Heart Failure, Chronic Heart Failure
Conditions
Keywords
remote patient monitoring, chronic heart failure, heart failure, non-adherent heart failure
Brief summary
This pragmatic randomized crossover trial looks at the effect(s) of using a remote patient monitoring device (Heartfelt device) with health alerts to monitor the development of peripheral edema in patients with heart failure (HF). The hypothesis is that this passive measurement method will lead to better data availability, which in turn will improve patient care and reduce hospitalizations for the management of worsening HF (HF hospitalizations, HFHs) in nonadherent participants with chronic HF. The study objectives are: * Primary objective: Establish if the Heartfelt device is safe to use and effective at reducing HFHs. * Secondary objectives: 1. Establish the effect of the Heartfelt device on data availability compared to existing remote monitoring devices. 2. Establish the effect of the Heartfelt device on HF clinical outcomes. Participants will need to: * Install the device in their home for at least a year and up to 4 additional years after. * Reply to remote patient monitoring phone calls to follow the care plans.
Detailed description
Participants will be recruited through Remote Patient Monitoring Providers (RPMPs) based in the US, using crossover randomization between standard care and Heartfelt device usage with health alerts sent to RPMPs. The RPMPs will follow a protocolized intervention when responding to raised alerts. The study design is a crossover randomization between: * Standard care (control), 162 days: The device is installed in the home and captures data but no health alerts are sent, and the RPMP does not receive data from the device. (Health alerts are generated and stored for review at the end of the study period to correlate with health issues which occurred during usual care). * Standard care + Heartfelt (intervention), 162 days: The device captures data and transmits volume measurements and health alerts to the RPMP for review. RPMPs follow a pre-specified alert protocol, potentially following up with the patient. At consent, patients get randomized to an install date over a period from 21 to 60 days following consent, subject to scheduling this at least 14 days after the most recent recorded hospital discharge at the time of consent, or the discharge from their current hospitalization if hospitalized at the time of consent. Total study length is 366 days (days 0 to 365). There are two 21-day washout periods to prevent carryover between study arms. The first washout period extends from day 0-20 inclusive, and the second from day 183-203 inclusive At the completion of both crossover arms, patients will be offered the opportunity to keep the device for as long as the study remains active (potentially up to an additional 4 years, for those patients recruited early in the study). During this long-term follow-up, the device will be placed in Intervention mode, with randomized 100-day periods where the device is switched into Standard care mode (control).
Interventions
Device installed in the patient's home and capturing foot volume data which are processed in the cloud.
OTHERStandard care
RPMPs in regular contact with patients and collecting health monitoring data as per their standard operating procedures.
DEVICEHeartfelt-guided care
Volume measurements and health alerts transmitted to the RPMP for review. RPMPs follow a pre-specified alert protocol, potentially following up with the patient.
OTHERQuestionnaires
Patients were presented with one or more optional questionnaires (some validated, some bespoke)
Sponsors
Heartfelt Technologies
Study design
Allocation
RANDOMIZED
Intervention model
CROSSOVER
Primary purpose
PREVENTION
Masking
SINGLE (Outcomes Assessor)
Masking description
Participants cannot be effectively blinded as they receive feedback from the RPMP in the intervention arm, and false alerts could alter the control arm's dynamics. The RPMP cannot be effectively blinded to study arms due to access to additional patient data (such as weighing scale data) to compare to any fake foot volume data that might be provided. The focus of blinding is the Clinical Events Committee (CEC), which will adjudicate if events recorded in the EHR/Claims data count as outcome events. There is no reason for the CEC to have access to logs of contact with patients or data measured with the Heartfelt device. Prior to EHR/Claims data-derived data being provided to the CEC, it will be cleansed of any such potentially-unblinding data. This ensures unbiased outcome assessment, with the CEC not accessing device alert data.
Intervention model description
This trial is a crossover randomized study with two arms: 1. Standard Care (control): Device is installed and data is captured but volume measurements and health alerts are not transmitted to the RPMP; 2. Standard Care + Heartfelt (intervention): Device is installed and data is captured, and volume measurements and health alerts are transmitted to the RPMP for review and potential follow-up (per a pre-specified alert protocol). Optional long-term follow-up where the device will be placed in Intervention mode, with randomized 100-day periods of Standard care mode (control).
Eligibility
Sex/Gender
ALL
Age
22 Years to No maximum
Healthy volunteers
No
Inclusion criteria
* Provision of signed and dated informed consent form (wet or digital signature) * Male or female, aged 22 to \[No maximum age\] * Diagnosed with Chronic Heart failure at least 2 months prior to randomization. * Exhibited peripheral edema on at least one HF-related hospitalization in the last 4 years (as documented in EHR). * Has been hospitalized for HF, received IV diuretics treatment or visited the ER for HF decompensation at least once in the last 6 months or twice in the last 12 months. * Participants who are treated with daily diuretics. * For females of reproductive potential: use of highly effective contraception for at least 1 month prior to screening and agreement to use such a method during study participation and for an additional 6 weeks after the end of the study. * Has failed to collect at least 50% of days over the last 180 of physiological data (incl. home weights and/or BP) from monitoring devices OR has been discontinued from remote patient monitoring due to non-adherence, OR, is considered by clinical team as non-adherent but does not have historical physiological data from monitoring devices. * Patients who are covered by an insurance plan that covers IDE-B costs (e.g. Medicare).
Exclusion criteria
* Participant has bandages to lower limbs every day * Participant has an amputation of both feet * Participant is a regular wheelchair user inside their home * Participant is bed-bound * Participant is of no fixed abode * Participant is taking part in a conflicting evaluation/study that could confound the results of this evaluation and/or impact clinical interventions and participant outcomes * Participant who was unable to have the device installed and activated within 90 days of the randomized installation date * Participant is unable to take diuretics * Participant is on a regular schedule of dialysis * Participant has a history of recurrent deep vein thrombosis (DVT) (two or more episodes within the last 12 months). * Participant has a history of recurrent cellulitis episodes (two or more episodes within the last 12 months). * Participant is prescribed diltiazem or verapamil on an ongoing basis. * Participant is pregnant or is not taking relevant birth control if of child-bearing potential
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Heart Failure Events (HFEs, a composite of HFH or Worsening HF event without hospitalization (as defined in [1-Abraham]) | 12 months, plus up to 48 months long term follow-up | Used to demonstrate whether the device effectively reduces HFHs and urgent HF visits. |
| Doubling in serum creatinine (sCr) from baseline with at least 2.0mg/dL | 12 months, plus up to 48 months long term follow-up | Used to ensure that increased diuretic treatment following alerts from the device does not lead to significant increase in AKI |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Cause of hospitalizations (HF or non-HF) | 12 months, plus up to 48 months long term follow-up | Used to review the lead time of health alerts, or the lack of health alerts generated prior to hospital admissions. |
| Date of increase in HF medications when the patient is home or in the community (not the detailed changes whilst on the ward) | 12 months, plus up to 48 months long term follow-up | This may include oral diuretic, IV diuretics and subcutaneous furosemide injections. Used to allow the determination if an outpatient appointment is HF related or not. |
| Death and cause of death from the clinical record (HF or non-HF as defined in [1-Abraham]) | 12 months, plus up to 48 months long term follow-up | Used to establish if using the device changes HF- or non-HF mortality. |
| Loss of independence (patient unable to live at home; for example, relocation to a care facility for an indefinite duration) | 12 months, plus up to 48 months long term follow-up | Used to establish if there has been an effect on the ability of the patient to live independently in their home. |
| Route of admission for hospitalization events (ICU, CCU, or ward admission),and discharge (home, skilled nursing facility and long term care facility). | 12 months, plus up to 48 months long term follow-up | Used to establish if using the device changes the scheduling and/or setting of care and/or route of admission. |
| Difference in data availability (device generated or contemporaneous self-report) | 12 months, plus up to 48 months long term follow-up | Used to measure the data availability from the Heartfelt device. One of the key findings from feasibility trials and pilot studies in the UK have demonstrated an increased data availability. Does this study in the US confirm the previous finding? |
| The setting of care events (inpatient or outpatient) | 12 months, plus up to 48 months long term follow-up | Used to establish if using the device changes the scheduling and/or setting of care and/or route of admission. |
| Date and time of health alerts generated by the Heartfelt device | 12 months, plus up to 48 months long term follow-up | Used to establish the response time to health alerts. |
| Date, time and method of contact made following reception of a health alert generated by the Heartfelt device | 12 months, plus up to 48 months long term follow-up | Used to establish the response time to health alerts. |
| Date and reason for study withdrawal | 12 months, plus up to 48 months long term follow-up | Used alongside the analysis of other right-censoring endpoints. |
| Scheduling of care events (scheduled or unscheduled) | 12 months, plus up to 48 months long term follow-up | Used to establish if using the device changes the scheduling and/or setting of care and/or route of admission. |
| The date and length (in hours) of hospital admission and re-admissions | 12 months, plus up to 48 months long term follow-up | Used to review the lead time of health alerts, or the lack of health alerts generated prior to hospital admissions. |
Other
| Measure | Time frame | Description |
|---|---|---|
| Optimal medical therapy score (if available) | 12 months, plus up to 48 months long term follow-up | Used to get an indication of medication adherence. |
| Presence of edema reported by clinicians during the clinical encounter (including grading of edema if available) | 12 months, plus up to 48 months long term follow-up | Used to validate the device's determination of edema status. |
| Data availability (device generated or contemporaneous self-report) in the 6 months prior to screening. | Retrospectively, over the 6 months prior to screening | Used to analyze the other endpoints based on this subgroup. |
| Device-related complications | 12 months, plus up to 48 months long term follow-up | Used to ensure that any device-related complications that occur after the main 12 months of the study are captured. |
| Date, time, reason and means of contact made with patients by RPMPs not due to Heartfelt health alerts (e.g. phone triage/counseling, repeat labs, etc) | 12 months, plus up to 48 months long term follow-up | Used to determine if any significant changes in outcomes might be due to increases in patient contacts overall. |
| NT Pro-BNP or BNP, with date if available in patient records | 12 months, plus up to 48 months long term follow-up | Used to correlate NT Pro-BNP health alert status. |
| Alcohol consumption | 12 months, plus up to 48 months long term follow-up | Used to determine if any significant changes in outcomes might be due to lifestyle choices and to analyze the other endpoints based on this subgroup. |
| Cannabis consumption | 12 months, plus up to 48 months long term follow-up | Used to determine if any significant changes in outcomes might be due to lifestyle choices and to analyze the other endpoints based on this subgroup. |
| Other recreational drug(s) consumption | 12 months, plus up to 48 months long term follow-up | Used to determine if any significant changes in outcomes might be due to lifestyle choices and to analyze the other endpoints based on this subgroup. |
| Cause of hospitalizations (ICD-10-CM codes) | 12 months, plus up to 48 months long term follow-up | Used to determine if a correlation could be identified between treatment effect and type of hospital stays; as well as sensitivity and specificity of the alert analysis. |
| Tobacco consumption | 12 months, plus up to 48 months long term follow-up | Used to determine if any significant changes in outcomes might be due to lifestyle choices and to analyze the other endpoints based on this subgroup. |
| Date and time of health alerts from other monitoring devices | 12 months, plus up to 48 months long term follow-up | Used to correlate health alerts from other medical devices and the Heartfelt device. |
| Symptoms diary (if available) | 12 months, plus up to 48 months long term follow-up | Used to establish patient usage of symptom diaries. |
| Questionnaire responses (if available) | 12 months, plus up to 48 months long term follow-up | Used to establish patient acceptability of the device. |
| Dates, types and length of stays in rehab facilities/nursing home/care homes | 12 months, plus up to 48 months long term follow-up | Recorded in hours if \<48h and in days if \>48h. Proxy used to identify trends in worsening of patient's health/independence/quality of life following participation in various arms of the study. |
| Dates and times of caregivers' shifts if home healthcare used | 12 months, plus up to 48 months long term follow-up | Proxy used to identify trends in worsening of patient's health/independence/quality of life following participation in various arms of the study. |
| HFHs in the 12 months prior to installation (as defined in [1-Abraham]) | Retrospectively, over the 12 months prior to device installation | Used to compare HF Hospitalization date before the study as a baseline period and during the study. |
| UHFV and other urgent healthcare usage in the 12 months prior to installation | Retrospectively, over the 12 months prior to device installation | Used to compare Heart Failure Visit numbers before the study as a baseline period and during the study. |
| Date and cause of Death | 12 months | Used to supplement data on HF-related death that is collected as a primary endpoint from clinical records of participants, as well as allowing the comparison to non-participants. |
| Ethnicity | At enrollment | Used to analyze the other endpoints based on this subgroup. |
| NYHA class | At enrollment | Used to analyze the other endpoints based on this subgroup. |
| Ejection fraction | At enrollment | Used to analyze the other endpoints based on this subgroup. |
| Comorbidities | At enrollment | Used to analyze the other endpoints based on this subgroup. |
| Date, prescription fulfillment and any modifications to dosage of HF medications (if available) | 12 months, plus up to 48 months long term follow-up | Used to get an indication of medication adherence. |
Countries
United States
Contacts
Primary ContactWH Wilson Tang, MD
Backup ContactAmparito Cunningham, MD. MPH.
Outcome results
None listed