Koen Smit

Background: Clinical decision support systems (CDSSs) are widely implemented in hospitals, yet their uptake in nursing practice remains inconsistent. While electronic health records continuously capture nursing decision data, these data are rarely used to support structured quality improvement. Decision mining offers a novel approach by reconstructing and visualizing decision logic from clinical decision logs to support retrospective reflection and protocol evaluation rather than real-time alerting. Before such tools can be integrated into quality improvement programs, their acceptance by clinical users must be empirically examined. Objective: This study aimed to conduct an early-stage, user-centered evaluation of a clinical decision mining tool for nursing quality improvement and to examine how nurses evaluate the factors influencing their intention to use the tool, using the unified theory of acceptance and use of technology (UTAUT) as a conceptual organizing framework. Methods: An exploratory qualitative design was used, using a theory-guided expert discussion as the primary method. A purposive sample of 5 quality improvement nurses from a large Dutch hospital participated in a 90-minute session comprising a structured prototype demonstration, a 12-item UTAUT-derived questionnaire administered as a facilitation instrument, and an expert discussion. Questionnaire scores were summarized descriptively by construct to anchor and organize the expert group discussion. Expert discussion data were analyzed using directed content analysis, with UTAUT constructs—performance expectancy, effort expectancy, attitude toward use, social influence, and behavioral intention—as sensitizing categories. Attitude toward use was included following UTAUT extensions validated in nursing informatics contexts. Results: Performance expectancy scores were consistently high, driven by perceived value for protocol evaluation and data-supported quality improvement discussions. Effort expectancy was moderate and variable, reflecting concerns about data interpretation and limited data literacy rather than interface usability. Attitude toward use showed the lowest construct-level scores and the widest variation, contingent on whether the tool was framed as learning-oriented rather than a monitoring or control mechanism. Behavioral intention was relatively high for targeted use, but participants positioned adoption in structured quality improvement meetings rather than daily in bedside practice, reflecting role differentiation rather than rejection. Social influence did not emerge as a substantive theme at this preimplementation stage. Conclusions: Nurses perceived strong value in the CDM tool’s capacity to make protocol adherence and decision variation transparent and discussable within quality improvement settings. Acceptance was conditional on a learning-oriented framing, interpretive scaffolding, and investment in data literacy. The findings suggest that governance framing and interpretive burden function as adoption determinants for retrospective decision support tools in ways not fully captured by standard UTAUT constructs. Retrospective decision mining may complement existing CDSSs not by adding alerts but by strengthening local data-driven learning cycles.

Although proactive services are widely discussed in e-government research as government-initiated, data-driven, and automated forms of service provision, little is known about how civil servants understand the concept in practice. This paper conducts an analysis of 21 semi-structured interviews with civil servants involved in policy, service delivery, and service design. Participants were asked to provide their own definitions of proactive services and examples from their organizations. The data were analyzed through multiple coding rounds. Three core properties of proactive public services emerged. First, governments take initiative in delivering services rather than waiting for citizen requests. Second, service delivery is initiated from a broad perspective on citizens' life situations. Third, service delivery is initiated earlier to prevent the emergence or escalation of problems. By identifying these three properties, this study contributes to the construct clarity of proactive public service delivery and broadens current understandings of what it means for public services to be proactive according to civil servants. While the first property aligns closely with existing proactive services literature, the other two properties of broader and early, preventive delivery extend current taxonomies. The findings also show that civil servants' conceptualizations of proactive services are in line with concepts from literature on proactivity as a human character trait, thereby bridging psychological and e-government perspectives on proactivity and offering a new perspective on current discussions in public organizations on becoming proactive.

System understanding is crucial for maintaining and reengineering software systems. This holds particularly for legacy systems, where information about the system is often not documented well and numerous changes over time have increased their complexity. These systems incur high maintenance costs, make it difficult to implement new functionalities, and pose challenges for reengineering activities such as modernizing parts of the system. Moreover, replacing these systems is often infeasible due to the risk of losing embedded business knowledge. Documenting the business logic embedded within such systems, for instance by means of models using Decision Model and Notation (DMN), could help address this challenge. However, the manual extraction of DMN models is difficult given the inherent complexity of these systems. In this regard, Large Language Models (LLMs) offer the potential to support the automation of this process by analyzing source code, leveraging contextual information, and performing without the need for extensive task-specific training data. In this study, we explore whether and how LLMs (GPT-4.1 and Gemini 2.5 Pro) can be applied for automated extraction of DMN models from source code. We applied the Design Science Research (DSR) methodology, supported by a theoretical literature review that identified requirements for LLM-based DMN extraction. In addition, we conducted experiments to determine prompts, model configurations, and the extent to which the extraction process could be automated. Our results indicate that LLM-based automated extraction of DMN models from source code can serve as an aid for understanding complex software systems.

BACKGROUND Pediatric oncology is characterized by highly protocolized but complex clinical pathways with high levels of variability. Gaining insight into real-world treatment delivery is essential for improving quality of care and research. Process mining offers a data-driven approach to reconstruct and analyze real-world clinical pathways using routinely collected data from healthcare systems. While process mining has been applied in adult oncology, its use in pediatric oncology has not yet been explored, despite its potential to improve quality of care. It is unclear which factors hinder or enable its application. OBJECTIVE This study aimed to identify barriers and facilitators influencing the application of process mining in pediatric oncology. METHODS A broad, systematic scoping review was conducted to identify barriers and facilitators for process mining in adult oncology. Identified factors were thematically mapped using the Human, Organization, and Technology–fit (HOT-fit) framework. Findings were then triangulated with barriers and facilitators identified during a retrospective case study applying process mining to clinical pathways of pediatric patients with neuroblastoma. RESULTS A total of 199 factors were identified across 34 themes in adult oncology. Technology-related factors were most prominent, particularly issues related to data quality, data complexity, and algorithm suitability, while human knowledge and expertise emerged as key facilitators. Many themes translated directly to the pediatric oncology setting, including pervasive data quality issues and high model complexity. Several additional aspects were identified, including limited awareness of organizational research, low patient numbers, and clinician perceptions of process mining. Facilitators unique to pediatric oncology included a highly protocolized care environment and strong international collaborations, which offer opportunities to mitigate limited data availability resulting from low patient numbers. CONCLUSIONS Process mining shows clear potential to provide insight into real-world pediatric oncology pathways, but its broader adoption is constrained by persistent technical, organizational, and human challenges. Advancing process mining in pediatric oncology will require both methodological development of process mining techniques and progress within oncology itself, including improved data awareness, data quality, and stronger engagement with organizational and data-driven research.

Operational decisions significantly influence organizational performance and individual well-being. Decision mining offers a method to discover and analyze decision logic from decision logs, enhancing decision-making processes. However, evaluating the quality of decision discovery algorithms remains a challenge. While precision, fitness, generalization, and simplicity are well-established quality dimensions in process mining, their adaptation to the decision mining domain is underexplored. This study adapts these four dimensions to the necessary characteristics of decision models, providing a framework for evaluating decision discovery algorithms. Using a design science research approach, we develop tailored metrics and functions and demonstrate their application through a practical example of environmental permit management modeled in Decision Model and Notation (DMN). Precision measures how the discovered decision model reproduces the observed fact types and values from the decision log, detecting over-specification in the decision model. Fitness evaluates how completely the decision model covers the behavior in the decision log, identifying missing or under-specified elements in the decision model. Generalization assesses the model’s robustness to unseen decision cases by quantifying how well the discovered decision model performs beyond the training data. Simplicity captures the complexity in the discovered decision model in relation to a human actor-specified threshold. These insights guide decision model improvements, contributing to higher transparency, accountability, and fairness in operational decision-making processes. This research bridges a gap in the body of knowledge by providing a concrete methodology for evaluating decision discovery algorithms. The results support organizations in aligning decision models with regulatory requirements and public values, while also laying a foundation for future research.

Background: Clinical decision support systems (CDSSs) have gained prominence in health care, aiding professionals in decision-making and improving patient outcomes. While physicians often use CDSSs for diagnosis and treatment optimization, nurses rely on these systems for tasks such as patient monitoring, prioritization, and care planning. In nursing practice, CDSSs can assist with timely detection of clinical deterioration, support infection control, and streamline care documentation. Despite their potential, the adoption and use of CDSSs by nurses face diverse challenges. Barriers such as alarm fatigue, limited usability, lack of integration with workflows, and insufficient training continue to undermine effective implementation. In contrast to the relatively extensive body of research on CDSS use by physicians, studies focusing on nurses remain limited, leaving a gap in understanding the unique facilitators and barriers they encounter. Objective: This study aimed to explore the facilitators and barriers influencing the adoption and use of CDSSs by nurses in hospitals, using an extended Fit Between Individuals, Tasks, and Technology (FITT) framework. Methods: A qualitative study was conducted using semistructured interviews with 22 nurses from across the Netherlands, representing 3 hospital types: general (n=9), top-clinical (n=12), and academic (n=1). The sample included a diverse mix of practicing nurses, nurses-in-training, and clinical nurse information officers, with clinical experience ranging from 1.5 to 38 years. Interview transcripts were analyzed thematically, beginning with an inductive coding approach to identify key factors. These were then categorized deductively using the extended FITT framework. In total, 988 code instances were examined. To ensure analytical rigor, the coding process was separately conducted by 2 researchers and reviewed by an expert panel. Results: A total of 26 distinct factors were identified, categorized into 4 FITT dimensions: technology-individual, technology-task, task-individual, and organizational context. Of these, 11 factors were facilitators (eg, cognition, clarification, and prevention), 7 were barriers (eg, alarm fatigue, poor design, and limited digital proficiency), and 8 were both facilitators and barriers depending on the context (eg, acceptance, workload, and training). In addition, key value tensions emerged, such as the balance between standardization and professional autonomy, and the trade-off between enhanced decision support and increased administrative burden. Conclusions: The findings underscore the complexity of CDSS adoption in nursing practice, highlighting the interaction of facilitators and barriers across FITT dimensions. Practical recommendations include participatory design processes, targeted training programs, advanced alert management systems, and strong organizational support. Addressing value tensions and aligning CDSS functionality with nurses’ workflows can enhance adoption and optimize patient outcomes.