There are currently around 150,000 persons living with HIV in France, and late diagnosis is an important public health problem. In addition to the individual therapeutic benefit, early diagnosis contributes, according to modeling, to primary infection prevention [1,2]. Early antiretroviral treatment initiation inhibits the multiplication of the virus, reducing the viral load and therefore the risks of future contamination. This strategy, however, requires extended screening [3,4]. Migrants (people born outside France and of non-French nationality at birth) are a population particularly at risk of the late diagnosis of HIV, as well as hepatitis B and C [5-7]. Migrants are also more prone to late diagnosis than nonmigrants [8-10]. The disproportionate risk of contracting HIV infection that migrants face in their host country is likely the result of a combination of factors, including stigma, increased risky behavior, and limited access to HIV prevention services [8]. Among the reasons for late diagnosis, the language barrier for non-French–speaking migrants might play an important role.

Research has been extensively conducted to investigate the language barrier among migrants in accessing care and prevention services and their consequences on health. Specifically, qualitative and quantitative studies carried out among different population of migrants in Australia [11], Canada [12,13], and England [14] found that lack of proficiency in English is a barrier to accessing testing for either HIV or hepatitis B or C. Those results have been confirmed in studies of health professionals caring for migrant patients in Belgium [15], England [14], and Australia [16] who state language barrier as a reason for not offering a screening test for those infectious diseases.

Beyond language proficiency issues, migrants may also face health literacy issues. The World Health Organization defines health literacy as “the cognitive and social skills which determine the motivation and ability of individuals to gain access to, understand and use information in ways which promote and maintain good health” [17]. Health literacy can aid a person in making informed choices, reducing health risks, and improving their quality of life [18]. Thus, health literacy is important to understand to what extent the communication barrier is due to the patient's low level of health literacy in their mother tongue and to what extent it is due to the language difference.

The STRADA study (Screening strategies for infectious diseases [tuberculosis, HIV, hepatitis C, hepatitis B] in the migrant population in France) is a current, ongoing, prospective, multicenter, observational study to assess the effectiveness of a strategy for screening HIV, hepatitis B virus (HBV), and hepatitis C virus (HCV) among migrants undergoing a medical examination at the French Office for Immigration and Integration (OFII) [19]. Eligible migrants undergoing a medical examination at OFII are offered rapid screening for the 3 viruses. During the informed consent process, participants are informed that the study is voluntary and independent from the residency permit. This screening is preceded by a short risk factor questionnaire, available in 11 languages (French, English, Arabic, Mandarin Chinese, Bengali, Russian, Lingala, Portuguese, Spanish, Turkish, Haitian Creole). This screening is not mandatory, and individuals are given the option to decline or accept such testing. Migrants who are invited to participate receive information in their own language; this information indicates that they are free to refuse this test, that their refusal to participate will not change anything regarding their residency permit or their level of care (during the medical check-up or in the future), that the participation and results of the tests are kept confidential and separate from administrative papers, and that in case of a positive result, they will be able to access free treatment in France. It is made clear to the participant that a positive result is not grounds for refusing a residency permit. Throughout STRADA, we set up a study of acceptability aimed at measuring the obstacles impeding screening so as to identify strategies for achieving a better acceptance rate. A 3-minute online form is filled out by the health practitioners to report practitioners’ reasons for not offering and patients’ motives for refusing testing [20-22]. The results from over a year show the proposition and acceptability rates are 87.1% and 49.9%, respectively, of the patients who attended the medical check-up. Impeded communication is reported in 29.6% of the reasons health professionals are not offering screening tests, of which 93.6% are related to language barriers (see Table 1). In addition, other cited obstacles to communication include illiteracy, the presence of informal translators (which might hinder screening where sensitive and confidential information is shared), or a general lack of understanding. Among migrants who are offered screening, a majority who refuse (38.0%) said they have already been screened, although health professionals do not mention if they asked when the last screening was performed or checked thoroughly if the screening was indeed performed for the 3 diseases (HIV, HBV, HCV). Communication barriers and low health literacy might make it difficult for health professionals to investigate whether migrants need a repeat screening. Impeded communication is reported in 7.5% of the reasons migrants refuse screening when it is offered. Other reasons for refusing screening include patients do not want screening or do not see any relevance (19.2%), and 5.2% feel either not at risk or not concerned (Table 2). Those reasons might be linked to a lack of knowledge regarding HIV and hepatitis risks and necessity for regular testing, which might be overcome with clear explanations by a health professional.

Overall, communication is a major issue in the implementation of screening, and, as such, a linguistic app targeted at improving how screening is proposed to patients could potentially meet an identified need, by reducing the language barrier on one hand and by providing motivational content to health care workers to help them explain the benefits of testing on the other hand.

Health care professionals use multiple solutions to communicate with migrants with a language barrier: informal interpreters, formal interpreters in person or over the phone, and general translation applications. Informal interpreters are a no-cost solution that enables building a relationship of trust but represent a burden for patients who have to adjust to their availability. Informal interpreters also commit more errors with serious clinical consequences [23,24] and do not allow frank communication [25,26]. Professional interpreters provide high-quality translations that result in a better quality of care [23,27]. Medical doctors and patients are generally satisfied with professional interpreters but are limited by their availability and number of languages [26]. Professional interpreters over the phone offer the same quality of translation with greater time flexibility and more languages. However, health care professionals report less satisfaction [28,29], longer consultation times [28], and more technical problems [29] than with in-person interpreters. Although less expensive than in-person interpreters, this service still has a cost that cannot be met by various health organizations [30].

General translation applications (eg, Google Translate, Lexilogos, iTranslate) have the advantage of being free of charge and available 24/7. Some studies have tested their efficacy in the medical context with mixed results. Studies have found the quality of translation ranging from similar to professional translation [31] to incomprehensible [32] or potentially dangerous [33], depending on the specific application and language tested. The translation of languages that are widely spoken tended to be of better quality than less widespread languages. Online translation apps are not appropriate for long or complex sentences [31], for critical situations, [34], in case of emergency, or if consent is needed [35]. They are also time consuming [36], not appropriate to use with patients who cannot read, and do not allow patients to respond [30].

Besides the general translation applications mentioned earlier, some applications or electronic tools have been specifically developed to be used in medical consultations to facilitate the dialogue between health care professionals and migrants with low language proficiency. Those medical translation applications have developed different features to overcome such drawbacks. They usually consist of a list of sentences commonly used in a medical setting, translated into several languages, either by bilingual researchers or by professional translators. Those sentences are often supported by an audio recording and/or culturally adapted pictures to illustrate them. Patients can rarely give feedback on the health professional’s sentences. Given the quantity and diversity of sentences that can be used in a medical setting, those applications are usually targeted to a specific medical setting, such as patient’s assessment [37]; emergency medicine [38] or anesthesia [39]; specific pathologies, such as asthma [40]; or a specific population, such as refugees [41-43]. No existing application has been developed to facilitate communication between migrants with a language barrier and health professionals regarding testing for HIV, HBV, and HCV.

The aim of the Apidé study (Electronic Application to promote screening among Migrants) is to develop and evaluate an app to assist with screening for HIV, HBV, and HCV among migrants who have low French proficiency. This will include a databank of phrases to offer, explain, and conduct a test for HIV, HBV, and HCV screening in several languages, with a voice version and pictograms, sociocultural adaptation, and adaptation to the patient's level of literacy. It will include a short evaluation questionnaire and items adapted according to the level of understanding. It will help health professionals test migrants who do not speak French and/or have a low level of health literacy. The tool will also include motivational content to improve acceptability, depending on the situation. For example, a frequent reason for migrants' refusal to screen was a lack of risk perception. The app will help overcome this objective using motivational content and explanations of the risks associated with a failure to be screened.

The study will take place in 3 parts that are detailed in the following paragraphs: development of the conceptual model, development of the app, and evaluation of the app (Figure 1). The approach we use is similar to the agile methodology for developing software, which is centered on users’ satisfaction and collaboration and consists of 10 stages: communication (with potential users), requirements gathering (from the demands of users), feasibility study, system analysis (of limitations and impact), software design and coding, testing , integration (of the different modules of the software), implementation (on users’ computers), operation and maintenance, and disposition [44]. In our study, the users are primarily health care workers (doctors, nurses, midwives) and other professionals involved in migrants’ testing, as they will use the app for themselves and guide migrants to use it.

In the first part of the project, we will develop the tool’s conceptual model based on a qualitative study of migrants and health care providers, a systematic review of the literature on electronic apps for migrants with a language barrier, and a focus group and modified Delphi survey with health care providers who propose screening tests to migrants. After conducting those different steps, we will have the app conceptual model, as well as language selection and a proposed layout for the app.

In the second part of the project, we will develop the app. The conceptual model will be translated by a service provider with experience in translation with cultural adaptation. All sentences will be voice-recorded in the chosen languages. The app will be developed, followed by a pilot study of the acceptability and usability of the app.

In the third part of the project, we will evaluate the acceptability and efficacy of the tool. The acceptability will be evaluated using the System Usability Scale (SUS) [45]. The efficacy of the tool will be evaluated in an epidemiological study set up in various settings where migrants go for medical consultations: OFII medical check-up, free hospital consultations (Permanences d’accès aux soins de santé [PASS]), and charity outreach that offers rapid tests. The chosen methodology is a stepped wedge randomized trial [46,47].

The study has received preliminary financing from the Agence Nationale de la Recherche contre le SIDA et les hépatites virales (French Agency for Research against AIDS and viral Hepatitis [ANRS]), followed by 1 year of funding from the Office Français de l’Immigration et Intégration (French agency for migration and integration [OFII]) and then by 3-year funding, including doctoral funding from ANRS. Further funding has been requested from private sponsors and Région Ile-de-France.

At the time of publication of this protocol, the initial qualitative study and systematic literature review have been completed. The study has been completed, and the results have been analyzed [48,49,73] for publication in a peer-reviewed journal. For the qualitative study, we interviewed 33 migrants undergoing a medical check-up at OFII. Migrants reported that difficult communication with a French doctor resulted in lack of confidence and lower compliance with treatment. Migrants with a French-speaking partner were either sidelined during the medical visits, being completely dependent on translation, or their partners helped them in learning new words needed for the medical visits. For migrants who preferred translation, the preference between physically present interpreters versus interpretation by phone or the use of applications was influenced by mainly 4 factors: perceived quality of translation (interviewees were divided, with some perceiving a human interpreter with knowledge of medical jargon most accurate and others trusting the application more), trust, intimacy, and empathy.

For the systematic review, we collected general information about the app, information about health literacy and cultural adaptation, information about the development of the app, evidence about the app’s acceptability and efficacy, and information about the use of apps. We included 61 articles presenting a total of 48 applications. About one-third of the applications (16/48) were designed solely to facilitate the interaction between migrants and a health care provider during a consultation, while the remaining two-thirds (32/48) were designed to promote health among migrants with a language barrier. Overall, the applications had good levels of acceptability, while only half had their efficacy evaluated. In those evaluations, the endpoints used are mostly related to reported behavior change and knowledge improvement, which is common for evaluations of health promotion programs.

Focus groups and Delphi consensus panel for selection of the tool’s content are currently underway.