Limitations to the 'revolutionary' findings of online studies
'Direct to consumer' research, using data obtained through increasingly popular online communities such as 23andMe, PatientsLikeMe and the Personal Genome Project, has methodological limitations that are known to epidemiological studies, including selection bias, information bias, and confounding. These limitations mean that the results and conclusions of research using these methods need to be interpreted with caution, according to a paper published in the journal PLoS Medicine.
Cecile Janssens, PhD, formerly of the Erasmus University Medical Center in The Netherlands and currently research professor of epidemiology in the Rollins School of Public Health, Emory University, and Peter Kraft, PhD, from the Harvard School of Public Health, argue that the findings of 'direct to consumer' research should be communicated in a way that is understandable, accurate, complete and not misleading. Researchers do often address methodological limitations in their scientific publications, but not on their websites, say the authors.
Studies relying on collections of self-reported data by self-selected participants raise critical questions that require further ethical analysis and public debate – for example, regarding the provision of adequate consent, the safeguarding of public trust, disclosure of commercial development of research results, and the sale of participants' data to third parties.
"We worry that overstating the conclusions that can be drawn from these resources may impinge on individual autonomy and informed consent," the authors say. "Only a responsible approach with realistic expectations about what can be done with and concluded from the data will benefit science in the long run."
The authors argue that clarity regarding the benefits of research using solicited personal data is particularly important when the data collected are also used for other purposes, such as selling participants' information to pharmaceutical and insurance companies.
"The potential for sharing participants' data with third parties as well as the commercial uses of research findings should be disclosed more explicitly to participants prior to consent," they conclude.