Overcoming Hurdles for Effective Heart Failure Telemonitoring

Remote patient monitoring for chronic conditions such as hypertension, diabetes and chronic obstructive pulmonary disease (COPD) have demonstrated successful outcomes. The evidence for heart failure (HF) monitoring, however, is less definitive. Yet, newer approaches may turn things around.

Complexity of heart failure

The Centers for Disease Control and Prevention estimate that remote patient monitoring of congestive HF, diabetes, COPD and chronic wounds or skin ulcers could reduce costs to the U.S. healthcare system by nearly $200 billion over the next 25 years. For HF telemonitoring to contribute its fair share to the savings, researchers will need to strengthen the evidence.

Klersy et al found cost savings associated with remote HF monitoring from 21 randomized controlled trials between 2000 and 2009 (mostly driven by reduced hospital admissions). However, they said the economic data were “scanty” and the approach to economic analyses were not uniform. They also noted the follow-up time was “limited" (Eur J Heart Fail 2011;online Feb 1).

Guy Paré, PhD, research chair of healthcare IT and professor of IT at HEC Montreal, found similar challenges with telemonitoring HF studies. He and colleagues analyzed 62 randomized controlled studies spanning from 1966 through 2008 (J Med Internet Res 2010;12[2]:e21). Overall, diabetics, asthmatics and hypertensive patients were able to improve their health, but findings for HF patients were “equivocal” and researchers called for larger and better designed trials.

Heart failure  is the most complex and perhaps the most unpredictable condition among those studied, which could explain the disparity in telemonitoring success; however, the literature is inconclusive, Paré says.

He notes several problems with HF telemonitoring studies. First, most do not extend beyond six months, “a considerable limitation.” Second, patients are typically compared with a control group, rather than their prior use of healthcare resources pre-intervention. Finally, the telemonitoring feedback does not usually empower patients to take appropriate actions for their health. “These systems need more than just preprogrammed protocols. They need to react in real time to the patient’s condition,” Paré says.

Paré and colleagues are planning two randomized controlled trials where patients will be followed for 12 months prior to the intervention and 12 months post-intervention. The technology being used has certain intelligent features that could impact self-care, Paré says. “My hypothesis is that the patients who are involved in managing their health will see a greater improvement in quality of life and clinical conditions, with reduced hospital readmissions.”

Following 50,000 patients remotely

In 2009, the Medicare Payment Advisory Commission estimated that 17.6 percent of all Medicare hospital admissions are readmissions, and a majority of them are avoidable. Readmissions cost $15 billion annually and if successfully prevented, could save Medicare $12 billion annually (Frontiers Health Serv Mgmt 2009;25[3]:3-10). The Veterans Health Administration (VHA) has shown remote patient monitoring saves money and reduces hospital admissions, albeit on a very large scale.

Between July 2003 and December 2007, the VHA introduced a home telehealth program across the U.S. During that time, the population being served increased from 2,000 to more than 31,000, and is expected to reach 50,000 this year (Telemed e-Health 2008;14[10]:1118-1126). Echoing Paré, Darkens et al wrote that “promoting patient self-management is a fundamental underpinning” of the telehealth model.

Implantable Devices Send Signals of Success

Studies of implantable cardioverter-defibrillators (ICDs) and cardiac resychronization therapy devices with defibrillators (CRT-Ds) that can wirelessly transmit diagnostic information from heart failure patients to providers are promising. In the CONNECT trial, Crossley et al randomized 1,997 patients to wireless remote data transmission with clinician alerts or in-office visits (J Am Coll Cardiol 2011;online Jan 19). The median time from a clinical event to a decision per patient was reduced from 22 days in the in-office arm to 4.6 days in the remote arm, and length of stay per hospitalization was reduced from four days to 3.3 days, respectively, an estimated savings of $1,659 per hospitalization. There was no difference in mortality between the two arms. In the ALTITUDE trial, Saxon et al examined a subset of 3,018 patients who also transmitted weight and blood pressure data an average of 2.5 times per week (Circulation 2010;122;2359-2367). The CRT-D patients transmitting these data had the lowest mortality risk compared with other networked CRT-D patients. Researchers hypothesized that improved outcomes could be due to patients “encouraged to be more aware and in touch with their health status.”

To help the telehealth effort, the VHA established a training center in 2004 and trained 1,500 care coordinators onsite and 5,000 through the internet in four years. In the referenced study, Darkens et al found an overall 25 percent reduction in bed days of care and a 20 percent reduction in numbers of admissions. Categorized by chronic condition, the telemonitoring system reduced healthcare utilization by 30 percent for patients with hypertension, 26 percent for HF, 21 percent for COPD and 20 percent for diabetes.

The program costs $1,600 per patient per year, compared with $13,121 for home-based primary care services per patient per year and $77,745 for market nursing home care per patient per year.

Researchers credited the success of the program to the VHA’s “comprehensive and systematic approach to the clinical, educational, technology and business processes” and called it a “groundbreaking achievement.” Translating that kind of success for HF patients to conventional healthcare systems presents a host of challenges, but one answer may lie with smartphones.

Monitoring goes mobile

Emily Seto, Msc, from the Centre for Global eHealth Innovation at the University Health Network in Toronto, and colleagues spent a year developing protocols and algorithms to remotely monitor HF patients via Bluetooth technology and smartphones. They recently completed a randomized controlled trial of the system (50 controls, 50 in the intervention arm) and will present findings at the American College of Cardiology conference in April.

The goal was to empower the patients to take better care of themselves. “If there were issues regarding a patient’s status, the system was intelligent enough to prompt them to take additional readings,” Seto says. “We found positive results in terms of self-care, clinical management and hospitalizations.”

These systems need to interact with patients, Seto says. Generally, more interactive protocols have fared better than systems where patients send data to providers and wait for a response. Using a telephone-based monitoring system, for example, Chaudhry et al reported no difference between the monitored and usual care groups in respect to recurrent hospitalization or death (N Engl J Med 2010;363:2301-2309), whereas a recent review of 25 randomized controlled trials was very favorable for telemonitoring with wireless transmission and only slightly favorable for structured telephone support (Cochrane Database Syst Rev 2010;Aug 4[8]).

When setting up such programs, facilities should consider several factors including: which components make a program effective; the optimal duration and intensity of monitoring; which patients benefit most from monitoring compared with face-to-face programs; and the sensitivity and specificity of responding to monitored events that could lead to a morbid or fatal event (Eur J Heart Fail 2011;13:115-126).

“The success of a system is highly dependent on the system itself, as it has to benefit all end-users including patients and clinicians,” Seto says.