health engagement, privacy and the quantified self: a free-ranging conversation with alexandra carmichael from curetogether
september 14, 2010 · by Fran Melmed
alex carmichael is a data devotee. she’s the director of the quantified self, co-founder of curetogether, subject in thomas goetz’s book the decision tree, and a research affiliate at the institute for the future.
i conducted an email interview with alex to talk specifically about curetogether and what companies could learn about health engagement and privacy from this health collaborative.
fm: can you give a quick overview of curetogether for those unfamiliar with who you are and what you offer the patient community?
ac: curetogether really started as an experiment, and has just taken off. curetogether is about quantifying the collective patient experience. while other patient support sites have focused on stories or information from experts, we focus on quantitative data across 576 patient-contributed conditions. individual data is kept private, but the anonymized aggregate data is shared openly to maximize discoveries.
at almost 1 million data points, curetogether has the largest available real-world comparative effectiveness database available to patients online. we help patients learn what treatments are available for their condition (based on what others are trying), and how popular and effective each treatment is reported to be. patients can also track any measures of their health over time, recording weight, stress levels, pain, and mood, and share this information with their doctors.
curetogether also helps people find and connect with others with whom they share multiple health conditions, which would otherwise be extremely difficult since most health sites are organized around individual conditions. imagine finding someone who shares 7 conditions with you and being able to ask him or her private questions.
finally, we have unique data on co-morbidities. we were recently able to replicate an association between infertility and asthma (and other conditions) using only patient-contributed data. as more patients add their voices, we’ll be able to provide more personalized information about what might work for whom. patients are helping each other, helping research go forward into their conditions and helping themselves.
fm: would you consider curetogether a social network?
ac: we’ve just added the ability to contact other members with conditions matching yours, so it’s more of a social network now than it has been, but we still keep the anonymity our members have requested, largely because we deal with sensitive chronic conditions that people don’t necessarily want to broadcast publicly.
fm: do you find the people who come to curetogether are already engaged in their health? or does being a part of your community engage them?
ac: the first one, definitely. we mostly hear from people who have tried everything, been to multiple doctors and are looking for new answers to their problems. they come to share all the things they’ve tried and hopefully get a few new ideas from others in the same boat.
fm: employers are looking for ways to engage employees in their health, whether currently living with a condition or not. what have you learned about health engagement that they could benefit from knowing?
ac: people are motivated. they just need the right tools and the right information at the right time. BJ foggof stanford’s persuasive technology lab calls this “put hot triggers in the path of motivated people.” (note: to learn more about fogg’s work, see fogg’s behavior grid or request a copy of the first publicly-available behavior guides.)
fm: i think you’re right about the right tools and the right information at the right time, and i also like fogg’s model. doing this is tricky when people are at different places or stages of change. can you elaborate on what you’ve learned about how to deliver the right tools and the right information to people who are at different places and different levels of knowledge and readiness to change?
ac: i think the key is to keep things as simple as possible. we’re inspired by the work of 37signals and edward tufte to eliminate any unnecessary words or graphics, and just show information in as clear a way as possible. it’s a matter of both design and adaptive learning—any health information system should refine what you see and how it guides you based on what it knows about you, like netflix or amazon do.
fm: i want to return to health privacy, which is always important—it’s a very individual decision, to share or to not. it’s a little more amped up with employee wellness efforts because of the employment relationship and employer-provided benefits. have you been surprised by how openly people are sharing such personal information?
ac: the only thing that’s publicly available is aggregate information, not individual data. having that layer of aggregation helps people feel freer to share personal information without worrying about future consequences.
fm: like you, employers look at aggregate data to figure out what it means for future health concerns or future benefits design. and also like you, they need to convey how they use this data while protecting individual privacy. how do you inform and assure users about their privacy?
fm: do you see a place where employers can be more supportive of what you’re doing or of their employees who are living with chronic conditions?
ac: yes, as more chronic conditions become destigmatized, and as people find health benefits from connecting online, employers and health plans will benefit from helping their employees find resources like curetogether. with the anonymity, employees can feel safe that their employers are not trying to spy on them but are genuinely trying to help.
for more on alexandra and curetogether:
- follow alex on twitter
- an interview by matthew cornell, a productivity consultant
- curetogether’s blog, the collective well
- curetogether’s story on changemakers