“It’s Not Just Developers”: Why Modelling Assistance Must Fit All Roles in Low-Code/No-Code Software Development for Digital Health
Wissensdatenbank Strukturen & Prozesse Systemauswahl & Implementierung Mensch Schulung & Digitale Kompetenz C.3: Künstliche Intelligenz-basierte Software-Fabrik für MedTech-AnwendungenLow-Code/No-Code software development isn't just for developers. The roles involved in digital health LCNC application development are diverse—and so are their needs. Understanding these roles and their pain points is essential to ensure that modelling assistance enhances quality and accelerates time to market for LCNC digital health software development.
Problem Description, Research Question, Relevance
Low-Code/No-Code (LCNC) tools—also known as Model-Driven Software Engineering (MDSE) tools—aim to streamline software development by modelling software rather than coding software. To assist their users, LCNC tools have introduced modelling assistants.
However, it is often unclear who exactly is involved in using these LCNC tools, who these assistants aim to support, and what the requirements from their users are. To address this, we pose two research questions:
- (RQ1)Who are the users/roles that require support in developing digital health software using LCNC tools? Identifying these roles provides insight into user profiles, expertise, and other factors that help shape more effective and tailored assistance solutions for faster and higher-quality LCNC software development.
- (RQ2)What are their specific needs? Once roles are known, their requirements become clearer. Without this understanding, it is difficult to ensure that LCNC tools and assistants address the pain points in software development.
Methods and Approach in the Project
To answer RQ1, we conducted a systematic mapping study [1] to identify—among other data—the roles reported in literature and practice that participate in LCNC software development, particularly those supported by modelling assistants.
To answer RQ2, we carried out focus groups [2] with participants across multiple roles to understand the needs and expectations of users interacting with LCNC modelling assistants.
Results and Findings
From over 3,000 research articles, we selected 58 relevant works and evaluated 17 LCNC tools available in the market. Our findings [1] reveal that:
- Most LCNC assistants are designed primarily to support software developers and designers.
- However, less technical roles, such as business analysts and end users (domain experts), are increasingly considered in LCNC modelling assistance proposals. In the context of digital health software development, these domain experts could be roles like patients, doctors, and nurses, whose input is invaluable during the LCNC development of digital health applications.
Additionally, based on our focus group discussions [2], we identified users priorities for modelling assistance in LCNC tools:
- Reducing the complexity of LCNC tools by automatically creating models.
- Using less technical languages to accommodate domain experts with limited software development experience.
- Enhancing transparency and traceability between their input and the effect on the final software.
Recommendations for Practice
- Acknowledge that digital health application development in LCNC contexts goes beyond technical roles.
- Do not overlook domain experts, such as doctors, nurses, and patients, who provide the essential non-technical knowledge that shapes meaningful digital health solutions.
- Identifying pain points, priorities, and role-specific needs in LCNC development is the cornerstone for assisting LCNC users in delivering and improving quality—democratizing in digital health software development for all roles.
Literature and Other Sources
[1] Mosquera, D., Ruiz, M., Pastor, O., & Spielberger, J. (2024). Understanding the landscape of software modelling assistants for MDSE tools: A systematic mapping. Information and Software Technology, 173, 107492. doi.org/10.1016/j.infsof.2024.107492
[2] Mosquera, D., Ruiz, M., Pastor, O., & Spielberger, J. (2022). Assisted-Modelling Requirements for Model-Driven Development Tools. International Conference on Research Challenges in Information Science, 458–474. doi.org/10.1007/978-3-031-05760-1_27
Citation
Mosquera, David & Ruiz, Marcela (2025). “It’s Not Just Developers”: Why Modelling Assistance Must Fit All Roles in Low-Code/No-Code Software Development for Digital Health. In Flagshipprojekt SHIFT. Wissensbeitrag C.3 (Nr. 2).