In this analysis of closed medico-legal cases in the CMPA repository we identified 15 cases where POCUS was a contributing factor towards medico-legal action. Almost half of the cases were due to physicians failing to perform POCUS when indicated. In contrast to previous literature that had not identified POCUS cases that resulted in a medico-legal action [12,13,14,15,16,17], we found a number of cases where POCUS use resulted in a medico-legal action due to issues with provider skill, sonographic approach, reporting, documentation, misdiagnosis, and inappropriate use. All cases resulted in findings against the physician involved. Five cases involved patient death.

Contrasting our study to existing literature [12,13,14,15,16,17] (which report civil-legal cases only), the increased number of medico-legal cases identified may be accountable by several factors. First, the CMPA repository is a comprehensive database capturing essentially all medico-legal cases against Canadian physicians, with data coded prospectively in a highly searchable way. Additionally, most of the cases were hospital or college complaints, which have a lower barrier to filing compared with civil litigation. This is important as this is the first study to examine hospital and college complaints, in addition to civil litigation. Thus, whereas patients and families may have felt POCUS was applied inappropriately and submitted a hospital or college complaint, it may not have met a threshold to proceed with civil litigation (Fig. 1).

Regardless, our study demonstrates an important reality for the twenty-first century acute care physician: failure to perform POCUS when indicated may result in medico-legal action. For some specialties and indications, POCUS may no longer be an adjunct to augment traditional bedside assessment but rather a core part of the diagnostic process itself. While POCUS infrastructure and expertise varies between different settings, as the evidence and use of POCUS grows, so too can the expectation that it is appropriately used when clinically indicated. This may be particularly relevant for specific use cases of POCUS with well-established diagnostic pathways including the Focused Assessment with Sonography in Trauma (FAST) exam during trauma resuscitation, or its use to diagnosis undifferentiated shock [18,19,20]. The excellent safety and diagnostic prowess of acute care POCUS has led to its endorsement by various societal guidelines [6,7,8,9,10].

Despite increasing adoption of POCUS in acute care medicine, adequate training and skill is necessary for its safe implementation, and in this analysis was flagged as a contributing factor for medico-legal action in several cases. One systems level approach to POCUS education being implemented at multiple institutions is to teach ultrasound physics, knobology, and anatomy in parallel with traditional medical school curriculum to provide learners with a base skillset that can be built on during further training [3,4,5]. Then, specialty specific POCUS training with a focus on interpretation and synthesis can be taught during residency. Although this approach will help ensure future generations of physicians have base competency in POCUS, for clinicians in practice, alternative educational approaches should be considered. These include informal or formal instruction from colleagues with POCUS expertise to help develop core skills. Alternatively, continuing medical education opportunities like POCUS courses, conferences, rotations, or fellowships are an excellent resource, however, may not be feasible for many physicians in practice. Some societies have suggested processes for credentialing and privileging practicing physicians, which may be helpful roadmaps for interested clinicians [6].

In addition to helping clinicians obtain and interpret POCUS images, formalized training helps clinicians appropriately integrate POCUS findings with other clinical information. In fact, diagnostic errors, deficient assessments, and failure to perform other indicated tests or interventions were the most common reasons for patient complaints in our study, indicating a failure to properly integrate POCUS into the diagnostic workup of patients. Ideally, POCUS findings should be integrated into patient care with a ‘Bayesian mindset’, meaning that the positive or negative finding on POCUS helps change the post-test probability of a pathology being present. This contrasts with an oversimplified view of POCUS where the presence or absence of findings on POCUS dictates whether a disease is present. This dichotomized view of POCUS is potentially dangerous, and in our experience seen more with novice POCUS practitioners. The visual nature of the medium may lend itself to a ‘seeing is believing’ phenomenon, which can lead some clinicians to place inappropriate weight on the POCUS findings, disregarding other competing clinical information.

Inappropriate integration of POCUS into practice may also result from a failure to understand the test characteristics for POCUS in that population. For instance, FAST scan has high specificity (> 98%) to detect intraperitoneal free fluid, however only moderate sensitivity (70–90%) with test characteristics varying between operators [18, 21, 22]. For a trauma patient with a very high pre-test probability for intrabdominal hemorrhage (e.g. 80%), even with a negative FAST scan (assume sensitivity of 70%), the post-test chance of intrabdominal hemorrhage is 55%. Overreliance on POCUS and failure to integrate other clinical information is a crucial pitfall to avoid.

Inadequate documentation led to medico-legal action and was a major theme in the contributing factor analysis. If a POCUS is performed, the indication, views acquired, findings, and interpretation should be recorded in the patient's chart. At a minimum, this should be written as a progress note or included as part of a consultation. A better practice, although not available at many centers, is to save images in an accessible archiving system, and then generate a written report to allow for accountability and communication between providers [23, 24]. Ideal practice would have all archived scans undergo quality assurance by a POCUS expert, with the amended reports subsequently uploaded into a patient’s electronic medical record. The practice of "shadow" scans where results are communicated by verbal handover between providers is not acceptable and may expose physicians to medico-legal risk.

Although this represents a preliminary analysis of Canadian medico-legal cases involving POCUS, we expect the number of medico-legal cases to grow in parallel with increased POCUS use across. A repeat analysis of this work in 5 or 10 years will be helpful to assess for evolving patterns in POCUS medico-legal risk. Furthermore, this study excluded procedural use of POCUS (e.g. central line insertion) which would be an important area for future research. Additionally, knowledge translation surrounding best practices in POCUS training, clinical integration, and documentation is needed to promote optimal POCUS use among physicians.

There are several important limitations: the search terms used may not have retrieved all medico-legal cases related to POCUS. To address this, the CMPA will now prospectively identify POCUS cases to facilitate future research. We omitted granular clinical details from the cases to protect patient and physician privacy, however this limits the analysis of factors leading to poor patient outcomes. This was unavoidable and was done in close collaboration with the CMPA to adhere to their rigorous privacy mandates. We recognize that this leaves unanswered questions, but feel this study still provides actionable take homes to improve patient safety. Finally, we have not included procedural POCUS, as we would lack granularity to distinguish between medico-legal action from the procedure itself, or the POCUS use.