Use of SMS-linked electronic surveys for COVID-19 case investigation and contact tracing - Marin County, CA, USA

Julia M Janssen; Alana McGrath; Rochelle Ereman; Patrick K Moonan; John E Oeltmann; Matthew Willis; Stephen A McCurdy

doi:10.1016/j.puhip.2021.100170

Use of SMS-linked electronic surveys for COVID-19 case investigation and contact tracing - Marin County, CA, USA

Public Health Pract (Oxf). 2021 Nov:2:100170. doi: 10.1016/j.puhip.2021.100170. Epub 2021 Jul 29.

Authors

Julia M Janssen¹, Alana McGrath², Rochelle Ereman², Patrick K Moonan³, John E Oeltmann³, Matthew Willis², Stephen A McCurdy⁴

Affiliations

¹ Epidemic Intelligence Service, U.S. Centers for Disease Control and Prevention, 1600 Clifton Road, Atlanta, GA, USA.
² Marin County Department of Health and Human Services, 3240, Kerner Blvd, San Rafael, CA, USA.
³ COVID-19 Response Team, U.S. Centers for Disease Control and Prevention, 1600 Clifton Road, Atlanta, GA, USA.
⁴ COVID-19 Case and Contact Investigation Unit, Marin County Department of Health and Human Services, 3240 Kerner Blvd, San Rafael, CA, USA.

Abstract

Objectives: We sought to quantify the proportion of contacts reported by persons with COVID-19 through a short message service (SMS)-linked survey in comparison to the proportion of contacts reported during a follow-up phone-interview. We also sought to assess improvement in contact tracing timeliness associated with sending SMS-linked surveys.

Study design: During December 4-15, 2020, persons identified as COVID-19 cases whose data was entered into Marin County's contact tracing database on even days received a SMS-linked survey and persons whose data was entered on odd days did not; all were called for case investigation and contact tracing. Chi-square test and Fisher's exact test were used to compare demographic data. Chi-square test was used to contrast categorical outcomes, and Wilcoxon's rank-sum test was used for continuous outcomes.

Results: Among 350 SMS-linked survey recipients, 85 (24%) responded and 4 (1%) reported contacts using the survey; an additional 303 contacts were reported during phone interviews. Without phone interviews, 99% of reported contacts would have been missed. There was no meaningful difference between study arms in the proportion of contacts notified within 48 h.

Conclusions: This SMS-linked survey had low participation and was not useful for identifying contacts. Phone interviews remained crucial for COVID-19 contact tracing.

Keywords: COVID-19; Case investigation; Contact tracing; Short message service (SMS); Surveys.