Discussion
Telehealth yielded clinically important interventions, with some titrations bringing patients to their target doses of medications for HFrEF. The 20 interventions made in the HT group can be largely attributed to the nurses’ efforts to alert clinicians to drug titrations or ACEi/ARB to ARNi transitions. Although the findings were not statistically significant, the difference in the number of drug therapy changes supports the use of the HT program for a GDMT optimization strategy. Patients may be difficult to titrate secondary to adverse effects that make medication initiation or titration inappropriate, such as hypotension and hyperkalemia, although this was not observed in this small sample size. Considering a mean HT enrollment of 5.3 months, many patients had adequate disease assessment and medication titration. Given that patients are discharged from the service once deemed appropriate, this decreases the burden on the patient and increases the utility and implementation of the HT program for other patients.
A surprising finding of this study was the lower rate of HF-related hospitalizations in the HT group. Although not the primary subject of interest in the study, it reinforced the importance of close health care professional follow-up for patients living with HF. Telehealth may improve communication and shared decision making over medication use. Because the finding was unanticipated, the rate of diuretic adjustments was not tracked.
Patients were reevaluated every 6 months for willingness to continue the program, adherence, and clinical needs. These results are similar to those of other trials that demonstrated improved rates of GDMT in the setting of pharmacist- or nurse-led HF treatment optimization.15,16 These positive results differ from other trials incorporating remote monitoring regarding patient continuation in HT programs. For example, in a study by Ding and colleagues, the withdrawal rate from their monitoring service was about 22%, while in our study only 1 patient withdrew from the HT program.11
The HT program resulted in fewer hospitalizations than the control arm. There were 6 HF-related hospitalizations in the control group, although 5 involved a single patient. Typically, such a patient would be encouraged to follow HT monitoring after just 1 HF-related hospitalization; however, the patient declined to participate.
Early optimization of GDMT in patients who were recently discharged from the hospital for an HF-related hospitalization yields a reduction in hospital rehospitalization.17 GDMT optimization has unequivocal benefits in HF outcomes. Unfortunately, the issues surrounding methodologies on how to best optimize GDMT are lacking. While HT has been found to be feasible in the aid of optimizing medical therapy, the TIM-HF trial concluded that remote monitoring services had no significant benefit in reducing mortality.7,8 On the other hand, the OptiLink HF Study showed that when clinicians respond to remote monitoring prompts from fluid index threshold crossing alerts, these interventions are associated with significantly improved clinical outcomes in patients with implantable cardioverter-defibrillators and advanced HF.9 In contrast to previous trials, the AMULET trial showed that remote monitoring compared with standard care significantly reduced the risk of HF hospitalization or cardiovascular death during the 12-month follow-up among patients with HF and LVEF ≤ 49% after an episode of acute exacerbation.10 Additionally, patients who received skilled home health services and participated in remote monitoring for their chronic HF experienced a reduction in all-cause 30-day readmission.18
Given the contrasting evidence regarding remote monitoring and variable modalities of implementing interventions, we investigated whether HT monitoring yields improvements in GDMT optimization. We found that HT nurses were able to nearly double the rate of interventions for patients with HFrEF. The HT program in providing expanded services will require adequate staffing responsibilities and support. The HT program is geared toward following a large, diverse patient population, such as those with chronic obstructive pulmonary disease, hypertension, and HF. We only evaluated services for patients with HFrEF, but the program also follows patients with HfmrEF and HfpEF. These patients were not included as GDMT optimization differs for patients with an LVEF > 40%.19,20
The lower rates of achieving target doses of GDMTs were likely obstructed by continuous use of initial drug doses and further limited by lack of follow-up. When compared with the rest of the VAAAHS, there was a greater effort to increase ARNi use in the HT group as 7 of 33 patients (21%) were started on ARNi compared with a background increase of ARNi use of 17%. There was a lower mortality rate observed in the HT group compared with the control group. One patient in each group died of unrelated causes, while 2 deaths in the control group were due to worsening HF. The difference in mortality is likely multifactorial, possibly related to the control group’s greater disease burden or higher mean age (75.2 years vs 72.6 years).
Limitations
This was an observational cohort design, which is subject to bias. Thus, the findings of this study are entirely hypothesis-generating and a randomized controlled trial would be necessary for clearer results. Second, low numbers of participants may have skewed the data set. Given the observational nature of the study, this nonetheless is a positive finding to support the HT program for assisting with HF monitoring and prompting drug interventions. Due to the low number of participants, a single patient may have skewed the results with 5 hospitalizations.
Conclusions
This pilot study demonstrates the applicability of HT monitoring to optimize veteran HFrEF GDMT. The HT program yielded numerically relevant interventions and fewer HF-related hospitalizations compared with the control arm. The study shows the feasibility of the program to safely optimize GDMT toward their target doses and may serve as an additional catalyst to further develop HT programs specifically targeted toward HF monitoring and management. Cost-savings analyses would likely need to demonstrate the cost utility of such a service.
Acknowledgments
We thank the home telehealth nursing staff for their assistance in data collection and enrollment of patients into the monitoring program.