of participants to detect meaningful differences in outcomes. In order to successfully answer their research questions, trialists must recruit volunteers to take part, often asking participants to accept risk and/or inconvenience, with the aim of improving treatment, care or prevention of disease for future patients.

Sharing results with trial participants is an ethical imperative [1] and is recommended by authorities that govern the conduct of clinical trials. The World Medical Association’s Declaration of Helsinki, which outlines the principles for ethical conduct of medical research involving human participants, states that ‘all medical research subjects should be given the option of being informed about the general outcome and results of the study’ [2]. In the UK, the Health Research Authority (HRA) recently published guidance saying that participants have the right to know the findings of research in which they have taken part, and that sharing results directly with participants can help ‘build trust, show respect and helps participants feel valued’ [3].

There is evidence from a broad array of studies that most participants want to be offered the opportunity to receive trial results, ranging from 88 to 98% in studies conducted across a range of diseases (cancers, idiopathic scoliosis, internal derangement of the knee, HIV) and geographical settings (including USA, UK, Canada and Uganda) [4,5,6,7,8].

Despite the moral obligation and clear demand from most participants to receive results, in practice, sharing results often does not happen, or is not done well. The UK HRA research transparency report in 2021 states that ‘90% of clinical trials have not told participants about findings’ [9]. A survey conducted in 2016 of authors of clinical trial results papers published in 2014–2015 found that only 27% of respondents reported disseminating results to participants, with only a further 13% planning to do so [10]. Even when it does happen, it may not be done in a way that participants can understand. The survey found that 40% of authors who had shared results with participants had shared academic publications, which are not written in a way that is easy for participants to understand [10]. Previous studies have reported that many participants struggle to understand trial results which are shared with them. For example, a study within the context of a breast cancer trial found that only 56% of participants said the results letter was easy to understand [11], and a survey of cancer trial participants found fewer than half reported fully understanding the results [7].

Sharing results with participants is a complex issue, with trialists facing considerable challenges including practical [10, 12,13,14,15,16,17] and resource barriers [10, 12,13,14,15, 18, 19] and concerns about the emotional impact of sharing results [4, 7, 11, 12, 14, 19,20,21,22]. It is important that it is done well, as there is potential for harm [13, 23,24,25]. Little work has been done to compare approaches to sharing results with participants. Show RESPECT assessed approaches to sharing results in a cluster randomised factorial trial, comparing an enhanced versus basic webpage; mailed printed summary versus no printed summary; and email list invitation versus no email list invitation, within the context of an ovarian cancer trial (ICON8) [26]. A major finding was that the mailed printed summary significantly improved patient satisfaction with how results were shared compared to a webpage with or without an email list invitation, without the printed summary [26]. It also showed that these approaches were feasible for site staff to implement [27]. However, Show RESPECT was carried out within only a single trial, in a particular clinical and geographical setting, with a particular set of results to communicate, so the generalisability of these results is unclear. It is likely that there is no one-size-fits-all approach to sharing results, so it is important that trial teams draw on guidance that offers a sound structure that can be adapted to fit the specific context and requirements of their own trial and trial participants.

One of the aims of the Show RESPECT study was to develop guidance for trialists, based on our results. We realised that findings from Show RESPECT [26, 27] could be used to derive an adaptable framework of factors for trialists to consider when planning how to share results with trial participants. This paper presents the framework, based on and illustrated with qualitative insights from data collected from site staff and patients who took part in the Show RESPECT study.

Methods

Show RESPECT was a mixed methods study, comprised of a factorial cluster randomised controlled trial within a trial to assess multiple approaches to communicating trial results, and an embedded explanatory qualitative study. The full protocol for the study is available online [28]. Results from Show RESPECT with regard to participant satisfaction with how the results were shared, and the resources required from sites and the clinical trials unit (CTU) to implement these approaches, have been reported previously [26, 27] and the qualitative results reported in this paper have been published as part of a doctoral thesis [29]. Show RESPECT took place within the ICON8 ovarian cancer chemotherapy trial (ISRCTN10356387) [30]. As the methods have been reported previously, we do not duplicate this information in this article, but have included it as Additional File 1. This paper reports qualitative results from data collected from trial participants and site staff who had been involved in sharing or receiving results. The Standards for Reporting Qualitative Research checklist for this paper can be found in Additional File 2.

Information about patient and public involvement in this study and the context of the ICON8 trial is available in Additional File 1.

Qualitative data collection

The main source of qualitative data for Show RESPECT was semi-structured interviews with ICON8 patients and site staff who had been involved in sharing the ICON8 results with patients. In addition to the qualitative interviews, qualitative data were collected by free-text questions on the questionnaires that were completed by patients (after receiving results) and site staff (immediately after sharing results and 2–3 months later). The topic guide, questionnaires, details of how these were administered, and researcher characteristics and reflexivity can be found in our previous publications [26, 27, 29, 31]. Further information about our qualitative data collection is available in Additional File 1.

Sampling and participants

We used a purposive sampling approach for the semi-structured interviews with both participants and site staff, allowing us to collect data from respondents with a range of characteristics that may be related to their experiences and views on sharing results. For ICON8 patients, this included age, education level, frequency of internet use and reported satisfaction with how the ICON8 results were shared, while for site staff this included their role and number of ICON8 patients at the hospital at which they work. For both groups, we included which interventions their hospital had been randomised to. Further information about our sampling approach and participants is available in Additional File 1.

Qualitative analysis

We used a reflexive thematic analysis approach [32], with a critical realist stance (taking the ontological position that an external reality exists that is independent of our beliefs and understanding, but that our knowledge of that external reality is influenced by our historical, social and cultural situation), to analyse the data. The findings reported in this paper are further findings from the analysis carried out for our previous papers [26, 27], rather than a separate re-analysis of the Show RESPECT data. Further details of our analysis methods are available in Additional File 1.

Developing the framework

We shared our findings around what influenced the experience of patients and site staff around receiving/sharing trial results at a patient and public involvement meeting with women who were taking part in ovarian cancer treatment trials. We held meetings with site staff who had been involved in sharing the ICON8 results with participants, and met with the ICON8 trial team. We also held seminars at three clinical trials units and presented our findings at a clinical trials conference. At these meetings, we discussed with these key stakeholders the implications our results, how they might be transferred to trials in other settings, and recommendations they would make for future trials. Based on the themes, sub-themes and high-level codes from our data, and the stakeholder discussions, we developed a long list of considerations that we believe trial teams should take into account when planning how to share results with trial participants, either because it came from our qualitative data, or was raised as an additional consideration during our discussions with stakeholders. We grouped related considerations together into categories and organised the categories so the initials spell a memorable phrase (SHOW RESPECT).

To explore how useful the framework was for a trial that was very different to the ICON8 trial in which we carried out Show RESPECT, we applied it to CHAPAS-4 (ISRCTN 22964075) [33]. This was done by a single researcher (AS) who worked with the study teams and (for CHAPAS-4) community representatives to consider the factors identified in the framework, and how they affect the communication of results to participants for these trials. AS had been involved in CHAPAS-4 from the proposal development stage, so was familiar with the study.

Results

A description of the patient and site staff participants in Show RESPECT is reported in our previous papers [26, 27]. A table showing a summary of their characteristics can be found as Additional File 3.

The adaptable framework of factors trialists should consider when planning how to share results with trial participants is shown in Table 1, with illustrative quotes from the Show RESPECT data. An editable template with the adaptable framework can be found online [34]. The framework covers supporting and preparing participants to receive results; how the communications tools will reach participants; who the trial participants are; what the results show; special considerations; who will provide results to participants; the expertise and resources the trial team have access to for sharing results; which communication tools will be used; and timing of results communication. Additional File 4 shows how the framework items relate to the qualitative themes, sub-themes and high-level codes from Show RESPECT. Additional File 5 illustrates these factors with findings from the qualitative interviews conducted during Show RESPECT.

Table 1 Factors to consider when planning how to share results with trial participants, with illustrative quotes from the Show RESPECT study
Full size table

Consideration needs to be given to supporting and preparing participants to receive trial results. This includes what participants are told when they join the trial, and immediately before receiving the trial results. It also includes how participants will be able to access support around dealing with the emotional aspects of processing the trial results, and finding answers to questions they have about the results and their implications. Patients in ICON8 differed in the extent to which they felt comfortable asking site staff for more information or clarification, and their confidence in searching for health information from other sources, such as online. Some patients were part of local support groups for people with cancer, whereas others felt they received sufficient support from family and friends. Still other patients were neither linked to support groups, nor had family or friends they could talk to about their cancer. In this context, both patients and site staff felt that links to further information and support might be useful for some patients (even if not themselves), particularly those with less access to support with processing the results.

Thought also needs to be given to how the communication tool(s) will reach participants, and the accessibility needs of your patient population. Alongside the question of how the results will reach participants is the question of where. Receiving results in the clinic may make support and clarification more easily available but provides less privacy and time for processing the results than patients receiving them at home.

Participant characteristics may affect the appropriateness of different communication approaches. The people taking part in the ICON8 trial were women with an average age of 67 by the time results were available. Four in ten of them used the internet and email less than daily [26]. In this context, printed summaries were viewed as being easy to access for all participants (including older participants and those who are not confident computer users). Other patient characteristics that may affect results communication include education level and health literacy. Non-written forms of communication (such as videos) may be useful for those who do not like to read. Consideration should also be given to what participants are likely to want to do with the results. Many patient interviewees kept folders with all the information they received about the ICON8 trial, to allow them to refer to it for future reference. Printed summaries of the results facilitated this, while email or webpages required printing. Printed summaries also made it easier to share results with others, such as family and friends.

The nature of the trial and its results also affects how results should be shared. ICON8 found no difference between the different chemotherapy schedules tested. In some ways, this made it easier for some patients to receive the results, as although they were disappointed that the trial did not find an improvement in treatment of ovarian cancer, no one was allocated to an inferior arm. The approaches used to share results in Show RESPECT were felt to be appropriate in this context. If the results had been different, with a clear difference between the arms, some patients and site staff felt that there may have been a need to communicate results to people in the group that had done less well overall in a more personal way. This may be less important in trials for less severe conditions than ovarian cancer, where participants have less riding on the results. Similarly, if the results are complex, they may need personal discussion to help patients to understand. One item in the framework that came from engaging with stakeholders rather than directly from the Show RESPECT data was that of ‘special considerations’ that need to be taken into account, such as if the trial had closed early, or experienced adverse media coverage. In ICON8, some patients wanted explanations on why the ICON8 results differed from those of previous similar trials in different settings. Patient and public involvement in the design of tools and processes is essential.

Communication of results takes place within the context of relationships that have developed over the course of the trial between patients and site staff. Participants in ICON8 have been in follow-up for 5 to 8 years, with regular clinic visits during that time. At sites where participants were seen by the same site staff each time, many developed close relationships, almost friendships. Where this was the case, site staff felt uncomfortable sharing the results without some degree of personalisation, so some added personalised cover notes, or called participants to let them know the study results were about to be disseminated. Communication of trial results should consider the strength of relationships developed between site staff and patients, for example allowing a degree of personalisation of how the results are shared where these relationships are close. Some staff at the largest sites did not know participants so well and felt uncomfortable telephoning patients out of the blue. There may be less need for personalisation in trials with shorter follow-up, or with follow-up that does not involve face-to-face visits with consistent staff over time, or when results come from staff other than those who had developed relationships with participants.

When considering the expertise and resources needed for sharing results with participants, thought needs to be given to the skills, staff time and budget needed for the development of the information product (e.g. writing the content in plain language, patient and public involvement, scientific review to ensure the summary is accurate, and technical skills required for the chosen communications tools); distribution of the results (e.g. site staff time for posting information, costs of distribution [e.g. printing, postage]) and supporting participants and dealing with queries. Our previous report describes the resources required from sites and the clinical trials unit for sharing the results in the ways tested in Show RESPECT [27]. Budget or staff time limitations may rule out certain approaches to sharing results, if they have not been included as part of the initial budget for and funding of the trial.

Choice of communication tools will be influenced by the factors described above. Patient and public involvement is important in helping to make these decisions. In addition to deciding what type of communication tool(s) to use, consideration needs to be given to the content of that tool. It should include the language(s) used and the appropriate reading level for the target audience (if a written tool is used). It should also include consideration of how to make the information attractive and easy to use (which may require input from design specialists). It may be appropriate to offer participants a choice of communication tools, possibly with different levels and forms of information.

The final factor for consideration is timing—when should the results be communicated to participants? This will depend on the point at which the research team are confident that the messages for participants are unlikely to change, and whether (and when) the results are likely to receive media or social media attention, to avoid participants finding out the results from others before hearing from the trial team. It should take into account when results will be released to other audiences (e.g. via conferences, peer-reviewed publications and public trial databases and registries), and associated embargoes and deadlines (such as the European Medicines Authority requirement to post a summary of results within a certain time period from the end of the trial).

Practical examples of the application of the framework to two very different trials can be found in Table 2: the ICON8 ovarian cancer trial in which Show RESPECT was conducted, and the CHAPAS-4 paediatric HIV treatment trial, which was conducted in Uganda, Zambia and Zimbabwe.

Table 2 Applying the SHOW RESPECT adaptable framework to the ICON8 and CHAPAS-4 trials
Full size table

Discussion

Summary of key findings

We propose several considerations when planning how to share results with participants in clinical trials. This includes how participants will be prepared and supported when receiving results and how the communication tool(s) will reach participants. Participant-related factors, such as demographics, education levels and computer literacy, alongside their health and expectations around receiving results, must also be considered. The trial results themselves (whether they will be considered as good, bad or neutral news by some or all participants, and their complexity) also need to be taken into account. Trials with more complex or potentially upsetting results (e.g. where the participants allocated to an arm did less well than participants allocated to other arms, or where one sub-group did less well than others) may need to offer participants additional support, for example through sharing results face-to-face or in individual video calls, or offering follow-up appointments or phone calls with doctors or research nurses if results are shared via written summaries. Trial results communication must also consider whether participants have developed relationships with site staff over the course of their participation, and how and from whom they are used to receiving communication about the trial. It may be appropriate to reflect this in some way, for example through allowing personalisation or one-to-one communication. The expertise and resources available to trial teams to communicate trial results is an important factor when deciding how this is done. Any communication tools used must reflect what the participants are likely to want to know and be understandable (using appropriate language(s) and reading levels) and accessible to the intended audience. It may be appropriate to provide a choice of tools, as different participants are likely to have different preferences and needs. The timing of when results are shared also needs to be carefully considered, avoiding participants finding out results from other sources prior to being informed by the trial team, if possible. Considering these factors, and involving patients and the public, can help develop communication tools and processes that are appropriate to the trial context, population and messages.

Strengths of this study

A key strength of this study is its integration of qualitative data from both site staff and ICON8 trial participants, giving us insight into the views of those who are responsible for sharing results, alongside those who have experienced receiving trial results. Many of the site staff who took part in Show RESPECT worked across many trials and were able to draw on their experience from other studies in addition to Show RESPECT. The qualitative data provide a rich understanding of the perspectives of ICON8 trial participants and site staff on the experience of receiving or sharing trial results. Applying an established theoretical model (the Information Seeking and Communication Model [35, 36]) increased our ‘information power’ [37], through synthesising existing knowledge, extending the sources of knowledge beyond our empirical data and explaining relations between different aspects of the empirical data in a coherent way [37]. Applying the model helped us to ground our conclusions in the context of existing knowledge about the process of information seeking and communication.

Discussion of our findings with key stakeholders working on a wide variety of trials allowed us to ensure the framework is applicable beyond the ovarian cancer setting. The applicability of our framework to very different trials is illustrated by the example of the CHAPAS-4 trial, shown in Table 2. Applying the framework was helpful for thinking through how to share results with participants in CHAPAS-4. The answers to the individual questions were very different from those for ICON8, and very different communication approaches were selected, but the considerations were all relevant. While the framework does not directly prescribe how to share results, having a structured framework to follow gave confidence that nothing important had been overlooked. We envisage the framework being most useful as the basis of discussion of ideas and plans between members of the trial team and patient representatives.

Limitations of this study

Show RESPECT was carried out within the context of a single trial, a limitation of this study, raising questions about the transferability of the findings to trials with different patient populations, diseases, results scenarios and settings. We acknowledge this possibility and emphasise that in this paper we focus on exploring factors that trialists should consider when preparing to share results with participants, rather than recommending that the approach that worked best within Show RESPECT should be used in trials with very different contexts or patient populations. We further acknowledge that we were not able to take account of ethnicity of respondents, nor on factors such as socio-economic background, as these data were not collected for this study. Very few of our patient participants reported having a first language other than English, but patient participants did report widely varying education levels.

The development of the framework was not a formal deliberative process—the framework is an output from our research that we believe will be of value to other researchers. However, we acknowledge that, given it is largely informed by evidence from a single trial, there may be considerations that we have missed that might be important for other trial contexts. We see this as being a starting point for improving practice in this area, but recognise that further refinement of the framework may be needed after it has been applied in a wide variety of trials. We invite readers to send us feedback around their experience of using the framework and will consider revising it in the future if further important considerations are identified or improvements need to be made.

Our results in the context of what was already known

Our adaptable framework of factors to consider when planning how to share results is similar to guidance released by the UK Health Research Authority in 2023 [3], after the Show RESPECT patient results had been published. These similarities are unsurprising, given that the results of Show RESPECT helped inform this guidance, and several authors of this paper were involved in developing them. The HRA guidance on what to consider covers:

  1. a)

    Who will receive the findings

  2. b)

    How you will communicate the findings

  3. c)

    Giving participants a choice

  4. d)

    Responsibility for communicating findings

  5. e)

    Exceptions

  6. f)

    When to communicate findings

  7. g)

    Evaluating your communication [3]

Our findings around giving participants a choice over whether to receive the results or not reinforces previous recommendations that a two-stage approach should be used, offering results and then providing them, rather than simply distributing results to all participants [38]. Choosing not to access results was, for some patients, a way of protecting themselves from potentially finding out that they missed out on the better treatment. This concept of people choosing what information to engage with or not as a protective mechanism is similar to findings from the BRACELET study, where some bereaved parents of babies who died while participating in a trial for very high-risk neonates advised that communication from the trial should be managed in a way that would suit any parents who felt that they might be upsetting for themselves or their partner [39].

Only by providing information in a way that is understandable to the intended audience can we meet the objectives of sharing research results. Care needs to be taken when preparing results summaries, to ensure they are comprehensible for participants. Previous research has found that much written information about clinical trials exceeds the average reading age [40]. The UK National Health Service Digital Service Manual style guide states that they aim for a reading age of 9–11 years old where possible [41]. Artificial intelligence can be used to help researchers produce plain language summaries, but these will still require review from both investigators and patient representatives to ensure the content is correct and appropriate for the intended audience.

Many of our findings align with findings from the RECAP study [42, 43], supporting the transferability of our framework to trials beyond the UK ovarian cancer setting in which our work took place.

Further research

Further research involving participants and site staff receiving and sharing trial results in trials with different patient populations, trial characteristics and results scenarios would be valuable for exploring the transferability of our findings to other contexts. Research is also needed to address how demographic factors such as geographical location, socio-economic status, ethnicity and different levels of language proficiency influence how results should be shared with participants.

Conclusion

To ensure that trials meet their moral obligations to participants to share trial results, trialists must consider how results should be shared with participants from the planning stage of trials, to ensure that adequate resources are budgeted for and included in agreements with sites. Relevant information about how results will be shared should be included in the Patient Information Sheet. When deciding how to share results with participants, trialists should consider the following factors: how to support and prepare participants to receive results, including whether to use an opt-in or opt-out approach and who will be available to answer participant questions; how the results will reach participants; the characteristics and expectations of participants in relation to the results; what the results show and how they are likely to be perceived by participants; special considerations; who will provide the results to participants; the expertise and resources available for sharing results; the communication tool(s) to be used; and the timing of results communication. Patient and public involvement is essential for planning how to share results with participants, identifying the outcomes and study results that are important and relevant to participants, and developing the content of results summaries to ensure they are written in a clear and sensitive manner.