VALIDITY AND RELIABILITY
The data practices in current self-experimentation projects rarely meet the standards of rigorous scientific research. Amateur self-experimentation studies are not randomized or blinded as in traditional clinical studies and often lack coherent data collection protocols. The data analysis and evaluation of findings is usually not subjected to a proper peer-review and rely on unstructured advice provided in online community forums or live meetups. That brings in the issue of limited scientific validity, which creates potential health risks for practitioners and weakens the acceptability of n=1 findings by health professionals. However, we would also like to question the very assumption that the expert way of knowing is always superior to that of amateurs. Would we lose some surprising and potentially valuable findings from peer-validated self-experiments by pushing for more disciplined data sharing practices? Should we only aim for the professionalization of amateur self-experimentation, or rather promote an agnostic approach to design that does not assume the superiority of one form of knowledge over another?
How can design solutions (interfaces, activities, tools) increase the adherence to more structured research protocols and baseline measures for data collection and evaluation? Could these protocols be crowdsourced, co-designed, and agreed upon by the community members, while ensuring scientific validity? How can we design tests comparing amateur and professional data collection?
Security and privacy
The data sharing practices in online health communities often benefit to corporate stakeholders rather than to the end users, which brings in certain privacy issues. Even the self-governed communities such as Quantified Self or Soylent that started as independent hobbyist endeavors often end up adopting a business ethos and monetizing the data shared over their services and products. While some suggest that all interested stakeholders could still benefit from such data sharing potential exploitation of users privacy is an important concern. Who can access and (re)use what type of data and to what end? Should we accept the idea of data sharing altruism and 'donating data for good as an intrinsically virtuous practice?
How can design help to balance altruistic data sharing intentions with market-led goals? Can design support transparent systems for data exploration that leaves interpretive control with the end users? How can we manage archival of the data and mitigate potential misuses by nefarious third parties? Should the design support further opening of self-experimentation data for professional health and pharmaceutical research at all?
Self-experimenters usually pay for their self-tracking devices, direct-to-consumer (DTC ) sequencing tests, and participation in crowdsourced health studies, which makes these services affordable only to certain socio-economic cohorts. For instance, the Quantified Self and Soylent groups are populated mostly by middle-aged white males. Not only does this skewness exacerbate the already problematic healthcare disparities, but it also limits the idea of crowdsourced health studies and trials as a source of demographically robust data. Thus, the limited access and demographic skewness are concerns both within the communities and on a broader social scale.
How can design support social robustness of self-experimental healthcare? Can design bring the self-experimentation practices closer to participatory models of design and support a "genuine participation"?