Rapid Cycle Deliberate Practice in Healthcare Simulation: a Scoping Review

Rapid cycle deliberate practice (RCDP) is a type of simulation-based medical education (SBME) where learners cycle between deliberate practice and directed feedback until skill mastery is achieved before progressing to subsequent learning objectives. This scoping review examines and summarizes the literature on RCDP, compares RCDP to other modes of instruction, and identifies knowledge gaps for future research. Of the 1224 articles identified, 23 studies met inclusion criteria. The studies varied in design, RCDP technique implementation strategies, and outcome measures. RCDP is associated with positive outcomes in immediate learner performance. It is unclear if RCDP is superior to traditional simulation.


Introduction
Simulation-based medical education (SBME) has been shown to improve knowledge, clinical skills, and behaviors when compared to traditional medical education [1]. It is also associated with improved downstream effects on patient care, patient outcomes, and reduced healthcare costs [2]. The most common or traditional method of simulation debriefing is facilitator-guided post-event debriefing; it occurs after a simulation scenario and is effective for achieving learning goals and understanding correct and incorrect actions [3,4]. Within-event debriefing involves interrupting the simulation scenario for short and focused debriefing events to allow for coaching in real time [3]. There is limited evidence that one method of simulation instruction is better than another.
Rapid cycle deliberate practice (RCDP), introduced by Hunt et al. in 2014, is a type of within event debriefing that "rapidly cycle(s) between deliberate practice and directed feedback until skill mastery is achieved" [5]. RCDP builds on the concepts of deliberate practice and mastery learning.
Deliberate practice, commonly used to gain expertise in music and sports, allows for repeated opportunities to practice a skill with directed feedback until learning objectives are met [6]. Mastery learning, which is associated with clinical skills retention and patient care improvement, consists of clear learning objectives, formative assessments, advancement to the next educational objective once mastery has been achieved, and continued practice [2]. In RCDP, learners meet predefined learning objectives before progressing to more challenging goals or scenarios that build upon previously mastered skills [5]. Hunt et al. describes three principles for RCDP [5]: 1. Apply the concepts of overlearning and automatization to create muscle memory by providing multiple opportunities for repetitive practice to "do it right." 2. Provide expert or evidence-based feedback efficiently. 3. Foster an environment for psychological safety to allow learners to enthusiastically welcome the feedback.
Hunt et al. found that pediatric resuscitation skills and critical care performance improved with RCDP simulation [5].
A systematic review was published in 2017 of RCDP that consisted of 15 resources, only two of which were published studies [7]. As such, it became a narrative review that reported implementation strategies and outcome measures. Results were limited and inconsistent, and the authors recommended conducting a similar review in 2-years time as more material would be available due to the rapid emergence of RCDP [7]. Given the limited results of the last review and the relatively new method of RCDP, the goal of this scoping review is to examine the current RCDP literature to help guide future research and identify gaps in RCDP and simulation. Ultimately, this may help shape the future of simulation instruction in healthcare.
The objectives of this scoping review are to examine and summarize the literature of RCDP in SBME, including outcome measures and implementation strategies; assess if RCDP leads to improved clinical performance of healthcare professionals and healthcare trainees when compared to other modes of instruction; and identify knowledge gaps and opportunities for future RCDP research.

Methods
A comprehensive search strategy was developed by a medical librarian to identify articles on rapid cycle deliberate practice (RCDP). The initial search strategy was developed for Ovid Medline and adapted from the one used by Taras and Everett using a combination of text words [7]. The search strategy was then customized for each database. Searches of the following databases were executed on October 13, 2020: Ovid MEDLINE ALL, Ovid EMBASE, Cochrane Database of Systematic Reviews (Ovid), Cochrane Central Register of Controlled Trials (Ovid), PsycINFO (Ovid), and CINAHL with Full Text (EBSCO). There were no restrictions on publication period. Limits were imposed for English language. No other limits were applied. See Appendix I for database search strategies. Additional search methods included screening reference lists of eligible studies.
The inclusion criteria were based on the PICO (population, intervention, comparison, outcome) approach. The target population was healthcare professionals and trainees. The intervention was RCDP healthcare simulation. The comparison was traditional simulation, other instruction, or no comparison. The outcomes examined were learner satisfaction, knowledge acquisition, clinical performance, implementation of skills in practice, quality of care, and patient outcomes. English language full-text published studies of any design were included.
The initial database search identified 2283 articles ( Fig. 1). One study was not captured by the data search and was found by screening the references of an eligible study. After duplicates were removed, 1224 studies were screened using title and abstract for relevance. The remaining 67 articles were screened using the full text to assess for eligibility based on the inclusion criteria, of which 44 were excluded. Common reasons for study exclusion were as follows:   [8]. Given the heterogeneity of the studies and the broad scope of this review, no critical appraisal of the evidence was performed.

Results
This review included 23 studies [5,. Table 1 outlines study characteristics, intervention, comparison, and outcomes. Nineteen studies were from the USA. Study designs were diverse; the most common study design was pre-/posttest (10 studies), and only five studies were randomized controlled studies. Most participants were residents (8 studies) with pediatric residents being most common (5 studies). The most common type of simulation scenarios that RCDP was used for was pediatric resuscitation (5 studies), pediatric critical care (4 studies), and adult resuscitation (4 studies).
Of the 23 studies, 15 studies directly assessed RCDP, while the remaining 8 studies had RCDP as part of mixed interventions. For example, Brown et al. studied a 1-day bootcamp where two RCDP scenarios were part of a larger intervention involving didactic lectures, case studies, and traditional simulation [12]. Most studies did not have a comparison; five studies compared RCDP to traditional simulation, two studies to standard training, one study to different training frequencies and standard curriculum, and one study to previous traditional simulation experience.
All studies used a similar definition of RCDP, as introduced by Hunt et al. in 2014, and used variations of pausing, micro-debriefing or coaching, and rewinding [5]. However, there were differences in how the RCDP technique was implemented as outlined in Table 2. Many studies had the first simulation run without interruption to act as a needs assessment. Studies with pre-determined pauses split the cases into smaller segments or learning chunks purposely, while other studies paused for praise or error correction.
During the pauses or completion of the short cases, microdebriefing, feedback, or coaching was completed in all studies. After the pause, rewinds occurred to either the beginning and/or to just before the pause. Most cases progressed, and 7 studies reported progressive difficulty. For example, Lemke et al. had medical objectives sequenced into progressively more difficult rounds [22]. Each subsequent round was built on previously mastered skills and escalated in difficulty. The supraventricular tachycardia case started with basic communication objectives, and the subsequent rounds required cardioversion and cardiac arrest management [22].
Outcome measures and results are listed in Table 1. The most common K1 outcomes were increased learner confidence (11 studies) and learner satisfaction (11 studies) in RCDP alone and with mixed interventions. At the K2 level, RCDP was associated with knowledge and skills acquisition demonstrated by significantly increased multiple choice question (MCQ) scores (4 studies), increased checklist scores (8 studies), decreased time to critical tasks (6 studies), and increased chest compression fraction (CCF) (2 studies). For example, a 2-day pre-hospital care provider course in a low resource setting consisted of didactic lectures, skills training, and multiple RCDP simulations. All participants found the course "useful", and it resulted in significant increases in MCQ scores (pretest 67%, posttest 85%, p < 0.001) and simulation checklist performance scores (pretest 42%, post-test 95%, p < 0.001) [20]. At the K3 level, a floor staff survey found that RCDP participants had perceived improvement in leadership, communication, and role identification [14]. Additionally, hospital clinical reports found that nursing staff were better able to manage cardiac arrest after RCDP training [21]. No K4 outcomes were reported.
When comparing interventions with RCDP to standard curriculums, RCDP was associated with a significant decrease in time to tasks, such as the proportion of time a cardiac arrest patient received no chest compressions (noflow fraction), time to defibrillation, and an impressive tenfold reduction in pre-shock pause [5]. A significantly larger CCF and decreased time to compression initiation were also found when examining RCDP compared to traditional BLS training [18]. Interestingly, this study did not find a significant difference in decreased time to defibrillation between the 2 groups, but defibrillation still occurred within the American Heart Association (AHA) guidelines of 180 s [18]. Sullivan et al. examined outcomes with different training frequencies and found that RCDP training for cardiac arrest skills at 2-and 3-month intervals were associated with significant decreased time to initiation of chest compressions and defibrillation when compared to 6-month intervals and standard AHA curriculum [28]. The authors concluded that the ideal time for RCDP resuscitation skills training is every 3 months [28]. When comparing RCDP to traditional simulation, the results were mixed. RCDP participants commented on liking RCDP more [13,22], but feeling more fatigued [22]. Rosman et al. did not find a significant difference in selfconfidence between the 2 groups [27]. The results were also mixed at the K2 level. Although RCDP was associated with decreased times to critical tasks [15,23], significantly higher checklist scores [15,16,23], and significantly higher simulation team assessment tool (STAT) human factors sub-section scores [22], there was no significant difference in overall STAT score improvement [22,27], and no difference in MCQ scores [15] between the two groups. When examining 3-4-month recall, checklist scores decreased in both groups with the RCDP groups having greater decreases in scores compared to the traditional simulation groups [15,23].

Discussion
To the best of our knowledge, this is the largest review to date that examines and summarizes the RCDP literature. This scoping review consists of 23 published studies, 21 more than the previous review [7]. The studies were heterogeneous in design, participants, simulation types, and outcome measures. Implementation strategies also varied. Interestingly, all studies referenced Hunt et al. for their definitions of RCDP [5]. They included the core techniques of RCDP, including within-event debriefing, deliberate practice, clear learning objectives, formative assessment, and pause-feedback-replay loops. However, the variations in the use of a needs assessment, pause timing, rewinding or restarting, case progression, and escalating difficulty make it difficult to draw general conclusions about RCDP. As RCDP becomes more popular, it is important that implementation and simulation design become standardized to better research its effects.
This review found that RCDP training is associated with multiple positive outcomes including learner satisfaction, increased confidence, knowledge and skills acquisition, and time to critical tasks. RCDP allows learners to receive timely direct feedback and multiple opportunities to "do it right"; they can incorporate feedback and try again immediately. Breaking cases down into "smaller chunks" maximizes learning without cognitive overload [13]. This approach likely works well for skills that are algorithmic, have defined protocols, or clear scripts. Although all studies reported positive K1/K2 outcomes, most studies did not compare RCDP to other interventions so their conclusions may be from the education itself or from the instruction technique. However, the K2 outcomes demonstrated, such as the direct observation of key tasks, may contribute to the overall competency of the learner [31]. Despite  [7,9,10]. This implies that RCDP simulation can be successfully employed to teach nontechnical abilities.
Since there is limited research on the impact of RCDP training outside of the learning environment, future studies should examine how RCDP interventions translate to clinical practice. This review examined comparisons between interventions with RCDP to standard curriculums and traditional simulation. It appears that adding RCDP to standard curriculums is overall beneficial, leading to multiple improvements in K2 outcomes. However, when compared to traditional simulation, the results are mixed. While some studies showed immediate improvements, three of the five studies that compared RCDP to traditional simulation did not show significant differences in all K2 outcome measures [15,22,27]. Furthermore, RCDP groups had greater decreases in scores at 3-4-month follow-ups compared to the traditional simulation groups even though these studies both showed significantly higher scores in the RCDP groups immediately after training [15,23]. This suggests that although RCDP has an immediate advantage in knowledge application, knowledge retention has not been demonstrated. Despite being able to apply feedback immediately with repetition to reinforce learned skills, it is possible that long-term "muscle memory" is not being created with RCDP. One study recommended that RCDP training should occur every 3 months to maintain skills and avoid skills decay observed at 6 months [28]. More studies are needed to compare RCDP to other types of SBME, and specifically how skills retention with RCDP compares with other modes of instruction. This review has not found that RCDP is superior to traditional simulation.
There are limitations to this review. Grey literature was not searched, and conference abstracts and poster presentations were excluded. Studies were also limited to the English language. Critical appraisal of study quality was not completed. Most studies were small and at a single site. Many of the studies had no comparisons, and the validity and reliability of the outcome measurement tools were not well described. It is possible that their outcomes may be related to other confounders.