1. Introduction

The prevalence of mental disorders has been steadily increasing, placing significant pressure on health systems and imposing a substantial economic burden on society [1,2]. Aggressive behavior is a common manifestation during the acute phase of mental illness [3,4] typically characterized by extreme anxiety, hyperexcitability, tension, hostility, and physical aggression. Such symptoms not only exacerbate patients’ physical and psychological distress but also increase energy expenditure, thereby elevating the risks of metabolic syndrome and cardiovascular complications, and in severe cases, directly threatening patient safety [5,6]. Intravenous infusion plays a vital role in replenishing nutrition and supporting recovery in agitated patients. However, compared with the general patient population, managing intravenous infusion in these individuals presents unique challenges, including difficulties in catheter insertion, higher risks during various procedural steps [7], and increased psychological stress and workload for caregivers [8]. Existing intravenous infusion management guidelines [9–12] are designed for general use and do not adequately address the specific needs of psychiatric patients with poor treatment adherence or cooperation. Consequently, these guidelines fail to fully meet the therapeutic requirements and quality control standards necessary for this population, highlighting a critical gap in clinical practice. Furthermore, intravenous infusion is a continuous process in which failure at any stage may compromise overall care quality. Donabedian’s theoretical model [13], which evaluates quality from the perspectives of structure, process, and outcome, provides a systematic framework for addressing this challenge. Previous studies applying Donabedian’s framework in nursing have demonstrated its effectiveness in improving both care processes and outcomes [14–16]. However, to date, no infusion management protocol tailored to psychiatric patients with aggressive behavior has been developed using this framework.

Therefore, this study addresses the existing gap by developing a scientifically rigorous, feasible, and comprehensive management protocol for intravenous infusion in patients with aggressive behavior due to mental disorders. This protocol is expected to enhance nursing quality, improve patient safety, and provide a structured approach for clinical practice.

2. Materials and Methods

3. Results

3.1. Results of expert consultation

3.1.1. Expert engagement and authority.

Both rounds of expert questionnaires achieved a 100% response rate. The suggestion rates were 53.33% in the first round and 20% in the second round. The authority coefficients (Cr) were 0.915 and 0.920, both exceeding the 0.80 threshold. (see Table 1).

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Expert engagement and authority (n = 15).

https://doi.org/10.1371/journal.pone.0326717.t001

3.1.2. Degree of consensus and coordination.

In the first round, the coefficients of variation (CV) for item importance ranged from 0 to 0.115, and Kendall’s W was 0.506. In the second round, CV values ranged from 0 to 0.107, and Kendall’s W increased to 0.628 (P < 0.001), reflecting improved consensus among experts (see Table 2).

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Degree of consensus and coordination among experts (n = 15).

https://doi.org/10.1371/journal.pone.0326717.t002

3.1.3. Expert modification suggestions.

In the first round of consultation, experts provided 32 suggestions. After merging similar comments, 7 items were revised, 1 item was combined, and 8 new items were added. In particular, items with borderline quantitative indicators or divergent expert opinions were refined through wording revisions, item merging, or retention for further evaluation in the second round, based on comprehensive consideration of expert feedback. The main modifications included: (1) expanding the item on infusion safety management to specify operational procedures, emergency response protocols, and annual drills; (2) clarifying ambiguous terms such as “staff education” (revised to “ideological education”), “experienced nurses” (defined as ≥10 years of service) [30], and “puncture method” (revised to “puncture process”); (3) aligning protective restraint nursing with the Expert Consensus on the Implementation and Discontinuation of Protective Restraint; (4) revising “soothing” to “verbal de-escalation and environmental modification techniques”; and (5) standardizing complication management according to the Standards for Intravenous Therapy Nursing Practices. In addition, several new elements were incorporated, such as establishing an emergency call system, specifying training content and assessment mechanisms, including agitation level assessment, detailing measures for peripheral venous protection, adding specialized risk management (e.g., violence, suicide, harm to others), specifying educational content, protecting patient privacy, merging overlapping items, and providing formulas for indicator calculations.

In the second round, further refinements were made. These included specifying a particular risk assessment tool for specialized risk management and reorganizing humanistic care items into pre-infusion, intra-infusion, and post-infusion phases to enhance clarity and practical applicability.

3.2. Intravenous infusion management protocol for psychiatric patients with aggressive behavior (see Table 3)

4. Discussion

This study, guided by Donabedian’s framework, developed an intravenous infusion management protocol for psychiatric patients with aggressive behavior through literature review, semi-structured interviews, and Delphi expert consultation. The high response rate, strong authority coefficients, and significant Kendall’s W values indicated reliable consensus. These results are consistent with prior methodological reviews confirming that structured Delphi processes with iterative rounds and transparent consensus indicators enhance scientific rigor [31,32]. Comparable Delphi studies in psychiatric emergencies have also demonstrated that context-specific consensus contributes to developing practical, safety-oriented protocols [33].

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. Intravenous Infusion Management Protocol for Psychiatric Patients with Aggressive behavior.

https://doi.org/10.1371/journal.pone.0326717.t003

Compared with existing infusion guidelines designed for general patient populations [7,34], our protocol responds to the unique needs of psychiatric patients, who often show poor cooperation and heightened risks. The incorporation of emergency call systems, risk management strategies, and detailed competency criteria mirrors findings from European Delphi consensus guidelines on suicidal crisis management, which emphasized structured risk assessment and phased care strategies [35]. By aligning structural, process, and outcome dimensions, our protocol ensures both technical precision and humanistic care, echoing evidence that Donabedian’s triad provides a robust model for nursing quality improvement [36].

From an implementation perspective, the proposed protocol is intended to be integrated into routine psychiatric nursing workflows rather than applied as an additional standalone intervention. The structure–process–outcome framework enables nursing managers to translate the protocol into standard operating procedures, role-based task allocation, and quality monitoring indicators that align with existing ward management systems. In clinical practice, the protocol can be operationalised through a phased implementation approach, beginning with core safety-related components such as risk assessment, environmental management, and staff training, followed by the gradual incorporation of multidisciplinary collaboration and outcome evaluation. Importantly, the protocol was designed with flexibility to accommodate resource-limited psychiatric wards. While multidisciplinary participation and advanced equipment are recommended at the structural level, the protocol prioritises essential nursing actions that can be implemented using existing personnel and basic infrastructure. Key elements, including non-pharmacological de-escalation strategies, standardized assessment tools, and environmental safety measures, can be applied without substantial additional resources. This adaptable design enhances the feasibility of implementation across psychiatric settings with varying levels of staffing and resource availability.

The expert-driven refinements incorporated during the consultation process—such as specifying standardized risk assessment tools, reorganizing humanistic care items according to infusion phases, and standardizing complication management—further support the clinical applicability of the protocol [37]. These features are consistent with existing evidence suggesting that structured training and clearly defined operational standards can enhance nurse confidence and patient safety in psychiatric settings. Nevertheless, several limitations should be acknowledged. First, the expert panel was drawn from tertiary hospitals in two provinces, which may limit the generalizability of the findings to other regions or levels of care. Second, although a high level of consensus was achieved, the protocol has not yet undergone empirical validation. Future research should therefore focus on stepwise empirical evaluation of the protocol. Initial feasibility studies and pilot implementations in psychiatric wards are warranted to assess practicality, staff adherence, and acceptability in routine clinical practice. Building on these findings, implementation research and outcome evaluations could further examine the protocol’s effectiveness in improving patient safety, reducing staff burden, and optimizing clinical outcomes. Relevant outcome measures may include infusion-related adverse events, unplanned catheter removal, treatment continuity, patient satisfaction, as well as indicators of nursing workload, occupational stress, and confidence in managing aggressive behavior. In addition, the transferability of the proposed protocol to other regions or international contexts may be influenced by cultural, organisational, and health-system factors. Variations in perceptions of aggressive behavior, ward management structures, staffing patterns, and resource availability across psychiatric settings may affect implementation of specific components. Nevertheless, the core principles of the protocol—namely structured risk assessment, process-oriented infusion management, and outcome-based quality monitoring—are grounded in the Donabedian framework and are broadly applicable. Context-specific adaptation is therefore recommended to ensure alignment with local clinical practices and health-system characteristics.

5. Conclusion

Based on Donabedian’s theory, this study has developed an intravenous infusion management protocol for psychiatric patients with Aggressive behavior that encompasses first-, second-, and third-level items. Covering the entire process of intravenous infusion management, the protocol is scientifically sound, reliable, targeted, and feasible—and it may serve as a valuable reference for clinical nursing practice in psychiatric settings. Although this study did not empirically validate the protocol, it provides a theoretically grounded and consensus-based framework for clinical application. Future research will focus on feasibility testing and pilot implementation in psychiatric wards, particularly to evaluate practicality, staff adherence, and preliminary clinical outcomes prior to large-scale application.

Supporting information

Acknowledgments

The authors thank the patients who agreed to participate in the study. Liying Liu and Liangdong Lin conceived and designed the study. Yujie Hao and Wanyu Zhang were responsible for data collection. Yan Gao and Yulian Wei contributed to the analysis of the data. Lina Wang, Honghui Wang, wrote the article.