Given the evidence that people are ready, willing, and able to engage in online discussions about cancer prevention and treatment, what steps are being taken to ensure that the U.S. (and the world) does not miss this latest opportunity for education and discovery?
The Pew Internet Project‘s latest report on health and health care, Cancer 2.0, provides evidence — and questions such as the one above — for the President’s Cancer Panel to consider.
This paper is a direct result of a discussion held on e-patients.net which convinced me that I needed to reiterate some past research findings and draw people’s attention to some upcoming data points. Thanks for all the input! I’ll let you know how the presentation goes tomorrow.
Update: It went well! Here are some notes from the other speakers (and stay tuned: I hope to add their slides, too).
Naz Sykes, executive director of the Dr. Susan Love Research Foundation, kicked off the meeting with her presentation: “Army of Women: A Paradigm-Changing Research Resource and a New Model for Democratizing Research.”
I enjoyed her forthright explanation of how Army of Women works, its strengths and shortcomings. It is a “just in time” resource for basic scientists and epidemiologists who need biological samples and/or information for breast cancer research (and yes, they intend to expand the model to other diseases). They want to help people go beyond fundraising, beyond pink ribbons, and partner with them for research. By doing so, they educate the public about clinical trial participation and scientific research. Check out armyofwomen.org.
Downsides: 85% Caucasian, 3% AfAm, 3% Latina, 1% Asian, 8% other. Recruitment is English-only because researchers aren’t prepared to work with Spanish-dominant respondents. They intend to optimize their site for mobile devices to take advantage of the popularity of wireless access among minorities. Privacy came up during the Q&A, but only briefly: “Privacy has never been an issue in our follow-up studies.”
Bradley Malin, director of the Health Information Privacy Laboratory at Vanderbilt, presented next: “Repurposing Clinical Data for Cancer Research with Formal Privacy Protections.”
As he put it, “Health care providers are stockpiling enormous amounts of patient information. Many want to shrae data for a variety of worthwhile purposes, such as cancer research. The big buzz word is ‘secondary use.'” But we need to address privacy issues now so we don’t let HIPAA and a misplaced fear of re-identification of individuals block our path. A headline for me was his critique of Paul Ohm’s “Broken Promises” article, which I wrote about last year. Malin says that Ohm exaggerated the threat and caused damage to the field by scaring people when, in fact, “distinguishable does not equal identifiable.”
His closing slide outlined some open challenges:
It takes time and energy to appropriately model risks…but we should do it.
- We need to train experts (my handwritten note: Understatement of the day!)
- May require national centers of excellence
- We need to support longitudinal studies, possibly across multiple datasets (aka, anonymous linking)
We need to determine what is an acceptable level of risk
- Dependent on data type
- Dependent on person
- Dependent on organization
As he said, “HIPAA treats everyone with a broken pinkie the same as everyone with breast cancer.”
One panelist asked if his proposals would add yet another bureaucratic layer to clinical trials. The opposite, Malin replied. This is for retrospective studies, not prospective, and actually simplifies the HIPAA burden by removing the need to repeatedly ask people’s permission to use the data.
Raj K. Puri, director of the FDA’s Division of Cellular and Gene Therapies, was next up with “Challenges and Opportunities in the Development of Cancer Vaccines and Immunotherapy Products.” I confess I didn’t take any notes — not because it wasn’t interesting, but because it was far from my area of expertise.
I spoke next and you can read my paper, Cancer 2.0, to get the gist of what I said.
Ellen Sigal, chair and founder of Friends of Cancer Research, then gave a tour of the problems facing cancer research, calling for a comprehensive plan and the breaking down of institutional barriers so that researchers can work together. Â I found it curious that her call for collaboration — either “Tear down the silos” or “Tear them down,” I’m not sure —Â is trademarked. I put my pen down after that.
Arthur Caplan, director of the Center for Bioethics at the University of Pennsylvania School of Medicine, spoke last on “The high cost of new cancer treatments: can evidence help us ration?” He was a quote machine:
- “Our values help us interpret the evidence.”
- “Politicians have shown no real stomach for data.”
- “A ‘good’ physician is a zealous advocate who provides options and gets informed consent. But we act as if evidence alone will end disputes.”
- “Autonomy beats futility. Evidence of no efficacy does not stop care in court decisions. And when you are treating dead people, you are pretty far out on the spectrum of evidence-based medicine.”
He calls for a “stewardship” model for the doctor-patient relationship, making it OK to not try every treatment and letting patients know that choosing to stop is just as courageous as fighting to the death.
I found his closing remarks particularly fascinating. As he put it, “We have a strong narrative of rescue in this country. We thrill to the Chilean miners’ rescue and we love pulling babies out of wells. We’re not so good at making mines safer and putting a cap on the well.” He asked, essentially, should we fund rescue of the few, the long shot? Or should we fund an increase in quality of life enhancement? Maybe, as he put it, we shouldn’t try to save every single person, at great public cost. Maybe private funding is the way to go: “Torture the rich and have them underwrite the costs of clinical trials for a chance at rescue.” He continued that we need to foster public discussion about where we spend our resources: our money, our time, our quality of life. If we don’t, he said, we can’t have an evidence-based system.
Great post, Susannah, thanks for the recap. I think that one point that Art Caplan made bears extending further. He alluded to prevention when he said that we are not excited about making mines safer or putting lids on wells. He could have taken this metaphor further into healthcare where we have the same aversion to preventive strategies. Yet this is where small investments may reap huge rewards. Though this is not specific to cancer, it is very applicable to carcinogenic exposures, from tobacco to chemicals.
Thanks again!
Yes, he did expand on that point and it’s a good one.
The alternative to the “rescue at all costs” model was expressed nicely in this short video, shot at TEDMED, featuring BJ Fogg:
http://www.youtube.com/watch?v=keFs5CspfJY
Essentially: “What’s the smallest thing we can do to help the biggest number of people, fastest?”
It’s a very interesting question, not just for ethicists and behavior change theorists, but for all of us.
Great paper and post as usual, Susannah. Completely agree that mobile is a game changer in the chronic disease space. It’s why I quit my job as a digital media exec to pursue participatory health. We’re at the very beginning of the intersection of health and mobile and the possibilities are staggering. We, the patients, are all walking networked devices, with access to connectivity, data, servers and people around the world. I often feel like the mobile part of mobile is underestimated, but it has the most potential in healthcare and medical research. Chronic illnesses, much like cell phones, are with us all the time. Symptoms don’t wait until we get back to our desktop computers. Understanding the hows and whys of chronic diseases can’t happen effectively without the participation of patients in their clinical setting, which is anywhere and everywhere. The collection of proactive, real-time data with time stamp and geo-location metadata is going to lead to areas of research that would have otherwise gone unexplored. Never in history has the medical community had access to a significant volume of this type of patient data, entered by the patient at the time of a symptom, medication dosage or meal, at the location of the event. We will have to make sure the incentives are aligned so that this data is properly filtered for research, but the possibilities are endless. An exciting time for those in the participatory health space.
Thanks, Brett!
I started my talk yesterday by asking how many people could put their hand on their cell phone – 100% of the people in the room (those who weren’t already busy tapping on their mobile devices).
In case you missed it, we hosted a somewhat intense discussion that touched on patient-entered data a couple of weeks ago:
http://pmedicine.org/epatients/archives/2010/11/the-future-of-health-robots-enchanted-objects-and-networks.html
And I know that ambient data collection was a key point in the internal debates we had when choosing grantees for Project HealthDesign. I’ll explore your site further, but if you have a chance to respond here, my question is: How much do you expect users to participate in the data entry? How much can be automated or made invisible to the user?
Hi Susannah,
Great questions – and the right ones to be asking ourselves. Use is our #1 focus.
Re: Participation – I have had 6 flare-ups of my condition in my life – 2 of which resulted in hospitalizations. Looking back, those hospitalizations could have been prevented if I was a more communicative patient (which is why GI Monitor now exists). So the question I constantly ask myself is, why was I a better communicator this time? Now – of course I am biased because I built the app, so of course I’m gonna use it. BUT before I built it, I searched for it and it didn’t exist. Had it existed, I would have used it. These are the times we live in. Mobile apps allow us to do things more efficiently. People will soon search for these apps just like they search the internet for health information now. So – to answer your question, I’d expect those who want to keep track of their conditions will enter data often if they’re aware of the app (we’re seeing that now with our users). It is extremely easy and the device is with you at all times. Furthermore, the benefits are recognizable. A little more on that here – http://wellapps.com/mobile-disease-management-study-nthe-right-1
Re: what data can be automated – depends on the definition of automation. Most ODLs need to be entered in some form or fashion (until we’re a Truman Show society). Metadata associated with these observations (time stamps, geo-location data, etc.) can be collected automatically in conjunction with the entered observation. The data entry will continue to get simpler with cameras, IVR (Interactive Voice Response) and speech to text, but technology is not yet at the point of understanding human feelings (although I’m pretty sure Apple is working on that).
The important point to note here is that how a person feels often dictates the course of treatment in a chronic condition. Physicians are tweaking treatments according to a conversation with the patient. And patients are terrible communicators. I got a degree from the #1 Communications school in the country (at the time – hope it still is), and I sucked at telling my doctor how I felt. It doesn’t have much to do with education. It was just plain old difficult to do. The bottom line is that available technology makes this transfer of information more accurate and efficient. The problem of an unmotivated patient is an entirely different discussion.
Hope that helps answer your questions.
Thanks, Brett! In case anyone reading this doesn’t know, ODL stands for Observations of Daily Living.
Here’s a great blog post by Julie Murchison to explain it further:
ODLs and Patient Engagement
http://projecthealthdesign.typepad.com/project_health_design/2010/04/odls-and-patient-engagement.html
Brett,
great comments!
I am a great fan of ODLs but I also see limits in their usefulness. The same applies to a lot of the structured data that is now being collected.
The next iteration of a rapid learning system will have to include these patients ODLs as well as a large mix of structured data bits but it will only succeed if it intimately associate them with the narratives from the same patients, in order to provide context to each.
Narratives remains the richest data sets from patients when you have active communities of informed patients. Unfortunately the cost of using and making these data sets in a computable format is the highest and therefore most efforts are made to bypass them.
I think it is a tremendous error. The GSTI-571 aka Gleevec aka Glivec aka Imatinib story that Susannah told in her presentation is essentially a story demonstrating the amazing power of patient narratives.
Gilles – Completely agree re: patient narratives. But we need to crawl before we can walk. The iPhone was a perfected iPod before it ever made a single call. :)
To prepare for the rise of patient narratives, our app. (GI Monitor) allows for free text notes to be entered alongside structured data. The flexibility of NOSQL DBs like Couch and Mongo – combined with the emerging Natural Language Processing tech will lead to amazing discoveries…I’m sure. But it all starts with patient data entry and we need to get the basics started asap.
First – it needs to be easy and user friendly since we’re introducing a new flow. Secondly, the patient needs to be recording data for the right reasons. Hopefully, to improve quality-of-life.
To me, physician care is up against a very simple communication barrier that mobile can help solve immediately with some structured data capture. There is no doubt that this will lead to narratives…but we shouldn’t jump to step 3 before we get step 1 right. Here is quick video essay I did to explain the first step towards better doc/patient communication using mobile. http://wellapps.com/patient-doc-communication-far-perfect-video-essay
Know yourself well,
Brett
Yes, he did expand on that point and it’s a good one. The alternative to the “rescue at all costs” model was expressed nicely in this short video, shot at TEDMED, featuring BJ Fogg: http://www.youtube.com/watch?v=keFs5CspfJY Essentially: “What’s the smallest thing we can do to help the biggest number of people, fastest?” It’s a very interesting question, not just for ethicists and behavior change theorists, but for all of us.