Summary: Asthmapolis, a Madison, Wisc.–based company, uses mobile technology to identify where and when asthma patients use their rescue inhalers to cope with symptoms. The technology helps patients realize when their disease is not adequately controlled while helping public health authorities identify the location and likely causes of asthma attacks.
By Sarah Klein
In 2009, David Van Sickle, Ph.D., launched Asthmapolis, a Madison, Wisc.–based company that uses mobile technology to pinpoint where and when asthma patients use their rescue inhalers to cope with asthma symptoms.1 Each time they use it, a GPS-enabled sensor sends out a signal that is captured in the company's database and analyzed to create a snapshot of inhaler use. The analysis, together with information on how to improve asthma control, is e-mailed to the patient on a weekly basis and plays a critical role in helping patients with poorly controlled asthma realize they need additional medical help. When these data are aggregated, they help researchers and public health authorities more precisely identify the causes of local asthma attacks and develop successful interventions to prevent them.
Quality Matters asked Van Sickle about his product and how mobile applications like it will improve health care quality.
QM: What prompted you to create this device?
Van Sickle: I had been working for the Centers for Disease Control and Prevention (CDC) in the Epidemic Intelligence Service, an arm of the agency that looks for unexpected clusters of diseases to figure out what is going on and how public health can address them. I found it frustrating that in the case of asthma public health surveillance had been largely limited to the retrospective analysis of the most severe exacerbations—people who were hospitalized or who died. We weren't learning much about the day-to-day morbidity that causes millions of unscheduled office visits and missed days of school and work each year.
I was also frustrated that despite all we'd learned about asthma and how to treat it, we hadn't been making much of a dent in the rates of emergency department visits and hospitalizations, nearly all of which are preventable. Unfortunately, the majority of people with asthma—something like 60 to 75 percent of patients in the U.S.—still don't have their symptoms under control. As a result we spend $56 billion on asthma each year. 2
QM: How does electronic monitoring of inhaler help?
Van Sickle: Use of a rescue inhaler—the type of inhaler patients turn to to relieve symptoms —is the most important signal we have of how someone is doing. It can be more reliable than the patient themselves, because the majority of patients with uncontrolled asthma—85 percent—will tell you they are doing fine. They've learned to live with and tolerate symptoms and don't realize a lot more could be done to prevent them and improve their quality of life.
Real-time monitoring of inhaler use also helps close the public health information gap by providing a view of where and when people are using rescue inhalers in a community. That data can be used to identify areas where environmental triggers, such as air pollution or workplace exposures, lead to frequent symptoms.
QM: How hard was it to create the device? You're a medical anthropologist by training.
Van Sickle: I pulled together a team of undergraduate engineering students and we built up four or five prototypes using cell phone parts and GPS receivers that we used to understand how such a system would work. That gave the project a little momentum. I took them to the CDC and said, "Here's an opportunity to think about public health surveillance for asthma in a different way. Would you be interested in doing an evaluation of it?" They agreed, sponsoring one evaluation in 2009 and another in 2010.
QM: What did you find out?
Van Sickle: We had 40 patients in each trial. As you would expect, the majority of them entered with uncontrolled asthma—more than half. In the course of the trial, we demonstrated that were able to rapidly assess a patients' level of control and use that information to help them improve their management. We were able to reduce the number of patients with uncontrolled asthma by 50 percent. Overall, 75 percent of the patients that took part improved their level of asthma control to some degree. We also learned a lot about how it works; that is, how the information from the inhaler can be used to improve management. There were a number of different ways: some people recognize for the first time that they have work-related asthma. Some are finally able to communicate with their physician that their asthma is not well controlled. Others identify new triggers in their daily life that lead to more frequent symptoms.
QM: What happened in the control group?
Van Sickle: These were simple (and limited) pre-post evaluations; there were no control groups in either study. Nevertheless, both were designed with a run-in period to establish a baseline; during this time, participants received no information or feedback from the sensor and we saw no change in their level of control. Patients only improved after we gave them online access and began sending them weekly e-mail reports. We're preparing now for our first large-scale randomized controlled trials later this year that will help us better answer some of the questions about how well it improves asthma outcomes. In the mean time, the technology provides a tool to help physicians remotely assess asthma control (and adherence, when used with a controller medication).
QM: There have been a number of tools developed over the years that rely on feedback and education to help patients gain control of their asthma symptoms, including video games for children. How do you engage patients with your product?
Van Sickle: For every patient, it is a little bit of a mystery about what is causing their symptoms to come and go. What we are hoping to do is set them on a path of discovery by providing feedback in the form of dashboards, maps, and charts. This gives them a view of what's happening to them at particular times of day and place and what's happening to other people around them. We developed the weekly e-mail report to summarize how well someone is doing and provide practical, data-driven guidance about how they could do better. For example, many patients are not aware that symptoms occurring at night may be a sign of worsening asthma. Saving time for users is important—and something I think some mobile app developers overlook. We tend to underestimate the day-to-day work of managing a chronic disease. We can't add to the burden of that. We have to figure out how to provide information and meaning that lightens the burden of managing asthma and that effectively serves as a force multiplier for patient. Unfortunately, I think what you are seeing now is a lot of the translation of the pen and paper approaches of the past to electronic interfaces.
QM: What about physicians and other providers? Have they been receptive to receiving and using the information?
Van Sickle: To be clear, in our first two projects, we didn't relay any information to providers. We were entirely patient focused. That was really driven by the medical school's institutional review board, which said if your product isn't functioning well, we don't really want that information to get into the clinical pathway. We had to show it was going to work reliably, which was a sensible approach. But now that we are about to roll out the product commercially, it's a totally different situation; still, a similar challenge exists. The most frequent refrain we hear from health systems is we don't want mobile health tools or remote monitoring to interfere with our existing workflow.
QM: What's their concern? That unintelligible data would be dumped on them?
Van Sickle: There's that. It's a challenge the whole industry faces: how do you give providers a way to quickly survey and shepherd a population of patients—see how their patients are doing, who is not well controlled, and who is worsening and needs more attention. There's also the issue of reimbursement. There is no payment or financial incentive to read e-mails and there's a strong disincentive to getting a stream of data that you don't have time to process. If you miss something, you're on the hook.
QM: How do you plan to overcome that?
Van Sickle: One of the first things we want to demonstrate is that we're a useful adjuvant to existing tools for managing asthma and understanding how your patients are doing. The tools physicians have now—such as quarterly questionnaires—don't provide a complete picture. We think objective, near real-time instruments that passively and objectively capture information about asthma from daily life can provide much more value. In addition there's a lot we can do just by helping a patient improve their management.
QM: So what's your business plan? Will you sell the product to patients? To insurers?
Van Sickle: To anyone who is at risk for the economic consequences of poorly controlled asthma. There is a pretty strong economic case to be made for improving asthma control. A patient with uncontrolled asthma requires an additional $3,000 to $4,000 a year in health care compared to a patient with controlled asthma.3 To lower the costs of asthma, you need a system to help you identify who those people are, that gets them engaged in management and that provides you with robust ways to monitor and evaluate whether they are improving. As a business, we provide health plans and clinicians and self-insured employers with those tools, for a tiny fraction of what they're paying to care for uncontrolled asthma.
QM: Do you see other uses for the data?
Van Sickle: We often run smaller projects with academic research teams and public health groups, where we provide them with technology to collect information on specific research questions. For example, right now we have a project getting started with the VA Puget Sound Health Care System applying our technology to understand patterns of exacerbations of chronic obstructive pulmonary disease. It's a comparable kind of disease state, with overlapping medications, and has similar patterns of unexplained exacerbations. Scientists and public health agencies are interested using the technology to understand health effects of a variety of environmental exposures. Characterizing respiratory symptoms resulting from things like red tide [algae bloom] events, or in the case of Hawaii we have a project where they are looking at the effect of volcanic sulfur dioxide—smog emitted by an active volcano. It has harmful respiratory health effects particularly in kids with asthma so you can use the sensors coupled with environmental sampling and modeling to better characterize and understand that burden and what you can do at a public health and policy level to minimize it.
QM: How do see mobile technology changing the way we conduct research?
Van Sickle: It's been happening out of necessity. Academic researchers and public health agencies increasingly struggle to get people to carve out time for surveys and in-depth examinations. In response, you're seeing the emergence of a complementary model there, which revolves around the collection of micro-scale events and data from daily life. Of course the other benefit is that ecologically these data are more valid than what you were getting before. And the aggregation of it creates new possibilities for research. Take asthma research, which has relied heavily on daily diaries. There are numerous studies to suggest that few of them (as low as 6 percent) are actually completed and that the information is often fabricated.4 When those kinds of instruments are at the core of how you capture data for scientific research, that is a big problem.
QM: What do you see as the future of mobile health technologies?
Van Sickle: It's an incredibly exciting time. For example, we're witnessing mobile technology completely change the way we think about eye exams. I'm referring to Ramesh Raskar's cell phone clip-on and software that allows users to determine their eyeglass prescription. Or Microskia, which created a device that turns a cell phone into a networked microscope. Medic uses cell phones to develop robust electronic medical record systems in low-income countries. These are examples of elegant, exciting technologies that sprang from the relentless work of passionate innovators who just will not give up. It is these folks who are going to eliminate the frustrating gaps between what we should be able to achieve in health care and our actual accomplishments.
QM: What challenges do you see ahead?
Van Sickle: I think a lot about the public health infrastructure and how our skills and tools measure up to those at work in other sectors, whether it's finance, or weather, or transportation. I'm not sure that were seeing enough people enter public health with the necessary data management skills. They are accustomed to doing static, retrospective analyses on accumulated datasets and to fielding evaluations on the back of those kinds of time-scales. Part of the problem is education and training. I'm concerned that public health students right now are not coming out of school with the skills they need to make sense of multiple data streams in real time. It is not trivial work—and it is not going to get easier. There is no trend toward fewer phones or less data. It is just going to get more complicated.
1 Rescue inhalers provide emergency relief when inhalers used for long-term asthma control do not fully relieve symptoms.
2 S. L. Barrett and T. A. Nurmagambetov, "Costs of Asthma in the United States: 2002-2007," Journal of Allergy Clinical Immunology, Jan. 2011 127(1):142–52. This estimate includes productivity losses from work days and school days lost because of morbidity and productivity losses from mortality. Together those represent 8 percent to 12 percent of annual costs between 2002 and 2007.
3 S. J. Szefler, "Advances in Pediatric Asthma in 2010: Addressing the Major Issues," Journal of Allergy and Clinical Immunology, Aug. 2011 127(1):102–15.
4 H. L. Yoos, H. Kitznian, A. McMullen et al., "Symptom Monitoring in Childhood Asthma: A Randomized Clinical Trial Comparing Peak Expiratory Flow Rate With Symptom Monitoring," Annals of Allergy, Asthma & Immunology, March 2011 88(3): 283–91; and P. Verschelden, A. Cartier, J L'Archeveque et al., "Compliance With and Accuracy of Daily Self-Assessment of Peak Expiratory Flows in Asthmatic Subjects Over a Three Month Period," European Respiratory Journal, May 1996 9(5):880–5.
