Research Consent Postscript
Two more research papers were published very recently (May 2022), evaluating the perceived effectiveness of Crisis Text Line in advance of the USA federal 988 crisis line rollout. They are companion papers, with the same 8 authors, except one additional author on the first paper. The authors include Dr. Anthony Pisani and Dr. Madelyn Gould, both highly respected and long-time advisory board members to Crisis Text Line. Both were among the 13 authors of the 2019 JMIR paper.
The first paper was published May 26, 2022 and cites to the 2019 paper:
“…Following carefully developed procedures to ensure privacy, appropriate data use, and other protections for texters (Pisani et al., 2019), CTL provided de-identified CC reports, texter surveys, and metadata…gathered from conversations initiated by texters between October 12, 2017, and October 11, 2018. CTL anonymized the data by removing personally identifiable information using natural language processing (NLP).”
The second paper was published May 22, 2022 and doesn’t cite directly to the 2019 JMIR paper.
Both papers appear to invoke the form of exemption used by federal agencies (federal exemption) where consent is not required for secondary research on data when the “…identity of the human subjects cannot readily be ascertained…” The federal exemption does not apply to Crisis Text Line, but Institutional Review Boards may grant exemption from consent along similar lines.
“The study’s protocol involving secondary analysis of de-identified data without access to links was considered to meet Federal and University criteria for exemption by the University of Rochester’s Institutional Review Board (IRB) and not to meet the definition of Human Subjects research requiring review by the New York State Psychiatric Institute/Columbia University Department of Psychiatry’s IRB.”
Crisis Text Line, and some researchers, are having it both ways with the millions of crisis conversations in their data store. They claim to have consent to do research, and they claim that consent is not needed to do the research.
[I feel that the federal regulation exempting research from consent, when data is anonymized, is not appropriate to invoke for crisis conversations. Also, when federal regulations don’t apply but Institutional Review Boards waive consent, I feel that is inappropriate in the crisis context. I’ll be writing separately about that.]
It’s easy to understand why Crisis Text Line has connections with Big Data. But these other supporters, from academia in particular, present a more complex, nuanced set of questions about deception, consent for research, and Crisis Text Line. Understanding and unraveling that story is important work that remains to be done.
The Deception about Conversation Privacy and Confidentiality
Persons using the service expect privacy. Comments have been made on social media by people who used the service and volunteers, expressing feelings of betrayal and violation. Trust in our systems of care, such as they are, has been damaged. People wonder: Where can I go? Where should I refer others?
While it’s true that some might support data storage and use for research, there are intractable problems. How is it possible to obtain informed consent for research that hasn’t yet been defined? How is research possible on a limited sample size (zero conversations so far with informed consent), and would informed consent infuse the data with bias? There’s a way to resolve these questions: Respect the conversations by staying out of them, except for quality control by humans.
The “Anonymized” Data Deception
The crisis moment is specifically unique to one. The crisis conversation, unique to two. It seems that every imaginable human suffering, in every possible combination, has happened, and is likely to be happening to someone right now. If viewed as one, mass anonymized devastating condition, we see hopelessness. If viewed one person at a time, we are invited to engage in one person’s unique story, worth listening to. Then, it becomes possible to offer a warm blanket of comfort and care in the crisis moment. One, and then another, like this, can do a whole world of good.
It’s easy to fall into a trap of lumping people together within categories: self-harm, anxiety, depression, persistent thoughts of ending one’s life. Presuming too much from a label is a form of erasure, a form of anonymizing the individual. In crisis conversations, there are no generic circumstances, no generic conversations. It’s all unique detail, personal to each person, and to the volunteers who answer.
This viewpoint stands on its own. The very serious concerns about whether it’s even possible to anonymize conversations, remain. Thank goodness for those who advocate there.
Here, I’m talking about the intimacy, the emotional vulnerability, the surrender of self that is required to open a possibility. A possibility for one human to help another. When someone answers that call, and offers to help hold that fragile space, then there are two, and that makes all the difference. There’s no room for anything else.
No More Deception
This means:
1. No research on the massive accumulation of crisis conversation transcripts. Without consent, this is a deception, a theft.
3. No more perpetual storage of crisis conversation transcripts. As if people’s lives and crises are being kept frozen in a cryogenic data lab for a future language processing breakthrough. No. Delete all conversations after six months, to allow reasonable time for quality control.
Please, no more deception.
Disclosures, More Information
b. For the curious, here’s the
opinion paper I shared, resulting in my termination.
c. In November 2021, I wrote a
letter of concern to the Journal of Medical Internet Research. At the time of this writing, it’s been shared with the paper’s authors. The concerns are under review at the time of this writing. I’ve been generously invited by JMIR to write a commentary on the 2019 paper, which I plan to submit by mid-August 2022.
August 20, 2022 update: The article has been updated to reflect that an additional author was on Crisis Text Line staff. Nitya Kanuri was the Open Data Collaborations Manager for Crisis Text Line. This affiliation was not declared in the 2019 JMIR paper.
August 31, 2022 update: To clarify that Crisis Text Line, Inc., as a private non-profit corporation, is not bound by terms of the federal exemption for consent.
