Editor’s note: In this final post for February’s ‘Openness Edition‘, Rachelle Annechino takes us on a journey with her to the homes of her research participants and asks some really important questions about the wild “foreign languages” (legalese/medical-ese) that supposedly produces “informed consent”, about the genesis of our understanding and practice of informed consent and challenges us to think about how we might redesign informed consent in our own projects.
________________________________________________________________________
One open window (Chris Downer) / CC BY-SA 2.0
Today I’m interviewing a couple of people who participate in a free program offered through a local hospital. The program mainly serves older adults who are dealing with a range of health issues, like diabetes, cancer, and arthritis. Many of the participants belong to groups that are affected by health disparities (or “preventable differences in the burden of disease, injury, violence, or opportunities to achieve optimal health that are experienced by socially disadvantaged populations” as defined by the US CDC [1]).
After hanging out at the hospital for a bit to check out the program, I go to the home of a woman in her 60s who couldn’t come to the hospital today. We talk about the study, its risks and benefits. It’s a small exploratory study, some semi-structured interviews; the hospital IRB gave it an expedited review.
The benefits, I explain, are that this might help improve the program or keep the program going. There aren’t really any direct benefits to you though. We wish we had something to give you to thank you for participating. Basically what we’ll do is just sit here and talk. A risk is that some of the questions could be uncomfortable, but we can skip anything you want. If it’s okay with you, I will record the interview. We won’t put your name on the recording or use your name in reports on the interviews.
We have this standard consent form that the hospital uses, I say. It’s kind of long. We can go over what’s in it together, and please feel free to take as much time as you want to look it over…
Et cetera. As I’m saying this stuff, I’m cautiously drawing out the consent form.
Which is eight pages long.
And crazy.
It’s written in some sort of academic/medical dialect of English. There are three separate places for the participant to sign.
The first page is about being an experimental research subject, and is full of language about “drugs,” “devices” and “medical treatments,” even though we’re not doing a clinical experiment, or anything involving drugs, devices or medical treatments. We’re interviewing people about a class. Nevertheless, everyone is supposed to sign this vaguely terrifying “human experimental research subjects” page separately.
Then there are the seven other pages. If I start to describe them, I’m pretty sure I’ll never stop ranting, so let’s just keep it at: there are SEVEN MORE PAGES. And don’t forget the two more places to sign!
Mmkay. Not gonna rant. But just a few things:
- This hospital wants more research that addresses health disparities, including research that looks at people’s experiences in context. Good for them; they want more than just lip service when it comes to figuring out health inequities.
- IRBs aim to prevent disparities in research practices, because like the Belmont report says, the burdens and benefits of research should be distributed justly [2], and this IRB takes that mission seriously. Good for them; they want to protect people that have historically been treated really (really, really) unjustly [3, 4, 5].
- Like health disparities, research disparities “are directly related to the historical and current unequal distribution of social, political, economic, and environmental resources. [1]”
So great, the IRB wants to support research that’s trying to make things better for people who bear the brunt of unequally distributed resources. Even better (?), it’s the IRB’s mission to protect people like the participants in this study.
And yet. Who is being protected by this consent form?
It’s not so well designed for, say, people who aren’t literate in medicalese, or for people who don’t trust institutions (because really!) [6]. Not so well designed for lots of issues that can occur more frequently among marginalized groups.
Informed consent for these interviews hasn’t really been about forms anyway. It has been about relationships between people, and trust.
Weirdly though, the form hasn’t actually been that bad to work with. Other consent forms I’ve used have been better designed from my point of view, but this one still informs people of the procedures that are in place and of the study’s risks and benefits. The thing is, informed consent for these interviews hasn’t really been about forms anyway. It has been about relationships between people, and trust.
One participant says that people come into this community all the time asking questions, and they never give anything back. She says she agreed to do this interview because the person who leads the program is giving something back. Her willingness to share information has little to do with me or my ridiculous consent form, but it is about a kind of openness or mutual exchange.
When I think about the ethics of information sharing, I’m not sure that the other forms and standard operating procedures I’ve encountered have been so different from this one. A lot of ethical practices in qualitative public health research — and perhaps in research with people more generally — are informed by bioethics, even when the language of clinical trials isn’t so obvious, and even when the research isn’t conducted in a traditional academic or institutional setting.
Bioethics tends to prioritize individual privacy and the individual’s right to be informed about what will happen to them (risks and benefits of experimental treatments). “Subjects” should only be identified in this model when there is an overwhelming interest in preventing the mass spread of disease. In contrast, journalistic and historical ethics privilege an audience’s right to be informed about what individuals — especially powerful individuals — are doing [7]. Anthropological ethics, meanwhile, cautions against revealing too much or too little about identity. Some secrets don’t want to be shared; some stories become theft when they are shared without credit.
Thick data is full of little details that make up the whole, and sometimes the right combination of a few details can make someone highly identifiable. In this way, qualitative researchers are intimately familiar with the problem of data re-identification that is sometimes ascribed to “big data.”
Doing qualitative or mixed methods research can highlight ways that standard human subjects “protections” aren’t always as straightforward as they may seem. When qualitative data needs to be de-identified, for example, researchers may sometimes go beyond removing pre-identified bits of information like names and addresses. Since qualitative data is less structured than quantitative data, the range of information that may be revealed is less specifiable ahead of time. Thick data is full of little details and bits of context that make up the whole, and sometimes the right combination of a few details — details other than, say, a name or address — can make someone highly identifiable. In this way, qualitative researchers are intimately familiar with the problem of data re-identification that is sometimes ascribed to “big data” [8].
Questions of identification are complicated by the ways people construct their own identities. In some contexts, people don’t want their stories to be de-linked from their names; they don’t want privacy… not exactly. In other contexts, being able to detach an identity from a story can free people to share information or parts of themselves that they would be afraid to reveal otherwise. Concealing information, in those situations, can paradoxically produce more openness.
New technologies add extra layers to longstanding tensions between openness and privacy. While many ideas about openness and privacy and all the spaces in between in social networks are familiar, the tools we have to negotiate exchanges between people are relatively new. Researchers who are exploring questions about consent and information sharing in relation to technology may also need to re-examine their ethical practices.
Suppose you designed an app to give research participants control over how their data was shared — a sort of Facebook for ethical research. What would the privacy settings look like?
Suppose you designed an app to give research participants control over how their data was shared — a sort of Facebook for ethical research, for example. Would that be practical in the contexts you’re working in? What would the privacy settings look like?
How would you redesign information sharing and informed consent in research? Can you Instagram your field notes?
_____
[1] CDC (2013) Health Disparities – Adolescent and School Health.
[2] National Institutes of Health (1979). The Belmont report: ethical principles and guidelines for the protection of human subjects of research. Bethesda, Maryland.
[3] Blue, E. (2009). The Strange Career of Leo Stanley: Remaking Manhood and Medicine at San Quentin State Penitentiary, 1913–1951. Pacific Historical Review, 78(2), 210–241. doi:10.1525/phr.2009.78.2.210
[4] CDC (2013) Tuskegee Study – Timeline.
[5] Veatch, R. M. (1997). Medical Ethics. Jones and Bartlett Publishers, Inc.
[6] Transcript – February 15, 1995. Eleventh Meeting of the Advisory Committee on Human Radiation Experiments (1995), Washington, DC.
[7] Fairchild, A. L., & Johns, D. M. (2012). Beyond Bioethics: Reckoning With the Public Health Paradigm. American Journal of Public Health, 102(8), 1447–1450. doi:10.2105/AJPH.2012.300661.
[8] Re-identification (2013) Electronic Privacy Information Center.
[9] Wang, T. (2012, August 2). Writing Live Fieldnotes: Towards a More Open Ethnography. Ethnography Matters.
________________________________________________________________________
Check out other posts from the Openness Edition: Jenna Burrel’s ‘#GoOpenAccess for the Ethnography Matters Community‘, Sarah Kendzior’s ‘On Legitimacy, Place and the Anthropology of the Internet‘, Juliano Spyer’s ‘YouTube “video tags” as an open survey tool‘ and An Xiao Mina et al’s ‘Designing for Stories: Working with Homeless Youth in Boyle Heights‘.
Ethnography Matters 
Great post, Rachelle.
I can certainly identify with some of your frustrations regarding consent forms. To be fair, here in the UK, the consent form you use is up to you, as long as it conforms to good practice as laid down by Ethics Committees run by our National Health Service (NHS).
The form I use is just 2 pages, and only requires one signature, so I’m struggling to see why 8 pages are needed! My own experience of these relates to getting a research project through Ethics approval (which is required both nationally and locally in the UK). I find that, provided you submit your research proposal in a very standard and normal way, Committees will be quite accommodating. But if you are using social-science based qual methods, they are less accommodating because it’s more unfamiliar to them.
As regards anonymisation, I do this for all proper nouns and names, as well as the participant themselves. e.g. the name of a pub, or a park, and of course a hospital would be redacted.Some health conditions are so rare and have such individual treatment regimes that re-identification is quite possible, even likely.
But the central thrust of your post – that complicated consent forms can work against patients, in that they create bureaucratic hurdles in the way of uncovering potential insight – is certainly something I feel too.
Thanks, Rick. Yeah, this is an unusual situation in my experience too. Different IRBs always have their own concerns and sometimes preferred language for certain items, but this was unique. As you say, if the IRB isn’t used to qual methods (and this one is used to clinical trials I expect), it’s harder. I’m hoping that there will be another iteration to this project and that we will be able to have some dialogue about consent. At this stage, it didn’t seem like that would be so productive.
But I do wonder sometimes if even the basic assumptions of anonymization are… well, I don’t know, often they make sense when you are doing one blanket thing with the data (as Tricia mentions below) and then releasing it, but maybe now there are opportunities to make anonymization controls more fine-grained, more individualized… Because people do sometimes want credit for their stories and insights, and certainly deserve that credit when it’s wanted. From a bioethics perspective focused on clinical trials, people aren’t generally interested in credit in the same way.
Also, yeah, anonymization is tough with multidimensional data, which can include qual data for sure. You can never guarantee complete anonymity, but I’ve done things like remove someone’s description of a hobby, because the combination of ethnicity, gender, the fact that the study was done in low SES neighborhoods in CA, and their take on their hobby could be pretty revealing.
“One participant says that people come into this community all the time asking questions, and they never give anything back.”
This is such an importnat point rachelle – It must feel so de-humanizing to participate in these studies and then feel as if you gave a lot of time without getting anything back. Like no updates, no info. I think what people want is just acknowledgment that their answers/time/input was helpful.
Do you think this is a common frustration among participants in your field of research?
Seems like a common frustration among marginalized communities in the US. Like In Native American communities there’s the long history of researchers coming in to “study the natives” and then leaving, either without providing anything in return, or leaving a trail of damage in their wake. Now a lot of Native American nations have their own IRBs, so things are different (and yet sometimes the same…)
The woman I was speaking to for this study mentioned outside researchers coming to poor communities ostensibly to “help”, but really, from her perspective, to get their research funded and write papers that don’t help anyone in the community. Which from my pov is often basically true, and makes me really interested in more open and “usable” research dissemination that lets participants engage more with findings (but then how much of a struggle will it be to get that kind of dissemination by some IRBs?)
On the other hand, sometimes research that is published in elite, closed journals can have very positive effects for getting helpful programs funded in a community, say, or for making policies that affect marginalized people suck less, etc. And it can also help build careers for people who genuinely care about that kind of work and are good at it, need some kind of support to do it, and need places to support conversations that aren’t popular/framed for popular audiences.
So another problem here is maybe a lack of visibility/openness about how that kind of stuff can be helpful… or maybe a lot of that changes as open access journals gain credibility?
“Suppose you designed an app to give research participants control over how their data was shared — a sort of Facebook for ethical research. What would the privacy settings look like?”
I like this question – and it’s got me thinking. But one of the issues with consents is that they are blanket consents. Like a question could be: can info from our interview be shared on twitter? yes or no or don’t know
The participant had to answer this question BEFORe the interview so she/he just answers no (cuz I woudl!). But if trust is developed with the researcher over time, maybe there is *some* info that could be shared – maybe some of it can be linked even to their twitter account.
But one of the issues with consents is that they are blanket consents. Like a question could be: can info from our interview be shared on twitter? yes or no or don’t know
The participant had to answer this question BEFORe the interview so she/he just answers no (cuz I woudl!). But if trust is developed with the researcher over time, maybe there is *some* info that could be shared – maybe some of it can be linked even to their twitter account.
For sure! It’s so hard and weird to think about it all. Or I find it hard and weird anyway.
Reminds me that when Yo-Shang and I did our project on mental maps of San Francisco, we had a release form for using people’s drawings of their neighborhoods. We would tell people about it at the beginning, and let them know that our suggestion was to wait until the end to decide what they wanted to do, since then they would know what they were agreeing to release or not. This way we figured they wouldn’t be worrying about it while they were drawing stuff, and yeah, I would never say “yes” to something like that ahead of time either.
I suppose a really fancy tool would let you issue consent and withdraw it at any time. Of course once something is released on the innernetz, it’s kind of out there, but still it might be nice to have that control. Or confusing and overwhelming? I dunno.
Such a good point about blanket consents.
“In some contexts, people don’t want their stories to be de-linked from their names; they don’t want privacy… not exactly”. I have frequently encountered the same situations while doing media ethnography. People not only want to be identified to the researcher (and in her consent forms) but also openly in the media. They want their ‘suffering stories’ to be told (in public) so that they could alleviate them. If ‘informed consent’ is about t “relationships between people, and trust”, I should not only identify the person rather promote her through publicizing her stories. Where would I be put in the ‘ethnics’ dynamics in such situation?
Thanks for pointing that out. Yeah, it’s tough.
Looking at the Belmont Report, there’s the “respect for persons” principle:
“To respect autonomy is to give weight to autonomous persons’ considered opinions and choices while refraining from obstructing their actions unless they are clearly detrimental to others. To show lack of respect for an autonomous agent is to repudiate that person’s considered judgments, to deny an individual the freedom to act on those considered judgments, or to withhold information necessary to make a considered judgment, when there are no compelling reasons to do so.”
This makes me think about whether there are ways to give individual research participants more access to informed choices about how/whether they are identified…