Patient values in patient-provider communication about participation in early phase clinical cancer trials: a qualitative analysis before and after implementation of an online value clarification tool intervention

Abstract

Background

Patients with advanced cancer who no longer have standard treatment options available may decide to participate in early phase clinical trials (i.e. experimental treatments with uncertain outcomes). Shared decision-making (SDM) models help to understand considerations that influence patients’ decision. Discussion of patient values is essential to SDM, but such communication is often limited in this context and may require new interventions. The OnVaCT intervention, consisting of a preparatory online value clarification tool (OnVaCT) for patients and communication training for oncologists, was previously developed to support SDM. This study aimed to qualitatively explore associations between patient values that are discussed between patients and oncologists during consultations about potential participation in early phase clinical trials before and after implementation of the OnVaCT intervention.

Methods

This study is part of a prospective multicentre nonrandomized controlled clinical trial and had a between-subjects design: pre-intervention patients received usual care, while post-intervention patients additionally received the OnVaCT. Oncologists participated in the communication training between study phases. Patients’ initial consultation on potential early phase clinical trial participation was recorded and transcribed verbatim. Applying a directed approach, two independent coders analysed the transcripts using an initial codebook based on previous studies. Steps of continuous evaluation and revision were repeated until data saturation was reached.

Results

Data saturation was reached after 32 patient-oncologist consultations (i.e. 17 pre-intervention and 15 post-intervention). The analysis revealed the values: hope, perseverance, quality or quantity of life, risk tolerance, trust in the healthcare system/professionals, autonomy, social adherence, altruism, corporeality, acceptance of one’s fate, and humanity. Patients in the pre-intervention phase tended to express values briefly and spontaneously. Oncologists acknowledged the importance of patients’ values, but generally only gave ‘contrasting’ examples of why some accept and others refuse to participate in trials. In the post-intervention phase, many oncologists referred to the OnVaCT and/or asked follow-up questions, while patients used longer phrases that combined multiple values, sometimes clearly indicating their weighing.

Conclusions

While all values were recognized in both study phases, our results have highlighted the different communication patterns around patient values in SDM for potential early phase clinical trial participation before and after implementation of the OnVaCT intervention. This study therefore provides a first (qualitative) indication that the OnVaCT intervention may support patients and oncologists in discussing their values.

Trial registration

Netherlands Trial Registry: NL7335, registered on July 17, 2018.

Background

Decision support technologies, or decision aids are evidence-based tools designed to support patients in reaching informed, value-based choices [1, 2] and have been widely recommend to assist patients in making difficult decisions. Patient values are desires, goals or beliefs that patients find important [3, 4] and are related to the perspectives that these patients have on the ‘good life’ [5]. The Ottawa Decision Support framework [6, 7] considers clarifying patient values a key part of decision support (technologies). Recent systematic reviews confirmed that “unclear values” are an important decisional limitation for patients [8]. Decision aids can improve decisional quality [9] as well as patient‐clinician communication and congruence between patients’ informed values and their chosen options [10]. Specifically, exposure to a decision aid with explicit value clarification resulted in a higher proportion of patients choosing an option congruent with their values [11]. Naturally, the International Patient Decision Aids Standards (IPDAS) strongly recommends the inclusion of value clarification in decision aids [1].

If value clarification methods (e.g. in decision aids) are used before and during a consultation, this could help patients to become more aware of their values and facilitate a shared decision-making (SDM) process [12]. SDM is “an approach where clinicians and patients share the best available evidence when faced with the task of making decisions, and where patients are supported to consider options, to achieve informed preferences” [13], and is highly recommended for decision-making between patients and providers [14, 15], especially when a choice is value-sensitive. Widely used SDM models consider eliciting patient values and preferences a key part of health- and treatment-related decision-making [12, 16, 17]. Having a discussion (e.g. during a doctor-patient consultation) on patient values is particularly important because patients rarely reach clarity of values and preferences without the help of others [18]. Moreover, discussing patient values can enhance patients’ involvement in the SDM process [19]. This aligns well with patient-centred care which means “providing care that is respectful of and responsive to individual patient preferences, needs, and values and ensuring that patient values guide all clinical decisions” [20, 21]. Indeed, one of SDM’s main goals is to deliver care in line with patients’ personal values [22, 23]. Discussing patient values becomes even more important when there is no clear ‘best’ or ‘worst’ option in the upcoming decision.

A decision that is – ideally – pre-eminently based on patient values, is the decision whether or not to participate in oncological early phase clinical trials, i.e. trials on experimental treatments with unknown outcomes in terms of e.g. effectiveness, toxicity and survival. Some patients who decide to participate in early phase clinical trials hope for clinical benefit [24] even while realizing that this chance may be small [25]. Patients could also decide not to participate but rather to focus on alleviating their existing symptoms and complaints by means of palliative care. Patients who participate in early phase clinical trials however may also use palliative care services in addition to the experimental trial with its many uncertainties. This complex context creates a difficult decision with no really ‘best’ option and consequently has major impact on the patients’ life. Integrating patient values into the SDM process could help ensure that patients spend the limited time towards the end of their life in the way that they want. In addition to hope, a previous systematic review [24] and a qualitative interview study [26] of our team found that important patient values impacting the decision upon early phase clinical trial participation are, amongst others, quality and quantity of life, acceptance (e.g. of their prospective death) and body preservation (e.g. maintaining fitness of the body, or preventing destruction of bodily senses).

However, while previous systematic reviews support the idea that SDM for early phase clinical trials needs to address both medical-technical information and patients’ values and preferences [27, 28], discussions about participation in early phase clinical trials tend to focus on explaining complex medical-technical information to patients [27, 28]. In this sense, oncologists currently appear to use an ‘informative model’ [29]. As these studies illustrate the currently limited integration of patient values, it remains important to support such a discussion. Aligning information with the subjective preferences of patients with advanced cancer may prevent unmet information needs [30]. However, moving towards an ‘interpretative model’ [29], in which oncologists take on the role of counsellor to help patients articulate and weigh their values, has its own challenges. Even when patients are empowered to overcome the difficulties of sharing their values [27], the challenge for oncologists remains to work with the patient during the consultation to appropriately clarify these values and how they relate to each other. Based on the literature on SDM and decision aids, a new intervention centred on value clarification could potentially provide a solution for these problems.

Following the Medical Research Council (MRC) framework for complex interventions [31, 32], we systematically developed the theory- and evidence-based OnVaCT intervention [33], consisting of an online value clarification tool (OnVaCT) for patients and communication training for oncologists. Together, these aimed to support patients and oncologists in clarifying and discussing patients’ personal values. In practice, we hypothesized that the OnVaCT intervention could lead to changes in the communication patterns, for instance by means of a (more) clear, explicit and/or extensive discussion on patient values and their relative weighing for individual patients. With this study, we want to unravel how the implementation of the OnVaCT intervention is actually reflected in the discussion of patient values in clinical practice, i.e. in doctor-patient consultations focusing on SDM for potential early phase clinical trials. More specifically, this study qualitatively analysed (in a between-subjects design, i.e. pre-intervention and post-intervention phase) the following research questions:

• RQ1. Which patient values are discussed (in context with each other) between patients and oncologists during consultations about potential early phase clinical trial participation before and after the implementation of the OnVaCT intervention?

• RQ2. How are patient values discussed between patients and oncologists during consultations about potential early phase clinical trial participation before and after the implementation of the OnVaCT intervention?

Method

Design

This study is a qualitative (content) analysis that is part of a larger prospective multicentre nonrandomized controlled clinical trial on SDM for early phase clinical trial participation that makes use of mixed-methods (Netherlands Trial Registry: NL7335) [34]. This study had a between-subjects design: patients included in the pre-intervention phase (i.e. before implementation of the OnVaCT intervention) functioned as ‘control group’ within current standard practice/usual care, while patients in the post-intervention phase (i.e. after implementation of the OnVaCT intervention) functioned as the ‘intervention group’ and were asked to use the OnVaCT before their consultation. Oncologists from three hospitals with large early phase clinical trial units in the Netherlands (i.e. Erasmus MC, Rotterdam; Netherlands Cancer Institute–Antoni van Leeuwenhoek, Amsterdam; and UMC Utrecht) participated in the communication training between the two study phases. The final OnVaCT intervention is summarized in Table 1 and described in detail elsewhere [33, 35]. Standards for Reporting Qualitative Research [36] were used to report the qualitative analysis.

Table 1 Summary of the OnVaCT intervention [33, 35]

Participants

The inclusion criteria were: diagnosed with advanced cancer and eligible for first participation in an early phase clinical trial (i.e. phase I or phase I/II); aged 18 years or older; sufficient command of the Dutch language; and written informed consent. Exclusion criteria were: cognitive impairment according to the medical record; no access to the Internet to fill out the online questionnaires (and use the OnVaCT); or participation in another part of the project (N = 13) [26].

Procedure

From February 18, 2019 up to December 18, 2020, patients were approached to participate in the pre-intervention phase (i.e. before implementation of the OnVaCT intervention). The study was put on-hold between March 16 and May 20, 2020 due to restrictive measures regarding the COVID-19 pandemic. Patients were approached to participate in the post-intervention phase between January 26, 2021 and August 31, 2022. Eligible patients were called by a nurse practitioner, trial secretary or researcher from their own hospital. Patients who verbally agreed, were sent an e-mail with the patient information and consent form by a researcher/assistant, and were called again to ask for preliminary consent. Written informed consents were signed and collected directly before the consultation. The consultation was (video- or) audiotaped. Local trial monitors later collected additional data regarding patients’ decision and their medical status from electronic patient records.

The OnVaCT intervention was implemented between December 18, 2020 and February 1, 2021. During this period, participating oncologists followed the communication training and the procedures (e.g. patient package) were adjusted. In the post-intervention phase, patients were sent the link to the OnVaCT via e-mail before their initial consultation, and were asked to use it in preparation to the consultation. Whether patients had used the OnVaCT prior to the consultation was verified in the transcripts and by checking whether they had filled out a questionnaire about the OnVaCT (results of this questionnaire will be reported elsewhere).

Sample

Starting with the pre-intervention data, we selected one recording for each participating oncologist who had at least one available recording in both study phases during every round of coding. We primarily aimed for an equal division of patients who accepted and refused trial participation (in total and per oncologist), and secondarily for a diverse group in terms of gender, age, and primary tumour location. We analysed pre-intervention consultations until saturation of themes was reached. Then, the entire procedure was repeated for the post-intervention phase. To avoid drop-out bias (i.e. to not miss relevant views and values from patients who could or would not comply to use the intervention), we applied an intention-to-treat approach [37]: patients included in the post-intervention phase were analysed as such, regardless of whether they had actually used the OnVaCT.

Data analysis

The qualitative software NVivo for Windows (release 1.4) was used. We applied a directed approach (which is both inductive and deductive) to thematically analyse the recordings [38]. An initial coding scheme with (sub)themes (see Table 2) was deductively created based on the findings from our previous systematic review [24] and interview study [26]. During the data analysis, the coding scheme was inductively adjusted (e.g. additions and/or omissions were made) based on the actual content of the communication (see below).

Table 2 Initial codebook

Selected recordings were transcribed verbatim and analysed by two coders with a background in communication (LL) and health sciences (MH). The data analysis of the transcripts was an ongoing, iterative process, with each transcript being double-coded by the two researchers independently. After every transcript, they compared their findings and resolved discrepancies in the codes by means of discussion. The progress of the coding process (including discrepancies and findings) was also discussed during regular meetings with two other researchers with a health communication (JW) and medical ethics (JG) background. If needed, adjustments to the codebook were made after establishing mutual agreement between the four researchers. Although not quantitatively verified, the two independent coders reached a good level of agreement after the first five transcripts, with generally no major discrepancies between them for the remaining consultations. Findings were peer-reviewed by two researchers/oncologists with a palliative care (CR) and early phase clinical trial (MJ) background. These steps of continuous evaluation and revision were repeated until data saturation was reached (i.e. no new themes arose) for both study phases separately. Supplementary Table 1 shows the final codebook.

When data saturation was reached for both study phases, a more ‘inductive’ phase of the analysis started in which the coders took a closer look at the encoded phrases per value to assess which values and subthemes were (not) discussed, the possible links between values, and which subthemes could be recognized into how these values were discussed by patients and oncologists. To enable analysis of values discussed in context with each other, each value was encoded together with the entire phrase in which it occurred, not to lose relevant sentences/questions/replies that preceded or followed the mentioned value. The two coders compared and discussed their labels and the observed links between different values, also together with the study team, until consensus was reached. Finally, the pre-intervention and post-intervention findings (that were in two separate NVivo files) were compared to each other by first comparing the patient values that were discussed (in context with each other) in the pre-intervention phase to those discussed (in context with each other) in the post-intervention phase, and then how these values were discussed in these two phases.

Results

Table 3 shows the characteristics of the selected 32 patients (i.e. seventeen pre-intervention and fifteen post-intervention) divided across nine oncologists (range: 2–4 recordings per oncologist). Patients were mostly male (N = 19) and between 36 and 78 years old. The consultations lasted between 28 and 53 min. In both study phases, approximately half of the patients decided to participate in an early phase clinical trial. Two post-intervention patients did not use the OnVaCT prior to the consultation. In the following paragraphs, we will first describe which of the patient values were discussed and then elaborate on the connections that were made between the values (RQ1), and then how oncologists and patients discussed patient values (RQ2). In each paragraph, the results from the pre-intervention and post-intervention phase are described consecutively to explore how the implementation of the OnVaCT intervention was reflected in these discussions.

Table 3 Patient characteristics

Which patient values were discussed

The consultations revealed the values hope, perseverance, quality or quantity of life, risk tolerance, trust, autonomy, social adherence, altruism, corporeality, accepting one’s fate, and humanity. Table 4 provides a more extensive description with quotes for each of the values that were used during consultations.

Table 4 Description of each patient value that was discussed in the pre-intervention and post-intervention phase

Several values were referred to in (almost) all pre-intervention discussions: quality of life was mentioned by all patients, and e.g. corporeality, risk tolerance and hope by most. In contrast, other values were mentioned only rarely, such as humanity, altruism, social adherence and trust. Figure 1 illustrates the main links that occurred between values in the pre-intervention phase. For instance, quality of life was mentioned together with accepting one’s fate (“It can of course just be a part of the process […] but I would still like it if I would just feel a little better”, subject 02), corporeality (“I’m fine […] You can stay put, but then you will become a ‘vegetable’, so I still have a reasonable stamina”, subject 03) and quantity of life (“I won’t live for many more years, I think […] I am already living more extensively, but yes, I think that I have to do that”, subject 09). Furthermore, hope and perseverance appeared to be connected, e.g. “Cons, I don’t really see because if you don’t do it, you’re certain that it’s over […] so I can actually only have certain benefits from it [trial participation]” (subject 03). These values were also sometimes referred to together with risk tolerance (as in the example), corporeality, social adherence, autonomy, altruism or trust. The only value that was not mentioned with any of the other values in the pre-intervention phase, was humanity (e.g. one patient mentioned that “[he did] not feel like being turned inside out again”, subject 02).

Fig. 1

Which patient values were discussed (together) in the pre-intervention phase*

* Disclaimer: during the coding of the transcripts/consultations, we encoded the entire phrase (e.g. with the preceding and consecutive sentence, or question/answer) in which a value was mentioned. This figure only provides a visualisation of values that occurred together in these phrases

In the post-intervention phase, similarly, quality of life was mentioned by all patients. Other values that were often referred to were risk tolerance, corporeality, autonomy and hope, whereas humanity, altruism and accepting one’s fate were, again, only occasionally mentioned. Figure 2 shows the main links between values in the post-intervention phase. Now, nearly all values were mentioned together with other values, including humanity, which was sometimes mentioned together with quality of life (e.g. “It still has to be humane, that is what I think. And I still enjoy everything” subject 19). Quality of life, hope, risk tolerance, quantity of life, autonomy and perseverance were mentioned along with all other values. There were multiple occasions in which several diverse values were discussed, for example:

“Of course the biggest question is, what can it bring me, mean to me? Yes, I’m actually already a little happy that I’m hearing from you that I have access to these, at least possible access to these studies. That already relieves a little tension. But well, besides, yes, I hope it helps me […] I have already hoped for four years that science catches up with the disease, and yes, that seems to be less likely and that is hard” (subject 22).

Fig. 2

Which patient values were discussed (together) in the post-intervention phase*

* Disclaimer: during the coding of the transcripts/consultations, we encoded the entire phrase (e.g. with the preceding and consecutive sentence, or question/answer) in which a value was mentioned. This figure only provides a visualisation of values that occurred together in these phrases

In this phrase, the patient consecutively referred to the values risk tolerance, trust, hope and accepting one’s fate. Furthermore, some patients very clearly weighed several values against each other, as illustrated in the following quote in which the patient particularly weighed quantity against quality of life:

“So then you’re talking about an prolongation of life of maybe a few months […], but if that’s it, then I don’t necessarily have to live six months longer. I prefer to have sufficient quality of life. It’s not about the duration of, well that too of course, but […] if I have to choose between the two, I would rather die half a year sooner” (subject 29).

There were a few relations between values that appeared in both the pre-intervention and post-intervention phase: quality of life was sometimes mentioned together with quantity of life or with corporeality; risk tolerance with trust or with hope; hope was also mentioned with perseverance or with altruism; and social adherence with autonomy.

How patient values were discussed

Based on the consultations, we distinguished four major ways in which patient values were discussed (see Supplementary Table 1 and below).

Questions and answers on patient values

Oncologists rarely asked direct questions about patients’ values at the beginning of the pre-intervention conversations. They sometimes indicated that the question “What do I want?” was important, but that it would be left for the time being and discussed later in the consultation. Occasionally, the oncologists explicitly asked the patient “What is important for you to strive for when we discuss treating the illness or not?” (oncologist A) or what they considered to be of importance in general. When patients mentioned values in the beginning of the consultation, they usually did so briefly, in response to a question from the oncologist about their cancer-related symptoms (treatments) and their current physical condition. Patients often mentioned how their cancer (treatment) had affected or was still affecting certain aspects of their life, such as hobbies, work, family and physical exercise. Sometimes patients shortly discussed their values in reaction to more open-ended questions, such as “How can I help you today?” (oncologist D). After mentioning values this early in the consultation, most oncologists continued with asking about the patients’ other (physical) symptoms, medical history or medication use, e.g. “Okay, clear. Do you have any complaints at the moment? How are you doing with eating?” (oncologist C). Later in the consultation, after providing information on early phase clinical trials, some oncologists asked patients for their (first) thoughts, e.g. “Now you have heard all of this, what do you think?” (oncologist I). Few patients then gave responses related to their values, usually by simply indicating what they had hoped for, e.g. “I had hoped that there was an experimental possibility” (subject 13). Oncologists sporadically asked a follow-up question. Besides, oncologists occasionally reacted to patients’ values by stating that it was clear what the patient meant or by saying that “these are really good considerations” (oncologist F).

In the post-intervention phase, similar to pre-intervention phase, oncologists sometimes indicated early in the consultation “What do I want” was an important question that would be addressed later. Furthermore, the discussion of patients’ symptoms and medical history remained an important topic and was, again, sometimes reason for patients to refer to their values. Besides, some oncologists already asked specific questions with regards to what was important to the patients in this phase of their life or “What is your motivation [for participation in an early phase clinical trial]” (oncologist A). In response, patients often described what was important to them in multiple sentences, sometimes explicitly explaining their weighing of values, for instance: “I know it asks a lot of me […] but I think, yes, well, I can get past that […] Only the weeks that I am really sick to death is nasty, but for the rest it’s, well, it’s uncomfortable, but I’m not somebody who says, well, …” (subject 26). In those cases, several oncologists now asked follow-up questions, which enabled a clear dialogue between patient (and relative) and oncologist. For instance, in some cases, the oncologists tried to clarify patients’ weighing of their values by figuring out what they were (not) willing to give up, e.g. “Yes and are there limitations, of which you say, this goes too far for me?” (oncologist A).

Continuing the dialogue

Most of the values in the pre-intervention phase were found in short, implicit expressions of patients in response to information from the oncologist about early phase clinical trials. Some patients responded immediately after or even during the information provision, often in the form of questions. For example, patients asked substantive (sometimes multiple or similar) questions about side effects, rather than stating their worries or wishes, e.g. “What are the side effects? […] But is that fatigue really extreme? […] And no other nausea or diarrhoea or…?” (subject 14). When the patient expressed worry about a certain aspect of the treatment, oncologists tried to explain more about the trial. Preferences and worries about the treatment were sometimes discussed at the end of the conversation, when the oncologist asked if the patient (and relatives) had any more questions, or if everything was clear.

In the post-intervention phase, discussions (i.e. in multiple sentences and with (follow-up) questions) about motivation and values often took place after the oncologist shared information about treatment options. Generally, oncologists explicitly mentioned that patients had a choice and that they had to make a well-considered decision based on what was important to them, for instance “And an experimental context comes with a bit of uncertainty, […] So that is a choice that you have to make for yourself […] Is that something that I want” (oncologist A). These remarks were sometimes combined with a concrete question to the patient, such as regarding what the patient thought about the option to participate in an early phase clinical trial. In other cases, patients already spontaneously responded, for instance when an oncologist emphasized the uncertainty of the situation: “You know that you’re starting your last phase of life. I can deal very well with that thus far” (subject 18). In the dialogue that followed, this patient and his relative together described that they wanted to contribute to science, but not if that required too much investment from the patient because that “would be a waste of time”. Similarly, other patients described multiple values in the same passage of the conversation, sometimes clearly providing their weighing towards each other. Some oncologists responded to such considerations by confirming/repeating them, or by asking follow-up questions.

Providing two ‘contrasting’ examples

In the pre-intervention phase, the main attempt oncologists made to discuss patient values generally happened after providing information about early phase clinical trials, in the form of an (extensive) story describing two ‘opposite’ examples of patients who accepted or declined early phase clinical trial participation based on different values, e.g.: “One says, guys, I’ve tried my best, […] it’s enough for me. […] And the other person can say, guys, I don’t care, […] if they offer me an inch, I’ll take the entire yard. The optimists, the fighters.” (oncologist D). At times, oncologists thereby mentioned that individual patients can feel differently about it, or that both perspectives can be right. Patients rarely spontaneously indicated how their personal values related to those the oncologist described. After the story, patients were generally asked to reflect about it (and their own thoughts) at home. Sometimes it was suggested to discuss everything with e.g. their close ones or general practitioner, for instance: “You don’t have to decide now, because […] such a decision is a big deal. I can imagine very well that you say, well, I want to think about it for a little while, or think about it together, talk about it with people” (oncologist E).

Some oncologists in the post-intervention phase also used opposite examples similar to those described above. In these occurrences, patients responded by indicating their personal considerations and values.

Discussing the OnVaCT

In the post-intervention phase, patient values were sometimes discussed in response to a question by the oncologist about the OnVaCT. Most (but not all) oncologists at least referred to the OnVaCT or the study in general, and often asked patients if they had “thought about it, or about things that are important” (oncologist C) to them personally. A few patients became emotional and had to cry when the OnVaCT was discussed, for instance after saying: “[character from OnVaCT] thought that quality of life, that is important. No longer wanting to keep going at all costs” (subject 21). When the oncologist asked about the OnVaCT, one patient indicated not to have used it, after which the oncologist gave an example of what was described in it and the patient immediately responded with “Yes, that is how I think about it myself” (subject 20). Another patient said that she did not recognize herself in the stories described in the OnVaCT, but when the oncologist asked “But if you had to tell that story for yourself, what is important to you?” (oncologist C), the patient was able to indicate what she missed in the OnVaCT and thereby explain what she valued for herself.

Discussion

This study aimed to qualitatively explore which (RQ1) and how (RQ2) patient values are discussed before and after the implementation of the OnVaCT intervention in communication about potential participation in early phase clinical cancer trials. Our analysis of consultations shows that similar values arose in the pre-intervention and post-intervention phase, i.e. hope, perseverance, quality or quantity of life, risk tolerance, trust in the healthcare system or professional, autonomy, social adherence, altruism, corporeality, accepting one’s fate, and humanity, with the difference that they were all discussed in context with each other in the post-intervention phase (RQ1). However, the way these values were discussed (i.e. the communication patterns) appeared to differ between the study phases (RQ2). In the pre-intervention phase, patient values were mainly found in short, spontaneous responses of patients to information or explanations provided by the oncologist. In the post-intervention phase, besides similar spontaneous responses, we also found a dialogue between patients and oncologists as well as a concrete weighing of values – often in relation to the OnVaCT. These longer phrases including several values (and their weighing against each other) may explain why almost all values appeared to be connected post-intervention. The implementation of the OnVaCT intervention appears to have changed the doctor-patient consultation both from the perspective of the patient, who talked extensively about values in long paragraphs, as well as from that of the oncologist, who asked concrete (follow-up) questions to explore individual patients’ values and their weighing. These questions may have potentially stimulated patients to tell in-depth about their values.

The results of this study seem to suggest that the intervention may have led to changes in the communication patterns. This would align well with previous studies on decision aids and communication skills trainings. For instance, decision aids (particularly those containing value clarification methods) had shown before to improve e.g. doctor-patient communication and congruence between patients’ informed values and their chosen options [10, 11]. Moreover, communication skills trainings for oncologists can positively impact communication behaviour [39] and SDM [40]. Specifically, assessing patients’ preferences for participation in a phase I clinical trial (e.g. ‘helping other cancer patients by participating in research’, ‘avoiding spending lots of hours in the clinic’) had shown to impact the strength of oncologists’ recommendations [41]. With regards to the OnVaCT intervention, we applied a synergetic approach [42, 43] in which the two different components of the intervention (i.e. the OnVaCT for patients and communication training for oncologists) are intended to enhance each other.

Previous studies have indicated that the discussion of patient values in current standard practice is often limited [19, 27, 28, 44, 45]. Our study is the first to provide a possible nuance to this claim. Already in the pre-intervention phase, we found that several, diverse patient values were (shortly) referred to during the consultation, for instance in response to general-inquiry questions [46,47,48] from oncologists at the beginning of the consultation (e.g. “how can I help you today?”). Such open-ended questions may encourage patients to “respond in their own terms and permitting the emergence of narratives based in lifeworld experience” [49], thereby exceeding a merely medical-technical discussion. This was also seen in the spontaneous responses from the patient, e.g. during history-taking. In that sense, when patients’ responses reflected their values, they appeared to go beyond the (medical-technical) agenda set by the oncologist [49]. However, oncologists often ignored patients’ explicit mentions or implicit cues [50, 51] on their values by continuing with history-taking questions or information provision, thereby non-explicitly reducing the space for further disclosure [50]. Still, oncologists addressed patient values when they described two opposite examples of what other patients in a similar situation might value. This is an example of ‘detached footing’ [52,53,54,55], (e.g. [56]) which may have enabled the oncologist to discuss this difficult topic while keeping a distance, i.e. without ‘labelling’ the patient to have certain values. This way of ‘option-listing’ by the oncologist could help create a moment of choice for their patients [57]. However, rather than asking follow-up questions and starting a dialogue about patients’ values, oncologists in the pre-intervention phase often recommended patients to further reflect at home. By doing so, oncologists explicitly reduced the space for further disclosure [50] of patient values.

Although oncologists seemed to acknowledge the importance of patient values in this complex decision and of following their personal wishes, an actual conversation on patient values was left to the patients and their proxies (e.g. relatives or general practitioner). This approach seems to align best with an ‘informative model’ in which patients are expected to know their personal values, and physicians only provide them with medical-technical information [29] and then offer them the choice to participate in an early phase clinical trial or not (i.e. the ‘logic of choice’ [58]). Still, in SDM, oncologists may be particularly equipped to discuss patient values: who would be better suited to help put the complex information into perspective than those specialized in explaining and conducting such trials, and who saw many patients in their practice? Yet, this required a new approach of the consultation in which oncologists try to understand these patients and discuss what matters to them.

With the OnVaCT intervention, we aimed to facilitate a discussion of patient values between patients and oncologists. The current analysis provides a first indication that the intervention may indeed offer such support. Based on the oncologists’ behaviour, the intervention may have helped them to become more sensitive to patient values, as they asked (follow-up) questions regarding patients’ personal wishes and beliefs. For instance, by asking patients what matters to them. By addressing the identification of patient values and goals of care [59] as a first step of SDM in line with the communication training, oncologists explicitly provided space for further disclosure of patient values [50]. With these probing questions, the oncologists seemed to apply an ‘interpretative model’ [29] in the post-intervention phase, thereby taking on the role of counsellor or advisor to help patients articulate and weigh their values. Moreover, by helping patients to find the right balance in their values, the oncologists may be better able to ‘care’ for these patients (i.e. the ‘logic of care’ [58]) as this approach enables the integration of all aspects of patients and their life with this life-limiting disease. When patients were asked by the oncologists what is important to them (with and without referring to the OnVaCT), most patients were able to explain their values. This even applied when patients did not use or recognize themselves in the OnVaCT narratives: patients who did not use it responded to narratives told by the oncologist, and patients who did not recognize themselves were able to indicate why this was the case and what they personally valued (i.e. what they missed in the OnVaCT). This aligns well with a narrative approach, according to which choices are “part of an effort to live a life that has coherence and meaning” [60]. In other words, decisions will likely have to suit patients’ personal life (stories). By using extended sequences to talk about their values, patients themselves seemed to be telling more of a story on their values too.

When comparing our results to previous studies in the OnVaCT project, there are mainly similarities: nearly all values and (sub-)themes from the systematic review [24] and interview study [26] occurred in clinical practice. Furthermore, similar to the interview study, we found that hope and perseverance were closely related. Regardless, this analysis added two new subthemes (i.e. ‘wanting to avoid inactivity’ under perseverance, and ‘(un)safety of the treatment’ under risk tolerance) to the existing set of values. A more substantial difference was that we did not find any occurrence of the subthemes ‘religious faith and/or being guided by God’ and ‘spiritual faith’ under accepting one’s fate. Besides a potential result of sampling, this could be explained by the fact that religious involvement in the Netherlands has declined over the past decade [61]. Nonetheless, religious or spiritual faith may still be important for at least a substantial minority of patients, which is why we have kept them as subthemes in our codebook.

Limitations

Although this study offers new insights into the discussion of patient values surrounding early phase clinical trial participation, our findings know some limitations. With the current qualitative content analysis and the study design (i.e. nonrandomized controlled trial), we cannot unquestionably prove that certain observed changes in communication patterns between the study periods are significant, nor that they are a direct result of our intervention. As our findings may have been susceptible to bias [62], they should be interpreted with caution. However, when conducted well, nonrandomized controlled trials can find similar outcomes as randomized controlled trials [63, 64]. The current analysis is, to the best of our knowledge, the first to examine patient values in communication on potential participation in early phase clinical trials, thereby providing initial insights into the clinical practice. In a next part of the OnVaCT project, we will take a closer look into specific SDM steps (including the discussion of patients’ preferences and goals of care) [59, 65], in order to further (quantitatively) explore the relationship of the OnVaCT intervention with SDM and decisional conflict [66].

Furthermore, we did not conduct a discourse analysis [see e.g. 67] nor narrative analysis [see e.g. 68]. Such analyses may be able to build on our findings, for instance by examining what and how language (e.g. in terms of vocabulary, grammar or behaviour) and stories (e.g. types of narratives) are used in this context. The existing discourse and conversation analytic literature that have been referred to above (i.e. [46, 47, 49,50,51,52,53,54,55, 57]) may provide inspiration for future analyses. Additionally, topic modelling [69] could be used as a means to verify the suggested links between (phrases of) values.

Lastly, all patients in this study were included in Dutch hospitals and were of Dutch nationality, thereby representing a ‘western world’-perspective. This points to the potentially limited applicability of our findings, since patient values can differ between individuals, countries or cultures. Additionally, we do not know whether maximum variability was achieved, for example in terms of patients’ religious and spiritual beliefs, as we did not include data on this. We still included patients and oncologists from several hospitals across the Netherlands, and aimed for sufficient spread in terms of patients’ decision, age and tumour type to substantiate the diversity of our sample, although only a small proportion of all available consultations were qualitatively analysed. Besides, for some oncologists, there were only limited available recordings in the pre-intervention and/or post-intervention phase. Although we only included oncologists with at least one available recording in both phases, this may have caused some bias in our results based on oncologists’ personal approach of these consultations. Additionally, only the main and/or clearest relations between values were indicated. Notwithstanding the insights these visualisations provide, for instance into which other values patients may be thinking about when they indicate a certain value, some relations may have been missed.

Implications for practice

Notwithstanding the limitations described above, we believe that the OnVaCT intervention may be of practical relevance for healthcare providers and patients facing this difficult decision together. For instance, the communication training could be used in the education of oncological specializations, such as for oncologists-in-training or research nurses, to help them improve the discussion of patient values as part of SDM for potential participation in early phase clinical trials. Besides, Dutch referring hospitals as well as specialized early phase clinical trial units can (continue to) share the OnVaCT with their patients to help them prepare to clarify their values during the upcoming SDM process. In order to implement the OnVaCT in other countries, sometimes cultural adaptation (especially of the values and narratives) may be required in addition to language translation. Meanwhile, it is important to note that it remains unclear which (or both) of the two intervention parts was responsible for the observed differences between the pre-intervention and post-intervention phase, i.e. what the effective ingredients were. Yet, with the complex SDM process in mind, we are confident that an intervention aimed at patients and oncologists was the only appropriate approach to support both parties in their shared process. For the sake of transparency/open science as well as to enable others to implement the OnVaCT intervention in clinical practice, an elaborate description of the intervention (i.e. a link to the OnVaCT and the documents used in the communication training) has been made available online [35].

Conclusions

This study has indicated the potential of the OnVaCT intervention (consisting of the OnVaCT for patients and communication training for oncologists) to support the integration of patient values into SDM consultations. In particular, this qualitative analysis has pointed to the different patterns surrounding patient values in communication on potential participation in early phase clinical trials. Besides, the current study provides nuance to previous claims regarding the limited discussion of patient values in oncology practice. All identified patient values were referred to already pre-intervention, with the difference that these values were all discussed in context with each other post-intervention. Furthermore, in the pre-intervention phase, most values were only found in short expressions of patients, e.g. in response to information from the oncologist, while the post-intervention phase again showed different communication patterns, with many oncologists referring to the OnVaCT and asking follow-up questions, and patients using phrases in which multiple values occurred. This analysis thus seems to provide qualitative evidence for the claim that the OnVaCT intervention can support the integration of patient values in consultations for early phase clinical trial participation.