Summary findings

Firstly, the finding that 72.1% of clients lacked recorded interpreter status highlights a significant gap in capturing relevant data. However, this is not a problem limited to TTad as it is apparent that there are issues in recording interpreter need or use in mental health care assessments (Young et al., Reference Young, Vicary, Tipton, Rodríguez Vicente, Napier, Hulme and Espinoza2023) or maternity care (MacLellan et al., Reference MacLellan, McNiven and Kenyon2024). Moreover, this practice is contrary to existing guidance that recommends recording a patient’s language and interpreting needs in their healthcare record and passing this information on when referring them to other health professionals (Office for Health Improvement and Disparities, 2021).

Of those that reported, the analysis identified a statistically significant association between clients’ waiting times and the use of an interpreter, with those utilising an interpreter experiencing longer waits than those who did not. We also identified significant associations between using an interpreter and outcome measures.

Although approximately 80% of clients who utilised an interpreter were seen within the TTad target of 6 weeks, this is notably lower than the 92% for clients who did not utilise an interpreter. The disparity suggests a potential inequity in timely access to care for clients with language support needs, with concomitant inequities in treatment outcomes. This delay likely reflects the service and organisational level barriers to interpreter-mediated therapy within TTad, including challenges with booking interpreters, sourcing interpreters for specific languages, and interpreter non-attendance reported by Kunorubwe et al. (Reference Kunorubwe, O’Leary and Wynne2025). Such disparities may not be limited to TTad; as clients have shared, they regularly wait longer than average to access an appointment or even to arrange an appointment across healthcare services (Brennan and Seymour, Reference Brennan and Seymour2024).

The analysis also indicates that clients who used an interpreter had lower recovery rates than those who did not. This finding stands in contrast to previous research, which has generally shown that interpreter-mediated therapy can be effective when appropriately delivered (Quiroz Molinares, Reference Quiroz Molinares2024; d’Ardenne et al., Reference d’Ardenne, Ruaro, Cestari, Fakhoury and Priebe2007; Schulz et al., Reference Schulz, Resick, Huber and Griffin2006; Woodward et al., Reference Woodward, Orengo-Aguayo, Stewart and Rheingold2020). These earlier studies have reported comparable or in some cases improved outcomes for clients receiving interpreter-mediated therapy, particularly when interpreters are professionally trained and integrated effectively into the therapeutic process. However, a key difference may lie in the context and population studied. Many of the aforementioned studies were conducted within specialist services, where therapists often have greater experience working cross-culturally and are more accustomed to collaborating with interpreters (Kunorubwe, Reference Kunorubwe2025). In contrast, the findings presented here reflect interpreter-mediated therapy within standard TTad services, where the infrastructure, training, and time required to support such work may be more limited. Importantly, in many of the prior studies showing equal or improved outcomes, participants may have received timely access to treatment as part of the research, potentially minimising delays between referral and treatment initiation. In routine practice in TTad, however, delays associated with arranging interpreter-mediated therapy may themselves contribute to poorer outcomes.

A particularly interesting finding involved the very small subset of clients (0.03% of the sample) who used friends or family members as interpreters, who proportionally showed higher than expected improvement rates. With such a small sample, we need to be mindful interpreting such data. It is therefore entirely possible that this is a statistical artefact, but future research might explore whether there are scenarios in which family and friend interpreters could lead to improved outcomes for some clients in certain circumstances. The use of family as informal interpreters is typically not advised in therapy due to concerns about limiting disclosure, communication errors, and the impact of personal biases (Bauer and Alegría, Reference Bauer and Alegría2010; Costa, Reference Costa2022a; Hadziabdic et al., Reference Hadziabdic, Albin, Heikkilä and Hjelm2014; Leanza et al., Reference Leanza, Miklavcic, Boivin, Rosenberg, Kirmayer, Guzder and Rousseau2013; Tribe and Morrissey, Reference Tribe and Morrissey2004). With healthcare interactions where there may be issues around domestic violence, child protection, mental health or sexual matters, communicating through family members may be counterproductive at best and, at worst, dangerous (Fazil and Beavan, Reference Fazil and Beavan2015). This is also compounded when considering that clients have expressed how the lack of access meant that they were more likely to have to rely on family members or children across healthcare, reliance creates risks interpreting sensitive or complex medical contexts, and may lead to poorer patient safety, miscommunication, missed appointments, or inappropriate care (Brennan and Seymour, Reference Brennan and Seymour2024).

The lower recovery rates observed could reflect delays in starting therapy due to issues with waiting times, or may be related to barriers to interpreter-mediated therapy in TTad, such as challenges associated with a change in the traditional therapy dynamic, lack of therapist training in working with interpreters, time pressures due to inflexible caseloads, and a reluctance or resistance to interpreter-mediated therapy (Kunorubwe et al., Reference Kunorubwe, O’Leary and Wynne2025). More broadly, this might be linked to wider inequalities in outcomes and a general absence of co-produced models of culturally adapted care. The NHS Race and Health Observatory (National Collaborating Centre for Mental Health, 2023) reported that clients from minoritised ethnic groups often present with greater symptom severity at the point of assessment and wait longer for assessment and treatment. Ethnic minority status is also linked to multiple disadvantages, with some outcome disparities reflecting structural factors experienced more acutely in these groups (Amati et al., Reference Amati, Green, Kitchin, Watt, Jones, AlRubaye, McCann and Greenfield2023). Clients from minority ethnic groups are less likely to complete treatment and show improvement (Ahmad et al., Reference Ahmad, McManus, Cooper, Hatch and Das-Munshi2022; Baker and Kirk-Wade, Reference Baker and Kirk-Wade2024), a trend that appears to be compounded when the clients have limited English proficiency (Njeru et al., Reference Njeru, DeJesus, St Sauver, Rutten, Jacobson, Wilson and Wieland2016). It is plausible that there is some overlap between minority ethnic status and need for interpretation in this sample, and poorer outcomes might therefore be partly due to the factors listed above. These delays and poorer outcomes are consistent with earlier studies (Costa and Briggs, Reference Costa and Briggs2014; Harwood et al., Reference Harwood, Rhead, Chui, Bakolis, Connor, Gazard and Hatch2023), which noted that interpreter-related delays and barriers can lead to disengagement and reduced therapeutic effectiveness. This is also of importance as increased duration of the waiting time has been associated with less favourable treatment outcome (van Dijk et al., Reference van Dijk, Meijer, van den Boogaard, Spijker, Ruhé and Peeters2023) and predicts drop-out (Carter et al., Reference Carter, Pannekoek, Fursland, Allen, Lampard and Byrne2012) more generally. Longer waiting times, especially those over 3 months, were associated with declines in outcomes for patients (Reichert and Jacobs, Reference Reichert and Jacobs2018). A qualitative study exploring young adults’ experiences of waiting lists found that participants reported relying on alternative support, experiencing impaired functioning, and developing negative beliefs and emotions, highlighting the broad psychological and practical impact of treatment delays (Punton et al., Reference Punton, Dodd and McNeill2022); 182 clients were recorded as Interpreter Didn’t Attend. While it is unclear from the data whether these clients were subsequently able to access interpreter-mediated therapy or how their care was managed following this disruption, the presence of such a sizeable group underscores a known barrier to equitable access. This aligns with broader concerns raised in Kunorubwe et al. (Reference Kunorubwe, O’Leary and Wynne2025), which identified interpreter non-attendance as a recurring issue that disrupts care, delays treatment, contributes to therapists’ frustration and may lead to disengagement from services.

Finally, to provide some context, it is helpful to consider the use of interpreters across other areas of healthcare. However, data on interpreter need and usage across NHS settings is limited. Young et al. (Reference Young, Vicary, Tipton, Rodríguez Vicente, Napier, Hulme and Espinoza2023) noted that it is not known how many people undergo mental health assessment with the use of an interpreter. MacLellan et al. (Reference MacLellan, McNiven and Kenyon2024) found that within maternity trusts, an average of 9% of people required an interpreter, compared with Gill et al. (Reference Gill, Beavan, Calvert and Freemantle2011) who found in a Birmingham Trust that 29% of consultations involved the use of an interpreter. Therefore, more research is needed to understand rates of interpreter use.

Clinical implications

The clinical implications of these results highlight several key areas for improvement and consideration in service delivery for clients requiring interpreters in mental health and therapeutic contexts.

The finding that clients utilising an interpreter tend to experience longer waiting times than those not utilising one underscores a potential barrier to timely access to therapy. This delay could have significant implications for client outcomes, particularly in mental health services where timely intervention is critical, supported by the poorer outcomes seen in these data. Importantly, this dataset only includes wait times to the initial assessment appointment, not to the start of treatment. As the assessment is typically used to determine the appropriate therapeutic intervention, clients are often placed on additional waiting lists thereafter. As such, potential further delays including between treatment sessions are not captured. This is relevant as waiting time targets for treatment are seen as too long by service users, which exacerbates distress, and increases drop-out and disengagement from treatment (Punton et al., Reference Punton, Dodd and McNeill2022). Therefore, clinicians, supervisors, services, and strategic leadership need to consider expanding interpreter services or adjusting scheduling practices to reduce waiting times for those needing interpreters, thereby improving the overall efficiency and accessibility of services.

The significant associations between rates of reliable improvement and reliable recovery between clients’ use of interpreters point to potential disparities in treatment outcomes, with 58.5% of those with an interpreter showing reliable improvement compared with 69.2% of those without an interpreter, and only 32.5% of those with an interpreter showing reliable recovery compared with 44.5% of those without an interpreter. The analysis also revealed that clients using professional interpreters had lower rates of reliable recovery (28.8%) compared with those using friends/family (38.6%) or other persons (39%). Interestingly, the group using friends or family as interpreters showed higher improvement rates than expected, despite literature advising against this practice due to concerns over dual roles, potential bias, and increased communication errors. This highlights how clinicians, interpreters, interpreting agencies, TTad services and strategic leadership need to assess the quality of the interpretation process, including ensuring that interpreters are adequately trained in the nuances of mental health terminology and therapy-related conversations.

While the majority of clients using an interpreter were seen within 6 weeks, the data suggest that targeted interventions are needed to ensure that those who wait longer are not disproportionately affected. Clients in the 6–18 weeks and 18+ weeks categories may experience deterioration in mental health, exacerbating the difficulty of treatment. It is important to acknowledge, however, that data on clients’ initial symptom severity are not available, which limits conclusions about the true impact of waiting times on outcomes and reliable change from this sample.

Clinical teams might consider implementing strategies to address the longer waits for those who utilise interpreters, such as offering additional resources or adjusting how appointments are scheduled to ensure quicker access. However, for these strategies to be successful, they would require robust support and resourcing from strategic leadership, including adequate funding for interpreter services, sufficient staffing to manage increased demand, and organisational buy-in to prioritise equity in access to timely care. Effective implementation of these changes would rely on collaboration between clinical teams and leadership to ensure that adequate resources are allocated and that systems are put in place to meet the needs of clients requiring interpreters.

Providing therapy in a client’s primary language is critical, particularly in mental health contexts where emotional nuance, cultural meaning, and identity are deeply embedded in language. Language is not merely a conduit for communication; it plays a central role in how individuals access, process, and express emotional experiences. Therapy in a non-primary language can disrupt this process and impose an additional cognitive burden on clients, whereas using a first language supports more accurate self-expression, emotional regulation, and therapeutic engagement (Tannenbaum and Har, Reference Tannenbaum and Har2020).

The lack of recording of interpreter status for 72.1% of clients in the dataset (n=127,877) has significant clinical implications. When such a fundamental aspect of service access is not routinely recorded, it becomes difficult to assess whether clients’ language needs are being identified and met. This data gap raises concerns about the visibility of linguistically minoritised groups within services and suggests that language support may be under-recognised or inconsistently documented in routine practice. Issues with inconsistent recording of language and interpreter use are also found in other areas such as Mental Health Act Assessments (Young et al., Reference Young, Vicary, Tipton, Rodríguez Vicente, Napier, Hulme and Espinoza2023) and within Maternity Trusts (MacLellan et al., Reference MacLellan, McNiven and Kenyon2024).

Moreover, the absence of these data hampers efforts to monitor equity, evaluate outcomes accurately, and implement service improvements. Without reliable recording, it is impossible to determine whether interpreter services are being offered consistently, declined, unavailable, or simply overlooked. Importantly, acknowledging and addressing language-based discrimination in healthcare is critical to shifting the culture away from the widespread yet often unacknowledged acceptance of substandard care for linguistically diverse clients (Tomkow et al., Reference Tomkow, Prager, Drinkwater, Morris and Farrington2023). Improving data quality and embedding routine assessments of language needs are necessary steps towards ensuring that language is not a hidden barrier to accessing equitable and effective mental health care.

This is not just limited to recording of language, but also the recording and use of information on the dialect spoken, given that errors in interpreter booking due to dialect mismatches are common within TTad, leading to missed appointments, ineffective communication, and impact on continuity of care (Kunorubwe et al., Reference Kunorubwe, O’Leary and Wynne2025). Such errors not only prolong waiting times but also risk further disengagement, particularly for clients already experiencing long delays. By consistently recording and reporting on dialects, services could reduce inefficiencies, ensure more appropriate interpreter matching, and ultimately support stronger therapeutic alliances. Better recording would also enable a full language/dialect analysis across TTad services, ensuring an evidence-based provision of spoken word interpreters. This simple adjustment to recording practices could have significant implications for treatment accessibility, engagement, and outcomes for linguistically minoritised clients.

Interpreter-mediated CBT should be embedded within training pathways rather than relegated to CPD. At present, many therapists receive little or no structured training on how to work effectively with interpreters (Kunorubwe et al., Reference Kunorubwe, O’Leary and Wynne2025), despite growing evidence that therapeutic processes and outcomes are significantly shaped by the quality of interpreter-mediated communication. A strategic approach from NHS England, accrediting bodies such as the BABCP (or others outside of the UK), and training institutions such as universities, would be to review core curricula and incorporate interpreter-related competencies into mandatory training requirements. This could include structured teaching, demonstrations, simulation-based learning, and, wherever possible, opportunities for trainees to engage with case studies specifically involving interpreter-mediated CBT. Embedding such training within the core curriculum would ensure that therapists enter the workforce equipped to deliver equitable care to linguistically diverse populations, while also aligning with recommendations that emphasise the role of organisations such as accrediting bodies and training institutions in creating optimum conditions for therapists and interpreters to work effectively together (Kunorubwe, Reference Kunorubwe2025).

In addition, it would be beneficial to address the lack of translated CBT materials that can act as a barrier to interpreter-mediated therapy with TTad (Kunorubwe et al., Reference Kunorubwe, O’Leary and Wynne2025). Therapists often rely on interpreters to directly translate English-language questionnaires during sessions, which raises concerns about the consistency and validity of such measures (Tutani et al., Reference Tutani, Eldred and Sykes2018). At a minimum, translated materials and resources in the client’s primary language should be shared with the interpreter prior to the session (d’Ardenne et al., Reference d’Ardenne, Ruaro, Cestari, Fakhoury and Priebe2007; Kunorubwe et al., Reference Kunorubwe, O’Leary and Wynne2025; Villalobos et al., Reference Villalobos, Orengo-Aguayo, Castellanos, Pastrana and Stewart2021).

Moreover, there is consideration as to what constitutes improvement. Williams (Reference Williams2015) argues that the IAPT framework is underpinned by positivist epistemological assumptions which emphasise quantifiable symptom change as the primary indicator of recovery. This emphasis on numerical outcome data shapes not only what is valued as ‘evidence’, but also what forms of experience are recognised, often ignoring more idiosyncratic, subjective, or contextually grounded meanings of recovery. Moreover, the outcome measures used are developed predominant in English speaking and Western contexts and may not fully capture expression of symptoms in linguistically diverse populations, and responses may be influenced by cultural norms, potentially limiting the comparability and accuracy of outcomes for some patients. For example, the PHQ-9 and GAD-7 have shown cultural biases, with certain items functioning differently across groups and potentially misrepresenting symptom severity; notably, the PHQ-9 item ‘psychomotor problems’ appeared culturally biased among Surinam Dutch males (Baas et al., Reference Baas, Cramer, Koeter, van de Lisdonk, van Weert and Schene2011), and Black/African American participants with high anxiety symptoms scored lower on the GAD-7 than other groups with comparable symptom levels (Parkerson et al., Reference Parkerson, Thibodeau, Brandt, Zvolensky and Asmundson2015). Therefore, there is a need for practitioners to carefully review psychometric data for accuracy and be mindful of how the interpretation process may have impacted client responses (Kunorubwe, Reference Kunorubwe2025) and in the longer term there is the need for the development and routine use of more culturally attuned routine outcome measures. Without culturally responsive outcome measures and translated therapeutic resources, services may unintentionally reinforce barriers to engagement and recovery, as clients are required to navigate therapy in frameworks that do not adequately reflect their cultural or linguistic contexts.

Strengths and limitations

A key strength of this study is its use of a large, routinely collected dataset from TTad services, allowing for the analysis of interpreter-mediated therapy in the context of everyday NHS clinical practice. Unlike many previous studies that have focused on smaller samples, often in specialist settings, this analysis reflects the real world. By focusing on routine care, the study enhances ecological validity and offers insights that are directly relevant to current service provision. Furthermore, the findings contribute to the broader discourse on healthcare inequities by highlighting how language needs intersect with other forms of structural disadvantage. This is particularly important given the longstanding under-representation of linguistically minoritised groups in mental health research. In addressing this significant evidence gap, the study offers an important first step towards better understanding interpreter-mediated therapy within TTad and sets the foundation for future research and targeted service improvements.

Despite the valuable insights provided by this study on interpreter-mediated therapy within the TTad program, several limitations must be acknowledged.

A significant challenge arises from the categorisation of interpreter use by NHS England. The available classifications: ‘Yes – Professional interpreter’, ‘Yes – Family member or friend’, ‘Yes – Another person’, ‘No – Interpreter not required’, ‘No – Interpreter was required but did not attend’, and ‘Not known (not recorded)’ lack specificity. The broad category of ‘Yes – Another person’ is particularly problematic, as it remains unclear whether this refers to another staff member, an ad hoc interpreter, or an unqualified individual. Additionally, ‘Not known (not recorded)’ represents the largest proportion of recorded data, 72.1% of the overall dataset, raising concerns about data quality, routine data collection practices, the visibility of language needs in service delivery and the potential under-reporting of interpreter needs. This gap limits the ability to assess the impact of different types of interpreters on therapeutic outcomes. As such, reported outcomes should be interpreted with caution.

Furthermore, there is no information on the quality of interpretation provided, which could significantly affect the therapeutic process. While best practice guidelines exist (Beck, Reference Beck2016; Beck et al., Reference Beck, Naz, Brooks and Jankowska2019; Costa, Reference Costa2022a; Kunorubwe, Reference Kunorubwe2025; Leanza et al., Reference Leanza, Miklavcic, Boivin, Rosenberg, Kirmayer, Guzder and Rousseau2013; Tribe and Lane, Reference Tribe and Lane2009; Tribe and Thompson, Reference Tribe and Thompson2017; Tribe and Thompson, Reference Tribe and Thompson2022), it is unclear whether these were followed by either the interpreter or the therapist. The reliance on informal interpreters, such as family members or untrained staff, is not recommended, as it may introduce biases or compromise confidentiality. However, the dataset does not allow for an analysis of these factors.

Another limitation is the lack of specificity regarding the CBT interventions delivered within TTad. The dataset does not provide details on the quality, structure, or supervisory processes involved when an interpreter was present. Additionally, information on the number of sessions attended is unavailable, making it difficult to assess engagement levels, treatment adherence, or other factors influencing treatment outcomes.

Importantly, there is no indication of whether treatments were culturally adapted or simply delivered with an interpreter in a manner that is not idiosyncratic, formulation driven or collaborative. Delivering CBT in a ‘copy and paste’ manner through an interpreter may be perceived as ineffective due to language barriers, when the underlying issue may be a lack of cultural adaptation (Kunorubwe, Reference Kunorubwe2025). This is particularly relevant, as culturally adapted interventions have been shown to improve acceptability, engagement and outcomes for diverse populations (Kunorubwe et al., Reference Kunorubwe, John, Molina, Davies, Gait, John, Roderique-Davies and Lancastle2024). Without this information, it is unclear whether interpreter-mediated therapy was tailored to meet the cultural and linguistic needs of service users.

The one-year time frame of the dataset also restricts the ability to analyse long-term trends or determine whether engagement with interpreter-mediated therapy and its effectiveness have changed over time. A broader timespan would allow for a more comprehensive understanding of patterns in service use and treatment outcomes. As the data included only patients who attended at least half of their Step 3 CBT appointments, we were unable to examine outcomes for those who dropped out or attended fewer sessions. This may bias the findings towards patients who completed a larger portion of treatment and limits the generalisability of the results to all referrals.

Additionally, the dataset lacks demographic information, making it impossible to account for other factors that may influence engagement and outcomes, such as socioeconomic status, migration history, or previous mental health treatment. This absence of data limits the ability to explore potential disparities in access to interpreter-mediated therapy.

Another key limitation is the absence of information on the range of languages represented in the dataset. Beyond English, there are approximately 7000 spoken languages worldwide (Leben, Reference Leben2018), some of which share historical and structural similarities. While certain widely spoken languages may be more readily interpreted, others may face interpreter shortages. A common barrier is the lack of available interpreters for some languages or dialects (Kunorubwe et al., Reference Kunorubwe, O’Leary and Wynne2025), which can leave some linguistic groups under-served, ultimately affecting access to and the quality of therapy. It remains unclear whether certain languages are better accommodated within TTad services or if some linguistic groups experience greater barriers to accessing interpreter-mediated CBT.

As no information is provided about the therapists, we are unable to offer any detailed report or analyse whether factors such as location, service structure, therapist experience, or other variables influence wating times or treatment outcomes. However, it is expected that therapists delivering therapy should be appropriately trained and supported, adhering to the best practice guidelines (National Collaborating Centre for Mental Health, 2024).

As no information was provided about the interpreters, we cannot report on their involvement or analyse factors such as interpreter experience, competence, amount of supervision, or whether the professional interpreter adhered to best practices. However, it is expected that interpreters involved in therapy are appropriately trained and supported, adhering to the best practice guidelines (National Collaborating Centre for Mental Health, 2024).

Additionally, there is a lack of data on patients’ initial symptom severity across different groups, which likely impacted on measures of reliable change and recovery. Without this information, we have limited ability to determine whether observed differences in outcomes are due to treatment effects, waiting times, or baseline differences in clinical need.

Finally, the dataset does not provide details on the specific disorders treated or whether disorder-specific measures were used to assess treatment outcomes. This limits the ability to determine whether certain mental health conditions are more amenable to interpreter-mediated therapy. Some disorders may be more responsive to interpreter-mediated CBT, whereas others may pose additional challenges.

Addressing these limitations would require improved data collection practices, including more precise categorisation of interpreter roles, quality measures, session level details, disorder specific outcomes, and cultural adaptations. Expanding the dataset beyond a single year and incorporating demographic and linguistic information would allow for a more nuanced understanding of interpreter-mediated therapy within TTad and its impact on patient outcomes.

Research recommendations

Service development and future research should focus on improving data collection and classification by ensuring more precise categorisation of interpreter roles within NHS frameworks and reducing the use of vague classifications like ‘Not known (not recorded)’. Assessing the quality of interpretation is also crucial, as interpreter training, adherence to best practice guidelines, and interpretation accuracy can significantly impact therapeutic outcomes.

Additionally, studies should examine whether CBT interventions delivered with interpreters are culturally adapted, as culturally tailored therapies have been shown to improve engagement and effectiveness. Longitudinal research is needed to track interpreter-mediated therapy trends over time, allowing for a better understanding of long-term engagement and treatment success. Incorporating demographic and linguistic factors into research would help identify disparities in access and effectiveness, particularly for under-served language groups. It would also be interesting to examine the impact of interpreter-mediated therapies of other modalities that are delivered in TTad.

Further investigation is also required to determine whether certain mental health conditions, such as anxiety and depression, are more responsive to interpreter-mediated CBT compared with disorders like PTSD or personality disorders. Best practices for interpreter-mediated therapy, including the role of pre-session briefings, post-session debriefings, and therapist–interpreter collaboration, should be explored to enhance service delivery. Additionally, studies should analyse the specific CBT interventions provided with an interpreter and the role of supervision in maintaining therapy quality. Research on language availability and accessibility would help address interpreter shortages for certain languages and dialects, ensuring equitable access to mental health services. Alternative language support methods, such as digital translation tools, multi-lingual therapy workbooks, and bilingual therapy models, should be investigated as potential solutions to barriers to interpreter-mediated therapy.

Finally, future research should also incorporate data on patients’ initial symptom severity and baseline needs, as these factors may influence both engagement with interpreter-mediated therapy and treatment outcomes. Studies examining reliable deterioration could help identify which patient groups are most at risk during waiting periods or therapy, informing targeted interventions and service planning to prevent worsening of mental health symptoms.

Addressing these research gaps will contribute to a more comprehensive understanding of interpreter-mediated therapy and its impact on mental health outcomes for linguistically diverse populations.