This study is grounded in four intersecting strands of scholarship that help explain how medical undergraduates manage terminology accuracy and meaning precision while composing in English in EMI settings. First, EMI research in higher education has consistently documented that studying content through English shapes students’ learning experiences and academic performance, particularly when disciplinary complexity and assessment demands are high (1, 2, 13). In Saudi medical education specifically, recent work has highlighted recurring challenges associated with EMI, including language-mediated difficulty with course demands and communication of disciplinary meaning (3, 4, 14, 15).

Second, second language (L2) writing scholarship conceptualizes writing as a recursive process in which writers plan, draft, and revise while continuously monitoring meaning and form (10, 11). Within this process view, accuracy work can be understood as ongoing decision-making that involves checking whether lexical choices and phrasing preserve intended meaning. This issue becomes especially salient when writers work with technical vocabulary under cognitive and time pressure (7, 11). CLT further suggests that when tasks involve high intrinsic load, such as complex medical content and dense terminology, writers may rely on structured routines and external supports to manage processing demands and reduce the risk of meaning distortion (6–9). This provides a principled basis for expecting “verification routines” to emerge in EMI medical writing as learners manage accuracy risk under high load.

Third, translanguaging theory offers a repertoire-based account of how multilingual writers and learners strategically mobilize linguistic resources to mediate meaning-making, reduce uncertainty, and sustain task progress (16, 17). In EMI contexts, pedagogical translanguaging has been discussed as a practical alternative to English-only orientations, with potential benefits for comprehension, participation, and learning—particularly when disciplinary content is complex (18, 51). In Saudi higher education, studies also suggest that students and teachers may view translanguaging as useful, even when classroom practices remain constrained by English-only expectations (19, 20). Although translanguaging research is well established in school literacy and bilingual education, evidence in higher education, particularly in disciplines where technical precision is central, remains less developed (18). Studies in science, technology, engineering, and mathematics (STEM)-related learning environments nonetheless indicate that multilingual resources can support disciplinary meaning-making by enabling learners to negotiate technical concepts and sustain participation under high language demands (21, 22). In medical writing specifically, this implies that L1-mediated meaning work may be especially consequential when terminology density and accuracy stakes are high.

Fourth, research on first language (L1) use and background-language resources in L2 writing indicates that multilingual writers often draw on their full linguistic repertoires for planning, meaning clarification, and lexical decision-making, especially when tasks require precision (23–26). Taken together, these perspectives frame terminology verification and meaning checking as central writing-process work in EMI medical education and guide the analytic attention to (a) how students identify and manage perceived accuracy risk during composing and (b) how instructional conditions (task requirements, feedback practices, and participation norms) shape the visibility of this meaning-verification work (2, 11, 16, 24, 27). Because much of this verification work is L1-mediated, translanguaging becomes especially relevant in early-stage medical writing, where students must stabilize concepts and terminology before producing precise English disciplinary meanings. This is particularly relevant for earlier-stage undergraduates, whose disciplinary discourse control is still developing and may depend more heavily on L1 resources for conceptual stabilization.

Building on this framework, prior research suggests that meaning-verification during L2 composing is often supported by background-language resources (e.g., planning in L1, conceptual clarification, lexical searching, and internal meaning checks), even when English-only expectations are present (23–27). In the context of medical writing, such verification work is plausibly intensified because students must align technical terminology with precise disciplinary meaning under cognitive load and assessment pressure (7, 11). This reinforces the view that L1 and translanguaging practices may operate as accuracy-protection mechanisms rather than as deviations from EMI goals, particularly when the instructional expectation is precise, discipline-appropriate English output. Accordingly, translanguaging can be conceptualized as supporting accuracy and meaning fidelity, not merely comprehension.

Recent strategy-focused research suggests that students’ ability to manage precision in medical and health-related writing can be strengthened through process-oriented instructional designs that make revision cycles visible, including structured peer feedback and iterative redrafting, rather than treating accuracy as a purely individual vocabulary task (28–30). Complementing these practices, feedback literacy research emphasizes that students need the capacity to interpret, evaluate, and act on feedback (including managing affective responses) in order to translate comments into concrete revisions (31–33). In parallel, recent work in medical English/EMP education highlights technology-supported approaches (e.g., structured web-based EMP learning and technology-enhanced, interactive practice) as ways to scaffold language development and provide additional opportunities for meaning-verification and terminology learning in time-constrained medical curricula (34, 35). Taken together, this literature positions strategies for overcoming medical writing challenges as embedded in iterative writing, feedback uptake, and supported opportunities to verify terminology and stabilize disciplinary meaning.

Recent evidence from medical and healthcare education shows that using a foreign language as the medium of instruction can introduce recurring language-mediated challenges that affect comprehension of disciplinary content, participation, and assessment performance, particularly when students transition from prior schooling in another medium of instruction (1, 2, 5). Recent syntheses focusing on healthcare education have similarly emphasized that EMI can increase linguistic and cognitive demands and may intensify stress-related experiences for non-native speakers (3, 36). In Arab and Gulf contexts, recent work likewise reports tensions between EMI policy expectations and students’ linguistic preparedness, with consequences for classroom participation and feedback-mediated learning (15, 37). Building on this broader picture, I focus here on Saudi Arabia, where EMI is institutionalized in medical education and where prior research has documented consequential EMI-related challenges for students and stakeholders (4, 14). Within healthcare education specifically, recent scholarship has called for closer attention to how EMI intersects with learning processes and educational outcomes in health professions contexts, underscoring the need for research that is discipline-attuned rather than language-generic (3). This makes it important to examine not only whether EMI is challenging, but how students manage disciplinary accuracy demands in concrete writing activity.

In Saudi health professions education specifically, available evidence suggests that EMI-related challenges are consequential for students, including difficulties connected to language demands in learning and assessment (14). Institution-level analyses also highlight practical constraints in implementing EMI in health sciences settings, including tensions between policy expectations and students’ preparedness (15). Saudi higher education studies also indicate that bilingual resources can be perceived as beneficial, even when classroom legitimacy is uneven and shaped by local norms and instructor practices (19, 20). Because many meaning-verification practices occur internally (e.g., mental translation) or privately (e.g., Arabic notes, peer mediation), they may remain largely invisible in final written products and therefore under-recognized in instructional design (24, 27).

Despite the relevance of these insights to EMI medical education, there remains limited qualitative evidence describing how undergraduate medical students manage accuracy and precision demands during English academic writing, what strategies they use to secure disciplinary meaning (particularly terminology work), and how course conditions (task requirements, instructor expectations, feedback practices, and participation norms) shape these strategies. This gap is especially salient in Saudi medical education, where EMI is institutionalized and where the Arabic-medium schooling background of most students makes cross-linguistic mediation a plausible component of learning and writing in the early undergraduate years (4, 14).

Accordingly, this study investigates Saudi undergraduate medical students’ experiences with terminology accuracy in English academic writing within an EMI context. It focuses on their reported challenges, strategies, and the role of instructional conditions-including course expectations and feedback-in shaping terminology-related writing practices. By foregrounding L1-mediated meaning verification in terminology work, the study also examines translanguaging’s contributions to disciplinary precision in EMI medical writing. It is guided by the following research questions:

This study adopted a qualitative, interpretive design to examine how Saudi medical students manage the accuracy demands of English academic writing in an EMI context, focusing on medical terminology and meaning precision. This design was appropriate because terminology- and meaning-focused writing strategies, including verification and accuracy monitoring, are typically internal and context-dependent processes best captured qualitatively (38). These processes are therefore better captured through students’ accounts and contextual classroom evidence than through decontextualized measures (38).

The study was conducted in an English academic writing course within an undergraduate medical program in Saudi Arabia, where assessed writing tasks required students to communicate disciplinary content in English with clarity, organization, and precision. Participants were recruited from two sections of the same course taught by instructors with different linguistic backgrounds. These two sections were selected because they were the available sections during the data-collection period and shared the same syllabus, learning outcomes, and assessed writing tasks. This allowed participants’ experiences to be examined within a common curricular structure. At the same time, these two sections provided naturally occurring contextual variation in instructional routines and feedback-treated as variation within the same setting rather than comparison groups. Instructor-related differences were used interpretively to contextualize students’ accounts of expectations, feedback, and participation norms. However, the analysis did not aim to attribute strategy differences to instructor background or to make between-section comparisons. The classroom environment was conceptualized as an ecological setting in which instructional routines and interactional norms can shape how writing-related strategies are enacted and discussed (39). Section membership was therefore used as contextual information during interpretation (e.g., when discussing expectations, feedback practices, and participation norms), rather than as a basis for between-group comparison.

I approached the study as a researcher with professional familiarity with EMI medical education in Saudi Arabia. This proximity supported context-sensitive interpretation of participants’ accounts and classroom norms, but it also required active reflexive attention to assumptions about students’ language practices and what counts as “accurate” medical writing. Throughout data collection and analysis, I used memo writing to document evolving interpretations, note moments of surprise or tension, and make explicit how my positioning may have shaped analytic decisions. These memos were revisited during theme refinement to check how interpretations were being constructed and to ensure that themes remained grounded in participants’ accounts rather than in prior expectations (40).

Participants were 15 Saudi second-year medical students enrolled in the writing course across the two sections. The study used purposive, criterion-based sampling to recruit students who were actively engaged in course writing tasks in English and were able to describe their writing processes and strategy use in detail. Purposive sampling is widely used in qualitative research to select information-rich cases relevant to the phenomenon under investigation (41). The sample size (n = 15) was appropriate for an interview-based qualitative design aimed at generating in-depth accounts of writing processes and strategy use. Dataset adequacy was evaluated during analysis, and later interviews primarily elaborated existing patterns rather than introducing substantively new ones (40). Recruitment aimed to represent students from both sections to capture variation in classroom routines and feedback practices that may shape writing strategies. To protect confidentiality, participants were assigned anonymized IDs. The letter indicates the course section (A or B), and the number indicates the individual participant within that section (e.g., A1–A7 for Section A; B1–B8 for Section B). Participants’ ages ranged from 19 to 21 years. All participants reported Arabic as their L1 and English as their L2, consistent with EFL educational contexts in Saudi higher education. Participants’ English proficiency was described as ranging from high-intermediate to advanced. This was based on their completion of a full year of intensive English coursework required by the program before entry into the medical program.

Data were analyzed using reflexive thematic analysis following Braun and Clarke (40, 44). The analysis was primarily inductive, allowing codes and themes to be developed from the dataset and then organized in relation to the research questions. Interview transcripts were treated as the primary analytic dataset because the study aimed to capture students’ reported writing processes, rationales, and strategy decision-making; dimensions that are often only partially visible through classroom observation alone. Observation fieldnotes were used to contextualize these accounts and to document instructional routines and writing-related episodes that supported, nuanced, or complicated students’ claims (42, 43). Fieldnotes were also consulted during theme refinement to sharpen interpretation of course expectations, participation norms, and feedback practices referenced in the interviews (40). Arabic occurred in only a few interview instances, typically for brief clarification. These segments were translated into English prior to coding and reporting using a meaning-based translation approach. Translations were checked against the original Arabic to confirm accuracy and preserve intended meaning and terminology. The translation process also drew on the researcher’s educational background and experience in Arabic–English translation.

Analyses were conducted on the combined dataset; section membership was used only as contextual information when interpreting how course expectations and feedback shaped strategy use. Analysis proceeded iteratively. As analysis progressed, the dataset adequacy was monitored by tracking whether new interviews contributed substantively new codes or whether they elaborated and strengthened existing patterns. In the later interviews, contributions primarily deepened and exemplified existing candidate themes rather than introducing new thematic patterns, supporting the decision that the dataset was adequate for the study aims (40). First, the dataset was read repeatedly for familiarization, with initial notes highlighting recurring challenges and strategy patterns. Second, transcripts were systematically coded with attention to function and timing in the writing process, focusing on terminology-related decision points, meaning verification, and accuracy monitoring during planning, drafting, and revision. Coding was accompanied by analytic memo writing to document emerging interpretations and track decisions about what constituted evidence for a candidate theme (40). Third, codes were examined for patterned co-occurrence and organized into candidate themes aligned with the three research questions. Themes were reviewed and refined by checking coherence across coded extracts and ensuring that each theme captured a distinct organizing concept with clear boundaries (40, 44). Candidate themes were refined by (a) merging overlapping patterns, (b) splitting themes that contained multiple organizing concepts, and (c) removing codes that did not contribute to a coherent pattern relevant to the research questions (40). Themes were bounded using two criteria: (1) each theme had to represent a patterned way participants described managing terminology/meaning precision (rather than isolated instances), and (2) the theme needed to be supported by multiple data segments across the dataset, with clear distinction from other themes (40). Finally, themes were defined and named to foreground how students manage precision and terminology in English writing and how course conditions shape strategy use. Findings were reported primarily through representative interview excerpts, with selective observation-note vignettes included to provide contextual grounding. ATLAS.ti was used to support data management, systematic retrieval of coded segments, and documentation of analytic decisions (45). Use of software facilitated organization and transparency; interpretive judgments remained grounded in the reflexive thematic analysis approach.

Across both sections, students consistently framed medical terminology as the most consequential source of “accuracy risk” in English medical writing: terminology was described as conceptually dense, highly specific, and tightly linked to disciplinary meaning, such that small lexical shifts could distort intended claims. This concern was not limited to “knowing vocabulary,” but reflected the perceived stakes of misrepresenting medical meaning when summarizing sources, describing mechanisms, or qualifying relationships. This aligns with disciplinary writing scholarship emphasizing that medical writing requires precise control of technical terminology and accurate representation of scientific meaning (11). In EMI settings, this precision burden is amplified because students must coordinate disciplinary reasoning, genre expectations, and language form while operating in an additional language (1, 2, 13). The present findings also align with work showing that EMI in Saudi health professions education can be associated with learning pressures related to specialized vocabulary demands, assessment performance, and confidence in academic participation (4, 14). These stakes are discipline-specific: in medical writing, terminology precision safeguards scientific meaning and can affect clinical reasoning development and assessment validity, whereas in general academic writing imprecision more often weakens clarity or argumentation rather than altering disciplinary meaning. Taken together, students’ accounts suggest that terminology selection operates as a high-stakes accuracy checkpoint in early medical disciplinary writing development, requiring deliberate monitoring to preserve meaning fidelity rather than serving as a peripheral language concern.

Students described a set of composing routines that were explicitly oriented to preserving meaning accuracy. These routines were distributed across composing stages, which is consistent with process models that conceptualize writing as recursive problem-solving involving planning, text generation, and review (10). Students’ emphasis on stabilizing ideas before English production and re-checking terminology when specificity increased can be interpreted through CLT: when tasks combine high intrinsic load (disciplinary complexity, dense terminology) with additional language demands, writers may adopt structured routines to reduce uncertainty and stabilize processing during production and revision (6–9).

From an L2 writing perspective, the reported use of Arabic for idea generation, lexical searching, conceptual clarification, and monitoring is consistent with evidence that multilingual writers draw on their stronger language when tasks become cognitively demanding or conceptually complex (23, 25, 26). Translanguaging scholarship provides an additional explanation by framing these practices as strategic mobilization of an integrated repertoire used to secure meaning and sustain progress rather than as interference between bounded codes (16, 17, 46). In this study, translanguaging functioned as a meaning-verification mechanism embedded in disciplinary composing, especially when terminology density and precision expectations were high, supporting the stabilization of concepts before and during English production. This interpretation is consistent with prior writing research showing that cross-linguistic meaning work can persist even under English-only expectations because it supports conceptual stability and rhetorical decision making in demanding tasks (24, 27, 47).

The findings also clarify how instructional conditions shape whether accuracy-oriented strategies are enacted publicly or privately. Students described English as the expected language for instructor-facing participation, while Arabic was more available in peer-facing interaction. This pattern is consistent with EMI research indicating that English-only norms often coexist with multilingual practices that move into less observable spaces, especially when classroom routines and participation structures reward public English performance (2, 51). Observation fieldnotes supported this interpretation by documenting routine participation structures, including whole-class English interaction paired with brief peer exchanges during task work, which contextualized why students reported internal translation and discreet peer mediation even when Arabic was not visible in whole-class interaction (39). Rather than indicating “absence” of L1 use, these patterns suggest a reallocation of translanguaging into backstage spaces where students can verify meaning with lower participation risk. Students’ accounts also suggest that visibility is tied to confidence and performance pressure. When teacher-facing participation is experienced as evaluative, students may increase self-monitoring and reduce willingness to externalize uncertainty. This is compatible with research linking language anxiety to reduced processing efficiency in a second language, particularly in high-stakes or stressful situations (48). Accordingly, instructional design that legitimizes verification as part of writing (even if L1 work remains private) may be especially consequential for earlier-stage students who are still developing disciplinary discourse control and terminology precision.

Taken together, the findings support a practical shift in how EMI medical writing is designed and supported. Students’ reports indicate that meaning verification and terminology control are not incidental strategies; they are the mechanism by which learners manage disciplinary accuracy in an additional language. A feasible instructional response is to build a structured accuracy-support model that makes verification an explicit part of the writing process while preserving the expectation that final writing is accurate, discipline-appropriate English.

Such a model can combine process-oriented writing principles with cognitive load considerations by sequencing work into three recurring checkpoints: concept clarification, terminology verification in context, and precision-focused revision (10). It can also leverage established vocabulary learning principles by requiring students to confirm terminology usage in context and maintain term consistency across sections of a text (49). Importantly, the model does not require public first-language use. Instead, it acknowledges that Arabic-mediated work may occur internally or peer-to-peer and focuses instructional attention on the quality of verification outcomes, meaning fidelity, and accountable decision making in the final text (16, 18). This approach also aligns with the study’s findings that students already engage in verification; instruction can make these practices more systematic, transparent, and learnable.

Collectively, this study provided qualitative insights into how Saudi medical students manage the accuracy demands of English academic writing in an EMI context, with particular attention to medical terminology work and the instructional conditions that shape strategy use. Students consistently positioned terminology accuracy and meaning precision as the central challenge in medical writing, and they described recursive verification practices across planning, drafting, and revision that helped them protect disciplinary meaning under high lexical and cognitive demands (7, 8, 10, 11). From a CLT perspective, these verification practices can be understood as routines that reduce avoidable processing demands and lower the risk of meaning distortion when students write complex medical content through an additional language (7, 8). Their accounts also indicate that cross-linguistic meaning work, including Arabic-mediated rehearsal and verification, functions as a routine mechanism for stabilizing concepts and monitoring accuracy during composing, consistent with L2 writing and translanguaging scholarship (16, 17, 23, 25, 26). At the same time, the visibility of these practices was shaped by participation norms and teacher-facing expectations, suggesting that English-only routines may relocate verification into private spaces rather than eliminate it (2, 51). In medical education terms, supporting terminology verification is not only a language intervention but also a way to protect clinical reasoning development and strengthen the validity and interpretability of writing-based assessments as evidence of disciplinary understanding.

This study relied primarily on interview accounts of composing practices, with observation fieldnotes used to contextualize participation norms and instructional routines. Accordingly, the findings foreground students’ reported strategy use and their perceptions of when strategies were visible in classroom participation, rather than offering a fine-grained interactional record of writing-in-action. Because observations were documented through fieldnotes rather than recordings, the observation data reflect selective, non-verbatim documentation and are subject to interpretive emphasis. As a result, some classroom events may have been under-documented, and certain features of interaction (e.g., exact wording, turn-by-turn negotiation, timing, and micro-repair sequences during verification) could not be examined in detail. Future studies could address this limitation by incorporating recorded classroom interaction where feasible and ethically approved, alongside complementary artifacts (e.g., task materials, feedback comments, or draft sequences) to better trace how terminology-related decisions unfold across composing stages.

The study also reflects a specific course context; transferability depends on similarity in assessment genres, feedback practices, and EMI participation norms. Since participants were second-year Saudi medical students in a particular academic writing course, the findings may not transfer fully to students at other year levels, other health disciplines, or EMI settings with different curricular demands, clinical exposure, or language policy expectations. Future research could broaden sampling across programs, institutions, and stages of training to examine whether the same challenge–strategy patterns emerge and to identify conditions under which terminology-related concerns become more or less salient.

Because the study draws on two sections taught by instructors with different linguistic backgrounds, some of the reported expectations, feedback routines, and participation norms may be instructor-specific rather than fully generalizable as course-wide norms. Therefore, the findings should be interpreted as situated within these instructional environments. Future work could examine instructor-related variation more explicitly through comparative designs across additional sections and instructors, as well as by testing targeted pedagogical supports (e.g., structured verification routines, feedback designs that encourage clarification-seeking) to evaluate whether they improve terminology precision without increasing cognitive burden. Mixed-methods designs may also be useful, combining surveys to map the prevalence of specific challenges and strategies with interviews and classroom-based methods to explain patterns in depth.