Since the advent of the Coronary Care Unit (CCU) in the early 1960’s, the field of cardiology has led in the introduction of new healthcare technology. Cardiac monitoring and resuscitation lowered the mortality rate for heart attack patients by 50 percent. This was followed by advances in cardiac imaging, prevention, monitoring techniques, and therapeutic interventions. The result has been an increase in our longevity.
However, with that longevity, comes a price to pay in terms of the diseases of aging. The most obvious is dementia, but heart disease, particularly heart failure, has been on the rise in recent decades. Today, there are an estimated 6 million Americans with clinical heart failure. It is the most common cause of hospitalization under the Medicare program as well as the largest expenditure. The result has been an attempt to shift the management of heart failure to the outpatient setting to improve cost effectiveness. The creation of programs for remote patient monitoring coupled with efforts at patient and caregiver education has been a challenge because reimbursement for these services has lagged behind even though they have proven to save money. Using the parameter of 30 day rehospitalization rates, successful monitoring programs have achieved a 30 percent reduction in rehospitalization with significant cost savings, but, at the same time, other programs using similar technology have failed to demonstrate benefit. What can explain the disparity in the results?
"The first attempts at home monitoring utilized a telephone land line, a scale, a pad and a pencil"
Let’s start with the technology. The measurement of daily weight with a scale has long been the surrogate for the fluid retention that is the hallmark of worsening heart failure. Keeping in mind that a gallon of water weighs a little more than 8 pounds, a sudden change in weight signals the onset of fluid retention which is followed by the development of the congestive symptoms. Anyone, however, who has worked in a hospital knows how unreliable this measurement can be. Variability of the equipment, standardization of the sensor, observer variability and recording of results are all sources of error. To address this problem, recent technology employing impedance devices, radar devices and pressure sensors has been introduced.
Impedance measurement was first attempted using surface electrodes to measure changes in electrical impedance as a reflection of thoracic fluid content. This technique proved too cumbersome to gain widespread acceptance. This concept, however, evolved into the incorporation of impedance measurement between the leads of implanted defibrillators. In this small select group of patients shown to be at risk of sudden death the results were too variable to entertain the widespread implantation of these devices.
Radar devices which use lung dielectric sensing, a more direct measure of lung fluid, are now being developed and introduced. They are non-invasive and can easily provide serial measurements of thoracic fluid content in both the inpatient and outpatient settings. The incorporation of these sensors into the standard monitoring platforms which will be discussed below may offer a significant advantage.
Implantable pressure monitors or MEMS (Micro Electro- Mechanical Sensor) have been gaining traction since being introduced nearly a decade ago. These devices, implanted in the pulmonary artery or left atrium, are designed to detect elevation of back pressure from the heart which is an early sign of decompensation. The MEMS device may provide an earlier warning signal, but there have been no head to head studies to prove superior outcomes. The implanted devices may also be more accurate but come with a set of invasive procedure risks.
So, we have an array of sensors ranging from the scale which costs $20 to the MEMS which costs $20,000. Besides cost and procedural risks, the implants also eliminate patient engagement from the process which may not be a good thing as we consider monitoring in the outpatient and home setting. We must now consider the human factors because technology without acceptance by the end-users always in mind will result in poor results.
The first attempts at home monitoring utilized a telephone land line, a scale, a pad and a pencil. These efforts proved ineffective because of the inaccuracy of the scales, the inability of the patient to accurately record or sometimes even see the numbers on the scale and the delay in transferring the information to a responsible clinician. Undaunted by these problems, we saw the introduction of monitoring platforms which added blood pressure, heart rate, oxygen saturation and a series of subjective questions about how the patient was feeling. Improvements to this system included Bluetooth technology for transfer of measurements from the sensors to the platform and wireless technology for transmission of data from the platform to a monitoring station. These upgrades on the data acquisition end coupled with data monitoring by dedicated nurses who were able to interact with the patients in real time improved the results. Studies have shown that structured telephone support with consistent providers create the best outcomes. The ultimate goal is that patient generated data (PGD) coupled with knowledgeable providers interacting with and educating patients could eventually lead to a state of self management for the patients. A lofty goal, not yet achieved.
An early lesson learned from the Coronary Care Unit experience was that smart alarm systems provided a false sense of security and that an unwatched cardiac monitor benefitted no one. In the world of remote patient monitoring, the same warning applies. Data integration, electronic records, artificial intelligence and improved technology are not enough. To have a fully functional system there must be engagement of physician leaders, administrators, skilled nurses, patients and caregivers. Performance standards must be developed and outcome studies must be performed in order to convince health insurers to provide reimbursement for these services. The available technology can provide actionable data. The challenge now is to engage the stakeholders to take action.