Whether case-based teaching combined with the flipped classroom is more valuable than traditional lecture-based teaching methods in clinical medical education: a systematic review and meta-analysis

Abstract

Background

An increasing body of evidence suggests that both flipped classroom and case-based teaching models outperform traditional teaching methods in clinical medical education. Both teaching methods share several advantages, including a student-centered approach, enhanced understanding and mastery of knowledge, the development of comprehensive skills, and an emphasis on interactive communication. Therefore, a teaching model that integrates both methods may achieve similar or even better educational outcomes. Although previous studies have demonstrated this, they are generally limited by small sample sizes and inconsistent evaluation criteria for teaching effectiveness. This study aims to provide a high-quality assessment of this teaching method through meta-analysis and systematic review, further confirming its value in the field of clinical medical education.

Methods

This study employed a meta-analysis to systematically and quantitatively compare the educational impact of a teaching model combining flipped classroom and case-based learning (Flipped Classroom Case Learning, FCCL) with traditional lecture-based learning (LBL) for clinical medical students. An extensive literature search was conducted to identify studies on the application of FCCL and LBL in clinical medical education, covering the period from inception to November 21, 2024. Databases searched included the Cochrane Library, Web of Science, Embase, and PubMed. A total of 12 studies were included, comprising 1,857 clinical medical students or residents. The quality of each included study was assessed using RevMan (Version 5.4). The primary outcome measures were students' theoretical scores and clinical analytical skills. The extracted data were subjected to meta-analysis using Stata 18.0 software.

Results

A total of 1,857 participants from 12 studies were included in the analysis. The meta-analysis revealed that the FCCL group demonstrated significantly superior theoretical scores (Cohen’s d = 0.60, 95% CI: 0.17 to 1.04, P = 0.01) and clinical analysis skills (Cohen’s d = 1.53, 95% CI: 0.86 to 2.19, P = 0.00) compared to the LBL group. The Cohen's d for clinical analytical skills was 1.53 (> 0.8), indicating a large effect size between the two groups, suggesting a significant difference between the FCCL and LBL groups in improving students' clinical analytical skills. In contrast, the Cohen's d for theoretical scores was 0.60(approximately 0.5), indicating a moderate effect size, suggesting a moderate difference between the FCCL and LBL groups in improving students' theoretical knowledge. Subgroup analyses indicated that Chinese students in the FCCL group exhibited better theoretical performance (SMD = 1.03, 95% CI: 0.30 to 1.77, P = 0.01; heterogeneity, P = 0.00) and clinical analysis skills (SMD = 2.54, 95% CI: 1.06 to 4.02, P = 0.00; heterogeneity, P = 0.00) compared to their LBL counterparts. Among students from Western countries, those in the FCCL group outperformed the LBL group in clinical analysis skills (SMD = 0.45, 95% CI: 0.26 to 0.63, P = 0.00; heterogeneity, P = 0.21). However, no significant difference in theoretical scores was observed between the two teaching methods in this subgroup (SMD = 0.10, 95% CI: -0.28 to 0.48, P = 0.60; heterogeneity, P = 0.00).

Conclusion

The present study demonstrates that FCCL is more effective than LBL in enhancing clinical medical students' theoretical scores and clinical analysis skills. These findings were generally consistent across different nationalities of the study populations, with the exception of students from Western countries, where no statistically significant difference was observed in theoretical scores between the two instructional methods. However, due to the inevitable methodological differences among the included studies, heterogeneity is difficult to eliminate, and the overall sample size is relatively small. Therefore, further randomized controlled trials with rigorous experimental designs are needed to confirm these conclusions.

Introduction

Clinical medicine is a discipline that demands a broad range of comprehensive skills. Effective clinicians are characterized by a strong theoretical foundation and advanced clinical analytical abilities. To enhance the theoretical learning and foster clinical reasoning skills among medical students, many institutions have implemented teaching reforms, introducing innovative instructional approaches such as case-based learning (CBL), flipped classroom (FC), and problem-based learning (PBL). These methods have been integrated across various stages of clinical education to assess their efficacy. A growing body of evidence supports the superior outcomes of these contemporary teaching methods compared to traditional approaches.

Active learning has been shown to enhance the effectiveness of both teaching and learning. Familiarizing students with course content prior to class and encouraging active participation in class discussions with peers can further facilitate active learning [1,2,3]. The flipped classroom (FC) model incorporates flexible, self-managed pre-class videos and in-class group activities, which serve to increase student engagement and interest [4]. FC is a versatile, multi-modal educational approach that demonstrates excellent compatibility with various learning styles. Similarly, case-based learning (CBL) is a guided inquiry method that enables students to link theoretical knowledge with real-world clinical scenarios, thus fostering the development of clinical reasoning skills [5]. In contrast, the LBL teaching model is relatively unidimensional, predominantly teacher-centered, with students passively receiving information. The curriculum places a significant emphasis on theoretical knowledge, offering limited opportunities for students to apply theory to practice. The teaching pace and content are relatively standardized, making it difficult to accommodate individual student differences. Furthermore, there is relatively little interaction between teachers and students, as well as among students themselves. The integration of FC and CBL into a combined methodology, termed Flipped Classroom Case Learning (FCCL), offers a promising approach to meet the complexities and demands of clinical medical education, potentially surpassing the efficacy of traditional lecture-based teaching (LBL). A large number of randomized controlled trials (RCTs) and meta-analyses on CBL and FC have demonstrated their advantages over LBL in various aspects of teaching outcomes, and the research in this area is well-established. However, these studies generally suffer from the limitation of analyzing these methods in isolation. In actual teaching, the situation is often more complex, with the sole use of CBL or FC being rare; instead, a combination of various teaching methods is more commonly employed. Therefore, conclusions drawn from studies that analyze these methods independently may have limited generalizability and fail to capture the synergistic effects between methods. In contrast, research on the combination of CBL and FC is better suited to meet the needs of teachers in diverse educational settings and aligns more closely with educational practice. As research on FCCL is relatively limited, we systematically reviewed and analyzed relevant experimental studies to assess its effectiveness, providing a theoretical basis for its widespread application in clinical medical education.

Methods

The meta-analysis was conducted in accordance with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines [6]. The study protocol is registered with PROSPERO (International Prospective Register of Systematic Reviews) under registration number CRD42024594162 (https://www.crd.york.ac.uk/prospero). As this meta-analysis utilized publicly available data from previously published studies, it did not require informed consent or ethical approval from the individuals included in the studies [7].

Search strategy

A comprehensive search was conducted across several electronic databases, including the Cochrane Library, Web of Science, Embase, and PubMed. The retrieval strategy involved searching using Medical Subject Headings (MeSH) and keywords such as “flipped classroom,” “case-based learning,” “clinical medicine,” and “case.” Our full Boolean search strings was:"case-based learning"OR"case"AND"flipped classroom"AND"clinical medicine."Additionally, a manual search of the references cited in the included studies, as well as relevant literature suggested by the databases, was performed to minimize the risk of data omission. The selection of literature was primarily carried out independently by two authors, with no communication between them during the process. After the screening was completed, the selections were cross-checked. In case of any disagreement regarding the inclusion or exclusion of a particular study, the issue was resolved through consultation with a third author. The most recent search was conducted on November 21, 2024. All duplicate records were removed using EndNote X9.3.3. No language restrictions were applied during the search process.

Selection criteria

Inclusion criteria: (1) types of participants: undergraduate or graduate student or resident in clinical medicine; (2) types of interventions: teaching methodology is flipped classroom case learning (FCCL); (3) types of controls: teaching methodology is lecture-based teaching (LBL); (4) observation results: theoretical scores and clinical analysis skills; (5) types of researches: randomized or quasi-randomized controlled trials (RCTs/Q-RCTs).

Exclusion criteria: (1) non-clinical medicine specialty.; (2) incomplete article data or no target data; (3) non-primary research (review, meta-analysis and so on), repeated studies, expert opinions and meeting summary.

Data extraction

Baseline information was extracted independently by both authors and the following information was collected: (a) Author, (b) Year, (c) Country, (d) Type of study, (e) Age (FCCL/LBL), (f) Specialty, (g) Number (FCCL/LBL), (h) Outcome (two objective outcome indicators: theoretical scores and clinical analysis skills). In cases of disagreement during data extraction, the issue was resolved through discussion with a third author to reach a consensus. The basic characteristics of the included studies are summarized in Table 1.

[IMAGE OMITTED: SEE PDF]

Quality evaluation of individual studies

Two authors independently evaluated the included studies for risk of bias and methodological quality. Any discrepancies in assessment were resolved through consultation with a third author. The Cochrane Collaboration’s risk of bias tool [8] was employed to assess the following domains: (1) selection bias (random sequence generation; allocation concealment), (2) performance bias (blinding of participants and personnel), (3) detection bias (blinding of outcome assessment), (4) attrition bias (incomplete outcome data), (5) reporting bias (selective reporting), and (6) other biases. Each study was categorized as having a"low","high", or"unclear"risk of bias for each criterion. When assessing the risk of bias, subjective judgment from the assessors is inevitable. To minimize this influence, in addition to at least two researchers independently evaluating the studies strictly according to the criteria and grading system of the Cochrane Collaboration’s Risk of Bias tool, each assessor was required to provide a detailed rationale for each judgment. This ensures traceability and allows for re-examination in case of disputes, thereby enhancing the transparency of the assessment process. In case of missing data or conflicting evidence, the original authors may be contacted to obtain or verify the information.

Statistical analysis

The extracted data were meta-analyzed and tested for heterogeneity using Stata 18.0 software. Subgroup analyses were conducted using standardized mean differences (SMD) and 95% confidence intervals (CIs), with P < 0.05 considered indicative of a statistically significant difference. The Der Simonian-Laird random-effects inverse variance model was applied to assess the effects. Combined effect sizes were reported as Cohen's d to account for potential small sample sizes [9]. In cases where studies reported multiple outcomes (e.g., immediate and delayed quiz scores), only the immediate posttest scores at the end of the course were used to mitigate the impact of factors such as forgetting. Heterogeneity across studies was assessed using the I2 statistic derived from Cochran’s Q test [10] and the tau-squared (T2) test. The I2 statistic was used to estimate the proportion of total variation across studies, with I2 values of 30%, 50%, and 75% indicating low, moderate, and high heterogeneity, respectively [11]. A T2 value greater than 1 was considered indicative of substantial heterogeneity between studies. If I2 < 50% and P > 0.10, indicating low inter-study heterogeneity, a fixed-effects model was employed for meta-analysis; otherwise, a random-effects model was used [12, 13]. In the presence of considerable heterogeneity, the random-effects model accounts for variability, adjusts weights, and relaxes assumptions, thereby offering more robust, conservative, and generalizable statistical inferences. It is the preferred approach for addressing heterogeneity in meta-analysis, particularly in research fields such as education and medicine. Consequently, when significant heterogeneity exists among the studies included in this research, we are more inclined to select the random-effects model due to its greater adaptability. In the presence of heterogeneity, sources of variation were explored through sensitivity and subgroup analyses. Funnel plots and Egger’s Z [14] were used to assess the presence of asymmetry, small-study effects, and potential publication bias [15]. The trim-and-fill method was applied to evaluate the impact of publication bias on the overall effect size [16].

Results

Studies selection and basic characteristics

A total of 4,397 documents were initially retrieved from multiple databases. After the removal of 3,780 duplicates, 528 additional documents were excluded based on the evaluation of titles and abstracts. The remaining 89 articles were further assessed by reviewing the full text according to predefined inclusion and exclusion criteria. Ultimately, 12 studies were included in the analysis [17,18,19,20,21,22,23,24,25,26,27,28]. The process of literature screening is illustrated in Fig. 1.

[IMAGE OMITTED: SEE PDF]

Methodological quality assessment and risk of bias of the included literature

Risk and bias were evaluated using the Cochrane Risk of Bias Tool. The risk assessment for the 12 included studies is presented in Fig. 2. The authors report the results of each quality assessment program, focusing on the proportion of cross-sectional studies. All studies included in the review provided complete data, which helps mitigate potential reporting bias. No instances of selective reporting or other forms of bias were identified. However, fewer than half of the studies provided detailed information on the methods used to generate random sequences, indicating a high risk of methodological bias. Overall, while most studies demonstrated a high risk of methodological quality, they were generally found to have a low risk of bias.

[IMAGE OMITTED: SEE PDF]

Determining which teaching method is better at improving theoretical scores of clinical medical students

A total of 1,473 students across eight studies were evaluated for their theoretical scores (FCCL group: 726, LBL group: 747). Significant statistical heterogeneity was observed among the studies (I2 = 93.29%, P = 0.00). Given the heterogeneity, a random effects model was applied for the meta-analysis. The pooled effect size revealed a statistically significant difference in theoretical scores between the two groups (Cohen’s d = 0.60, 95% CI: 0.17 to 1.04, P = 0.01) (Fig. 3a). The Cohen's d for theoretical scores was 0.60(approximately 0.5), indicating a moderate effect size, suggesting a moderate difference between the FCCL and LBL groups in improving students'theoretical knowledge. Students in the FCCL group demonstrated superior theoretical knowledge compared to those in the LBL group.

Determine which teaching method is better at improving clinical analysis skills of clinical medical students

[IMAGE OMITTED: SEE PDF]

A total of 1,358 students across four studies were assessed for their clinical analysis skills (FCCL group: 672, LBL group: 686). Significant statistical heterogeneity was observed among the studies (I2 = 96.61%, P < 0.001). As a result, a random effects model was employed for the meta-analysis. The pooled effect size revealed a statistically significant difference in clinical analysis skills between the two groups (Cohen’s d = 1.53, 95% CI: 0.86 to 2.19, P = 0.00) (Fig. 3b). The Cohen's d for clinical analytical skills was 1.53 (> 0.8), indicating a large effect size between the two groups, suggesting a significant difference between the FCCL and LBL groups in improving students'clinical analytical skills. Students in the FCCL group demonstrated superior clinical analysis skills compared to those in the LBL group.

Subgroup analysis by country of the study population

In this study, the theoretical scores and clinical analysis skills of students from different countries were examined to evaluate the generalizability of the teaching effectiveness of the FCCL and LBL methods. A comprehensive analysis of the results revealed that Chinese students in the FCCL group outperformed those in the LBL group in both theoretical scores (SMD = 1.03, 95% CI: 0.30 to 1.77, P = 0.01; heterogeneity, P = 0.00) and clinical analysis skills (SMD = 2.54, 95% CI: 1.06 to 4.02, P < 0.001; heterogeneity, P = 0.00). Among students from Western countries, those in the FCCL group demonstrated superior clinical analysis skills compared to the LBL group (SMD = 0.45, 95% CI: 0.26 to 0.63, P < 0.001; heterogeneity, P = 0.21), whereas no statistically significant differences were observed between the two teaching methods in terms of theoretical scores (SMD = 0.10, 95% CI: −0.28 to 0.48, P = 0.60; heterogeneity, P = 0.00). The comparative effectiveness of the two teaching methods for students from different countries is illustrated in Fig. 4.

[IMAGE OMITTED: SEE PDF]

Sensitivity analysis

Given the substantial heterogeneity observed in the included studies, sensitivity analyses were conducted to evaluate the robustness of the combined effect size. For each outcome indicator, the remaining studies were sequentially excluded, and the analysis was re-performed (Fig. 5). The results indicated that the combined effect sizes for theoretical scores remained within their initial 95% confidence intervals (CIs) after the exclusion of each study. In contrast, the combined effect size for clinical analysis skills, after excluding the study by Cai et al. [17], fell outside its initial 95% CI, suggesting that the result may not be stable. However, the statistical significance of the effect remained consistent with the original analysis. Further, we identified that the studies by Cai et al. [17] and Yang et al. [24] exerted the greatest influence on the Cohen’s d for clinical analysis skills. After excluding these two studies, the I2 statistic for the combined Cohen’s d decreased to 1.4%, and the heterogeneity p-value increased to 0.419. The revised combined effect size was Cohen’s d = 0.50 (95% CI: 0.38 to 0.62, P < 0.001), which did not alter the outcome of the original analysis. Upon reviewing the original studies, it was noted that neither Cai et al. [17] nor Yang et al. [24] reported explicit grouping methods, and there was a significant gender imbalance in the samples of Cai et al.’s control group and Yang et al.’s overall sample. Therefore, these two studies exhibit higher heterogeneity compared to the remaining studies, and their exclusion has a significant impact on the study results.

[IMAGE OMITTED: SEE PDF]

Publication bias

Funnel plots were initially employed to visually assess the publication bias of the studies included in the two combined effect sizes (Fig. 6). The symmetry of the funnel plot for theoretical scores (Fig. 6A) appeared relatively well-maintained, suggesting minimal publication bias. In contrast, the funnel plot for clinical analysis skills (Fig. 6B) exhibited clear asymmetry. However, it is important to note that this visual assessment is inherently subjective. Moreover, the limited number of studies included in each composite effect size hindered a definitive evaluation of the funnel plot's accuracy [31]. To address this limitation, we performed Egger's linear regression test to further evaluate publication bias for both outcome indicators. The results indicated no significant publication bias for theoretical scores (P >|t|= 0.281) but suggested the presence of publication bias for clinical analysis skills (P >|t|= 0.035). Despite these findings, sensitivity analysis through the"trim and fill"method revealed no substantial changes in the combined effect size after the removal of individual studies, thereby supporting the robustness and reliability of our results.

[IMAGE OMITTED: SEE PDF]

Discussion

Summary of main findings

This study presents a systematic review and meta-analysis of the impact of integrating flipped classroom (FC) and case-based learning (CBL) methodologies (FCCL) on the knowledge acquisition of clinical medicine students in recent years. The meta-analysis included 11 studies that assessed two key outcome measures—theoretical scores and clinical analysis skills—which are widely regarded as critical indicators of clinical competence. Our findings suggest that the FCCL approach is more effective than traditional lecture-based learning (LBL) in enhancing students'mastery of medical knowledge and clinical analysis abilities. Subgroup analyses indicated that both Chinese and Western students in the FCCL group outperformed their LBL counterparts in terms of clinical analysis skills. Furthermore, Chinese students in the FCCL group achieved significantly higher theoretical scores compared to those in the LBL group, a difference not observed among Western students. However, the reliability of these results requires further validation through additional educational experiments.

In recent years, there has been a shift in teaching methodologies from a traditional, teacher-centered, lecture-based model to a student-centered, active learning approach. This transformation has been particularly evident in medical education, with the adoption of strategies such as flipped classrooms and case-based teaching gaining increasing popularity [24, 29, 30]. Han and Klein emphasized that providing students with preparatory materials before class, coupled with independent study, enables them to review content and enhance their knowledge, representing an effective instructional approach [31]. The flipped classroom (FC) model, in particular, capitalizes on this strategy by distributing instructional videos and resources in advance, allowing students to engage in self-paced, personalized learning and assume greater responsibility for their educational progress [32]. This model also optimizes classroom time, as students are better prepared and more confident when engaging in class activities. Furthermore, active participation in discussions and collaborative learning fosters deeper skill acquisition and knowledge retention compared to traditional lecture-based instruction [33]. Flipped Classroom (FC) approaches facilitate the reinforcement of students'ability to recognize clinical signs and symptoms [7]. Case-based Learning (CBL), a specific form of Problem-Based Learning (PBL), is a student-centered pedagogy that is driven by clinical cases. It bridges the gap between theoretical knowledge and clinical practice [34], mirroring the decision-making processes encountered in real clinical environments [35]. This method enables students to apply abstract theoretical concepts to complex clinical scenarios, fostering early adaptation to clinical practice. Over time, CBL has evolved to become diverse and can be combined with other teaching methods [36]. FC, being a multi-modal educational approach, exhibits a high degree of compatibility with other teaching formats. Consequently, the integration of FC and CBL into a combined pedagogical model, referred to as Flipped Classroom Case Learning (FCCL), is likely to yield synergistic benefits, producing a teaching effect where"1 + 1 > 2."Students receiving LBL teaching are generally passive recipients of knowledge, leading to higher rates of forgetting. In contrast, students taught through FCCL engage in self-directed learning of theoretical knowledge prior to class and participate in discussions and analyses of real cases during class. This integration of abstract theory with actual cases enhances understanding and memory, resulting in long-term retention of theoretical knowledge. FCCL provides numerous opportunities for clinical case analysis, allowing students to encounter a variety of case descriptions, diagnostic processes, and treatment strategies. This helps to develop clinical thinking, enabling students to identify causes from symptoms, formulate appropriate treatment plans, and enhance clinical skills through practice, thereby ensuring long-term retention of clinical analysis abilities. FCCL also fosters communication and collaboration skills, as students exchange ideas and share experiences in group discussions, which is critical in clinical team settings. The retention of long-term theoretical knowledge and clinical analysis skills lays the foundation for cultivating outstanding clinicians, ensuring their continuous professional development throughout their careers. Empirical evidence supports this, with students taught through FCCL outperforming those in traditional lecture-based learning (LBL) in both theoretical knowledge and clinical analysis skills. These findings suggest that FCCL holds considerable promise as a viable alternative to traditional lecture-based instruction.

Subgroup analyses indicate that the differences across countries represented in the student sample may have contributed significantly to the variability in the results regarding clinical analysis skills. Specifically, the studies by Mohsen et al. [22] and Lewis et al. [20] appeared to be the primary factors behind the lack of a statistically significant difference in theoretical knowledge between students exposed to FCCL and those subjected to LBL in Western countries. Notably, one course in Mohsen et al.'s study [22] reported that students in the LBL group outperformed those in the FCCL group in terms of theoretical knowledge. We hypothesize that this outcome can be attributed to the specialized nature of both studies. In these studies, surgical students were selected as participants, and the demanding nature of surgical internships—characterized by high workload and student competition—necessitates careful consideration of teaching methods that optimize students’ time and energy expenditure for effective learning [37]. While FCCL necessitates adequate pre-course preparation, students often prioritize clinical practice over theoretical learning due to time conflicts stemming from their demanding schedules. As a result, they may not fully engage with or benefit from classroom instruction. Consequently, the application of FCCL in surgical education in Western countries requires careful consideration. However, it is important to interpret the findings of the subgroup analysis with caution. The number of studies included in our analysis is limited, and publication bias may be present. These studies also exhibit methodological biases, including small sample sizes, unclear sample grouping methods, and limitations in the representativeness or comprehensiveness of the exam content. Future research should include more high-quality randomized controlled trials to further assess its accuracy and reliability.

Strengths and limitations of the review

An extensive search was conducted across multiple databases to identify relevant studies. Outcome indicators were selected based on their ability to best reflect the competencies of clinical medicine students. The substantial heterogeneity observed across the included studies is a significant concern. To address this, subgroup and sensitivity analyses were performed, revealing that the variability could be attributed to factors such as substantial differences in sample sizes, imbalances in the male-to-female ratio, geographical variations among student populations, and discrepancies in the types of examinations or assessment methods employed across different institutions. Despite these efforts, the study’s findings, particularly with regard to clinical analytical skills, remain somewhat unstable. These factors undoubtedly raise concerns about the accuracy, stability, and generalizability of the results. It is recommended that clinical medical educators consider these variables comprehensively, interpret the study’s results critically, and apply them contextually, aligning with real-world clinical settings, rather than adopting them rigidly. Additionally, the funnel plot indicates the potential presence of publication bias in our findings. Although the Egger's test and the trim-and-fill method were employed to mitigate this issue, the relatively small number of studies included in the meta-analysis may constrain the accuracy of these assessments. We encourage future clinical educators to contribute to this area of research and recommend increasing sample sizes in future studies to enhance statistical power and minimize the impact of publication bias. Moreover, the exclusive focus on students'objective performance, without incorporating subjective assessments from students or teachers, may limit the comprehensiveness of our findings. Despite these limitations, this meta-analysis provides valuable insights into the role of FCCL in the context of clinical medical education.

Implications for future research

Our study provides preliminary evidence suggesting that FCCL model may offer superior effectiveness compared to the LBL approach in clinical medical education. However, further rigorous investigations are necessary to confirm its reliability and generalizability. Future research could explore the applicability of various case types across different medical disciplines, as well as the customization of FCCL models tailored to the specific learning needs and characteristics of students in diverse clinical specialties. This would aim to enhance students'learning satisfaction and sense of achievement. Additionally, the implementation of the FCCL model places greater demands on instructors in terms of preparing learning materials, such as instructional videos and case designs. Further exploration is needed to support teachers in adapting to this new pedagogical approach and in enhancing their teaching competence. Future studies should also investigate the integration of advanced educational technologies—such as online learning platforms and virtual reality tools—to augment case-based teaching and the flipped classroom methodology.

Conclusions

In this study, we compared the educational effectiveness of FCCL and LBL in clinical medical education. Our results indicate that FCCL offers greater benefits than LBL in enhancing both theoretical knowledge and clinical analytical skills among medical students. Therefore, we believe that there is a theoretical basis for administering FCCL in the field of teaching clinical medicine, except for the need for careful selection of surgical education in Western countries. It is important to note that the number of studies included in this research is limited, there is a potential publication bias, the overall sample size is relatively small, and there is considerable heterogeneity among the studies, all of which may impact the accuracy of this study. To further enhance the credibility of this study, researchers in the field of clinical medical education could demonstrate the generalizability of the FCCL teaching method by expanding its application across various disciplines, such as obstetrics, gynecology, pediatrics, and others. Additionally, adopting more rigorous experimental designs, such as randomization through random number tables and blind assessment of exam scores, could improve the accuracy of the findings. The extensibility of FCCL can also be improved by evaluating student and faculty satisfaction through feedback surveys.

Data Availability

All data generated or analyzed during this study are included in this published article. The Stata raw dataset can be provided on request. The corresponding author, Zhi-Gang Sun, will provide additional data, if requested.

Abbreviations

FCCL:

Flipped Classroom Case Learning

LBL:

Lecture-based Learning

FC:

Flipped classroom

CBL:

Case-based Learning

RCT:

Randomized Controlled Trial

Q-RCT:

Quasi-randomized Controlled Trial

TS:

Theoretical Scores

CAL:

Clinical Analysis Skills

N:

None

PRISMA:

Preferred Reporting Items for Systematic Reviews and Meta-Analyses

SMD:

Standardized Mean Difference

CI:

Confidence Interval

T2 :

Tau-squared

References

Freeman S, Eddy SL, McDonough M, Smith MK, Okoroafor N, Jordt H, Wenderoth MP. Active learning increases student performance in science, engineering, and mathematics. Proc Natl Acad Sci U S A. 2014J 10;111(23):8410–5.

Prober CG, Heath C. Lecture halls without lectures–a proposal for medical education. N Engl J Med. 2012;366(18):1657–9.

Mehta NB, Hull AL, Young JB, Stoller JK. Just imagine: new paradigms for medical education. Acad Med. 2013;88(10):1418–23.

Ramnanan CJ, Pound LD. Advances in medical education and practice: Student perceptions of the flipped classroom. Adv Med Educ Pract. 2017;8:63–73.

Thistlethwaite JE, Davies D, Ekeocha S, Kidd JM, MacDougall C, Matthews P, Purkis J, Clay D. The effectiveness of case-based learning in health professional education. A BEME systematic review: BEME Guide No. 23. Med Teach. 2012;34(6): e421–44.

Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, Ioannidis JP, Clarke M, Devereaux PJ, Kleijnen J, Moher D. The PRISMA statement for reporting systematic reviews and meta-nalyses of studies that evaluate healthcare interventions: explanation and elaboration. BMJ. 2009;21(339):b2700.

Hu R, Gao H, Ye Y, Ni Z, Jiang N, Jiang X. Effectiveness of flipped classrooms in Chinese baccalaureate nursing education: A meta-analysis of randomized controlled trials. Int J Nurs Stud. 2018;79:94–103.

Higgins JP, Altman DG, Gøtzsche PC, Jüni P, Moher D, Oxman AD, Savovic J, Schulz KF, Weeks L, Sterne JA; Cochrane Bias Methods Group; Cochrane Statistical Methods Group. The Cochrane Collaboration's tool for assessing risk of bias in randomised trials. BMJ. 2011 18;343:d5928.

Higgins J, Green S. Cochrane Handbook for systematic reviews of interventions. Chichester: John Wiley & Sons; 2011.

Higgins JPT, Thompson SG. Quantifying heterogeneity in a meta-­analysis. Stat Med. 2002;21:1539–58.

Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ. 2003;327(7414):557–60.

DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials. 1986;7(3):177–88.

Mantel N, Haenszel W. Statistical aspects of the analysis of data from retrospective studies of disease. J Natl Cancer Inst. 1959;22(4):719–48.

Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ. 1997S;315(7109):629–34.

Guyatt GH, Oxman AD, Montori V, Vist G, Kunz R, Brozek J, Alonso-Coello P, Djulbegovic B, Atkins D, Falck-Ytter Y, Williams JW Jr, Meerpohl J, Norris SL, Akl EA, Schünemann HJ. GRADE guidelines: 5. Rating the quality of evidence--publication bias. J Clin Epidemiol. 2011;64(12):1277–82.

Duval S, Tweedie R, Trim TR. Trim and fill: a simple funnel-­plot-­based method of testing and adjusting for publication bias in meta-­analysis. Biometrics. 2000;56:455–63.

Cai L, Li YL, Hu XY, Li R. Implementation of flipped classroom combined with case-based learning: A promising and effective teaching modality in undergraduate pathology education. Medicine (Baltimore). 2022;101(5):e28782.

Qian Q, Yan Y, Xue F, Lin J, Zhang F, Zhao J. Coronavirus Disease 2019 (COVID-19) Learning Online: A Flipped Classroom Based on Micro-Learning Combined with Case-Based Learning in Undergraduate Medical Students. Adv Med Educ Pract. 2021;28(12):835–42.

Riddell J, Jhun P, Fung CC, Comes J, Sawtelle S, Tabatabai R, Joseph D, Shoenberger J, Chen E, Fee C, Swadron SP. Does the Flipped Classroom Improve Learning in Graduate Medical Education? J Grad Med Educ. 2017;9(4):491–6.

Lewis CE, Chen DC, Relan A. Implementation of a flipped classroom approach to promote active learning in the third-year surgery clerkship. Am J Surg. 2018;215(2):298–303.

Lu F, Luo Z, Huang T, Lv X, Wang H, Wang Y, Yin H. Effectiveness evaluation of flipped classroom in emergency medicine online teaching for medical undergraduates. Med Teach. 2024;46(5):689–96.

Kolahdouzan M, Mahmoudieh M, Rasti M, Omid A, Rostami A, Yamani N. The effect of case-based teaching and flipped classroom methods in comparison with lecture method on learning and satisfaction of internship students in surgery. J Educ Health Promot. 2020;28(9):256.

Cao YX, Xia SL, Zhu ZY, Zeng FR, Li HN, Zhang TT, Liu YJ. Exploring lemology teaching with “internet plus” flipped classroom pedagogy. BMC Med Educ. 2023;23(1):341.

Yang F, Lin W, Wang Y. Flipped classroom combined with case-based learning is an effective teaching modality in nephrology clerkship. BMC Med Educ. 2021;21(1):276.

Marshall AM, Conroy ZE. Effective and Time-Efficient Implementation of a Flipped-Classroom in Preclinical Medical Education. Med Sci Educ. 2022;32(4):811–7.

Gray MM, Dadiz R, Izatt S, Gillam-Krakauer M, Carbajal MM, Johnston LC, Payne A, Vasquez MM, Bonachea EM, Karpen H, Falck AJ, Chess PR, Huber M, French H. Comparison of knowledge acquisition and retention following traditional didactic vs. flipped classroom education utilizing a standardized national curriculum: a randomized controlled trial. J Perinatol. 2022;42(11):1512–1518.

Chan PP, Lee VWY, Yam JCS, Brelén ME, Chu WK, Wan KH, Chen LJ, Tham CC, Pang CP. Flipped Classroom Case Learning vs Traditional Lecture-Based Learning in Medical School Ophthalmology Education: A Randomized Trial. Acad Med. 2023;98(9):1053–61.

Hu B, Wang L, Wu J, Zhu L, Chen Z. A combination of case-based learning with flipped classroom improved performance of medical students in nephrology bedside teaching. BMC Med Educ. 2024;24(1):995.

Gillette C, Rudolph M, Kimble C, Rockich-Winston N, Smith L, Broedel-Zaugg K. A Meta-Analysis of Outcomes Comparing Flipped Classroom and Lecture. Am J Pharm Educ. 2018;82(5):6898.

Diel RJ, Yom KH, Ramirez D, Alawa K, Cheng J, Dawoud S, Snyder MR, Kemp PS. Flipped ophthalmology classroom augmented with case-based learning. Digit J Ophthalmol. 2021;27(1):1–5.

Han E, Klein KC. Pre-Class Learning Methods for Flipped Classrooms. Am J Pharm Educ. 2019;83(1):6922.

Shiau S, Kahn LG, Platt J, Li C, Guzman JT, Kornhauser ZG, Keyes KM, Martins SS. Evaluation of a flipped classroom approach to learning introductory epidemiology. BMC Med Educ. 2018;18(1):63.

El-Miedany Y, El-Gaafary M, El-Aroussy N, Youssef S. Flipped Learning: Can Rheumatology Lead the Shift in Medical Education? Curr Rheumatol Rev. 2019;15(1):67–73.

Demarco R, Hayward L, Lynch M. Nursing students’ experiences with and strategic approaches to case-based instruction: a replication and comparison study between two disciplines. J Nurs Educ. 2002;41(4):165–74.

Hudson JN, Buckley P. An evaluation of case-based teaching: evidence for continuing benefit and realization of aims. Adv Physiol Educ. 2004;28(1–4):15–22.

Cen XY, Hua Y, Niu S, Yu T. Application of case-based learning in medical student education: a meta-analysis. Eur Rev Med Pharmacol Sci. 2021;25(8):3173–81.

Taylor DC, Hamdy H. Adult learning theories: implications for learning and teaching in medical education: AMEE Guide No. 83. Med Teach. 2013;35(11): e1561–72.