Home Telemonitoring of Community-Dwelling Heart Failure Patients After Home Care Discharge

Background

Heart failure (HF) is a progressive chronic disease and is among the most commonly diagnosed chronic diseases of Medicare beneficiaries, with the risk for older adults doubling every ten years. ^1,2 HF is typically characterized by recurrent periods of clinical exacerbation marked by high rates of emergency department (ED) and inpatient hospital utilization and is often followed by discharge to a certified home healthcare program for monitoring at least one 60-day episode of care, which can be renewed if the patient is considered to be homebound and requiring continued skilled nursing care. ³ Once the home care episode is complete, the patient returns to outpatient care until the next exacerbation and likely hospital admission. This costly, inefficient management cycle leads to exorbitant healthcare costs, as well as poor health outcomes, including decreased quality of life (QoL) and functional status. ^{4 –7}

The Hospital Readmissions Reduction Program of the Affordable Care Act was implemented in 2012 to reduce readmission rates. HF, with a 30-day all-cause (Medicare) rehospitalization rate of 24.7%, was targeted by this program in an effort to control costs and decrease length of stay (LOS). ^6,7 Because this program required Centers for Medicare and Medicaid Services (CMS) to reduce payments to hospitals with excess readmissions, health systems have been actively exploring different models to reduce readmission for HF. ^{8 –11}

One of the most promising approaches to controlling costs and improving QoL is home telemonitoring (HTM). Designed to identify cardiac exacerbative symptoms through physiologic indicator monitoring, the goal is to facilitate improved management through timely treatment adjustments in response to changes in weight, heart rate, lung sounds, and blood pressure, as well as to improve patient/provider communication. ^12,13

Multiple meta-analyses of HTM interventions in HF have concluded that HTM for HF is effective in reducing the risk of all-cause mortality and HF-related hospitalizations; Kitsiou's overview of systematic reviews found that HTM reduces the relative risk of HF-related hospitalizations (0.64–0.86), as well as all-cause mortality (0.60–0.85) compared with usual care. ^14,15 This review also found a great deal of variability in outcomes associated with home telehealth technology, specifically finding that only telemonitoring using mobile or automated devices effectively reduced the risk of HF-related hospitalizations and all-cause mortality, as opposed to computer-generated phone calls or live virtual video visits between patient and provider. ¹⁴

HTM has been found to improve outcomes such as QoL, depression, anxiety, and satisfaction, while all-cause costs and utilization for telemonitored patients have not consistently shown significant improvement. ^{16 –23} Furthermore, multiple comorbidities and poor adherence have been found to have a negative effect on outcomes. ²⁴

The previous work of our research team found that HTM, provided to patients directly upon hospital discharge as part of a comprehensive home healthcare program, did not significantly improve hospitalization rates, time to first readmission, LOS, and costs to Medicare at 30 or 90 days compared to traditional home care. HTM did however result in decreased costs while delivering more patient “touches,” although not significantly. ²⁵

Given the fact that both traditional home care and HTM provide the same “active ingredient”—nursing visits and vital signs monitoring—it may come as no surprise that significant differences in outcomes such as rehospitalization between groups were not found. Indeed, in a systematic review and meta-analysis of transitional care interventions to prevent readmission in HF failure patients, Feltner et al. found that home care programs reduced both all-cause and HF-specific readmission, as well as mortality, over 3–6 months. ²⁶

Unfortunately, multiple meta-analyses have found that while many studies of telemonitoring in home care are conducted immediately after the patient is discharged from the hospital, a large number of studies fail to report whether home care is being delivered. As evidenced by the Feltner study, the effect of home care may serve as a significant confounder to telemonitoring effectiveness, for both intervention and control groups. ²⁶ In the intervention (telemonitored) group, patients may be receiving a confusing assortment of services: two different nurses providing visits and vital signs monitoring: one live (home care) and one virtual (telemonitoring). Conversely, the control group (home care) still receives the active ingredients of telemonitoring: nursing visits and vital signs monitoring.

The purpose of the present study was to attempt to isolate the effects of telemonitoring by enrolling patients in the outpatient setting, after completion of home care, when their condition was no longer being closely followed by a nurse monitor. We hypothesized that HTM would result in improvements in QoL, hospital utilization, and LOS.

Methodology

This study followed 89 Medicare HF patients with a New York Heart Association (NYHA) classification of 1–2, for a period of 6 months immediately following completion of home care. All patients approached had been using telehealth technology as part of their home care protocol. Patients were approached by a telehealth nurse in their homes and offered study participation. If consented to participation, they were randomized to either the HTM group or the standard comprehensive outpatient management (COM) group. This study was IRB approved (#11–006).

The specific aim of this study was to compare the following outcome variables between groups over the 6-month period: ED visits (primary), inpatient utilization (primary), LOS, QoL, and costs.

Three data sources were accessed to obtain reliable utilization data, including: (1) medical records, (2) patient self-report, and (3) health system financial reports. All three sources were available for patients seen within our health system network. For patients hospitalized outside of our network (n = 11%), no financial data were available, and utilization data were solely based upon patient self-report.

Subjects were recruited from a population of patients with HF upon discharge from home care. Specific inclusion criteria were as follows: Medicare coverage, a primary or secondary diagnosis of HF, a NYHA class of 1–2 within the last 6 months, and a patient age of 18 years or older. Subjects were also required to correctly answer a brief research review questionnaire to assess the patient's ability to understand the study process and to perform a variety of telemedicine competency tasks, with or without caregiver assistance.

Patients were randomized by the health system's Biostatistics Department to either the HTM group or COM group. Those randomized to HTM were provided with a telehealth patient station (American TeleCare Incorporated, LifeView Patient Station), which was installed in their home by the health system's Telehealth Installation and Patient Orientation Specialist. HTM patients were asked to participate in a weekly virtual nursing visit and to monitor their symptoms and vital signs on a daily basis, transmitting their vital signs to the telehealth nurse through encrypted transmission. The technology utilized in this study was an FDA-approved computerized monitoring device, which connected the patient's residence, through wireless air card, broadband, or standard telephone line, to a nursing provider station. Electronic peripherals included a video monitor, blood pressure cuff, stethoscope, weight scale, and pulse oximetry monitor.

Telehealth nursing staff monitored patient data on weekdays and conducted a weekly video visit, during which the nurse checked vital signs and listened for any abnormal lung sounds using stethoscope. The stethoscope was either held by the patient or caregiver to locations on the chest and back, as designated by a reference chart. In addition, the nursing staff would discuss adherence with diet and exercise recommendations during the previous week and how they may have affected the patients' vital signs. HTM patients were also asked about ED and inpatient utilization (primary outcome variable), medication adherence, medication changes, and symptoms such as dyspnea, fatigue, and edema during their weekly visit.

Vital signs feedback reports were generated in the primary language of the patient (English or in Spanish), allowing the patient and the telehealth nurse to easily identify out-of-range values. If values for key indicators were outside the normal range, the patient/caregiver was: (1) instructed to remeasure values, (2) contacted by their healthcare practitioner to discuss/revise treatment plan (i.e., diuretic or antihypertensive), (3) asked to perform a telemonitoring visit, or (4) instructed to call 911 for urgent matters. During the weekend/holidays, patients were instructed to report to the ED if symptoms of HF exacerbated.

Patients randomized to COM were followed by their primary cardiologist or HF clinic, at the discretion of the clinician and contacted by the telehealth nurse on a weekly basis to collect self-reported hospital utilization data to maintain a comparable frequency of contact.

Results

In all, 89 patients participated in the study: 32% male, 76% White, 18% Black, 3% Hispanic, and 2% Asian. Average age at enrollment was 81.4 for HTM and 84.9 for COM.

Discussion

It is interesting to note that while HTM patients utilized the ED significantly less often than COM patients (binary analysis), this positive effect of HTM was not observed in the inpatient setting. While this lack of effect in the inpatient setting may simply be a function of small sample size (given the trends in the data), it may also indicate that while HTM can help to reduce ED visits, it may not be powerful enough to avert the hospitalization of patients that are more severely ill.

We hypothesized that QoL, as measured by the MLHFQ, would significantly improve over time for HTM patients—and while both groups showed significant improvement over time, the HTM group improved significantly more than the COM group, indicating a greater benefit.

Given that about one-third of the HTM group had low adherence, we conducted a post hoc analysis to look for differences between high and low adherence subgroups within HTM, and we did see striking differences between high adherent HTM patients versus low adherent groups for both the mean number of hospitalizations and LOS. However, this study did not utilize an “active” control group (i.e., phone management), so there was no way to accurately compare adherence between HTM and COM. We can only note that higher adherence HTM patients had better outcomes than lower adherence patients. Future studies should clearly explore ways to address adherence, so that the intervention itself, and not group assignment, is being accurately measured. ²⁸

Although this study did not measure technology and telehealth staff charges, overall acute care utilization costs over 6 months varied widely, ranging from $0 to $485,762, with a total mean cost of $38,990 (SD = $69,031) and $50,943 (SD = $98,519) for HTM and COM, respectively. Given that current HTM costs have been estimated at roughly $1,600 per year while traditional home care visits are estimated at over $16,000 and average HF-related hospitalizations are estimated at over $14,000 per admission, these savings, if supported by larger studies, represent a substantial reduction in costs for healthcare systems who offer home-based technological support. ^29,30 As technology becomes smarter and smaller (phone-based and wireless), these cost savings will likely become even greater. ³¹

Conclusion

While the data show lower overall utilization and lower costs for the HTM group, several of these group differences did not reach statistical significance. Although telehealth has the potential to help patients and healthcare providers to identify exacerbations and enable patients to participate and understand how to best manage their disease, more research is needed to conclusively determine how to best utilize this technology and manage the unique needs of each patient.