Risk Factors for False?Negative Covid?19 Testing in the Emergency Department: A Retrospective Cohort Study

Abstract

Research Question: In adults who present to the emergency department for Covid?19 testing, what clinical characteristics are associated with false?negative Covid?19 test results?

Background and Significance: Covid?19 has made enormous impacts on human health and society. Prior to Covid?19 vaccines and effective treatment strategies, medical and public health officials focused on reducing the morbidity and mortality of infections through early detection and social isolation. These strategies relied on dependable tests that could accurately confirm or exclude the disease. Early in the pandemic, important testing parameters were poorly understood. The lack of complete information on test accuracy had tremendous consequences. Covid?19 test results influence behavior, as those with a positive test result are more likely to socially isolate and cancel in?person plans, thereby limiting disease spread. Negative test results, however, often cause people to participate in their daily activities, like going to work or school inperson. False?negative Covid?19 test results can lead to devastating repercussions, as people with active Covid?19 infection may unknowingly infect others and to forgo other strategies to treat their own disease. The purpose of this study was to identify risk factors for potential false?negative Covid?19 tests. This information could be used to inform patient management, preserve limited resources, and educate patients to reduce the burden of Covid?19.

Materials and Methods: This was a multi?center retrospective cohort study of patients who presented to the emergency department at one of 12 study hospitals in Texas, USA from June 1 to August 31, 2020 who were tested for Covid?19. Eligible patients received an initial negative Covid?19 test by reverse transcriptase chain reaction (RT?PCR), followed by a second, positive Covid?19 RT?PCR test within 30 days. We used Wilcoxon signed?rank testing (for categorical variables) and paired t?testing (for continuous variables) to examine clinical factors that were different between the first and second visits.

Results: After screening 23,687 emergency department patient encounters, a total of 88 patients were included in the study. Patient?reported symptom duration was shorter for the first ED visit compared to the second (2.6 ¿ 0.3 days versus 3.6 ¿ 0.4 days, p?value=0.02). The findings from the first ED encounter that were most commonly associated with Covid?19 in our sample were increased lymphocyte count (35.2%), increased body temperature (32.6%), feeling nauseous (29.5%), and difficulty breathing (27.9%). Compared to the first visit to the ED, patients in our sample had differences in the second ED visit, which included the following: hypoxia (13.6% in the second visit versus 4.6% in the first, p?value: 0.005), abnormal infiltrates on chest radiography (59.7% in the second visit versus 25.9% in the first visit, p?value < 0.001), and elevated aspartate aminotransferase (29.1% in the second visit compared to 9.1% in the first visit, p?value <0.001).

Conclusion: Healthcare providers in the emergency department should understand the factors that may be associated with false negative Covid?19 tests. In our analysis, the biggest factor was a short duration of symptoms (symptoms for less than 3 days). During this window period, the main clinical features were fever, nausea, and difficulty breathing. Additionally, laboratory investigations showing an increased white blood cell count, especially lymphocytes, was also associated with a false?negative result. This information could be used to guide patient management by re?testing in the ED, encouraging patients to re?test at home or another healthcare facility in 1?2 days, and urging patients to social distance while awaiting repeat testing and diagnostic confirmation.