Can you comment on the COVID PCR monitoring program at Amazon or other similar large companies? Should workers be tested randomly or is there an epidemiologic formula to ensure the proper sample size and frequency of testing?

Until a safe and effective vaccine is broadly available, the cornerstone of a workplace COVID-19 prevention program is reducing the presence and spread of SARS-CoV-2. This includes symptom screening, paid sick and quarantine leave, supporting work from home when feasible, increased distancing, reducing contact time, use of face coverings (including respirators in appropriate settings), and increasing airflow. Employers should also encourage employees to follow public health recommendations and practice safe behaviors outside of work to reduce the risk of infection for themselves, their families, and co-workers. Testing to detect asymptomatic and pre-symptomatic workers can be considered to supplement these program elements but has significant costs and limitations that need to be considered.

Factors to consider in deciding whether to pursue a testing program include:
  • Type and performance of test - RT-PCR and antigen tests are increasingly available. In general, RT-PCR tests are more sensitive and therefore more likely to be positive in people infected with SARS-CoV-2. However, even sensitive tests may not perform well during the pre-symptomatic period. In one study, the probability of a false-negative one day before symptoms developed was 67%.1
  • Personnel RequirementsWhether the samples need to be obtained by medical personnel
  • CLIAFor in-house testing, whether a certified lab is in place or whether a new lab must pursue certification under the Clinical Laboratory Improvement Act (CLIA)  
  • Time to receive results – To be useful in identifying workers who need to be isolated, test results need to be available quickly. Tests conducted in-house may have a shorter turn-around time. Since tests that are sent out may take a day or more before results are available, it is essential to understand the time-frame for results.
  • Cost – The cost of testing varies, with some tests priced at $100 or more. RT-PCR tests are usually more expensive than antigen tests. Depending on the size of the workforce and the frequency of testing, program costs can be substantial.
  • Risk for workplace transmission – Identifying an asymptomatic or pre-symptomatic employee through testing creates an opportunity for isolation prior to symptoms. This would be most helpful in a setting where it is more challenging to limit spread to co-workers or others. 
  • Consequences of workplace transmission – The greater the risk to the business from an infected worker, the greater the value from early detection. The National Basketball Association (NBA) combined extensive testing and maintained a 14-day quarantine before starting play in their restricted access setting (the “bubble”). Similarly, employers who send groups of employees to work together in remote locations, such as mining, oil platforms, and fishing vessels, face the risk of the entire group becoming infected, putting the employees and the work at risk.3 Congregate living facilities, including skilled nursing facilities and prisons, face a high risk of spread if employees are infected. Colleges have also shown high rates of spread, forcing many schools to send students home within weeks of starting classes (although some colleges with extensive testing programs have been able to continue a modified residential program). 
  • Rate of infection in the surrounding community – The higher the local prevalence, the greater the chances of identifying pre-symptomatic and asymptomatic cases through screening.

There is no simple formula for deciding whether to pursue testing and at what frequency. If testing is pursued, more frequent testing has the potential to detect more cases, but at increasing cost. The growing availability of less expensive antigen tests may increase employers’ interest in pursuing testing, as a lower cost test may allow for more frequent testing. A modeling study suggests that successful surveillance, including time to first detection and outbreak control, depends largely on frequency of testing and the speed of reporting, and is only marginally improved by high test sensitivity.

If a testing program is pursued, a robust employee education program is needed to encourage continued safe behaviors and to help employees understand how to interpret positive and negative results. False negative results may lead people to believe that they are not infectious. False positive results are less common but can lead to unnecessary worry and the need for isolation. No published studies have assessed perception of test results from workplace screening programs. 

Outside of healthcare and skilled nursing facilities, there are few publications on workplace testing programs. At this time, it appears that few large employers have embarked on a broad employee testing program. Amazon, on their blog, reported that they had screened 1,372,000 Amazon and Whole Foods Market front-line employees across the U.S. employed at any time from March 1 to September 19, 2020. They indicated that 19,816 employees have tested positive or been presumed positive for COVID-19—42% lower than what they calculate to be the expected number.5 It is not clear what role their testing program played in any reduction in cases

Each employer needs to weigh all of the relevant factors to determine the strategy that best fits their situation. This strategy should be periodically reassessed as the factors for consideration change.


  1. Kucirka LM, Lauer SA, Laeyendecker O, Boon D, Lessler J. Variation in False-Negative Rate of Reverse Transcriptase Polymerase Chain Reaction-Based SARS-CoV-2 Tests by Time Since Exposure. Ann Intern Med. 2020;173(4):262-267. doi:10.7326/M20-1495
  2. Bontemps T. In documents, NBA details coronavirus testing protocols, including 2-week resting period for positive tests., June 16, 2020.
  3. CDC. SARS-CoV-2 Testing Strategy: Considerations for Non-Healthcare Workplaces. July 22, 2020.
  4. Larremore DB, Wilder B, Lester E, et al. Test sensitivity is secondary to frequency and turnaround time for COVID-19 surveillance. Preprint. medRxiv. 2020;2020.06.22.20136309. Published 2020 Jun 27. doi:10.1101/2020.06.22.20136309
  5. Amazon. Update on COVID-19 Testing. The Amazon blog, October 1, 2020.

Return to Forum

The Forum does not necessarily represent an official ACOEM position. The Forum is intended for health professionals and is not intended to provide medical or legal advice, including illness prevention, diagnosis or treatment, or regulatory compliance. Such advice should be obtained directly from a physician and/or attorney. Questions are answered with the best available data or recommendations at the time.