Improving Self-Confidence of Military Medical Providers With Joint Procedure Simulation: A Pilot Study

ABSTRACT

Introduction

Military medical providers are a unique population that encounter different environments across the world. From hospital clinics to war zones, these providers must perform procedures and rely on their training and skill to help their patients. This pilot study aimed to assess the self-confidence of military medical providers performing joint aspiration and injection before and after a simulation workshop in both clinical and austere settings.

Methods

In 2016, 25 military physicians from various military facilities participated in a 1-hour knee arthrocentesis and injection and shoulder injection workshop. Education was provided on the knee and shoulder anatomy and various approaches to performing the procedures before the hands-on portion of the workshop. Surveys assessing self-reported confidence levels by performing the procedures in the clinic and austere settings were completed before and after simulation training.

Results

The results were analyzed and grouped based on the provider experience level, simulation environment, and specific procedure performed. There was a statistical significance seen in the shoulder arthrocentesis group, which included all participating providers, with a P-value of <.01 in the clinic setting and a P-value of <.001 in the austere setting. In the knee aspiration simulation, there were also improvements in the provider confidence, but it was not statistically significant with P-values of .36 and .14 in the clinical and austere settings, respectively.

Conclusion

Simulation training can lead to increased medical provider self-confidence in performing musculoskeletal joint aspirations and injections in both clinic and austere settings. The military medicine demographics have had little research in joint injections and provider confidence to date. This pilot study was one of the first to evaluate this unique population. The methods used in this study, and the positive data collected on provider confidence, can be used in larger studies, encompassing other medical providers to increase the confidence of providers throughout various fields of medicine.

INTRODUCTION

Musculoskeletal disorders are common complaints in the primary care setting nationwide, accounting for upward of 23% of appointments.¹ Studies focusing on young- to middle-aged adults are reporting upward of 60% of appointments being related to musculoskeletal complaints.² The U.S. military is well within this age range, with the average age of enlisted service members being in the late 20s and officers being in their early 30s. In the military, musculoskeletal pain and injuries are one of the leading causes for lost duty time and disability.^3,4 In the deployed setting, military members report musculoskeletal pain and injuries, with knees being reported to have more injuries than shoulders in one study completed by soldiers deployed to Afghanistan.⁵ In a 2017 analysis of active duty soldiers in the Army, more than half were noted to have at least one injury, with overuse injuries being a significant proportion of these joint and musculoskeletal injuries.⁶ This study also found knees and lower back injuries to be more common than shoulder injuries. In the civilian counterpart, arthritic complaints alone, related to age and joint overuse leading to increased inflammation over time, have an annual average of 46.4 million according to 2003–2005 NHIS surveys.^4,7 An expedited diagnosis can be made through the arthrocentesis of synovial fluid followed by a therapeutic intra-articular steroid injection if indicated. Arthrocentesis has been found to be useful in identifying gout, septic arthritis, and pseudogout and categorizing the fluid as inflammatory or noninflammatory processes, all of which help in diagnosis and treatment.⁸ Unfortunately, primary care providers (PCPs), which can include physicians, physician associates, and nurse practitioners, often refer this select group of patients for further evaluation by a subspecialist provider. This is likely due to the lack of provider confidence in their ability to perform joint injection and aspiraton, secondary to the lack of training or infrequent use in the outpatient setting. Specifically looking at physicians with an internal medicine residency training, the American Board of Internal Medicine does not require joint injections or aspirations as a graduation requirement.⁹ Internists going through this training are required to understand and explain the procedure to patients. Individual institutions are then allowed to adjust their own “in-house” requirements for graduation. With no formal requirement for arthrocentesis and injections, there have been a diminished number of procedures being performed by internists coming out of residency who are in clinical practice.¹⁰

U.S. military providers are deployed all over the world, encountering different environments that are oftentimes a significant distance from the nearest hospital. In the austere or deployed setting, poor confidence in performing an arthrocentesis can be costly to the medical system because of the logistical difficulty in obtaining a subspecialist for a procedure and transport of the patient. Delayed or missed diagnosis of potentially destructive pathologies, such as septic arthritis, may also lead to catastrophic results, including joint destruction, sepsis, or death, and a reduction in the fighting force.

The objective of this pilot study was to assess military medical providers’ procedural confidence in their ability to perform a shoulder injection and knee joint injection and aspiration before and after simulation training in both clinical and austere settings. Knee and shoulder joints were chosen after a literature search revealed that these are some of the more common injuries sustained by U.S. military members. Clinical and austere environments were assessed in order to determine true independent procedural confidence in remote settings as compared to a large hospital or clinic environment.

METHODS

In 2016, a total of 25 military physicians with an internal medicine training participated in a 1-hour musculoskeletal simulator workshop. There was only one session provided that all 25 participants attended at once. The 25 medical providers included a mix of 13 staff internists and trainees along with 12 subspecialists. Of the staff internists and trainees group, there were 10 resident trainees and 3 graduated staff internists. The 12 internal medicine subspecialists included 2 cardiologists, 3 endocrinologists, 2 pulmonologists, 2 gastroenterologists, 1 infectious disease specialist, 1 rheumatologist, and 1 allergist. The workshop was led by a board-certified rheumatologist and consisted of identifying anatomical landmarks and conducting a needle aspiration and injection of the knee and a shoulder injection. The introduction to the course began with the importance and use of joint aspirations and injections. Education was given on the types of needles that may be used for the different types of aspirations and injections, as well as the medications that can be given for the injections. Anatomical landmarks were identified both on the simulation models and on the attendees themselves. Each participant was required to practice both on themselves and with at least one partner to identify shoulder and knee anatomy and landmarks for joint aspiration and injection. The knee model focused on medial and lateral knee anatomy and approach, with appropriate needle placement sites around the patella identified and confirmed by the course instructor. For the shoulder model, anterior and posterior anatomy was identified, with a focus on identifying anatomical landmarks such as humeral head, distal clavicle, and coracoid process. Anterior and posterior approaches were taught, with a focus on injections into the glenohumeral joint. After practicing and identifying anatomical landmarks, two simulation models were used to practice joint aspiration and injections. The models used were the Knee Aspiration Model #70013 and Shoulder Injection Model #30010 by LIMBS & THINGS. Under the guidance of the teaching rheumatologist leading the session, each participant was individually monitored identifying landmarks, performing a knee aspiration and injection first, followed by a shoulder injection. Injections were given with normal saline. Proper technique was a focus. After being monitored for proper technique for aspiration and injections, participants were allowed to practice several times under further guidance, with individualized guidance provided as needed. All aspirations and injections were completed on the models by LIMBS & THINGS. Appropriate needle placement on the knee aspiration was confirmed as a result of the model containing a synovial gel mimicking agent that would be seen in the needle. Appropriate needle placement for the shoulder injection was confirmed by the shoulder simulation model having a built-in sensor that would turn on a light when the needle was in the correct joint space. No human models were used for practice. Ultrasounds were not used in this session. The 1-hour session was guided by participants’ questions and answers, with the majority of the time being hands-on practice. Notably, the competency of providers performing the procedure was not directly measured. Instead, a major focus was on proper technique, hands-on practice, and reported provider self-confidence. All participants that completed the course received one-on-one feedback from the instructor to successfully complete at least one knee arthrocentesis and injection and shoulder injection.

Confidence was measured presimulation and postsimulation using an identical survey [see supplement] with a 10-point Likert-type scale with 1 representing the lowest confidence and 10 representing the highest confidence. Figure 1 is a summarized list of the statements that participants completed using this scale (see supplement for the original survey). Data were collected for both a clinic environment and a deployed setting. Results were statistically analyzed using a Wilcoxon rank sum test in order to assess if simulation training had a statistically significant improvement in clinicians’ overall confidence. The total group was divided into a combination of staff internists with resident trainees group [group 1] and a medical subspecialists group [group 2] to determine if there was a difference in confidence.

FIGURE 1.

Joint simulation survey summary of statements. A summarized version of the 10 statements included in the survey given to the participants before and after the 1-hour knee and shoulder joint arthocentesis training session. The statements were measured on a Likert scale, with 1 corresponding to “strongly disagree” and 10 to “strongly agree.” For the original survey given to the session’s participants, please see the supplementary material.

A postsimulation survey using a 10-point Likert scale for the assessment of the utility of simulation training for joint aspiration/injection of the shoulder and knee to be applied in the clinic and austere environment was also evaluated (see supplement for original survey).

RESULTS

Group 1 consisted of a mix of 13 internal medicine staff and residents; group 2 consisted of 12 medical subspecialists. All of the 25 providers participated in the prejoint simulation surveys [n = 25] and 12 internal medicine staff and trainees and 11 subspecialists participated in the postjoint simulation survey [n = 23]. The two providers (two endocrinologists) that did not participate in the postjoint simulation survey left the class early because of a time conflict with another course. The change in confidence was measured in three categories consisting of total participants, group 1, and group 2 with confidence assessed in the clinic and austere settings. Data were assessed for statistical significance and are presented as mean ± for standard deviation.

All participants as a whole demonstrated a significant statistical change during the shoulder simulation in the presimulation to postsimulation training confidence scores (Table I). The clinic setting revealed a presurvey score of 4.6 ± 2.9 with a postsimulation score of 6.8 ± 2.1 [P < .01]. In the austere setting, a presurvey score of 4.6 ± 2.8 and postsurvey score of 7.3 ± 2.0 [P < .001] was observed. Improvement in confidence was observed in the knee simulation but was not statistically significant. In the clinic setting for the knee simulation, presurvey score was 6.5 ± 3.1 and postsurvey score was 7.7 ± 1.7 [P = .36]. The austere setting had a presurvey score of 6.5 ± 3.0 with postsurvey score of 8.0 ± 1.6 [P = .14].

TABLE I.

Comfort Levels Measured in All Participants on the Knee and Shoulder Simulations in Both Clinic and Austere Environments

All participants	Preworkshop mean comfort level, mean (±SD)	Postworkshop mean comfort level, mean (±SD)	P value
Clinic environment
Shoulder	4.6 (±2.9)	6.8 (±2.1)	<.01*
Knee	6.5 (±3.1)	7.7 (±1.7)	.36
Austere environment
Shoulder	4.6 (±2.8)	7.3 (±2.0)	<.001*
Knee	6.5 (±3.0)	8.0 (±1.6)	.14

Preworkshop and postworkshop mean comfort levels are reported on a scale of 1–10, with 10 being the highest confidence.

*

Indicates clinical significance.

Group 1 presimulation to postsimulation workshop confidence scores for the shoulder in the austere setting demonstrated statistical significance, but not in the clinical setting (Table II). The clinic environment revealed presurvey results of 5.2 ± 2.6, and postworkshop survey results of 6.9 ± 2.1 [P = .10]. The austere environment had a presurvey score of 5.2 ± 2.6 with a postsimulation score of 7.5 ± 1.9, [P < .01]. The knee simulation demonstrated confidence improvement in both environments, but the results were not statistically significant. The knee simulation in the clinic environment resulted in a preworkshop survey score of 7.5 ± 2.6 and postworkshop score of 7.7 ± 1.6 [P = .78]. In the austere setting, the knee workshop revealed a presurvey score of 7.3 ± 2.6 and postsurvey score of 8.0 ± 1.4 [P = .69].

TABLE II.

Comfort Levels Measured in Internists and Resident Physicians on the Knee and Shoulder Simulations in Both the Clinic and Austere Environment

Internists and resident physicians	Preworkshop mean comfort level	Postworkshop mean comfort level	P value
Clinic environment
Shoulder	5.2 (±2.6)	6.9 (±2.1)	.10
Knee	7.5 (±2.6)	7.7 (±1.6)	.78
Austere environment
Shoulder	5.2 (±2.6)	7.5 (±1.9)	<.01*
Knee	7.3 (±2.6)	8.0 (±1.4)	.69

Preworksop and postworkshop mean comfort levels are reported on a scale of 1–10, with 10 being the highest confidence.

*

Indicates clinical significance.

Group 2 demonstrated a statistically significant change comparing the before and after surveys from the shoulder simulation in the austere environment (Table III). All other surveys within this group of medical staff were not statistically significant. The deployed setting shoulder presurvey score was 4.1 ± 3.1, with a postsimulation score of 7.1 ± 2.1 [P < .05]. The clinical setting presurvey score was 4.0 ± 3.1, with a postsurvey of 6.6 ± 2.3 [P = .05]. The knee simulation found improvement but not statistically significant change. Clinical confidence had a pre-knee workshop score of 5.5 ± 3.5, with a postsurvey score of 7.8 ± 1.9 [P = .13]. The austere setting had a presurvey score of 5.7 ± 3.3, with a postsurvey score of 8.0 ± 1.8 [P = .69].

TABLE III.

Comfort Levels Measured for Medical Subspecialists on the Knee and Shoulder Simulations in Both the Clinic and Austere Environment

Medical subspecialists	Preworkshop mean comfort level	Postworkshop mean comfort level	P value
Clinic environment
Shoulder	4.0 (±3.1)	6.6 (±2.3)	.05
Knee	5.5 (±3.5)	7.8 (±1.9)	.13
Austere environment
Shoulder	4.1 (±3.1)	7.1 (±2.1)	<.05*
Knee	5.7 (±3.3)	8.0 (±1.8)	.10

Preworksop and postworkshop mean comfort levels are reported on a scale of 1–10, with 10 being the highest confidence.

*

Indicates clinical significance.

A final analysis of the 25 clinicians’ opinions on the utility of simulation training was conducted, with a collective score of 9.3 ± 0.90. Group 1 yielded a score of 9.4 ± 0.80 and group 2 yielded a score of 9.1 ± 0.90.

DISCUSSION

Simulation training is a growing technique used in many fields in which participants can practice real-life skills on models or in virtual reality (VR) platforms. It has become part of medical training and has improved the clinical comfort and confidence of medical students, primary care providers, and various specialists.^11–16 Going beyond comfort level, the quality of procedures based on simulation training versus hands-on patient care training has also shown to be equivalent.¹⁷ This suggests that simulation training can be used instead of direct practice on patients to improve self-confidence and competency. In turn, improved self-confidence and competency may lead to better patient outcomes and, depending on the context, physician safety in performing certain procedures.¹⁵

Joint simulation training improves medical knowledge and technical skills of training physicians. The benefit of simulation and VR training likely comes from the repetition of the skills, which, in turn, improves physician competence in performing the particular procedure.¹⁸ In the 1-hour simulation session discussed in this article, the participants were evaluated based solely on self-confidence rather than competency. This was chosen because of time constraints, variability in their background training, and inability to have long-term follow-up on all the participants. However, each participant received one-on-one time with the instructor during the 1-hour session in which their techniques were refined to successfully perform at least one knee arthrocentesis and injection and one shoulder injection. Regardless of the limitations of this study, there currently is a growing area of research of provider self-confidence improvement through simulation training.

Medical professionals, such as nurses, physician associates, and physicians, are using simulation training from the time they are in their respective trainings until they are actively in practice.^15,19 This applies both in the military and civilian medical fields. However, there have been no studies dedicated to the thousands of military medical providers who encounter a variety of environments and patients. Performing procedures well is critical in maintaining good patient care. Recent studies have shown that procedural volume within military treatment facilities can vary, which is an opportunity for simulation procedures and VR to be used and enable providers to maintain procedural readiness.^20,21 This pilot study aimed to evaluate changes in confidence for procedural operation, with a particular focus on military medical providers, in the hopes that improved confidence would further support the use of and need for simulation training in all military facilities both in clinic and austere environments. By being able to perform arthrocentesis and injections in the field, the fighting force may be better preserved with no need to send patients to a higher echelon of care.

Evaluating the self-confidence changes in austere environments was a unique aspect of this pilot study not previously done. Military medical providers are placed in combat situations far from the clinic and training environments, which presents many challenges. Because of the lack of resources and distance from supporting hospitals and clinic settings, military medical providers must be willing and capable to perform procedures. These providers must be willing to take risks in austere environments, or else the patient may suffer. In comparison to their civilian medical provider counterparts, these austere environments with little access to formal medicine are likely less frequently encountered. In our select group of medical providers studied, self-confidence increases were seen throughout the study for both knee and shoulder aspirations and injections in both austere and clinic environments, albeit some were not statistically significant. The cause for the lack of statistical significance was likely the small population studied in the 1-hour workshop. Future studies are planned to involve a larger group of participants of varying backgrounds. Future studies will also plan on having a follow-up workshop to see if self-confidence is sustained. Outside of the training environment, future studies will need to encompass in-the-field testing on joint models for military medicine. These studies will likely provide significant improvement to overall self-confidence and help many soldiers in austere environments abroad.

Within the data collected in this study, the shoulder simulation showed consistent improvement in presimulation to postsimulation self-confidence, regardless of the group being measured. The authors would like to suggest that the improvement in confidence on shoulder procedures versus knee procedures is likely related to the lack of procedural volume on shoulders as compared to knees. This would explain the lower overall scores for preworkshop mean comfort levels for the shoulder simulation as compared to the knee simulation (Tables I–III). Symptomatic knee osteoarthritis and injury are more common than shoulder arthritis and injury. This is likely related to age, obesity, and, in the military, increased exercise as compared to the general population.^22,23 For military medical providers, this leads to more frequent exposure and need for procedures on knees rather than shoulders, which is reflected in the higher provider comfort levels overall found in the current study. This idea is supported by studies in deployed settings of soldiers reporting more injuries in the knees as compared to shoulders, along with overall outpatient visits being reported more frequently for knees over shoulder pain.^5,6

Another possibility for the difference in comfort levels between shoulder and knee procedures could be the difference in anatomy and provider comfort identifying landmarks. Other studies comparing shoulder and knee injections have shown less accuracy in shoulder procedures than knee injections.²⁴ A potential cause for this is the different approaches that can be taken for a shoulder injection (anterior versus posterior) as compared to the anatomy of the knee being more symmetric, limiting the different procedural approaches.²⁵ Models used in the workshop were anatomically correct, but it is not uncommon for patients to present with different anatomies based on natural variation or previous injury or procedures, further adding to the variability in the preworkshop and postworkshop comfort levels seen. The last possibility explored for the differences in preworkshop and postworkshop comfort levels is the number of procedures that providers had performed in the past. Although this was not directly measured, higher-trained subspecialists likely skewed the data with their more time in practice, more opportunities for procedures, and potentially more hands-on specific jobs as compared to the medical trainees.

The final observation collected from this pilot study was that all military providers found simulation to be a useful tool. This supports the desire for more training and further studies to evaluate the usefulness of simulation workshops. Studies have shown that teaching practical skills in knee aspiration, and likely other procedures, early in medical careers has long-term benefit in improving clinician skills.¹⁴ With the ever-growing body of evidence pointing to the usefulness of teaching practical skills hands-on, the hope will be that all physicians universally will receive training like that presented in this research.

CONCLUSIONS

Simulation training is a useful and desired tool by military medical providers. This unique population faces resource-limited environments that providers outside of the military may also encounter on a daily basis. The information collected in this pilot study can be extrapolated and possibly applied to medical professionals in similar environments, especially other members of the medical team that may perform similar procedures, such as physician associates. Knowledge obtained from this study in regards to improving provider confidence in performing procedures can be used in guiding further training, education, and research studies with other medical providers, especially those in rural and austere environments, to further medical education and patient care as a whole.

ACKNOWLEDGMENTS

V.M. has conducted significant research about the case, a significant amount of planning and writing up of the case, and the interpretation of the data; Z.A. has discussed the planning and reporting of the case with inputs in image formatting as well as image interpretation; J.M. contributed to editing, review, and table and figure formatting. K.S.E. has contributed with editing, review, and validation. J.R. has contributed to planning, research, and idea of the case and made a significant contribution to data acquisition and image analysis. All authors had access to the data and approved the final manuscript.

SUPPLEMENTARY MATERIAL

SUPPLEMENTARY MATERIAL is available at Military Medicine online.

FUNDING

No funding was received for this study.

CONFLICT OF INTEREST STATEMENT

The authors declare that they have no competing interests.

ETHICS APPROVAL AND CONSENT TO PARTICIPATE

Protocol #218111 was approved by the hospital institutional review board. Consent was obtained from all providers for participation.

Author notes

The views expressed in the abstract/manuscript are those of the authors and do not reflect the official policy or position of the Department of the Army, DoD, or the U.S. Government.