Volume 29, Number 2—February 2023
Dispatch
Estimated Cases Averted by COVID-19 Digital Exposure Notification, Pennsylvania, USA, November 8, 2020–January 2, 2021
, Gabriel Rainisch, A-Mac Harris, Jared Shinabery, Muneeza Iqbal, Amar Pallavaram, Stacy Hilton, Saugat Karki, Patrick K. Moonan, John E. Oeltmann, and Martin I. Meltzer
Table
Reported and estimated program metrics in a study of estimated cases averted by COVID-19 digital exposure notification, Pennsylvania, USA, November 8, 2020–January 2, 2021*
| Program | Reported metrics |
|---|---|
| Standard case investigation and contact tracing |
|
| Cases interviewed, no. (% total cases) | 77,477 (20) |
| Cases named >1 contact, no. (% interviewed cases) | 32,648 (42) |
| No. contacts named | 48,615 |
| Contacts notified and monitored, no. (% identified contacts) | 26,203 (54) |
| Contacts notified but not monitored, no. (% identified contacts) | 418 (1) |
| Timing of case interview, d† | 5 |
| Timing of contact notification, d‡ |
6 |
| Digital exposure notification | |
| Median no. active daily users (% total population)§ | 356,835 (3.2) |
| Cases interviewed and identified as app user, no. (% total cases) | 786 (0.2) |
| No. validation codes generated (% cases that had the app installed) | 579 (74) |
| No. validation codes claimed and certified (% cases that had the app installed) | 390 (50) |
| Timing of digital notification, d‡ |
6 |
| Estimated program effectiveness# | |
| Cases and contacts isolated or quarantined, %** | 7–11.7 |
| Days from infection to isolation or quarantine | 10 |
*Data excludes Philadelphia County. CICT, case investigation and contact tracing. †Reported average number of days from specimen collection to case interview. ‡Reported average number of days from specimen collection to contact notification. §For Android users, the total number of devices that were turned on >1 time in the past 30 d. For iOS users, the total number of devices with >1 session within 30 d of the selected day. During the study period, only persons >18 years of age were eligible to download and activate the digital notification application on their smartphone devices; thus, data provided is equivalent to 4.0% of the eligible population. #Calculations provided in the Appendix . **Includes contacts that later become cases. The range reflects the lowest and highest values across 18 studied scenarios of compliance with quarantine guidelines and the degree of overlap between notifications received via the COVID Alert PA app and by Pennsylvania Department of Health staff members (Appendix Tables 4, 5). The low-value results from a scenario assuming 50% of digital notifications were sent to contacts that were already notified by Department of Health staff members and 10% of notified contacts followed quarantine guidance. The high-value results from a scenario assuming all digital notifications were sent to contacts that were not notified via standard CICT and 50% of notified contacts followed quarantine guidance.