A revealing picture of how personal health
information searches become the property
of private corporations.
BY TIMOTHY LIBERT
PRIVACY ONLINE IS an increasingly popular field of
study, yet it remains poorly defined. “Privacy” itself is
a word that changes according to location, context,
and culture. Additionally, the Web is a vast landscape
of specialized sites and activities that may only apply
to a minority of users—making defining widely shared
privacy concerns difficult. Likewise, as technologies
and services proliferate, the line between on- and
offline is increasingly blurred. Researchers attempting
to make sense of this rapidly changing environment
are frequently stymied by such factors.
on the Web
Therefore, the ideal object of study is
one that is inherently sensitive in nature, applies to the majority of users,
and readily lends itself to analysis.
The study of health privacy on the Web
meets all of these criteria.
Health information has been regarded as sensitive since the time of the ancient Greeks. In the 5th century B.C., physicians taking the Hippocratic Oath were
required to swear that: Whatever I see
or hear in the lives of my patients...I
will keep secret, as considering all
such things to be private. 21 This oath is
still in use today, and the importance
of health privacy remains universally
recognized. However, as health-infor-mation seeking has moved online, the
privacy of a doctor’s office has been
traded in for the silent intrusion of behavioral tracking. This tracking provides a valuable vantage point from
which to observe how established cultural norms and technological innovations are at odds.
Online health privacy is an issue
that affects the majority of Internet
users. According to the Pew Research
Center, 72% of adult Internet users in
the U.S. go online to learn about medical conditions. 9 Yet only 13% of these
begin their search at health-specific
sites. In fact, health information may
be found on a wide spectrum of sites
ranging from newspapers, discussion
forums, to research institutions. In
order to discover the full range of sites
users may visit when seeking health
information, I used a search engine to
˽ Over 90% of the 80,142 health-related
Web pages initiate HTTP requests
to third-parties, oftentimes outside
the view of the user.
˽ Some 70% of third-party requests transmit
information on specific symptoms,
treatments, and diseases in the URI string.
˽ Page visitors are at risk of their health
interests being publicly identified as well
as being blindly discriminated against by
˽ Extant policy and legal protections
are few in number and weak in effect,
demonstrating a need for interventions.