Insights into population behavior during the COVID-19 pandemic from cell phone mobility data and manifold learning
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (ScreenIT)
Abstract
Understanding the complex interplay between human behavior, disease transmission and non-pharmaceutical interventions during the COVID-19 pandemic could provide valuable insights with which to focus future public health efforts. Cell phone mobility data offer a modern measurement instrument to investigate human mobility and behavior at an unprecedented scale. We investigate aggregated and anonymized mobility data, which measure how populations at the census-block-group geographic scale stayed at home in California, Georgia, Texas and Washington from the beginning of the pandemic. Using manifold learning techniques, we show that a low-dimensional embedding enables the identification of patterns of mobility behavior that align with stay-at-home orders, correlate with socioeconomic factors, cluster geographically, reveal subpopulations that probably migrated out of urban areas and, importantly, link to COVID-19 case counts. The analysis and approach provide local epidemiologists a framework for interpreting mobility data and behavior to inform policy makers’ decision-making aimed at curbing the spread of COVID-19.
Article activity feed
-
-
SciScore for 10.1101/2020.10.31.20223776: (What is this?)
Please note, not all rigor criteria are appropriate for all manuscripts.
Table 1: Rigor
NIH rigor criteria are not applicable to paper type.Table 2: Resources
Software and Algorithms Sentences Resources The Laplacian Eigenmaps algorithm was implemented using the SpectralEmbedding function from sklearn.manifold module of scikit-learn package [30] in Python 3. Pythonsuggested: (IPython, RRID:SCR_001658)To implement the trustworthiness metric, we used the function trustworthiness from sklearn.manifold of scikit-learn package [30] in Python 3 with default parameters (5 neighbors, to capture the local structure). scikit-learnsuggested: (scikit-learn, RRID:SCR_002577)3.4 Statistical Testing: To test the difference of the socio-economic covariates distributions between clusters, we used the … SciScore for 10.1101/2020.10.31.20223776: (What is this?)
Please note, not all rigor criteria are appropriate for all manuscripts.
Table 1: Rigor
NIH rigor criteria are not applicable to paper type.Table 2: Resources
Software and Algorithms Sentences Resources The Laplacian Eigenmaps algorithm was implemented using the SpectralEmbedding function from sklearn.manifold module of scikit-learn package [30] in Python 3. Pythonsuggested: (IPython, RRID:SCR_001658)To implement the trustworthiness metric, we used the function trustworthiness from sklearn.manifold of scikit-learn package [30] in Python 3 with default parameters (5 neighbors, to capture the local structure). scikit-learnsuggested: (scikit-learn, RRID:SCR_002577)3.4 Statistical Testing: To test the difference of the socio-economic covariates distributions between clusters, we used the Kolmogorov-Smirnov [38, 39] test as implemented in kstest function of scipy.stats package in Python 3. scipysuggested: (SciPy, RRID:SCR_008058)Results from OddPub: We did not detect open data. We also did not detect open code. Researchers are encouraged to share open data when possible (see Nature blog).
Results from LimitationRecognizer: We detected the following sentences addressing limitations in the study:We acknowledge several limitations of the mobility data and challenges in linking behavior to demographic variables. SafeGraph aggregates mobility data from many uncoordinated sources on the locations of millions of cell phones. These phones are not systematically tracked, and the GPS data might not be precise. The data are then aggregated by census block group and filtered to preserve the privacy of the mobile device owners. It is difficult to ascertain how well a set of mobility data represents the general population [5, 6]. Different states, and segments of the population, have different levels of coverage that are hard to correct for [45]. This is further complicated by likely gaps in coverage for high-risk populations such as migrant agricultural workers. However, the associations we found between mobility and other factors are consistent with those found in other datasets and are quite plausible. We studied the fraction of mobile devices that stayed at home each day, but this is just one metric than can be derived from the mobility data. Other measures, such as the mean length of time spent outside the home, the distance traveled from the home, or even the number of trips to stores, could provide additional insight into the population’s response to the pandemic. The demographic data in this study was from the 2018 American Community Survey, which we believe generally reflects the population in 2020 but might not accurately characterize the demographics of the most rapid...
Results from TrialIdentifier: No clinical trial numbers were referenced.
Results from Barzooka: We did not find any issues relating to the usage of bar graphs.
Results from JetFighter: We did not find any issues relating to colormaps.
Results from rtransparent:- Thank you for including a conflict of interest statement. Authors are encouraged to include this statement when submitting to a journal.
- Thank you for including a funding statement. Authors are encouraged to include this statement when submitting to a journal.
- No protocol registration statement was detected.
-
