As news about the coronavirus (nCov) that has spread from Wuhan, China, continues to emerge, Marc Lipsitch, professor of epidemiology at Harvard T.H. Chan School of Public Health and director of the Center for Communicable Disease Dynamics, offered a “tweetorial” on key questions and terminology in infectious disease epidemiology to help make sense of current information about the virus. Highlights are below. See @CCDD_HSPH for the complete thread.
How and when did the outbreak start?
Analyses of nCoV viruses suggest it was introduced in November or December 2019. The similarities in the viruses suggest only one or a small number of introductions from animals.
Combined with data on people’s travel history and exposure, genomic data can help distinguish between imported cases and local transmission. Most cases outside of China have been imported, but local transmission is beginning to be reported.
How do we know the number of cases?
Reports from health facilities and government agencies are key sources of information. But at the beginning of an outbreak, even if diagnostics are available quickly (like for nCoV), the total number of cases will be uncertain.
If we assume many cases are missed in the outbreak’s epicenter (Wuhan), but detection is near 100% in international travelers, case incidence in travelers combined with daily probability of travel and mean detection time can be used to estimate total number of cases at the epicenter.
Active surveillance, which involves testing an (ideally representative) sample of the population exhibiting symptoms, can be used to estimate total number cases.
What does R0 mean, and what does it tell us?
R0, the basic reproductive rate or number, is the value that summarizes how contagious a pathogen is.
R0 is the average number of people one case will infect if it is introduced into an entirely susceptible population. If R0 >1 each infected person will transmit to >1 person, creating epidemic potential.
R0 does not give us any information on the total number of people who are currently infected. It is also not a measure of disease severity. It only tells us on average how many people each person will infect, not how severe those infections will be.
At the beginning of an outbreak, estimates are challenging given the limited data.
How do we contain an outbreak? What makes containment harder or easier?
Until vaccines and treatment are available, we must rely on nonpharmaceutical interventions. These include measures to decrease contacts, such as monitoring for symptoms, isolation and quarantine.
Infectiousness before or without symptoms makes control harder because cases may not be identified before transmission, or may be missed completely yet may transmit. Quarantine of contacts can reduce these effects but has feasibility and social liberty costs.
Why is understanding the full clinical spectrum—from severe cases to cases with few or no symptoms—of a novel infectious pathogen relevant for public health response to an outbreak?
Severe cases, while leading to hospitalization and potential death, are more likely to be detected and reported. Infected individuals with no or few symptoms are more likely to remain unnoticed. If these individuals contribute to transmission, the outbreak is harder to control.
On the other hand, if mild or asymptomatic cases are common and do not contribute much to transmission, then they will aid control because fewer individuals will need care, and their infections will likely give them immunity to reinfection—at least for some time
Culled from harvard.edu