Prevalent practices amongst healthcare professionals in paediatric settings in using medical interpreters for families with limited national language proficiency: A narrative scoping review

Abstract

Background

Where language barriers exist, trained medical interpreters provide more accurate and comprehensive communication between clinicians and service users compared to ad hoc interpreters or no interpreter at all. The nature and extent of research evidence about when and how medical interpreters are used by health professionals in paediatric settings to mediate communication with families who do not speak the national language proficiently have not yet been published.

Aim

To examine the potential size and scope of available research literature about the prevalent practices of healthcare professionals in using medical interpreters in paediatric settings for families with limited national spoken language proficiency.

Methods

We used an established framework for scoping reviews, including suggested modifications by JBI. Precise eligibility criteria were identified, and a comprehensive search strategy was applied. Sources of evidence were confined to primary research that spoke specifically to the research question. The process of study selection and results are presented narratively and in tabular form. Two reviewers independently engaged in the screening process, and two reviewers undertook a full paper review of the included articles, with a third consulted where an additional opinion was required.

Results

We included 21 studies published between 2000-2022, most using (quantitative) survey methods. Key results captured in six themes are as follows: medical interpreters tended to be underused by health professionals, especially nurses; time was the greatest impediment to their use, but so too were difficulties with the perceived quality of interpreter services; technology was used for medical interpreting at some study sites with mixed responses; use of unauthorised interpreters was widespread; language barriers impacted on the care provider and family relationship; and cost of interpreter services tended not to be reported as a barrier to their use.

Conclusions

Further scholarship is needed to understand the use of medical interpreter services with reference to the complexity of health work, as well as the social context of inequality for many families with limited national language proficiency accessing paediatric healthcare. Recent developments in video technology offer promise in improving interpreter use but require oversight for quality. We have highlighted the need for better training for and regulation of medical interpreters and clinical guidelines on the use of interpreters by healthcare professionals.

1. Introduction and background

Most high-income countries across the globe have experienced increasing levels of migration in recent decades (United Nations, 2021) and are challenged to provide an adequate standard of healthcare for migrants who do not speak the receiving country's primary spoken language (Lebano et al., 2020). Effective communication has long been perceived as a basis for good health service provider-user interactions, and lapses of communication can result in poor health outcomes (Ratna, 2019). Compromised understanding has been found to occur even where both provider and service user speak the same language, so clearly, a language barrier exacerbates communication difficulties substantially (Fatahi, 2019). In paediatric settings, the engagement of families with limited national language proficiency is required to navigate communication about the child's condition, and while medical interpreters as communication brokers may facilitate the process, the nature and extent of research evidence about when and how they are used by healthcare professionals has not yet been published. This article presents a scoping review of literature on the prevalent practices of healthcare professionals in using medical interpreter services in paediatric settings to map available literature on the topic and identify key concepts relating to it. We provide a brief background on what is known about the effectiveness of medical interpreters in enhancing communication in healthcare interactions, as well as the extent to which legal jurisdictions financially support medical interpreter services, before we shift the focus to the subject of the scoping review.

Available evidence on the effectiveness of interpreter services in healthcare settings is limited, and what is available has yielded mixed results (Brabdl et al., 2020). Overall though, medical interpreter services in clinical settings where they work well have been found to result in more accurate and comprehensive communication between clinicians and service users (Gutman et al., 2018; Brabdl et al., 2020). Outcome measures have varied from rather weak indicators such as service users’ satisfaction with interpreter services (Bagchi et al., 2011) to more objective measures that identified how much service users understood compared to having no interpreter present (Gutman et al., 2018; Fennig and Denov, 2021). Flores et al. (2012) measured medical interpreter errors and found that accuracy in interpretation was linked to training; interpreters with a higher number of training hours had interpretations with fewer errors. An important finding of that study was that there was a significant reduction in errors for professionally-trained interpreters compared with ad hoc interpreters or the absence of an interpreter. With reference to health promotion and illness prevention, provision of interpreter services has had positive results: Jacobs et al. (2001) found that service users with limited national language proficiency engaged to a greater degree with prevention services when an interpreter was provided as part of the service. It should be noted, however, that interpreter qualifications appear to be limited and poorly regulated (Jungner et al., 2021), and no source was located that has mapped practices and regulations in this regard across countries.

The extent to which legal jurisdictions financially support medical interpreter services varies, and some have laws that provide for this. In the United States of America (USA), guidance is provided by the Department of Health and Human Services, but a standard has yet to be established for the use of professional interpreters (Brabdl et al., 2020). In Sweden and Norway, interpreter costs are shared by the state and service provider, while in France these costs are met solely by the service providers (Jungner et al., 2021). In Denmark, a fee for interpreting services is charged to those residing in the country for over three years, although minors are exempt from the rigid criteria associated with the legislation (Dungu et al 2019). In Ireland, there is no allocated budget for interpreter services; funding of these is provided from the general budget of the clinical site (HSE, 2022) and hence competes with other budgetary demands of the service. In this review, we considered the extent to which the cost of interpreters influenced health care provider practices in using them, along with other emerging themes.

2. Methods

A scoping review was chosen as a suitable approach for synthesising research evidence in this area (Munn et al. 2018; Aromataris and Munn, 2020). This evidence synthesis review was carried out following the framework initially outlined by Arksey and O'Malley (2005) and subsequently refined by Levac et al. (2010) and JBI (Tricco et al., 2016); (Peters et al., 2022). The review was carried out and reported in accordance with the Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews checklist (PRISMA-ScR). (A protocol was developed and followed but not published owing to time constraints.) The PRISMA checklist for this study is shown in Fig. 1. The five stages of this scoping methodology are outlined below.

Fig. 1.

Prisma flow chart.

2.1. Stage 1: identification of the research question

Identifying the research question was necessary for directing the evidence synthesis and determining the relevant studies to be identified and selected (Arksey and O'Malley 2005). The research question following the Participant, Concept, and Context (PCC) format was as follows: What are the prevalent practices amongst healthcare professionals (Participants) regarding the use of medical interpreters (Concept) for families with limited national language proficiency (Concept) in paediatric health settings (Context)?

2.2. Stage 2: identification of relevant studies

Key terms were selected to locate studies pertinent to the research question outlined and are provided in Table 1.

Table 1.

Key search terms.

Participants	Concept	Context
Healthcare professional*, healthcare worker*, Health care professional*, nurse*, doctor*, physician*, NCHD*, Consultant*, CNS, CNF, ANP, Nurse specialist*, Nurse educator*, nurse facilitator*	Medical interpreter*, interpreter*, face to face interpreter*/interpretation*, phone interpreter*/interpretation*, interpretation service*, translation*, ad hoc interpreter*, Google, google translate, app, translation app*, family and friends interpreter*, LEP families, non-English speaking, Limited English proficiency, Limited English.	Paediatric hospital*, child* OR paed*, p*ed*OR infant* OR “young adult*” OR adolescent* OR infant* “young child*” tertiary hospital*, acute care, acute hospital*, intensive care, emergency, A&E, Accident and Emergency, Emergency department

Inclusion criteria were primary empirical studies in paediatric settings that included any healthcare professionals published from January 2000 to April 2022, with a precise study focus on the topic of interest. The year 2000 as a starting point was chosen to limit systematic search results to a period of technological advancement in the healthcare system. Specifically, only papers published in peer-reviewed academic journals presenting the findings of original primary research were included. Owing to the restricted timeframe related to the completion of an academic degree, we decided to exclude other search strategies such as ancestry techniques and grey literature. Studies that reported on any of the following were included: families with limited national language proficiency; medical interpreter use; healthcare professionals’ experiences with medical interpreters; healthcare professionals overcoming language barriers; use of ad hoc interpreters; use of family and friends as interpreters; and use of Google translate and translation applications. Excluded studies were those: published before 2000; published in languages other than English; based on populations other than healthcare professionals; and publications other than full research papers. (Review articles, reviews of reviews, systematic reviews, and meta-analyses, editorials and opinion pieces were excluded.)

Using the search terms and key words established for each concept (Table 1), the following six databases whose scope aligned with the review question were searched: Pubmed, EMBASE, CINAHL, PsycInfo, Social Sciences Citation Index (SSCI), and ScieLo. Incorporating Boolean operators OR, AND, and truncation, search strings were developed for each search term. MeSH terms were also utilised for the relevant databases. Search strings were formatted and inputted into a table by the team. The team liaised with an experienced librarian who rendered expert advice and enhanced the search strings. This facilitated a more optimal and efficient search. The search strings were inputted into the selected databases, and their results were exported into an EndNote library.

2.3. Stage 3: study selection

Following the clear set eligibility criteria, studies were screened and selected for review suitability following a staged screening process. A total of 27,965 articles were retrieved from the six databases and were imported into COVIDENCE for article screening and selection. COVIDENCE is an online review screening application that aims to promote a systematic approach to evidence review and enables easy screening across multiple reviewers, facilitating more effective collaboration (Babineau, 2014). After duplicate removal, 11,529 records remained. The titles and abstracts of these records were screened for relevance. Of these, 1,180 were determined to be directly related to the review question, and their full texts were reviewed. Following the application of inclusion criteria, a total of 1,155 papers were eliminated, leaving 25 studies for extraction. Two reviewers independently assessed the articles to determine study eligibility at both the title/abstract and full-text screening stages; any disagreements were further deliberated or the opinion of the third reviewer was sought.

2.4. Stage 4: data charting

The fourth stage of the scoping review framework was to organize the data from the selected articles. Quality appraisal of the studies was carried out using Critical Appraisal Skills Programme (CASP) check-lists (Critical Appraisal Skills Programme, 2022) and the Mixed Method Appraisal Tool (MMAT) (Hong et al., 2018) was utilized for quantitative studies. All twenty-one studies were of sufficient quality for inclusion in the analysis.

On completion of screening and establishment of the final studies for review (N=21), the team extracted the data from these studies using a modified and adapted version of the JBI data charting form to suit the review. The modified data extraction table included the following columns: author, year, title, country, settings, language, study participants, interpretation language, methodological approach (including conclusions and recommendations), study design, study aim, data analysis, and sample size (including general characteristics such as age and any special characteristics related to the nature of the study). Table 2 provides a more concise version of this form.

Table 2.

Individual study overview: Aim, methodological approach, sample, settings and key findings of included studies.

Author, year, country	Aim	Methodological approach	Setting and sample	Key findings
Abbe et al. (2006) USA1	To explore this challenge from the perspectives of a sample of pediatric oncologists, interpreters, and Spanish-speaking parents of children with newly diagnosed leukemia.	Quantitative Survey questionnaire that included open-ended questions	Two non-profit private paediatric hospitals (oncology) 37 oncologists 17 medical interpreters 17 parents	•Concern expressed by all participant groups re communicating with a language barrier. •Physicians concerned re accuracy and completeness of interpretations, complexity of information delivered and loss of control and confidence over information. •Interpreters concerned re information overload and healthcare professionals’ lack of understanding of disadvantaged position of limited national language proficiency families. •Parents concerned re understanding information and missing meaning.
Burbano O'Leary et al. (2003) USA	To describe perceptions of how a lack of house staff Spanish proficiency adversely affects communication with Spanish-speaking families with limited English proficiency.	Quantitative Survey questionnaire	Paediatric hospital 59 junior doctors	•Over half of trainees used their inadequate language (Spanish) skills to communicate often or daily with limited English proficiency service-users. •Over half believed limited English proficiency families did not understand their child's diagnosis. •Large majority reported avoiding communication with limited English proficiency families. •All respondents agreed that hospital interpreters were effective, but three-quarters used hospital interpreters ‘never’ or just ‘sometimes’. •Over half called on language proficient colleagues ‘often’ or ‘daily’ to interpret.
Cheng et al. (2021) USA	To increase understanding of current practices and perceptions of family-centered rounds by providers for limited English-proficient families relative to English-proficient families.	Qualitative Ethnographic observation of healthcare professionals’ rounds Semi-structured interviews Grounded theory	Paediatric inpatient wards tertiary hospital 10 limited English proficiency families 5 nurses 7 junior doctors 5 senior doctors 5 interpreters	•Interpreters not always present at rounds, did not interpret accurately, and did not interpret everything communicated. •Healthcare professionals recommended having medical interpreters as part of the healthcare team and standardise practices. •Coordination required to organize medical interpreters seen as a major barrier.
Choe et al. (2019) USA	To identify barriers to and drivers of effective interpreter service use when caring for hospitalized limited English proficiency children from the perspectives of pediatric medical providers and interpreters.	Qualitative Four group level assessment a structured participatory approach) Analysed into themes and generated action items.	Paediatric wards at a medical centre 10 nurses and nursing assistants 36 physicians, 18 interpreters	•Difficulties in accessing medical interpreter services. •Uncertainty among healthcare professionals re communicating with limited English proficiency families. •Unclear and inconsistent role expectations of healthcare professionals. •Unmet family engagement expectations. •Facilitators of effective communications included a team-based approach between healthcare professionals and medical interpreters; empathy for service users; and using effective family centered communication strategies.
Dungu et al. (2019) Denmark	To describe language barriers and the use of professional interpretation in two paediatric emergency units, including the medical professionals’ knowledge, experiences and practices. A further aim was to explore whether language barriers affected clinical management of children and adolescents.	Quantitative Two survey questionnaires	Two paediatric emergency departments Questionnaire (1) – All paediatric admissions over a 3-month period N=399 healthcare professionals Questionnaire (2) 17 nurses 8 doctors	•Language barriers present in over a third of non-Danish native admissions over the study period. •Almost three-quarters of healthcare professionals reported never using medical interpreters for encounters with limited Danish service-users. •Family or friends preferred as interpreters and extensively used. •Almost half of respondents to children rights to medical interpretation.
Gil et al. (2016) USA	To describe the level of satisfaction and experiences of parents and health care providers who used the limited English proficiency Patient Family Advocate while receiving or providing care.	Quantitative Survey questionnaire that included an open-ended question	Paediatric oncology 10 nurses 12 parents 5 doctors	•Use of limited English proficiency patient Patient Family Advocate results in high levels of satisfaction for parents and healthcare professionals. •Role reported to extend beyond just interpretation. •Continuity of care and safety improved. •Positive effects on communication, trust and connectedness reported.
Guerrero et al. (2018) USA	To describe physician perceptions of differences in limited English proficient pediatric encounters and the behavioral adaptations they make to provide quality care to limited English proficiency pediatric patients.	Qualitative Semi-structured interviews Content analysis	Paediatric oncology units 6 family physicians 5 pediatricians	•Language barriers created difficulties with trust, rapport, continuity of care, structure and flow, patient assessment. •Physicians made both positive and negative adaptations in interactions with limited English proficiency families, including not addressing lower priority aspects such as prevention.
Gutman, et al. (2018) USA	To describe the characteristics of emergency department discharge communication for limited English proficiency families and to assess whether the use of a professional interpreter was associated with provider communication quality during emergency department discharge.	Quantitative analysis of video-recorded discharge encounters involving limited English proficiency service-user and healthcare professionals	Paediatric emergency department 47 limited English proficiency patients (101 interactions with 42 healthcare professionals – healthcare professionals not specified)	•Almost one third of communications occurred without a medical interpreter, even though healthcare professionals were aware that the study was about use of medical interpreters. •Discharge instructions often omitted medication dosing and administration info, follow up instructions. •Return precautions and majority of encounters omitted •Medical interpreter use was associated with better discharge communication.
Gutman et al. (2020) USA	To analyse the effect of remote interpretation (telephone or video) on pediatric emergency provider attitudes and behaviors around professional interpretation.	Quantitative Cross-sectional questionnaire (embedded in randomized controlled trial with limited English proficiency service-users assigned to telephone or video interpretation)	Paediatric emergency department 161 physicians and nurse practitioners	•Frequent lapses in profession interpretation reported. •Healthcare professionals reported that deferred and delayed communication because a medical interpreter was needed were frequent. •No significant differences found in terms of healthcare professionals’ perspectives on interpreter skill and technical difficulties between telephone and video interpretations. •Yet interpretations via video found to be more satisfactory to healthcare professionals than telephone.
Hernandez et al. (2014) USA	To assess residents’ self-efficacy, satisfaction and frustration in the care of limited English proficient families and to identify individual and programmatic factors associated with the above outcomes.	Quantitative Survey questionnaire that included open-ended questions	Multi-site 271 junior doctors (in paediatric site)	•Half of doctors rated their self-efficacy in dealing with limited English proficiency families as low. •Over half reported that most of their encounters with limited English proficiency families were satisfying. •Respondents reported inefficiency in using MIs and lack of trust with interpreter skills frequently generating frustration.
Jungner et al. (2019) Sweden	To investigate communication over language barriers in pediatric oncology care. Specifically, how language barriers are overcome in different types of communication situations, how healthcare professionals relate to such language barriers, to what extent are professional interpreters or other communicational tools used, and to what extent are other individuals used to translate.	Quantitative Cross sectional survey that included open-ended questions	Paediatric oncology multi-site 151 nurses, 54 doctors 62 nursing assistants	•Most common medium used to overcome language barriers were using medical interpreters. •Family members, siblings, child themselves, colleagues used as well as self regardless of level of foreign language proficiency. •All healthcare professionals reported that care relations, patient safety and family engagement in care improved when medical interpreters were used. •The most common way to use interpreter was on-site. •Nurses used medical interpreters less frequently than doctors.
Jungner et al. (2021) Sweden	To investigate the reasons for not using interpreters to secure patient-safe communication. Methods: Healthcare personnel at six paediatric oncology centres in Sweden responded to the Communication over Language Barriers questionnaire. Descriptive and comparative analyses were performed.	Quantitative Survey questionnaire that included open-ended questions (separate analysis from study wider above)	Paediatric oncology centres multi-site 151 nurses, 54 doctors 62 nursing assistants	•Barriers to using medical interpreters were lack of time, especially in emergency situations. •Over a third of nurses and half of doctors perceived interpreters’ ability to correctly interpret medical terminology as ‘fairly bad’ or ‘bad’. •A sizeable minority of healthcare professionals unsure if service-user had received the correct information. •Significant differencess in evaluating interpreters’ abilities between healthcare professionals with and without education in using medical interpreters. •Financial constraints not perceived as a barrier to using interpreters.
Kuo et al. (2007) USA	To examine pediatricians’ provision of language services to patients with limited English pro- ficiency and the pediatrician, practice, and state characteristics associated with use of these services.	Quantitative Survey questionnaire	Multi-site cross-state (USA) 835 physicians, all fellows of American Academy of Pediatrics	•Less than half physicians utilised interpreters. •Over a quarter had used telephone interpreters. •Just over a third provided translated written material. •Over two-thirds reported frequently using bilingual family members and over half bilingual staff as interpreters. •More likely to use medical interpreters in US states with 3rd party reimbursement for interpreter services.
Kynoe et al. (2020) Norway	To explore how communication in neonatal intensive care units between immigrant mothers and nurses take place without having a common language, and how these mothers experience their stay at the unit.	Qualitative Semi-structured interviews Thematic and hermeneutic analysis	Neonatal intensive care unit 6 nurse-mother encounters observed 2 focus group interviews with 8 nurses	•Medical interpreters present for medical consultations but rarely for mother-nurse interactions. •Hands-on guidance tended to be provided by nurses, as well as body language, guesswork, and trial and error attempts at communication. •Both mothers and nurses expressed a desire to use interpreters more frequently. •Nurses used pictorial communication boards but use of this interfered with the conversation.
Lion et al. (2021) USA	To identify factors associated with professional interpreter use during pediatric emergency department visits.	Quantitative Randomised controlled trial telephone versus video interpretation for limited English proficiency families. Quantitative analysis of video recordings	Paediatric emergency department 312 communication events from 50 emergency department visits by limited English proficiency service-users (28 telephone arm, 22 video arm).	•Medical interpreters were not used all the time, even though staff knew that patients were recruited to an interpretation study. •Doctors utilised interpreters more than nurses did. •Medical interpreter was used for over a third of communications overall. •Medical interpreters used most often for detailed histories and least often for procedures and medication administration. •Medical interpreter use was inconsistent over course of emergency department visit. •Assignment to video rather than telephone interpretation was associated with greater use of medical interpreters among doctors and nurse practitioners but not nurses.
Marcus, et al. (2020) USA	To examine nurses’ positive and negative perceptions of the technology used in language interpretation for patients who have limited English proficiency.	Quantitative Survey questionnaire that included open-ended questions	Emergency department and Mother & Baby unit at two separate hospitals 47 nurses	•Video interpretation was preferred to telephone interpretation, and found to be less time consuming for nurses and more effective. •More than half of nurses perceived telephone systems as increasing stress and diverting nurses time away from patients. •A fairly small minority viewed video interpretation as impacting negatively on patient care or practice.
Patriksson et al. (2017) Sweden	To explore the experiences of health care professionals in Swedish neonatal care units regarding communication with parents of foreign origin who have difficulty understanding and speaking Swedish.	Qualitative Open-ended interviews Inductive content analysis	Five neonatal care units 25 nurses 25 nursing assistants 10 physicians	•Participants experienced inability to perform work properly, lack of guidelines, dependence on others and used their own strategies to get by. •Healthcare professionals reported powerlessness and frustration. •Physicians used interpreters more than nurses and nursing assistants. •Assistive technology and body language were adopted by nurses.
Patriksson et al. (2019) Sweden	To examine healthcare professionals’ use of interpreters and awareness of local guidelines for interpreted communication in neonatal care.	Cross-sectional survey	Thirty-eight neonatal care units 484 nurses 320 nursing assistants 54 physicians	•All healthcare professional groups reported low awareness of guidelines. •Authorised interpreters were used more for medical communications, and a lot less for nursing communications. •Physicians used authorised interpreters especially in emergency situations. •Physicians used unauthorised interpreters less than other healthcare workers. •Overall there was insufficient awareness of guidelines on interpreter-use among healthcare professionals. •Awareness of guidelines on interpreter-use was highest amongst nurses.
Pines et al. (2020) Australia	To investigate (a) when healthcare professionals consider it appropriate to use family members as interpreters and (b) what characteristics of family members healthcare professionals believe make them suitable to act as interpreters.	Qualitative Semi-structured Interviews Thematic analysis	Neonatal and paediatric department (single site) 51 nurses/midwives 11 doctors, 7 allied health professionals	•Family members were used for emergencies and for basic communication. •Some healthcare professionals highlighted a preference for a family as their first option over medical interpreter but were reluctant use them because of official policies (and a small minority actually did anyway) but knew was against hospital policy. •Healthcare professionals deemed suitable family members as ad-hoc interpreters to be those with basic English proficiency and medical knowledge, those with whom they had a good/close relationship and the child if aged >12.
Stephen (2021) USA	To understand pediatric nurses’ experiences in caring for patients and families with limited English proficiency and to explore how nurses navigate the communication gap.	Qualitative Semi-structured interviews Phenomen-ological analysis	Paediatric inpatient setting 15 paediatric nurses	•Nurses experienced emotional distress, frustrations dissatisfaction, sadness and defeat regarding communication challenges, as well as emotional and mental exhaustion. •To overcome these challenges, nurses used care-planning, assessing families’ level of understanding, utilising non-verbal cues and body language, written materials, whiteboards, and applications, ie, Google translate. •Medical interpreters were sought only for 'important' communications. •Missed connection was experienced and therapeutic relationship with families was compromised
Williams et al. (2018) (United Kingdom)	To understand the perspectives of healthcare professionals and interpreters in relation to working with and caring for non-English speaking families accessing National Health Service paediatric tertiary health care services.	Qualitative Two focus groups: one with healthcare professionals and one with interpreters Seven (unspecified) individual interviews Framework analysis	Paediatric wards tertiary setting 5 (unspecified) healthcare professionals (Focus group) 7 healthcare professionals (individual interviews) 11 interpreters	•Challenges found included uncertainty regarding the role of the interpreter (whether a team member or not) and time constraints. •Difficulties with the interpreter withholding culturally sensitive information, eg, regarding gender matters. •Providing pyschosocial support and engaging patients were reportedly challenging. •Navigating communication with limited English proficiency families was highly complex.

¹United States of America

2.5. Stage 5: collation, summarising, and reporting of results

The final stage of the Arksey and O'Malley (2005) framework was to organize the relevant findings into themes, prioritising the results based on relevance to the research question and quality of the study. The 21 included studies were reviewed for consistency as well as differences, and ultimately, their results were aggregated into themes that formed the basis of understanding existing knowledge on prevalent practices amongst healthcare professionals on the use of medical interpreters for families with limited national language proficiency in paediatric healthcare settings. Quantifying the strength of these themes would not have added value to the review, since individual studies addressed the subject of the themes to varying degrees; several had a strong focus on one or other of the themes with limited reference to other themes, while some studies offered a brief account of a wide range of issues that mediated the themes.

3. Results

Within the 21 studies included, the scope of the review spanned a population of 1,344 doctors; 813 nurses; 407 nurse assistants; 10 nurses and nurse assistants (aggregated); 111 unspecified healthcare professionals; 161 physicians and nurse practitioners (aggregated); 51 medical interpreters; and 45 families/service users. Mixed populations of healthcare professionals, parents/families, and medical interpreters in various combinations were used in over a third of studies (38.1%, n=8), while a lower proportion sampled a mix of healthcare professionals (33.3%, n=7); fewer again focused on physicians only (19%, n=4) or nurses only (9.5%, n=2). Researchers in a scoping review of language barriers between nurses and patients similarly found that studies of nurses as the primary study population were very limited (Gerchow et al., 2021). Where study sampling extended beyond healthcare professionals (i.e., included interpreters, parents, etc.) aspects of these studies that focused on the prevalent practices of healthcare professionals were foregrounded in this scoping study to ensure that the research question drove the review. For an overview of the aggregated characteristics of the study, see Table 3.

Table 3.

Aggregated characteristics of the included studies.

Research Designs All included studies	Number (Percent*) N=21 (100%)	Included Studies
Qualitative studies	N= 8 (38%)
Methodological approach/analytical technique	Below, % of qualitative studies
Grounded theory, ethnography	n=1 (12.5%)	Cheng et al. (2021)
Group level assessment (GLA)	n=1 (12.5%)	Choe et al. (2019)
Content analysis/ inductive content analysis	n=2 (25%)	Guerrero et al. (2018), Patriksson et al. (2017)
Thematic and hermeneutic analysis	n=1 (12.5%)	Kynoe et al. (2020)
Phenomenological analysis	n=1 (12.5%)	Stephen (2021)
Thematic analysis	n=1 (12.5%)	Pines et al. (2020)
Framework analysis	n=1 (12.5%)	Williams et al. (2018)
Quantitative studies Methodological approach/analytical technique	N=13 (62%) Below, % of quantitative studies
Survey questionnaire that included open-ended question(s)	n=6 (46.2%)	Marcus, et al. (2020), Hernandez et al. (2014), Gil et al. (2016) Abbe et al. (2006), Jungner et al. (2019), Jungner et al. (2021)
Survey questionnaire	n=5 (38.5%)	Burbano O'Leary et al. (2003), Dungu et al. (2019), Gutman et al. (2020),Kuo et al. (2007), Patriksson et al. (2019)
Quantitative analysis of video-recorded interactions (observation)	n=2 (15.3%)	Gutman, C. et al. (2018), Lion et al. (2021)
Data gathering techniques	N=21 (100%)
Direct observation and semi-structured interviews	n=2 (9.5%)	Cheng et al. (2021), Kynoe et al. (2020)
Group interviews	n=1 (4.8%)	Choe et al. (2019)
Semi-structured interviews	n=3 (14.3%)	Guerrero et al. (2018), Stephen (2021), Pines et al. (2020)
Open-ended interviews	n=1 (4.8%)	Patriksson et al. (2017)
Individual (unspecified) interviews and focus groups	n=1 (4.8%)	Williams et al. (2018)
Self-reported questionnaires	n=11 (52.3%)	Marcus, et al. (2020), Hernandez et al. (2014), Gil et al. (2016),Abbe et al. (2006), Jungner et al. (2019), Jungner et al. (2021) Burbano O'Leary et al. (2003), Dungu et al. (2019) Gutman et al. (2020), Kuo et al. (2007, Patriksson et al. (2019)
Indirect observation (video recordings)	n=2 (9.5%)	Gutman, C. et al. (2018) Lion et al. (2021)
Population	N=21 (100%)
Nurses only	n=2 (9.5%)	Stephen (2021) Marcus, et al. (2020)
Physicians only	n=4 (19%)	Guerrero et al. (2018), Burbano O'Leary et al. (2003),Hernandez et al. (2014), Kuo et al. (2007)
Mixed population of HCPs	n=7 (33.3%)	Dungu et al. (2019, Gutman et al. (2020), Jungner et al. (2019) Jungner et al. (2021), Patriksson et al. (2017) Patriksson et al. (2019),Pines et al. (2020)
Mixed population HPCs, service-users, medical interpreters	N=8 (38.1%)	Abbe et al. (2006), Cheng et al. (2021, Choe et al. 2019 Gutman, C. et al. (2018), Kynoe et al. (2020), Lion et al. (2021) Williams et al. (2018), Gil et al. (2016)
Country	N=21 (100%)
USA	n=13 (62%)	Abbe et al. (2006), Burbano O'Leary et al. (2003) Cheng et al. (2021), Choe et al. (2019), Gil et al. (2016), Guerrero et al. (2018), Gutman et al. (2018) Gutman et al. (2020), Hernandez et al. (2014) Kuo et al. (2007), Lion et al. (2021), Marcus, et al. (2020), Stephen (2021)
Denmark	n=1 (4.8%)	Dungu et al. (2019)
Sweden	N=4 (19%))	Jungner et al. (2019), Jungner et al. (2021), Patriksson et al. (2017), Patriksson et al. (2019)
United Kingdom	n=1 (4.8%)	Williams et al. (2018)
Norway	n=1 (4.8%)	Kynoe et al. (2020)
Australia	n=1 (4.8%)	Pines et al. (2020)

^⁎Numbers do not always add to exactly 100% because of rounding up or down

In terms of the types of paediatric settings covered in the studies, these varied: Seven studies were conducted in mixed, multiple, or unspecified paediatric locations (33%); four in paediatric emergency units (19.1%); five in oncology units (23.8%); three in neonatal units (14.3%); and two in neonatal and another paediatric location (9.5%) See Table 2 for settings.

In terms of methodological approach adopted, the majority used quantitative strategies (62%, n=13), with the remainder using qualitative approaches (38%, n=8). Of the quantitative studies, almost half had at least one open-ended question (38.5%, n=5) (these were sometimes loosely described by the authors as ‘mixed methods’, but the qualitative component tended to be minimal). As expected, of the quantitative studies, fewer (15.3%, n=2) reported observation as the data gathering technique compared to questionnaires (84.7%, n=11). All of the eight qualitative studies used either individual or group interviews, with two (25% of qualitative studies) combining observation and interviews. A wide variety of analytical strategies was used across the eight qualitative studies (see Table 3). In terms of study location, the majority were conducted in the USA (62%, n=13), followed by Sweden (19%, n=4), with just one study from each of the following countries: Norway, United Kingdom, Denmark, and Australia.

The six themes identified across the 21 studies are as follows: 1. Use, efficacy, and quality of medical interpreter services; 2. time factors associated with medical interpretation; 3. the use of technology in medical interpretation; 4. use of unauthorised interpreters; 5. impact of a language barrier on the healthcare professional-family relationship; and 6. cost of medical interpretation services. (See Table 4).

Table 4.

Themes and sub-themes from included studies.

Themes and sub-themes	Selected included studies that refer to the theme subject
Use, efficacy, and quality of medical interpreter services •Use of medical interpreter services •Underuse linked to lack of health professionals’ training •Issues with efficacy and quality of interpreter services	Lion et al. (2021), Cheng et al. (2020), Dungu et al. (2019), Gil et al. (2016), Jungner et al. (2019), Jungner et al. (2021), Kynoe et al. (2019), Patriksson et al. (2019), Abbe et al. (2006), Choe et al. (2019), Hernandez et al. (2013), Williams et al. (2018), Kuo et al. (2007)
Time factors associated with medical interpretation •Time required to organize interpreter services •Wait-time for interpreter services •Time needed to complete an interpretation encounter •Impact of technology on time relating to interpretation	Cheng et al. (2020), Jungner et al. (2021), Stephen (2021), Williams et al., (2018), Gutman et al. (2020), Choe et al. (2019), Marcus et al. (2020), Abbe et al. (2006), Guerrero et al. (2018), Gutman et al. (2018), Hernandez et al. (2013), Lion et al. (2021), Patriksson et al. (2017), Burbano O'Leary et al. (2006)
The use of technology in medical interpretation •Telephone and video interpretation services •Health professionals’ practices relating to translation applications and other media	Marcus et al. (2020), Gutman et al. (2020), Gutman et al. (2018), Lion et al. (2021), Burbano O'Leary et al. (2006), Lion et al. (2021), Patriksson et al. (2017), Cheng et al. (2020), Choe et al. (2019), Jungner et al. (2019), Kynoe et al. (2020), Patriksson et al. (2017), Patriksson et al. (2019)
Use of unauthorised interpreters •Use of family and friends as ad hoc interpreters •Use of professionals’ peers as ad hoc interpreters	Jungner et al. (2019), Kuo et al (2007), Pines et al. (2020), Burbano O'Leary et al. (2006), Dungu et al. (2019)
Impact of a language barrier on the healthcare professional-family relationship •Responses of healthcare professionals to language barriers with families •Health professionals’ practices to enhance relations with LEP families	Gill et al. (2016), Guerrero et al. (2018), Kynoe et al. (2019), Patriksson et al. (2017), Stephen (2021), Cheng et al. (2020), Choe et al (2019), Hernandez et al. (2013), Cheng et al. (2020)
Cost of medical interpretation services •Impact of cost of interpretation services on health professionals’ practices	Jungner et al. (2021), Kuo et al. (2007), Marcus et al. (2020), Dungu et al (2019)

3.1. Theme 1: use, efficacy, and quality of medical interpreter services

This theme refers to how and when medical interpreters tended to be used in paediatric settings and the training of healthcare professionals in using such services, as well as the reported efficacy and quality of interpreter services.

3.1.1. Use of medical interpreter services

As in all healthcare settings, in paediatric settings, medical interpreter services need to be organized as required, and a very consistent prevalent practice among healthcare professionals across the studies was a hesitance to arrange such services (Burbano O'Leary et al., 2006; Kuo et al., 2007; Jungner et al., 2019, 2021; Lion et al., 2021; Cheng et al., 2020, Dungu et al., 2019), even for high-risk activities such as the administration of medication and in undertaking medical procedures (Lion et al., 2021). Interpreter use has been found to be lower among nurses (other than nurse practitioners [Lion et al 2021]) compared to doctors (Jungner et al., 2019; Lion et al., 2021; Patriksson et al., 2019; Kynoe et al., 2019) in planned care other than emergencies (Patriksson et al., 2019). Nurses have been found to improvise by physical gestures, actions, picture drawings, etc. (Patriksson et al., 2017; Kynoe et al., 2019; Stephen, 2021), and in one study (Jungner et al., 2019), nurses utilised only pre-booked interpreters already on-site and available, unlike doctors who utilized interpreters when they needed them.

An impediment to medical interpreter use has been the difficulty for healthcare professionals to understand how the system of scheduling interpreters works (Choe et al., 2020) and the lack of integration of interpreter services at clinical sites (Dungu et al., 2019). Unpredictable availability and access to interpreters have also been noted (Cheng et al., 2021; Williams et al., 2018). In one study where interpreter services were integrated within the clinical site by a dedicated trained interpreter who simultaneously acted as a family advocate, self-reported outcomes in terms of standard of communication and trust were positive (Gil et al., 2019); however, this was confined to a clinical site where the vast majority of families with limited national language proficiency spoke just one language.

3.1.2. Underuse linked to lack of healthcare professionals’ training

Training of healthcare professionals on knowledge and procedures in the use of medical interpreters has been proposed, as this has been found to be inadequate (Dungu et al., 2019; Jungner et al., 2021; Hernandez, 2013). Instances have been reported where a mismatch between the language of the interpreter and family occurred because the exact language was incorrectly recorded on admission (Cheng et al., 2020). One study found a lack of awareness among healthcare professionals of minors’ right to medical interpretation (Dungu et al., 2019).

3.1.3. Issues with efficacy and quality of interpreter services

Where medical interpreter services are used, healthcare professionals have reported some difficulties with the efficacy and quality of services (Hernandez, et al., 2013; Abbe et al., 2006; Cheng et al., 2020; Pines et al., 2020; Guerrero et al., 2018; Williams et al., 2018; Jungner et al., 2021). Some issues are interpreter-specific based on the healthcare professionals’ faith or confidence in the ability and approach of the interpreter. These issues include healthcare professionals’ reservations about: misinterpretations, incorrect translations, and inabilities to translate medical terminology (Jungner et al., 2021); incomplete translations (Cheng et al., 2020; Williams et al., 2018); uncertainty regarding interpreters’ competency levels and accuracy to translate (Hernandez, et al., 2013; Abbe et al., 2006); rigid structure and ‘diluted’ dialogue (Guerrero et al., 2018); lack of agreement around the interpreter role as advocate or mere translator (Williams et al. 2018); and lack of professionalism (Pines et al., 2020; Jungner et al., 2021).

Healthcare professionals’ suggestions for service improvements include the hiring of more interpreters, increasing access to telephone and video interpretation services (Cheng et al., 2021), and establishing in-house interpreters (Choe et al., 2019; Williams et al., 2018).

3.2. Theme 2: time factors associated with medical interpretation

Theme 2 captures prevalent practices of health professionals in paediatric settings relating to medical interpreter use that focus on the concept of time. It includes reports of the time needed to organise interpreter services; wait time to initiate the service; and time needed to complete an encounter, as well as the impact of the use of technology on time.

3.2.1. Time required to organize interpreter services

The underuse of medical interpreter services (identified in the previous theme) found in paediatric settings was related to the subject of this second theme, namely, time. Across the 21 studies reviewed, the most consistent factor raised by healthcare professionals that reportedly impacted on their practices regarding use of medical interpreters was time. Interpreter services were time consuming to organize (Patriksson et al., 2017; Stephen, 2021, Williams et al., 2018; Jungner et al., 2021, Gutman et al., 2020; Choe et al., 2019; Marcus et al., 2020 Guerrero et al., 2018; Hernandez, et al., 2013) and particularly challenging in an emergency setting where time-sensitive decisions needed to be made very rapidly and prearranged interpreter services were not feasible (Pines et al., 2020; Jungner et al., 2021).

3.2.2. Wait-time for interpreter services

Wait time for the interpreter service to initiate has also been problematic (Abbe et al., 2006; Burbano O'Leary et al., 2006, Stephen, 2021). One study found that even where booking was made in advance, teams persistently experienced the service as unpredictable (Cheng et al., 2021). Dovetailing the timing of ward rounds with the presence of an interpreter has also been found to be problematic, in some cases resulting in the interpretation session being abandoned altogether (Cheng et al., 2020). In addition, the time for interpretation has been found to interrupt the workflow (Stephen, 2021; Cheng et al., 2020). There have also been instances where interpreters have been left waiting to fit in with medical team consultations (Cheng et al., 2021).

3.2.3. Time needed to compete an interpretation encounter

The time it takes to complete an interpreter-facilitated encounter has also featured in studies that found that this takes considerably longer than a regular encounter (Guerrero et al., 2018; Hernandez, et al., 2013). It has been reported that healthcare professionals need to speak clearly and slowly in shorter sentences than usual to enable interpreters to capture the information to convey, requiring extra time (Patriksson et al. 2017).

3.2.4. Impact of technology on time relating to interpretation

To save medical interpretation and wait times, technology to communicate with families with limited national language proficiency, especially video remote interpreting systems, have been considered (Marcus et al., 2020; Gutman et al., 2018, Lion et al., 2021). These video systems have been found by healthcare professionals to be less time consuming than telephone communication (Marcus et al., 2020; Lion et al., 2021). However, the introduction of new technology takes clinicians time to learn and requires training, adding to workload in the short term (Marcus et al., 2020), and using a trained medical interpreter, even for a remote session, takes time to organise (Gutman et al., 2018). Where medical interpretation sessions have been structured around set booking systems, either via phone or online, an adverse impact was that healthcare professionals allowed insufficient time for effective translation (according to the interpreters) (Williams et al. 2018). In the study in question (Williams et al. 2018), communication with families with limited national language proficiency was limited to set slots, and interaction outside these slots was avoided by healthcare professionals.

3.3. Theme 3: the use of technology in medical interpretation

Theme 3 relates to findings from the review on when and how technology and media are used for medical interpretation in paediatric settings.

Several studies in this scoping review provide some understanding of healthcare professionals’ prevalent practices relating to the use of technology to facilitate medical interpretation in paediatric settings (Marcus et al., 2020; Gutman et al., 2020; Gutman et al., 2018; Lion et al. 2021; Burbano O'Leary et al., 2006; Lion et al., 2021; Patriksson et al., 2017; Cheng et al., 2020; Choe et al., 2019; Jungner et al., 2019). Use of technology for interpretation purposes appears to vary across paediatric locations.

3.3.1. Telephone and video interpretation services

A number of studies have compared medical interpretation services provided by telephone with those provided by video. While some studies have found telephone interpreter services to be unsatisfactory (Burbono O'Leary et al., 2006; Patriksson et al., 2017), healthcare professionals have commended them for providing services at short notice in critical situations and for enabling confidentiality in small minority communities where the interpreter may be known in the minority network (Patriksson et al., 2017).

Overall, while Gutman et al.’s (2020) randomised control trial found no significant differences in terms of interpreter skill and technical difficulties between telephone and video interpretations, the latter have been found to be more satisfactory to healthcare professionals (Marcus et al., 2020; Gutman et al., 2020). Video remote interpreting has been credited with enhancing care standards without compromising healthcare professionals’ autonomy (Marcus et al., 2020). Several studies credited the superiority of video over telephone to the fact that eye contact, non-verbal cues, and nuances are captured on video, resembling more closely an in-person encounter than a telephone communication (Marcus et al., 2020; Patriksson et al., 2017). Conversely, the lack of visual cues as happens with telephone interpretation can result in conversations being misunderstood (Patriksson et al., 2017). However, connectivity problems with using videos have been found to be a deterrent to their use and acceptance (Marcus et al., 2020; Choe et al., 2019). One study found that doctors expressed frustration with both telephone and video technologies because of problems with the reliability of equipment, with locating it when needed, and with lack of familiarity with using it (Choe et al., 2019).

3.3.2. Healthcare professionals’ practices relating to translation applications and other media

Healthcare professionals have found online translation applications like Google translate to be useful in a limited range of communication situations (Patriksson et al., 2017; Patriksson et al., 2019; Kynoe et al., 2020), such as in answering brief questions (Kynoe et al., 2020) and as a “quick fix” for immediate communication (Patriksson et al., 2017). Nurses have reported it to be “somewhat helpful”, albeit with some translation inaccuracies (Kynoe et al., 2020). However translation applications have been deemed to be no substitute for a medical interpreter (Patriksson et al., 2017). Media like glossaries and key-word charts have been found to be less useful (Kynoe et al., 2020).

3.4. Theme 4: use of unauthorised interpreters

Theme 4 relates to the review's findings on the use of unauthorised interpreters, such as family, friends, and professional peers as interpreters in paediatric settings.

3.4.1. Use of family and friends as ad hoc interpreters

Studies report that, in the absence of professional medical interpreters, healthcare professionals often resort to ad hoc interpreters; e.g., family members, friends, siblings (Burbano O'Leary et al., 2006; Abbe et al., 2006; Kynoe et al., 2020; Pines et al., 2020; Patriksson et al., 2019; Dungu et al., 2019; Jungner et al., 2019), and even the child themselves interpreting for their family (Pines et al., 2020; Jungner et al., 2019), with mixed responses. One study found that more than a third of healthcare professionals reported that family/friends were the preferred type of interpreter, a higher proportion than those whose preference was for a face-to-face or telephone interpreter (Dungu et al., 2019). Positive reports from healthcare professionals relate to using ad hoc translators for relaying basic day-to-day information at the bedside, but not for communicating information like diagnoses, or end of life and medico-legal conversations for reasons relating to emotional distress, privacy, confidentiality, and translation accuracy (Pines et al., 2020). Specifically, unauthorized interpreters tended to be used more commonly for information about nursing care and less so for communication about medical treatment (Patriksson et al., 2019). However, the use of ad hoc interpreters by healthcare professionals in delicate and sensitive situations has been reported owing to time constraints or in emergency situations or where the ad hoc interpreter has demonstrated a strong and close relationship with the child and family (Pines et al., 2020).

In a national USA survey of paediatricians, those in small and rural practices used ad hoc interpreters (especially family members) significantly more than the rest of the sample, independent of whether costs were covered by a third party (Kuo et al., 2007). The reasons for this were not established, but availability of medical interpreters may have been a factor (Kuo et al., 2007).

In some healthcare settings, policies prevail that prohibit the use of unauthorsied interpreters from medical interpretation (Pines et al., 2020) or at least strongly discourage their use (Gutman et al., 2020).

3.4.2. Use of healthcare professional peers as ad hoc interpreters

Healthcare professionals themselves or colleagues proficient in the language have also been used as ad hoc interpreters (Burbano O'Leary et al., 2006; Kuo et al., 2007; Jungner et al., 2019). In Jungner et al.’s (2019) study, healthcare professionals reported that a colleague ‘sometimes’ translated in encounters with families with limited national language proficiency.

3.5. Theme 5: impact of a language barrier on the healthcare professional-family relationship

Theme 5 relates to the impact of a language barrier on the relationship between the health provider and the family, health professionals’ responses to these barriers and the practices they report using to enhance relations.

Without adequate interpretation services in communicating with families with limited national language proficiency, the language barrier has been found to impact on healthcare professionals’ ability to develop a relationship and rapport with the families of children in their care (Patriksson et al., 2017; Choe et al., 2019; Gil et al., 2016; Jungner et al., 2021; Cheng et al., 2020). The loss of communication nuances in translation and the disconnection has also been found to place a strain on the healthcare professional-family relationship dynamic and hinder trust (Jungner et al., 2021; Guerrero et al., 2018). Challenges and lack of clarity predominantly occur regarding provision of emotional support and interpretation (Choe et al., 2019; Williams et al., 2018).

3.5.1. Responses of healthcare professionals to language barriers with families

The impact of language barriers on healthcare professionals has been frustration (Hernandez, et al., 2013; Patriksson et al., 2017; Guerrero et al., 2018; Stephen, 2021); feelings of inadequacy in delivering effective care (Hernandez et al., 2013), and uncertainty (Guerrero et al., 2018), as well as dissatisfaction, sadness, and defeat (Stephen, 2021). More generally, healthcare professionals find dealing with families with limited national language proficiency daunting and overwhelming (Choe et al., 2019), which results in their avoiding and limiting interactions with such families, leading to amplified feelings of neglect and isolation for these service users (Stephen, 2021).

While deficits in medical interpretation services were often framed as impediments to desirable healthcare professional-family relationships and connectedness across the studies, some researchers found a sense of apathy on the part of healthcare professionals for making special accommodations for families with limited national language proficiency. Examples include healthcare professionals expressing dismay that these families could not speak the country's language (Williams et al., 2018) and lack of patience when answering their questions (Choe et al., 2019).

3.5.2. Healthcare professionals’ practices to enhance relations with families with limited national language proficiency

In the absence of adequate professional medical interpretation, healthcare professionals have been found to use strategies to demonstrate engagement with the family, such as expressing interest in the child's family life outside their illness, by making an effort to learn a few words in the families’ native language (Guerrero et al., 2018), or by smiling and using an empathic voice tone (Kynoe et al., 2020). In one USA study, an interpreter service was integrated with a family advocate role at a clinical site where a very high proportion of families spoke Spanish and was found to have positive impact on relations between healthcare professionals and service users (Gil et al., 2016).

3.6. Theme 6: cost of medical interpretation services

Theme 6 relates to costs associated with medical interpreters.

3.6.1. Impact of cost of interpretation services on healthcare professionals’ practices

A small minority of studies referred to the subject of cost relating to medical interpretation and its potential impact on the practices of healthcare professionals working in paediatric care. A Swedish study found that the cost of interpretation was not considered a barrier to interpreter use, with the vast majority of healthcare professionals in this study reporting that economic restriction was ‘never’ a reason for not using medical interpreters, even in the context where healthcare professional were occupationally socialized to be frugal with resources (Jungner et al., 2021). In a USA study, nurses were found to be undecided as to whether or not telephone interpreter services increased overall hospitalisation costs (Marcus et al., 2020). In Dungu et al's study (2019), just one of the 25 healthcare professionals studied reported that lower cost influenced the type of interpreter used. Researchers from the USA have found that, where language services were reimbursed by a third party, professional interpreters were more likely to be used (Kuo et al., 2007), suggesting that in this context at least, cost did matter.

4. Limitations

This scoping review was limited to understanding the practices of healthcare providers from their own perspectives and did not include studies where either medical interpreters themselves or service providers comprised the study population. Clearly, a fuller picture of interpreter services would have been gained from considering all perspectives, but the analysis would have become diluted had we included multiple populations. In addition, only primary research articles were included (and thus not wider literature; e.g., grey papers) owing to the timeframe of this project.

5. Discussion

The foregoing scoping review aimed to explore primary research into the prevalent practices of healthcare professionals in using medical interpreters for families with limited national language proficiency in paediatric healthcare settings. This review identified 21 studies that met the inclusion criteria, indicating that there is a very limited amount of research available specific to paediatric settings over a fairly extended period (2000-22) considering the degree of migration that has occurred over this period. The studies were concentrated in the USA (62%) and Scandinavia (19% when different data sets from the same broad study were accounted for), with just one study each in the United Kingdom and Australia.

With regard to the extent to which findings of our review reflect findings from similar scoping or systematic reviews in clinical settings beyond paediatrics, the capacity to make comparisons is limited by our finding no reviews with the same focus as ours on wider patient groups. Reviews on language barriers that cover a broader span of service users have tended to focus on access to healthcare in the context of language barriers and the impact of medical interpreter use on clinical care. Nonetheless, Ramirez et al.’s (2008) review on interpreter use in emergency departments (in general) yielded similar findings to our review based specifically on paediatric settings, such as an under-utilization of interpreter services at emergency departments. The findings of studies included in Gerchow et al.’s (2021) review of language barriers between nurses and patients (inclusive of, but extending beyond those in paediatric settings) also reflected those of our review, namely the concern nurses expressed about the accuracy of interpreter translations. No review covering wider clinical settings was located that provided information on how children were used to interpret on behalf of parents in the case of adult service users.

Returning to our scoping review, regardless of the country of the study site, there were recurring themes across the studies that crossed geographical boundaries. The substance of these themes featured to a greater or lesser extent in most of the reviewed studies, while a small number of themes were specific to a just a few. We found that prevalent practices of healthcare professionals in using medical interpreters for families with limited national language proficiency in paediatric settings include interpreter underuse owing mainly to the additional time required to organise and engage with the service, and to a lesser extent, poor experiences with interpreter services. Healthcare providers, especially nurses, have been found to improvise in managing communication with families with limited national language proficiency to engage them in care without using formal interpretation services, and ad hoc interpreter use was a common practice. Practices in using technology for interpretation varied, and where used, video tended to be favoured over telephone, provided the former was reliable and user-friendly. Cost of interpretation services as an influence on health provider practices actually featured very little, and where it did, the influence of this varied according to the location of the study.

There were clear interconnections between several of the themes that featured in the studies reviewed, particularly between the underuse of interpreter services (Theme 1); time restraints (Theme 2); and the impact of a language barrier on the healthcare professional-family relationship (Theme 5). The themes are discussed together to draw out their linkages and the complexity of using interpreter services in clinical settings. Other key findings of the scoping review were woven in as appropriate or revisited further on.

A major impediment to using medical interpreter services was time. Interpreter services need to be organised, which takes staff time; while the style of interpreters can vary from simply translating to a more interactive approach (Meeuwesen et al. 2010), interpretations require pauses to allow the translation to happen. Verification for understanding is needed, taking more time than a regular encounter. It is widely accepted that clinical sites are busy places and ideally would require additional staff to accommodate the extra time associated with interpretation. However, this is unlikely to address all of the challenges relating to the structural complexity of clinical work. A related and more fundamental problem at clinical sites was the challenge of coordinating the interpreter presence with a suitable time for clinical staff in the context of the normal ebb and flow of clinical units. This was found to be a difficulty for a range of healthcare professionals at most healthcare settings.

The underuse of interpreters by nurses, a consistent finding, merits special consideration here, and most likely reflects their more continuous presence with patients. Solutions like increasing staff ratios or providing more interpreters are likely to be of limited success in meeting all of the challenges relating to nurses’ underuse of interpreters because of the very nature of nursing work. The need for consistency in nursing allocations to families with limited national language proficiency and for nurses to book the same interpreter was acknowledged in the some of the studies (Choe et al., 2019; Cheng et al., 2021), but overall, analyses of the inherent difficulties of using interpreter services to convey nursing communication was limited to a small number of studies (Stephen 2021, Kynoe et al., 2020). More than two-thirds of included studies used very broad samples inclusive of multiple healthcare professionals and sometimes parents and interpreters, diluting the space available for a critical account of the complexity of using interpreter services with reference to nursing work. There is a lack of research into hospital nurses' usage patterns of interpreter services, and clarification is needed about when it is essential for hospital nurses to use interpreters with patients and when it is not.

In some studies, we found useful insights into the interpersonal engagement by nurses with families with limited national language proficiency in the absence of an interpreter, such as using non-verbal cues to depict engagement; e.g., smiling when appropriate (Kynoe et al., 2020) or remaining in the patient's room longer to enhance the parents’ comfort (Stephen, 2021). Both Stephen's (2021) and Kynoe et al.’s (2020) analyses were the most developed with reference to providing insights into the challenges of holistic nursing in the context of language barriers. There is a need for future research to further consider the challenges of interpreter use for nursing, with reference to existing nursing theory that emphasises the importance of embedded signals of care and connection through body language and “gestural messages” that convey interest and compassion (Peplau, 1997), p.164. Theorists such as Kitson (2018, p.101) have drawn attention to the nurse-patient relationship as an integrated connected relationship and not a “one-time event” but “requires the ability to authentically engage in rapid moments . . .” Links to a nursing theory such as Kitson's would enhance an understanding of the challenges in using medical interpreters in delivering nursing care in the context of language barriers.

Studies reviewed, and indeed beyond those reviewed (Fatahi et al., 2010), note that there are aspects of nursing work where a clear cognitive understanding of information by the service user is vital for patient safety and positive health outcomes. Kynoe et al. (2020, p.2228) observed that in these situations, nurses had challenges conveying “information in a deeper sense”. In Stephen's (2021, p. 694) study, nurses used interpreters for communicating “official” information, laboratory or x-ray results, and changes to physician orders, but not for routine care planning. Medical interpreters, however, are not the sole solution; rather they may be viewed as a necessary supplement to continuous nursing engagement with families with limited national language proficiency.

If we return to the practices of health professionals more generally in the context of language barriers, some studies (Cheng et al., 2021; Williams et al., 2018) have reported that healthcare professionals clustered education sessions and briefings into a single interpreter booking. This was found to lead to information overload for the families, resulting in shallow and rushed interpretation sessions. It seems that limited national language proficiency service users feel the impact of disjointed relationships with healthcare providers; in a scoping review of service user experiences, (Yeheskel and Rawal, 2019) identified studies where families with limited national language proficiency expressed dissatisfaction with a lack of continuity of care.

The included studies have proposed solutions to aspects of interpreter use by healthcare professionals that are amenable to change. There is a need to strengthen organisational systems for using medical interpreters in paediatric units. Suggestions included increasing the time allocation for nurses caring for children from families with limited national language proficiency (Jungner et al., 2021) and better training of healthcare professionals in how to organize and use interpreters (Dungu et al., 2019; Jungner et al., 2021; Hernandez, 2013). Findings of the included studies also identified the need for improved and easier booking processes for obtaining medical interpreters, which requires careful organizational oversight. The use of on-site interpreter systems has been proposed (Jungner et al., 2021) and found to work well (Choe et al., 2019), but this is really feasible only where all or most families with limited national language proficiency speak the same language, which is not the case at many clinical sites.

On the whole, included studies tended not to frame their data analyses with reference to the broader complex context of family dynamics and cultural practices, the complexity of paediatric contexts that involve parents and multidisciplinary teams, and wider structural inequalities that mediate the lives of many families with limited national language proficiency. Williams et al (2018) was a notable exception here, where good insights were offered into the interconnection between the immediate and wider factors that related to communication challenges with families with limited national language proficiency.

Turning to the use of technology by healthcare professionals for interpretation purposes, these are likely to strongly influence the future practices of healthcare professionals following the Covid-19 pandemic, which saw dramatic improvements in remote video communication. Moreover, communicating by video has become normalised for both work and non-work communication. Even before the pandemic, studies reviewed indicated that video remote interpreting systems were generally well-received by healthcare professionals and held promise. Advancements in technological video systems since the pandemic potentially now mean rapid access to a range of remote interpreter services at geographical locations far from the clinical sites at which the children of families with limited national language proficiency are being cared for. However, as emerged in the studies reviewed, for technologically-mediated systems to be accepted, they must be reliable and user-friendly, and healthcare professionals must know how to operate them. This requires an organisational investment in equipment and training, and importantly, in identifying and sourcing high quality medical interpretation services to address the deficits in interpretation quality that studies in this review identified.

The few studies that reported on whether cost influenced interpreter use by practitioners suggested that it tended not to (with one exception), but the difficulties that were experienced with the quality of interpretation and lack of training in how to use interpreter services may signal cost saving measures higher up at institutional level. Institutional investment by health provider organisations may well determine the modes of interpreting (e.g., remote or face-to-face), as well as the standard of interpretation services used; more cost-effective services may not provide the highest quality interpretation. A Swiss study of primary care practitioners found that lack of financial coverage of interpreter services was a deterrent to their use (Jaeger et al., 2019). Further research, especially qualitative research with its capacity to gain an in-depth understanding of organisational dynamics, is required to consider the impact in clinical settings of informal messages about costs and interpreter services. This may provide insights into whether there is an implicit understanding that interpreter services incur costs that might subtly influence practitioner interpreter use.

6. Conclusion

Overall, the studies reviewed provided a fairly consistent body of empirical evidence about the prevalent practices of healthcare professionals in using interpreter services in paediatric settings. Further scholarship is needed to understand the use of medical interpreter services with reference to the complexity of health work, as well as the social context of inequality for many families with limited national language proficiency accessing paediatric healthcare. Going forward, remote interpreting offers enormous potential, but its success will rest as much with the standard of interpreting as with the reliability and user-friendliness of the technology itself. Our review also highlights the need for better training for and regulation of medical interpreters and national standards of practice for medical interpreting in Ireland and across Europe with a focus on service users with limited national language proficiency.

What is already known

• •Most high-income countries across the globe have experienced increasing levels of migration in recent decades and need to provide healthcare to migrants who do not speak the receiving country's spoken language.
• •Available evidence on the effectiveness of interpreter services in healthcare settings is limited, but overall, where language barriers exist, trained medical interpreters have been found to provide more accurate and comprehensive communication between clinicians and service users compared to ad hoc interpreters or no interpreter at all.
• •In paediatric settings, the engagement of families with limited national language proficiency is required to navigate communication about the child's condition; however, the nature and extent of research evidence about when and how medical interpreters are used by health professionals to mediate communication with these families has not yet been published.

What this paper adds

• •We identified a consistent pattern across the studies of underuse of medical interpreters in paediatric settings owing to: issues with efficacy and quality of the interpreter services; lack of staff training in how to use them; time restraints for organizing, waiting for, and using the services; and at a small number of locations, the cost of services.
• •Nurses were less likely to use medical interpreters than doctors, and we considered this finding with reference to the way in which nursing work is structured.
• •We have highlighted the importance of technology for future medical interpretation, given recent rapid advances in use and user acceptance of video.

Funding

The research received funding from the University College Dublin Sr. Antoinette Kelleher Scholarship Award scheme

Declaration of Competing Interest

None