The Connected Patient Disconnect
Having been supported by several small business grants from the National Cancer Institute to create online interventions for cancer patients, I have been learning gradually about commercialization models to get our work out to the public. I am dismayed about the major disconnect between eHealth entrepreneurs and eHealth intervention researchers (my personal reference group).
Last year I attended Stanford Medicine X and last week I did a demo of one of our web sites at Health 2.0 in Santa Clara. Both times, I was struck by the assumption in the IT developer and consumer community that giving people realtime feedback about their health will automatically result in major positive changes in behavior, not to mention cost savings for insurers.
The Connected Patient movement seems particularly naïve to me. Psychologists have been using self-monitoring, i.e. recording behaviors such as smoking, eating, and exercise, for at least 30 years to promote behavior change. First we used paper-and-pencil diaries, but researchers like Saul Schiffman quickly adapted the first handheld computers to prompt people to record their behaviors in realtime, greatly increasing the accuracy and power of self-monitoring.
As technology has advanced, so have our means of self-monitoring. Overall, however, the technology matters far less than the procedure itself. For most people, tracking their smoking, calories, mood, or steps does change unhealthy behaviors somewhat, for a limited period of time. A small group of highly educated, motivated people is more successful in using self-monitoring to make larger, more lasting changes.
I was reminded of this last year in a seminar on tracking at Stanford Medicine X, when a concierge physician from San Francisco and several of his patients talked about being empowered to change their health by using feedback from various types of sensors. One had paid out of pocket for a continuous blood glucose monitor since his insurance would not cover the costs to use it for his Type II diabetes.
Another doggedly demanded access to the data from his cardiac defibrillator. They believed their experiences heralded a sea change in health care in the United States. I am all for empowering patients with knowledge, tracking tools, and social support.
However, if knowledge and feedback was all it took to change unhealthy behaviors, psychologists would be superfluous in the world
As I listened this year to Peter Diamandis, M.D. talk about using wireless sensors for fitness(along with his disclosure that he arrived at Health 2.0 on his own personal jet), I thought about the statistic that at least 20% of Americans are functionally illiterate—and of course there is a huge correlation between social class and health. How many people living on food stamps in our inner cities are wearing Fitbits®?
How can we make eHealth effective for a wider group of people? Researchers in eHealth interventions combine self-monitoring with an array of other intervention features, such as tailoring messages specifically to the barriers that are limiting behavior change for an individual patient, incorporating cognitive-behavioral therapy exercises, and experimenting to find the most cost-effective way to supplement a web site or app with automated feedback or human contact in the form of email, phone, videoconferencing, or face-to-face visits.
Our research is not judged by the number of people who “like” an intervention on Facebook, being included in an insurance company’s collection of “vetted” apps, or even the number of people who use a site or app more than once, but by carefully documenting actual health outcomes, such as quitting smoking, weight loss, sleep quality, fitness parameters, or changes on questionnaires measuring depression, anxiety, or even (in my case) sexual function and satisfaction.
We analyze our data to see if recordings of the amount and type of usage of an intervention actually predict whether it works. This is not an easy type of research to do. I was listening to a lecture this morning by Deborah Tate of the University of North Carolina, an expert on using eHealth interventions to treat obesity. She pointed out that in the period of time between submitting a grant and getting it funded, the technology we propose to use often becomes outmoded (especially these days when getting grant funding is increasingly difficult).
However, it is the study design that is crucial, not whether the research is ultimately done using smartphones or Google glasses.
I would love to see more communication between the developers, gamification experts, innovators, and healthcare organizations that attend Health 2.0 and the community of behavioral scientists who present work at meetings such as the Society of Behavioral Medicine (SBM) or the International Society for Research on Internet Interventions (ISRII).
As researchers, we need to learn how to move more nimbly, keep up with changing technical possibilities, and how to scale up our interventions to make them commercially viable. But we also have a body of knowledge that your community needs if we are going to realize the potential of these new technologies to change health.
I am not speaking without personal experience, either. This year I was diagnosed with breast cancer, and after staring for a day or two at Adjuvant Online!, a decision tool that shows the costs and benefits of having chemotherapy for a woman with a hormone-positive tumor, I decided to pass up 6 months of personalized toxic chemicals to achieve a 5% decrease in my risk of cancer recurrence and at least a 2% further increase in my cardiovascular health risk over the next 10 years.
Instead I am finally tackling the massive amount of weight I gained after having my son at age 40 and returning to work fulltime (he is now 21, so my excuse is long gone). I have lost 40 pounds, use calorie-counting and exercise-recording apps on my iPhone, and wear a fitness tracker.
However, I still need to lose 30 more pounds to get to a truly healthy BMI, and I can feel my resolve flagging. Help! I think I need a psychologist!
Leslie R. Schover is a professor of behavioral science at the UT MD Anderson Cancer Center. This op-ed originally appeared on The Health Care Blog.
