Point-of-Care Ultrasound in Internal Medicine: A National Survey of Educational Leadership

Ultrasound is a useful tool for enhancing the safety of bedside procedures. The most noted of these is the placement of central venous catheters,1 which in 2001 was listed by the Agency for Healthcare Research and Quality as 1 of 11 key recommendations for improving patient safety.2 Ultrasound use improves the safety and success of many other bedside procedures, including the placement of arterial lines,3 paracentesis,4,5 thoracentesis,6,7 arthrocentesis,8 and the incision and drainage of superficial abscesses.9,10 

Point-of-care ultrasound has emerged as a powerful tool for bedside physical assessment. Although the use of ultrasound as a diagnostic instrument by radiologists and cardiologists is well-established, another paradigm has emerged in which clinicians use ultrasound to answer limited point-of-care questions and integrate their findings into their assessment and management, analogous to the use of a stethoscope.11 Emergency physicians were the first to adopt this paradigm, and others have followed, including critical care physicians,11 anesthesiologists,12,13 physiatrists,14 and rheumatologists.15 It has been speculated that ultrasound machines could become the first new diagnostic instrument to be added to the internist's black bag in generations.16 

At the same time, relatively little is known about which ultrasound applications internists believe are the most useful, what ultrasound skills internal medicine (IM) residency programs are teaching their trainees, or what barriers may exist to the teaching or use of ultrasound in these settings. Prior experiences have shown that conflicts with traditional imagers,16 liability and risk,16 time constraints,17–19 and credentialing requirements13,19 may be barriers to the teaching and use of point-of-care ultrasound.

We surveyed IM educators to gain a better understanding of their views on the utility of ultrasound for procedures and diagnosis at the point of care and to determine what ultrasound training residents are currently receiving. We also sought to identify barriers to the adoption of point-of-care ultrasound in IM residency programs and clinical practice.

In February 2012, a 27-question Web-based survey (created using surveymonkey.com) was sent to all 2200 members of the Association of Program Directors in Internal Medicine (APDIM). Members of APDIM include faculty in educational roles (program directors, associate and assistant program directors, and core faculty), program coordinators, fellows, and residents. Three notifications of the survey link were made. Participation was voluntary and anonymous.

Survey items (provided as online supplemental material) were developed by the authors and tested with the University of Minnesota General Internal Medicine Scholarship Committee, which resulted in the rewording of some unclear questions. The survey included questions about respondents' perceptions of the usefulness of ultrasound for various procedures (central venous lines, arterial lines, paracentesis, and thoracentesis) and for point-of-care evaluations. Responses used a 5-point Likert scale (1  =  not at all useful, 5  =  very useful). Participants who identified themselves as program directors were asked about current ultrasound use by their residents and about their program's existing or anticipated future educational efforts to teach ultrasound, including the curriculum's content, duration, timing during residency, and instructors. They were also asked about other residency or fellowship programs at their institution that might teach bedside ultrasound.

The survey also asked respondents to rate potential barriers to the implementation of point-of-care ultrasound curricula using a 5-point Likert scale (1  =  no barrier, 5  =  very large barrier). For each barrier, an average score for all respondents was calculated. This study was determined to be exempt from review by the University of Minnesota Human Subjects Committee.

Of the 2200 surveys distributed to APDIM members, 234 were returned, yielding a response rate of 11%. Among the 234 respondents, 231 indicated their role in a training program. Ninety-eight respondents (42%) were associate program directors, 54 (23%) were program directors, 38 (16%) were other faculty members, and 15 (6%) were assistant program directors. The other 26 respondents (11%) included educational faculty (department chairs and educational chairs) and trainees.

Respondents' average ratings for the usefulness of point-of-care ultrasound for specific procedures and diagnostic applications are shown in figure 1. The mean usefulness scores for procedures ranged from 3.9 to 4.8; scores were highest (indicating greatest usefulness) for central venous lines (4.8), thoracentesis (4.7), and paracentesis (4.6). The mean usefulness scores for diagnostic applications of ultrasound ranged from 3.5 to 3.9; scores were highest for assessment of the lungs (4.0), inferior vena cava (3.9), and heart (3.9).

Of the respondents, 161 reported the applications for which residents use ultrasound (figure 2). The greatest use of ultrasound was reported for central venous catheter placement (n  =  159; 99%), followed by thoracentesis (n  =  99; 61%) and paracentesis (n  =  94; 58%). A smaller number of respondents reported that residents currently use ultrasound for diagnostic evaluation.

Of 177 respondents, 45 (25%) reported that a formal ultrasound curriculum existed at their institution. Of those who did not report having a formal curriculum, 46 (35%) indicated plans to implement one in the next 12 months. Of the respondents indicating a current or planned curriculum, 79 of 84 (95%) reported that they plan to teach ultrasound-guided central venous catheter placement, followed by paracentesis (57 of 76; 75%) and thoracentesis (58 of 78; 74%). Use of ultrasound to evaluate the inferior vena cava was the most commonly taught diagnostic application.

Program directors with current or planned ultrasound curricula reported that an average of 10 hours is or will be spent on didactics and 13 hours in hands-on training. In many cases, training is or will be spread over the course of a residency. Critical care physicians (55 of 72 respondents; 76%) and hospitalists (38 of 66 respondents; 58%) make up a large proportion of the instructors for these courses.

Of 82 respondents, 52 (64%) reported that residents were required to complete training before using ultrasound for procedures. Of the 61 respondents who indicated that diagnostic ultrasound was being taught, 33 (54%) reported that training was required before using ultrasound in diagnosis. Few programs (7 of 80; 8%) retain images to provide feedback to learners.

Mean scores for perceived barriers to teaching and/or implementing point-of-care ultrasound are presented in the t a b l e. Overall, respondents indicated that the cost and time to train faculty members (mean scores  =  2.82 and 2.79, respectively) were the largest barriers to implementation.

Our findings show a substantial interest in point-of-care ultrasound among IM educators. Almost all respondents reported that residents use ultrasound for central lines, and many also reported its use in thoracentesis, paracentesis, and other procedures. Many reported that ultrasound could be useful as a point-of-care diagnostic tool. Although only 25% of respondents indicated that their programs currently have formal curricula to teach these skills, others reported plans to implement training within the next year. To our knowledge, this is the first study documenting this interest in the use of ultrasound on a national level.

Our findings suggest that although barriers exist to the implementation of ultrasonography in IM residency training settings, most are perceived to be low to moderate. Respondents indicated that legal barriers and fear of litigation were not major barriers; this finding is consistent with an analysis in emergency medicine, which found only 1 lawsuit related to bedside ultrasonography in more than 20 years of use. This lawsuit was notable, however, in that it was not for misapplication but for failure to use point-of-care ultrasound.20 Lack of interest by residents was not a barrier, which is consistent with a study that indicated there was strong interest in this technology among IM trainees.21 

Respondents indicated that residents may be using point-of-care ultrasound to make patient care decisions without any training or quality assurance mechanisms. The American Medical Association has stated that appropriate training and quality assurance/quality improvement are mandatory for users of ultrasound.22 Guidelines for such training and procedures have been developed in emergency23 and critical care medicine,24,25 but to our knowledge none exist for IM. Further collaboration among ultrasound-trained faculty and the establishment of guidelines and standards by governing professional societies may be helpful in facilitating the safe advancement of ultrasound use in IM.

Our study is limited by the survey's low response rate (11% overall and 15% for program directors). The low response rate may be partially explained by the fact that the instructions for the survey asked only program directors and academic internists to complete the survey, thereby excluding many of those receiving it (program coordinators, fellows, and residents). It is possible that respondents were more enthusiastic about the use of ultrasound than nonresponders, and this would limit the generalizability of our results. We also did not pilot test our survey, and it is possible that respondents misunderstood or misinterpreted questions.

Our study demonstrated interest in point-of-care ultrasound by IM educators, and an interest in teaching ultrasound skills to IM residents. Further research is needed to determine the ultrasound skills most useful for IM trainees and to develop competency-based curricula as well as quality improvement and quality assurance mechanisms.