Introduction

Ethical dilemmas in qualitative research are pervasive and demand ongoing reflection and resolution throughout the research process (Taquette & Borges Da Matta Souza, 2022). Given that all participants in research possess some degree of vulnerability (Horowitz et al., 2002), all research requires safeguarding their well-being and upholding core ethical principles. Following Horowitz et al. (2002), vulnerability in research can be understood as a heightened risk of harm or reduced capacity to protect one’s own interests, arising from personal circumstances, situational factors, or the research context itself. Research involving individuals in vulnerable situations, such as those affected by violence or mental health challenges, poses particularly complex ethical demands (Bredal et al., 2024; Findley et al., 2024; O’Brien et al., 2022), necessitating robust protective measures (Block et al., 2013; Downes et al., 2014). Vulnerability should be considered situational and research-induced, not merely a fixed state or set of defined traits (Bracken-Roche et al., 2017; Findley et al., 2024; Gordon, 2020).

Post-World War II ethical frameworks, notably the Nuremberg Code (1947) and the Declaration of Helsinki (World Medical Association, 2025), established voluntary informed consent and participant rights as foundational principles. The Belmont Report (1979) further articulated the principles of respect for persons, beneficence, and justice, principles especially vital in research involving vulnerable populations (Watts et al., 2001). However, while these frameworks provide essential ethical foundations, they offer limited guidance for navigating ethical dilemmas that arise from relational, contextual, and processual dynamics that unfold during ongoing qualitative research.

Guillemin and Gillam (2004) extend this discussion by distinguishing between procedural ethics, which encompasses formal approvals, protocols, and safeguards established prior to the study, and ethics in practice, which involves navigating ethically important moments that arise during the research process itself. They highlight that this distinction is particularly relevant in qualitative research conducted in real-world, relational settings, where ethical challenges often emerge unpredictably and require situated judgment rather than rule-based solutions. Guillemin and Gillam (2004, p. 265) describe these moments as indeterminate situations in which “the moment of response is an ethically important moment for there is the possibility that a wrong could be done”.

While ethical risks must be carefully considered, research shows that participants in vulnerable situations often may benefit from participation when safeguards are in place (Alexander, 2010; Alexander et al., 2018). Excluding such individuals from research may be paternalistic and marginalizing, undermining autonomy and silencing experiences that are central to improving knowledge and practice (Alexander et al., 2018; Fisher, 2012; World Health Organization, 2016). Including participants from vulnerable contexts can therefore provide ethically and scientifically valuable insights, particularly when research designs are attentive to relational dynamics and participant safety (Bredal et al., 2024; Rojas-Rozo et al., 2024).

Research on violence, particularly in intimate partner and family contexts, presents distinct ethical and methodological challenges (Berry, 2009; Gelles, 2017; Sobočan et al., 2023; Watts et al., 2001; World Health Organization, 2016). While ethics committees and such formal guidelines play an essential role, ethical practice in this field often relies on researchers’ situated judgment in response to unfolding, relational dynamics (Johnsson et al., 2014). More generic ethical frameworks may therefore offer limited guidance for navigating such complex situations, risking oversimplification (Nordentoft & Kappel, 2011). This points to a need for empirically grounded, context-sensitive knowledge that illuminates how ethical dilemmas are experienced, interpreted, and addressed in specific research designs and settings.

Positioning the Study and Identifying the Research Gap

To address this need, the study is situated within the Norwegian Family Counselling Services (NFCS), a publicly funded service that offers free couple and family therapy without referral and is characterized by low thresholds for access and early engagement in relational difficulties. NFCS operates 42 offices nationwide, employs approximately 500 therapists, and serves a large and diverse client population, with 50,637 new cases registered in 2024 (Norwegian Directorate for Children, Youth and Family Affairs, 2024).

This article reports on an interlinked companion study conducted within a multiperspective qualitative design examining ongoing couple therapy in cases where violence became a central concern (Snellingen et al., 2025). The related article explores clients’ and therapists’ experiences and perspectives on how intimate partner violence (IPV) becomes the main focus of conjoint therapy in cases where it was not initially disclosed. In contrast, the present study examines the ethical dimensions of participation in the research process itself.

While the two studies are closely interrelated and draw on the same research setting and participants, they address analytically distinct research questions and draw on different segments of the interview material. This analytic distinction was reflected in the interview design, in which questions for each study were organized into clearly demarcated sections within a single interview.

Against this backdrop, the study engages with a growing body of literature on research ethics in IPV by empirically examining ethical challenges as they unfold within a multiperspective research design conducted alongside ongoing couple therapy.

Within the broader literature on research ethics in IPV, ethical challenges are extensively documented, particularly with regard to participant safety, disclosure of violence, emotional distress, and informed consent (Berry, 2009; Watts et al., 2001; World Health Organization, 2016). Within this body of literature, a growing number of studies have examined ethical and methodological issues associated with various research designs, including interviews with clients or therapists and dyadic interviews with both partners. This work has highlighted challenges such as protecting internal confidentiality, managing safe recruitment, navigating power asymmetries between partners, and handling the analytic and reporting implications of interdependent accounts (Eisikovits & Koren, 2010; Ummel & Achille, 2016). Research has further shown that dyadic interview formats, whether joint or separate, may shape what participants disclose, how they position themselves relationally, and how sensitive information is produced and represented in research (Love et al., 2021).

However, far less research has examined ethical dilemmas and challenges associated with research designs that move beyond individual interviews with clients or therapists, or dyadic interviews with both partners. This gap is particularly evident in multiperspective studies that include multiple, interrelated participants occupying different roles and relationships, and that are conducted alongside ongoing couple therapy where research and clinical processes unfold in parallel. Such designs introduce additional ethical complexity related to confidentiality, consent, safety, and role boundaries that cannot be adequately captured through individual or dyadic perspectives alone. Examining these dilemmas empirically within an institutional context, such as the NFCS, is therefore essential for developing ethically responsive research practices in this field.

Aim and Research Question

This study aims to empirically examine the ethical dilemmas that arise in multiperspective research on violence in couple therapy when such research is conducted alongside ongoing therapeutic processes, thereby providing practice-informed insights into a research context in which existing ethical guidelines often remain abstract and underspecified. In this article, a multiperspective approach refers to a qualitative design that gathers data from multiple parties involved in the same relational process (Larkin et al., 2019), explicitly including both partners in the couple and their therapist. While this design offers the potential for richer relational insight, it also introduces additional ethical challenges that are qualitatively distinct from those encountered in single- or dyadic-perspective studies.

Because these challenges can be understood as situational, relational, and, at times, research-induced rather than as fixed individual characteristics, there is a need for empirically grounded knowledge to support ethically sound decision-making within such designs. This article addresses that need by examining ethical dilemmas as they unfolded in practice in a multiperspective study of ongoing couple therapy in which violence became a central concern. The guiding research question for this study is:

What ethical dilemmas and challenges emerge when exploring the experiences of clients and their therapists in a multiperspective qualitative study conducted alongside ongoing couple therapy in which partner violence has become the central issue?

Ethical Considerations in this Study

Building on Guillemin and Gillam’s (2004) distinction between procedural ethics and ethics in practice, this study adhered to formal ethical protocols while remaining responsive to emergent ethical moments.

Procedural ethics were addressed through compliance with the Declaration of Helsinki and national data protection requirements, including review by the Norwegian Centre for Research Data (ref. 999687) and the organization’s Data Protection Officer. The Regional Committee for Medical and Health Research Ethics (South-East Norway) reviewed the project and determined that it fell outside the scope of the Act on Medical and Health Research (2021), thus not requiring formal approval (ref. 192736).

The ethical and legal context of the NFCS is particularly relevant to this study. Therapists in the NFCS are bound by professional confidentiality under the Family Counselling Act (1997). At the same time, Norwegian law establishes a duty to avert or report serious criminal acts in close relationships. According to Sect. 196 of the Norwegian Penal Code (2005), all adults are required to act if it appears certain or most likely that serious violence has occurred or will occur and can be averted. This duty applies selectively and prospectively, rather than to all instances of violence, and requires professional judgment regarding the imminence and severity of the harm. In addition, Sect. 13−2 of the Child Welfare Act (2021) establishes a mandatory duty for public authorities, including NFCS therapists, to notify child welfare services when there is reason to believe that a child is being subjected to maltreatment or serious neglect.

These legal and institutional conditions form the backdrop against which both therapy and research are conducted in the NFCS, and they shape how ethical dilemmas are encountered and managed in the practices of research.

Ethics in practice were therefore central to this qualitative, multiperspective study involving both partners and their therapist during ongoing therapy, where violence was a concern. In such settings, ethically important moments (Guillemin & Gillam, 2004) are likely to arise and cannot be fully anticipated through procedural planning alone. This made ongoing reflexivity and contextual judgment essential, as further elaborated in the Results and Discussion sections.

Methodology

For this study, we adopted a multiperspective design within the framework of Interpretative Phenomenological Analysis (IPA) (Larkin et al., 2019). While traditional IPA focuses on single-perspective, homogeneous samples to enable in-depth analysis (Smith et al., 2021), complex relational phenomena, such as couple therapy involving violence, may benefit from a broader and more systemic approach. By including both partners and their therapists in separate interviews, this approach aimed to generate a rich and nuanced understanding of the individual experiences and relational dynamics at play.

Multiperspective IPA design has been successfully applied to examine relational dynamics, professional identities, and challenges in healthcare and education settings (Borg Xuereb et al., 2016; Chowdhury & Winder, 2022; Mjøsund et al., 2017). Extending beyond single-perspective IPA, this approach allows for the examination of interrelated viewpoints while maintaining an idiographic commitment to lived experience (McInally & Gray-Brunton, 2021).

Recruitment Process

The sample consisted of eight clients from four couples, along with five therapists who were directly involved in their respective couple therapy cases.

Therapists were recruited via office leaders, who were contacted through email and video meetings. Upon receiving approval from these office leaders, researchers approached individual therapists, obtained consent, and held preparatory meetings to clarify recruitment procedures and address safety concerns. Given that the study was conducted as part of a public-sector PhD project within the NFCS, some degree of prior professional contact between the researchers and participating therapists was unavoidable. The first and second authors had previously interacted with most of the therapists through various professional development initiatives in the NFCS. This pre-existing familiarity, involving both roles and prior relationships, was explicitly acknowledged and discussed at the outset and conclusion of the therapist interviews.

Recruiting therapists was straightforward, but attracting clients took more effort. Therapists noted two main obstacles. First, many eligible couples were early in therapy, described as “too chaotic” for new elements like this research project. Additionally, severe violence in some cases made it difficult and unethical to discuss the study with the clients. The second challenge was obtaining informed consent from all individuals in the therapeutic system. Therapists found that “clients who have been victims of violence” were eager to participate, while “clients who used violent behavior” often hesitated or chose not to participate. These dynamics influenced which couples ultimately met the inclusion criteria.

Participating therapists first informed relevant clients about the study and inquired about whether they were interested in participating. Clients who expressed interest provided initial consent, which permitted the researchers to contact them with further information. Interviews were scheduled only if both partners in the couple and their therapist consented, reflecting the multiperspective design and safeguarding confidentiality. Participation was voluntary, and both oral and written information emphasized that declining or withdrawing would have no consequences for ongoing therapy. The consent materials also explained how data would be stored, anonymized, and used for research purposes. Prior to the interviews, participants were informed about confidentiality boundaries, the study’s safety protocol, and legal obligations related to child welfare notification and the duty to avert imminent serious criminal acts.

To support informed and ongoing consent, a staged consent process with four formal points for clarification and re-consent was established, alongside continuous attention to emerging questions or concerns:

1. Initial information and consent: Once inclusion criteria were met, the therapist introduced the study and obtained initial written consent, allowing the researcher to make contact.
2. Telephone clarification: Participants were offered a telephone consultation with the researcher to address questions or concerns prior to the interview.
3. Interview re-consent: On the interview day, the study was reintroduced, questions were addressed, and participants signed a second informed consent form.
4. Post-interview check-in: At the end of the interview, participants were invited to reflect on their experience and were given the opportunity to clarify or withdraw consent for use of the data.

The study employed three inclusion criteria: (1) no information about violence at intake or in referrals, (2) participants confirmed that violence emerged as an issue after the start of therapy, and (3) both partners in the couple and their therapist consented to participate.

Participants

To safeguard both internal confidentiality (between participants within the same therapy case) and external confidentiality, information in this section is handled with particular care. To reduce the risk of disclosure, demographic characteristics are reported only in aggregated form, and therapists are not linked to specific couples in the presentation of data.

The study included four heterosexual couples and the therapists involved in their ongoing couple therapy. All couples had children living at home (nine children in total) and were in long-term relationships averaging more than ten years. In one case, the couple was seen by two co-therapists, resulting in a total of five participating therapists. All therapists had formal training in couple and family therapy, specific experience with cases involving violence, and were members of their offices’ designated violence teams. Descriptive characteristics of clients and therapists are presented in Tables 1 and 2.

All participants reported episodes of less severe physical violence within the year prior to the interviews, including behaviors such as pinching, holding, or pushing. In three cases, participants also described, in their research interview, one incident each that, as researchers, we would classify as more severe forms of violence, occurring at different points in time, ranging from an earlier historical episode from a decade ago to more recent events.

Based on our assessment at the time of the interviews, none of the disclosed situations met the legal thresholds for activating safety protocols or mandatory reporting under Sect. 196 of the Norwegian Penal Code (2005). Similarly, there were no indications of child maltreatment, neglect, or risk to life or health that would have required notification under Sect. 13–2 of the Norwegian Child Welfare Act (2021).

Data Collection

Each partner and their therapist were interviewed separately to ensure confidentiality and promote open reflection. Although the interviews were conducted sequentially, we explicitly did not cross-reference or share information between them to ensure confidentiality. This approach limited participants’ opportunities to comment on or challenge each other’s accounts. The first and second authors conducted all interviews.

A semi-structured interview format was used to ensure consistency while allowing flexibility. Specific questions about research participation were included at the beginning and end of the interviews. Some examples of questions were: What motivated you to participate in this research interview? How was the recruitment and interview process,and how would you characterize your overall experience? Was there anything specific that arose during the interview that resonated with you or that you believe could be useful to incorporate into your therapy sessions? Interviews lasted between 1.5 and 2.5 h, were audio-recorded, and transcribed verbatim.

Data Analysis

Data were analyzed using an IPA framework, applied flexibly to accommodate the multiperspective design (Larkin et al., 2019; Smith et al., 2021). The multiperspective IPA process involves iterative, layered interpretation that begins with detailed analysis of each participant’s data, respecting IPA’s idiographic focus on personal experiences. These individual analyses were then synthesized to identify patterns, divergences, and intersections across perspectives in multiple stages:

The analysis followed nine iterative steps, serving as a framework to capture key elements and allow continual refinement and insight integration: (1) Exploring the Researcher’s Orientation and Potential Bias: Reflexive practices involved identifying personal biases and assumptions, with self-reflective interviews and diaries facilitating continuous introspection and adjustment. (2) Immersion in the Data and Exploratory Noting: Transcripts were read multiple times, with exploratory notes capturing initial thoughts and patterns before formal coding. (3) Multiple Readings with Different Emphases: Transcripts were analyzed from three perspectives: descriptive, linguistic, and interpretative, merging participants’ descriptions with researchers’ interpretations and insights. (4) Identifying,Highlighting,and Clustering Quotes: Key quotes were grouped by thematic relevance, illustrating theoretical and practical connections. (5) Constructing Experiential Statements: Each experiential statement sought to capture and describe participants’ lived experiences, maintaining fidelity to their accounts while identifying broader themes. (6) Analyzing Connections Between Participants in the Same Case: This within-case analysis explored the perspectives of couples and their therapists, considering roles, gender, and relational dynamics. (7) Searching for Connections Across Group Themes: Connections were examined across groups, identifying overarching themes relevant to different perspectives within groups (“clients who have used violence,” “clients who have been victims of violence,” and “therapists”) and through group comparisons contrasting clients’ and therapists’ experiences. (8) Ensuring Comprehensive Coverage: The final review of transcripts ensured that all significant themes were captured. (9) Connecting to Research Questions: We linked the experiences of clients and therapists from the broader study to this study’s research question, examining ethical dilemmas and challenges within a multiperspective research design with both partners in a couple and their therapist in ongoing couple therapy, where violence has become a central concern.

The first author led the initial analysis, while the second and third authors reviewed emerging themes and contributed to analytic refinement. This collaborative process enhanced rigor and trustworthiness by grounding interpretations in the data while incorporating multiple analytic perspectives (Smith et al., 2021).

Results

Drawing on interviews with both partners and therapists, the analysis identified four key themes that illuminate central ethical dilemmas in conducting multiperspective research alongside ongoing couple therapy: (1) When Understanding Accumulates: Confidentiality and Safety as Emerging Ethical Tensions, (2) Motivation,Recruitment,and Power in Interconnected Relationships, (3) Consent in Motion: An Iterative and Ongoing Process, and (4) Research Ripples: Reflexive Spillovers into Ongoing Therapy? Participant quotes are used throughout to ground the analysis in the lived experiences of clients and therapists.

Each theme is introduced with a brief analytic framing that outlines the ethical dynamics it captures and its analytic scope. While the themes are presented separately for clarity, they are closely interconnected and often surface simultaneously during the research process, reflecting the layered, relational, and evolving nature of ethical challenges in this context.

Theme 1: When Understanding Accumulates: Confidentiality and Safety as Emerging Ethical Tensions

This theme focuses on how ethical tensions related to confidentiality and safety sometimes emerged and intensified as understanding accumulated through separate interviews with interconnected participants. Rather than arising from a single disclosure or decision, these tensions developed over time as information was gathered from multiple participants within the same therapy case, while insights from each interview had to remain strictly non-communicating across perspectives.

Analytically, the theme centers on balancing researchers’ investigative curiosity with the need to maintain confidentiality and ensure participant safety, a dilemma that became increasingly salient as knowledge accumulated across interviews. In a multiperspective design, this dilemma is shaped not only by what participants disclose, but by how insights from separate interviews interact and compound to form an expanded understanding of the case. The theme comprises two interconnected sub-themes: Balancing Curiosity and Confidentiality and Managing Safety while Building an Accumulated Understanding. These dilemmas and challenges become especially clear in a multiperspective design when researchers must ensure that data from one participant is not shared with others involved in the same case. At the same time, the accumulation of understanding across interviews may reveal potential risks to participants’ safety and well-being that are not apparent within a single interview, thereby intensifying ethical tensions over time.

Sub-theme 1: Balancing Curiosity and Confidentiality

A central ethical dilemma in this study was how to balance researchers’ curiosity with the need to protect confidentiality. In a multiperspective design, researchers explore the same phenomenon from different perspectives. However, in accordance with the informed consent, we as researchers could not disclose information shared in one interview to participants in another. This raises questions about how to explore sensitive issues without revealing prior knowledge gained from previous interviews within the same case. For example, Sara (client) withheld details of violent episodes from her therapist but spoke openly during the research interview:

Sara: I’m a very honest person, but I could say that we haven’t described what happened. We’ve just touched on the fact that all these things are happening, but we haven’t talked about what’s actually happening.

Interviewer 1: You’re thinking about your three sessions with the therapist?

Sara: Yes, we haven’t talked about it. You know more than our therapist about the events because I haven’t wanted to revisit them, as they’re really irrelevant to us.

In the research interview, her disclosure generated both curiosity and constraint. It raised curiosity about how these events might shape each partner’s experiences and the therapist’s clinical approach. At the same time, it imposed constraints on how to pursue follow-up questions in subsequent interviews with her partner or therapist. Probing these issues, either directly or indirectly, risked breaching confidentiality and, in some instances, potentially compromising safety. This illustrates a dilemma unique to multiperspective designs, in which researchers must navigate interconnected narratives from multiple participants across separate interviews, whereas in single-perspective or dyadic designs (e.g., both partners in the same interview), such unintended disclosure is not an issue. A similar situation arose when a therapist, interviewed after we had spoken separately with both partners in the couple, reflected aloud on whether some of our questions pointed to something that the therapist might have overlooked in therapy:

Therapist: “When you’ve asked about that (if the therapist experiences that the clients can speak freely), I’ve started wondering, is there something I’ve missed, something that’s not being said? There are always stories that don’t get told, so how can I get to those? Right? Should I have individual conversations, or should I bring it up somehow and make it a topic in conjoint sessions?”

These examples illustrate a fundamental ethical challenge in multiperspective research. While seeking a more profound understanding by exploring a phenomenon from multiple viewpoints, researchers may, in sum, unintentionally refer to information gathered from other interviews, either explicitly or implicitly. Here, the dilemma was operational: how to ask analytically relevant questions while ensuring that neither the content nor the framing signaled prior knowledge on the part of another participant. At its core, this dilemma required us to continually balance the pursuit of insight with the obligation to protect trust, confidentiality, and participant safety.

Sub-theme 2: Managing Safety while Building an Accumulated Understanding

The examples above illustrate the ethical tension inherent in “balancing curiosity with confidentiality,” while simultaneously underscoring participant safety and welfare as the primary concerns in research ethics. In this study, no disclosures reached the level of imminent danger or compromised well-being that would have required the activation of formal safety protocols or mandatory reporting. Nevertheless, challenges arose when clients shared information with researchers that their therapists were not yet aware of, thereby exposing potential blind spots in ongoing risk assessments. Such gaps may partly reflect that most cases were still in the early stages of therapy, with only a few sessions completed. These moments underscore the ethical complexity involved and raise essential questions about how researchers should respond when their understanding exceeds that of the therapist(s), particularly when this may affect whether existing safety measures are sufficient and whether legal obligations to report are met.

As researchers conducting a multiperspective study, we engaged in an iterative inquiry process that alternated between interviews, gathered information from multiple perspectives, compared accounts, and developed an accumulated understanding of each case. This process at times raised concerns about participant welfare, prompting us to reflect on our role and responsibility in safeguarding participants even when disclosures did not meet the thresholds for mandatory reporting. In response, we chose to address certain of our concerns directly within the interviews. Toward the end of each interview, we invited participants into an open dialogue about their participation and whether there was anything they wished to bring forward in therapy. In some cases, we shared selected reflections, particularly those related to their safety and support needs.

One example illustrates this approach: a client described that her partner, in a heated argument a decade earlier, had threatened her life. While she reported no severe violence since, she acknowledged lingering effects, sometimes reliving the incident when triggered. Earlier in the interview, she had dismissed the need to share this with her therapist. We did not assess any imminent risk but viewed the information as potentially valuable to address in therapy, and we encouraged her to reflect on whether to raise it in an individual session.

Interviewer 1: First of all, thank you very much for the interview. Secondly, the last thing I want to mention is that this doesn’t have to occur in a joint session, but if you have an individual session with your therapist, I would suggest it, and of course, you can decide for yourself whether it feels right, because only you know what’s best for you. Perhaps mentioning that specific experience that sometimes triggers you and that you end up reliving. Not with the intention that your therapist will “confront” your partner, but because it might be important for your partner to understand that when he becomes angry, it’s not always the current situation that affects you most, but rather that it brings back those past images, leaving you overwhelmed…It may also be beneficial for the therapist to be aware of this. However, again, that’s something you must decide: “Does it feel right for me to bring this up with my therapist”, perhaps in a one-on-one session? What is the first thing that comes to mind when I say this?

A few similar situations emerged across interviews, requiring continuous ethical reflexivity that combined adherence to legal obligations with an ongoing commitment to safeguarding participant well-being.

Across these cases, the key analytic feature was that “accumulated understanding” from separate interviews could outpace the therapist’s current risk picture, creating a recurring need to assess whether and how to invite safety-relevant reflections back into therapy without breaching confidentiality or overstepping the researcher’s role.

Theme 2: Motivation, Recruitment, and Power in Interconnected Relationships

This theme focuses on how motivation to participate, recruitment processes, and power relations intersect in a multiperspective study of ongoing couple therapy addressing violence. In this context, aspects of participants’ motivation for participation appeared to be entangled with relational loyalties, expectations, and perceived obligations, complicating the concept of voluntariness. The theme encompasses two sub-themes: Within the Couple, which describes how relational dynamics between partners may influence decisions about participation, and Clients–Therapist–Researchers, which outlines the ethical implications of existing professional relationships and interactions between participants and researchers. Together, these sub-themes illustrate how recruitment and consent unfolded within a network of interconnected relationships rather than as discrete, individual acts.

Sub-theme 1: Within the Couple

In several instances, the partner who had been exposed to violence, usually the woman in this study, expressed greater motivation to participate. Their motivation often reflected a hope that the research might benefit other couples facing similar challenges. Tove (client) expressed altruistic reasons for participating: “The work they do here is very, very important and valuable…and if the study that my partner and I are part of now can help others who are struggling, then I want to be part of it. So, that’s it.”

In contrast, some partners who had used violence, primarily the men in this study, participated voluntarily but appeared less motivated to engage fully in the study. Although they stressed that their participation was voluntary, their involvement at times seemed driven more by a desire to accommodate their partner rather than by their own engagement with the research aims, as illustrated below.

Interviewer 2: How was it for you to be asked by your therapist to participate in this research project?

Per: Uh, no, it was utterly unproblematic.

Interviewer 2: What made you say yes?

Per: Well, it was mainly because it allows the public to look at all aspects and learn something, that’s at least what my wife was thinking.

Several women described therapy as their last effort to save the relationship, sometimes framing continued participation as contingent on their partner’s engagement. Sara explained:

Sara: Continue here? He knows that if he doesn’t continue therapy, there’s no future for him and me. So, he’s in a very pressured situation, but it’s also important to me that he does it because he feels that it is necessary for himself, that he needs it.

Although couple therapy and research interviews are distinct processes, such statements raise ethical concerns about the genuineness of a partner’s consent to participate in research, especially when participation appears intertwined with relational ultimatums linked to continuing therapy. Analytically, this sub-theme highlights how consent decisions are also embedded in ongoing relational negotiations rather than made as isolated individual choices. In addition, participants’ stance on informed consent could shift over time, requiring ongoing attention.

In one interview, when uncertainty arose, we paused to address the issue and verify informed consent. Once confirmed, we also suggested a check-in at the end of the interview.

Interviewer 1: Okay, was there anything else you were wondering about?

X: No.

Interviewer 1: Alright, then we’ll continue, and we can talk about it […] at the end of the interview. Does that sound okay?

X: Yes.

Across these cases, uncertainty about participation was addressed situationally, with consent treated as something to be revisited in response to emerging relational pressures.

Sub-theme 2: Clients–Therapist–Researchers

Another layer of complexity arose during the recruitment process. Clients were invited to participate by their therapists, some of whom had professional connections with the researchers. Some clients described their therapist’s recommendation as influential in their decision to participate, as illustrated below:

Interviewer 1: I’m curious now, when you say that you feel discomfort talking about this, yet here you are, sitting with us and sharing?

Anna: Yes, I know. But the therapist was kind of like, “I think this could be a good idea” (laughs). So, I trust my therapist, and we want things to get better. I want things to get better for myself and for him. If this (participating in the research study) can help me, my partner, and our relationship, then those are selfish reasons, I guess, yes, mm (laughs).

This raises questions about whether some clients agreed partly to accommodate their therapist, potentially prioritizing relational loyalty over their own comfort. Therapists, in turn, may have felt implicit pressure to recruit clients because of their professional relationships with the researchers. What distinguishes this sub-theme analytically is the triangulated nature of influence, in which consent was shaped not only by dyadic therapist–client relations but also, to some degree, by therapists’ perceived obligations and interconnections with the researchers.

These dynamics were addressed explicitly in the interviews, with attention to participants’ motivations and experiences of being asked to participate. While no participant described these dynamics as overtly problematic, making them explicit and creating space to reflect on them supported a more informed and transparent consent process.

Theme 3: Consent in Motion: An Iterative and Ongoing Process

This theme conceptualizes informed consent as an ongoing and iterative process rather than a one-time event. Analytically, it captures how consent in a multiperspective design remained in motion, as participants’ understanding of the study shifted over time in relation to experiences within the research encounter and in their broader personal and therapeutic contexts.

In this study, consent was revisited at several points: therapists first introduced the study and obtained initial consent; clients were then offered a telephone consultation; on the interview day, informed consent was reconfirmed; and finally, a post-interview check-in provided participants with an opportunity to raise concerns or withdraw their consent at the end of the interview. However, participants’ sense of consent did not remain static across these stages. Uncertainty emerged at different points in the participation process, both as new issues surfaced during interviews and as participants’ feelings about, and interpretations of, their involvement evolved over time. This theme encompasses two perspectives: The Clients’ Perspectives and The Therapists’ Perspectives.

The Clients’ Perspectives

Many participants lived under stressful and uncertain circumstances, including the early phases of therapy, fears of further conflicts, concerns about children, and anxiety about their future and relationships. Being asked to participate sometimes intensified feelings of vulnerability or stigma, as Anna described: “I think it was terrible, you know, to be asked to be interviewed about violence or the fact that we are or have been physically. It’s painful to think about the situation we find ourselves in.” Despite this, she added, “I think it’s good to be able to contribute.”

The emotional strain associated with these circumstances raised questions about how much information participants could realistically absorb during the initial consent process. Some participants found it difficult to recall or fully grasp the study’s purpose and scope when they arrived for their interview, despite having received written information and preparatory conversations. Tim described agreeing with little reflection: “I said yes, from a point of view, well, I guess I could do that like, if, if someone wants us to do this, then yeah…I mean, I guess, I didn’t exactly know what to expect from this interview session.”

For some participants, uncertainty became more pronounced as the interview unfolded and touched on sensitive topics that had not been anticipated. Tim, for example, initially expressed no concerns but later raised questions about the potential consequences of disclosure: “What impact will arise from discussing these matters…how much should I disclose about what’s happened, you know? How could that impact our lives, the situation with the kids, and everything else?”

What distinguishes these accounts in a multiperspective context is that uncertainty was not only about participation per se, but about how disclosures made in one interview might relate to, or be taken up alongside, information provided by other participants in the same case. Because researchers could not know in advance which issues would become salient, consent could not be treated as fully settled at the outset.

Across interviews, consent was therefore managed as a situational and revisitable process, responsive to participants’ emerging concerns and evolving sense of what participation might entail in their specific context.

The Therapists’ Perspectives

Recruiting therapists also raised ethical questions, underscoring informed consent as a dynamic, iterative process shaped by professional roles and the relational proximity to the researchers. A recurring concern was whether the study aimed to evaluate therapeutic practice or function as an “effects study.” One therapist described how clarifying the exploratory nature of the study influenced her decision to participate. Lina (therapist): “I really appreciate how you’ve clarified earlier that this isn’t about evaluating how ‘good’ we are, but rather about exploring ways of working with this topic…”.

Some therapists also expressed uncertainty about how interview data, including sensitive client information, would be handled and who might have access to it:

Maria (therapist): Will anyone else, like your leader, my leader, or anyone else, be able to connect this back to me?

Interviewer 1: No, no one other than the two of us. Once this interview is anonymized and turned into text, it won’t be linked to any geography or person.

In a multiperspective design, therapists’ consent was shaped not only by their own participation but also by their responsibility toward clients whose interviews formed part of the same analytic case. This dual positioning meant that questions about confidentiality, data handling, and anonymity carried particular weight, as therapists considered both personal and professional implications. Across therapist interviews, consent emerged as something that had to be reaffirmed as the implications of multiperspective participation became clearer over time.

Viewed together, the experiences of clients and therapists highlight the importance of treating informed consent as a continuous process. Analytically, the findings show that in multiperspective research, consent is not only temporal (revisited over time) but also relational, as participants’ understanding of their involvement is shaped by how their interview relates to others within the same case and by how their understanding of implications evolved and became clearer over time.

Theme 4: Research Ripples: Reflexive Spillovers into Ongoing Therapy?

This theme explores how research interviews may generate subtle “ripples” that shape clients’ and therapists’ reflections and meaning-making, and that may subsequently be carried into ongoing therapy. Rather than positioning research participation as an intervention, the analysis attends to how interviews functioned as an additional reflective space alongside therapy.

Analytically, the theme captures how participation in the research sometimes prompted new reflections, articulations, or shifts in awareness that could resonate beyond the interview context. These influences were not uniform or predictable but emerged differently across participants and over time. To describe this complexity, the theme is presented from two perspectives: The Clients’ Perspectives and The Therapists’ Perspectives.

The Clients’ Perspectives

Research interviews often prompted clients to reflect on their situations, relationships, and future choices. The focused, open-ended questions, combined with the researcher’s attentive stance, helped some clients articulate thoughts that had previously remained unspoken in other contexts. Many described that the interview facilitated new awareness or connections that they perceived as potentially relevant for their therapy process. For instance, Sara initially expressed reluctance to bring past events into therapy: “Well, I do think it’s kind of important, yeah, but I’ve also said that I’d rather not talk too much about things that have happened because it’s really difficult in so many ways.” However, at the end of the interview, when asked whether any of the topics discussed had raised new awareness or could be relevant to bring into future conjoint sessions, she reflected: “Maybe putting a bit more focus on the tough and difficult things, so that eventually, we can talk about them constructively.”

In some interviews, clients not only articulated issues they hoped to address in therapy but also appeared more engaged and reflective toward the end of the interview. Per, who had used violence, described how the interview prompted self-reflection:

I think you always learn something from everything, but something that stands out to me immediately, the most important thing for me is realizing, is that I need to be more open to my partner’s experiences and really meet her where she is. She probably feels like she’s not being understood in the way she needs, even though I say I understand. She likely doesn’t feel that I truly do.

Across client interviews, these “ripples” were characterized by shifts in reflection and may indicate that research interviews, to some degree, shape how clients approach therapy by potentially foregrounding and bringing new issues or perspectives into therapeutic conversations.

The Therapists’ Perspectives

For therapists, the research interviews provided a space to step back from clinical work and reflect on ongoing cases. Several therapists described the interviews as resembling “case supervision,” offering an opportunity to re-evaluate their approaches, question assumptions, and consider alternative ways of working.

It’s always helpful to talk through a process because it makes you more aware of things, so I think it has opened up some new perspectives. Since we’re two therapists in this case, we can constantly talk things over, and we write notes together to make sure we document everything properly. But I think I will become a bit clearer about that. I wish to start working with them individually sooner than we originally planned, even though they prefer joint sessions. Yeah, just to check things out from a safety perspective.

In most cases, therapists handled the cases independently, without a co-therapist. Several described the research interview as providing reflective space that was otherwise difficult to access in everyday practice, particularly in complex cases involving violence. Sunniva reflected:

I think, first and foremost, it’s great to have the opportunity to reflect on cases like this because they’re so challenging. Honestly, I wish I could have interviews like this for all these kinds of challenging cases more often because they offer such a valuable chance for reflection. It’s a reality that there are so many perspectives to consider, and trying to approach them clearly and balanced can feel overwhelming…that’s why this interview provides such an important opportunity for reflection.

Analytically, therapists’ accounts emphasize how research interviews served as a separate reflective space, connected to but distinct from clinical practice. These reflections did not constitute supervision or guidance, yet they nevertheless influenced therapists’ thinking about safety, pacing, and therapeutic structure.

Across perspectives, the key analytic feature of this theme was that research interviews created reflective “ripples” that participants carried back into therapy, highlighting the interconnectedness between research and clinical processes in multiperspective studies conducted alongside ongoing treatment.

Discussion

Conducting qualitative, multiperspective research in the context of ongoing couple therapy involving violence presents ethical complexities that often extend beyond and may encompass aspects different from those encountered in single- or dyadic design studies. By including both partners and their therapist, the design not only creates potential for richer insights but also introduces different forms of vulnerability, as participants’ perspectives are inherently interconnected. The four themes: (1) When Understanding Accumulates: Confidentiality and Safety as Emerging Ethical Tensions, (2) Motivation,Recruitment,and Power in Interconnected Relationships, (3) Consent in Motion: An Iterative and Ongoing Process, and (4) Research Ripples: Reflexive Spillovers into Ongoing Therapy? highlight two overarching dimensions relevant to ethical challenges in this context.

First, the findings underscore the relational and systemic nature of ethical dilemmas in multiperspective research, involving multiple stakeholders embedded in overlapping relationships and power dynamics. Second, the findings indicate that ethical challenges emerge dynamically during the research process and cannot be fully anticipated through procedural review alone. These challenges can be understood as ethical “branching points” that require researchers to move from ethical sensitivity toward context-aware action. Therefore, the discussion focuses on (1) relational and systemic dimensions of ethical practice in multiperspective research, and (2) how researchers may navigate ethical branching points through context-aware action.

Relational Ethics and a Systemic Approach in Multiperspective Research

Relational Ethics: Responsibility and Dialogue

In qualitative, multiperspective studies involving ongoing therapy, relational dynamics between researchers, clients, and therapists become particularly intricate. As Guillemin and Gillam (2004) distinguish, research ethics encompasses both procedural ethics and ethics in practice. While both are essential, our findings primarily speak to ethics in practice, where unforeseen, ethically charged situations arise during data collection.

Relational ethics provides a useful lens for navigating such situations by framing ethical decision-making as negotiated, ongoing, and context-sensitive. For a broader, in-depth elaboration, see Ellis (2007), Fisher (2012), and McNamee (2015). We draw on relational ethics to interpret moments in the study where ethical responsibility emerged through interaction, dialogue, and responsiveness to participants’ evolving needs. For example, when participants expressed distress during interviews, researchers paused to check well-being, invited feedback, and reopened dialogue about participation and support needs. Such moments illustrate how ethical responsibility extended beyond data collection to sustaining participant agency and safety within the research encounter. At times, this also involved revisiting informed consent as participants’ experiences and understanding evolved throughout the research process.

Similarly, when concerns arose about participant safety and well-being that did not meet the legal threshold for mandatory action, researchers remained ethically engaged by inviting reflection and dialogue rather than treating responsibility as exhausted by procedural compliance. These findings highlight relational ethics as an orientation toward sustained ethical engagement rather than discrete ethical acts.

Systemic Approach: Context, Vulnerability, and Boundaries

While relational ethics emphasizes the relational aspects of ethical practice, our results suggest that a systemic approach can usefully expand and complement the scope to include the multiple systems and contexts that shape interactions. As von Bertalanffy (1968) argued, systems cannot be understood solely from their individual parts; subsystems are nested within larger wholes, and new properties emerge from their interactions with the surrounding environment. Adopting a systemic approach to research ethics encourages reflection on interconnectedness rather than treating issues in isolation (Gordon, 2020).

In this study, clients, therapists, couples, and researchers were embedded in nested systems that shaped how ethical decisions reverberated across roles and relationships. Ethical actions directed toward one participant often had implications for others, underscoring that ethical decisions rarely remain isolated in multiperspective designs. This perspective also reframes vulnerability as dynamic and relational rather than as a fixed individual characteristic (Bracken-Roche et al., 2017; Delor & Hubert, 2000). Across the themes, vulnerability shifted in relation to role, context, and timing, underscoring the need for sustained, process-sensitive ethical attentiveness rather than solely reliance on static or pre-defined categorizations of vulnerability.

From a systemic perspective, the “ripples” between research interviews and therapy observed in this study can be understood as part of a broader ecology of influence rather than as unintended interference. Research interviews thus became one element within a wider network of interactions shaping therapeutic processes. The ethical challenge, therefore, was not whether research influenced therapy, but how and to what degree such influence could be navigated reflexively and transparently. This should not be read as positioning research as a therapeutic intervention, but as acknowledging influence as an inevitable condition requiring ethical navigation.

Similarly, balancing researcher curiosity with confidentiality across interviews requires ongoing boundary awareness. Viewing confidentiality systemically, across participants within the same case, broadened ethical sensitivity and supported context-aware action when established agreements were tested in practice.

From Ethical Sensitivity to Context-Aware Action in Ongoing Research

The second overarching dimension concerns how ethical sensitivity was translated into context-aware action. Based on the findings, this process involved (1) recognizing ethical branching points, (2) interpreting and forming an impression of these through reflective and dialogical processes, and (3) mobilizing resources to respond with context-aware actions.

Recognizing Ethical Branching Points

Ethical dilemmas arose at multiple stages of the research process. The concept of ethical “branching points” builds on Guillemin and Gillam’s (2004) notion of ethically important moments, emphasizing junctures where multiple courses of action are possible, and decisions carry relational consequences. In this study, ‘ethical branching points’ is used as a heuristic to organize ethical sense-making, rather than as a comprehensive model of ethical decision-making.

Such branching points were not always immediately recognizable. They often emerged as uncertainty or unease and became ethically salient only as their consequences unfolded over time. Examples included disclosures unknown to therapists (Theme 1), relational pressures shaping consent (Theme 2), and evolving uncertainties about participation (Theme 3). These findings underscore the need for sustained ethical sensitivity throughout the research process.

Interpreting and Forming an Impression Through Reflective and Dialogical Processes

Navigating ethical branching points required interpretation rather than rule application. This process parallels clinical sense-making, in which practitioners interpret complex, indeterminate situations through reflection and dialogue (Schön, 2017; Shön & Rein, 1994; Snellingen et al., 2024).

Drawing on Schön and Rein (1994, s. 24), ethical dilemmas in qualitative research rarely appear as clear-cut problems, but rather as messy, indeterminate situations that require researchers to distinguish, name, and frame aspects of a situation in order to form an initial impression.

From a Gadamerian hermeneutic perspective, researchers cannot approach ethical dilemmas as neutral observers. Instead, they interpret branching points through their prejudices and horizons of understanding (Gadamer, 1998). Gadamer argues that prejudice is not an obstacle, but rather a necessary precondition for understanding. Our socially and historically situated horizon not only enables sense-making but also directs how we respond in practice. Although impressions may evolve, they provide a provisional orientation for ethical action. This, in turn, requires ongoing reflexivity, a sustained examination of assumptions, biases, and responses (Finlay, 2012), supported by dialogue with participants, colleagues, and stakeholders to co-create context-sensitive ethical judgments in complex, evolving situations.

Mobilizing Resources to Respond with Context-Aware Action

Recognizing and forming an impression of ethical branching points should not merely invite reflection but also lead to responsive, context-aware action. These moments demand engagement, echoing Biesta’s (2015) notion that some situations “call us” to ethical responsiveness. When participants expressed distress or uncertainty, researchers paused, clarified, and engaged in dialogue about participation and support needs. These actions were situational and relational rather than procedural, drawing on ethical sensitivity, systemic awareness, and tolerance for uncertainty.

Mobilizing resources requires not only ethical knowledge but the capacity to apply it in complex encounters. This highlights the importance of training and supervision that integrates ethical theory with lived dilemmas (Banks, 2016; Israel & Hay, 2006). Some of the ethical challenges in this multiperspective research design on couple violence often resemble “wicked problems” (Rittel & Webber, 1973): complex, value-laden, and lacking simple solutions. To address them responsibly, researchers may not only identify ethical branching points but also activate relational and systemic resources, ensuring that ethical sensitivity is consistently transformed into context-aware action.

Implications

Implications for Researchers

For researchers conducting multiperspective studies alongside ongoing therapy, the findings can highlight several practice-relevant implications:

• Ethical dilemmas often emerge from the accumulation of understanding across separate interviews rather than from any single interaction, requiring heightened vigilance regarding confidentiality and indirect disclosure.
• Informed consent should be treated as relational and revisitable and should not be understood as a single event but as an ongoing process that extends throughout the research process, particularly as participants’ understanding evolves and their stance may change over time. This underscores the importance of multiple, deliberately structured check-ins throughout the research process, as well as ongoing sensitivity.
• Ethical branching points may be subtle and emotionally charged; developing tolerance for uncertainty and dialogical competence is therefore central to ethical practice.
• Structured opportunities for reflection, supervision, and peer dialogue are essential to support researchers in translating ethical sensitivity into context-aware action.
• Where research is embedded in clinical services, ethical responsiveness should be supported not only at the individual level but also through institutional arrangements that provide time, consultation, and shared responsibility for ethical decision-making.

Implications for Clinicians

For clinicians who participate in multiperspective research together with their clients, the findings may point to several distinct considerations:

• Research participation may invite clients into additional reflection on experiences, concerns, or meanings that they may later consider bringing into therapy, underscoring how research interviews can intersect with clients’ ongoing sense-making and therapeutic processes.
• For therapists, awareness that research interviews can function as a parallel reflective space, without constituting clinical supervision, may support greater reflexivity and transparency when noticing shifts in focus, engagement, or emerging themes that they may later consider introducing into the ongoing therapeutic process.
• Clinicians may experience uncertainty about what research interviews entail for their clients, underscoring the importance of transparent communication about roles, boundaries, and confidentiality across research and clinical contexts.
• Clarifying expectations and maintaining open dialogue with researchers can support ethical collaboration while preserving clinical responsibility and autonomy.
• Beyond the immediate research context, the findings may also be clinically relevant when therapists temporarily or strategically divide naturally occurring systems (such as couples or families) into individual sessions, as is often done when violence becomes a concern. Attending to these dynamics may help clinicians remain mindful of how individual conversations interact with conjoint therapeutic work.

Reflections on Strengths, Limitations, and Future Research

By integrating client and therapist perspectives, this study offers a nuanced understanding of ethical challenges in multiperspective research conducted alongside ongoing couple therapy involving violence. Including multiple perspectives enhanced analytic depth and contextual sensitivity, allowing ethical dilemmas to be examined as relational and situational rather than as isolated methodological issues.

Several limitations warrant consideration. The sample was small and context-specific, potentially limiting the transferability of the findings. In line with the epistemological foundations of Interpretative Phenomenological Analysis, the aim of this study was not to make statistical generalizations but to offer an in-depth, idiographic account of how ethical dilemmas are experienced and navigated within a particular context (Smith et al., 2021). The findings should therefore be understood as situated and illustrative, with potential for theoretical or conceptual transferability rather than direct applicability to all therapeutic or research settings (Smith & Nizza, 2022). These ethical challenges may also differ in countries with other legal frameworks, welfare systems, professional mandates, or cultural norms surrounding therapy, research participation, and IPV.

Researchers’ dual roles within the service organization may also have influenced recruitment and disclosure, although participating therapists reported that this affiliation enhanced trust and openness. At the same time, this insider positioning raises important questions about the interconnection between therapy and research, particularly regarding how clients may have understood and navigated the boundaries between these roles, including whether they experienced implicit pressure to participate to maintain access to services. Although participation was clearly communicated as voluntary and separate from therapy and was followed by several check-ins, clients’ interpretations of voluntariness and role boundaries may nevertheless have shaped both the consent process and the disclosures made during interviews. This dual role may also have influenced how therapists and clients framed their accounts, given the researchers’ familiarity with the service context. Relatedly, selection bias remains a possibility, as participants who consented to take part may differ systematically from those who declined, particularly given that clients who used violent behavior were more likely to hesitate or opt out of participation.

An additional limitation concerns the potential influence of language and translation on meaning-making, which is particularly relevant given the study’s hermeneutic-phenomenological focus. All interviews were conducted in Norwegian, except one in English, and translation was approached as an interpretative and reflexive component of the analytic process. Collaborative translation reviews were used to address tone, context, and experiential nuance, with support from bilingual authors and a native English-speaking supervisor. Nevertheless, translation choices may have shaped how participants’ experiences and ethical dilemmas were represented and should be understood as part of the study’s interpretative layer.

Future research could further examine how ethical dilemmas and branching points unfold across different contexts and over time, including how participants themselves make sense of these moments as they evolve. Cross-cultural and cross-institutional studies may be particularly valuable for exploring how ethical sense-making, vulnerability, and researcher–participant relationships are shaped within different legal, organizational, and cultural frameworks. Longitudinal designs could also contribute to a deeper understanding of how ethical considerations develop as therapeutic and research processes intersect over time.

Conclusion

Taken together, this study underscores the importance of research ethics approaches that attend to the relational, systemic, and temporal dimensions of qualitative inquiry when exploring IPV through multiperspective designs conducted alongside ongoing therapy. The findings suggest that ethical challenges in such contexts should not be understood solely as isolated dilemmas or as issues that can be resolved through procedural safeguards alone. Instead, ethical challenges often take the form of emerging ethical branching points: interrelated concerns that unfold through participation, relationships, and the accumulation of knowledge over time, in which different courses of action carry distinct ethical implications. Such challenges call for ongoing ethical sensitivity and context-aware action.

By foregrounding how ethical dilemmas are lived, negotiated, and enacted within research conducted in parallel with ongoing therapeutic processes, the study contributes to a more practice-near and process-sensitive understanding of research ethics in qualitative multiperspective studies. Viewing ethical responsibility through this lens highlights ethical practice as a matter of continuous interpretation, dialogue, and judgment, rather than the application of predefined rules. This perspective invites continued critical reflection on how ethically robust research designs can be developed and sustained in contexts in which research and clinical work unfold side by side.