We performed a survey-based evaluation of Family Medicine residents’ perceptions of POCUS impact on medical decision-making in a Level 1 Hospital in Zambia. We found their perceptions of this impact to be large for the patients that they scanned.

POCUS assisted and changed management in the majority of patient encounters. POCUS was found to assist medical management almost universally, likely due to the fact that both positive and negative scan findings often assisted the clinical team’s decision-making. POCUS was also found to change medical management in approximately 2/3 of patient encounters, likely a reflection of the relative unavailability of alternative imaging options and other diagnostics (such as most laboratory tests) for many patients at our study site. Therefore, for the vast majority of patient who received POCUS exams, residents perceived POCUS to be impactful.

Although we did not ask residents to record specific diagnoses or scanning indications in the Patient Scan Questionnaire, anecdotally residents reported that IVC, pulmonary, cardiac, and FASH (Focused Assessment with Sonography for HIV-associated tuberculosis) scans were found to be very helpful [6, 8]. IVC scanning assisted frequently in estimating central venous pressure (CVP) and fluid tolerance for shock patients, and when combined with pulmonary scanning, assisted in the evaluation of patients with undifferentiated respiratory failure. Residents reported that both conditions were commonly encountered, hence the high numbers of IVC scans performed in this study. Given the hospital’s X-ray machine was frequently non-functional, residents reported that pulmonary ultrasound proved useful in the evaluation of respiratory patients and management of pleural effusions—which are commonly encountered in this setting. Finally, they found cardiac echo useful in assessment of patients with likely congestive heart failure, which is underdiagnosed in our setting, and FASH scanning empowered them to make a presumptive diagnosis of extra-pulmonary tuberculosis, which can be difficult to diagnose definitively.

The potential impact of POCUS on patient outcomes in LMICs is tremendous [2, 25]. While a variety of POCUS education methods have been implemented in LMICs previously, the findings from this study suggest that integration of POCUS education directly into residency training curricula in LMICs is both feasible and perceived by resident physicians as having large impact in their clinical care. Indeed, POCUS education at the residency level may prove to be a useful tool for POCUS to be more broadly adopted—and regularly utilized—by clinicians practicing in LMICs. In addition, expanded availability of handheld devices with whole-body scan abilities, low power requirements, and ultra-portability will likely further support the utilization of POCUS in these settings [2].

Integrating POCUS education at the residency level in LMICs has multiple benefits. First, clinicians in residency are still developing their medical skills, allowing them to embrace POCUS and incorporate it into their patient care more easily than seasoned clinicians. Second, residency is a formative time, when clinicians develop many habits that persist long-term. By incorporating POCUS into their residency training, resident clinicians will more likely incorporate POCUS into their practice and utilize it throughout their medical careers—potentially impacting the outcomes of thousands of future patients [5, 21]. Even if these clinicians cannot afford to purchase their own handheld devices, many will have access to an ultrasound machine of some sort at their future clinical sites. Finally, incorporating POCUS education at the residency level allows for the possibility of longitudinal POCUS training with direct supervision and ample opportunities for hands-on practice—features which are often unavailable in short-term POCUS trainings [21].

How can POCUS training be effectively and sustainably integrated into LMIC residency programs? One option includes the sending of POCUS-trained physicians to serve as visiting faculty at such residency programs for a longer period (e.g., 1 year or more), which allows for longitudinal training of both residents and local faculty. Ideally, interested local faculty can also be trained as “POCUS champions” and serve as POCUS educators for their own programs in the future—a “train the trainers” model [1, 4]. An alternative is longitudinal remote POCUS education, involving both online didactics and remote mentorship sessions spread over a longer time (months to > 1 year). Online platforms enhancing effective remote mentorship are already available. Finally, shorter term options such as a POCUS away rotation or a short-term workshop are possible but would likely have greater impact if paired with some form of long-term POCUS mentorship. Whatever the method, the features of an optimal POCUS educational program in LMICs would include longitudinal duration, direct supervision, opportunities for hands-on practice, and long-term mentorship—similar to what is most effective in high-resource settings [3, 9].

There are several limitations to this study. First, the survey data contained both subjective and objective data points. Survey items #5a and #6a, which ask if the scan assisted or changed management, respectively, are somewhat subjective. Improved survey item design to make the survey more objective would be beneficial for future studies. In addition, the potential for bias is significant for several reasons. First, the survey was administered to residents only after the scans were performed, without baseline data present. A more objective attempt at quantifying POCUS’s impact, such as comparing the differential diagnosis and/or the patient plan using surveys both pre- and post-scan, may serve to eliminate some of this bias. Also, a resident’s pre-existing clinical acumen may influence the degree to which she views POCUS as having assisted her management, and this was not accounted for in this study. Controlling for residents’ training levels may strengthen future study. Next, the questionnaires were not anonymous, potentially adding additional bias to the survey results. Finally, the study’s design places it at significant risk for reporting bias, given visiting faculty served to supervise the residents’ scans, assist in image interpretation, and subsequently administer the survey to the same residents. Those residents may have felt pressure to answer survey questions more positively that they would have otherwise, possibly inflating the study’s results. Ideally similar surveys in future studies would be performed by an alternative non-faculty data collector.

The study used surrogate markers of patient-level impact, not direct measures. Patient-level indicators, such as mortality, complication rates, additional procedures, etc., were not directly recorded, which limits the study’s applicability to patient care. Ideally future studies would involve chart review or other means of tracking patient-level impact. In addition, patients were selected to be scanned based off of clinical necessity, as determined by the clinical team—which can be a subjective process at times. This could potentially open the data to selection bias. The study’s sample size was relatively small (n = 366 patients scanned), which was mostly a reflection of the faculty’s schedule limitations and residents’ restricted access to the ultrasound probes.

In addition, the data contains very limited numbers of Obstetrics/Gynecology (OB/GYN) and trauma (Extended Focused Assessment with Sonography in Trauma, or eFAST) scans—indications, where ultrasound is generally found to be very useful—which was likely multifactorial. First, faculty schedule limitations and the hospital’s request that faculty lead the departments of Internal Medicine and Pediatrics resulted in less availability to scan OB/GYN and trauma patients. Adding to the paucity of eFAST scans was the fact that the hospital’s Emergency Department referred all trauma patients to the region’s tertiary center, as they lacked a surgeon capable of managing severe trauma cases, so these patients were not admitted to our facility. Likely exacerbating the paucity of OB/GYN scans was the fact that residents on the OB/GYN spent most of their time performing surgical obstetric cases, with most non-operative deliveries being managed solely by midwives. Since our program’s residents were not spending as much time on Labor & Delivery, they did not contact faculty regularly for POCUS scans there. This asymmetry in scan types may have skewed the survey results. Adjustment of faculty schedules—or provision of additional POCUS faculty and probes—to ensure consistent device availability and appropriate supervision in all patient care settings would improve this imbalance.

Over the course of the study, IVC scans were likely relied upon too heavily and sometimes performed as stand-alone scans, which may have limited diagnostic reliability [14, 16]. Future studies should utilize IVC scanning only in conjunction with cardiac [20] and/or pulmonary scanning [11] and only for evidence-based indications.

The POCUS scans in this study were performed exclusively with two handheld devices, which may have reduced image quality as compared to larger, less portable ultrasound machines. Ideally, future studies would utilize more sophisticated ultrasound equipment. Utilizing residency faculty—who are intimately involved in the team’s patient care—as the POCUS experts may have introduced a level of bias in image interpretation. Future studies should utilize an external, unbiased POCUS expert. Finally, data analysis was performed using only descriptive statistics for this study. Future studies of this type would benefit from more robust data collection and use of inferential statistics, so that any differences could be evaluated for statistical significance and further inferences made.

This study suggests that resident physicians in LMICs perceive POCUS to be very useful in their medical decision-making for whom they deemed a POCUS scan necessary. Though this study’s sample is small, these data may support the advancement of POCUS education in residency programs throughout LMICs, which may be an ideal strategy to promote widespread utilization of POCUS in low-resource settings globally. Further analytical studies with larger sample sizes which evaluate impact using patient-level indicators will be necessary to further characterize the effect of POCUS education in residency programs in LMICs.