Influence of six teaching methods on critical thinking ability of nursing students: a protocol for systematic review and network meta-analysis

Abstract

Introduction

To enhance nursing students’ critical thinking abilities, numerous educators have explored alternative teaching methods. While meta-analyses have confirmed that various approaches are effective in developing critical thinking, consensus regarding their comparative effectiveness remains elusive. Furthermore, few investigations have directly contrasted the outcomes across these methods, highlighting the necessity to undertake a comprehensive evaluation of their impact on nursing students’ critical thinking skills. Accordingly, this study aims to assess the effects of six teaching methods on nursing students’ critical thinking abilities.

Methods

A comprehensive literature search will be carried out up to May 2025 across various databases, such as PubMed, Embase, The Cochrane Library, Web of Science, OVID, CNKI, Wanfang Database and the China Biological Literature Database (CBM). The search strategy will specifically target randomised controlled trials meeting predefined inclusion criteria. Two independent reviewers will screen the selected studies and extract pertinent data. The methodological quality of the included studies will be assessed using the Cochrane risk of bias tool. A network meta-analysis will then be performed using Stata software, incorporating the following analytical components: heterogeneity, network evidence diagrams, publication bias plots, league tables, forest plots, subgroup analyses or meta-regression and sensitivity analyses. The Grading of Recommendations Assessment, Development and Evaluation (GRADE) system will be leveraged to appraise the overall quality of evidence related to critical thinking abilities across all compared interventions.

Ethics and dissemination

No formal research ethics approval is required. The results will be submitted to a peer-reviewed journal for publication.

PROSPERO registration number

CRD42024618735.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• The study selection, data extraction and quality assessment will be performed independently by two authors.

• To minimise potential bias, subgroup analyses will be conducted by categorising studies based on different scales, intervention timing, duration of intervention, curriculum type, country and educational level.

• This study will include only randomised controlled trials, which may limit the generalisability of findings.

• Variations in intervention timing and disciplinary focus may increase heterogeneity and introduce bias.

• Although data will be retrieved using both database searches and manual supplementary methods, some grey literature may still be missed.

Introduction

At the end of the last century, the American Association of Colleges of Nursing emphasised the importance of cultivating critical thinking, identifying it as an essential competency for all nursing students.¹ Since then, nursing experts have widely acknowledged the significance and urgency of fostering critical thinking abilities among nursing students.² With the continuous evolution of the medical field, nursing professionals are confronted with increasingly complex healthcare demands and expanded roles in patient care, which underscores the necessity for nurses to possess strong critical thinking abilities.³

Critical thinking refers to a comprehensive cognitive process directed towards problem-solving. It entails the impartial gathering, evaluation and analysis of relevant information, along with continuous questioning, reflection and assessment, ultimately leading to well-reasoned decisions through logical deduction and inductive reasoning.⁴ However, traditional approaches to nursing education have predominantly emphasised knowledge transmission, often neglecting the development and nurturing of critical thinking abilities among nursing students.⁵ As healthcare continues to evolve, the demand for core nursing competencies has increased, positioning critical thinking as a fundamental competency for graduating nursing undergraduates and a central objective of higher nursing education.⁶

In recent years, an increasing number of nursing educators have become aware of the shortcomings inherent in traditional teaching methods, prompting them to explore alternative pedagogical approaches aimed at enhancing nursing students’ critical thinking abilities. To date, several meta-analyses have demonstrated that innovative instructional strategies—such as problem-based learning,⁷ blended learning,⁸ concept mapping,⁹ high-fidelity simulation,¹⁰ team-based learning¹¹ and flipped learning¹²—are more effective than conventional teaching methods in fostering critical thinking skills among nursing students. However, traditional meta-analyses are limited in their ability to identify the most effective teaching method due to the diversity of available educational interventions. Network meta-analysis, which represents an advancement over traditional meta-analysis, allows for the simultaneous comparison of multiple interventions by extending beyond pairwise group analyses. This approach enables a comprehensive evaluation and ranking of all relevant interventions within the same body of evidence, thus supporting the identification of the most efficacious option.¹³ Given the current lack of consensus regarding the relative effectiveness of various teaching methods and the paucity of direct comparative studies, it is both necessary and valuable to evaluate the impact of different educational approaches on the development of critical thinking abilities in nursing education.

To the best of our knowledge, no published pedagogical mesh meta-analyses have specifically evaluated the critical thinking abilities of nursing students. Given the current demands in nursing education, this network meta-analysis aims to provide robust evidence concerning the effectiveness of six teaching methods in enhancing nursing students’ critical thinking abilities, thereby offering actionable insights for nursing educators to refine and improve teaching strategies.

Materials and methods

Protocol and registration

This protocol is registered in the PROSPERO International prospective register of systematic

reviews (registration number: CRD42024618735) and reported based on Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) statement.¹⁴ The study is scheduled to start on 1 May 2025 and is expected to be completed by 30 November 2025.

Inclusion and exclusion criteria

The inclusion criteria will be established as follows:

1. Participants: Nursing students currently enrolled in academic programmes.

2. Interventions: The intervention group will use educational strategies such as problem-based learning, team-based learning, flipped-class learning, blended learning, high-fidelity simulation teaching and concept mapping.

3. Comparisons: The control group will receive traditional teaching strategies, including lectures and demonstration teaching, or one of the six teaching methods that will be included in the intervention group.

4. Outcomes: Critical thinking ability, without any restrictions on the type of measurement scale.

5. Study design: Randomised controlled trials (RCTs).

Exclusion criteria will be as follows:

1. Studies without full texts, with incomplete details or from which data could not be extracted.

2. Duplicate publications.

3. Conference abstracts.

4. Studies in which the experimental group received a combination of two or more teaching methods.

5. Studies related to practicum nursing students.

Note: We decided to exclude those who were already engaged in internships. The rationale for this exclusion is the difference in learning environments between nursing interns and full-time on-campus nursing students. On-campus students receive instruction from academic faculty within structured classroom settings focused on specific courses, whereas interns typically develop critical thinking skills through clinical teaching rounds led by hospital-based instructors. This discrepancy may introduce heterogeneity into the study results.

Search strategy

Searches will be conducted in databases including PubMed, Embase, The Cochrane Library, Web of Science, OVID, ERIC (Education Resources Information Centre), CNKI, Wanfang Database and the China Biological Literature Database (CBM). A combination of subject headings and free-text terms will be used to develop the search strategy. In addition, references from included studies will be reviewed, and manual searches of printed materials and related citations will be carried out to expand the search scope and identify additional relevant literature. The search strategies for electronic databases are shown in online supplemental file 1. The detailed search strategy for Web of Science is as follows:

#1 TS=Problem-Based Learning OR Problem Based Learning OR Problem-Based Curriculum OR Problem-Based Curricula OR Problem Based Curricula OR Experiential Learning OR Active Learning OR PBL.

#2 TS=Team-based learning OR Team based learning OR TBL.

#3 TS=Flipped teach* OR Flipped class* OR Flipped learn* OR Inverted class* OR Inverted learn* OR Inverted teach* OR Upside down class* OR Upside down lesson OR Reverse learn* OR Reverse instruction OR Flipped instruction OR Inverted instruction.

#4 TS=Blended teaching OR Mixed teaching OR Blended learning OR Mixed learning.

#5 TS=High Fidelity Simulation Training OR High-fidelity simulation OR HFS.

#6 TS=Concept map.

#7 TS=Nurse* OR Nurse student OR Nurse education OR Nurse.

#8 TS=Critical thinking OR Critical reflection OR Clinical thinking OR Clinical application.

#9 #1 AND #7 AND #8.

#10 #2 AND #7 AND #8.

#11 #3 AND #7 AND #8.

#12 #4 AND #7 AND #8.

#13 #5 AND #7 AND #8.

#14 #6 AND #7 AND #8.

Literature selection

Two researchers (LYJ and XJW) will use EndNote V.X8 reference management software to screen titles and abstracts and remove duplicate records. The selection process will proceed through a stepwise review of titles, abstracts and full texts in accordance with the predefined inclusion criteria. Studies that meet these criteria will undergo a detailed evaluation, during which relevant data will be extracted. In instances of disagreement, a third reviewer (LJ) will be consulted to facilitate consensus. The specific process and the result of searching and screening are shown in figure 1.

Figure 1. PRISMA (Preferred Reporting Items for Systematic Review and Meta-Analysis) flow chart of study selection. CBM, China Biological Literature Database.

Data extraction

Data extraction will be performed independently by two researchers (ZCJ and WLX) using a predefined template. Afterwards, their findings will be cross-verified. Any discrepancies will be resolved through discussion with a third researcher (WHC), who will render the final decision. The extracted data will encompass basic information about the literature (including title, authors and year of publication); fundamental characteristics of the study subjects (such as gender, age, sample size and intervention measures); key elements for assessing the risk of bias; and data relevant to the outcome measures of interest.

Risk of bias assessment

Two researchers (ZCJ and WLX) will assess the risk of bias in the included randomised controlled trials (RCTs) using the standardised tool described in the Cochrane Handbook for Systematic Reviews of Interventions.¹⁵ The assessment will evaluate the following key domains: (1) random sequence generation; (2) allocation concealment; (3) blinding of participants and intervention providers; (4) blinding of outcome assessors; (5) completeness of outcome data; (6) selective outcome reporting; and (7) other sources of potential bias.

Statistical analysis

This study will conduct statistical analyses within a frequentist framework, using Stata V.16.0 software to compute the effect sizes along with their 95% CIs.¹⁶ Given that critical thinking ability is a continuous variable, the effect sizes will be calculated using either the mean difference or the standardised mean difference, depending on the measurement scales across studies.

Heterogeneity testing

For each outcome, prediction interval plots will be generated.¹⁷ If the prediction intervals cross the line of no effect, this suggests the presence of heterogeneity among the studies, and a random-effects model will be applied for the network meta-analysis. Conversely, if there is no significant heterogeneity, a fixed-effects model will be used.

Similarity testing

The experimental design will incorporate both clinical and methodological consistency criteria. To ensure clinical homogeneity, we will rigorously define the fundamental characteristics of all included studies, including but not limited to participant demographics, intervention protocols and primary outcome measures. Regarding methodological uniformity, the selection process will exclusively consider studies that meet or exceed a quality assessment threshold of Grade B.

Consistency testing

The inconsistency test will primarily evaluate the concordance between direct and indirect comparison results. Since no closed loops will be formed for the three outcome measures in this study, a design-by-treatment interaction model will be implemented.

Drawing network evidence diagrams

Network evidence diagrams will be used to depict the relationships among various intervention measures. The thickness of the interconnecting lines will represent the number of studies making direct comparisons between interventions, while the nodal dimensions will scale proportionally with the corresponding sample sizes. A connection between two nodes will serve as evidence for a direct comparison between the respective treatment measures; conversely, the absence of such a connection will indicate a lack of direct comparative evidence.

Drawing publication bias plots

Should funnel plots exhibit symmetry, this indicates minimal evidence of small-study effects or publication bias. Conversely, observed asymmetry suggests significant small-study effects or publication bias.¹⁸

Network meta-analysis

League tables along with forest plots will be employed to present the comparative effectiveness rankings between any two treatment methods. The surface under the cumulative ranking curve (SUCRA) value will serve to identify the potentially optimal intervention approach.¹⁹ SUCRA values range between 0 and 1, where a value of 1 designates the best intervention efficacy and 0 indicates the poorest.²⁰

Subgroup analysis or meta-regression

Subgroup analysis or meta-regression will be performed to explore potential sources of heterogeneity, including (1) different measurement scales (eg, the California Critical Thinking Disposition Inventory (CCTDI), the California Critical Thinking Skills Test (CCTST) and the Critical Thinking Scale (CTS)); (2) intervention timing; (3) duration of the intervention; (4) curriculum differences; (5) country context; and (6) the educational level of nursing students across the respective study settings.

Sensitivity analysis

To verify the robustness of the study conclusions, a sensitivity analysis will be conducted by systematically excluding individual studies in sequence to evaluate the impact of sample size on the synthesis findings. If the heterogeneity results remain statistically consistent before and after this sensitivity analysis, the outcomes will be considered methodologically robust.

Strength of evidence

On the basis of the network meta-analysis findings, the GRADE approach will be employed to evaluate the quality of evidence pertaining to critical thinking abilities. This will entail a systematic evaluation across five domains: risk of bias, inconsistency, imprecision, indirectness and publication bias. Subsequently, the quality of evidence will be categorised into four distinct levels: high, moderate, low and very low.²¹

Patient and public involvement

No patients or members of the public will be involved in the design of this systematic review protocol or network meta-analysis.