Fresh Territory for Bioethics: Silicon Valley
Research
Susan Gilbert, 10/28/2015

Fresh Territory for Bioethics: Silicon Valley

(Research) Permanent link

Biomedical researchers are increasingly looking to Silicon Valley for access to human subjects, and Silicon Valley is looking to biomedical researchers for new ventures. These relationships could be a boon to medicine, but they also raise questions about how well-informed the consent process is and how securely the privacy of the subjects’ identity and data is kept. Other than a few quotes in the popular press, bioethicists have had little to say on the topic, although those whom I have spoken with agree that more attention is warranted. And so, in the current issue of the Hastings Center Report, I invited readers to write commentaries for Bioethics Forum.

Since the journal went to press, more has happened, adding to the case for examining how social media and technology companies are changing the way human subjects research, and especially genetic research, are conducted.  The latest news has been about 23andMe, the consumer genetic testing company supported by Google, and about Google itself.

This month, 23andme raised $115 million to be used to expand its consumer genetic testing business and its laboratory research capacity. The company is already involved in research. It sells drug companies and academic institutions access to its research portal, which has a database of more than 650,000 people, customers of its genetic testing service who have consented to make their data available for research. This data consists of deidentified genetic profiles and “more than 225,000 phenotypic data points, including demographic, clinical, family history information and more.”

23andMe says that its research portal is “designed to disrupt the way that genomic studies are conducted by empowering scientists to make research discoveries in a faster and more targeted way than ever before.”

The company’s capacity for disruption is likely to grow because of another development. Last week, 23andMe announced that it would resume providing some health-related genetic information to consumers. The Food and Drug Administration gave the company approval to do so nearly two years after ordering it to stop offering health-related genetic tests because of concerns about the accuracy of the results. With the green light from the FDA, 23andMe is bound to attract new customers – whose rate of increase had declined by more than half in the last two years, when the company could offer only ancestry testing – and many of the new customers will no doubt agree to have their data added to the research database.

Meanwhile, more disruption of the way research is conducted could soon take place at Google. It recently started a life sciences research program and is hiring “biomedical superstars,” including Thomas Insel, director of the National Institute of Mental Health. “Nurtured by Google’s expertise in data analytics and engineering, the biology team is expected to create miniaturized electronic devices and to use these and other means to collect and analyse more health data, more continuously, than is possible today,” according to an article in Nature last week.

Other social media and tech companies have entered the health research arena. This year, Apple launched ResearchKit, an open source software platform that enables researchers to gather medical information from study participants’ iPhones and other Apple devices. Facebook developed an app called Genes for Good for a University of Michigan study on the genetic causes of disease. Volunteers, who must have Facebook accounts, use the app to sign up for the study, answer health-related questions, see summaries of their health information, and consent to donate saliva samples for genetic analysis. Volunteers can also use the app to tell their Facebook friends that they have joined the study and invite them to do so too.

It’s easy to see the potential benefits of these research arrangements. Scientists can tap into the large social networks of Facebook and Apple and the huge databases of 23andMe and Google to quickly and inexpensively recruit thousands of research subjects and gather the vast amounts of data. This capacity is essential for making headway on the national Precision Medicine Initiative, which aims to tailor the prevention and treatment of diseases to an individual’s genes, environment, and lifestyle.  But what are the ethical gray areas or gaps?

23andMe says that all of its research activities are governed by an institutional review board. But how well can the privacy of research subjects be protected when their data are being sold to other institutions? A new study found privacy risks from a type of web server used for genomic data-sharing. 

Some experts have also questioned the informed consent process of studies conducted with the Silicon Valley companies. What piqued the concern of Paul Appelbaum, a Hastings Center Fellow and director of the Division of Law, Ethics, and Psychiatry at Columbia University, was a statement by Anne Wojcicki, founder of 23andMe, saying that 85 percent to 90 percent of its customers for genetic testing agree to have their data used for research. “That consent rate is quite high,” Appelbaum told the San Jose Mercury News. “Most studies conducted in academic settings have much lower consent rates. That raises the question of how well 23andMe describes what it is that they are asking permission for.”

I wonder if the feel-good marketing of some of the studies could interfere with informed consent.  On their websites, 23andMe and Genes for Good entice volunteers with the prospect that their participation in research could help “make a difference” and advance “the common good.” Might the desire to help people blindside potential participants from paying attention to the terms of consent to participate?

We don’t know, but we should find out. Social media and communications technology are increasingly integral to daily life, and so it’s no surprise that they’re also becoming embedded in biomedical research. Is special oversight warranted? If so, what should it be? I reiterate my invitation to bioethicists to write for Bioethics Forum on this new and evolving development in research ethics.

Susan Gilbert is the public affairs and communications manager of The Hastings Center and editor of Bioethics Forum.

Posted by Susan Gilbert at 10/28/2015 04:00:30 PM | 


Readability of 23andme.com documents
Mark Hochhauser, Ph.D.
Readability Consultant/IRB Member


As a follow-up to Susan Gilbert’s comments about 23andme.com, I analyzed the readability of the site’s Consent Document and Full Privacy Notice. For improved reading and comprehension, Federal agencies typically recommend that consent forms be written at a 6th – 8th grade reading level.

Consent Document
Based on ReadabilityStudio 2012 software, using five readability formulas (Flesch-Kincaid, Fry, Gunning Fog, and Flesch Reading Ease, the 23andMe Consent Document (2,102 words) is written at about a grade 15.9 reading level—about 4th year college reading level. The Flesch Reading Ease formula scored it at 28 (Very Difficult), where 0 = very difficult, 100 = very easy.

Full Privacy Notice
Using the same readability analysis, the 23andMe Full Privacy Statement (5,715 words) is written at about a grade 17 reading level—about 1st -2nd year graduate school reading level. The Flesch Reading Ease formula scored it at 22 (Very Difficult).

Conclusion
At an average adult reading speed of about 250 words per minute, readers need about 30 minutes to read the 7,817 words in these two documents. Based on 2014 U.S. Census (Educational Attainment) data, about 19% of US adults have a Bachelor’s Degree, about 10% have a Master’s Degree or higher. Most US adults will have a very hard time reading and understanding these two 23andme.com documents.
Posted by: Mark Hochhauser ( Email ) at 10/29/2015 7:04 AM


Leave a comment
Name *
Email *
Homepage
Comment