25 May 2020

Contact-tracing apps face serious adoption obstacles

Sarah Kreps, Baobao Zhang, and Nina McMurry
Source Link

If public-health authorities’ worst predictions come true, COVID-19 may never disappear. That means the world will have to live with the virus and develop effective treatments and measures to contain the virus.

Mobile contact-tracing technology has emerged as one such measure to track population movements and alert individuals when they come into contact with an infected person. But such technology faces enormous obstacles. In order for such tools to be effective, some 60 percent of the population needs to opt-in and use them. With the novel coronavirus continuing to spread in the United States and major American universities and technology companies actively developing digital contact-tracing tools, understanding whether the American people would be willing to use such technology to stem the outbreak has never been more important.

But Americans continue to be deeply skeptical of such technology. In a nationally representative study of 2,000 Americans between April 30 and May 1, 2020, we found that just over 30 percent of Americans indicated they would download and use a mobile contact-tracing app, raising questions about whether such technology will be adopted widely enough to be effective. In a bit of good news for developers, support among Americans for digital contact tracing tends to increase with stronger privacy protections.




In countries where digital contact-tracing efforts are already underway, widespread adoption continues to be a major problem. In Singapore, where authorities in March rolled out one of the first contact-tracing apps, up-take has only reached 20 percent. Iceland announced recently week that 38 percent of its population had downloaded its version, the highest level of voluntary adoption of an automated contact tracing app to date.

Skepticism toward digital contact tracing was not equal across racial groups, a disparity that if such apps are rolled out in the United States could result in further inequities in health outcomes among marginalized communities. African Americans, a group already disproportionately affected by the virus, expressed greater skepticism about digital contact tracing, with less than 20 percent supporting government roll-out of these apps compared to 36 percent of white Americans. When told about a hypothetical contact-tracing app, only 22 percent of African-American smartphone users indicated that they would be likely to download the app, compared with 32 percent of white Americans. African Americans also reported greater concern about usage of their personal data. In light of these concerns, widespread testing, traditional contact tracing, and targeted outreach to marginalized communities will continue to be important.

But there is reason to believe that support for digital contact tracing will improve, especially as the disease continues to spread. Those surveyed were more likely to opt in if they knew someone who had tested positive for COVID-19. As the number of cases continue to increase in the United States, so too will the number of people supportive of these tools.

Our survey indicates significant levels of distrust toward the technology companies and public health authorities administering contact-tracing technology, but when we described the way that some companies are trying to build contact-tracing tools with privacy and security in mind, we find significantly higher levels of support for apps. While some European countries and Singapore are using a centralized model for storing COVID-related health data generated by apps, others, such as the Apple-Google API currently under development, envision a decentralized system of storage. The latter system also relies on Bluetooth signal exchanges that do not reveal the identity or location of the users, helping safeguard against the potential misuse of data. Among our respondents, only 27 percent expressed willingness to download a hypothetical app with centralized data storage, while 34 percent were willing when the app was described as using decentralized storage. Apps using Bluetooth technology were supported by 32 percent of respondents, compared to only 27 percent for hypothetical apps using GPS.

The technology nonetheless elicits visceral reactions about whether it can adequately preserve privacy. Over half of respondents in our survey predicted that contact-tracing apps would make tech companies too powerful and would violate their privacy. In open-ended answers, respondents frequently cited concerns about the threats to democratic governance. Representative statements include an individual who referenced the hazard to “our democracy from the tech companies tracing who we contact, what we think and who we know.” Another respondent shared that “I would rather limit my activity than keep records to the extent that might be accessed by other people.” In general, respondents expressed wariness that either Big Tech or the government would be tracking their every move. One individual called digital contact tracing “another step toward socialism.”

This skepticism isn’t surprising—just consider the position of the Cambridge Analytica scandal or the Snowden revelations in the public imagination—but the public-health community need not pay the price for governments’ and tech firms’ prior sins. App developers should recognize the importance of the privacy-preserving measures and follow through on decentralized data storage while dropping considerations of GPS location tracking. Our analysis points to the need for public education campaigns that clarify what the tools are and, especially, what they are not doing.

Furthermore, Congress might consider passing legislation, as Australia has done, that would impose fines for data misuse. If legislation is not feasible, state public health authorities and tech companies should establish a multi-stakeholder board to oversee the technology. That board should include not only public health experts but also advocates for groups that are disproportionately harmed by COVID-19 or could be disproportionately harmed by the introduction of digital contact tracing. 

Contact-tracing apps are not meant to replace other public health measures, nor should they. Indeed, beyond the concern about effectiveness and security of these apps, public health officials should realize that digital contact tracing introduces new challenges of access and equity. The technology on which it relies—cell phones—is not spread evenly across societies but fragmented. The tool will not help populations who have lower levels of smartphone ownership such as the elderly or homeless, who are also particularly vulnerable to the virus.

Research is required to determine whether the apps are effective and whether they generate significant amounts of false positives or negatives, so that the technology can improve. Its roll-out should of course be as accurate as possible, but our research shows that public confidence and in turn efficacy will grow if, or as, it proves its worth.

Sarah Kreps is a professor of government and Milstein Faculty Fellow in Technology and Humanity at Cornell University.

Baobao Zhang is a postdoctoral fellow in the Department of Political Science at the Massachusetts Institute of Technology.

Nina McMurry is a PhD candidate in the Department of Political Science at the Massachusetts Institute of Technology.

No comments: