Using cellphone mobility data to make sense of infection heterogeneity

EPID 684
Spatial Epidemiology
4/14/2022


Jon Zelner
[email protected]
epibayes.io

Agenda

  • What are the uses and challenges provided by 📱 mobility data ?

  • How can mobility data tell us about patterns of social segregation and inequality?

  • Smoothing in 2D

Mobility data have become more prominent during the COVID-19 pandemic

Relating mobility to COVID-19 spread in the early days of the pandemic (1)

What can mobility data capture that is relevant to understanding infectious disease transmission?

Mobility data have revealed potential drivers of inequities in COVID-19 risk

Changes in mobility by income decile in early 2020 in the U.S. from Chang et al. (2)

Have become important inputs to theoretical models

Mobility data were used to simulate SARS-CoV-2 transmission under different housing policy scenarios in (3)

Cell phone data come in different shapes and sizes

Relationships between population covered and cellphone data type (4)

A lot of processing and analysis choices take place between data collection and use

Using cellphone data to craft mobility networks in Italy during COVID-19 (5)

What might be worrisome about the rise of mobility data in epidemiology?

  • Privacy 🕵️

  • May drive technocratic vs. principled approaches to policy and decision-making

  • Focus in on micro factors with sufficient attention to the macro drivers of micro behavior.

  • Data may not be representative of variation in intensity and nature of contact.

  • Simulation models built off of these data may appear more authoritative but have no more predictive usefulness than less-complex models.

Mobility Neighborhood Disadvantage and COVID-19 🦠 risk

What is the distinction between residential and mobility neighborhood disadvantage?

How do Levy at all conceptualize the meaning of MND and RND for risk? (6)

What is triple neighborhood disadvantage(TND)?

  • Extends the concept of neighborhood disadvantage to include connected locations.

  • This includes neighborhood of residence, neighborhoods visited by people in a given location, and neighborhoods sending visitors to the neighborhood of interest.

  • MND has been shown to be associated with increased violence risk after adjusting for RND (7,8).

MND predicts increased COVID-19 cases after adjustment for RND

MND & RND are strong - but variable - mediators of associations with race/ethnicty

Mediation of neighborhood percent Black residents by RND, MND

These effects vary across groups

Mediation of neighborhood percent Hispanic residents by RND, MND

Smoothing in 2 Dimensions 🗺️

Final hands-on!

https://sph-umich.shinyapps.io/smoothing/

Also: The last time you have to hear me talk about this Lisa Simpson video!

Next time

  • Last class! (🍪)

  • Come prepared with 1-3 references you think your classmates (present and future) should read.

  • Will compile an annotated bibliography that I will post on the course site for your reference and for anyone else who might be interested.

  • Wrap-up discussion about how the course went.

  • Open office hrs for balance of time for anyone who wants to check in about projects etc.

References

1.
Glanz J, Carey B, Holder J, et al. Where America Didn’t Stay Home Even as the Virus Spread. The New York Times [electronic article]. 2020;(https://www.nytimes.com/interactive/2020/04/02/us/coronavirus-social-distancing.html). (Accessed April 11, 2022)
2.
Chang S, Pierson E, Koh PW, et al. Mobility network models of COVID-19 explain inequities and inform reopening. Nature [electronic article]. 2020;1–8. (http://www.nature.com/articles/s41586-020-2923-3). (Accessed November 10, 2020)
3.
Nande A, Sheen J, Walters EL, et al. The effect of eviction moratoria on the transmission of SARS-CoV-2. Nature Communications [electronic article]. 2021;12(1):2274. (https://www.nature.com/articles/s41467-021-22521-5). (Accessed December 27, 2021)
4.
Grantz KH, Meredith HR, Cummings DAT, et al. The use of mobile phone data to inform analysis of COVID-19 pandemic epidemiology. Nature Communications [electronic article]. 2020;11(1):4961. (http://www.nature.com/articles/s41467-020-18190-5). (Accessed October 5, 2020)
5.
Pepe E, Bajardi P, Gauvin L, et al. COVID-19 outbreak response, a dataset to assess mobility changes in Italy following national lockdown. Scientific Data [electronic article]. 2020;7(1):230. (http://www.nature.com/articles/s41597-020-00575-2). (Accessed April 12, 2022)
6.
Levy BL, Vachuska K, Subramanian SV, et al. Neighborhood socioeconomic inequality based on everyday mobility predicts COVID-19 infection in San Francisco, Seattle, and Wisconsin. Science Advances [electronic article]. 2022;8(7):eabl3825. (https://www.science.org/doi/10.1126/sciadv.abl3825). (Accessed February 18, 2022)
7.
Levy BL, Phillips NE, Sampson RJ. Triple Disadvantage: Neighborhood Networks of Everyday Urban Mobility and Violence in U.S. Cities. American Sociological Review [electronic article]. 2020;85(6):925–956. (https://doi.org/10.1177/0003122420972323). (Accessed April 13, 2022)
8.
Sampson RJ, Levy BL. Beyond Residential Segregation: Mobility-Based Connectedness and Rates of Violence in Large Cities. Race and Social Problems [electronic article]. 2020;12(1):77–86. (https://doi.org/10.1007/s12552-019-09273-0). (Accessed April 11, 2022)