Orthopedics, a field deeply rooted in anatomical mastery, biomechanical logic, and dynamic surgical decision-making, faces persistent challenges in fostering well-rounded competencies. Traditional lecture-based learning (LBL), which emphasizes rote memorization of concepts like fracture classifications or implant protocols, struggles to bridge the gap between theory and real-world application. Studies have revealed that under LBL, orthopedic residents exhibit a 31% error rate in biomechanical analysis of complex cases and significantly lag in managing postoperative complications compared to active learning approaches (1). In China’s current system, the fourth year of medical school focuses on theory, while the fifth shifts to clinical practice, yet both phases rely heavily on standardized curricula. This rigid structure stifles critical thinking and teamwork skills—72% of trainees report limited exposure to real surgical challenges during internships, as lessons remain textbook-bound (2). This “theory-practice divide” contributes to burnout (affecting 18% of students) and skill gaps: only 43% of graduates from top Chinese medical schools meet competency standards in emerging technologies like robot-assisted surgery (3), far below global benchmarks. Thus, reforming orthopedic education to integrate active learning, dynamic feedback, and adaptability to new technologies has become urgent.

To address these limitations, innovative strategies like virtual reality (VR), 3D interactive systems, and flipped classrooms (FC) are transforming training. VR simulations replicate real surgical environments, boosting skill transfer efficiency. Vallée et al. (1) found VR-trained residents completed arthroscopic procedures 42% faster, with 14% higher accuracy in tissue identification than traditional apprenticeship trainees. Xue et al. (3) developed a 3D Training System that integrates real imaging with live anatomical guides, tripling femoral neck fracture classification accuracy while reducing cognitive load by 28%. Meanwhile, FC combined with team-based learning (TBL) redefines roles: pre-class self-study, in-case debates, and post-surgery debriefs raised clinical decision-making scores by 9.2% and complication prediction accuracy to 72.3% (4). A study of 130 students found that podcast/videos users significantly outperformed text users in posttests and knowledge gain, with higher approval ratings for podcasts/videos (5). Recent quasi-experimental work further supports the efficacy of integrated digital approaches; for instance, massive open online course (MOOC)–virtual simulation combinations have demonstrated significant improvements in surgical skill acquisition, including wound debridement and basic operative techniques (6, 7). Beyond building muscle memory for procedures, these tools provide visual data and instant feedback to correct errors—opportunities missing in traditional teaching.

Despite their clear benefits, debates persist regarding the best strategies. A key issue is the lack of direct comparisons between methods. Moreover, conflicting conclusions across meta-analyses could limit the scalable implementation of innovative teaching methods in educational and clinical environments (8–11). Despite the suggestion that specific teaching environments should undergo thorough evaluation when introducing new teaching strategies (12), there remains a dearth of research to substantiate this notion. Here, systematic reviews and network meta-analysis (NMA)—considered the highest level of evidence—can help by comparing multiple methods at once. In this study, we employ NMA integrated with statistical ranking to systematically evaluate eight mainstream educational approaches (LBL, problem-based learning (PBL), video, VR, 3D, 3D + PBL, FC and FC + TBL) in fostering theoretical knowledge assimilation and clinical skill proficiency. The findings aim to establish a data-driven decision-making framework for optimizing resource allocation across diverse educational and clinical contexts.

The systematic review and NMA adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) extension statement (13), with the objective of assessing and comparing the efficacy of eight individual teaching strategies in enhancing orthopedic education by examining three specific indicators (14). This review did not require ethical approval as it utilized data from published studies, and no detailed participant information was made public.

Recently, progress in computer technology and changes in the healthcare service system have made it challenging for medical students trained in traditional settings to fully adapt to the growing needs of public healthcare (28). Numerous innovative teaching methods have gained widespread adoption worldwide and shown promising results in improving teaching efficacy. However, there is uncertainty regarding the most effective teaching strategy for enhancing orthopedic teaching effectiveness due to the lack of direct comparative evidence. Compared to traditional pairwise meta-analysis, the NMA approach offers a more straightforward method and yields more insightful information (29). Consequently, we undertook this NMA to assess the impact of various teaching methods on student performance and satisfaction, considering theoretical exam scores, practical or clinical skill assessments, and student satisfaction ratings. By synthesizing all available evidence from 11 RCTs involving 690 medical students or residents in orthopaedics, our findings revealed that VR stands out as the most effective method for enhancing the practical or clinical skills of medical students or residents, whereas PBL has proven to offer a more effective overall learning experience throughout the educational journey.

The training of orthopedic interns has shifted from a primary focus on theoretical mastery to an emphasis on clinical application. Traditional LBL teaching tends to concentrate on imparting theoretical knowledge, which can hinder the development of students’ creativity and individual traits, and overlooks their subjective initiative and potential (30). Previous meta-analyses have found that PBL can improve knowledge scores compared to LBL. However, due to the relatively limited research on orthopedic education, no meta-analysis has been conducted to compare the effects of FC combined with TBL (FC + TBL) versus LBL on orthopedic teaching outcomes. While research indicates that both TBL and FC are more effective in enhancing student learning compared to traditional LBL, each method has its limitations (31, 32). Some students have provided feedback indicating that although the FC offers high-quality instructional resources, it demands strong self-learning abilities and offers limited opportunities for student interaction, potentially hindering the learning effectiveness of certain students (33, 34). Others have highlighted that TBL requires a substantial amount of time to master and necessitates a consistent schedule among team members (15). However, when these two approaches are integrated, they compensate for each other’s shortcomings, enabling students to engage with one another, acquire a broad range of knowledge in a brief period, and allocate more time to preparing for in-class problem discussions. More crucially, this combination fosters group discussions and critical thinking, enhances active communication abilities, and promotes a more relaxed atmosphere in the classroom (35). The apparent advantage of FC + TBL may arise from synergistic pedagogical mechanisms. The FC fosters self-directed cognitive preparation, while TBL promotes collaborative knowledge application through structured problem-solving. This combination aligns with constructivist theory—where learners actively construct knowledge through experience—and social interdependence theory, which posits that positive group interdependence enhances motivation, accountability, and deep learning. In our study, when assessing the impact of orthopedic teaching on exam scores, we also found that FC + TBL is the most effective teaching method, with an 81.73% likelihood of effectiveness. This is consistent with the findings of previous study (36).

The clinical skill test scores, which assess students’ competence in essential surgical techniques, offer a direct indication of the efficacy of hands-on training. Multiple systematic reviews have examined the advantages of incorporating VR technology into orthopaedic training programs. These reviews have consistently shown that VR notably improves both the theoretical understanding and practical abilities of learners, ranging from beginners to seasoned professionals like orthopaedists and neurosurgeons (37). Particularly, the integration of VR into surgical training for spinal interventions marks a novel approach to honing residents’ skills and enhancing their self-assurance (21). Our NMA revealed that VR significantly outperforms traditional LBL significantly (SMD = 6.88, 95% CI: 1.05 to 12.13), with a 62.07% probability of being the most effective method for improving procedural or clinical skill test scores. These results further highlight VR’s potential as a highly efficient and cost-effective training tool, effectively bridging the gap between theoretical knowledge and practical application.

A crucial metric for assessing teaching methods is student satisfaction. Prior systematic reviews have indicated that students in the PBL group exhibit greater interest in and satisfaction with teaching (38). Our analysis revealed that according to the SUCRA analysis, there is a 61.53% probability that PBL is the most effective approach for enhancing student satisfaction scores and subjective evaluations. PBL encourages students’ proactive participation, fosters their learning abilities, and boosts their enthusiasm for learning. Norman et al. (39) demonstrated that PBL enhances students’ learning interest, self-learning capabilities, and sustains these interests over time. Another study found that students in the PBL group outperformed those in the traditional teaching group in terms of professional knowledge and classroom satisfaction (p < 0.05) (22). Likewise, Ren et al. (40) reported that PBL significantly improved satisfaction among both students and teachers, aligning with our meta-analysis findings. Nevertheless, several pivotal factors, such as small group sizes and realistic case scenarios, influence teaching satisfaction (41).

This research consolidates and strengthens the existing evidence base in the academic literature on orthopedic education, advocating for innovative teaching approaches. To our understanding, previous systematic reviews and NMAs have assessed the impact of novel teaching methods on particular majors and curricula, including medicine (27), nursing (42), and pharmacology (43). However, our study represents the inaugural systematic review and NMA to explore the effects of innovative teaching strategies specifically in orthopaedics, thereby addressing a gap in orthopedic education. Orthopaedics differs from other surgical specialties (e.g., neurosurgery, general surgery) in several critical ways that shape educational needs: (1) it emphasizes hands-on psychomotor skills (e.g., fracture reduction, implant placement, arthroscopic manipulation) that require precise spatial reasoning and muscle memory—skills that may respond differently to instructional strategies than the cognitive or procedural focus of other fields; (2) orthopaedic training covers a broad spectrum of subspecialties (e.g., trauma, spine, sports medicine, joint replacement), each with distinct technical and decision-making demands; (3) orthopaedic trainees often balance high operative volume with didactic learning, creating unique time constraints that influence the applicability of teaching strategies (e.g., concise, simulation-based training may be more feasible than lengthy group discussions). Additionally, orthopaedic surgeons-in-training face distinct challenges, such as adapting to rapidly evolving implant technologies and navigating the physical demands of surgical procedures, which differ from the training priorities of other surgical specialists (e.g., neurosurgeons may focus more on imaging interpretation and microsurgical precision). By focusing exclusively on orthopaedics, our study provides targeted evidence for educators in this specialty, who previously lacked a synthesized overview of which innovative strategies align with their trainees’ unique needs. This specificity enhances the translatability of our findings to clinical practice, making it a core strength of the study. The NMA findings offer a ranked order of eight teaching strategies based on their effectiveness in mesh fixation, guiding curriculum designers and educators in adopting these novel instructional techniques. Furthermore, to ensure comprehensiveness and minimize selection bias, two authors independently conducted the study selection, data extraction, and quality assessment processes.

While FC + TBL and VR show promise, their real-world adoption requires careful consideration of feasibility. VR implementation entails significant costs for hardware, software licensing, and faculty training. FC + TBL demands curriculum redesign and skilled facilitation. We recommend phased integration—starting with pilot programs in well-resourced institutions—and the development of open-access TBL modules to support scalability in low-resource settings.

Despite its strengths, our study also has some limitations that should be acknowledged. Firstly, this NMA assesses only seven distinct novel teaching strategies, excluding other modes and various combinations of methodologies. Each teaching method incorporates a limited number of studies, with a maximum of five. The intricacy stemming from the amalgamation of diverse teaching approaches may impede the conduct of the NMA. Nonetheless, this NMA can aid in pinpointing effective teaching strategies and optimizing course arrangement. Secondly, the nature of the instructional approach precludes blinding of both students and instructors. Consequently, students may modify their behavior when aware of being studied under different teaching strategies, thereby affecting the reliability of evidence in RCTs. This limitation is an inherent and crucial shortcoming in all primary studies included in this meta-analysis. Given that pre-registration of protocols is not mandatory in educational research, the bias related to selective reporting of results does not apply when assessing literature quality, which impacts the quality evaluation of the included studies. Thirdly, this study is restricted to English-language publications, potentially overlooking trials with negative outcomes that may exist in extensive national databases, such as those in Chinese literature (44). Fourthly, differences in baseline characteristics among students, including residents and medical students in the included studies, the broad design framework, and variations in teacher expertise levels may introduce a certain degree of heterogeneity (45). Moreover, NMA differs from direct comparison and carries an additional risk of bias, particularly when few direct comparisons exist between the novel teaching strategies in our NMA, potentially leading to imprecision. Significant heterogeneity and imprecision in the data substantially degrade the quality of evidence, reducing the accuracy of results. A further limitation is the overrepresentation of Chinese orthopaedic residents, which may limit generalizability to Western or low-resource settings. Educational and cultural differences may affect engagement with strategies like TBL (46). Generational factors also matter: our predominantly Gen Z cohort tends to favor technology-enhanced tools like VR, which may not reflect preferences in other regions or age groups (47). Gender disparities in orthopaedics (e.g., lower female representation in some settings) could further influence participation in collaborative methods (48). Moreover, promising interventions such as VR and high-fidelity 3D simulations require resources often unavailable in low- and middle-income countries. Future research should test low-cost alternatives (e.g., free online modules, low-fidelity simulations) and include more diverse international samples to improve external validity. These contextual factors likely contributed to the modest effects observed for some innovative strategies, underscoring their context-dependent efficacy. In addition, our analysis included 11 RCTs involving 690 participants; however, many comparisons—such as VR for theoretical knowledge—were informed by a single study. This sparsity may reduce the precision and stability of effect estimates for certain nodes. Although sensitivity analyses excluding high-risk-of-bias studies yielded consistent rankings, future large-scale, head-to-head randomized trials are essential to validate these findings. Lastly, this study does not comprehensively capture other vital qualities and skills of orthopedic residents or students in the “outcomes” section, such as case analysis ability, social and communication skills, problem-solving and self-learning capabilities, and subjective enthusiasm. Future research should involve direct comparisons among novel teaching strategies to identify optimal educational approaches for orthopedic education.

In conclusion, our NMA have provided valuable insights into the effectiveness of novel teaching strategies in orthopaedics education. Among the strategies evaluated, FC + TBL and VR emerged as the most effective methods for improving theoretical test scores and procedural or clinical skill scores, respectively. However, it is important to note that the differences among the teaching strategies were relatively minor, and more research is needed to fully understand their relative advantages and disadvantages. Future studies should focus on larger sample sizes, diverse student populations, and a wider range of outcome measures to provide a more comprehensive evaluation of teaching strategies in orthopaedics education. Overall, our findings have important implications for curriculum designers and educators seeking to optimize the learning experience for orthopaedics students or residents.