Workplace violence constitutes a global public health problem particularly prevalent in settings characterized by high levels of interpersonal interaction, with the health care sector being the most affected [1,2]. In this context, the situation is especially concerning due to the sustained increase in violent incidents over the past decade, with an even more pronounced rise during the recent COVID-19 pandemic [3,4].

The consequences of workplace violence extend beyond the individual, affecting health care workers’ well-being, patient safety, and team dynamics, ultimately reducing the quality of services provided by health care systems [5,6]. Among health care professionals, exposure to workplace violence has been associated with higher levels of burnout, stress, and psychological trauma, as well as decreased job satisfaction [7-9], which, in turn, translates into a greater intention to leave the job [9]. Moreover, workplace violence generates significant costs for health care systems by increasing absenteeism, staff turnover, and the need for replacement personnel, thereby negatively impacting productivity and the economic resources of health care institutions [10,11].

Violence in health care settings may be physical, verbal, or psychological, with the latter being the most frequent and least visible forms [12,13]. Despite their high prevalence, a large number of workplace violence incidents are not formally reported or recorded in official reporting systems, limiting accurate knowledge of their true magnitude and hindering the implementation of preventive and supportive measures [6,14,15].

The problem is particularly acute in emergency departments, which are consistently identified as high-risk settings for workplace violence. Previous studies have reported high rates of verbal and physical aggression among emergency health care professionals, with verbal aggression being the most prevalent form and frequently underreported [16,17]. Global evidence also indicates a higher prevalence of workplace violence in emergency and psychiatric settings than in other hospital services [13,18,19].

In Spain, workplace violence against health care professionals represents a well-documented and growing public health concern. Studies conducted in hospital settings report a high prevalence of aggression, particularly affecting nursing staff and frontline professionals, with nonphysical forms such as verbal and psychological abuse being the most frequent [20,21]. This problem is especially evident in high-demand areas such as emergency departments, where exposure to aggression is common, often involves patients and relatives, and remains substantially underreported [9,13,20].

Within the Andalusian Health Service (Servicio Andaluz de Salud), specific protocols for the prevention of and response to aggression against health care professionals have been developed [22]. These institutional measures are currently implemented in the study setting, namely, the Almería Health District, where health care professionals working in emergency and critical care units have access to formal reporting systems, legal support, and preventive protocols. However, aggressive incidents continue to occur, and their psychological impact is not always addressed in an early and systematic manner. This situation highlights the need for complementary tools that facilitate immediate and confidential access to guidance, psychological support, and self-care strategies, contributing to a comprehensive approach to workplace violence in health care settings and reinforcing the effectiveness of existing protocols [13,23].

In this context, the development of innovative digital solutions aimed at supporting health care professionals who have experienced workplace aggression emerges as a promising strategy. Conversational agents based on structured and rule-based architectures have demonstrated utility in health care by providing information, guidance, and emotional support in an accessible and continuous manner [24-26]. There is recent evidence suggesting that this type of technological assistance can be effectively integrated into workplace violence prevention programs, enhancing their reach and effectiveness [26].

However, currently available tools are not usually specifically designed to address workplace violence in health care settings from a comprehensive, professional-centered perspective. Within this framework, Sanidad Segura has been developed as a pioneering technological tool aimed at bridging this gap by integrating assistance, emotional support, and referral to legal and psychological resources, offering accessible, safe, and complementary support to existing institutional measures against violence toward health care professionals.

The development of Sanidad Segura is guided by the updated Medical Research Council framework for complex interventions. A preliminary program theory underpins the intervention, whereby chatbot components (emotional support, legal guidance, incident reporting, and psychoeducational content) are expected to improve perceived support, reduce barriers to reporting, and enhance help-seeking behaviors. Sanidad Segura is designed as a postincident support tool intended to be used after exposure to workplace aggression rather than during acute violent situations. The chatbot provides structured guidance on emotional coping strategies, incident documentation, legal rights, and access to institutional and psychological support services.

The platform operates through a structured conversational workflow based on predefined modules. After accessing the chatbot, users are guided through a sequence of options tailored to their needs following a workplace aggression incident. These modules include (1) emotional support and validation, (2) guidance on how to document and report the incident, (3) information on legal rights and institutional procedures, and (4) access to psychological and occupational support resources. Users interact with the chatbot by selecting options or entering short text inputs, which are categorized using predefined intents and routed through decision tree pathways. The chatbot does not generate free-text responses autonomously; instead, it delivers standardized, expert-validated content adapted to user input.

This study aims to design, develop, and validate this chatbot (Sanidad Segura) to assist health care professionals who experience workplace aggression and evaluate its usability, readability, and exploratory indicators of perceived usefulness and support in a pilot study.

The study has been funded for a 2-year period starting on December 18, 2024. This study will generate both quantitative and qualitative outcomes to evaluate the chatbot. Quantitative outcomes will include usability scores measured using the SUS and readability scores assessed using the Inflesz scale, along with participants’ sociodemographic characteristics.

Data collection was conducted between February and May 2026. A final convenience sample of 105 participants was obtained. At the time of manuscript submission, data analysis had been completed using SPSS Statistics (version 30.0; IBM Corp). The analyses included descriptive statistics, group comparisons, and correlation analyses as part of the pilot study. The results are expected to be published in an open-access journal in September 2026, with a potential second publication derived from the qualitative analysis anticipated in 2027. Qualitative findings will be derived from semistructured interviews, allowing for an in-depth exploration of user experience. Thematic analysis will be used to identify key themes related to usability, user experience, and suggestions for improvement.

Finally, quantitative and qualitative findings will be integrated through a convergent mixed methods approach using triangulation techniques to provide a comprehensive understanding of the usability, acceptability, and readability of the chatbot.

Workplace violence against health care professionals remains a critical and escalating global challenge. Recent evidence indicates that exposure to verbal threats, physical assaults, and psychological aggression continues to rise, with substantial negative consequences for health care workers’ mental health, job satisfaction, professional retention, and quality of patient care [25]. Studies conducted over the last years consistently report associations between workplace violence and increased levels of anxiety, depression, burnout, and reduced organizational commitment [10].

The growing awareness of workplace violence as a critical threat for health care professionals underscores the need for innovative, accessible, and evidence-based strategies to support those affected [2]. Despite the implementation of institutional prevention programs and legal frameworks, underreporting persists, largely due to fear of retaliation; normalization of violence; and lack of accessible, confidential reporting mechanisms. This study will focus on the design, development, and validation of Sanidad Segura, a chatbot developed to assist health care professionals who experience workplace aggression, and evaluate its usability, readability, and exploratory indicators related to perceived usefulness and support in a pilot study. Recent studies suggest that conversational agents can provide timely guidance, psychoeducational support, and facilitate help-seeking behaviors, particularly in contexts where time constraints and stigma limit access to conventional support pathways [25,37].

Sanidad Segura is designed to function as both an individual support tool and an organizational intelligence resource. At the individual level, the chatbot will enable confidential identification and reporting of aggression-related incidents while providing tailored medical, psychological, and legal guidance. Early access to structured support following exposure to violence has been shown to mitigate long-term psychological distress and reduce the likelihood of chronic mental health issues [11]. The interactive and anonymous nature of chatbot-based interventions may lower barriers to disclosure, particularly among professionals who would otherwise refrain from formal reporting. From an organizational perspective, integrating Sanidad Segura into existing occupational health frameworks may strengthen the culture of safety and reporting. There is evidence from recent health care management studies suggesting that institutions with proactive reporting systems and transparent response mechanisms demonstrate improved staff trust, reduced turnover intention, and enhanced patient safety outcomes [10]. In addition, Sanidad Segura is expected to provide valuable information on workplace violence in health care, thereby supporting organizational learning and prevention strategies. On the basis of notification records (reported events), Sanidad Segura can generate aggregated insights into the incidence and prevalence of workplace aggression. These data enable the identification of vulnerable professional groups and high-risk clinical settings, thereby informing targeted prevention strategies, training initiatives, and more efficient resource allocation. This evidence will address the lack of careful planning, resource management, and ongoing evaluations to ensure effective implementation and sustain impact in reducing workplace violence across health care settings [38], and it will provide empirical support for institutional and legislative decision-making related to workplace violence prevention.

The methodological design of this study emphasizes usability, readability, and acceptability as critical prerequisites for effective digital health implementation. Previous studies evaluating health care chatbots indicate that high usability scores are strongly linked to sustained engagement and perceived usefulness, underscoring the relevance of usability for the successful adoption of digital health interventions [25]. In fact, the use of validated instruments such as the SUS is consistent with current best practices in digital intervention development and aligns with the tools used in other comparable studies to assess usability [39,40].

This is the first study in the country aiming to design, develop, and validate a chatbot supporting health care professionals facing workplace aggression. This protocol represents a significant step toward the development of an integrated digital framework for the detection, management, and prevention of workplace violence in health care settings. By combining individual support with organizational data, Sanidad Segura has the potential to reduce the burden of workplace aggression, improve professional well-being, and inform evidence-based prevention strategies. If effective, Sanidad Segura could be integrated into the health care system and scaled to other regions nationwide. Future studies will be required to evaluate effectiveness, long-term impact, and scalability across diverse health care contexts.

This study is positioned as a feasibility and usability study rather than a full effectiveness evaluation. Qualitative findings complement the limitations of traditional usability instruments such as the SUS. Safety considerations are addressed through controlled design and escalation pathways.

It is important to note that Sanidad Segura is not based on generative artificial intelligence models but rather on a controlled rule-based system, which enhances safety, predictability, and alignment with validated institutional protocols.