“Warning! You’re entering a sick zone”. The construction of risk and privacy implications of disease tracking apps
| Published date | 14 October 2019 |
| Pages | 1046-1062 |
| Date | 14 October 2019 |
| DOI | https://doi.org/10.1108/OIR-03-2018-0075 |
| Author | Scott S.D. Mitchell |
| Subject Matter | Library & information science,Information behaviour & retrieval,Collection building & management,Bibliometrics,Databases,Information & knowledge management,Information & communications technology,Internet,Records management & preservation,Document management |
“Warning! You’re entering a
sick zone”
The construction of risk and privacy
implications of disease tracking apps
Scott S.D. Mitchell
Department of Communication, Carleton University, Ottawa, Canada
Abstract
Purpose –Traditional public health methods for tracking contagious diseases are increasingly
complemented with digital tools, which use data mining, analytics and crowdsourcing to predict disease
outbreaks. In recent years, alongside these public health tools, commercial mobile apps such as Sickweather
have also been released. Sickweather collects information from across the web, as well as self-reports from
users, so that people can see who is sick in their neighborhood. The purpose of this paper is to examine the
privacy and surveillance implications of digital disease tracking tools.
Design/methodology/approach –The author performed a content and platform analysis of two apps,
Sickweather and HealthMap, by using them for three months, taking regular screenshots and keeping a
detailed user journal. This analysis was guided by the walkthrough method and a cultural-historical activity
theory framework, taking note of imagery and other content, but also the app functionalities, including
characteristics of membership, “rules”and parameters of community mobilization and engagement,
monetization and moderation. This allowed me to study HealthMap and Sickweather as modes of governance
that allow for (and depend upon) certain actions and particular activity systems.
Findings –Draw on concepts of network power, the surveillance assemblage, and Deleuze’s control societies,
as well as the data gathered from the content and platform analysis, the author argues that disease
tracking apps construct disease threat as omnipresent and urgent, compelling users to submit personal
information –including sensitive health data –with little oversight or regulation.
Originality/value –Disease tracking mobile apps are growing in popularity yet have received little
attention, particularly regarding privacy concerns or the construction of disease risk.
Keywords Surveillance, Privacy, Mobile apps, Data, Disease tracking, Surveillance medicine
Paper type Research paper
Introduction
Imagine wakingup in the morning and, along with the day’sweatherandbreakingnews,you
get an update about your riskof contracting a contagious disease. It warns youabout which
illnesses are prevalent in your neighborhood, just as a weather forecast might tell you it is
warm and sunny with a chance of afternoon showers. A live map warns when you enter a
“zone of contagion,”orwhen a sick person draws near; a numericalscore corresponds to your
calculatedrisk of becoming diseased. This is nota vision of the future: it is today’s technology.
The app and website Sickweather –a so-called “Facebook for hypocho ndriacs”–collects
informationfrom social media and across theweb, as well as self-reports from its users,so that
people can see who is sick in their neighborhood. In 2011, the site reportedly detected an
outbreak of whooping cough two weeks before public health officials (Miller, 2014).
As the Sickweather website proclaims, “Just as Doppler radar scans the skies for
indicators of bad weather, Sickweather scans social networks for indicators of illness.”
The app identifies “sick zones”and helps “keep you and your family safe”with its
SickScore, an estimation of your current risk of contracting disease[1]. A future version of
the app will allow users to see which of their friends are talking about being sick on
Facebook or Twitter (Kotenko, 2013). Surprisingly, few concerns have been raised about
potential privacy infringements, or the capacity for Sickweather to panic a public already
primed by the media to overreact to news of contagious diseases (Greenberg, 2014;
Online Information Review
Vol. 43 No. 6, 2019
pp. 1046-1062
© Emerald PublishingLimited
1468-4527
DOI 10.1108/OIR-03-2018-0075
Received 9 March 2018
Revised 12 June 2018
Accepted 14 August 2018
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/1468-4527.htm
1046
OIR
43,6
SteelFisher et al., 2015). Caduff (2014) describes how tools such as Sickweather contribute to
a“networked communication environment of contemporary ‘epidemic intelligence’” and
questions whether this results in an “inevitable condition of incomplete information”that
“creates a chronic desire for more information”(pp. 39-40).
This paper explores digital disease tracking tools, focusing on platforms that depend
on contributions from networks of users, or “crowdsourcing.”For health- and fitness-
related apps and websites, there are concerns around privacy and surveillance, with little
oversight or regulation surrounding the collection, storage, transmission and ownership
of personal information (Helm and Georgatos, 2014; Huckvale et al., 2015). Google’s
company DeepMind, which specializes in developing artificial intelligence, recently
announced that it was working with the National Health Service (NHS) in England to build
an app called Streams that will be able to help medical professionals monitor kidney
patients. DeepMind’s foray into medical databases has raised much concern. Rather than
focusing on a few thousand patients with kidney injuries, which many had assumed
would be the case, the company is accessing the NHS records of 1.6m patients who use
three hospitals run by the Royal Free NHS trust. The records include information from the
patients’medical history such as HIV status, drug overdoses and abortions, alongside logs
of day-today hospital activity including records of the location and statusof patients, who
visits them, and real-time data and historical records from critical care and emergency
departments. DeepMind claims that it needs access to the entire patient database to
produce accurate medical alerts and potentially diagnose diseases sooner. Yet many are
concerned that DeepMind’s database could allow for much more than the original stated
purpose. Further, with these data being controlled by one of the largest companies in the
world, there are fears that Google could quickly establish a monopoly over health
analytics (Shead, 2016; Kopstein, 2016).
Overview of disease tracking apps: past to present
Sickweather is far from the first digital disease surveillance tool: perhaps the most
well-known is the now-defunct Google Flu Trends, which tracked users’search terms and
other online activity to “nowcast”the flu –essentially, forecasting the flu the same way the
weather is monitored –based solely on people’s searches. When people are sick, many of
them search for flu-related information, which can supposedly be used as a proxy for overall
flu prevalence or spread. When combined with information from the Centers for Disease
Control and Prevention (CDC), accurate estimates could be provided up to two weeks earlier
than using CDC data alone (Lazer and Kennedy, 2015). Yet by 2013, Google Flu Trends, as
many media outlets reported, had “failed spectacularly,”over-predicting the prevalence of
the flu by more than 50 percent. The so-called poster child of big data approaches to disease
tracking had come up short. Google’s algorithm did not properly account for certain
seasonal trends in search query volume, and it did not account for changes in search
behavior over time. This was foreshadowed in October 2011, when Flu Trends data were
notably skewed due to pop singer Rihanna tweeting about having the flu, leading to a spike
in search queries from fans who were curious about the celebrity’s health (Lazer and
Kennedy, 2015).
Since the rise and fall of Flu Trends, other disease tracking apps and services have been
introduced, in many cases supplementing their algorithms with “small data”approaches
(more traditional epidemiological methods, surveys, health reporting and so on) to avoid the
failures of Google’s early attempt at disease tracking (Lazer et al., 2014). An example of a
“big data success story”is HealthMap –an online system created by researchers at Boston’s
Children’s Hospital –which collects and analyzes information from a variety of platforms,
including social media, online news and travel sites, to provide early detection and
surveillance of disease outbreaks. The creators say that the primary goal is to provide
1047
Privacy
implication
of disease
tracking apps
To continue reading
Request your trial