Occurrence and Timing of Subsequent SARS-CoV-2 RT-PCR Positivity Among Initially Negative Patients

Dustin R. Long, Saurabh Gombar, Catherine A. Hogan, Alexander L. Greninger, Vikas N. O’Reilly-Shah, Chloe Bryson‐Cahn, Bryan Stevens, Arjun Rustagi, Keith R. Jerome, Christina S. Kong, James L. Zehnder, Nigam H. Shah, Noel S. Weiss, Benjamin A. Pinsky, Jacob E. SunshinemedRxiv 2020 · DOI

Abstract

Two independent research teams report the similar finding that, among initially negative patients subjected to repeat SARS-CoV-2 RT-PCR testing, the occurrence of a newly positive result within 7 days is uncommon. These observations suggest that false negative results at the time of initial presentation do occur, but potentially at a lower frequency than is currently believed. Although it is not possible to infer the clinical sensitivity of NP SARS-CoV-2 RT-PCR testing using these data, they may be used in combination with other reports to guide the use and interpretation of this common testing modality.

Introduction

During the initial surge of the COVID-19 pandemic in early 2020, the medical community faced significant uncertainty regarding the reliability of nasopharyngeal (NP) swab RT-PCR testing. Concerns about high false-negative rates—potentially driven by low viral shedding, suboptimal sampling techniques, or the timing of the test relative to symptom onset—created widespread "false negative anxiety." This uncertainty complicated critical clinical decisions, including the management of personal protective equipment (PPE), patient isolation protocols, and the clearance of healthcare workers for return to duty. This study seeks to quantify the frequency of "discordant" results—where an initial negative test is followed by a positive result within a short timeframe—to provide a data-driven assessment of test performance in a real-world clinical setting.

Methods

This retrospective cohort study utilized clinical data from two independent healthcare systems: the University of Washington (UW) and Stanford Health Care. The researchers tracked all patients undergoing initial NP swab testing between March 2 and April 7, 2020.

The primary metric was the "conversion rate," defined as the proportion of initially negative patients who tested positive upon re-testing within a seven-day window. Testing was conducted using a variety of platforms, including laboratory-developed tests (LDTs) and commercial assays such as the Hologic Panther Fusion, Roche, and DiaSorin systems. Additionally, the researchers analyzed cycle threshold (Ct) values for the positive re-tests; in RT-PCR, higher Ct values generally indicate lower viral loads.

Results

The study dataset comprised 23,126 RT-PCR tests from 20,912 unique patients (8,977 at UW and 11,935 at Stanford). Approximately 9% of the total population tested positive on their initial swab.

Of the patients who initially tested negative, 626 were re-tested within the seven-day observation window. The findings were remarkably consistent across both sites:

  • Conversion Rate: Only 3.5% of those re-tested became positive (4.1% at UW and 2.8% at Stanford).
  • Stability: 96.5% of patients who warranted a re-evaluation for any reason remained negative on all subsequent tests within the observation window.
  • Symptomatic Subgroup: At UW, among patients re-tested specifically for persistent or worsening symptoms, the conversion rate was only 4.3%.
  • Timing and Viral Load: The mean interval between the initial negative test and the first positive re-test was 4.0 days. The mean Ct value for these newly positive results was 28.5, suggesting that many of these cases involved patients with lower viral loads, likely at the very beginning or end of their infectious period.

Discussion

These results suggest that the clinical reliability of NP SARS-CoV-2 RT-PCR is higher than early anecdotal reports and media coverage suggested. The low conversion rate, even among patients with persistent symptoms, indicates that such symptoms in a test-negative individual are more likely due to other respiratory pathogens—such as Influenza A or Rhinovirus—than a missed COVID-19 diagnosis.

However, several limitations must be acknowledged. First, selection bias is inherent, as the study only captures data for patients who clinicians chose to re-test. The status of the thousands of negative patients who did not undergo repeat testing remains unknown; in most cases, this likely reflects the resolution of symptoms or an alternative diagnosis. Second, it is impossible to definitively distinguish between an initial false negative and a newly acquired infection occurring within the seven-day window. Finally, the absence of a "gold standard" confirmatory mechanism, such as serology, prevents the calculation of absolute clinical sensitivity.

Conclusion

While false-negative results for SARS-CoV-2 RT-PCR do occur, this large-scale analysis demonstrates that they are relatively uncommon in clinical practice. These findings provide much-needed reassurance that a negative result, when integrated with clinical judgment, is a highly reliable indicator for patient management. Future work incorporating orthogonal diagnostic approaches will be necessary to further refine the quantification of test performance.

Summary updated

Abstract

Two independent research teams report the similar finding that, among initially negative patients subjected to repeat SARS-CoV-2 RT-PCR testing, the occurrence of a newly positive result within 7 days is uncommon. These observations suggest that false negative results at the time of initial presentation do occur, but potentially at a lower frequency than is currently believed. Although it is not possible to infer the clinical sensitivity of NP SARS-CoV-2 RT-PCR testing using these data, they may be used in combination with other reports to guide the use and interpretation of this common testing modality.
BibTeX
@misc{long2020occurrence,
  author = {Dustin R. Long and Saurabh Gombar and Catherine A. Hogan and Alexander L. Greninger and Vikas N. O’Reilly-Shah and Chloe Bryson‐Cahn and Bryan Stevens and Arjun Rustagi and Keith R. Jerome and Christina S. Kong and James L. Zehnder and Nigam H. Shah and Noel S. Weiss and Benjamin A. Pinsky and Jacob E. Sunshine},
  title = {{Occurrence and Timing of Subsequent SARS-CoV-2 RT-PCR Positivity Among Initially Negative Patients}},
  year = {2020},
  doi = {10.1101/2020.05.03.20089151},
  howpublished = {medRxiv},
}

Loading related content...

Figure