The drawback of two years of basic science and two years of clinical training is that students may not recognize the importance of the basic biomedical sciences to clinical practice [1]. Although the integrated curriculum appears to be a solution, it has significantly shortened the time available to teach anatomy and embryology [2]. In recent years, concerns have been raised about medical students' knowledge of anatomy, their preparedness for clinical practice, and the impact on the quality and safety of the healthcare they provide [3].

The use of problem-based learning (PBL) [4], case-based learning (CBL) [5], or other small-group active learning, such as TBL [6], combined with exposure to patients through ambulatory clinic attendance, integrated clinical practice into the preclinical years. These diverse approaches further contributed to the development of curriculum integration in the preclinical years, both horizontally and vertically. At the end of each course or module, assessments included a written multiple-choice question (MCQ) exam and a relevant laboratory practical exam.

Cooper Medical School of Rowan University (CMSRU) enrolled its first class of 50 students in 2012. The school was approved to admit more students, increasing the class size to 100. As a result, the faculty size was also expanded. Preclinical courses engaged students in active, case-based, participatory learning in small groups; the cases incorporated clinical inquiries and fostered the habit of self-regulated, self-directed learning. The preclinical curriculum encompassed horizontally and vertically integrated organ system-based courses/modules. Classes were short, especially in anatomy, and students needed to assimilate the required knowledge and understanding in fewer dissection hours. The integrated curriculum was touted as a cornerstone of modern medical education and practice; the information students were expected to learn and master also expanded rapidly.

In CMSRU's curriculum, assessment consisted of MCQ written examinations with clinical vignettes, while practical examinations used short-answer questions. After the spring 2013 course delivery, anatomy instructors in the Skin and Musculoskeletal (SMS) course noted a challenge: first-year medical students performed poorly on practical examinations. The practical examination required students to identify a tagged structure and answer a primary, secondary, or tertiary question about it. Faculty realized that students correctly identified structures on the cadavers they dissected, but could not identify them on the others because they spent very little time reviewing cadavers they had not dissected and lacked familiarity with those structures. Faculty accepted the challenge of enhancing students' ability to recognize structures across all cadavers by implementing pedagogical changes, including the use of the Retrieval Practice Laboratory Review (RPLR) or Retrieval Review (RR). The RR sessions provided students with formative feedback and peer-to-peer teaching, helping them address knowledge gaps before exams. This study examines how retrieval practice, through peer-to-peer teaching with the content, improves students' knowledge of anatomy, resulting in better performance and enhanced learning.

The Rowan University Institutional Review Board approved the protocol for this study as an 'exempt' study (ID: Pro2018002430). The amount of time dedicated to teaching anatomy varied for each subject module in CMSRU's integrated organ system-based courses curriculum, and the laboratory dissection time changed accordingly. The anatomy laboratory was open 24/7 for additional study hours. Each dissecting group consisted of five randomly selected students; four full-time faculty members taught the lab. In addition to cadavers, the school provided anatomical models and bones, which were also used in the practical exam. The traditional 'tag or spotter' test was criticized for testing rudimentary levels of knowledge [7]. Hence, pure identification of a tagged structure was minimized in favor of higher cognitive level testing involving function and clinical application types of questions [8].

Our anatomy practical required to 1) name the tagged structure (remembering); 2) know the surrounding anatomy to recognize the structure (understanding) and its function (comprehension); and 3) understand the function of the structure (application) or what functional deficit occurs if the structure is damaged (analyzing) and, finally, identify which clinical condition is associated with that deficit (evaluating). Parallel to the integrated curriculum, the practical examination of anatomy followed Bloom's Taxonomy [9].

An Independent Samples t-test was conducted to assess differences in practical examination performance in the Skin and Musculoskeletal (SMS) course before and after the implementation of RR sessions (Table 1). The analysis revealed that medical students' practical examination scores before RR implementation {Mean (SD) = 77.5 (12.4)} were significantly lower than post-RR scores {Mean (SD) = 85.19 (5.72)}; t(438) = -3.25, p < .001.

To further examine the impact of RR across cohorts, a one-way between-subjects ANOVA was conducted to compare practical examination scores across six undergraduate medical education class years. The analysis showed a significant effect of RR on student performance at p < .001 [F(5, 434) = 5.20, p = .000]. Post hoc Tukey HSD comparisons indicated that the 2013 course delivery, which did not experience RR, had a significantly lower mean score (77.50) than the courses delivered in 2014 (85.19), 2015 (86.77), 2016 (84.59), 2017 (85.75), and 2018 (85.90). However, although the 2014 cohort did not have access to RR, their performance was not significantly different from that of post-RR cohorts. This raises questions about what additional factors might have contributed to student performance beyond the introduction of RR. It may be that the 2014 class was better prepared, given that medical school requires a higher level of knowledge. It is also possible that the second class performed well without RR due to the overall quality of the class and other factors, such as well-prepared students.

The number of students who did not pass the practical examination before and after RR implementation is illustrated in Figure 1. Notably, no students failed the 2015 or 2018 practical examinations. This trend suggests that, while RR may have played a role in improving overall performance, other factors, such as curricular adjustments, student preparation habits, and faculty support, may have also contributed.

The present study's results demonstrate that the RR introduction at CMSRU had a positive outcome: student performance improved after implementing this innovative approach, despite the reduced anatomy teaching time. These findings are significant. Such reductions in anatomy curricular time were not unique to medical schools in the United States [2]; they also occurred in North and South American countries [11,12], as well as the United Kingdom and the Republic of Ireland [13]. By utilizing different strategies, the challenge became an opportunity. Other approaches to improving learning outcomes supported the assertion that 'where there is a will, there is a way' [14,15]. Conceptually, the RR is a different approach, but Hofer found that using checklists in a gross anatomy laboratory improved learning outcomes and dissection quality [14]. Rao Bhagavathula utilized a pre-laboratory assignment to improve academic performance and student engagement in a neuroanatomy course [15].

The anatomy practical examinations are much different from all other exam types; students are expected to be familiar with many anatomical structures in different student-dissected cadavers. Additionally, 1) students are expected to know the clinical significance of most of these structures and 2), where applicable, the embryological origin of and congenital disabilities related to some of the structures. In the last two decades, the NBME licensing examination questions have evolved to become based on a horizontally and vertically integrated curriculum. In addition to developing a curriculum for 21st-century medical students, assessments should incorporate other subjects that connect to knowledge, comprehension, and integration [16].

The use of RR sessions was an innovative method for addressing the problem of student failures on the SMS anatomy practical examinations, which was attributed to poor learning and a lack of diligent preparation for the type of testing. Participation in RR resulted in students becoming comfortable with all the cadavers used in the practical examination and gaining familiarity with the structures that they were expected to focus on in both learning and preparation for the assessment, while engaging in peer-to-peer teaching, all of which led to a deeper understanding of the content. Studying the effect of active and passive repetitive learning activities in the retention of anatomy knowledge, Kooloos concluded that the repetition of anatomical knowledge in any form is beneficial for students and will likely improve student outcomes in a curriculum that builds on prior knowledge [17]. Minhas noted that, during initial learning, active learning methods outperform passive ones because of improved student engagement and attitude [18].

Constructivist teaching and theory are grounded in the belief that learning occurs when learners actively engage in the process of constructing meaning and knowledge, rather than passively receiving information. Learners create their own meaning and understanding. Constructivism as a theory can be effective in teaching and educational contexts because students learn through experience rather than solely from textbooks. They are encouraged to apply their critical thinking, deductive reasoning, and analytical abilities to articulate their thoughts and develop solutions.

The RR allowed students to engage in peer-to-peer teaching and simultaneously observe and learn how other students approached structure identification by looking at surrounding structures and differentiating similar structures, such as nerves and blood vessels. This peer-to-peer teaching is identical to 'developing medical students as teachers.' In addition, RR required students to review all cadavers systematically. It is essential to highlight that each dissection group rapidly became a cohesive and productive team, going through the cycle of forming, norming, storming, and performing [19]. This is crucial for successful team interactions, development of trust, professionalism, and communication [6,20].

The RR sessions also provided students with formative feedback, allowing them to address any deficits in knowledge before examinations. According to Rowland, practical examinations are more suitable for testing anatomical knowledge than written ones [21]. As described by Smith, basic factual understanding is a key component of learning; when applied to core concepts, basic building blocks, which can be classed as 'identity,' are essential in learning anatomy [22]. The end goal of medical school is clinical practice, which requires strong clinical reasoning skills. A surface approach to anatomy is significantly related to a low achievement level and little development of clinical reasoning skills; a deep or strategic approach is associated with a high attainment level. Students enjoy the RR and find it helpful in developing their understanding of anatomy and relationships between structures.

Formative RP feedback allowed students to address knowledge deficits before practical examinations. A surface approach to learning anatomy resulted in lower achievement levels; a strategic approach resulted in higher attainment levels, and the RP provided this strategic approach.