Study area
The study was conducted at the pediatric ED of the University Hospital of Bern, Switzerland. This institution provides the full range of tertiary emergency care to children and adolescents aged 0 to 16 years including surgical, traumatological and pediatric conditions. It provided emergency care to 23,274 patients in 2021. The hospital is part of the Swiss Hospitals for Equity Network [32] since 2021. A phone interpreter service is available at the facility around the clock. It is free of charge for patients and the costs are covered by the department.
Study design
This study was a pre- and post-intervention study analyzing the use of interpreter services for LLP families (Figure 1). It included families presenting to the ED between April 1st and June 30th 2021 (pre-intervention period) and between October 1st and December 5th 2021 (post-intervention period).
During the three months between the data collection periods, a package of interventions was implemented in the ED. The primary outcome was defined as the proportion of LLP families that received an interpreter before and after the package of interventions (Figure 1). Secondary outcomes were the comparison of the self-reported versus the assessed language proficiency of caregivers, and their knowledge about the interpreter service.
Study population
From all patients presenting to the ED during the data collection periods, administrative health records were screened and those fulfilling all inclusion criteria were identified. Inclusion criteria were: i) nationality with national language other than German, French or English ii) not presenting on the COVID-19 track. All patients fulfilling these two criteria were systematically screened for their caregiver’s language proficiency within one week after consultation. This was done by phone call interviews with the person who had accompanied the patient during the consultation. In the case of two caregivers present at the consultation, the one with better language skills was screened. A score, validated for the classification of language proficiency was used ranging from A1 (very limited language proficiency) to C2 (excellent language proficiency) [33]. Caregivers with a language level of A1 or A2 (limited language proficiency) or those asking for an interpreter by themselves were defined as LLP families. Patients with a good language proficiency ≥ level B1, were defined as language-proficient families. Those not answering several phone calls or not giving informed consent for the screening were excluded from the final analysis (Figure 2).
Intervention
With the aim of increasing the use of interpreter services at the pediatric ED, a package of different activities was provided to the entire emergency team between July 1st and September 30th 2021. It consisted of the following three parts:
Blended transcultural training
All health workers were asked to complete the national department of health’s official e-learning module of about one hour on transcultural competence in healthcare. The in-person part of the transcultural training was conducted on four different dates for all medical staff. The training took two hours and was conducted by the study team and an expert on intercultural communication from the Swiss Red Cross. The training’s first part focused on awareness raising and on equality and equity in healthcare. It emphasized the ability to communicate as a precondition for equity. It also provided practical instructions and discussions about how to use the ABC screening tool to assess the need for interpreter services in all LLP families at the point of triage and how to order the interpreter services (Figure 3).
Continuous awareness raising
To keep the awareness up and to integrate the new skills into daily routine, several reminders were placed in the departments’ weekly news. The topic was the theme of the month in the form of a visual display of information at the workplace (mid-September-mid-October 2021). Finally, the topic was highlighted during team meetings and the information was integrated in the regular orientation program for new health workers at the department.
Introduction of a language proficiency pathway
Originally created at the Western Sydney Local Health District, the ABC-tool is a systematic tool to detect the need for a professional interpreter [34]. The tool was adapted to the local context by the study team (Figure 3). The health workers at triage categorized individuals into different language competency groups. If the caregiver was comfortable answering the screening-questions, the family was classified as language proficient. If the caregiver did not understand or was not able to answer the questions, the family was classified as LLP. “Interpreter-yes" was noted in the personal electronic health record. If the health worker at triage was not sure about the comprehension and communication skills, the caregiver was asked if he or she wanted an interpreter. If the answer was “yes”, “Interpreter-yes” was also noted in the personal electronic health record. After triage, the responsible physician or nurse actively offered and booked an interpreter online for families with a noted "yes" in the record or also if communication problems had arisen in the meantime.
Data collection and analysis
The following variables were extracted from administrative health records: nationality, age, gender and date of visit. The data on interpreter use was extracted from the electronic bill of interpreter services every month, including the intervention period. Two members of the study team were trained in the use of the validated language proficiency classification of the Goethe institute [33]. They performed phone call screenings with the families meeting the inclusion criteria. The families identified as those with LLP were called with a professional interpreter and asked for consent to participate in the study. From those agreeing to participate, the following variables were collected during a phone interview: native language, assessed language proficiency, self-reported language proficiency and details about the interpreter use. Deidentified data was transferred to a REDcap-database (Vanderbilt University/ Version 11.1.4 2022). STATA (Stata/MP Version 16.1. 2020) was used for the statistical analysis and generation of graphs.