The Effect of Test Timing on the Probability of Positive SARS-CoV-2 Swab Test Results: Mixed Model Approach.

BACKGROUND: During the COVID-19 pandemic, swab tests proved to be effective in containing the infection and served as a means for early diagnosis and contact tracing. However, little evidence exists regarding the correct timing for the execution of the swab test, especially for asymptomatic individuals and health care workers. OBJECTIVE: The objective of this study was to analyze changes in the positive findings over time in individual SARS-CoV-2 swab tests during a health surveillance program. METHODS: The study was conducted with 2071 health care workers at the University Hospital of Verona, with a known date of close contact with a patient with COVID-19, between February 29 and April 17, 2020. The health care workers underwent a health surveillance program with repeated swab tests to track their virological status. A generalized additive mixed model was used to investigate how the probability of a positive test result changes over time since the last known date of close contact, in an overall sample of individuals who tested positive for COVID-19 and in a subset of individuals with an initial negative swab test finding before being proven positive, to assess different surveillance time intervals. RESULTS: Among the 2071 health care workers in this study, 191 (9.2%) tested positive for COVID-19, and 103 (54%) were asymptomatic with no differences based on sex or age. Among 49 (25.7%) cases, the initial swab test yielded negative findings after close contact with a patient with COVID-19. Sex, age, symptoms, and the time of sampling were not different between individuals with an initial negative swab test finding and those who initially tested positive after close contact. In the overall sample, the estimated probability of testing positive was 0.74 on day 1 after close contact, which increased to 0.77 between days 5 and 8. In the 3 different scenarios for scheduled repeated testing intervals (3, 5, and 7 days) in the subgroup of individuals with an initially negative swab test finding, the probability peaked on the sixth, ninth and tenth, and 13th and 14th days, respectively. CONCLUSIONS: Swab tests can initially yield false-negative outcomes. The probability of testing positive increases from day 1, peaking between days 5 and 8 after close contact with a patient with COVID-19. Early testing, especially in this final time window, is recommended together with a health surveillance program scheduled in close intervals.

Comparing swab- and different symptoms-based strategies to ascertain COVID-19 recovery in healthcare workers: a cost-effectiveness analysis.

OBJECTIVE: Given the human and economic cost of the COVID-19 pandemic, protecting healthcare workers (HCW) and ensuring continuity of care is critical. The aim of this study is to evaluate the cost-effectiveness of different strategies to ascertain COVID-19 recovery in HCWs. METHODS: Data were collected from the hospital health surveillance program on HCWs at the University Hospital of Verona between 29/02/2020 and 14/04/2021. The diagnosis of SARS-CoV-2 infection and the assessment of the recovery were made through RT-PCR on oro-nasopharyngeal swab-sample. Recovery time and probability were estimated through Kaplan-Meier estimate. For each recovery assessment strategy costs (laboratory diagnostics and human resources), expressed in local currency (euro-€), and working days saved (WDS-effectiveness) were estimated. A decision-tree was created where each knot was a time point scheduled by the different recovery assessment strategies. A Monte Carlo simulation method was used, and probabilistic sensitivity analysis assessed the effect of input uncertainty. RESULTS: In the study period 916 (9.9%) HCWs tested positive. Recovery time through symptom-based strategy (21 days 0.95 CI 16-24) was significantly lower compared to swab-based one (25 days 0.95 CI 23-28, p < 0.001). The swab-based strategy was dominated by all symptoms-based ones. Symptoms-based with a swab on days 14 and 17 had an ICER of 2 €/WDS and 27 €/WDS compared to the one scheduled on days 10 and 17 and with only one swab on the 17th day. CONCLUSIONS: Scheduling swabs on days 14 and 17 in a symptom-based strategy was the most cost-effective, saving 7.5 more working days than the standard one with swabs on days 10 and 17.