Lymelight: forecasting Lyme disease risk using web search data
Abstract Lyme disease is the most common tick-borne disease in the Northern Hemisphere. Existing estimates of Lyme disease spread are delayed a year or more. We introduce Lymelight—a new method for monitoring the incidence of Lyme disease in real-time. We use a machine-learned classifier of web sear...
Main Authors: | , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Nature Publishing Group
2020-02-01
|
Series: | npj Digital Medicine |
Online Access: | https://doi.org/10.1038/s41746-020-0222-x |
id |
doaj-7607b3dae73f4c24804ed0fc3026e323 |
---|---|
record_format |
Article |
spelling |
doaj-7607b3dae73f4c24804ed0fc3026e3232021-02-23T09:42:33ZengNature Publishing Groupnpj Digital Medicine2398-63522020-02-013111210.1038/s41746-020-0222-xLymelight: forecasting Lyme disease risk using web search dataAdam Sadilek0Yulin Hswen1Shailesh Bavadekar2Tomer Shekel3John S. Brownstein4Evgeniy Gabrilovich5GoogleDepartment of Social and Behavioral Sciences, Harvard T.H. Chan School of Public HealthGoogleGoogleComputational Epidemiology Lab, Boston Children’s HospitalGoogleAbstract Lyme disease is the most common tick-borne disease in the Northern Hemisphere. Existing estimates of Lyme disease spread are delayed a year or more. We introduce Lymelight—a new method for monitoring the incidence of Lyme disease in real-time. We use a machine-learned classifier of web search sessions to estimate the number of individuals who search for possible Lyme disease symptoms in a given geographical area for two years, 2014 and 2015. We evaluate Lymelight using the official case count data from CDC and find a 92% correlation (p < 0.001) at county level. Importantly, using web search data allows us not only to assess the incidence of the disease, but also to examine the appropriateness of treatments subsequently searched for by the users. Public health implications of our work include monitoring the spread of vector-borne diseases in a timely and scalable manner, complementing existing approaches through real-time detection, which can enable more timely interventions. Our analysis of treatment searches may also help reduce misdiagnosis of the disease.https://doi.org/10.1038/s41746-020-0222-x |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Adam Sadilek Yulin Hswen Shailesh Bavadekar Tomer Shekel John S. Brownstein Evgeniy Gabrilovich |
spellingShingle |
Adam Sadilek Yulin Hswen Shailesh Bavadekar Tomer Shekel John S. Brownstein Evgeniy Gabrilovich Lymelight: forecasting Lyme disease risk using web search data npj Digital Medicine |
author_facet |
Adam Sadilek Yulin Hswen Shailesh Bavadekar Tomer Shekel John S. Brownstein Evgeniy Gabrilovich |
author_sort |
Adam Sadilek |
title |
Lymelight: forecasting Lyme disease risk using web search data |
title_short |
Lymelight: forecasting Lyme disease risk using web search data |
title_full |
Lymelight: forecasting Lyme disease risk using web search data |
title_fullStr |
Lymelight: forecasting Lyme disease risk using web search data |
title_full_unstemmed |
Lymelight: forecasting Lyme disease risk using web search data |
title_sort |
lymelight: forecasting lyme disease risk using web search data |
publisher |
Nature Publishing Group |
series |
npj Digital Medicine |
issn |
2398-6352 |
publishDate |
2020-02-01 |
description |
Abstract Lyme disease is the most common tick-borne disease in the Northern Hemisphere. Existing estimates of Lyme disease spread are delayed a year or more. We introduce Lymelight—a new method for monitoring the incidence of Lyme disease in real-time. We use a machine-learned classifier of web search sessions to estimate the number of individuals who search for possible Lyme disease symptoms in a given geographical area for two years, 2014 and 2015. We evaluate Lymelight using the official case count data from CDC and find a 92% correlation (p < 0.001) at county level. Importantly, using web search data allows us not only to assess the incidence of the disease, but also to examine the appropriateness of treatments subsequently searched for by the users. Public health implications of our work include monitoring the spread of vector-borne diseases in a timely and scalable manner, complementing existing approaches through real-time detection, which can enable more timely interventions. Our analysis of treatment searches may also help reduce misdiagnosis of the disease. |
url |
https://doi.org/10.1038/s41746-020-0222-x |
work_keys_str_mv |
AT adamsadilek lymelightforecastinglymediseaseriskusingwebsearchdata AT yulinhswen lymelightforecastinglymediseaseriskusingwebsearchdata AT shaileshbavadekar lymelightforecastinglymediseaseriskusingwebsearchdata AT tomershekel lymelightforecastinglymediseaseriskusingwebsearchdata AT johnsbrownstein lymelightforecastinglymediseaseriskusingwebsearchdata AT evgeniygabrilovich lymelightforecastinglymediseaseriskusingwebsearchdata |
_version_ |
1714850553426608128 |