Pandemics: Early Detection, Networks, Spreaders - The a16z Show Recap

Podcast: The a16z Show

Published: 2020-05-23

Duration: 2046

Guests: Nicholas Christakis

What Happened

Pandemics are predictable in occurrence, but not in intensity or timing. Jorge Conde, a16z general partner, explains that early detection, rather than relying on delayed warnings from traditional systems like the CDC flu trackers, is crucial. The analogy is drawn to weather forecasts that arrive too late to inform decisions.

Nicholas Christakis, a sociologist and physician, shares insights from the H1N1 pandemic about using social network 'sensors' for early detection of disease spread. His lab developed the Hunala app, which uses crowdsourced data to assess personal risk of contracting respiratory illnesses, functioning like Waze but for health.

The role of network centrality in predicting pandemics is highlighted. Individuals with extensive social connections can act as early indicators of disease spread, akin to canaries in coal mines. This concept is utilized in the Hunala app, which improves its effectiveness as more users join.

Christakis discusses the potential of mobility data and population flows, such as those observed in China, to predict the spread of COVID-19. This data, combined with real-time analysis, could significantly enhance pandemic response and resource allocation.

The episode examines the concept of superspreaders, individuals who are more likely to contract and transmit diseases due to their high degree of social interaction. While predicting specific superspreaders is uncertain, understanding their role can improve epidemic forecasting.

Christakis also expresses optimism about human nature, emphasizing qualities like cooperation and teaching as discussed in his book 'Blueprint'. He believes these traits can outweigh negative human behaviors, supporting the argument with evidence from his research.

Key Insights

• Pandemics can be anticipated in terms of occurrence but not in their intensity or precise timing, making early detection vital for effective response.
• The Hunala app leverages social network data to provide personalized risk assessments of contracting respiratory illnesses, functioning similarly to Waze for health.
• Network centrality plays a crucial role in early pandemic detection, where individuals with numerous social connections serve as key indicators of disease spread.
• Mobility data and population flows, like those observed in China during COVID-19, offer valuable insights for predicting the spread of pandemics and optimizing resource allocation.