Skip to main content

COVID-19 Contact Tracing in Two Counties – North Carolina, June-July 2020

Our take —

In two neighboring North Carolina counties, no close contacts were identified in a large proportion of confirmed COVID-19 cases, over one-fourth of enumerated close contacts were unreachable, and there was a long delay between case identification and contact notification. When case counts are high, health departments may lack the capacity to conduct timely and thorough contact investigations. The low response rates may also reflect public reluctance to subject contacts to disruptive isolation and quarantine procedures.

Study design


Study population and setting

Between June and July 2020, confirmed COVID-19 cases in Mecklenburg (n=7,116) and Randolph Counties (n=589) in North Carolina, USA were asked to report close contacts. Three attempts were made to contact each enumerated close contact; if reached, each contact was referred for COVID-19 testing, irrespective of new-onset signs or symptoms. Close contacts were considered lost to follow-up after three failed communication attempts.

Summary of Main Findings

Among laboratory-confirmed COVID-19 cases, 52% in Mecklenburg and 66% in Randolph completed the initial case investigation and named close contacts. Of note, 23% of index COVID-19 cases in Mecklenburg were lostto follow-up. A total of 13,401 and 1,146 close contacts were elicited from index cases in Mecklenburg and Randolph, respectively. Of these, one-fourth (25%) in Mecklenburg and nearly half (47%) in Randolph were unreachable. Overall, the proportion of enumerated contacts who were reached, counseled to quarantine, and monitored daily for symptoms by the health department was 73% in Mecklenburg and 48% in Randolph. In Mecklenburg, 137 (1.0%) contacts had laboratory-confirmed SARS-CoV-2 infection (the total number of tests was not reported), while 69 (6%) contacts tested positive in Randolph (293 contacts (26%) were tested). The median time from index case specimen collection to contact notification and quarantining was 6 days in both counties.

Study Strengths

The authors considered contacts from all confirmed COVID-19 cases in these counties during the study period, and compared results (e.g., proportion of cases with enumerated contacts, time to contact investigation) in two neighboring counties of North Carolina with variable population density and COVID-19 cases per capita.


Information about the types of close contacts (e.g., household, occupational) elicited from index cases was not available or reported. Increased testing and rising cases per capita in both counties may have bottlenecked the health system; attribution of the 6-day median time from index case specimen collection to contact notification to contact tracing performance is, therefore, incomplete. Because the investigators did not collect information on health department capacity (i.e., number of contact tracers per case), no inferences can be made about the potential impact of health system capacity on the outcomes of contact tracing. The number of contacts tested in Mecklenburg county was not reported. Lastly, given the study’s short time interval and inclusion of data from only two counties, these results may not be generalizable to other local health departments or reflect performance of these same counties at other points in time.

Value added

This is among the first studies to describe the testing and tracing continuum from index case identification to counseling and monitoring of contacts in the United States.

This review was posted on: 13 October 2020