Heart Failure With Reduced Ejection Fraction
Conditions
Brief summary
A randomized controlled trial to compare the efficacy of an electronic health record-based alert informing providers about evidence-based medications for HFrEF versus usual care (no alert) in outpatient clinics across a single health system.
Detailed description
Data from clinical trials suggest that pharmacological therapies prescribed at appropriate doses will lead to dramatic improvements in survival and hospitalization rates in patients with heart failure with reduced ejection fraction (HFrEF). Consequently, major cardiovascular societies assign the highest level of recommendation to use these therapies in all eligible patients. However, data from several registries over the last three decades has failed to see use of these evidence based therapies at levels noted in clinical trials, despite aggressive guideline recommendations and promotion by thought leaders in the field. It remains unclear as to why many patients with HFrEF are not on evidence-based therapies, and why the percentages are consistent across national registries over time. One explanation might be that providers know the data regarding evidence-based therapies, but the therapies only benefit a narrow population. Another factor might be a lack of knowledge among providers about the appropriate management of HFrEF patients. A simple way to test this hypothesis is to examine whether electronic health record (EHR) based best practice advisories (BPAs) can increase use of evidence based therapies. If found to be effective, these low cost interventions can be rapidly applied across large healthcare systems. This study will conduct a randomized controlled trial across outpatient clinics within a single health system comparing the effectiveness of an EHR-based alerting system that informs practitioners about what evidence-based medications they can prescribe for HFrEF patients versus usual care (no alert). One hundred eligible unique providers will be randomized to an intervention in which an alert will appear for all eligible patients with HFrEF, or to a control group in which no alert appears and usual care will continue, with a target patient enrollment of 1,310. The primary outcome for the trial will be the proportion of patients with HFrEF with an increase in evidence based medical therapies for HFrEF (beta-blockers, ACE-I/ARB/ARNI, MRA, SGLT2i). Secondary outcomes will include 30-day hospital admission rates, 30-day ED visits, one year all-cause mortality, and total 6 month healthcare costs.
Interventions
OTHERBest practice alert for the notification of patient HFrEF and recommended evidence-based therapies (NO drugs are being administered in this trial)
Providers will receive a best practice alert for each of their eligible patients upon opening of the order entry screen in the patient's medical record. The alert will inform the provider to the presence of HFrEF and of the patient's current left ventricular ejection fraction and current evidence-based medications for HFrEF. It will also provide access to an order set with recommended evidence-based HFrEF therapies as well as a link to the best available guideline-recommended information regarding the treatment of heart failure.
Sponsors
AstraZeneca
Yale University
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
SUPPORTIVE_CARE
Masking
TRIPLE (Subject, Investigator, Outcomes Assessor)
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
No
Inclusion criteria
Patient Inclusion Criteria: * Age 18 or over * Seen in eligible internal medicine or cardiology clinic * Left ventricular ejection fraction less than or equal to 40% * Registered in the Yale Heart Failure Registry Patient
Exclusion criteria
* Opted out of EHR-based research * Under hospice care * Already receiving each targeted class of evidence-based HFrEF medical therapy Selection of Providers: * Practicing at an eligible internal medicine or cardiology clinic * High frequency of visits by eligible patients based on retrospective chart review
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Percentage of Patients With Heart Failure With Reduced Ejection Fraction (HFrEF) With an Increase in Prescribed HFrEF Therapy | Assessed from the date of randomization to 30 days post-randomization | Assessed as an increase in the number of prescribed targeted evidence-based therapies for HFrEF, including beta-blockers, ACEi, ARBs, ARNIs, MRAs, and SGLT2is. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Rate of One-year All-cause Mortality | Assessed from the date of randomization to the date of death from any cause, up to 365 days post-randomization | Assessed as the number of patients who expired from randomization up to one year from any cause. |
| Percentage of Patients on Beta Blockers | Assessed from the date of randomization to 30 days post-randomization | Assessed as the number of patients with prescribed beta blockers |
| Percentage of Patients on ACE Inhibitors | Assessed from the date of randomization to 30 days post-randomization | Assessed as the number of patients with a prescribed ACEi |
| Percentage of Patients on ARBs | Assessed from the date of randomization to 30 days post-randomization | Assessed as the number of patients with a prescribed ARB |
| Percentage of Patients on ARNIs | Assessed from the date of randomization to 30 days post-randomization | Assessed as the number of patients with a prescribed ARNI |
| Percentage of Patients on MRAs | Assessed from the date of randomization to 30 days post-randomization | Assessed as the number of patients with a prescribed MRA |
| Percentage of Patients on SGLT2 Inhibitors | Assessed from the date of randomization to 30 days post-randomization | Assessed as the number of patients with a prescribed SGLT2i |
| Rate of 30-day All-cause Emergency Department Visits | Assessed from the date of randomization to the date of ED/ER admission, up to 30 days post-randomization | Number of participants with an emergency department visit within 30 days of randomization. |
| Rate of 30-day Hospital Admission | Assessed from the date of randomization to the date of hospital admission, up to 30 days post-randomization | Number of participants with a hospitalization within 30 days of randomization |
| Percentage of Filled Prescriptions | Assessed 6 months post-randomization | Proportion of prescriptions filled as assessed by SureScripts |
| Medication Dose of Any Prescribed Beta Blocker | Assessed at 6 months post-randomization | — |
| Medication Dose of Any Prescribed ACEi | Assessed at 6 months post-randomization | — |
| Medication Dose of Any Prescribed ARB | Assessed at 6 months post-randomization | — |
| Medication Dose of Any Prescribed ARNI | Assessed at 6 months post-randomization | — |
| Medication Dose of Any Prescribed MRA | Assessed at 6 months post-randomization | — |
| Medication Dose of Any Prescribed SGLT2 Inhibitor | Assessed at 6 months post-randomization | — |
| Total Six-month Healthcare Costs | Assessed from the date of randomization to 6 months post-randomization | Assessed as total direct and indirect cost of patient care from randomization to 6 months post randomization. |
Countries
United States
Participant flow
Recruitment details
100 providers were identified and consented to the study. 93 were randomized to the intervention group (alert) (n=46) or to usual care (no alert) group (n=47). This created 93 clusters (providers) to which eligible patient participants were assigned on their first eligible clinic visit. Eligible patient subjects were automatically identified, enrolled and randomized by our best practice alert based on their provider's randomization status (685 to the intervention group and 625 to usual care).
Pre-assignment details
7 provider subjects were excluded prior to randomization as patient recruitment finished before these providers saw an eligible patient.
Participants by arm
| Arm | Count |
|---|---|
| EHR-based Alert Providers will receive a best practice alert for each of their eligible patients upon opening of the order entry screen in the patient's medical record. The alert will inform the provider to the presence of HFrEF and of the patient's current left ventricular ejection fraction and current evidence-based medications for HFrEF. It will also provide access to an order set with recommended evidence-based HFrEF therapies as well as a link to the best available guideline-recommended information regarding the treatment of heart failure.
Best practice alert for the notification of patient HFrEF and recommended evidence-based therapies (NO drugs are being administered in this trial): Providers will receive a best practice alert for each of their eligible patients upon opening of the order entry screen in the patient's medical record. The alert will inform the provider to the presence of HFrEF and of the patient's current left ventricular ejection fraction and current evidence-based medications for HFrEF. It will also provide access to an order set with recommended evidence-based HFrEF therapies as well as a link to the best available guideline-recommended information regarding the treatment of heart failure. | 685 |
| Usual Care Providers will not receive an alert and will proceed with usual care. | 625 |
| Total | 1,310 |
Baseline characteristics
| Characteristic | Usual Care | EHR-based Alert | Total |
|---|---|---|---|
| Age, Continuous | 72 years | 72 years | 72 years |
| Baseline Blood Pressure Diastolic blood pressure | 70 mmHg | 70 mmHg | 70 mmHg |
| Baseline Blood Pressure Systolic blood pressure | 118 mmHg | 120 mmHg | 120 mmHg |
| Baseline Creatinine | 1.2 mg/dL | 1.2 mg/dL | 1.2 mg/dL |
| Baseline estimated Glomerular Filtration Rate (eGFR) | 58.4 ml/min/1.73m^2 | 56.9 ml/min/1.73m^2 | 57.8 ml/min/1.73m^2 |
| Baseline Heart Rate | 74 beats/minute | 74 beats/minute | 74 beats/minute |
| Baseline Left Ventricular Ejection Fraction (LVEF) | 32 percent ejection fraction | 32 percent ejection fraction | 32 percent ejection fraction |
| Baseline Potassium | 4.3 mEq/L | 4.2 mEq/L | 4.3 mEq/L |
| Ethnicity (NIH/OMB) Hispanic or Latino | 58 Participants | 67 Participants | 125 Participants |
| Ethnicity (NIH/OMB) Not Hispanic or Latino | 559 Participants | 614 Participants | 1173 Participants |
| Ethnicity (NIH/OMB) Unknown or Not Reported | 8 Participants | 4 Participants | 12 Participants |
| Medicaid/Medicare Enrollee | 582 Participants | 535 Participants | 1117 Participants |
| Medical History Atrial Arrhythmias | 342 Participants | 349 Participants | 691 Participants |
| Medical History Chronic Kidney Disease (CKD) | 212 Participants | 241 Participants | 453 Participants |
| Medical History Chronic Pulmonary Disorders | 199 Participants | 202 Participants | 401 Participants |
| Medical History Coronary Artery Disease (CAD) | 208 Participants | 262 Participants | 470 Participants |
| Medical History Depression | 99 Participants | 124 Participants | 223 Participants |
| Medical History Diabetes | 235 Participants | 260 Participants | 495 Participants |
| Medical History Dyslipidemia | 500 Participants | 524 Participants | 1024 Participants |
| Medical History Hypertension | 531 Participants | 583 Participants | 1114 Participants |
| Medical History Obesity | 130 Participants | 145 Participants | 275 Participants |
| Medical History Peripheral Vascular Disease | 392 Participants | 386 Participants | 778 Participants |
| Number of patients on GDMT classes at baseline ACE-I/ARB/ARNI | 437 Participants | 489 Participants | 926 Participants |
| Number of patients on GDMT classes at baseline Aldosterone Receptor Antagonist (MRA) | 186 Participants | 193 Participants | 379 Participants |
| Number of patients on GDMT classes at baseline Angiotensin Converting Enzyme Inhibitor (ACE-I) | 96 Participants | 93 Participants | 189 Participants |
| Number of patients on GDMT classes at baseline Angiotensin Receptor Blocker (ARB) | 91 Participants | 102 Participants | 193 Participants |
| Number of patients on GDMT classes at baseline Angiotensin Receptor/Neprilysin Inhibitor (ARNI) | 250 Participants | 294 Participants | 544 Participants |
| Number of patients on GDMT classes at baseline Beta blockers | 517 Participants | 581 Participants | 1098 Participants |
| Number of patients on GDMT classes at baseline Sodium-Glucose cotransporter-2 inhibitor (SGLT2i) | 62 Participants | 83 Participants | 145 Participants |
| Patients per provider type Number of patients seen by Advanced Practice Providers | 187 Participants | 218 Participants | 405 Participants |
| Patients per provider type Number of patients seen by physician | 438 Participants | 467 Participants | 905 Participants |
| Race (NIH/OMB) American Indian or Alaska Native | 1 Participants | 1 Participants | 2 Participants |
| Race (NIH/OMB) Asian | 11 Participants | 9 Participants | 20 Participants |
| Race (NIH/OMB) Black or African American | 118 Participants | 119 Participants | 237 Participants |
| Race (NIH/OMB) More than one race | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) Native Hawaiian or Other Pacific Islander | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) Unknown or Not Reported | 48 Participants | 64 Participants | 112 Participants |
| Race (NIH/OMB) White | 447 Participants | 492 Participants | 939 Participants |
| Sex: Female, Male Female | 195 Participants | 207 Participants | 402 Participants |
| Sex: Female, Male Male | 430 Participants | 478 Participants | 908 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk |
|---|---|---|
| deaths Total, all-cause mortality | 75 / 685 | 68 / 625 |
| other Total, other adverse events | 0 / 685 | 0 / 625 |
| serious Total, serious adverse events | 0 / 685 | 0 / 625 |
Outcome results
Primary
Percentage of Patients With Heart Failure With Reduced Ejection Fraction (HFrEF) With an Increase in Prescribed HFrEF Therapy
Assessed as an increase in the number of prescribed targeted evidence-based therapies for HFrEF, including beta-blockers, ACEi, ARBs, ARNIs, MRAs, and SGLT2is.
Time frame: Assessed from the date of randomization to 30 days post-randomization
Population: Observed in patient subjects only.
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| EHR-based Alert | Percentage of Patients With Heart Failure With Reduced Ejection Fraction (HFrEF) With an Increase in Prescribed HFrEF Therapy | 176 Participants |
| Usual Care | Percentage of Patients With Heart Failure With Reduced Ejection Fraction (HFrEF) With an Increase in Prescribed HFrEF Therapy | 117 Participants |
Secondary
Medication Dose of Any Prescribed ACEi
Time frame: Assessed at 6 months post-randomization
Population: Surescripts data is inaccurate and does not capture all patient prescription fills, thus this data cannot currently be obtained.
Secondary
Medication Dose of Any Prescribed ARB
Time frame: Assessed at 6 months post-randomization
Population: Surescripts data is inaccurate and does not capture all patient prescription fills, thus this data cannot currently be obtained.
Secondary
Medication Dose of Any Prescribed ARNI
Time frame: Assessed at 6 months post-randomization
Population: Surescripts data is inaccurate and does not capture all patient prescription fills, thus this data cannot currently be obtained.
Secondary
Medication Dose of Any Prescribed Beta Blocker
Time frame: Assessed at 6 months post-randomization
Population: Surescripts data is inaccurate and does not capture all patient prescription fills, thus this data cannot currently be obtained.
Secondary
Medication Dose of Any Prescribed MRA
Time frame: Assessed at 6 months post-randomization
Population: Surescripts data is inaccurate and does not capture all patient prescription fills, thus this data cannot currently be obtained.
Secondary
Medication Dose of Any Prescribed SGLT2 Inhibitor
Time frame: Assessed at 6 months post-randomization
Population: Surescripts data is inaccurate and does not capture all patient prescription fills, thus this data cannot currently be obtained.
Secondary
Percentage of Filled Prescriptions
Proportion of prescriptions filled as assessed by SureScripts
Time frame: Assessed 6 months post-randomization
Population: Surescripts data is inaccurate and does not capture all patient prescription fills, thus this data cannot currently be obtained.
Secondary
Percentage of Patients on ACE Inhibitors
Assessed as the number of patients with a prescribed ACEi
Time frame: Assessed from the date of randomization to 30 days post-randomization
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| EHR-based Alert | Percentage of Patients on ACE Inhibitors | 83 Participants |
| Usual Care | Percentage of Patients on ACE Inhibitors | 92 Participants |
Secondary
Percentage of Patients on ARBs
Assessed as the number of patients with a prescribed ARB
Time frame: Assessed from the date of randomization to 30 days post-randomization
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| EHR-based Alert | Percentage of Patients on ARBs | 111 Participants |
| Usual Care | Percentage of Patients on ARBs | 85 Participants |
Secondary
Percentage of Patients on ARNIs
Assessed as the number of patients with a prescribed ARNI
Time frame: Assessed from the date of randomization to 30 days post-randomization
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| EHR-based Alert | Percentage of Patients on ARNIs | 334 Participants |
| Usual Care | Percentage of Patients on ARNIs | 285 Participants |
Secondary
Percentage of Patients on Beta Blockers
Assessed as the number of patients with prescribed beta blockers
Time frame: Assessed from the date of randomization to 30 days post-randomization
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| EHR-based Alert | Percentage of Patients on Beta Blockers | 606 Participants |
| Usual Care | Percentage of Patients on Beta Blockers | 521 Participants |
Secondary
Percentage of Patients on MRAs
Assessed as the number of patients with a prescribed MRA
Time frame: Assessed from the date of randomization to 30 days post-randomization
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| EHR-based Alert | Percentage of Patients on MRAs | 232 Participants |
| Usual Care | Percentage of Patients on MRAs | 203 Participants |
Secondary
Percentage of Patients on SGLT2 Inhibitors
Assessed as the number of patients with a prescribed SGLT2i
Time frame: Assessed from the date of randomization to 30 days post-randomization
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| EHR-based Alert | Percentage of Patients on SGLT2 Inhibitors | 144 Participants |
| Usual Care | Percentage of Patients on SGLT2 Inhibitors | 103 Participants |
Secondary
Rate of 30-day All-cause Emergency Department Visits
Number of participants with an emergency department visit within 30 days of randomization.
Time frame: Assessed from the date of randomization to the date of ED/ER admission, up to 30 days post-randomization
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| EHR-based Alert | Rate of 30-day All-cause Emergency Department Visits | 86 Participants |
| Usual Care | Rate of 30-day All-cause Emergency Department Visits | 77 Participants |
Secondary
Rate of 30-day Hospital Admission
Number of participants with a hospitalization within 30 days of randomization
Time frame: Assessed from the date of randomization to the date of hospital admission, up to 30 days post-randomization
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| EHR-based Alert | Rate of 30-day Hospital Admission | 47 Participants |
| Usual Care | Rate of 30-day Hospital Admission | 63 Participants |
Secondary
Rate of One-year All-cause Mortality
Assessed as the number of patients who expired from randomization up to one year from any cause.
Time frame: Assessed from the date of randomization to the date of death from any cause, up to 365 days post-randomization
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| EHR-based Alert | Rate of One-year All-cause Mortality | 75 Participants |
| Usual Care | Rate of One-year All-cause Mortality | 68 Participants |
Secondary
Total Six-month Healthcare Costs
Assessed as total direct and indirect cost of patient care from randomization to 6 months post randomization.
Time frame: Assessed from the date of randomization to 6 months post-randomization
Population: The data necessary to analyze this outcome was unable to be released from the university analytics team, therefore we cannot report cost results. The Finance Administration at Yale did not grant approval for access to this data. The data was not collected and will not be accessed in the future.