Early-stage COVID-19 pandemic observations on pulmonary embolism using nationwide multi-institutional data harvesting

Abstract

We introduce a multi-institutional data harvesting (MIDH) method for longitudinal observation of medical imaging utilization and reporting. By tracking both large-scale utilization and clinical imaging results data, the MIDH approach is targeted at measuring surrogates for important disease-related observational quantities over time. To quantitatively investigate its clinical applicability, we performed a retrospective multi-institutional study encompassing 13 healthcare systems throughout the United States before and after the 2020 COVID-19 pandemic. Using repurposed software infrastructure of a commercial AI-based image analysis service, we harvested data on medical imaging service requests and radiology reports for 40,037 computed tomography pulmonary angiograms (CTPA) to evaluate for pulmonary embolism (PE). Specifically, we compared two 70-day observational periods, namely (i) a pre-pandemic control period from 11/25/2019 through 2/2/2020, and (ii) a period during the early COVID-19 pandemic from 3/8/2020 through 5/16/2020. Natural language processing (NLP) on final radiology reports served as the ground truth for identifying positive PE cases, where we found an NLP accuracy of 98% for classifying radiology reports as positive or negative for PE based on a manual review of 2,400 radiology reports. Fewer CTPA exams were performed during the early COVID-19 pandemic than during the pre-pandemic period (9806 vs. 12,106). However, the PE positivity rate was significantly higher (11.6 vs. 9.9%, p < 10−4) with an excess of 92 PE cases during the early COVID-19 outbreak, i.e., ~1.3 daily PE cases more than statistically expected. Our results suggest that MIDH can contribute value as an exploratory tool, aiming at a better understanding of pandemic-related effects on healthcare.

CTPA ordering trends in local emergency departments: are they increasing and did they increase as a result of COVID-19?

“…The positivity rates in this study were similar to the overall positivity rates found within Canada, with other studies reporting positivity rates around 15-20% [3,[28][29][30]. In the United States, these rates are typically lower, at approximately 10% or less [27,31,32]. The reasons behind this are unknown, but may be related to differences in healthcare funding and structure or litigation concerns.…”

“…In this case, there was no significant difference in the overall number of CTPA studies ordered; however, there was a significant increase in the positivity rate (for acute pulmonary embolism) of these studies, indicating a correlation with the timing of the COVID-19 pandemic. COVID-19 infection induces a prothrombotic state [13] and another study examining a larger dataset over a shorter time period found similar results of increased positivity rates since the onset of the COVID-19 pandemic [27]. However, other possible contributing factors could include the more sedentary lifestyles that afflicted many people during quarantine periods, or a lack of healthcare utilization during the earlier stages of disease leading up to pulmonary embolism due to a widespread decrease in non-urgent healthcare services during the beginning of the pandemic.…”

A Pictorial Essay Describing the CT Imaging Features of COVID-19 Cases throughout the Pandemic with a Special Focus on Lung Manifestations and Extrapulmonary Vascular Abdominal Complications

“…Radiology case reviews reported high rates of PE, ranging from 17% and 50%, especially in patients who were admitted to the ICU [133] (Figures 20-22). Radiology case reviews reported high rates of PE, ranging from 17% and 50%, especially in patients who were admitted to the ICU [133] (Figures 20-22). Radiology case reviews reported high rates of PE, ranging from 17% and 50%, especially in patients who were admitted to the ICU [133] (Figures 20-22).…”

Tracking the impact of global iodinated contrast agent shortage on radiology: analysis of CT exam volumes at a major US healthcare

“…By tracking both large-scale utilization and clinical imaging results data, the MIDH approach is targeted at measuring surrogates for important disease-related observational quantities over time. We have been able to demonstrate its applicability to track changes in the observed prevalence of pulmonary embolism on CT examinations during the early pandemic outbreak of COVID-19 in 2020 [4].…”

“…We have recently developed a novel Multi-Institutional Data Harvesting (MIDH) approach for longitudinal observation of medical imaging utilization and reporting that uses repurposed IT infrastructure of AI-based image analysis software for tracking the utilization of specific imaging services [4]. By tracking both large-scale utilization and clinical imaging results data, the MIDH approach is targeted at measuring surrogates for important disease-related observational quantities over time.…”