Monitoring Arriving Travelers in California at the Start of COVID-19, 2020

When the new coronavirus threatened to reach the United States early in 2020, federal officials began screening travelers from China — and later Iran — and passing their contact details to states for follow-up. The California Department of Public Health (CDPH) described its experience in CDC's MMWR.

How the system worked

On December 31, 2019, Chinese authorities reported a novel coronavirus in people with pneumonia in Wuhan. An executive order limiting travel from China took effect on February 3, 2020. U.S. citizens, lawful permanent residents and their families who had been in China in the previous 14 days could still enter, but all flights from China were directed to 11 U.S. airports.

• At the airport, Customs and Border Protection agents asked arriving travelers about symptoms, collected their demographic and contact information and gave them self-monitoring instructions; CDC oversaw further screening of anyone with symptoms.
• CDC forwarded every arriving traveler's information, sick or not, to state health departments through its Epi-X network. From March 5, people who had been in Iran within 14 days were added.
• Local health departments were asked, if they had the resources, to contact travelers, interview them about symptoms and exposures, and oversee 14 days of quarantine, self-monitoring or both, following CDC risk criteria.

California's workload

From February 3, CDPH moved medical officers, epidemiologists and other staff onto a Return Traveler Monitoring team. Between February 5, when notifications began arriving, and the program's end on March 17, it processed:

• 2,266 notifications covering 12,061 travelers — a median of 39 notifications a day, and 1,431 travelers a week;
• 1,694 staff hours, the equivalent of six people working full time for 7 weeks, 576 of them (34%) outside normal working hours. The time spent by local health departments wasn't measured.

The data were messy. Of the records, 1,523 (13%) had errors staff could identify:

• 1,135 (75% of those) lacked a correct U.S. phone number;
• 603 (40%) were duplicates;
• 487 (32%) had too little location information, or the traveler lived outside California.

Many names looked misspelled and some birth dates were out of range, with no flight manifests to check against. After fixing what they could, staff sent 11,574 records (96%) to 51 of California's 61 local health jurisdictions — from one traveler in some to 4,852 in another.

What it found

Of those 11,574 travelers, only three could be matched by name and birth date to any of the 26,182 COVID-19 cases reported in California by April 15. Two had come from Iran and were tested a few days after arriving. The third had come from China but wasn't tested until March 30, about six weeks after returning — after local follow-up would have ended. Nationally, only 14 confirmed cases had been found among some 268,000 screened travelers by April 21, 2020.

Despite the intense effort, screening did not keep COVID-19 out of California. The state's first case with no known link to a traveler or another patient was reported on February 26. Several counties issued shelter-in-place orders on March 16, and CDPH ended traveler monitoring the next day.

Why it fell short

• Incomplete and inaccurate information from the federal screening delayed or prevented contact with travelers.
• The sheer numbers overwhelmed resources — compare the 2014–2015 Ebola monitoring, when California followed a median of 21 travelers a week over 17 months.
• Spread before or without symptoms. Ebola is contagious only after obvious symptoms appear, which makes monitoring effective; COVID-19 can spread before symptoms or without them, limiting what airport screening can catch.

The study had limits too: infected travelers without symptoms, or who never sought testing, wouldn't show up as cases; errors in names and birth dates hampered matching; and as community spread grew, it became harder to tell whether an infection came from travel.

Lessons

• Better passenger data — such as airlines providing electronic flight manifests to federal officials to pass on to states — and flexible tools like text messaging would help reach at-risk travelers quickly for testing and contact tracing, as would extra staff for health departments with limited capacity.
• Traveler monitoring is likely most useful early, when the goal is to prevent introductions and delay community spread. Once transmission is local, health departments must weigh its cost against mitigation measures.
• Later in a pandemic, after mitigation brings transmission down, focused monitoring with accurate traveler information and more public health staff might help keep case numbers low if new waves arrive.

Travelers processed by CDPH each day, by country of travel (MMWR).

Sources

• Myers JF, Snyder RE, Porse CC, et al. "Identification and Monitoring of International Travelers During the Initial Phase of an Outbreak of COVID-19 — California, February 3–March 17, 2020," MMWR Vol. 69, No. 19, CDC; a work of the United States government in the public domain.