Smart Moves? E-patient Systems Rise Amid Skepticism

OncologyLive, June 2011, Volume 12, Issue 6

During the past 20 years, disease management systems have mushroomed in medicine

During the past 20 years, disease management (DM) systems have mushroomed in medicine, expanding from call centers staffed by nurses to electronic devices that collect streams of data to pump into doctors’ offices. Heart patients are strapping on wristband monitors, and diabetics are checking their glucose meters; there’s even an “intelligence toilet” system that tracks several health indicators.

Now, as many cancers evolve into chronic diseases, momentum is building for DM systems in oncology that, among other things, remind patients to get diagnostic tests, help them manage adverse events, and talk them through their bouts with depression. One trade group estimates that nearly a fifth of cancer patients already use some form of DM and, collectively, spend $250 million a year on it.

Amid the growth, though, questions persist about whether DM can deliver on its goals: better quality of life and treatment outcomes for patients, more efficient systems for busy physician practices, and cost efficiencies for healthcare payers. Indeed, studies conducted thus far have concluded that DM does nothing to cut medical bills and little to improve health outcomes in chronic illnesses far less complex than cancer.

"As much as the studies show questionable benefits, the alternative of having people hung out to dry is not acceptable to most health plans and employers. That’s why you have this growth."

—Al Lewis

—Al Lewis, president of the Disease Management Purchasing Consortium International

Such disappointments have spurred efforts to design better DM tools, but they have hardly dampened overall enthusiasm. The idea that you can keep patients healthier with better communication and smart tools makes so much intuitive sense that the DM market keeps booming.

Annual revenues for independent DM vendors have jumped from $78 million in 1997 to $1.2 billion in 2006 and to $2.8 billion today, according to the Disease Management Purchasing Consortium. Add in the money that insurance companies spend on their own systems and it’s a $5 billion industry, the group says.

Looking forward, analysts expect DM to grow 10% a year. Oncology DM is growing faster still—even without experimental technology that would boost growth higher and software systems that would allow oncology practices to make sense of the data collected.

“As much as the studies show questionable benefits, the alternative of having people hung out to dry is not acceptable to most health plans and employers,” said Al Lewis, president of the consortium. “That’s why you have this growth.”

Human Touch Is Part of Electronic Advantage

The concept of DM began 2 decades ago with cardiology patients. Several small trials identified patients who came to major academic hospitals with heart failure. After these patients were sent home, the teams they’d worked with at the hospital followed up regularly, both over the phone and face to face, asking what patients had done over the past few days and reminding them of what they needed to do over the next few days.

Repeat hospital visits fell. Patients developed healthier habits. Medical costs dropped. And a movement was born. DM began to boom—not because doctors embraced it, but because employers and insurance companies thought it might cut costs by keeping chronically ill patients out of the hospital.

Rather than using the hospital staff for the follow-up work, the larger programs created by insurers or DM vendors—those that are being increasingly employed in oncology—find it far more efficient to use teams of nurses who work in call centers and do all of their work over the phone.

But such changes, DM skeptics say, eliminated benefits that came with initial trials, where doctors and nurses tapped their relationships with patients and knowledge of their needs.

“After a decade of market experimentation with limited success, new thinking is called for in the design of DM programs,” argued Brenda R. Motheral, PhD, in a review of telephone-based models in The American Journal of Managed Care (2011;17(1):e10-e16).

“Targeting of specific patients and activity combinations based on risk, actionability, and treatment and program effectiveness has the potential to deliver a cost-saving DM program if combined with an outreach model that brings the care manager much closer to the patient and physician,” she said.

—Scott Josephs, MD

"We can see when two medications prescribed by different doctors will interact. We can see if patients actually fill the prescriptions that doctors write them. We can see when a patient is due for a certain test and remind them to get it."

—Scott Josephs, MD, a national medical officer at Cigna Corporation

Indeed, Miaskowski et al found pain intensity scores decreased significantly for patients with cancer randomly assigned to an intervention program while patients in standard care experienced an increase (J Clin Oncol. 2004;22(9):1713-1720).

The trial, which included 174 patients, tested a program in which specially trained nurses taught patients how to use a pillbox and discuss pain with their doctors through 2 in-home visits and 3 phone calls.

Insurers counter that some big benefits come from operating their own DM systems, rather than relying on doctors to do whatever is necessary.

“We can see the whole patient in a way that’s hard for individual doctors, and that allows us to provide real value,” said Scott Josephs, MD, national medical officer for Total Health Management at Cigna Corporation.

“We can see when two medications prescribed by different doctors will interact. We can see if patients actually fill the prescriptions that doctors write them. We can see when a patient is due for a certain test and remind them to get it.”

High-Tech Approaches Grow

But the real value of DM may be yet to come, starting with technology designed to convince patients with cancer to take their medicine.

In 1990, virtually all cancer treatment occurred at a doctor’s office. By 2008, 20% of all money spent on oncology went to oral cancer medications, and the number was growing by 25% a year, according to independent pharmacy consultant Barbara Hawes, RpH, MBA.

Oral cancer treatments, generally designed to be taken daily at home over long periods of time, have one major drawback: Many patients do not take medication exactly as directed. Ruddy et al found that adherence and persistence rates were as low as 16%, depending upon the type of cancer and the medication (CA Cancer J Clin 2009;59:56-66).

It is a problem that everyone wants to solve.

Barbara Rapchak, founder of Leap of Faith Technologies, developed eMedonline (inset).

For example, the National Cancer Institute paid Leap of Faith Technologies, a small company in suburban Chicago, to develop a system that uses smart phones, radio-frequency identification technology, and the Internet to ensure patients take their medicines.

The Leap of Faith system, called eMedonline, stores all the information about what a patient needs to take and when. It then alerts the patient at the proper times and has the patient scan the medication into the phone as he takes it. Failure to take the right medication at the right time leads to a phone call, first an automated one, and, if necessary, a live one.

The eMedonline system has boosted compliance to >95% in trials and is ready for commercial use.

“It can handle any amount of complexity,” said Barbara Rapchak, who founded Leap of Faith 21 years ago. “We had one patient on 27 meds a day, all intricately timed.”

The eMedonline system sounds an alert and forces the patient to scan the medication at the correct time. Other tools take automation further still.

Doctors ask patients with many diseases, including some types of cancer, to track their weight and report any significant changes. The traditional tracking tool was a handwritten log, while the traditional reporting tool was a telephone. Now there are scales that automatically send figures to computer programs that pass all the information on to doctors.

The smart toilet system, which a Japanese company sells for up to $6000, can track and report not only the sugar in urine but also the weight, body fat, and blood pressure of any user. For children, there’s a blood glucose monitoring device that plugs into a videogame system so young patients can register their readings while they play.

The eMedonline system, left, tracks patient care. MIT researchers work on implantable nanosensors, right.

Nanosensors Among Future Developments

Looking ahead, researchers are developing tools that require even less of users.

Researchers around the globe are working on nanoparticles that continuously monitor cancer patients for substances related to their particular cancer and transmit wireless information about what they find.

Implantable devices are already commercially available for some conditions. Pacemakers have been gathering data for well over a decade, and the first implantable blood sugar sensor hit the market in 2005.

Disease Management in Oncology Settings

The Core Component

Call Centers: The bedrock idea of disease management (DM) has always been that greater communication will help patients take better care of themselves while providing medical professionals with more of the information they need to spot and treat problems early. Rather than paying a patient’s own doctors and nurses to call regularly, insurance companies pay DM vendors to maintain call centers stocked with nurses who are connected to patients at intervals determined by software.

Available but Still Unusual

Medication Trackers: The single most important home treatment for most diseases is prescription medication, properly taken. Unfortunately, surveys suggest roughly half of all patients don’t take medicine properly if left to their own devices, which is why the market abounds with high-tech helpers. Beyond the plastic day-of-the-week pillboxes, vendors have developed systems that call a patient’s phone when it’s time to take a pill, pill bottles that e-mail the doctor if the patient doesn’t open them frequently enough, and even in-pill microchips that transmit information about when the patient swallows them.

Networked Sensors: If a medical device exists, some vendor has devised a version that connects with a computer every time it is used, stores each measurement in a journal of some kind, and alerts someone about worrisome numbers. Scales, thermometers, pedometers, heart rate monitors, blood pressure cuffs—the list is endless. Adoption has been slow because doctors don’t yet have the technology to sort so much data and because many devices work only with their own software rather than larger DM systems.

Well-Being Calendar Software: Each disease comes with its own therapeutic recommendations, dietary restrictions, and key indicators, which might be weight or blood pressure, or any other biological marker. Smart programs can be tweaked to keep motivated patients organized and motivate reluctant ones. These programs ask patients if they’re experiencing problems that are common to individual diseases and their treatments. They also help coordinate treatment for multiple chronic diseases, pointing out that the vigorous exercise that helps a diabetic might not be as beneficial for the same patient’s heart condition.

Down the Road

Automatic Implantable Devices: Some exist already, but the future will see tiny products designed to rest somewhere inside the human body gathering data, looking for problems, and streaming information out to computers.

Data Analysis Software: The mountain of data that such devices provide will overwhelm physicians’ capacity for analysis—and his or her time. The real value will come with software that can analyze patterns in data from thousands of patients and spot clinically important trends.

Cancer presents a more difficult challenge because researchers are still seeking the relevant biomarkers for many types of tumors, and even markers that are known may appear in different parts of the body that are often quite inaccessible.

Still, doctors and scientists at the David H. Koch Institute for Integrative Cancer Research at the Massachusetts Institute of Technology think they’ll change oncology with implantable nanosensors that will tell doctors exactly how they need to treat individual tumors and evaluate whether the treatment worked.

The nanoparticles are made of iron oxide. Antibodies are attached to the surface of the particles. The specific antibodies are chosen to bind to proteins that are secreted by the tumor cells.

Michael C. Cima, PhD

“The radiation dose that is appropriate for a tumor depends on the concentration of oxygen in the tumor,” said Michael C. Cima, PhD, an engineering professor who leads the project. “If we have a sensor there, we can measure the oxygen before deciding on the radiation and then get the dosage right.”

After treatment, implantable devices could stay in the patient for years, looking for any sign of new tumors. Cima hopes such products will begin clinical use in 5 years.

Software, Privacy Challenges Loom

The sheer number of networked devices on the market and in the works will generate an astonishing amount of data and create a massive potential for improving patient health. But even the biggest boosters of DM worry that much of that potential will go untapped because the various tools implemented by medical practices and insurance companies won’t talk to one another.

“Very little is truly automated right now,” said Lewis, of the Disease Management Consortium. “No DM software receives information from the pharmacy about a new heart medication the patient just got, calculates that it interacts with his cancer medications, and alerts doctors automatically. A human must pick up the phone and that limits how much information gets shared.”

Privacy laws also limit benefits.

“The nurses at one insurer’s DM system might spend a couple years figuring out a great system to motivate a particular patient to live healthy, but if that patient gets a new job, the nurses at the next insurer’s system will have to start from scratch because insurers can’t share information,” Lewis said. “It makes sense to limit what insurers can say to one another, but regulators have a real opportunity to improve patient health if they can make rules that allow practical information to be shared while still protecting privacy.”

Solving such problems requires not only that dozens of manufacturers agree on standards, but also that politicians and regulators find smart ways to update privacy safeguards for the information age.

If DM makers can solve these problems— insurers, researchers, and observers agree—it will grow faster still.

Andrew D. Smith is a freelance technology journalist based in New Jersey.