Rapid Genotyping in a Los Angeles County Hepatitis A Outbreak, 2018–2019

Los Angeles County covers 4,058 square miles and is home to about 10 million people, an estimated 59,000 (0.6%) of whom are homeless on any given night. In November 2018 the county's Department of Public Health (DPH) learned of a hepatitis A case in a person experiencing homelessness. What followed, described in CDC's Morbidity and Mortality Weekly Report, showed how genotyping the virus early, while there were still only a few cases, can make an outbreak response faster and better targeted.

The first cases

On November 10, 2018, an acute hepatitis A case was reported in a person experiencing homelessness who used methamphetamine (patient A). Patient A had been brought by ambulance from a street intersection (intersection X) to a hospital emergency department but left against medical advice and could not be found for an interview. Three days later, another homeless person who used methamphetamine (patient B) was seen at a second hospital and diagnosed with hepatitis A.

Patient B told investigators about using public restrooms in restaurants and stores at intersection X, named patient A as an ill contact, and named an acquaintance (patient C) who had been reported with hepatitis A a month earlier but never interviewed. Patient C also spent time at the intersection, lived unsheltered nearby and had shared drug equipment with patient A. The state laboratory genotyped the virus from patients A and B and found the same genotype IB strain, called CA Cluster A, matching a recent outbreak strain.

Enhanced surveillance

With three linked cases, DPH set up enhanced surveillance across a 50-square-mile area bounded by four major freeways, on the assumption that the roads would limit people's movements. It immediately sought and held every blood sample positive for hepatitis A antibodies (anti-HAV IgM) from people living in the area, and sent qualifying samples to the state laboratory for genotyping. This continued until 100 days — twice the incubation period — had passed without a new outbreak case.

A case counted as outbreak-associated if the person lived or spent time in the area between October 15, 2018, and April 29, 2019, and either carried the CA Cluster A strain or was linked to someone who did.

What surveillance found

Nineteen anti-HAV IgM-positive cases were reported from the area between November 10, 2018, and April 29, 2019. Ten met neither the national surveillance definition nor the outbreak definition. Five met the surveillance definition but not the outbreak one; two met both; and two met the outbreak definition only.

• A senior living campus. Four more outbreak cases — patients E, G, H and J — occurred among people with no history of homelessness or illegal drug use, all connected to a senior living campus within two blocks of intersection X: two residents, a staff member and a visitor. Patients E, H and J carried the outbreak strain. Patients G and J lacked symptoms of acute hepatitis and so did not meet the national surveillance definition, but J had the outbreak strain and G was linked to E. Two of the four reported visiting businesses at intersection X.
• A case ruled out. Patient K, who was homeless, used methamphetamine and lived near the intersection, was first counted as part of the outbreak — but genotyping showed a different strain, and the case was reclassified.

In all, seven people had outbreak-associated infections. Five were hospitalized, and none died. DPH declared the outbreak over on April 29, 2019.

The response

• Alerts. DPH told the county's health care providers to watch for hepatitis A among people experiencing homelessness or using drugs and to report suspected cases immediately.
• Vaccination. From November 22, DPH targeted vaccination at people with similar risk factors where patients A, B and C had lived, and from the week of December 17 held clinics for residents and staff of the senior living campus. From November 22, 2018, to March 13, 2019, it gave 857 doses of hepatitis A vaccine at the campus, drug treatment centers, food pantries and homeless shelters.
• Restaurants. Environmental health staff visited 22 restaurants near the intersection and the campus to check sanitation and hygiene and educate staff, and emailed information to every restaurant in the two ZIP codes where patients A, B and C had spent time while infectious.

What genotyping added

Since 2016, large hepatitis A outbreaks across the United States have disproportionately affected people experiencing homelessness or using drugs, and genotyping had mostly been used after the fact. Here it was used in near real time, and it helped in three ways:

1. Narrowing the response. Ruling out patient K shortened enhanced surveillance by about 3 weeks and let DPH investigate that case as a separate chain of transmission.
2. Linking unconnected cases. Matching strains tied the senior living campus to the cluster among homeless people despite no other obvious link, pointing to shared community spaces such as public restrooms and prompting sanitation outreach to businesses.
3. Confirming extra cases. It confirmed infections in people who did not meet the national case definition, allowing earlier control measures.

It is not easy. Serum for sequencing must be processed and frozen within 72 hours of collection, which commercial laboratories typically do only on request, and routine case investigation often takes longer. The extra staff time and shipping were manageable here because the effort was limited in time and place; with widespread transmission, it might not be feasible.

Limitations

• The CA Cluster A strain causes many hepatitis A outbreaks nationally and may be endemic, so the senior living cases could represent a separate chain of transmission.
• Genotype results depend on epidemiologic context, and patients may not have disclosed all risk factors.
• Mishandled specimens or samples collected long after symptoms began can reduce the sensitivity of molecular testing, so some positive antibody results may have been misclassified.

Sources

• Meredith Haddix, Rachel Civen, Jill K. Hacker, Will Probert, Sarah New, Nicole Green, Peera Hemarajata and Prabhu Gounder, "Use of Molecular Epidemiology to Inform Response to a Hepatitis A Outbreak — Los Angeles County, California, October 2018–April 2019," Morbidity and Mortality Weekly Report, CDC: https://www.cdc.gov/mmwr/volumes/69/wr/mm6926a3.htm
• The report's figure, prepared by the county and state health departments, is not reproduced.