‘Anonymous’ research: what you need to know
Volunteers who take part in medical and social research are the unsung heroes of the modern age. Clinical trials pave the way for ground-breaking treatments such as Truvada, which might prevent HIV transmission. Similarly, social research such as the ‘Growing up in Australia’ longitudinal study guides public funding and policy to improve the lives of future generations.
It is accepted that ‘good’, ethical research safeguards participants’ privacy by ensuring they can remain anonymous and that their information is kept confidential, especially if people are asked to disclose sensitive or secret information. It is common for researchers to ‘de-identify’ data by removing ‘personally identifying information’ such as name, date of birth, or address, and ensuring these can’t be seen by people outside the research team. However, emerging technologies could allow hackers to exploit de-identified data to identify the individual participant in question. This has weighty implications for participants, scientists, and the organisations which undertake human research.
Universities and research institutions such as the University of Melbourne enforce strict processes to protect human research participants and their data. Ethics committees require that volunteers only take part if they have given ‘informed consent’. As part of this, participants should understand any risks related to their privacy.
Genome hacking: easier than you think
Increasing use of DNA data in medical research presents a raft of new challenges. Each individual’s DNA sequence is unique – and so is potentially identifying information. Biomedical researchers have habitually shared genetic data in global databases to maximize the usefulness of their work, acting in the belief that there is no way to link these DNA sequences to the individual they came from.
However Yaniv Erlich, at the Whitehead Institute for Biomedical Research, has shown that it is possible to match anonymous genetic data to an individual. DNA profiling services such as YSearch allow people to use their own genome data to trace unknown relatives. Erlich demonstrates that a hacker could subvert one of these online services to match DNA to a family line, and potentially to an individual. Asked what’s involved in identifying an anonymous genome donor, Erlich told Science Magazine “it’s easier than you think. There are more and more enabling tools, and more and more data out there”.
What your data says about you
Mundane pieces of information can also shatter anonymity. Combining data such as gender, postcode and date of birth can create a match to a unique individual, as Harvard researcher Latanya Sweeney has powerfully demonstrated.
Even the words we write can be used to identify us. A team of researchers at Drexel and George Mason universities have shown that software to analyse online posts and compare them with works by known authors can unmask anonymous writers with chilling accuracy.
Planning for the unknown
It is clear that researchers cannot ignore the possibility of future threats to privacy, nor guarantee complete anonymity for research participants. They must address emerging threats and weigh them seriously with the benefits of sharing research data to increase the pace of scientific advances and reduce inefficiency, waste and duplication. For research institutions the stakes are high. Online media means that a single slip could rapidly undermine a well-established reputation and damage the public goodwill that is essential for people to take part in research.
It is important that potential volunteers should be able to engage with these issues as thinking, critical partners in the research. The rise of social media means that people have increasingly sophisticated understandings of personal privacy issues, and the trade-offs involved in sharing information.
Research participants should always feel that they and their information are treated with respect and integrity. In fact, they should get to feel like superheroes.
What questions do you think someone should ask before they agree to take part in research?
Do you worry about technological advances endangering your privacy, or your family’s privacy?