Multilingualism in Psychotherapists in Training in Germany: Results of a Nationwide Cross-Sectional Study

Design

The cross-sectional, nationwide study was implemented and enabled through the University Medical Center Hamburg-Eppendorf (UKE). The survey was conducted among psychotherapists-in-training in Germany (all with a master's degree in psychology or similar) as an online survey, implemented on SoSci Survey (Leiner, 2024).

Recruitment

A survey link and cover letter were emailed to all 238 publicly identifiable institutes for psychotherapy training in Germany with a request to forward the email to their trainees. In addition, the link was spread in forums and social media groups used by the target group. A small incentive in the form of the possibility of winning 1 of 20 online book vouchers (worth €15 each) was offered to increase participation rate.

Informed Consent and Data Protection

At the beginning of the survey, participants received written information about the study's purpose, data privacy policy and an ethical statement including confidentiality, informed consent and voluntary participation according to the principles expressed in the Declaration of Helsinki (World Medical Association, 2013). Participants were informed of the aim, background and duration of the survey prior to taking part. All participants provided written informed consent for the survey to be evaluated, and the results of the study to be published anonymously.

Data Collection

Data were collected between February and March 2019. To increase the participation rate, a reminder email was sent after two weeks to all 238 institutes for psychotherapy training.

Instrument Development and Pretesting

The applied instrument was developed based on prior paper-and-pencil questionnaires (Mösko et al., 2024; Wagner, 2018). Part of the instrument was the questionnaire on ‘Attitudes towards working with an Interpreter in Health Care Services (AIHC-20)’, which consisted of 20 items (e.g. ‘I could imagine working more frequently with interpreters if access were easier’) assessed with a 5-point Likert-scale and was developed as a generic instrument for different healthcare services. The developed instrument was evaluated by six experts in the fields of community interpreting, healthcare services, legal interpretation, intercultural training, psychology and migration research. After modification, the instrument was pretested for usability by two people from the target group as an offline-paper version and by five people from the target group as an online version.

Questionnaire

The final questionnaire consisted of 48 items and was divided into seven sections. The first section collected sociodemographic data (city, age, gender) and the second collected information on psychotherapy training details. Sections three to five focused on the trainees’ migration background, language capacities and use (e.g. ‘In which languages other than German would you feel confident offering treatment?’), cross-cultural encounters (e.g. ‘Do you have contact with people from different cultural backgrounds within your family and circle of friends?’), relevance of language barriers in psychotherapeutic treatment and interpreter use. AIHC-20 was used as the sixth section. The last section included a question about preferences for interpreting and items assessing the need for further training.

The AIHC-20 was tested psychometrically. All item difficulties ranged between .37 and .75, except for item 12 (.80). The corrected item–total correlation value of the items should be above .5, which was the case for items 2, 3, 6, 7, 8, 14 and 15; items 4 and 5 were under .3. The internal consistency of the scale was assessed by Cronbach's alpha (α = .866) which can be considered as good.

The Kaiser–Meyer–Olkin criterion value was .882 and the Bartlett test of sphericity showed a highly significant result (p < .001), which allowed us to conduct a principal component analysis with varimax-rotation. The number of selected factors was determined by conducting a parallel analysis. Factor analysis extracted two factors, which explained 41.28% of the total variance. Because the items subsumed under the factors were heterogeneous, meaningful classification and interpretation was not possible. The assessment was conducted on a single-item basis, instead.

Data Analysis

Only completed questionnaires were considered for the analysis. People who discontinued the online questionnaire prematurely were excluded. The data were analysed using SPSS Version 28. Descriptive analyses were calculated to characterise the current lingual and cultural diversity of German psychotherapists-in-training. After checking the test requirements, chi-square tests were used for exploratory questions and between-group differences were tested using two-tailed t-tests for metric data and the Mann–Whitney U-test for ordinal scale data. A regression analysis predicted the use of professional interpreters and the treatment of patients with limited German language skills (predictor variable: number of cross-cultural encounters). All analyses were conducted at a significance level of p = .05

Sample

Of the 997 psychotherapy trainees who participated, 194 were excluded due to incomplete data sets, leaving 803 valid responses (age: M = 32.49 years, SD = 6.78 years) (Table 1). There were participants from each of the 16 German federal states. Most of the participants (74%; n = 598) had not worked as a psychologist before starting training.

OPEN IN VIEWER

Table 1.

Sample Description (n = 803).

Variable	Characteristic	Number	Valid Percentage
Gender	Male	108	13
	Female	691	86
	Diverse	4	1
AgeAge groups (years)	M = 32 years
	22–25	58	7
	26–30	332	41
	31–35	223	28
	36–40	90	11
	41–45	48	6
	46–50	29	4
	>50	23	3
Patient target group	Children and youth	187	23
	Adults	591	74
	Both	31	4
Training process (years)	M = 3.30 yearsSD = 1.965
Training type	Cognitive–behavioural therapy	591	74
	Depth psychology	115	14
	Psychoanalysis	80	10
	Systemic therapy	14	2
Migration background	Foreign birthplace	130	16
	At least one parent born abroad	199	25
	Migration background (total)	217	27
First learned language	German	628	78
	Other language	175	22

Migration Background and Lingual Diversity Among Psychotherapists-in-Training

Some 217 participants (27%) reported a so-called migration background, defined as being born outside Germany or having at least one parent who was born abroad (Table 1).

The 217 psychotherapists with a migration background themselves or whose parents were born abroad came from 65 different countries of origin, most often from Russia (n = 16), Bulgaria (n = 10), Poland and Ukraine (each n = 9), Romania (n = 8), Turkey (n = 7), Iran or Italy (each n = 6).

Some 175 (22%) participants had a first learned language other than German, with 42 (5.2%) having learned German bilingually. Overall, 38 different languages were reported. The ten most common languages were Russian, Turkish (each 4%), Polish and English (each 2%), Bulgarian, Italian, Persian, Romanian and Spanish (each 1%) and Dutch (<1%).

A total of 99% of the participants (n = 791) stated that they had learned at least one other language in addition to their first learned language (not including German, Latin, Old Greek, or sign language). Of those, 304 (38%) participants spoke two additional languages, 260 (33%) three additional languages, and 97 (12%) more than three languages. Altogether, 44 languages were reported. The ten most common languages were English (97%), French (63%), Spanish (34%), Italian (9%), Dutch (8%), Russian (6%), Afrikaans (5%), Swedish (3%), and Turkish and Arabic (each 2%).

Multilingual Treatment

Some 609 (76%) participants reported that they would conduct psychotherapy in a language other than German. In total, 30 treatment languages were reported, including English (82%), French (9%), Spanish (7%), Dutch and Russian (each 5%), Turkish (4%), Italian (3%), Bulgarian, Portuguese, Romanian and Polish (each with 1%).

Significantly more of the participants with a migration background stated that they would conduct psychotherapy in a language other than German compared with those without a migration background (χ² (df = 1) = 29.885, p < .001; n = 803).

Patients with Limited Language Skills

Some 238 (69%) participants had contact to at least one patient with insufficient German language skills for diagnostic and treatment discussion during the past six months. On average, 7.5% (SD = 12.58, n = 779) of patients who had been treated during this period had insufficient German language skills. Participants with a migration background treated significantly more patients with limited German language skills (M = 10%, SD = 16%) compared with participants without a migration background (M = 6%, SD = 11%) (t(778) = 3.799; p < .001). Those who had worked as a psychologist before starting training treated significantly more patients with limited German language skills (M = 10%, SD = 15%) than participants without prior work experience (M = 7%, SD = 12%) [t(788) = −2.821, p = .005]. Furthermore, the number of cross-cultural encounters predicted the treatment of patients with limited German language skills (F(1, 784) = 11.820, p < .001) with β = .122 (t(784) = 3.438, p < .001).

Ninety-six (12%) sample members had refused treatment because of a language barrier in the past six months. There was a significant difference between those who had work experience as psychologists before starting training and those without prior work experience (16% vs 11%; χ² (df = 1) = 4.488, p = .034; n = 803).

Working with Professional Interpreters

In total, 420 (52%) participants already had experience of working with professional interpreters; 208 (26%) participants had already worked with an interpreter during their psychotherapy training; and 293 (37%) in outside psychological or psychotherapeutic training/work (jobs, internships or voluntary work before or parallel to the current training). Most of the trainees stated that they would prefer to work with interpreters on-site (94%) rather than via telephone (2%) or video (4%). Participants with a migration background more often worked with interpreters outside psychotherapy training compared with participants without a migration background (χ² (df = 1) = 18.721, p < .001; n = 803). In addition, participants with prior work experience were more likely to have worked with an interpreter outside psychotherapy training compared with those without prior work experience (χ² (df = 1) = 10.419, p = .001; n = 803). Moreover, prior use of professional interpreters was predicted by participants’ number of cross-cultural encounters (F(1, 801) = 43.438, p < .001). More cross-cultural encounters were significantly associated with a higher frequency of interpreter use (β = −.227, t(801) = −6.591, p ≤ .001).

Attitudes Towards Working with an Interpreter

A descriptive analysis of the AIHC-20 items is given in Table 2.

OPEN IN VIEWER

Table 2.

Frequencies Concerning the Attitudes Towards Working With an Interpreter.

		Totally disagree (%)	Disagree (%)	Neutral (%)	Agree (%)	Totally agree (%)
	Organisational barrier
1	The time and organisational effort doubles when including an interpreter. I can’t afford this.	9	31	30	24	6
2	I could imagine working more frequently with interpreters if access were easier.	2	10	20	42	26
3	If the financing of interpreters were clearly regulated, I would integrate them more often in my work.	4	9	21	37	30
4	I could imagine working more frequently with interpreters if there were a concrete manual for clinical–practical work with interpreters.	6	14	23	37	20
	Qualifications
5	I do not involve interpreters because they are not sufficiently trained for psychological work.	26	35	21	15	3
6	I could imagine working more frequently with interpreters if I were trained for work with interpreters.	5	18	22	38	17
7	I am concerned that when including interpreters, they will intervene.	5	25	30	28	12
	Therapeutic relationship
8	Interpreters interfere with building a relationship of trust between the patient and me.	3	25	33	34	5
9	Inclusion of interpreters hinders the construction of the therapeutic relationship.	2	18	30	37	13
10	I feel excluded when interpreters are included because I do not speak the same language as the patient and the interpreter.	21	40	23	14	2
	General therapeutic process
11	Interpreters calm the treatment situation because the patient and I have breaks to think during the interpreting process.	5	24	44	35	5
12	I appreciate patients increased participation in the conversation when interpreters are included.	1	3	12	43	42
13	Inclusion of interpreters could facilitate diagnostics.	2	6	24	46	22
	Specific therapeutic process
14	I am concerned that patients do not address intimate topics when interpreters are included.	3	17	30	35	15
15	If I include interpreters, I fear that information is lost or that there are mistakes in interpreting.	3	16	29	35	17
16	Cultural misunderstanding could be exposed and addressed through inclusion of interpreters.	<1	5	25	48	21
	Emotional assessment
17	I am uncomfortable when I do not know exactly what is spoken in the patient's language.	8	22	24	34	13
18	I feel assured when working with interpreters.	10	27	34	22	7
	General
19	When a language barrier exists, inclusion of interpreters is the only opportunity to conduct successful psychotherapeutic treatment.	1	6	18	41	34
20	Inclusion of interpreters reduces the patients’ motivation to learn German.	34	36	19	9	3

Training in ‘Working with an Interpreter’

A total of 579 (72%) participants stated that training in ‘psychotherapy with an interpreter’ would be useful in their further training. By contrast, only 48 (6%) participants reported that they had already participated in similar training.

Instrument

The self-developed questionnaire measuring ‘Attitudes towards working with an Interpreter in Health Care Services (AIHC-20)’ was the first German language instrument of its kind. However, factor analysis showed that there was no useful interpretation of the suggested underlying two-factor solution, and so further adaptations are needed in future to develop a valid and reliable instrument for assessing health professionals’ attitudes to working with an interpreter.

Language Barrier

The majority of trainees (69%) had worked with patients who did not speak sufficient German to receive psychotherapeutic treatment.

A personal migration background, work experience before starting psychotherapeutic training and number of trainees’ cross-cultural encounters were associated with having treated more patients who had limited German language skills. Indeed, studies report that therapists with a personal migration background are more often consulted by migrants than are therapists without a migration background (Kirkcaldy et al., 2006; Mösko et al., 2013) and that migrants prefer language-concordant treatment (Villalobos et al., 2016). First, therapists with a personal migration background might be perceived as having a better understanding of the importance of language for the therapeutic process (Pavlenko, 2006) and being more culturally open towards this patient group. Second, cross-cultural encounters have been shown to be associated with increased intercultural competence and cultural intelligence (Engle & Crowne, 2014; Holmes & O’Neill, 2012), which might explain the higher treatment rate for patients with limited German language skills. The therapists may feel more confident about and familiar with treatment of this patient group. Moreover, one might guess that the positive feedback they may receive from patients could encourage them to offer more therapy options of this type. Third, work experience has been shown to be significantly associated with perceived therapeutic mastery (Dawson, 2018; Orlinsky et al., 1999). This could explain the correlation between work experience and the number of patients with limited language skills that have been treated. Work experience might boost therapists’ confidence in dealing with treatment situations that are perceived as difficult, such as treatment involving a language barrier.

Treatments were refused by 12% of trainees because of a language barrier. Similar can be observed in other parts of the healthcare system. For example, about one-fifth of clinicians in outpatient diabetes care have refused patient treatment because of a perceived language barrier (Mösko et al., 2024). Refusals by trainees may result from the unavailability of a professional interpreter (Jensen et al., 2017). Furthermore, a lack of supervisors for treatments with an interpreter within psychotherapy training may be one explanation for the refusal to treat patients with limited German language skills. Because a lack of treatment can lead to worsening symptoms, a chronification of the mental disorder and an increased risk for suicidality, this problem has to be addressed as a matter of urgency. Regular funding of interpreting services in the healthcare system and the inclusion of modules on ‘interpreting in psychotherapy’ in training curricula are exemplary measures to counteract this.

Multilingual Treatment

Some 38 first learned languages other than the official language were reported. Twenty-two per cent of participants reported a first learned language other than German, with Russian and Turkish being the two most prominent languages. Also including second language skills, 76% of participants could imagine conducting psychotherapy in a language other than German, with 30 possible treatment languages reported in total. This highlights the trainees’ general potential to address multilingual needs in psychotherapy. Although 99% of participants stated they had learned a second language, almost one-quarter did not feel comfortable with the idea of offering psychotherapy in a language than German. In addition to subjectively perceived insufficient language skills, feelings of insecurity and unprofessionalism when having to treat patients in another language might be an explanation for this discrepancy, because psychotherapy training is taught in German only. Participants who were bilingual reported feeling like a different person when using different languages (Bager-Charleson et al., 2017; Costa & Dewaele, 2012; Dewaele & Costa, 2013) and that the expression of emotions felt closer in one's first learned language (Verkerk et al., 2023). This highlights the importance of language for identity and might explain the caution shown by psychotherapists practising their profession in another language.

An estimated 200 different languages are currently spoken in Germany (Gogolin & Krüger-Potratz, 2012). The languages that trainees offered for multilingual treatment were far from being linguistically diverse. Most of the participants (82%) would offer psychotherapy in English. Offering treatment in Russian and Turkish was reported by at least 5% of the current sample.

Although our findings indicate the general existence of linguistic diversity among psychotherapists in Germany, there is still a huge discrepancy with the patients’ actual needs. The latest supply report of the National Working Group for Psychosocial Counselling for Refugees and Victims of Torture (BAfF e.V., 2023) has shown that more than half of all counselling/therapy sessions in refugee counselling centres in Germany had to be conducted with the assistance of an professional interpreter because communication in German, English or another common language was not possible. Thus, multilingual psychotherapists alone will not be able to sufficiently cover patients’ heterogeneous linguistic needs. Therefore, because migrant patient groups are considered a particularly vulnerable population (Li et al., 2019), the use of professional interpreters is inevitable to ensure access and good quality of treatment for patients speaking any language.

Use of Professional Interpreters

Compared with 26% of survey participants with previous experience of working with professional interpreters, 69% reported a language barrier in their treatment within the past six months. Different factors may be considered to explain this. In Germany, for example, there remains a lack of accessibility and funding opportunities, as well as a lack of qualified interpreters and training institution awareness (Karliner et al., 2007; Penka et al., 2012), although psychotherapy with interpreters has been shown to be effective (Brandl et al., 2020; Brune et al., 2011; Fennig & Denov, 2021). Whether those psychotherapist trainees who encountered a language barrier in their treatment, but did not include a professional interpreter resorted to other ways of communication (e.g. translation apps), or whether those patients were left untreated, remains unanswered. At the same time, the numbers illustrate the inevitability of including professional interpreters to guarantee qualified healthcare for patients with limited language skills.

Participants with a personal migration background, as well as those with more frequent cross-cultural encounters in their everyday life, more often worked with interpreters outside training, which indicates an increased cultural openness that has also been found in earlier studies (Engle & Crowne, 2014; Holmes & O’Neill, 2012). Trainees showed a clear preference (94%) for working with interpreters on-site compared with via telephone or video. These results agree with studies that showed a preference for face-to-face consultations with interpreters in the MHC sector (Kluge et al., 2012) and the somatic healthcare sector (Mösko et al., 2013), whereas other healthcare services (e.g. emergency services) make more frequent use of quickly available telephone interpreting (Kluge et al., 2012).

Attitudes Towards Working with an Interpreter

The psychotherapist trainees’ attitudes towards working with professional interpreters are somewhat diverse. The majority of the sample agreed that including an interpreter is the only opportunity to conduct successful psychotherapy if a language barrier exists. It may, for example, facilitate the diagnostic process and prevent or address cultural misunderstandings. Concerning organisational efforts, more than two-thirds considered working with an interpreter more frequently if access were easier and financing clearly regulated, which is in line with earlier studies on reasons for a lack of interpreter services in German healthcare institutions (Dauvrin et al., 2012).

The need for training for interpreters working in a triad (Kluge & Kassim, 2006; Kuay et al., 2015; Leanza et al., 2013; Miller et al., 2005; Morina & Maier, 2009; O'Hara & Akinsulure-Smith, 2011) was supported by the current results. Half of the participants mentioned concerns of losing information and feared mistakes in translation. Translation mistakes have been shown to depend on the interpreter training; professional interpreters make fewer translation mistakes compared with ad hoc interpreters (Karliner et al., 2007; Vasquez & Javier, 1991) Thus, a minimal qualification standard for interpreters (Breitsprecher et al., 2020) may enhance treatment quality and at the same time alleviate therapists’ concerns.

In addition, more than half of the participants agreed they would work with an interpreter more frequently if they had received prior training or if there were a concrete manual for clinical–practical work with interpreters. Although most trainees disagreed on the concrete fear of feeling excluded when an interpreter is involved (Brisset et al., 2013; Miller et al., 2005; Morina & Maier, 2009), almost half mentioned feelings of discomfort when they could not fully understand what was being said in the patient's language. Therefore, the findings clearly indicate a need to include modules on working with interpreters in the training curricula for future psychotherapists and other healthcare professions (von Lersner et al., 2016; Mews et al., 2018).

In relation to this, the study participants reported motivation and interest in including the topic ‘working together with interpreters’ in psychotherapy training. In this domain, little training has been offered to date, although the relevance and need has been reported previously (Machleidt et al., 2005; Miller et al., 2005).

Limitations

This study has a few limitations that should be considered. First, it is likely that people with a migration background were overrepresented in this sample. Compared with the general German population at the time (24%) (Statistisches Bundesamt, 2017), the percentage of people with a migration background was slightly higher (27%). In addition, people with a migration background are still underrepresented at universities and in obtaining higher work-related qualifications (Haasen, 2000; Studentenwerk, 2012).The online survey was forwarded to all psychotherapy training institutes in Germany. It is possible that trainees with a migration background felt more drawn to the study topic and showed a greater willingness to participate in the study, owing to, for example, personal experiences or the experiences of close relatives. At the same time, the comparatively high percentage of trainees reporting previous experience in working with a professional interpreter indicates that those psychotherapist trainees with a special interest in multilingual psychotherapy and the use of interpreters were more likely to participate in the survey. However, given the otherwise similarity to other psychotherapy training samples (Nübling et al., 2019), we still consider our findings to be conclusive. Although the sample may have been more interested in this topic than average, our results show a significant gap in multilingual psychotherapy provision. It is likely that this gap is even greater, meaning that the general population of psychotherapists-in-training might not be as culturally open as our sample. An overestimation of this treatment gap, however, is unlikely given the current sample. Also, to obtain a valid instrument for measuring therapists’ attitudes towards working with interpreters, the self-developed AIHC-20 questionnaire needs further major revisions and validations. Because this was not the main focus of our study, the questionnaire could not be revised within the scope of this research project. However, because an instrument like this does not yet exist, we highly recommend its further development.

Conclusions

This explorative study on multilingualism in psychotherapists-in-training and their attitudes and experiences of working with interpreters sheds light on an understudied research field that is gaining increasing importance in society. Even if future psychotherapists are more linguistically and culturally diverse, that will not be enough to cover patients’ linguistic needs. So far, multilingual treatments are mainly offered in European languages, whereas the patients’ linguistic needs are mostly different and considerably more diverse. Although there already exist a substantial number of future psychotherapists who seem to be open to working with interpreters, this number is far from being sufficient to cover the actual demand. Therefore, patients speaking languages other than German or English are underserved in MHC services, so far. On the one hand, this means that more effort should be invested in training psychotherapists with greater linguistic diversity in the future. Heterogeneous language acquisition programmes could be offered as part of the studies and multilingual therapy could be practised early (e.g. through role-play). In addition, more egalitarian access to academic education would be a prerequisite to enabling more people with a migration background to study psychology/psychotherapy, which may indirectly promote multilingual therapy.

On the other hand, the use of interpreters in cases of a language barrier has to become a matter of course. Critical reasons underlying the overall too infrequent use of interpreters are related to access, financing and training. In Germany as well as many other countries, costs incurred in connection with interpreting services in the healthcare sector are not reimbursed and have to be covered by the practitioner or the patient. Services such as community interpreting services attempt to close this gap, but are far from being available everywhere and to a sufficient extent. In addition to standardised training courses for interpreters to ensure the quality of translation services, psychotherapist training on the inclusion of interpreters in psychotherapy is useful to reduce fears and strengthen confidence in dealing with the triadic situation. The current study shows that a large proportion of the psychotherapy trainees would be open to such training. In general, a personal migration background, prior work experience as a psychologist and cross-cultural encounters in everyday life are positively associated with serving patients with a language barrier and working with interpreters. This underlines, among other things, the need to increase diversity in the psychotherapeutic profession.

Implications for the Future

Owing to increasing migration flows, there is currently and will be a future need for more linguistically and culturally diverse psychotherapy globally. Psychotherapy students with a migration background should be supported in using their diverse language proficiencies throughout their studies and training. In addition, the second learned languages of trainees could be further promoted during study and training to strengthen their linguistic competences. To increase the use of professional interpreters in psychological treatments, political efforts are needed to improve access and funding. Access to and financing of professional interpreting is still mainly dependent on local initiatives; national regulations and financial support are missing. Psychometrically tested instruments assessing attitudes towards working with interpreters should be further developed. Furthermore, training on how to work effectively with an interpreter in (mental) healthcare services should be integrated into the curricula for trainees and videos (Hanft-Robert et al., 2023) could be helpful in this regard. In addition, training for supervisors should be provided to develop a culture of acceptance for this relatively new profession in MHC services. Finally, the occupational group of interpreters needs to recruit more professionals by establishing basic qualification programmes including an independent exam and an appropriate payment for their valuable work. Together, this will lead to greater use of professional interpreters and better access to and quality of MHC for people facing language barriers.