I was sitting up in the middle of the night last night, pondering (ok, stressing about) the enormity of work that lies ahead of me for the upcoming week when an email arrived in my work email box and caught my interest.
Now, normally I would never have opened an email like this. Spammy-looking emails generally go right to the trash. Maybe it was the fact that it was 2 am. Maybe it was the vodka martini I’d had earlier in the evening. Maybe it was the fact that it was sitting alone in a list of read messages. Something inspired me to open it and, thus, we have inspiration for today’s post.
The sender of the email was the “Citizen Science Newsletter” and the title read “Only Hours Left to Sequence your Microbiome with Citizen Science”. I have pdf-ed the uBiome email if you’d like to read it in its entirety yourself.
The group soliciting me to sequence my microbiome is called “uBiome” and they look pretty excited about people doing science.
I’m pretty excited about regular folks doing science too, so I explored their website a bit. This group has been featured by a lot of media outlets and it seems that folks are pretty excited about their work. Even the NIH is excited about them.
The link in their email takes you to a website to donate to their cause and they have more than surpassed their funding goal.
But, why I am concerned about this project?
I should preface the rest of this post by telling you that I have been doing research with human subjects and patients for a long time. About 12 years, to be more precise. I take the privilege of being able to do human research very seriously. It is important to me to do research that protects the safety and dignity of my participants. More importantly, it is important to me that participants in my studies be informed about what their participation in my studies entails. What are the risks? What will be done to them? How will their information be used? What safeguards are in place to protect their rights?
And this brings me to what I believe to be the potentially dubious ethics of this citizen science project. One of the first questions I ask when I see any scientific project involving collecting data from humans is, “What institutional review board (IRB) is monitoring this project?” An IRB is a group that is specifically charged with protecting the rights of human research participants. The legal framework that dictates the necessary use of an IRB for any project receiving federal funding or affiliated with an investigational new drug application stems from the major abuses perpetrated by Nazi physicians during Word War II and scientists and physicians affiliated with the Tuskegee experiments. The work that I have conducted while affiliated with universities and with pharmaceutical companies has all been overseen by an IRB. I will certainly concede to all of you that the IRB process is not perfect, but I do believe that it is a necessary and largely beneficial process.
From their website, the uBiome project does not appear to be affiliated with an IRB and I find no mention of an informed consent process.
Why is informed consent process important? Informed consent assures that human research participants are fully aware of what their participation entails. It also makes participants aware of any relationships the investigators have that might influence their conduct. With both of these considerations in mind, I can see several places in which the uBiome project might be ethically dubious in a way that would be have been picked up on and probably prevented by an IRB.
1) Participation in the uBiome project requires the payment of a fee.
The coupling of payment with participation is problematic. It biases your data by limiting participation to people who can afford to participate. uBiome appears to have donated 150 kits, but that’s a trivial portion of their sample size. When I conduct research, our informed consent notifies participants that they will not be charged for participation in research. In no way will their revocation of consent influence their relationship with our institution and they are free to revoke their consent at any time. Participants in the uBiome project do not appear to have that right.
Now, let’s be clear that I am not necessarily saying that uBiome shouldn’t crowdsource their funding. I am saying that data collection and crowdsourcing should be separated, even if that limits the efficacy of the crowdsourcing.
2) There is no clear statement about what will be done with the data or samples.
Let’s imagine I collect blood. I am required to disclose to our participants exactly what I am analyzing the sample for. I have to disclose whether I will share the sample with anyone and whether an individual’s identity will be tied to the sample. I also have to have a plan for destroying samples. None of this appears to be disclosed by the uBiome folks. It could be that the particularly unique chlamydia sequence they find in your vaginal swab (that you didn’t know you had) would be shared with hundreds of other researchers, ultimately ending up incorporated into some GMO tomato somewhere that your mother adds to her spaghetti sauce. Maybe you’re ok with being the person that changes the face of chlamydia research and tomato growing, but you should have to right to make that decision. It’s not enough to know that people can go to their database, but will they actively send your information to people?
3) There is no apparent plan for how your identity will be protected.
The site states “Your data is open to the world… if you choose. Your data is yours — you can download it, share it, do whatever you want with it. We encourage you to opt-in to share your data with our scientists, but we respect your privacy and will not force you to do so. The data is also anonymized and private.” It’s not clear to me how this process works. Do you have the choice to “opt out” before or after you find out you have science-changing chlamydia? This is the sort of question an IRB would ask.
They also state “uBiome will never release any of your personal identifying data. We will never even store your data in such a way that anyone could figure out who you are from it.”
I suspect, if they are directly linking your ability to access your results either through email or a website that this is not true. If it is true, I would certainly like to hear more about it…
With my projects, we have a very rigorous data protection plan. We disclose all of the potentially identifying information that we collect. Birth dates. Medical history. Everything. We disclose both to the participant and to the IRB the locations where their name is linked to their data. We disclose how long we will keep those materials and how we will destroy them. We disclose the encryption on our computers, phones, and other devices that prevents people from accessing identifying information about our participants. The uBiome project’s data protection plan appears to amount to “Trust us! We love science!”
4) It’s not clear what conflicts the major players in the project might have.
The site states “our team will analyze the results to explore scientific research questions such as how the microbiome influences human health and disease”. The major players in the project appear to be a graduate student, an entrepreneur, a postdoc, a couple of pharmaceutical company employees, a physician, and a couple of independent investigators. How are their relationships influencing their work on this project? Will any of the university-affiliated staff be using your data for publication or will the pharmaceutical companies be using your information for drug development? That may be difficult without IRB approval. Are either of these groups either funding the project or receiving money from it? You should know about the financial relationships that surround a project that might influence your participation. There’s not guarantee that a citizen science project has to tell you any of that.
5) The benefits you’ll receive are grossly over-stated
These are the statements on the site that bother me the most:
“uBiome is not a diagnostic test. However, we can give you valuable information about your microbiome that you can use to learn more about your health. You and your doctor can discuss the results of the test and determine the best way to proceed.”
“First, you can use it to understand what is going on in your body. If you have a tricky health condition and would like more information about your body, your uBiome profile might add useful additional data points. If you are going on a diet, experiencing a bowel flare-up, or changing your lifestyle, we can help you see how your microbiome is before and after. If you would like to know how your microbiome compares to others or to yourself over time (or both), we can help with that too.
Also, the information you contribute allows all of us to learn more about the microbiome, and, going forward, to better understand and possibly cure disease.”
Granted, the investigators do state that uBiome is not a diagnostic test, but they couple it with statements that the results should be discussed with the participant’s doctor and that this information may be useful data in treating difficult disease. It gives the participant hope that their doctor will be able to decode some information from the data. I’ve posed this to a few of my physician colleagues this morning and they agree that they are not sure how they would use this data, if at all. They would probably treat their patient according to evidence based guidelines.
This is especially troubling when an individual has to pay for their data. It’s not hard to make the leap to the scenario where people pay $1337 for the kit with the hope that they will learn something that helps them treat their difficult illness only to find themselves none the wiser and about a thousand dollars poorer.
To the best of my knowledge, there are no medical guidelines for the use of microbiome data and uBiome does not have a medical pathology-grade laboratory. The project should tell people that the uBiome is not a diagnostic test. End of story.
6) There is no statement as to what your risks of participation are.
I am sure that swabbing a region of the body is generally harmless, but it’s also not hard to jump to the scenario where people learn something they may not have realized they would learn. For example, it’s not difficult to imagine a scenario where a mother swabs herself, her husband, and her child and learns that the husband and child share a common STD. Or, where someone swabs a partner purposefully, using this information to make a judgement, but receives a contaminated result.
In our research projects, we must disclose any risks of participation and we must give individuals the right to tell us what information they want to know and what they don’t want to know.
7) Children are included, but it’s not apparent how they are protected.
This brings me to my final concern (for now). It’s not apparent how children are treated, other than an answer to a FAQ about whether you can swab your child that essentially says “Sure!” Can you swab someone else’s child? Can you swab your disabled family member? A prisoner?
I note these groups, among others, because they are specially protected groups in research because of their difficulty providing consent. An IRB takes special care in evaluating research using these groups because they, historically, have been subjected to abuse. There is no evidence to me that consideration of these groups has been given by this project in the same way that it would be considered in a classically-conducted research study.
In conclusion, the project investigators state:
“We are funding this project through citizen science for several reasons. First, we want to bring this technology directly to the public as quickly as possible. Second, we want to prove that citizen science is a viable way to do science, and an alternative to slower, more traditional methods. And finally, going directly to the public does not prevent more traditional funding later — we just wanted to get you all involved as soon as possible!”
I want the average citizen to be excited about science. That’s why I do stuff like blogging. I also think it is valuable when people feel invested in science, but I am frightened by the phrase “we want to prove that citizen science is a viable way to do science, and an alternative to slower, more traditional methods.” Faster science shouldn’t be the motivation for citizen science especially when some aspects of the “traditional methods” exist for a reason – to protect human research subjects and guide investigators in the responsible and ethical conduct of research.
I worry that the use of human subjects by citizen scientists has the potential to return us to an era where human subjects abuses were more prevalent. At the extreme, I worry that “citizen science” projects may provide a route for scientists affiliated with universities and pharmaceutical industries (individuals who should know better) to circumvent the IRB process when it is convenient for them.
I worry that, for as much as we think “citizen science” and “crowdsourcing” is the way of the future, we are actually taking a huge step backward.