Interventions to develop clinical judgment among nurses: A systematic review with narrative synthesis

Abstract

Aim

To explore interventions developed to improve nurses’ clinical judgment.

Background

Clinical judgment can assist nurses in assessing clinical situations, identifying and preventing problems, and making effective decisions about patient care. Studies on developing clinical judgment among nurses thus far been limited and heterogeneous.

Design

A systematic review with narrative synthesis.

Methods

Eight databases (PubMed, CINAHL, PsycInfo, Scopus, Web of Science, Cochrane Library and ProQuest Dissertations and Theses databases) were systematically searched for studies published until May 2024. A total of 13 studies satisfied the inclusion criteria. Joanna Briggs Institute’s critical appraisal tools were used to assess the quality of the selected studies, whereas the Mixed Methods Appraisal Tool was used for mixed methods studies.

Results

Overall, 13 studies were included for analysis. Simulation was the most used type of intervention, whereas Tanner clinical judgment model was the most used framework, and Lasater Clinical Judgment Rubric was the most used tool for exploring the development of nurses’ clinical judgment skills. Among the 13 interventions analyzed, 11 were found to be effective.

Conclusions

Simulation teaching strategies using Tanner’s clinical judgment model and Lasater Clinical Judgment Rubric satisfactorily develop clinical judgment among nurses. The findings of this systematic review underscore the dearth of nursing research exploring the efficacy of interventions designed to enhance clinical judgment among general registered nurses.

Reporting Method

The systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA).

Patient or Public Contribution

No patient or public contribution.

1 Introduction

Nurses represent the largest group of professionals involved in patient care. Given the nature of their work, which involves continuous care, they interact most closely with patients ( ^{American Nurses Association, 2021; Lange, 2000; Planas-Campmany & Icart-Isern, 2014; World Health Organization, 2017}). Evidence has shown that excellent nursing care directly impacts the clinical progression of the patient during their care process ( ^{International Council of Nurses, 2020; Sebastián-Viana and Gil-Hernández, 2007}).

One of the essential elements that conditions the quality of care is the nurse’s ability to develop sound clinical judgment across all issues involved in the delivery of patient care. As such, clinical judgment is a fundamental process of nursing care that requires clinical knowledge and skills to make informed and effective decisions in patient care. Considering that clinical judgment is based on nurses’ theoretical and practical knowledge, it can assist in assessing clinical situations, identifying and preventing problems, and making effective decisions about patient care. This approach enables nurses to provide quality nursing care and prioritize patient care over all other possibilities ( ^{Manetti, 2018; Tanner, 1994}).

1.1 Background

The concept of clinical judgment has been extensively studied. The most widely utilized definition for clinical judgment is that provided by Christine Tanner ( ^{Tanner, 2006}): “interpretation or conclusion about a patient’s health needs, concerns or problems and/or the decision to take action (or not) using or modifying standard approaches or improvising new ones as deemed appropriate for the patient’s response”. This author also designed a model to describe the process undertaken by nurses in developing clinical judgment, which consists of four phases, namely noticing, interpreting, responding, and reflecting.

Unfortunately, the term clinical judgment can generate confusion or uncertainty given that it is closely related, even confused, with other associated terms, such as decision-making, clinical reasoning, critical thinking, or problem solving. To address this concern, three concept analyses have been published, which, based on Tanner’s definition and model, have proposed a method for differentiating between these related terms. On the one hand, ^{Van Graan et al. (2016)} conducted a concept analysis in the context of a South African setting, subsequently defining clinical judgment as “a complex cognitive ability to assess patient needs, adaptation of current treatment protocols as well as new treatment strategies, prevention of side effects by being proactive rather than reactive within the clinical nursing environment”. Later, ^{Manetti (2018)} defined clinical judgment as a “cognitive process in which the nurse forms a holistic assessment of a patient’s situation. Critical thinking, clinical reasoning, practical wisdom and intuition are used in the decision-making process that follows”. Finally, ^{Connor et al. (2022)} published another concept analysis that defined clinical judgment as “a reflective and reasoning process that draws on all available data, is informed by a broad knowledge base, and results in the formation of a clinical conclusion”.

Christine Lasater ( ^{Lasater, 2007}) created an assessment rubric to evaluate clinical judgment. The four phases of Tanner’s clinical judgment model, namely noticing, interpreting, responding, and reflecting, served as the foundation for the development of the Lasater Clinical Judgment Rubric (LCJR), which contains four phases with 11 dimensions. Effective noticing involves focused observation, recognizing deviations from expected patterns, and information seeking. The second phase, or effective interpreting, involves prioritizing data and making sense of data. The third phase, or effective responding, involves calm, confident manner, and clear communication with well-planned interventions/flexibility and being skillful. Finally, the fourth phase or effective reflecting addresses evaluation/self-analysis and commitment to improvement.

Proper training that enhances the development of clinical judgment among nurses is fundamental to nursing development and practice. Evidence suggests that educational strategies designed to enhance clinical judgment can influence what a nurse brings to a patient's situation ( ^{Cappelletti et al., 2014}). Multiple interventions especially aimed at addressing such training among student nurses have been published in the literature, some of which have proposed clinical case simulation ( ^{Hallin et al., 2016; Kim, 2014; Lindsey and Jenkins, 2013}), concept mapping ( ^{Gerdeman et al., 2013}), and narrative and reflective journal writing as effective tools. However, interventions developed for nurses have been marginal and less localized.

Preliminary literature searches on studies concerning nurses have shown that general programs to improve clinical judgment ( ^{Marshall et al., 2001}), teaching-learning strategies aimed at advanced practice nurses ( ^{Benner et al., 1996}), and programs aimed at specific populations [i.e., acute nurses ( ^{Barron-Kagan, 2016}) and surgical nurses ( ^{Duff et al., 2014}) have been published. The most widely cited type of intervention in the literature has been simulation programs ( ^{Cantrell et al., 2022; D’Cunha et al., 2021; Faulkner et al., 2023; Padilha et al., 2019; Reece-Jones and Maguire, 2000; Sportsman et al., 2009; Thompson et al., 2012}), especially those aimed at addressing specific clinical situations or nursing specialties, similar to those used in student nurses. Moreover, several isolated studies have used reflective narratives ( ^{Graham-Hannah, 2016}), computer programs ( ^{Barken et al., 2017}), and serious games ( ^{Blanié et al., 2020}) or escape rooms as an educational method ( ^{Dacanay et al., 2021}). However, such studies tend to focus on specific unit areas and do not address the various circumstances experienced in the general field of daily clinical nursing.

Considering this variety, a systematic review of the literature was conducted to identify interventions developed so far to improve clinical judgment among nurses.

2 Methods
2.1 Objective

The current review aimed to identify interventions developed so far to improve clinical judgment among nurses.

2.2 Design

A systematic review with narrative synthesis.

2.3 Search methods/study screening and selection

The literature search was conducted for studies published in PubMed, CINAHL, PsycInfo, Scopus, Web of Science, Cochrane Library and ProQuest Dissertations, and Theses databases until May 2024. Each search was conducted using the MESH/Thesaurus/Headings terms from each database. The following terms were used in the searches: Nurses (Mesh + Ti/Abs) OR Nursing Staff (Mesh + Ti/Abs) AND Clinical reasoning (Mesh + Ti/Abs) OR Clinical judgment (Ti/Abs) OR Clinical judgment (Ti/Abs). No limits were placed on the searches.

Considering the presence of many articles on interventions that did not specifically mentioned the term “intervention,” the search strategy did not include this word. The articles were then filtered manually.

The inclusion criteria were as follows: (1) studies exploring interventions aimed at improving clinical judgment in any nursing setting; (2) studies aiming to improve clinical judgment among nurses; (3) studies with a population of registered nurses; and (4) studies published in English and Spanish. The exclusion criteria were as follows: (1) studies focused on nursing students and (2) those in which we had no access to the full-text.

Article selection was conducted using the COVIDENCE platform. All articles were independently peer reviewed. During the title/abstract review phase, all articles were reviewed in two parts: first by the lead author (M. M.) and second by three other authors, dividing the articles to be reviewed between them (A. Ch., V. S., and M. Ch.). Discrepancies between both sides were resolved by a fifth author (C. O.). Similarly, during the full-text review phase, all included articles were reviewed in two parts: first by the lead author (M. M.) and second by three other authors, dividing the articles to be reviewed between them (C. O., V. S., and M. Ch.), with discrepancies being resolved by a fifth author (A. Ch.).

2.4 Search outcomes

A total of 5841 articles were initially obtained, of which 3049 were discarded due to repetition. In the first review of the title/abstract, 2586 articles were eliminated, leaving 206 to be filtered by full-text review. Finally, after the full-text review, 13 articles were included for analysis. ^{Fig. 1} shows the article selection process (PRISMA).

2.5 Data extraction

The first author (M.M.) extracted all data from the 13 studies included for analysis and contacted the second and last authors (C.O. and A.Ch.) to discuss and clarify any doubts about the articles. The following data were extracted from each study: author name, publication year, country, aim, design/methods, sample, intervention framework, evaluation tools, intervention, and results.

2.6 Quality appraisal

The selected studies were evaluated according to the Joana Briggs Institute (JBI) critical appraisal tools. This method allows for different quality assessments for each type of scientific study. In the case of mixed-method studies, we used the Mixed Methods Appraisal Tool (MMAT).

2.7 Data synthesis

Owing to the heterogeneity of the studies, a meta-analysis could not be conducted. Hence, the results were presented in a narrative format following the Guidance on the Conduct of Narrative Synthesis in Systematic Reviews ( ^{Popay et al., 2006}). This guide proposes a four-step process involving (1) developing a theory for how, why, and for whom the intervention (i.e., clinical judgment in this case) works; (2) developing a preliminary synthesis of the findings obtained from the included studies; (3) exploring relationships in the data; (4) and assessing the robustness of the synthesis. The characteristics of the 13 selected studies are summarized in ^{Table 1}. Thereafter, the studies were grouped according to type of intervention, sample, assessment tool, and impact/effectiveness.

3 Results
3.1 Characteristics of the studies

The characteristics of the 13 studies included for analysis, including objective, design/methods, intervention, evaluation tool and results are summarized in ^{Table 1}. The included studies were conducted in five countries: the United States ( ^{Brown et al., 2022; Cantrell et al., 2021; Monagle et al., 2018; Schmehl, 2019; Wynn, 2011; Zehler and Severi, 2022; Shinnick and Cabrera-Mino, 2021; Letcher et al., 2017; Franks, 2020}), Malaysia ( ^{Foo et al., 2017}), Canada ( ^{Lavoie et al., 2013}), Korea ( ^{Kim et al., 2018}) and China ( ^{Luo et al., 2021}). Among the 13 studies, 11 were quantitative studies ( ^{Brown et al., 2022; Cantrell et al., 2021; Foo et al., 2017; Schmehl, 2019; Wynn, 2011; Zehler and Severi, 2022; Shinnick and Cabrera-Mino, 2021; Letcher et al., 2017; Kim et al., 2018; Luo et al., 2021; Franks, 2020}), one was a mix-method study ( ^{Monagle et al., 2018}), and one was a qualitative study ( ^{Lavoie et al., 2013}).

Although the quality of the included studies was generally high, direct comparisons were limited given the heterogeneous methodologies employed by each study. All studies had scored a minimum of 77.7 % in the JBI or MMAT tools, except for the pilot study of ^{Shinnick and Cabrera-Mino, (2021)}. However, considering the pilot nature of the study, a lower score (i.e., 50 %) was expected and acceptable.

3.2 Categories

Our results were grouped into the following four categories to synthesize the selected studies: type of intervention, sample, assessment tool, and impact of the intervention/effectiveness ( ^{Table 2}).

3.2.1 Type of intervention

All interventions identified in the 13 studies are defined in ^{Table 3}.

3.2.1.1 Simulation

Among the 13 selected studies, 10 included simulation as their principal type of intervention. Of these 10 studies, four involved simulation programs ( ^{Cantrell et al., 2021; Wynn, 2011; Shinnick & Cabrera-Mino, 2021; Franks, 2020}), whereas the other six used simulation as the basis of the intervention with some differences. One was a high-fidelity simulation (HFS) program ( ^{Schmehl, 2019}), one combined HFS with reflective debriefing ( ^{Lavoie et al., 2013}), one was a virtual simulation program (NovEx) ( ^{Brown et al., 2022}), one was a simulation-based learning ( ^{Letcher et al., 2017}), and one was a simulation-based and peer-learning study ( ^{Kim et al., 2018}). The last one of the studies was a comparison between HFS, virtual simulation, and a case study ( ^{Luo et al., 2021}).

Among the four studies on simulation programs, one ( ^{Cantrell et al., 2021}) used a simulation program with four steps: (1) prebriefing session, (2) implementation of a scenario, (3) debriefing session with a video recording of their performance, and (4) self-rating by participants. ^{Wynn (2011)} used a human patient simulator of a patient with diabetes in which nurses had to diagnose and administer treatment according to different scenarios. ^{Shinnick and Cabrera-Mino (2021)} created a 12-min simulation of a patient (SimMan 3 G™, Laerdal Medical) in a hospital setting with experts observing the expected reactions of participants. ^{Franks (2020)} intervention consisted in a 4-h class including a pretest, didactics, simulation, a posttest, and debriefing.

HFS were used in two studies. The first study by ^{Schmehl (2019)} used a structured framework designed to evaluate clinical judgment, as well as a manikin to provide realistic patient responses, symptoms, and vital signs requiring participant interaction. In the other study, ^{Lavoie et al. (2013)} combined HFS with reflective debriefing wherein the participants to recognized, analyzed, and introduced changes to the patient’s situation and then attended a 45-min debriefing in the form of a group discussion.

The study by ^{Brown et al. (2022)} showed that a virtual simulation program (NovEx) helped replicate real clinical situations in different pathophysiological modules that challenged participants to problem-solve 18 clinical conditions and prioritize evidence-base nursing practices to drive actions and interventions.

^{Letcher et al. (2017)} used simulation-based learning that combined three simulation-based sessions with a high-fidelity manikin focused on distinct physiologic and structured debriefing.

^{Kim et al. (2018)} combined simulation-based and peer-learning. This particular study consisted in two scenarios using high-fidelity simulators and standardized patients with problems and a debriefing session that facilitated discussion of the nurses’ feelings and experience with handovers. Thereafter, they were asked to listen to their peers’ and provide positive and negative feedback about each other’s performance.

Finally, this review included a comparison study of three type of interventions, namely high-fidelity simulation, virtual simulation, and a case study ( ^{Luo et al., 2021}). The three groups used the same four scenarios with an hour per scenario but with a different process than before. The HFS group participated in a case discussion, simulation with a high-fidelity manikin and actual medical equipment, and debriefing; the virtual simulation group participated in a virtual simulation with a platform computer; and the case study group received lectures and participated in a case study of the scenarios.

3.2.1.2 Other type of interventions

^{Foo et al. (2017)} used cased-based learning (CBL), which consisted of 2 days (16 h) of workshops followed by 2 weeks of practicum. The workshop topics focused on the nursing process and its application in a clinical setting (using an actual CBL scenario), data interpretation from an actual CBL scenario, problem solving, critical thinking, decision-making, communication skills, nursing ethics, reflections regarding nursing practice, and the LCJR. During the 2 weeks of practicum sessions, case discussions in groups using the nursing process and three complete case studies were conducted.

The study by ^{Zehler and Severi (2022)} used game based learning (GBL) composed of multiple stations closely reflecting the creation of a television show, with teams competing on timed activities to win points. Each activity was related to a pertinent assessment, intervention, or evaluation of a patient experiencing postpartum hemorrhage.

^{Monagle et al. (2018)} used reflection in their study, which consisted of a structured reflection intervention wherein the experimental groups attended three sessions, one in-service session early in the study and two structured reflection sessions.

3.2.2 Sample

All samples selected from the 13 studies included registered nurses, which was one of the inclusion criteria for study inclusion.

Five studies were conducted on nurses with little or no work experience. Accordingly, ^{Cantrell et al. (2021)} included an intervention in a group consisting of 43 registered nurses with a Bachelor of Science in nursing in a suburban health center system who had not been employed before and were enrolled in a residency program. Their ages ranged from 22 to 32 years (average, 23.88 years), and 81.4 % were female. ^{Kim et al. (2018)} used a sample of 55 new graduate nurses from a nursing college who completed their studies in 2015, had no clinical experience, and expected to begin work in a university hospital. The mean age of the sample was 24 years, and 90 % were female (50 women, 5 men). ^{Monagle et al., 2018} included a sample of 74 new graduate nurses aged between 21 and 48 years (average, 29 years) who had < 3 months of experience. Among them, 56 were women, 13 were men, and 5 identified as other gender. They were distributed throughout the medical/surgical (47), critical care (19), and other areas (4). ^{Foo et al. (2017)} included 80 registered nurses from two district hospitals with a mean age of 35.85 ± 8.95. More than half (56.4 %) of the registered nurses had secondary education level, whereas 53.8 % had more than 6 years of working experience. ^{Franks (2020)} included a small group of six new graduate nurses (4 women, 2 men) who worked in the emergency department of a hospital with ≤ 6 months of nursing experience.

Two studies included a sample of new and experienced nurses. Accordingly, ^{Shinnick & Cabrera-Mino (2021)} used a mixed group of 28 individuals, among whom 13 were senior prelicensure students recruited from one course (Medical-Surgical III) and 15 were expert nurses in the adult intensive care unit with at least 5 years of experience. ^{Letcher et al. (2017)} conducted a study over a 2-year period on 130 neonatal intensive care unit (NICU) nurses whose experience in critically ill neonate care ranged from 6 months to 30 years.

Three studies focused on nurses who were part of a training program. Accordingly, the doctoral thesis by ^{Schmehl (2019)} included a group of 33 registered nurses with diverse academic backgrounds and similar ages who were enrolled in a residency program at an acute care facility where they were employed. ^{Zehler and Severi (2022)} included 57 registered nurses working in a hospital with 24 obstetric beds and over 60 nurses as a part of a mandatory continuing education session for annual competency support. All of the included participants were identified as female, and 53 % were certified within their specialty of midwifery. The pilot study by ^{Lavoie et al. (2013)} included nurses who were about to complete their orientation program in an intensive care unit at a francophone teaching hospital.

^{Wynn (2011)} recruited 20 registered nurses who worked in acute inpatient psychiatry of a medical center. ^{Luo et al. (2021)} included 59 registered nurses (average age, 22.5 years; 73.9 % female) who graduated from a university college in 2018 and were working at a level A tertiary hospital after passing the Chinese licensure examination for nurses. Finally, the study by ^{Brown et al. (2022)} included a sample of 41 registered nurses, with no other information about the population having been provided.

3.2.3 Assessment tools

Five different types of assessment tools were identified, all of which are detailed in ^{Table 4}.

3.2.3.1 LCJR

The most frequently used assessment tool was the LCJR ( ^{Lasater, 2007}), which is based on Tanner’s model ( ^{Tanner, 2006}), and was used in nine of the 13 studies. Five studies, namely ^{Cantrell et al. (2021)}, ^{Wynn (2011)}, ^{Kim et al. (2018)}, ^{Luo et al. (2021)} and ^{Schmehl (2019)}, used the original LCJR, as well as the Test of Clinical Judgment Ability. Four studies incorporated LCJR with some modifications to adapt the tool to their study. Accordingly, Foo ML et al. (2017) used a modified LCJR adapted from ^{Lasater (2007)} who agreed to the use of the questionnaire with modification provided that the original tool was acknowledged. Effective responding in the original rubric consists of four criteria, of which three required direct observation of each participant and were deleted, leaving one item. ^{Monagle et al., 2018} used the LCJR instead of a measurement scale as the structured framework to reflect on and report anecdotal reflections from practice. ^{Shinnick & Cabrera-Mino (2021)} used a modified LCJR in which four domains (information seeking, clear communication, evaluation/self-analysis, and commitment to improvement) were eliminated for being impertinent to the content of the simulation, changing the overall possible maximum score to 28 points. Finally, ^{Letcher et al. (2017)} utilized the LCJR adapted with author guidance to include an explanatory section for participant selection and comment on one self-rated dimension.

3.2.3.2 Other tools

The other four studies used different tools. In particular, ^{Zehler and Severi (2022)} used the clinical judgment model applied to postpartum hemorrhage (recognition, analysis, action, and evaluation). Meanwhile, ^{Brown et al. (2022)} designed their own tool for their study, which consisted of assessing virtual clinical performance, medication errors, sentinel events, and failure to rescue. ^{Lavoie et al. (2013)} used the adapted reflective debriefing (inspired by ^{Nielsen et al., 2007}). Finally, ^{Franks (2020)} utilized the Creighton Competency Evaluation Instrument (C-CEI).

3.2.4 Impact of the interventions/effectiveness
3.2.4.1 Effective interventions

The results from ^{Wynn’s (2011)} study suggested that the intervention was effective in increasing the psychiatric nurses’ clinical judgment. The difference between pre/posttest mean scores was statistically significant. Medical emergencies involving patients with diabetes decreased from 55 % to 20 %, suggesting that the education provided to nursing staff might have improved the quality of care provided to veteran who suffered from medical conditions associated with diabetes.

^{Shinnick & Cabrera-Mino (2021)} compared two groups, namely novice and expert nurses, and found that years of nursing experience was the only independent predictor of clinical judgment on the LCJR based on stepwise linear regression. Significant differences in age and years as a nurse were identified between the groups, but no significant difference in the number of prior simulation experiences was noticed between novices and experts.

^{Franks (2020)} indicated that the average learning gained from the training was 28.3 % in the C-CEI tool. Among the six participants, four had an increase of 34,25 % in their post-test, with participant number 3 having a higher score in the pretest, and participant number 2 having the same pre-and posttest scores.

^{Schmehl’s (2019)} HFS program established a relationship between project interventions and clinical practice. Correlational analysis of the posttest scores demonstrated a significant statistical effect of the intervention.

Lavoie et al.’s (2013) qualitative results showed positive perceptions regarding learning and satisfaction. In particular, participants reported that reflection positively influenced their prioritization and organization of care and enhanced their nursing assessment capacities and global clinical judgment. Moreover, debriefing positively influenced their understanding of how they reached decisions. The observers noted that the intervention can serve as an excellent integrative care exercise.

^{Brown et al. (2022)} demonstrated a 373 % improvement in post assessment performance during the simulation, a 25 % reduction in medication errors, and an 85 % decreased in medication errors causing adverse events. Sentinel events decreased by 85 %, whereas failure to rescue decreased by 51 %.

^{Letcher et al. (2017)} showed improvement in clinical judgment over time (year 2) for both self and rater ratings using the LCJR. They concluded that simulation-based learning can effectively improve clinical judgment for NICU nurses.

^{Kim et al. (2018)} revealed that the simulation-based and peer-learning program improved clinical judgment. Although the immediate effects were not significant in both groups, the latent effects scores showed statistically significant improvements. The average score for clinical judgment was significantly higher in the simulation-based training group than in the peer-learning training group. In the simulation-based training group, the difference between latent and immediate effects on clinical judgment was statistically significant (p = .001). In the peer-learning training group, the differences between latent and immediate effects on clinical judgment were also statistically significant (p = .009).

The comparison study by ^{Luo et al. (2021)}, which analyzed the impact of three types of learning modalities, showed the highest level of clinical judgment in the virtual simulation group, followed by the HFS group and CS group.

Foo et al.’s (2017) study on CBL demonstrated significant differences in mean scores for clinical judgment skills between the experimental and control groups after the intervention. Hence, the authors concluded that CBL was effective in improving clinical judgment skills.

The study by ^{Zehler and Severi (2022)} found that GBL activities successfully improved clinical judgment ability. In particular, registered nurses’ overall total scores for performance improved significantly (p > 0.5) together with their ability to recognize, analyze, manage, and evaluate postpartum hemorrhage scenarios. A significantly change in the mean total pretest and posttest scores was observed.

3.2.4.2 Noneffective interventions

^{Cantrell et al. (2021)} showed that simulation program had a considerable effect on the intervention group’s clinical judgment, although no significant differences were found.

^{Monagle et al., 2018} revealed that the reflection intervention showed no significant effect on clinical judgment. However, the authors concluded that the use of reflection exercises had a positive impact on new graduate nurses.

4 Discussion

To the best of our knowledge, this has been the first systematic review that addressed the types, samples, tools, and effectiveness of the interventions developed so far to improve clinical judgment among nurses. Although Gonzalez & Nielsen ( ^{Gonzalez and Nielsen, 2024}) published an integrative review addressing interventions that develop clinical judgment, their population comprised student nurses.

The studies included in the current systematic review were conducted in five different countries belonging to two different continents (America and Asia). Remarkably, none of the studies were conducted in Europe and Oceania considering the amount of nursing research currently published in such regions and the authoritative scientific journals that exist there.

The current review identified four different types of intervention teaching strategies that promoted clinical judgment: simulation ( ^{Brown et al., 2022; Cantrell et al., 2021; Schmehl, 2019; Wynn, 2011; Shinnick & Cabrera-Mino, 2021; Lavoie et al., 2013}; ^{Letcher et al., 2017}; ^{Kim et al., 2018}; ^{Luo et al., 2021}; ^{Franks, 2020}), case-based learning ( ^{Foo et al., 2017}), GBL ( ^{Zehler and Severi, 2022}), and reflection ( ^{Monagle et al., 2018}).

The most prominent among the se identified strategies are simulation interventions, which had been used in 10 studies. Indeed, several studies had demonstrated that simulation was an effective teaching strategy ( ^{Brown et al., 2022; Schmehl, 2019; Wynn, 2011; Shinnick and Cabrera-Mino, 2021; Lavoie et al., 2013}; ^{Letcher et al., 2017}; ^{Kim et al., 2018}; ^{Luo et al., 2021}; ^{Franks, 2020}). Although simulation was used alone in many cases ( ^{Cantrell et al., 2021; Wynn, 2011; Shinnick & Cabrera-Mino, 2021; Franks, 2020}), some studies did combine it with other types of teaching methods or had made some modifications, such as HFS ( ^{Lavoie et al., 2013; Luo et al., 2021; Schmehl, 2019}), virtual simulation ( ^{Brown et al., 2022; Luo et al., 2021}), and simulation-based learning ( ^{Kim et al., 2018; Letcher et al., 2017}). This method has been so widely used for teaching due to its recognized benefits ( ^{Gaba, 2004}). In particular, simulation allows nurses to gain experience with clinical scenarios without risking patient wellbeing, promotes teamwork, facilitates feedback or debriefing, and provides a variety of contexts or scenarios that can be adjusted to the teaching objective ( ^{Koukourikos et al., 2021}). Simulation has been demonstrated to be a potent teaching and learning method for developing clinical judgment ( ^{Kaddoura et al., 2016}; ^{Yang, 2019}). Moreover, one study showed that this type of teaching strategy increases nurses’ confidence in making clinical judgments ( ^{McCaughey and Traynor, 2010}).

Surprisingly, our review did not find interventions that included innovative educational strategies that are currently in vogue in nursing education. For example, ^{Lewis and Bryan (2021)} included simulations, case studies, flipping the classroom, and debates to enhance the adult learner experience. Gamification has also been a booming effective learning strategy in recent years ( ^{Aktaş et al., 2024; Gokalp et al., 2024; Rushdan et al., 2024}), and artificial intelligence is becoming increasingly determinant in nursing education ( ^{Akutay et al., 2024; Lifshits and Rosenberg, 2024}). The aforementioned innovative techniques have been identified as effective in applying andragogy techniques in nurses ( ^{Lewis and Bryan, 2021}).

All included studies involved registered nurses, which was one of the inclusion criteria for study selection. Some studies were centered on new graduates with little to no experience ( ^{Cantrell et al., 2021; Foo et al., 2017; Franks, 2020; Kim et al., 2018; Monagle et al., 2018}), whereas others combined novice nurses with experienced ones ( ^{Shinnick & Cabrera-Mino, 2021; Letcher et al., 2017}). The study by ^{Shinnick & Cabrera-Mino (2021)}, which included both new graduates and experienced nurses adjusted according to age, showed that years of nursing experience was the only predictor of clinical judgment. In this regard, Patricia Benner’s From Novice to Expert model ( ^{Benner, 1984}) remains enlightening as it classifies nurses into five levels of expertise. On one hand, novice nurses are beginners who rely on rules and guidelines to perform tasks, with the lack of experience limiting their understanding of clinical situations. However, expert nurses have a deeper understanding of their practice, can manage complex situations, and have intuitive skills that allow them to respond effectively. These levels proposed by Benner are consistent with the results obtained by ^{Shinnick & Cabrera-Mino (2021)}, which showed better results in experienced nurses after adjusting clinical judgment according to clinical experience. This idea suggests that the development of teaching strategies should differ depending on whether the sample comprises student or registered nurses and should consider whether such nurses are fresh graduates or experienced.

Our findings highlight the lack of nursing research on interventions that facilitate the development of clinical judgment in registered nurses working in general hospital units. Specific interventions with specialist nurses or nurses working in specialized areas were identified in areas such as midwifery ( ^{Zehler and Severi, 2022}), psychiatry ( ^{Wynn, 2011}), neonatal ( ^{Letcher et al., 2017}) and intensive care units ( ^{Shinnick & Cabrera-Mino, 2021}), but only two studies addressed general nurses ( ^{Schmehl, 2019; Zehler and Severi, 2022}). Given that the majority of the nurse population works in non-specialist positions, it would be necessary to develop more studies that include this population.

While our study has concentrated on nurses, it is important to note that the majority of studies aimed at fostering clinical judgment have been conducted so far with nursing students, integrating clinical judgment-based teaching strategies as part of nursing courses or orientation programs. A scoping review conducted by ^{Bussard et al. (2024)}, which addressed current practices for assessing clinical judgment in student nursing and fresh graduate nurses (first year of practice after graduation), found that the most prominent types of intervention was, again, simulation, which was used in 21 of the 52 studies. Other interventions used included educational programs, concept maps, and problem-based learning. However, in the aforementioned review by ^{Gonzalez and Nielsen (2024)}, which was also conducted on student nurses, none of the interventions were based on simulation. Instead, concept mapping, concept-based learning, and debriefing were the most utilized type of interventions among students in the mentioned review. This difference in interventions between both reviews draws attention and shows that multiple strategies can be used with student nurses.

The majority of studies identified in this systematic review employed Tanner’s definition and model for clinical judgment. Nevertheless, the concept of clinical judgment in the literature has been considered, in many cases, as synonymous to other terms that share similarities but cannot be used interchangeably, such as critical thinking, clinical decision-making, and problem solving. Given this obscurity in concepts, various authors have conducted a concept analysis of the term ( ^{Connor et al., 2022; Manetti, 2018; Van Graan et al., 2016}). The study published by ^{Connor et al. (2022)} provided an interesting definition of the concept of clinical judgment. A comparison of Connor et al.’s (2022) definition with that provided by Christine Tanner ( ^{Tanner, 2006}) 16 years earlier may potentially highlight some differences. First, although Connor et al. defined the concept as a process, Tanner defined it as an interpretation or conclusion, which has different implications considering that a process takes somewhat longer, whereas an interpretation or conclusion seems shorter or more specific. Secondly, Tanner’s definition appears to be more complete given its inclusion of patient’s health needs, concerns, or problems and methods to act (or not), which Connor et al. failed to specify. Finally, Connor et al.’s definition includes nurses’ knowledge, whereas Tanner refers to a “modification of the standard or improvisation of new ones.” Considering that these standards or improvisations are based on nurses’ prior knowledge, both authors take this into account.

In this study, Tanner’s Clinical Judgment Model ( ^{Tanner, 2006}) was the most used framework for the interventions ( ^{Brown et al., 2022; Cantrell et al., 2021; Foo et al., 2017; Monagle et al., 2018; Zehler and Severi, 2022; Shinnick & Cabrera-Mino, 2021; Lavoie et al., 2013}; ^{Letcher et al., 2017}; ^{Kim et al., 2018}; ^{Luo et al., 2021}; ^{Franks, 2020}), whereas the LCJR ( ^{Lasater, 2007}) was the most used tool for measuring improvements in clinical judgment ( ^{Cantrell et al., 2021; Foo et al., 2017; Monagle et al., 2018; Wynn, 2011; Shinnick & Cabrera-Mino, 2021; Letcher et al., 2017}; ^{Kim et al., 2018}; ^{Luo et al., 2021}). Interestingly, many cases used modified versions of the original LCJR considering that it did not for not fulfill all items, whereas other studies employed ad-hoc designed tools based on LCJR. The study by ^{Bussard et al. (2024)}, which analyzed the concept and tools related to clinical judgment in 52 studies on student and fresh graduate nurses, found that Tanner’s model and the LCJR were also the most prominently used concept and tool, respectively. A study published by Jesse et al. ( ^{Jessee et al., 2023}), which aimed to explore clinical judgment models and teaching strategies that promote clinical judgment among prelicensure nursing students, found that 65 % of the programs used Tanner’s model. These findings highlight the fact that this model had been specifically developed by a nurse for teaching nursing, which makes its use quite easy and clear. Moreover, this model consists of four clearly differentiated phases, allowing for a breakdown of clinical judgment and the separate assessment of these phases. Furthermore, an evaluation rubric based on Tanner’s model, namely the Lasater Clinical Judgment Rubric, which facilitates work and interpretation, had been elaborated. Other models have been used in the literature, such as ^{Dickison et al. (2019)} National Council of State Boards of Nursing-Clinical Judgment Measurement Model, that identifies six interrelated steps in the process: recognize cues, analyze cues, prioritize hypotheses, generate solutions, take action and evaluate outcomes. Given that this framework was developed for the valid measurement of clinical judgment and decision-making, it may be more suited for use as an assessment model rather than a teaching model.

In light of the impact of the interventions developed in the analyzed studies, it is notable that the simulation interventions utilized in the 13 studies demonstrated effectiveness. The similarities and differences in the structure of the developed intervention may provide insights about the use of simulation to promote clinical judgment. Debriefing seemed to be an essential element for the effectiveness of the intervention and was present in seven of the simulation interventions ( ^{Cantrell et al., 2021; Lavoie et al., 2013; Letcher et al., 2017; Monagle et al., 2018; Schmehl, 2019; Wynn, 2011; Franks, 2020}) wherein the actions carried out were discussed and reflected on to help nurses to improve their skills. However, some of them had prebriefing sessions ( ^{Cantrell et al., 2021; Monagle et al., 2018}) that provided information regarding the intervention and helped nurses prepare for the same. Regarding the completion of the questionnaires, some interventions used self-reported questionnaires ( ^{Cantrell et al., 2021; Franks, 2020; Letcher et al., 2017; Schmehl, 2019}), whereas others were complimented by the observers ( ^{Brown et al., 2022; Foo et al., 2017; Wynn, 2011; Zehler and Severi, 2022; Shinnick & Cabrera-Mino, 2021; Lavoie et al., 2013}; ^{Letcher et al., 2017}; ^{Kim et al., 2018}; ^{Luo et al., 2021}). Although the perceptions of the person receiving the intervention are important, the objectivity of the observer provides more reliability to the assessment.

Based on these results, interventions such as case-based learning ( ^{Foo et al., 2017}) and game based-learning ( ^{Zehler and Severi, 2022}) seem to be effective in improving clinical judgment. However, given that only one intervention for each type was included in the review, we found it difficult to extrapolate our results. Other types of interventions used in the studies included reflection ( ^{Monagle et al., 2018}), but did not demonstrate significant results. Ultimately, our results demonstrate that simulation interventions appear to be effective in developing clinical judgment ( ^{Brown et al., 2022; Foo et al., 2017; Schmehl, 2019; Wynn, 2011; Zehler and Severi, 2022; Shinnick & Cabrera-Mino, 2021; Lavoie et al., 2013}; ^{Letcher et al., 2017}; ^{Kim et al., 2018}; ^{Luo et al., 2021}; ^{Franks, 2020}).

5 Conclusions

The current systematic review offers a narrative synthesis of nursing interventions published with the objective of developing clinical judgment among nurses. In light of the paucity and heterogeneity of studies included in this review, it is evident that further research among general nurses is required. The results demonstrated that the identified interventions were effective in improving clinical judgment in the majority of cases. Simulation was the most prevalent technique, while Tanner's framework and LCJR rubric were the most utilized tools for evaluating the interventions. Future research should concentrate on the creation of programs centered on promoting the clinical judgment of general registered nurses. This should include the implementation of effective innovative interventions and the development of appropriate measurement tools to assess the efficacy of these interventions and to improve the quality of clinical nursing practice.

6 Limitations

The use of similar terms as synonyms complicated the identification of articles to be included in the systematic review. The heterogeneity of the results of this synthesis must be considered before generalizing our results. Moreover, we need to highlight the fact that different samples of nurses were included herein, from new graduates to experts. Given that the modified version of the LCJR was used in many cases, this variety in measurements may complicate the interpretation, analysis, and comparison of the results.

Funding

This study did not receive any specific grant from funding agencies.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

None.

Registration

This protocol has been published in the International Prospective Register of Systematic Reviews (PROSPERO) database (No. CRD42024411773).

Table 1

Characteristics of the studies.

First author, date	Study design	Population (N)	Intervention framework	Intervention	Evaluation	Results	Quality appraisal
Schmehl, 2019	RCT - Quantitative, pretest posttest, randomized	33 RN	QSEN competencies and associated KSAs. NLN/Jeffries simulation	High fidelity simulation	LCJR. Test of Clinical Judgment Ability	Correlational analysis of the LCJR post-test scores demonstrated a significant statistical effect.	JBI 13/13
Foo et al., 2017	Quasi-experimental pre and posttest	80 RN: 30 experimental, 41 control group	Tanner's model	Case-based learning (CBL)	Modified LCJR	Effective to increase the level of clinical judgment skills: statistically differences in mean scores of the clinical judgment skills between the experimental and control groups post intervention.	JBI 9/9
Cantrell et al., 2021	Quasi experimental, partially randomized	43 RN: 17 intervention, 26 control group	Tanner’s model	Simulation	LCJR	No statistically significant differences found.	JBI 9/9
Zehler and Severi, 2022	Replication study - Quasi-experimental	57 RN	Clinical Judgment Model	Game based learning	Clinical judgment model applied to PPH	Increase the clinical judgment ability, significantly (p > 0.5) improved in the overall total scores and in their ability to recognize, analyse, act on, and evaluate PPH scenarios; mean total test score changed significantly from the pre-test 11.44 (SD = 0.78) to the post-test 12.39 (SD = 0.14.).	JBI 7/9
Brown et al., 2022	Quasi experimental, pre/post	41 residents	Tanner’s model	Virtual simulation program (NovEx)	Virtual clinical performance, medication errors, sentinel events, and FTR	373 % (12,1 % to 57,2 %) of improvement in post-assessment in the simulation; a 25 % reduction in medication errors and medication errors causing adverse events decreased by 85 %, sentinel events decreased by 85 %, (from 29.9 % to 4.4 %), and Failure To Rescue (FTR) decreased by 51 % (from 62.8 % to 30.8 %).	JBI 6/6
Wynn, 2011	Quasi experimental pre/post	20 RN working in acute inpatient psychiatry	CEPT: Clinical Excellence Through Evidence Based Practice.	Simulation	LCJR	Effective in increasing the psychiatric nurses clinical judgment, difference between pre/post-test mean scores statistically significant; medical emergencies involving patients with diabetes decreased from 55 % to 20 %, suggesting that the education provided to nursing staff may have improved the quality of care to veterans diabetic medical conditions.	JBI 6/6
Monagle et al., 2018	Mixed-methods, pretest/posttest	N = 74 in two groups	Tanner's model	Reflection	LCJR	No significant differences between the groups but positive impact.	MMAT 5/5
Shinnick and Cabrera-Mino, 2021	Pilot study, prospective, two-group, comparative study	28 RN, 13 novice nurses, 15 expert nurses	Tanner’s model	Simulation	LCJR	Years of nursing experience is the only independent predictor of clinical judgment on the LCJR, established in the stepwise linear regression. Significant difference between groups for age (Novice, 25.38 ± 6.14; Expert, 38.80 ± 10.07; p < .01) and years as a nurse (Novice=0; Expert=11.75 ± 9.02; p < .01, range 5–32), but no significant difference for number of prior simulation experiences.	JBI 6/12
Lavoie et al., 2013	Pilot study, qualitative	N = 5 nurses	Tanner's model	High fidelity simulation + Reflective debriefing	Adapted Reflective Debriefing (inspired by Nielsen et al., 2007)	Participants: positive perception of learning and satisfaction; reported that reflection influenced positively their priorization and organization of care, enhanced their nursing assessment capacities and their global clinical judgment. Also debriefing influence positively in understand how they reached decisions. Observers: the intervention was an excellent integrative care exercise.	JBI 8/10
Letcher et al., 2017	Quasi-experimental pre-post	130 NICU RNs over a two-year period. (1: 32, 2: 98).	Tanner's model	Simulation-Based Learning	LCJR	Trended patterns improvement over time for clinical judgment (Year 2) for self and evaluator ratings using the LCJR, they conclude that simulation-based learning can be effective in improving clinical judgment for NICU nurses.	JBI 8/9
Kim et al., 2018	Quasi-experimental nonequivalent control group post-test	N = 55 new graduate nurses	Tanner's model	Simulation-based and peer-learning	LCJR	Immediate effects: no significantly in both groups. Latent effects: statistically significant improvements: the average score for in the simulation-based group was significantly higher than in the peer-learning group (33,46 ± 5,09 vs. 30,14 ± 6,08, p = .033); In both groups difference statistically significant between latent and immediate effects: the simulation-based training group 9.32 (p = .001) and peer-learning training group 4.00 (p = .009).	JBI 7/9
Luo et al., 2021	Quasi-experimental with post-intervention	N = 59 RN: 19 HFS, 20 VS, 20 CS	Tanner's model. NLN Jeffries simulation theory.	High Fidelity Simulation vs. Virtual Simulation vs. Case Study	LCJR	Virtual simulation group higher level of clinical judgment (LCJR of 31.94 ± 5.12) followed by the HFS group (27.94 ± 5.64) and CS group (25.79 ± 4.41).	JBI 7/9
Franks, 2020	Quasi-experimental	6 new graduate nurses - emergency department	Tanner's model	Simulation	The Creighton Competency Evaluation Instrument (C-CEI)	Average learning gained from the training was 28.3 % in the C-CEI tool; from six participants, four had an increase of 34,25, participant number 3 had a higher score in the pre-test, and participant number 2 had the same score pre and post-test.	JBI 8/9

Table 2

Results.

Fist author, date	Type of intervention	Sample (N)	Assessment tool	Impact on clinical judgment
Schmehl, 2019	HFS	33	LCJR	Yes
Foo et al., 2017	CBL	80	LCJR	Yes
Cantrell et al., 2021	Simulation	43	LCJR	No
Zehler & Severi, 2022	GBL	57	CJ model PPH	Yes
Brown et al., 2022	Virtual simulation	41	Own	Yes
Wynn, 2011	Simulation	20	LCJR	Yes
Monagle et al., 2018	Reflection	74	LCJR modified	No
Shinnick & Cabrera-Mino, 2021	Simulation	28	LCJR modified	Yes
Lavoie et al., 2013	HFS + Reflective debriefing	5	Adapted reflective debriefing	Yes
Letcher et al., 2017	Simulation-based learning	130	LCJR modified	Yes
Kim et al., 2018	Simulation-based + peer-learning	55	LCJR	Yes
Luo et al., 2021	HFS/VS/CS	59	LCJR	Yes
Franks, 2020	Simulation	6	C-CEI	Yes
	Simulation		LCJR	Improve Clinical Judgment
	Others		Others	Not improve Clinical Judgment

Table 3

Type of interventions - definitions.

Type of interventions: definitions
Simulation	Teaching technique used by healthcare professionals to replace real experiences with guided experiences that replicate situations of the real world in a fully interactive manner ( Gaba, 2004, Koukourikos et al., 2021).
High fidelity simulation	Type of simulation that uses a structured framework designed to evaluate clinical judgment. An electronic manikin is utilized to provide realistic patient responses, symptoms and vital signs, requiring participant interaction ( Hansen & Bratt, 2017).
Case-based learning (CBL)	Case-based learning (CBL) is an instructional method with the use of case studies; facilitating student’s learning and self-decision in a field ( Kaddoura, 2011).
Game-based learning (GBL)	“Activity presided over by precise rules that involve varying degrees of chance, in which, players compete through the use of knowledge or skill in attempts to reach specified goals” ( Peddle, 2011).
Reflection	“Reflective experience in which participants used prompts to individually and collectively reflect on patient care experiences to describe their clinical judgment” ( Monagle et al., 2018).

Table 4

Assessment tools - definitions.

Assessment tools: definitions
Lasater Clinical Judgment Rubric (LCJR)	Based on the 4 phases of Tanner`s (2006) Clinical Judgment Model: noticing, interpreting, responding, and reflecting. Each of these four phases were further described by dimensions: - Effective noticing involves: focused observation, recognizing deviations from expected patterns and information seeking. - Effective interpreting involves: priorizing data and making sense of data - Effective responding involves: calm, confident manner, clear communication, well-planned intervention/flexibility and being skillful. - Effective reflecting involves: evaluation/self-analysis and commitment to improvement.Eleven domains of assessment are scored as Exemplary (4 points), Accomplished (3 points), Developing (2 points), or Beginning (1 point) for a total possible score of 44 points ( Lasater, 2007).
Test of Clinical Judgment Ability	Used selected questions originated from the work of Lasater et al., fill in the blank, short answer and multiple choice and were constructed by their original author to align with the LCJR categories.
Clinical judgment model applied to PPH	Phases: recognition, analysis, action and evaluation. Consists in 13-question next generation nursing style pre-/post-test to evaluate RN knowledge of critical PPH content and ability to notice and act upon patient changes. Test questions reflected each aspect of the Clinical Judgment Model.
The Adapted Reflective Debriefing	Inspired by Nielsen et al., 2007. Guide for reflection assessment based on Tanner’s model. This guide takes into account Tanner’s model: “I notice, I interpret, I respond and reflection with the role as a nurse, the previous experiences, the emotions and the formal knowledge of the nurse”. ( Lavoie et al., 2013).
Creighton Competency Evaluation Instrument (C-CEI)	23-item evaluation tool organized into four categories: assessment, communication, clinical judgment, and patient safety modified from an existing instrument, the Creighton Simulation Evaluation Instrument, for use in the National Council of States Boards of Nursing National Simulation Study (NCSBN NSS).