Ultrasound Curriculum Implementation into a State-Wide Family Medicine Clerkship

Abstract

Introduction

While point of care ultrasound (POCUS) integration into clinical clerkships provides unique educational experiences for students, there are barriers to implementation, particularly in a distributed campus medical school model in clerkships where the faculty do not often perform POCUS, like family medicine (FM). The objective of this paper is to describe the implementation and evaluation of a POCUS curriculum in an FM core clinical clerkship in a state-wide medical school campus model.

Methods

Seventeen Philips Lumify Ultrasound Systems were used in 20 clerkship sites with the requirement that students obtain abdominal aortic and inferior vena cava (IVC) images on patients evaluated during their rotation. Images were de-identified, transmitted to a university cloud–based storage account, and scored by medical school ultrasound faculty.

Results

Students were able to obtain adequate images of the aorta and IVC without direct ultrasound–trained faculty at the performance site. Of the 183 students, 119 (65%) were able to successfully submit images for scoring with failure to upload to the cloud-based storage account as the most common reason students were unsuccessful (42%). The majority of students (62%) scored in the top quartile of image quality scoring with the percentage of those scoring in the top quartile improving during the academic year from 57% in the first four rotations to 67% in the last four rotations.

Conclusion

Barriers to implementation of a POCUS curriculum into a FM clerkship in a distributed campus medical school model can be challenging due to equipment availability and issues around the successful transmission of images. These challenges can be overcome however with sufficient attention to implementation design that includes equipment sharing protocols and technical options that enhance the ease of image transmission.

Introduction

Point of care ultrasound (POCUS) is the use of ultrasound at a patient’s bedside, typically for diagnostic and procedural purposes. POCUS is a powerful tool for the clinician as it can be employed in a variety of clinical environments and can be safely utilized by any properly trained healthcare provider, including mid-level providers, nurses, and medical students [1]. A large body of evidence demonstrates that using POCUS improves clinical outcomes, reduces failure and complication rates during procedures, rapidly narrows differential diagnoses, shortens time to definitive treatment, lowers cost, and reduces the use of computed tomography (CT) images thereby reducing patient exposure to radiation [2–5]. According to the American Academy of Family Physicians, POCUS is the biggest advance in bedside diagnosis since the advent of the stethoscope 200 years ago [6].

In recent years, ultrasound has been integrated into medical school education, particularly in the pre-clinical years, during which time POCUS has been used in conjunction with instruction in anatomy and physiology and in the development of physical exam skills to aid students in understanding and reinforcing these concepts [7]. Incorporation into clinical clerkships allows students to experience hands on ultrasound evaluation at a patient’s bedside, reinforcing concepts learned in the pre-clinical years in patients with pathology and also aiding students in understanding the clinical medicine concepts of the rotation.

Research has demonstrated that ultrasound imaging of certain pathologies can be learned successfully within a discreet time frame as is the case with a clinical clerkship [8–10]. Kobal et al., for example, demonstrated that after completing a brief training in cardiac ultrasound, medical students were able to correctly diagnose 75% of cardiac pathologies among 61 patients with known cardiac disease [9]. Krause et al. showed that after completing 1 hour of training on the FAST exam, third-year medical students are able to increase the accuracy and speed in which they can identify and interpret image results [10]. These results show that when provided training and access to clinical ultrasound, undergraduate medical education (UME) students can successfully learn and apply these skills to patient care [8].

Incorporation of POCUS into the clinical rotations of medical school has different barriers and challenges than incorporation into pre-clinical curricula. In a pre-clinical setting, expensive resources such as ultrasound equipment can be shared among many students, and ultrasound education faculty have large groups, in one place, to teach and ultrasound trained faculty can be identified and utilized in an educational program. Also, processes such as giving feedback on acquired images, didactic education on how to use POCUS, and assuring appropriate use on patients can be easily instituted as the director has control over who and why a patient is scanned. When students are on clinical rotations, in contrast, they are taught by faculty trained in the specialty of the rotation. In many specialties, POCUS has little penetration, meaning there is no ultrasound equipment for the student to use, the faculty are unfamiliar with POCUS, and POCUS is not commonly used in their clinical practice. In a distributed campus model, where students utilize clinical sites that are unaffiliated with the medical school’s hospital system and are often separated geographically, these barriers are magnified. Providing equipment not owned by the practice, transmission of images for grades, and delivering didactic education are all made much more difficult.

According to the Liaison Committee on Medical Education (LCME), all students must have an equivalent rotation experience regardless of the rotation site. Thus, to have POCUS as a requirement of the rotation, all students have to have access to similar resources, including both educational instruction and equipment access. Having sites that are disparate in equipment and educational opportunities is not acceptable.

Beginning in 2016, integration of a POCUS curriculum into clinical clerkships began with the emergency medicine (EM) and obstetrics and gynecology (OB/GYN) clerkships. These clerkships were selected for initiation of an integrated ultrasound curriculum as both of these clerkships generally have ultrasound equipment and ultrasound trained faculty. The family medicine (FM) clerkship integration was subsequently started in 2017. The objective of this paper is to describe the implementation and evaluation of an ultrasound curriculum into a FM core clinical clerkship in a distributed campus UME model.

Methods/Implementation

The medical school participating in this research employs a distributed campus model, such that UME students (approximately 240 per year) complete their training at five campuses and more than 200 clinical sites across the state. The dean of each campus coordinates with regional hospitals, outpatient facilities, and physicians to establish clinical clerkship sites at which the students rotate during their third and fourth years of medical school. Each campus has a clerkship director who oversees the individual sites at which students rotate, taking care to ensure that students’ clerkship experiences align with those at other campuses, including adhering to standardized goals and objectives for the clerkship. Each site also has a director who is responsible for managing the logistics of the site, including providing adequate resources for rotating students. In addition, all sites use a standardized format for each rotation including weekly quizzes, didactic sessions conducted via teleconferencing, and a comprehensive student evaluation process at the conclusion of the rotation. This research was determined exempt by the institution’s institutional review board (IRB) under Health and Human Services Exempt Category 1.

The medical school curriculum office assigns students to one of up to 21 individual sites across the state for the FM clerkship, ranging in size from small, single-physician clinics to mid-size clinics, to a tertiary care academic medical center. In planning for the POCUS integration into the FM clerkship, each site was surveyed to determine the feasibility of using POCUS, taking into consideration such details as equipment availability, faculty familiarity with ultrasound, frequency of student rotation at the site, and status of Wi-Fi connectivity (which would enable students to transmit their images to the hub site for grading). While every site was found to have Wi-Fi capabilities, only one of the sites had an US machine in house and a FM physician who was familiar and comfortable with using POCUS. Because the in house US machine was not portable and not a Lumify (and therefore did not provide the same experience the rest of the students had), the results from these students are not included in the analysis. One site chose not to participate in the project. Students assigned to this site performed their assignment at another site in close proximity and had access to the same equipment, training, and assessment as all other students in the project.

In order to address the issue of lack of equipment at the remaining 20 locations used during the academic year, the medical school’s Center for Ultrasound Education (CUSE) purchased 17 android tablet app-based Philips Lumify Portable Ultrasound systems (S4-1 transducer) for students to use during their FM rotation. The Lumify app was programmed to upload images to the institutional cloud–based storage account (Box) and included educational videos and an ultrasound ebook on POCUS topics. Included with the android tablet was a PDF-based patient consent informing the patient that the images obtained by the student were for educational purposes only and not for diagnostic decision-making. This consent process by students is part of standard educational practice during clerkships. The CUSE was the hub site, coordinating the distribution of the ultrasound systems, collecting the submitted images, and providing evaluation and feedback to the student and clerkship director.

As part of POCUS integration in the clinical curriculum, the FM clerkship was assigned abdominal aorta imaging as part of abdominal aortic aneurysm screening (AAA) and inferior vena cava evaluation (IVC) for intravascular volume assessment. During the pre-clinical years, ultrasound was an integrated component of the year-long physical diagnosis modules of first and second year and included instructional training on these exams (AAA and IVC). During the clerkship orientation at the end of second year, prior to staring clerkship rotations, students were given the opportunity to review all of the ultrasound exams that would be required in the clerkship rotations including hands-on practice with standardized patients as well as instruction in how to use the Philips Lumify Portable Ultrasound system. This project was determined exempt by the institutional review board (IRB) under Health and Human Services Exempt Category 1. Data used for research purposes were first deidentified by an honest broker as instructed by the IRB. Students collected written electronic consent from the patient acknowledging that the ultrasound exam was only for educational purposes and that no diagnostic information would be obtained that would benefit the patient. If concerning medical issues are found by the student, the physician in charge will ensure that appropriate follow-up is scheduled for the patient.

Students were instructed to obtain ultrasound images of the aorta and the cine-loop of the IVC on at least three patients evaluated during clinical duties. Clinical education regarding AAA screening guidelines and fluid hydration was incorporated into the patient screening module as part of the weekly 2.5-hour videoconference students participated in during the FM clerkship. POCUS objectives and tasks were a required component of the clerkship, but student performance was not included in the overall FM grade during the initial year of the integration. Each student had access to the Lumify system for 3 weeks of the 6-week rotation.

Ultrasound Image Evaluation

Ultrasound images were transferred electronically from the Lumify app to the institutional cloud–based storage account (Box). As a quality assurance check on the process of submitting images electronically to the cloud-based storage account, students recorded in the learning management system (LMS) that the images had been obtained and transmitted. No patient identifiers or history were included with the images, and students were instructed to enter their name as the patient’s name in order to identify the images and link them back to the student for feedback and assessment purposes. A single faculty member of the CUSE provided comments and a standardized score of the images according to a grading rubric (see Table 1); the same grading rubric was used during the physical diagnosis ultrasound assignments in the first 2 years of medical school. The rubric focused on image acquisition and organ identification and provided a score that was used to monitor student progress as a group. This score was not part of the grade the student received for the rotation. Submitted images were evaluated on image quality including appropriate labeling of organs, use of gain and depth, proper probe selection, and correct orientation and not for pathology or diagnosis.

Table 1.

Standardized feedback rubric

Aorta/IVC point-of-care ultrasound exam
25 points	Score	12 points	Score	0 points	Score
Acquires and correctly identifies aorta (above SMA) with anterior/posterior measurement on trans ultrasound image (label image: AORTA PROX)		Acquires and correctly identifies aorta (above SMA) with anterior/posterior measurement on trans ultrasound image with limited errors (label image: AORTA PROX)		Unable to acquire and correctly identify aorta (above SMA) with anterior/posterior measurement on trans ultrasound image (label image: AORTA PROX)
Acquires and correctly identifies aorta (at SMA) with anterior/posterior measurement on trans ultrasound image (label image: AORTA MID)		Acquires and correctly identifies aorta (at SMA) with anterior/posterior measurement on trans ultrasound image with limited errors (label image: AORTA MID)		Unable to acquire and correctly identify aorta (at SMA) with anterior/posterior measurement on trans ultrasound image (label image: AORTA MID)
Acquires and correctly identifies aorta (above bifurcation) with anterior/posterior measurement on trans ultrasound image (label image: AORTA DISTAL)		Acquires and correctly identifies aorta (above bifurcation) with anterior/posterior measurement on trans ultrasound image with limited errors (label image: AORTA DISTAL)		Unable to acquire and correctly identify aorta (above bifurcation) with anterior/posterior measurement on trans ultrasound image (label image: AORTA DISTAL)
Acquires and correctly identifies IVC with hepatic vein on long ultrasound VIDEO CLIP (label IVC long)		Acquires and correctly identifies IVC with hepatic vein on long ultrasound VIDEO CLIP with limited errors (label IVC long)		Unable to acquire and correctly identify IVC with hepatic vein on long ultrasound VIDEO CLIP (label IVC long)

Total points _____/total points is the sum of your scores (maximum score = 100)

Results

In general, in each rotation, the majority of students (119/183, 65%) were able to successfully complete the assignment (see Table 2) as defined by submitting all four images to the cloud-based storage account and uploading the appropriate documentation to the learning management system. Ultimately, 35% of students were not able to complete the assignment (see Table 2).

Table 2.

Students able and unable to complete the assignment

	No. of students	No. of able to complete	No. of unable to complete
Rotation 1	19	14	5
Rotation 2	25	9	16
Rotation 3	20	11	9
Rotation 4	28	19	9
Rotation 5	21	16	5
Rotation 6	22	20	2
Rotation 7	24	14	10
Rotation 8	24	16	8
Total	183	119	64
Percentage	100%	65%	35%

The reasons students were unable to submit their images successfully were the following: 42% (27/64) failed to upload the images to the cloud-based storage account from the Lumify app, 19% (12/64) did not label the images appropriately with their name before transmission to the cloud-based storage account and therefore the images could not be assigned to the student, and 5% (3/64) opted to show the images to their preceptor directly and did not submit to the CUSE for evaluation as they had been instructed (see Table 3). These three instances (when the students opted to show their images to their preceptor) occurred during the first rotation, and this issue was corrected by providing more explicit instructions for image submission before the start of rotation 2.

Table 3.

Reasons students were unsuccessful with image submission

	No. of students unsuccessful with image submission	No. of sent to LMS but no images in account	No. of unidentifiable images (failed to use their name as patient name)	No. of showed to preceptor not sent to LMS or account
Rotation 1	5	2	0	3
Rotation 2	16	8	8	0
Rotation 3	9	6	3	0
Rotation 4	9	8	1	0
Rotation 5	5	0	0	0
Rotation 6	2	0	0	0
Rotation 7	10	0	0	0
Rotation 8	8	3	0	0
Total	64	27	12	3
Percentage	100%	42%	19%	5%

Of the 119 students who were able to submit images to be scored (see Table 4), the majority (62%) scored in the top quartile (76–100% range). A combined 32% scored in the second and third quartiles (16% in each quartile), and 6% scored in the bottom quartile.

Table 4.

Performance of students who successfully submitted images by quartile. Percentages rounded to the nearest half percent

	No. of students	0–25%	26–50%	51–75%	76–100%
Rotation 1	14	3	4	1	6
Rotation 2	9	2	2	1	4
Rotation 3	11	0	4	2	5
Rotation 4	19	0	1	3	15
Rotation 5	16	0	1	2	13
Rotation 6	20	1	3	6	10
Rotation 7	14	1	1	2	10
Rotation 8	16	0	3	2	11
Total	119	7	19	19	74
Percentage	100%	6%	16%	16%	62%

Discussion

The goal of the family medicine POCUS implementation is to train students to become better future physicians by integrating the use of ultrasound as a clinical tool into core clinical rotations. Integration into the clinical rotations (while arguably the most difficult component of the medical school curriculum into which to integrate such a curriculum) was recognized as the most important as it was observed that there is a decline in sonography skills among medical students during the clinical years. By providing a physical task, such as performing an AAA screening, POCUS provides a cognitive scaffold on which to build and remember clinical medical knowledge such as screening guidelines and the pathology of aneurysms.

The barriers to implementation of a POCUS curriculum to clerkships are substantial [11]. Prior publications, such as the University of South Carolina’s The evolution of an integrated ultrasound curriculum (iUSC) for medical students: 9-year experience, discuss the integration of ultrasound into the UME curriculum [12]. However, this model focuses on the integration of ultrasound into a single campus system, one where the medical school has governance over the sites used to train medical students. Many medical schools, however, utilize multiple campuses and/or clinical sites, making a standardized ultrasound curriculum more challenging to implement. Moreover, multiple campuses and clinical sites often preclude the standardization of equipment which introduces technology barriers for image storage and transmission. Further, since many of the clinical faculty in a distributed campus model are not members of the medical school associated health system, issues of liability with students using ultrasound are more of a barrier than in those models covered legally by a single entity.

The Philips Lumify Ultrasound System was chosen specifically to address the barriers of standardization of equipment, image storage, and image transmission. This ultraportable ultrasound system is android-based, an operating system most medical students are familiar with using or can easily learn, unlike many radiology-focused ultrasound systems. This app-based system includes multiple mechanisms for traditional image transmission such as PACS integration and for non-traditional image transmission including cloud-based storage sites like Box and e-mail, all of which are HIPAA (Health Insurance Portability and Accountability Act) compliant. This is an important feature as a proprietary cloud-based image archive was not needed, which reduced the cost of implementation as well as simplified the image review process. Additionally, the android platform is an advantage over a computer-based or proprietary-based ultrasound operating system. The android tablet format also allowed for installation of customized apps which enabled us to use a universal patient participation consent, which, when completed, automatically uploaded to the institutional cloud–based storage account (Box). The android format also allowed us to include on the tablets educational materials such as institutionally created videos as well as commercially available apps such as the One Minute Ultrasound app.

While ultraportable ultrasound systems such as the Philips Lumify are less expensive than traditional portable ultrasound systems, the number of systems needed to cover multiple sites can still be a significant barrier. We were able to provide an ultrasound system to each student for a minimum of 3 weeks of their 6-week rotation using only 17 systems to cover 20 sites. At larger sites, where multiple students were rotating during a given rotation, the students could share a single unit during the rotation. In more rural or remote sites, the unit was assigned to the location throughout the course of the year to be used by one student at a time during each rotation. When the unit was not going to be in use for an extended period of time (more than a few weeks), a student would bring the ultrasound unit back to the medical school at the end of their rotation and the unit would be stored in the Clerkship Coordinator’s office until it was needed by another student. There were only two occasions during the year in which the clerkship coordinator needed to ship a unit to the hub site as returning it to the medical school was not feasible for a student between rotations.

To maximize the learning experience while decreasing the cost as much as possible, we limited the ultrasound system to a single transducer, the S4-1 or “cardiac” phased array transducer. This transducer adequately images abdominal structures as well as cardiac structures, making it a near-universal transducer for teaching purposes. Since the Philips Lumify cost is based “per transducer,” by using a single transducer, costs were minimized per ultrasound system.

In this initial pilot, we encountered several technological and implementation hurdles. There were technological issues with image transmission. Approximately 15% (27/183) of the images failed to transmit to the institutional cloud-based storage account (Box). By having the student report completion in the LMS, we were able to determine that the images were transmitted but not uploaded into Box. The Lumify app does not give notification confirming successful upload. This notification function is a DICOM standard, known as “storage commitment,” and is an acknowledgement of the handshake between the imaging device’s export of the image and the PACS acknowledgement that the image has arrived and has stored successfully. This type of PACS functionality would be difficult to implement in this app-based program using non-PACS storage such as the intuitional cloud–based storage account, and the Lumify app does not have this functionality. On investigation, we determined the most likely cause of the failure to upload was due to the wi-fi not being connected at the time of image transmission.

Another technological issue was loss of the log-in token as part of the cloud storage app preventing the upload to the cloud-based storage account (Box). Acquired images remain on the ultrasound tablet until specifically deleted, even when not transmitted. Once, it was discovered that image transmission to cloud-based storage account was an issue when there was documentation in the LMS, but no images in the cloud-based storage account, students were instructed to transmit their images via alternate mechanisms. This explains why the number of successful image transmissions increased with later rotations. Our current process has been modified to use the e-mail function in the Lumify app to have students e-mail the images to their institutional e-mail account where they can then directly upload them to the LMS. This method allows the student to immediately determine if the images exported successfully as well as to include documentation regarding the patient history and to indicate if they identified and treated any pathology. This e-mail method automatically removes any patient identifiers and is HIPPA compliant.

Acceptance of student performed ultrasound by the clerkship faculty was not a barrier. The online patient consent, approved through the university, alleviated faculty concerns with liability and patient confusion with ultrasound imaging by a student. Faculty participating in the image acquisition was minimal, though most faculty either viewed the clerkship educational videos on the ultrasound exams or participated in a faculty orientation course on POCUS. While the ultrasound exams were not done to evaluate for pathology, all images were reviewed during scoring for any potential pathology such as aortic aneurysm. Findings such as potential masses or lesions which may be cancerous are outside the scope of POCUS and hence outside the scope of student-performed educational ultrasound. Only one abnormal finding was discovered during image review, iliac artery dilation. In this case, the family physician was notified, but since only de-identified images were transmitted, only the date and time of the examination and the student performing the exam could be provided. The notification process could possibly be improved by using a code to retrospectively identify patients in the family practice location.

Overall, the pilot year of the program was successful with the majority of students able to successfully submit POCUS-acquired images that adequately identified and measured the appropriate anatomical structures. The scoring of the images improved throughout the year. In the first four rotations 30/53 (57%), students scored in the top quartile versus the last four rotations of the academic year where 44/66 (67%) scored in the top quartile. Prior to implementation, our expectation was that students would have a decrease in ultrasound proficiency as demonstrated by worse scoring as the year progressed as they became more remote from formalized, dedicated ultrasound training. However, the opposite was observed. This was likely due to a compounding effect as the students were exposed to POCUS in other clerkships during the year.

This project demonstrates UME students can obtain adequate ultrasound images of the aorta and the IVC on patients in a clinical setting during a clinical clerkship. The other components of POCUS such as image interpretation for pathology and clinical integration of POCUS into clinical practice require higher levels of competency which take more education and practice, much of which will be gained during residency. It remains to be evaluated if the addition of POCUS to clinical clerkships improves scoring on standardized testing or if it improves the clinical understanding of POCUS-related topics.

Conclusion

Integration of POCUS into a FM clerkship in a distributed campus model has unique barriers. The use of new ultraportable ultrasound systems helps to overcome these barriers and provides a resource for remote education, image export for review, and provides educational consent to allow practice patients to participate in student learning using ultrasound. Students’ ability to obtain appropriate images without real-time oversite and instruction can be done with a high degree of success, and this skill appears to improve the more the curriculum is integrated into the clinical rotations.

Compliance with Ethical Standards

Conflict of Interest

The authors declare that there is no conflict of interest.

Ethical Approval

This study was determined by the IRB to be exempt.

Informed Consent

This study was determined by the IRB to be exempt and therefore no need for informed consent.