Heart failure (HF) is a leading cause of preventable hospitalization and death in the VA and many patients fall short of self-care goals. Numerous efficacy trials have shown that HF care management supported by health information technology (i.e., HITCM) can improve patients' outcomes, although VA care managers in 'real-world' health systems are often overwhelmed by the need to provide monitoring and behavior change services. Informal caregivers (ICGs) may help to fill the gaps in VA care management and enhance support for patients' treatment adherence, behavior changes, and symptom monitoring. The challenge is to identify ways to leverage assistance from ICGs who lack the resources to fill this role effectively.
We evaluated the impact of extending the reach of VA care management by incorporating a protocol-driven model for improved monitoring and self-management support by a CarePartner (CP). CPs were adult children or friends living outside the Veteran's home who were willing to play a structured role to support self-care. The specific aims of the trial were: 1) to determine whether an intervention that uses automated patient monitoring and behavior change calls with follow-up to HF patients' care manager and CP (HITCM+CP) improves key patient-centered outcomes relative to a system that only uses the same technology to support patients' care management (HITCM-only). Outcomes of interest included patients' health-related quality of life, mental health, health service use, and mortality risk; 2) to evaluate the impact of HITCM+CP on patients' self-care behaviors compared to HITCM-only; and 3) to determine whether the intervention increases the quality and quantity of support for HF patients' self-care compared to HITCM-only.
A total of 372 HF patient-CP pairs were recruited from the Louis B. Stokes VA Healthcare System and 13 affiliated community based outpatient clinics or CBOCs in Cleveland, Ohio. Patients with a diagnosis of heart failure and at least one outpatient visit within the previous 12 months were identified as potentially eligible. Half (N=189) were randomized to the intervention group, and half (N=183) were randomized to the control group. Patients received weekly automated telephone assessment and behavior change calls for 12 months. The weekly patient calls included recorded information and questions that the patient answered using a telephone keypad. During each automated assessment, patients were asked about their overall health status, heart failure symptoms, self-management behaviors (such as weighing themselves and monitoring their salt and fluid intake), and medication availability and adherence. For patients in both study arms, a care manager monitored patients' assessment results via a secure website and received automated, structured fax reports concerning urgent health problems. In the HITCM+CP group, patients' CPs also received structured and tailored e-mail reports based on the patient's automated calls. The feedback included information about what the patient reported during their most recent assessment call along with suggestions for how the CarePartner could help their patient-partner with their HF-related self-care. For example, if a patient reported that s/he was not taking their heart failure medicine exactly as prescribed, the patient's CarePartner received an email report letting them know what problem the patient reported, why it was important, and how the CarePartner could help. CarePartners used a structured protocol to review the patient's assessment results, identify self-care goals and barriers, and ensure that the patient's in-home caregivers and healthcare team remained involved. All patients and CPs completed quantitative surveys at baseline and were surveyed again at 6 and 12 months. The study included a mixed-methods approach (including quantitative surveys plus qualitative interviews with patients, CPs, and clinicians) to evaluate intervention use and the service's potential for translation. The primary outcome was HF-related quality of life at 12 months. Secondary outcomes included self-care behavior, patient-CP relationship indicators, hospitalization, and death.
A total of 372 patients were enrolled. Participants' average age was 68 years, 99% were male, 51% had some college education, 32% had an annual income less than $15,000, and 84% completed follow-up. Study participants completed 89% of their weekly automated calls, with assessment call completion rates that were similar across groups defined by baseline sociodemographic and clinical characteristics. Overall satisfaction with the automated calling system was high. Preliminary results indicate that participants in both groups receiving automated assessment and self-care support calls improved from baseline to the 6-month and 12-month follow-up on most survey measures of the patient's functioning, perceived health status, and HF self-care. On average both physical and mental composite summary measures of the SF-12 improved from baseline to follow-up in both groups, and participants' reported both improved heart failure-related quality of life and less depression symptoms at follow-up compared to baseline. Confidence in managing heart failure as measured by the Revised Heart Failure Self-care Behavior Scale (BEES-HF) scale also improved in both groups, and participants reported improved CarePartner relationships. Qualitative analyses suggest that many patients and CPs felt closer and more informed as a result of being in the study. Analyses of intervention control differences are ongoing but suggest that intervention engagement was greater when patients' baseline caregiver relationships were stronger. To date, we have not identified statistically significant overall differences in intervention versus control-group physical functioning, mental functioning, HF self-management, or health service use at six or 12 months. Subgroup analyses are continuing with support from our Center and will be reported in mid-2013.
We found that it is technically feasible to deliver automated telephone monitoring and behavior change calls weekly to heart failure patients in a major VA Healthcare System. We found that patients completed a high percentage (89%) of their weekly telephone assessments (even after 12 months of automated calls) and reported very high satisfaction with the system and the CarePartner program in general. Patients in both groups receiving the automated intervention improved over time on multiple patient-centered outcomes; preliminary analyses suggest no differential intervention impact associated with HITCM+CP. We found that structured automated calls are a feasible way to increase informal support for heart failure patients' self-care management and may serve as an effective enhancement to scarce clinical resources.
- Piette JD, Rosland AM, Marinec NS, Striplin D, Bernstein SJ, Silveira MJ. Engagement with automated patient monitoring and self-management support calls: experience with a thousand chronically ill patients. Medical care. 2013 Mar 1; 51(3):216-23.