Prospective evaluation of cardiac ultrasound performance by general internal medicine physicians during a 6-month faculty development curriculum

Point-of-care ultrasound (POCUS) is clinician-performed ultrasonography used to guide real-time diagnostic and management decisions. POCUS training is rapidly expanding in undergraduate and graduate medical education. Twenty-eight percent of medical schools offer POCUS curricula [1], including several well-established, vertically-integrated programs [2–4]. A survey of Association of Program Directors in Internal Medicine members found that 25% of respondents reported formal POCUS curricula, with another 25% planning to implement programs within a year [5]. Educational leaders in these studies identified faculty development as a major barrier to broader adoption of POCUS training [1, 5].

Despite the increasing need for POCUS-trained faculty in medical education, limited research has investigated training programs for general internal medicine (GIM) faculty. One 10-week faculty program resulted in improved self-reported confidence and exam scores, but image acquisition skills were not assessed [6].

POCUS training can improve short-term outcomes, but knowledge and skills decay without ongoing support. A study of internal medicine physicians who completed a 3-year POCUS curriculum during residency demonstrated poor retention of cardiac ultrasound skills after more than 1 year of nonuse [7]. Surgical residents and medical students who participated in a 2-h cardiac ultrasound course had significant decay in their knowledge and skills within 1 month [8]. Studies of internal medicine residents found knowledge deteriorated in the months following workshop-based training [9], but that longitudinal support prevented knowledge decay [10].

This study investigated the impact of a longitudinal POCUS curriculum on GIM faculty’s performance of focused cardiac ultrasound (FCU). We hypothesized that at the end of the study period, GIM faculty performance of FCU would be resilient to skills decay and comparable to cardiology fellows’ performance.

Four of nine GIM faculty were female (44%) and eight (89%) had no prior POCUS training. Eight participants (89%) completed all three sessions. Faculty earned an average of 30.6 CME hours (SD 8.6 h) over the 6 months. The average attendance at voluntary curricular components (workshops, didactics, and image review sessions) over the 6 months was 68% (range 38–88%). FCU efficiency score data are displayed in Table 1. GIM faculty efficiency scores were maintained over the 6-month study period (baseline 2.2, SE 1.0 vs. 6 months 3.8, SE 1.0, p = 0.076). There was no difference between mean GIM faculty and cardiology fellow efficiency scores during any of the three assessment sessions (Fig. 2). DCSs performed significantly better than GIM faculty and cardiology fellows during all three assessments (p < 0.0001).

Baseline			3 months			6 months
GIM	CF	p	GIM	CF	p	GIM	CF	p
2.2 (1.0)	3.7 (1.2)	0.69	3.3 (1.0)	4.7 (1.2)	0.77	3.8 (1.0)	4.4 (1.2)	1.00

GIM	DCS	p	GIM	DCS	p	GIM	DCS	p
2.2 (1.0)	11.7 (1.2)	< 0.001	3.3 (1.01)	16.0 (1.2)	< 0.001	3.8 (1.0)	14.3 (1.2)	< 0.001

CF	DCS	p	CF	DCS	p	CF	DCS	p
3.7 (1.2)	11.7 (1.2)	< 0.001	4.7 (1.16)	16.0 (1.2)	< 0.001	4.4 (1.2)	14.32 (1.2)	< 0.001

Data reported as model-adjusted means (SE). Comparisons reported using adjusted p value

GIM general internal medicine faculty, CF cardiology fellows, DCS diagnostic cardiac sonographers, SE standard error

GIM faculty’s exam duration decreased during the study period (baseline 17.5 min, SE 1.5 vs. 6 months 13.0, SE 1.5 min, p < .001). Combining the three sessions, GIM faculty and cardiology fellow exam duration were comparable across the three sessions (faculty 15.3 min, SE 1.4 vs. fellows 13.8 min, SE 1.8, p = 0.64).

GIM faculty participating in a voluntary, multi-modal curriculum maintained their FCU acquisition skills over a 6-month period. GIM faculty efficiency scores were similar to those previously reported by critical care physicians with American Society of Echocardiography level III certification (although the latter performed FCU on mechanically ventilated patients) [12]. Our findings highlight the importance of longitudinal support when teaching FCU to novice faculty. FCU performance requires integration of a complex set of psychomotor skills, which cannot be mastered without regular practice and feedback. Prior studies of novice cardiology fellows [13] and hospitalists [14] found that FCU skills can be developed in weeks to months with frequent expert coaching. The duration of training required to achieve proficiency likely depends on institutional resources and needs. In the present study, GIM faculty did not have dedicated time and was concomitantly learning non-cardiac POCUS applications, requiring a longer curriculum.

Our primary outcome was a measure of efficiency, which Gaudet et al. [12] have argued is a “hallmark of expertise”. Efficiency has been used as an assessment variable in prior FCU performance studies [15] and is important for several reasons. First, POCUS must be economical for integration into clinicians’ daily workflow. Exam efficiency is also vital for the prompt assessment of time-sensitive clinical conditions, such as shock. Finally, task efficiency may precede improvements in performance quality. This phenomenon has been observed in prior FCU studies [12, 13] and performance research for other fine-motor skills, such as endoscopy [16].

While GIM faculty had efficiency scores similar to cardiology fellows, their total image quality scores were lower. This reason for this is likely multifactorial. GIM faculty may not have engaged in enough deliberate practice over the 6 months to advance their image acquisition skills [17]. Additionally, cardiology fellows’ extensive experience in interpreting diagnostic echocardiography studies may have contributed to their superior scores. It should also be noted that our FCU protocol included views which are more technically challenging for novices [15]. Some of these views, such as apical four chambers, are excluded from commonly-used FCU exam protocols [18]. Although their total image quality scores were lower, GIM faculty performed comparably to cardiology fellows for overall diagnostic quality. This may be the most clinically relevant outcome, as the primary purpose of POCUS is to quickly gain information to guide clinical decision making.

This study adds to the existing literature in several ways. It is one of the first papers to examine the educational outcomes of a longitudinal POCUS curriculum for GIM faculty. Studies of residents and fellows may not be applicable to faculty, given differences in clinical experience, time availability, and willingness to adopt innovations [19]. Additionally, we assessed performance of a complete FCU exam protocol including all recommend views [20] and utilized a validated scoring instrument. Finally, the baseline assessment occurred after an intensive introductory workshop, allowing us to evaluate the impact of the longitudinal portion of the curriculum independent of the introductory training.

There are limitations to this study. It was conducted at a single institution with few participants. Repeated scanning of the same SPs may have led to higher scores due to familiarity, although this was likely limited by the 3-month washout periods. Furthermore, only the GIM faculty group showed improvement across all testing periods. We were unable to accurately track the number of exams performed by each participant during the curriculum to look for correlation with performance outcomes. Additionally, our study did not assess image interpretation skills, which typically develop more quickly than image acquisition skills [21].

In conclusion, GIM faculty participating in a longitudinal POCUS curriculum maintained their FCU acquisition skills over a 6-month period. Efficiency scores were comparable to cardiology fellows, but overall image quality was lower. Future studies will need to examine if GIM performance of FCU is sustainable beyond 6 months.