Privacy Implications of Health Information Seeking on the Web

Abstract

This article investigates privacy risks to those visiting health- related web pages. The population of pages analyzed is derived from the 50 top search results for 1,986 common diseases. This yielded a total population of 80,124 unique pages which were analyzed for the presence of third-party HTTP requests. 91% of pages were found to make requests to third parties. Investigation of URIs revealed that 70% of HTTP Referer strings contained information exposing specific conditions, treatments, and diseases. This presents a risk to users in the form of personal identification and blind discrimination. An examination of extant government and corporate policies reveals that users are insufficiently protected from such risks.