| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
University of California Davis Sacramento, CA 95817
Cedar Sinai Medical Center Los Angeles, CA 90033
American College of Emergency Physicians Charlotte, NC
28232
Society for Academic Emergency Medicine Manhassett, NY
11732
We read with interest the study published by Hertzberg et al. [1] and would like to comment on several components. We agree with premise that there is a paucity of evidence to support any group's contention for the number of examinations needed to be "competent." We applaud the authors for making an attempt to measure the level of competency for radiology residents. However we wish to make several observations and criticisms of the methods and resulting conclusions.
First, this is a study of 10 residents at one institution and the results appear to reflect more on this institution's training practices. For more external validity to be granted to the results, a larger and more varied sample would be needed. To our reading, this model is more an assessment of first-year radiology residents at their institution.
Second, were outcome criteria (e.g. errors) defined a priori and based on published learning objectives, or was it a post hoc determination of which findings were significant or minor? This distinction is important because it appears that the investigators did the determination of errors between the resident's examination and the attending physician or sonographer's examination without any blinding. Certainly this introduces the potential for significant bias. In other words, if the authors' intent was to show that there were more significant errors at 50 examinations than at 200, without blinding, they could be more critical of earlier examinations.
Third, was the measurement model Hertzberg et al. [1] were using tested to be reliable and valid by any other published study or was it merely developed for this trial? We suspect that the latter is the case because the residents were not measured against standardized examination models and examinations were not evenly distributed among the 16 anatomic areas described; consequently, the trial was evaluating both the resident learning curve and the ability of their test to measure resident learning. As with any study, one must be careful in applying a measurement tool without first showing it to be valid or reliable. The authors should realize that they were comparing different outcomes for each resident because residents did not scan the same patients.
It appears more that this study is a sounding board for criticism of other specialities' use of sonography. The discussion section appears to have more of an editorial flavor than scientific appraisal of sonographic proficiency. To our understanding, this study appears to be a case of "apples and oranges." If one was to ignore the previously described methodologic issues of the study and merely focus on the conclusion, it could be said that for radiologists who offer comprehensive sonographic studies, 200 examinations are too few. For other physicians who perform limited, goal-directed bedside sonography to answer focused questions the answer is still unknown.
The American College of Emergency Physicians has endorsed six primary applications for limited sonography. This study by Hertzberg et al. [1] evaluated 16 areas, most of which would not be within the interest of limited bedside sonography. We agree that this skill cannot be learned in a "weekend course." The skill must be learned over time and the exact number of scans needed to be proficient in a particular area is still unknown. The sonography skills are taught in many residency rotations including obstetrics and gynecology, surgery, family practice, emergency medicine, urology, and ophthalmology, to name a few. To think that only one specialty has the exclusive domain of such a useful technology is sophomoric.
The American Medical Association's policy H-230.960 [2] that each specialty society should decide the necessary training requirements for sonography proficiency only strengthens the position that each specialty can determine what is useful for its practitioners.
References
Duke University Medical Center Durham, NC 27710
We appreciate the interest Dr. Rose and colleagues have shown in our recent article [1] and would like to take this opportunity to respond to their comments. Although our study was conducted at a single institution, we believe it is unlikely that resident performance would have been substantially better in a multiinstitutional setting. The radiology residency program at our institution is widely regarded as one of the premier training programs in radiology in the country, and our residents are uniformly high-achieving individuals with long records of academic success. Residents at our institution receive in-depth training in sonography from faculty with extensive teaching, research, and clinical experience. The faculty has a collective experience of more than 75 years performing and teaching sonography. Every resident entering this study was given one-on-one training and supervised hands-on scanning experience with highly skilled certified sonographers. Thus, we believe our results reflect a "best case" scenario, in which a superb resident group was trained by certified sonographers and a seasoned faculty who specialize in sonography. We doubt such conditions are replicated in all institutions.
Outcome criteria were defined a priori and detailed extensively before beginning the investigation. Our criteria for grading errors are similar to those used in other publications and investigative contexts [1, 2]. Specifically, discrepancies were considered minor if "unlikely to have affected imaging work-up, laboratory studies, medical, or surgical treatment" and were considered clinically significant if the discrepancy was "likely to have required modification in medical or surgical treatment or additional imaging or laboratory studies to disprove or confirm" [1]. Likewise, the scanning protocols by which resident performance was judged were based on standardized criteria for examinations established by the American Institute of Ultrasound in Medicine (a multispecialty professional organization) and the American College of Radiology.
Finally, we strongly disagree with the notion that for some physicians an abbreviated sonogram is an acceptable option, whereas other physicians should be held to a higher standard of care. The medical community must ensure that words such as "focused" and "goal-directed" do not become merely code words for substandard and inadequate work. Published sonographic protocols have been carefully generated through the experience and consensus of expert sonologists from multiple specialties with the intent that all patients receive at least a minimum standard of care. Adherence to these protocols by competent practitioners should greatly reduce the likelihood that important findings are inadvertently missed and should also ensure that a single limited—possibly misguided—diagnosis is not the only one considered. It would seem that patients in acute distress are hardly the ones who should receive abbreviated studies performed by individuals with limited training.
In short, diagnostic competence with sonography depends crucially on operator skill and methodic technique. Our results underscore the need for thorough and intensive training for physicians who perform and interpret these challenging studies.
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  C. L. Moore, S. Gregg, and M. Lambert Performance, Training, Quality Assurance, and Reimbursement of Emergency Physician-Performed Ultrasonography at Academic Medical Centers J. Ultrasound Med., April 1, 2004; 23(4): 459 - 466. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
