Understanding the Role of Clinical Decision Support Systems Among Hospital Nurses Using the FITT (Fit Between Individuals, Tasks, and Technology) Framework: Qualitative Study

Introduction

Information technology (IT) has become increasingly important in all industries, including health care. In the past 25 years, a wide variety of information systems were invented and implemented to aid health care professionals in decision-making, including clinical decision support systems (CDSSs). CDSSs are intended to improve health care delivery by enhancing medical decisions with targeted clinical knowledge, patient information, and other health information [1,2]. Examples of CDSSs used in health care are diagnosing health problems and medication-prescribing [3,4]. One of the use cases for using a CDSS is the prevention of medical errors [2-4]. Overall, the role of CDSSs in the health care sector has received substantial recognition, significantly impacting patient care quality, safety, and efficiency [2,5]. One critical user group of these systems is nurses, whose central role in direct patient care, monitoring, and coordination between health care teams makes CDSS integration essential for improving workflow efficiency and patient outcomes.

The use of CDSSs diverges significantly among various health care professionals in hospitals, such as physicians and nurses, according to their distinct roles and responsibilities. Physicians predominantly use CDSSs to aid diagnostic reasoning and to optimize treatment plans [2,6]. In this context, physicians rely on these systems to filter vast amounts of clinical data to gather precise diagnoses, suggest potential treatment options, and predict patient responses to different treatments. On the other hand, nurses employ CDSSs for a slightly different set of tasks. As emphasized in a study by Ruland and Bakken [7], nurses face significant challenges in managing complex care tasks, including continuous patient monitoring, timely detection of clinical deterioration, and prioritizing care under time pressure. CDSSs can help address these challenges by offering timely, structured support in areas such as care planning, infection control, and wound management. A widely used application of CDSSs is the filtering of alarms in intensive care for neonatal intensive care units, where the system helps distinguish clinically relevant alerts, such as those related to infusion pumps or critical vital signs, from nonurgent ones caused by, for instance, loose cables.

Despite these demonstrated benefits, the literature on CDSS adoption by nurses remains relatively limited and fragmented. A recent rapid review synthesizing 21 studies concluded that while CDSSs improve nursing decision-making, workflow efficiency, and patient outcomes, they also present persistent barriers such as alert fatigue, workflow misalignment, and limited adherence to system recommendations [4,8,9]. While existing research provides valuable insights into specific functionalities and adoption barriers, there is still a lack of systematic understanding of how CDSSs fit into the day-to-day clinical realities of nurses. Recent qualitative and mixed methods reviews call for more in-depth exploration of nurses’ experiences and perspectives, as these are often underrepresented compared to physician-centric studies [8].

To better capture this complexity, we use the extended Fit Between Individuals, Tasks, and Technology (FITT) framework, which states that the success of health IT adoption depends on the alignment between user characteristics, task demands, technology features, and organizational context (Figure 1) [10,11]. The FITT model has been applied and validated in various health IT studies, including electronic health record (EHR) and telemonitoring adoption, and is particularly well suited to understanding technology adoption in complex, team-based clinical environments [10,12-14]. Compared to behavioral models such as the Unified Theory of Acceptance and Use of Technology (UTAUT), FITT places greater emphasis on the practical interaction between people, systems, and work processes, making it especially relevant for exploring the nursing context [10,15]. The disconnect between the theoretical advantages of CDSSs and their real-world application in nursing highlights the urgent need to address these challenges [16].

Prior reviews cover clinicians broadly or cover nonhospital settings, with limited qualitative evidence from hospital nurses and little integration of findings within an explicit implementation lens [8,17]. Framework and feature scoping reviews inform what may help adoption but not how ward-level nursing work experiences map onto domains such as FITT [18,19]. We contribute nurse-focused qualitative evidence, map facilitators, barriers, and value tensions onto FITT, and compare patterns across Dutch hospital types.

This paper aims to address the gaps in our understanding of the potential complex and relatively unexplored dynamics associated with decision support systems usage in a nursing context. Through the extended FITT framework, we explore how individual nurse factors, executing the task of decision-making, and the supporting technology interact to facilitate or hinder effective decision support system usage.

‎

Therefore, the research question addressed in this paper is “What are the facilitators and barriers for hospital nurses regarding the use of a clinical decision support system?”

From a theoretical perspective, the goal is to contribute to the body of knowledge that informs health care policy and system design, fostering environments where technology effectively supports and enhances patient care delivery in hospitals. For example, studies have shown that user-centered design and tailored technology adaptation enhance the usability and effectiveness of CDSSs [6]. By using the extended FITT framework, we can identify how CDSSs can be better tailored to fit the specific needs of users, such as nurses and physicians, reducing barriers and enhancing facilitators for more seamless integration into clinical workflows [2]. This approach ensures that technology aligns with real-world practices, leading to improved patient outcomes [10].

From a practical perspective, this study aims to contribute to the limited body of research on the implementation and use of CDSSs in a hospital nursing context. By offering insights into practical strategies to overcome barriers and leverage facilitators, this research provides valuable guidance for (quality) nurses, health care IT developers, and management. A better understanding of CDSS design and adoption specifically tailored to nurses’ needs can support better integrated workflows and enhance both nursing practice and patient outcomes.

Methods

Design

In this qualitative study, semistructured interviewing was chosen as the primary data collection method to operationalize the FITT framework because it allows for in-depth exploration of nurses’ experiences, perceptions, and challenges with CDSSs in a hospital setting. Given the complex nature of the interactions between individuals, tasks, and technology, interviews offer a nuanced understanding that is essential for capturing the subjective experiences that other methods, such as secondary sources, may not provide.

The study specifically focuses on a CDSS in a hospital setting. Our research methodology was designed with an orientation toward the extended FITT framework, which suggests that successful technology implementation and use is largely determined by the fit between 3 aspects and the organizational context [10,11]. This framework was particularly relevant to our goal of identifying facilitators and barriers in the use of CDSSs by nurses. By focusing on the alignment between the individual users (nurses), their tasks, the technology itself, and the organizational context, we are able to explore how CDSSs either support or hinder their work. This approach enabled us to find specific factors that act as facilitators, enhancing system usability and integration, as well as barriers that impede effective use.

Setting

The interviews were conducted face-to-face at the participants’ own hospital in the Netherlands. In the Dutch health care system, hospitals are generally categorized into 3 types: general hospitals, top-clinical hospitals, and academic hospitals. General hospitals provide a broad range of standard medical services and are typically smaller in scale. Top-clinical hospitals offer more specialized care and often serve as regional centers for complex procedures; they are also typically affiliated with teaching programs. Academic hospitals are university-affiliated institutions that combine advanced patient care with medical research and education. This diversity in hospital types was considered in the sampling strategy to ensure a range of organizational contexts.

Participants

In this study, we contacted a total of 25 nurses for an interview from different hospital types, with each nurse having varying levels of experience using CDSSs. Participants were selected using a combination of purposive and convenience sampling. We opted for a larger sample size because, while saturation in qualitative studies typically occurs with 6-12 interviews, more complex contexts often require 15-30 interviews to ensure comprehensive coverage of the subject matter [20]. In total, 22 nurses accepted the invitation for an interview. Saturation was monitored during the data collection process and was considered reached after approximately 18 interviews. The remaining interviews confirmed that no new themes were emerging across hospital types and professional roles. The remaining interviews added nuance or illustrative quotations only, and the codebook remained stable after the 18th interview, supporting the judgment that thematic saturation had been achieved. This diversity in backgrounds, roles, and hospital types allowed for a broad range of perspectives to be captured. The participants included practicing nurses, nurse-in-training, and clinical nurse information officers (CNIOs). Their experience with CDSSs varied considerably: some used specific CDSS tools regularly (eg, risk scoring), while others only knew the term CDSS or were exposed indirectly, such as through implementation projects or digital protocols. By incorporating nurses with experience ranging from 1 year to over 3 decades, we ensured the data reflects both novice and seasoned professionals’ experiences with CDSS usage. This variation in expertise, hospital types (general, top-clinical, and academic), and professional roles provides a rich and comprehensive understanding of the facilitators and barriers associated with CDSS use.

Data Collection

A semistructured interview guide was developed based on the key constructs of the extended FITT framework: individual, task, technology, and organizational context. The selection of interview questions was informed by prior studies using the (extended) FITT framework in CDSS research [10,12-14], as well as broader models of technology adoption in health care, such as the UTAUT model (Table 1). This theoretical foundation ensured the alignment of our interview protocol with existing literature while allowing for an in-depth exploration of barriers and facilitators to CDSS use in hospital-based nursing practice. A semistructured approach was chosen to ensure flexibility, allowing participants to elaborate on their experiences while maintaining alignment with the research objectives. This method facilitated the identification of context-specific insights regarding CDSS implementation in nursing workflows.

Interviews were conducted in person alone at the participating hospital by one of three trained interviewers (Susan Bruggeling, Judit Putnoki, and Johanna Zuidhoff) with backgrounds in nursing and health informatics under the supervision of 2 researchers (MB and TVH). Their familiarity with clinical environments and training in health informatics helped build rapport with participants and ensure relevance of follow-up questions. At the same time, their background may have introduced implicit expectations about functionalities or the value of a digital system, potentially influencing how (follow-up) questions were asked or interpreted. This risk was mitigated through joint interview training and the use of a semistructured interview guide. All interviewers completed a training session together before the data collection commenced and conducted 2 pilot interviews each to refine the protocol, ensuring question clarity, consistency, and methodological rigor. These pilot interviews, which were discussed together, also allowed for the evaluation of question validity and comprehensibility from a nursing perspective. These interviews were excluded from the data analysis. A standardized interview protocol helped ensure consistency and minimize bias. The interviewers did not have a prior personal or extended professional relationship with the participants, except for limited professional contact in isolated cases.

Each interview lasted 30 to 60 minutes and was audio-recorded with participants’ written informed consent. The participant was interviewed a single time. Field notes from the interviewer were not gathered during the interview. Data collection occurred over a 4-month period between September 2022 and December 2022, ensuring representation from nurses across different shifts, departments, and experience levels.

Data Analysis

The recorded interviews were transcribed verbatim and were then analyzed. The transcribed interview was sent back to the participant for a check, but no member check was conducted. Our analysis process is based on the 6 steps outlined by Braun and Clarke [21].

Steps 1 and 2, following Braun and Clarke’s approach, consisted of open coding conducted with an inductive first approach to identify factors directly emerging from the data. This initial step allowed us to ground the findings in participant experiences without being constrained by predefined frameworks, ensuring that all relevant elements from the interviews were captured.

Steps 3 and 4 from Braun and Clarke were simultaneously started in which the identified factors were directly after they were identified systematically plotted onto the FITT framework in seven categories using a deductive approach: (1) the fit and (2) no fit between “technology and individual,” (3) fit and (4) no fit between “task and technology,” (5) fit and (6) no fit between “individual and task,” and (7) “organizational context.” This 2-step simultaneous process ensured a balance between inductive coding for emerging factors and deductive coding rooted in the FITT framework. After we started with step 5, the identified factors were categorized as facilitators or barriers. This step involved analyzing how each factor influenced the adoption and implementation of CDSSs within the context of the FITT framework. In the last step (6), facilitators were defined as elements that promoted or enhanced CDSS use, while barriers were identified as factors that hindered the usage of CDSSs.

The analysis was conducted in Atlas.ti (v24; Lumivero) by 2 researchers (MB and SB) independently to reduce interpretive bias and to ensure the reliability of the findings. The researchers had a background in health care and informatics. Any discrepancies between the researchers were resolved after the coding through discussion by the 2 researchers until consensus was reached. After the analysis, an external expert session was conducted to validate the identified codes found by 2 independent researchers (MB and SB). The session lasted for 2 hours and involved 5 participants: 2 researchers (TVH and DS) with expertise in the health care domain, 1 researcher (KS) with experience in qualitative coding, 1 PhD student (MB), and 1 research assistant (SB). During the session, the group critically reviewed the codes, discussed their relevance, and ensured that they accurately reflected the data to find consensus. This collaborative approach allowed for a robust validation process, ensuring that the coding framework was aligned with the domain expertise of the participants [22].

Following the thematic analysis, we conducted a post hoc analysis to explore the value tensions inherent in the identified factors. Value tensions refer to conflicting priorities, needs, or expectations that emerge during the adoption and implementation of CDSSs, often reflecting deeper ethical, organizational, or professional dilemmas [23]. The value tensions were analyzed by re-examining the coded data through the lens of key ethical and practical trade-offs reported in prior literature on health informatics and digital health care adoption.

Ethical Considerations

This research was conducted following Dutch human subject regulations. Since the Dutch Medical Research Involving Human Subjects Act (in Dutch “WMO”) did not apply, no official ethical approval was required. To confirm this assessment, the research team completed the decision tree for non-WMO research, as provided by the medical ethics review committee, which is aligned with national ethical guidelines [24]. The outcome confirmed that the study was classified as non-WMO. The act applies only to medical research that infringes on a subject’s physical or psychological integrity [25].

Although no official ethical approval was required, all necessary precautions were taken to protect participants’ anonymity and confidentiality. All participants were informed about the purpose and procedures of the study before their participation, including the nature of the questions to be asked, the intended use of the information they provided, and the fact that the data would be anonymized and kept confidential. The participants did not receive (financial) compensation for the interview. The data are stored according to a data management plan for this study and are not used in other studies. All participants had the opportunity to read back the transcript of the interview to check their data.

Written informed consent was obtained from all participants. They were assured that their participation was voluntary, and they had the right to withdraw from the study at any time without any repercussions. Care was taken to ensure that the interview setting was comfortable and nonintimidating, with interviews scheduled at times convenient for the participants, within their own hospital, to minimize disruption to their work schedules.

During the data analysis, all identifying information was removed to ensure participant anonymity. Access to the data was restricted to the research team, and all data were securely stored in accordance with the General Data Protection Regulation. The findings are reported in a way that individual participants could not be identified, ensuring their privacy and confidentiality throughout the research process.

Results

Participants’ Characteristics

A total of 22 nurses participated in the study. Participants were drawn from general, top-clinical, and academic hospitals across the Netherlands, ensuring variation in organizational settings. The sample included nurses with a range of clinical backgrounds and levels of experience, including practicing nurses, nurses-in-training (n=7), CNIOs, and medical technology officers. Participants’ experience with CDSSs varied, from those who used such systems daily to those involved in implementation or training efforts.

Participants were recruited from general, top-clinical, and academic hospitals, reflecting the range of hospital types in the Dutch health care system. Table 2 summarizes the characteristics of the participants.

The Fit of Clinical Decision Support Systems

Thematic analysis revealed 26 distinct factors that influence the adoption and use of CDSSs by nurses. These factors were categorized across the 4 FITT domains: technology-individual fit, technology-task fit, task-individual fit, and a redefined category of organizational context. During the coding process, 988 code instances were identified (Table 3). This number reflects the volume and depth of qualitative data analyzed across all interviews, including recurring themes and insights. However, our reporting focuses on the consolidated factors, as these represent the core themes relevant to understanding facilitators and barriers. Each factor was assigned based on thematic relevance rather than frequency, in line with qualitative best practices [21].

We examined the identified factors to determine whether each factor serves as a facilitator (coded as “fit”), a barrier (coded as “no fit”), or both (coded as “fit and no fit”). Our analysis revealed that 10 factors function exclusively as facilitators, 7 factors operate solely as barriers, and 8 factors were coded as both facilitators and barriers. Table 4 presents the identified factors with an explanation.

Cross Synthesis

Taken together, patterns were role- and context-sensitive. Nurses-in-training described CDSSs as a scaffolding for decision-making and knowledge updating; experienced nurses stressed alert noise, extra clicks, and loss of flexibility; and CNIOs framed success around structured training and content governance. By hospital type, general hospitals more often emphasized basic usability and resource limits, while top-clinical hospitals highlighted interoperability, device or EHR integration, and alarm management. These differences help explain why the same functions could operate as facilitators in one setting and barriers in another.

Value Tensions Across FITT Categories

In the final stage of our analysis, we examined how the identified factors shown in 4 interacted across the FITT domains and gave rise to underlying value tension situations where competing priorities or goals created friction. These tensions were not predefined but were identified inductively by examining patterns across coded excerpts that involved contrasting experiences or trade-offs described by participants and discussed by the researchers.

One key relationship is the interaction between autonomy (technology-individual fit) and restriction of freedom (technology-task fit). Participants described how CDSSs use could enhance autonomy by reducing their reliance on colleagues and supporting independent decision-making. However, they also noted that these same systems sometimes imposed constraints through rigid protocols or limited customizability, illustrating a tension between increased autonomy and loss of flexibility. For example, nurses expressed concerns about being unable to deviate from standardized recommendations when clinical judgment suggested an alternative course of action. This tension illustrates the need for CDSS designs that balance standardization with flexibility, allowing nurses to exercise their professional judgment while maintaining consistency in care delivery.

Another notable tension arises between uniformity (task-individual fit) and safety (technology-task fit). Uniformity, achieved through standardized processes and documentation, was viewed as a facilitator for reducing variability in care and ensuring consistent outcomes. However, overly rigid standardization may overlook unique patient circumstances, potentially compromising safety in complex or atypical cases. For example, nurses reported that while CDSSs provided valuable guidance for routine cases, their inability to account for nuanced patient conditions sometimes limited their applicability. Addressing this requires systems that allow for both adherence to best practices and customization to fit individual patient needs.

The relationship between management support (organizational context) and training (technology-individual fit) highlights the critical role of leadership in sustaining user engagement and system adoption. Strong leadership was identified as essential for providing adequate training resources and ensuring alignment between institutional goals and technology implementation. However, the absence of visible support from management often led to gaps in training, leaving nurses feeling unprepared to use CDSSs effectively. This finding emphasizes the interconnectedness of organizational and individual factors and the importance of a cohesive approach to system adoption.

The dual nature of workload (task-individual fit) also demonstrates how factors can function as both facilitators and barriers depending on context. CDSSs can reduce workload by streamlining routine tasks, such as documentation or early warning detection. Equally, it can introduce additional administrative burdens, such as repetitive data entry or managing excessive alerts. This tension highlights the need for workflow optimization, ensuring that CDSSs enhance efficiency without creating new inefficiencies.

A broader value tension lies in the interplay between patient-focused care (technology-task fit) and contact (task-individual fit). While CDSSs facilitate patient-centered care by supporting shared decision-making and improving health outcomes, participants expressed concerns that increased reliance on technology might reduce personal interactions with patients. Nurses reported that time spent navigating systems could detract from meaningful patient engagement, particularly in time at bed parts of care. Balancing these priorities requires designing systems that integrate seamlessly into workflows, minimizing the time spent on system interaction and maximizing direct patient care.

Lastly, the findings suggest that alarm fatigue (technology-task fit) interacts with cognition (technology-individual fit) in ways that affect user engagement. While CDSSs enhance cognition by simplifying data and providing actionable insights, excessive or irrelevant alerts can overwhelm nurses and reduce their ability to focus on critical information. This underscores the need for smarter alert algorithms that prioritize relevant notifications, ensuring nurses are directed toward meaningful tasks.

Discussion

Principal Results

This study examined the facilitators and barriers influencing the adoption and use of CDSSs within nursing practice in Dutch hospitals. The findings highlight 11 facilitators, 7 barriers, and 8 factors that are both facilitators and barriers. All factors were categorized according to the FITT framework. Beyond identifying these individual facilitators and barriers, the findings highlight value tensions, situations where competing priorities or perspectives create conflict within or between categories.

Integration With Prior Research

The identified facilitators, barriers, and value tensions in CDSS adoption among nurses align with existing theories of health informatics adoption, such as the multidimensional nature of success described in Brender et al [26]. Similar to our use of the extended FITT framework, Brender et al emphasized that clinical and decision support systems are influenced not only by technical quality but also by organizational, behavioral, and management factors. In our study, the importance of factors like workflow alignment, user training, and leadership support shows their identified success dimensions: user involvement, management commitment, and fit with the organizational context. Moreover, several barriers we observed—such as alarm fatigue, lack of digital proficiency, and increased workload—reflect the types of failure criteria Brender et al highlight, including overloading the user, underestimating user acceptance, and not understanding the impact on work procedures. This reinforces earlier conclusions from our own rapid review on CDSS adoption in nursing, which identified similar facilitators and barriers, and emphasized the importance of aligning CDSSs with the values and realities of nursing practice [9]. Together, these findings demonstrate that while technology continues to advance, many foundational barriers to adoption remain persistent in practice.

Our findings also align with those of Meunier et al [8], who found that lack of integration with workflows, poor usability, and insufficient stakeholder engagement were persistent barriers in CDSS implementation, particularly in nursing and primary care settings.

Topaz et al [16] further reported that nurses often experience a misalignment between electronic health systems and their professional identity. Our findings present this, especially in relation to tensions between standardization and autonomy, or efficiency and meaningful patient contact. These relational and ethical aspects of CDSS use, rarely captured in earlier models, provide a novel perspective on adoption challenges.

However, our study also contributes new insights by connecting identified value tensions to the FITT framework. The identified tensions highlight the ethical and relational dimensions of CDSS adoption specific to nursing, suggesting that adoption is not only a technical or structural issue, but also a negotiation of values embedded in care practices.

This comparison underscores that while technology has evolved, many of the fundamental implementation challenges identified by earlier studies remain. Despite the passage of time, barriers related to user experience in IT systems, organizational change, and professional identity continue to shape CDSS adoption. This contribution lies in explicitly mapping these tensions across the FITT framework, offering a more dynamic view of the complexity of IT adoption. This reaffirms the need for more context-sensitive, participatory, and ethically adjusted approaches to implementation.

Reflection on FITT Dimensions

The following section reflects on the main findings across each FITT dimension.

Technology-Individual Fit

The findings revealed that acceptance of CDSSs is highly contingent on trust and user involvement in the system’s design. While participants recognized the potential of CDSSs to streamline workflows and enhance decision-making, lack of involvement in system development emerged as a major barrier. This underscores the need for participatory design processes that engage nurses early in system creation. Additionally, digital proficiency remains a challenge for some, particularly for nurses with limited prior experience with technology, emphasizing the importance of targeted training programs. Experience-level asymmetries were clear: nurses-in-training emphasized cognitive scaffolding and confidence gains, whereas experienced nurses more often raised concerns about autonomy loss, supporting role sensitive design and training pathways rather than one-size-fits-all rollouts. Cognition and clarification were significant facilitators, as CDSSs simplified complex data and provided timely knowledge updates. These factors indicate that well-designed systems can enhance clinical confidence and reduce cognitive overload. However, inadequate support and inconsistent training hindered system adoption for many participants, suggesting that sustained organizational investment in user education and technical support is essential for long-term success.

Technology-Task Fit

While CDSSs demonstrated substantial potential to enhance information supply, continuous monitoring, and reporting, barriers such as alarm fatigue and poor design remained pervasive. Frequent, irrelevant alerts desensitize users and pose risks to patient safety, highlighting the need for advanced alert prioritization algorithms. Similarly, poorly aligned system interfaces exacerbated user frustration, emphasizing the importance of usability testing tailored to clinical workflows.

The system’s ability to support patient-focused care was identified as a key facilitator, enabling shared decision-making and improving patient outcomes. However, concerns around restriction of freedom and safety suggest that overly rigid protocols and potential system errors may undermine professional autonomy and trust. Developers must balance standardization with flexibility to address unique patient needs.

Alarm fatigue, as identified in this study, aligns with findings by Abell et al [27] which emphasize its impact on user desensitization. However, our research adds nuance by highlighting that nurses in critical care settings perceive this overload as exacerbated by the lack of integration between CDSS alerts and existing workflows. These findings reframe alert fatigue from a user-resilience issue to an organizational signal-engineering problem (thresholds, routing, escalation, and device or EHR integration), consistent with evidence that unmanaged alarms desensitize even experienced staff [28-30].

This underscores the importance of addressing workflow alignment alongside technical improvements.

Task-Individual Fit

Participants valued CDSSs for promoting accountability and uniformity, which reduced variability and enhanced care quality. However, the dual impact of workload emerged as a central theme. While CDSSs streamlined some tasks, additional administrative burdens often negated these efficiency gains. Future implementations should aim to optimize workflows by minimizing redundant data entry and integrating CDSSs seamlessly into existing processes.

The system’s role in prevention was widely acknowledged, particularly in its ability to identify early warning signs. However, concerns about work ethic highlight the risk of overreliance on CDSSs, potentially diminishing professional engagement and critical thinking. These findings suggest a need to reinforce the complementary role of CDSSs rather than allowing them to dominate clinical decision-making.

The interaction between accountability and workload presents another tension in this category. While accountability is facilitated by CDSSs through standardized documentation, the added burden of maintaining these standards may increase workload. Accordingly, standardization aids accountability but can also constrain professional judgment in atypical cases; guardrails with explicit override (and rationale capture) may offer a better balance than hard locks [26]. This dual effect reflects the need for careful workflow design to balance efficiency with accountability.

Organizational Context

Organizational factors played a significant role in shaping CDSS adoption. Strong management support was a key facilitator, as engaged leadership provided the necessary resources and alignment with institutional goals. In addition, limited budgets also mentioned as cash, hindered system upgrades and training opportunities, excessively affecting resource-constrained settings. Effective change management emerged as both a facilitator and a barrier. While structured planning and clear communication facilitated smoother transitions, poor management of change processes created resistance and uncertainty among staff. This highlights the importance of robust implementation frameworks that address user concerns and foster collective ownership of CDSSs.

The findings align with Kujala et al [10], who emphasized the role of strong leadership in enabling successful CDSS implementation. However, our study highlights a unique aspect: the critical role of consistent communication in sustaining nurse engagement throughout the implementation process. This nuance underscores the need for transparent leadership at every stage of adoption.

Taken together, these recommendations move beyond confirmation by specifying where and why common remedies (eg, more training and more standardization) succeed or fail in everyday nursing work, and by redirecting levers from individual users to organizational signal- and workflow-engineering.

Implications for Practice

Based on the findings, several recommendations emerge for the successful adoption and integration of CDSSs into nursing practice. The recommendations are ordered from “quick wins” to “structured investments.”

Quick Wins

The first quick win is to improve acceptance and usability; it is essential to involve nurses at all stages of CDSS development and implementation. Providing opportunities for nurses to share their insights ensures that the system aligns with clinical workflows and minimizes resistance stemming from lack of involvement [31].

Targeted training programs that are tailored to the diverse digital proficiency levels among nurses are also critical. These programs should emphasize the complementary role of CDSSs, reinforcing critical thinking and professional autonomy while addressing knowledge gaps, particularly for less experienced nurses. Continuous education can help ensure that nurses feel confident and competent in their interactions with CDSSs, for example, by integrating data-driven competencies into nursing studies as mentioned by the American Association of Colleges of Nursing [32].

The last quick win is that organizations must also ensure nurses have access to technical and operational support. Clear communication channels and structured feedback loops are essential to identify and resolve user challenges. Allocating dedicated training time for nurses to engage with CDSSs and related tools before implementation can possibly further enhance their effectiveness and ease of use. Leadership support is also critical for developing a culture of innovation. Leaders must prioritize CDSS adoption by demonstrating its value to clinical practice and allocating resources.

Medium-Term Investments

A medium-term investment is the preventive potential of CDSSs, which should be leveraged to drive adoption by integrating tools that alert nurses to early warning signs of patient deterioration. Highlighting this capability can underscore the system’s role in improving patient outcomes. It is essential to address ethical and professional concerns by reinforcing the role of CDSSs as a support tool rather than a replacement for clinical expertise. By emphasizing its ability to enhance decision-making while preserving the human aspects of care, organizations can mitigate fears of overreliance and loss of meaningful work, as mentioned in multiple studies [33,34]. A responsible implementation approach that considers these elements, for example, value-sensitive design [35], can be used to ensure that CDSSs contribute positively to nursing practice and patient care.

Another medium-term investment that can be mitigated is alarm fatigue, a barrier to CDSS adoption [28-30]. This can be mitigated through the incorporation of advanced alert management systems. Developers should implement smarter algorithms that prioritize clinically relevant alerts, distinguishing between critical and noncritical notifications. Such improvements would reduce cognitive overload and enhance patient safety. Similarly, usability issues must be addressed through usability testing with frontline nurses during the design phase. Intuitive systems that integrate seamlessly into existing workflows can reduce redundant administrative tasks and minimize the burden of data entry, ultimately enhancing efficiency.

CDSS developers should aim to balance standardization and flexibility in system design. While CDSSs can enhance uniformity in care delivery, they must also accommodate unique clinical situations by allowing for overrides and customization. Supporting decision-making in this way ensures that nurses maintain professional autonomy and can reflect on past decisions to learn from previous cases.

Long-Term Structural Investments

While many actions are primarily enacted at the ward or team level, their impact ultimately depends on several system-level preconditions in Dutch hospitals. Hospitals should ensure that CDSS content and handovers align with nationally endorsed information standards for electronic exchange; plan integrations on a standards-based roadmap so device-EHR-CDSS interfaces are reliable and monitored; formalize hospital-wide governance for alerts and decision support (clear ownership of default settings, change control, and routine review of alert volumes, overrides, and time-to-action); and schedule go-lives with protected training time and backfill, acknowledging ongoing nurse-staffing pressures.

Limitations

This study provides insights into the adoption and use of CDSSs by nurses in hospital settings; however, several limitations should be acknowledged.

First, the study focused exclusively on nurses’ perspectives, offering a detailed exploration of their experiences and challenges with CDSSs. While this single-disciplinary focus was essential to understand the specific interactions between nurses and CDSSs, it does not fully capture the interdisciplinary nature of CDSS usage in hospitals. Input from other stakeholders, such as physicians, administrators, and patients, could provide a more holistic understanding of the factors influencing adoption and implementation.

Second, while the extended FITT framework incorporated organizational context, this study did not explore specific organizational factors such as workflow dynamics, hierarchical structures, or institutional policies that may shape CDSS adoption. These organizational elements likely influence facilitators and barriers in significant ways, and future studies should consider a more granular analysis of these factors, focusing specifically on the organizational context.

Third, the qualitative design of this study enabled in-depth insights into nurses' lived experiences but limits the generalizability of the findings. Although the sample included nurses with diverse roles and experience levels from different hospital types, the findings remain context specific. Quantitative or mixed methods approaches could validate these findings and provide broader generalization. Furthermore, while this study was conducted within the Dutch health care system, many of the identified facilitators and barriers, such as alarm fatigue, training, usability, and leadership support, are relevant across other health systems. Nonetheless, differences in health care funding models, workforce structures, and digital infrastructure may influence the generalizability of our findings. In addition, while the data collection was over 2 years ago, insights remain applicable to ongoing implementation efforts as the core experiences and challenges described by nurses remain highly relevant.

Last, a principal limitation concerns our overlap of inductive coding and deductive FITT mapping, as during the analysis of the interviews, the factors and the fit or no fit aspects were coded simultaneously, combining the first two phases. Although this provided efficiency, we acknowledge that conducting open coding and deductive categorization in parallel may have introduced some risk of bias or reduced the emergence of unexpected themes. To address this, we initially applied an open inductive approach first to allow themes to surface directly from the data before aligning them with the FITT framework. This approach was further strengthened by an expert validation session to ensure that no relevant themes were constrained or overlooked. We acknowledge that subtle or emergent themes may still be underrepresented because of this approach. Nevertheless, we recommend that future studies consider separating inductive and deductive phases more clearly to enhance methodological rigor and reduce interpretive bias.

Future Research Directions

Future research could expand this study by incorporating quantitative data collection to allow for broader generalization of the findings. Quantitative data could provide more precise insights into the extent and impact of various facilitators and barriers. Additionally, further research could explore variations in barriers across different settings or user groups, which may inform more tailored implementation strategies that address specific needs or challenges in CDSS usage.

Another promising future research direction is the application of decision mining as a quality assessment instrument. By retrospectively analyzing decisions made by nurses when using CDSSs, decision mining can offer valuable insights into decision-making patterns, the quality of clinical judgments, and potential areas for improvement. This retrospective approach not only provides a method to assess the effectiveness of CDSSs but also identifies where barriers may have impacted decision quality. These insights can be used to refine both the CDSS itself and the support systems around its implementation, ensuring that barriers, such as design or usability issues, are addressed. In doing so, decision mining could play a key role in optimizing CDSSs for nursing practice, improving both their adoption and effectiveness in enhancing patient care.

Conclusions

This study highlights the factors influencing the adoption and use of CDSSs by nurses in hospitals, emphasizing the importance of achieving a fit between individuals, tasks, technology, and organizational contexts. The extended FITT framework provided a valuable lens for identifying facilitators and barriers that can guide practical improvements. By addressing critical areas such as user engagement, training, system usability, and organizational support, CDSS implementation can be more effectively tailored to meet the needs of nurses. These findings contribute to a growing understanding of how to optimize CDSS integration into nursing workflows, ensuring it enhances clinical decision-making and improves patient outcomes.