Implementing a digital communication assistance tool to collect the medical history of refugee patients: DICTUM Friedland - an action-oriented mixed methods study protocol

In this study, we explore DCATs usability, accuracy and the helpfulness as well as its impact on patient’s utilisation of medical services (re-consultations). By doing so, perspectives from all stakeholders (GPs, refugee patients and healthcare system insurers/providers) are taken into account.

The results of this study are expected to indicate the effectiveness of DCAT with respect to improving the communication with medical professionals and consequently medical outcomes for non-German speaking patients of the healthcare encounter. The study is located at a refugee camp in Lower Saxony in Germany. A daily GPs consultation hour at a primary healthcare centre in the camp is offered for acute medical conditions for all camp residents.

The first study phase follows an action-oriented approach, meaning that observation outcomes can be used as an input to improve the content and implementation process of DCAT. This pragmatic approach allows us to make adjustments within a complex intervention in a barely known field, in response to unexpected problems and new opportunities that arise throughout the study. In contrast to other interventional procedures, the action research method requires close collaboration between researchers and other involved parties in the field [14, 15]. This approach also has proved its validity in evaluating health-related issues among refugees and asylum seekers [16, 17]. The “action” follows a circular process consisting of planning, implementation, observation and reflection, which is repeated several times. Since it is not possible to ascertain which challenges may be faced during the testing, it is difficult to determine ahead of time the duration of the action research method. This initial study phase ends with a successful three week preliminary testing with patients supported by intensive field observation and technical backup. Following this testing, the final version of DCAT will be fixed and will remain unchanged during the second study phase.

For a better understanding of DCAT impact, potential and limitations, a mixed methods research design is to be combined with quantitative and qualitative research methods in the second study phase [18, 19]. The quantitative approach includes DCAT-users and non-DCAT users and therefore offers the possibility to compare both groups in an interventional study design.

The primary outcome will be the consultation rate of patients who used DCAT compared with those who did not use this digital translation tool. As we found in a previous pilot study, the consultation rate of refugee patients at the primary healthcare centre is higher, especially compared to the non-refugee population, and this might be a result of miscommunication. Also, consultation rates are key values for healthcare costs.

As a secondary outcome, the following are to be assessed:

For a deeper insight into the acceptance of DCAT users, we will apply qualitative research methods, namely:

The research design of the DCAT Study is shown in Fig. 1.

The study takes place in the Friedland transit camp “Grenzdurchgangslager Friedland”. The camp was established in 1945 in the middle of Germany, Lower Saxony. Since its foundation, it has received over 4 million people. Nowadays, it accommodates two groups of refugees: resettlement refugees (who have already gone through the refugee-status determination process) and asylum seekers (who have applied for asylum after arrival in Germany and are waiting for the decision regarding their refugee-status). Currently, there are 200–400 resettlement refugees and asylum seekers mainly from the middle east and north Africa countries [23]. The primary healthcare centre in Friedland is an outpatient clinic staffed by a team of six GPs offering medical consultations to refugees and asylum seekers during weekdays. These doctors work in co-operation with nursing staff from Malteser International, which also operates the facility outside regular consultation hours [24].

According to the Services for Asylum Seekers Act (AsylBLG sections 4 and 6) [2], only primary healthcare services for acute medical conditions are insured [11]. These conditions represent diseases and complaints commonly examined by primary healthcare doctors (medical emergencies, acute conditions, pregnancy and vaccinations) [25].

The digital communication assistance tool, which will be implemented and evaluated in the study, was developed by aidminutes GmbH, a former study group at Leuphana University in Lüneburg and the Department of General Practice at the Göttingen University Medical Centre.

DCAT is able to collect medical histories in 13 different languages/dialects (Modern Standard Arabic, Arabic Syrian, Arabic Egyptian, Arabic Tunisian, Arabic Moroccan, Turkish, Persian, Kurdish Sorani, Kurdish Kurmanji, Kurdish Feyli, Pashto Kandahari, Pashto Mazurka) and generate a translated summary in German for the German-speaking doctors (see Figs. 2 and 3).

The questioning structures are partly based on the concept of “programmed diagnostics in general medicine” by Braun and Mader [14], a cybernetic approach for gaining information on common signs and symptoms as well as diagnostic procedures. Several GPs with significant experience in the medical care of migrants have developed the DCAT medical history algorithm and overseen the whole programming process. Professional interpreters translated the DCAT content in its entirety. Using DCAT, the patient is asked to specify a symptom (e.g. a headache) or concern (e.g. prescription of a medication). It is possible to include additional symptoms and order them by importance. At the same time, the symptoms input can be elaborated on through further inquiry and via the possibility of visual information.

Participating in the study and utilising DCAT is voluntary. Written informed consent is collected from each patient (or from the parents for patients under 18 years) willing to participate before using DCAT. All participants have the right to withdraw from the study at any point and all their collected information is immediately deleted.

Since the DCAT study is restricted to non-German-speaking patients, who might also have difficulties in reading and writing, audio- and visual content in the chosen language has been incorporated into DCAT [15, 16].

The potential participants are informed of the DCAT study directly on their arrival in Friedland camp through interpreters and social service workers. There are also posters and flyers (available in different languages) describing DCAT distributed across various sites in Friedland. Due to the structure of the referral system in Friedland, which mandates that asylum seekers need to get a permit from the responsible social service worker before being referred to the primary healthcare centre, we also included the social service bureau as a recruiting location in this study (see Fig. 4).

The sample size calculation is based on the average numbers of patients in the last 12 months in Friedland. Due to fluctuations in refugee flows (ranging from 285 to 636/month), it is difficult to predict how many patients will be treated at the primary healthcare centre during the study period and meet the inclusion criteria, especially regarding the available languages. Despite these uncertainties, during the six months recruitment period we aim to include a total sample size of 880 participants (study group: 440 patients using DCAT, 440 patients not using DCAT). The sample size calculation is based on the consultation rates for medical treatment at the primary healthcare centre in Friedland as this is the primary outcome of the study protocol. Our previous research in the preparation phase revealed an average consultation rate for medical treatment of 2.65 (SD: 3.44) consultations per patient. Assuming a 10% dropout rate, to measure a minimal difference of 0.65 consultations per patient, 5% significance (two-sided) and a standard deviation of 3.5, 20 patients should be recruited per week to use DCAT to achieve the identified sample target.

Every patient in the DCAT study will receive a unique digital readable label (QR-Code), which is placed on his/her identification card (ID) in order to pseudonymise patient’s data. The social service workers issue these IDs to all refugees and asylum seekers after their arrival to Friedland. DCAT is launched after scanning the QR-Code. Therefore, the printed report, which has the QR-Code, can only be reassigned back to the patient by medical staff at the primary healthcare centre. Thus in case a report is lost, no one can identify it, which safeguards the patient’s personal as well as medical data.

Personal information allowing patient identification (e.g. first and last name) is not collected by DCAT. The collected data is not shared with other authorities or third parties. All other information not described above, e.g. health-related information given by the patient through DCAT is deleted permanently. The pseudonymised patient notes are evaluated by the research team and coded according to diagnosis (ICD-10 codes) [22].

All quantitative and qualitative data is only accessible by the study team. The study team is familiar with data privacy regulations and committed to data protection principles. A data monitoring committee is not needed in this trial because of the relatively short study period and the minimal risks to patient safety using the DCAT in comparison to randomised controlled trials on drugs and biologics.

All collected data will be permanently deleted ten years after the study.

The quantitative data will be analysed using SPSS (version 25.0) on the basis of descriptives (i.e., frequencies and means). Differences between patients of the study group and the control group will be tested for statistical significance. Furthermore, differences between patients using DCAT and their physicians regarding their opinions on the perceived quality of communication and mutual understanding throughout the consultation will be tested for significance.

The qualitative data (interviews with GPs, nurses and patients) will be audio-recorded. Two members of the team will code the interview transcriptions and analyse them by the summarising qualitative content analysis [31].