Clinical decision support system for clinical nurses decision-making on nurse-to-patient assignment: a scoping review protocol

Background

Nurse resource planning in hospitals is a very complex dynamics which comprises a few steps.1 Punnakitikashem suggested four stages regarding nurse planning: (1) nurse budgeting, (2) nurse scheduling, (3) nurse staffing and (4) nurse assignment.2 Nurse-patient assignment (NPA) is described as the process of aligning a nurse with a specific patient for a designated duration.3 Compared with other stages, the NPA stage has not received sufficient attention in previous research.4 Optimal nurse-to-patient assignment is a pivotal factor in healthcare delivery, directly impacting patient outcomes and nursing workloads.5 Previous researchers have reported that a positive perception of appropriate NPAs positively affects job satisfaction, intent to stay and nurses’ perceived quality of patient care.6 Another researcher also found a positive correlation between positive perception of patient assignment and nurses’ perceived nursing performance.7 In contrast, when nurses perceive that their assignments are not appropriate, it could lead to a negative effect on nursing care and nurses’ job satisfaction.8

In general, the NPA process is carried out at the beginning of each shift, and the decision-making for each shift is conducted by the charge nurse of the previous shift. This process is highly intricate requiring critical thinking competency,9 as there are numerous factors to be considered. Allen3 identified these factors in the NPA process through interviews with charge nurses in a non-profit hospital. In summary, mainly three categories were outlined: patient factors (demographics, acuity and length of stay), environmental factors (nurse-patient ratio, location and current staffing) and nurse factors (nurses’ competence, relationships, nurse demographics and preferences).3 Another group of researchers also identified six factors: human resources, physical resources, work design and technology, administrative practices, patient severity and patient turnover.10 Also, the comparative importance among factors varies according to the settings which makes the NPA process much more complex. In the emergency room, triage nurses’ main consideration is patients’ condition, patient flow and nurses’ experience,11 while a home care setting places importance on nursing continuity when deciding patient assignment.12 Likewise, the decision process of the NPA is so complicated; however, this process solely depends on the charge nurse’s judgement. The NPA process is often a manual process in which the charge nurse must sort through multiple decision criteria in a limited amount of time.3 Charge nurses bear the responsibility for overseeing a specific shift, encompassing various duties such as staffing management, assigning nurses, ensuring the quality and safety of patient care, providing teaching and counselling, and handling administrative tasks.13 14 Among their 8 hours of work time, many charge nurses were reported to spend considerable time (more than 30 min) on the assigning process, which could cause inefficiency in nursing work for themselves.15 Furthermore, it is noteworthy that the rotation of charge nurses takes place every other shift; the practice of the NPA decision-making varies based on their own circumstances.16 Given its subjective and complex nature, this process carries a consistent risk of uneven workload distribution among nurses, which could lead to a negative effect on a nurse’s job satisfaction.5 The American Nurses Association (ANA) has recommended that in cases where nurses perceive the distribution of patient acuity among their colleagues as inappropriate, they should consider refusing the assignment.17 However, in actual clinical settings, the nursing organisational culture often tends to be hierarchical,18 and in general, nurses who are assigned patients are often junior in terms of experience, compared with charge nurses. This structure makes it challenging to refuse NPA decisions made by charge nurses. Therefore, there is a growing need to apply objective procedures or methods that can overcome the subjectivity inherent in the decision-making process.

To support the complicated decision-making process for the charge nurse’s assignment, Cathro proposed a practical guide for patient assignments, which encompasses three key factors: patient acuity, nursing continuity and safety concerns.19 Allen also provided effective steps for NPAs adding insight into setting priorities among the considerations, adjusting the assignment as the condition is constantly changing and evaluating the assignment.20 Furthermore, researchers have investigated more precise methods for accurately assessing patient acuity to help fair and optimal assignment.21–23 While this practical guideline proves valuable for charge nurses, it does have limitations, notably in its inability to resolve subjectivity issues since it relies on individual nurses’ judgement. Also, methodological development of measuring patient acuity has its limitations, as it could not capture the nurse or environmental factors, such as the distance between the assigned patients and nurses’ experience.

Some researchers have developed clinical decision support systems (CDSSs) for this process to address subjectivity and complexity in this context.1 24 CDSS, a form of health information technology, is a software system designed to enhance healthcare delivery by suggesting optimal decisions using patient and clinical health data.25 Due to the rapid advancement of health information technologies in hospitals,26 patient acuity score and healthcare professionals’ workload, which are key considerations in NPA, are automatically registered and processed in electronic medical records.27 28 This has laid the foundation for developing CDSSs aimed at optimising NPAs by considering various health information. Using the CDSSs in NPA can enhance nursing care efficiency, promote care continuity, and result in a more equitable workload distribution for nurses. However, to date, there is no comprehensive study that synthesises the use of informatics approaches applied for optimising NPAs. Therefore, the purpose of this scoping review is to compile articles that discuss the CDSSs designed for the NPA process.

Methods

Scoping reviews are useful tools to identify current knowledge in a given field and to examine how research is conducted on a certain topic.29 Scoping reviews allow for a comprehensive overview of the existing evidence, accommodating the evolving nature of the field and the heterogeneity across diverse research settings.29 Since applying CDSSs in NPA is a relatively recent development, and its application may vary across diverse nursing settings, a scoping review is deemed more appropriate than a systematic literature review.

Our review will follow the six stages of the scoping review suggested by Arksey and O’Malley: (1) identifying the research question, (2) identifying the relevant studies, (3) selecting studies, (4) extracting the data, (5) collating, summarising and reporting the outcomes and (6) performing a consultation exercise.30 We will report the results according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis extension for Scoping Reviews (PRISMA-ScR) checklist.31 The protocol has been preregistered through Open Science Framework (https://osf.io/ymxz7). Covidence (Veritas Health Innovation, Melbourne, Australia), a software system for managing systematic reviews, and Microsoft Excel will be used by both reviewers to support screening, extracting and monitoring of the research process.

Stage 1: identifying the research question

Before developing the research questions, the authors have operationally defined the following concepts:

NPA: This is the matching of a nurse with a specific number of patients for a designated period.3

CDSS: This is operationalised to include any electronic system designed to assist nurses’ decision-making processes in assigning patients to nurses. As the assignment problem is based on optimisation,32 we included the development of optimisation modelling in this concept.

Optimisation model: It is defined as a health information data-driven model aimed at identifying the best solution to the NPA problem employing various methods, such as mathematical methods, artificial intelligence, machine learning and stochastic programming.

As the scoping review summarises the breadth of evidence on the aforementioned research topic, the research question should be broad; therefore, we started with the question, ‘What is known about the application of the CDSSs on the NPA process?’. An iterative approach was developed for the research questions through a preliminary literature review on the topic and consultation with the research team. Particularly, this scoping review will answer the following questions:

  1. What kinds of CDSSs have been used to support clinical nurses’ decision-making during the NPA process in hospital inpatient settings?

  2. What methods are used in optimising NPA in CDSSs?

  3. What are the primary factors to be considered in developing CDSSs for the NPA process?

  4. How was the CDSS for the NPA process validated, and how well did the system improve the NPA process on the nurse and patient sides?

Stage 2: identifying the relevant studies

A comprehensive literature search will be conducted in six international bibliographic databases (ie, MEDLINE via PubMed, EMBASE via Ovid, CINAHL via EBSCOhost, IEEE Xplore, Scopus, ProQuest Dissertations and Theses Global). The search terms are tailored to each database. The research team planned the search strategy in consultation with a professional librarian to identify a comprehensive list of relevant literature specific to our research. The search strategies for databases will be presented in online supplemental file 1. A time filter was applied, specifically set to the year 2000, as early research on computerised programming NPA processes emerged around 2002.33 The search is scheduled to be conducted until the end of February 2024.

Supplemental material

Stage 3: selecting studies

Covidence software for systematic reviews will be used for selection. The search results from each base will be imported into Covidence, and duplicated literature will be removed automatically. Two reviewers (HK and DL) will independently review titles and screen abstracts with prespecified inclusion criteria. Then, the same researchers will review the full texts of the included studies using the same inclusion and exclusion criteria and will decide whether to include the studies in the analysis. In this process, any discrepancies that arise will be resolved through discussion between the reviewers or, if necessary, by involving an additional reviewer to provide further insights and consensus. We have planned to complete the research selection process by April 2024.

Eligibility criteria

The eligibility criteria are outlined in table 1. This review will include studies that developed a CDSS for NPA and reported its considerations and outcomes. This involves examining considerations, implementation processes and the overall impact. Studies with insufficient data on reporting CDSSs and those targeting users who are not nurses will be excluded. Further, articles that do not present electronic CDSSs, including those centred on patient acuity scale development or documented NPA guidelines development, will also be excluded. Regarding the context, only studies conducted in hospital inpatient settings will be included. Therefore, research that was conducted in communities or schools and outpatient settings will be excluded. We will consider original research regardless of its study design and include articles written in English with full-text availability. We will exclude literature reviews, study protocols, editorials and studies with inaccessible full texts.

Table 1

Inclusion and exclusion criteria for study selection

Stage 4: extracting the data

The results of the studies will be charted using Microsoft Excel. Key information will be extracted and synthesised within a charting table that has been created and tested by the research team. The data that will be extracted include general information about the relevant articles (ie, year, authors, publication type, study design and geographical location), characteristics of the population and the number of data, characteristics of CDSSs (ie, mode of delivery, development details, considering factors, validation methods, outcomes and limitations) and settings (ie, hospital level of care, number of beds and type of unit).

We will initiate a pilot test employing the charting table. Two researchers will independently extract data from 20% of the included studies and subsequently compare their respective extractions. If necessary, adjustments to the charting table will be made during the data extraction phase, and we will provide a detailed account of any modifications made. The data extraction from the selected studies will be conducted until July 2024.

  • General information: Year of publication, author(s), publication type(s), study design and geographical location

  • Population: Characteristics of populations and sample (or data) size

  • Concept:

    • CDSSs description

    • Mode of delivery (eg, integrated into elctronic medical records (EMRs), web-based and others)

    • Methods for developing CDSSs (eg, AI-based methods, machine learning methods, integer linear programming and others)

    • Consideration(s): Description of criteria and restraints of CDSSs

    • Validation method(s)

    • Outcome(s) from CDSSs utilisation

    • Limitation(s)

  • Context:

    • Hospital level of care

    • Number of beds

    • Type of unit (eg, intensive care unit, medical-surgical unit, oncology unit and others)

Stage 5: collating, summarising and reporting the outcomes

We will present a table addressing a summary of the findings of the eligible research and describe the findings as a narrative report. For quantifiable data, descriptive statistics will be reported and qualitative data will be synthesised by narrative format. The evidence will be reported based on the PRISMA-ScR. This stage will be conducted in December 2024.

Stage 6: performing a consultation exercise

Arksey and O’Malley delineated consultation as an optional stage in the scoping review methodology.30 However, in our review, we will actively integrate consultations with stakeholders, encompassing charge nurses, nurse professors and computer scientists. After analysis of the key findings in the scoping review, we plan to initiate a consultation stage with stakeholders. This deliberate step is designed to achieve a thorough understanding of the extent to which our preliminary findings resonate with the distinctive experiences of the stakeholders involved.

Quality assessment of included studies

The evaluation of quality and the assessment of bias risk will not be conducted as it is not part of the scoping review methodology.34

Time schedule

This study will be completed by December 2024. Figure 1 details the research plan.

Figure 1
Figure 1

Scoping review research schedule.

  • November to December 2024: identifying the research question

  • January to February 2024: identifying the relevant studies

  • March to April 2024: study selection completion

  • May to July 2024: extracting the data

  • August to December 2024: collating, summarising and reporting of outcomes. If needed, a refinement process will be conducted based on the insights gleaned from the consultations with stakeholders.

  • October 2024: consultation exercise with stakeholders

Patient and public involvement

No patients or members of the public were involved in this study.

This post was originally published on https://bmjopen.bmj.com