Efficacy of clinical evaluations for COVID-19 on the front line

To the Editor:

In the midst of the COVID-19 pandemic, there remains limited availability of Food and Drug Administration-approved tests for presence of the SARS-CoV-2 agent [1, 2]. Even as testing capacity expands, optimization of resource utilization in the healthcare setting remains a significant priority [3, 4]. Thus, the vast majority of front line work being done to evaluate for possible COVID-19 is highly dependent on the clinical assessment of a presenting patient’s signs and symptoms. The extent to which current clinical assessments are effective, in the era of rapidly evolving local and professional guidelines, is not entirely clear.

We conducted a retrospective review of patients assessed for possible COVID-19 illness at our urban medical center in Los Angeles, California. The institutional review board deemed the study exempt. We carefully reviewed all clinical records to ascertain the provider’s level of clinical suspicion for COVID-19 illness and compared these assessments with available results of SARS-CoV-2 testing, in addition to longitudinal data on clinical outcomes. We found that the vast majority of patients (96% of N = 25) clinically assessed to have a low probability of COVID-19 illness were subsequently confirmed to have either a negative SARS-CoV-2 test result or, in the absence of testing, clinical stability without any further concern for COVID-19 illness (Table 1). All clinical assessments were performed by a physician, with some (16%) conducted by a nurse practitioner or physician assistant in conjunction with physician supervision.

Table 1 Characteristics, testing status, and clinical outcomes for N = 25 patients

In the absence of widespread readily available access to SARS-CoV-2 testing, clinical assessment is and will remain the standard of care for initially determining probability of COVID-19 illness and, in turn, appropriateness for receiving testing—especially in areas where testing availability is limited. This case series from an urban medical center suggests that despite the rapidly evolving body of knowledge around COVID-19 illness and its variable presentations among affected patients, clinical provider assessment of high versus low probability of active infection can be relatively reliable. This case series further supports the hypothesis that a well-informed clinical assessment, with or without concurrent access to rapid point-of-care SARS-CoV-2 testing, could be leveraged to more efficiently triage patients [5]—even those with medical comorbidities whose chronic illness burden may appear to pose a diagnostic challenge at the outset. In effect, a clinical evaluation that does not rely on viral testing results may be very accurate and substantially aid in ongoing efforts to conserve and appropriately prioritize the use of medical resources. Use of sound clinical judgment can also facilitate consideration of alternative diagnostic explanations.