Infusing Needed Iron to Target Insufficiency in Adults Treated for Evidence of Heart Failure

The INITIATE-HF study will be managed at the Kaiser Permanente Northern California (KPNC) Division of Research and will include participants from KPNC medical centers. Up to 21 medical centers will be randomly assigned into two groups, with balanced populations between the groups based on demographic and heart failure-related characteristics. In medical centers assigned to the intervention group, physicians of eligible patients with heart failure and iron deficiency (i.e., TSAT \<20%) will be sent information from current KPNC and national clinical practice guidelines about the use of IV iron therapy in the setting of heart failure. In medical centers assigned to the control group, physicians will not be contacted about their patients with iron deficiency or receive information about current guidelines about the use of IV iron (i.e., usual care). If a patient with heart failure is admitted to a medical center assigned to the intervention group and has iron deficiency, their treating doctor will be notified about their patient's iron stores lab results, information from KPNC and national guidelines about the use of IV iron therapy, and details about the available IV iron formulations and recommended dosages for their consideration. The study team will not have any direct contact with patients at any of the medical centers assigned to either the intervention or control group and will not be involved in any clinical care decision-making for any patients, as that will reside with each patient's responsible physician and care team.

The intention-to-treat (ITT) population will consist of patients meeting all eligibility criteria at both intervention and control sites. The per-protocol population will include data from all patients at intervention sites. All primary and secondary endpoints will be analyzed using mixed effects Cox proportional hazards regression for point and interval estimation of the intervention effect on the primary outcome and on each of the secondary outcomes. Similarly, mixed effects Poisson regression will be additionally performed to analyze the effect of the intervention in relation to secondary outcomes when characterized as counts: all-cause and heart failure-specific outpatient encounters (i.e., clinic and telehealth visits), emergency department visits, and hospitalizations.

Interaction analyses will be performed to assess for potential heterogeneity of treatment effects based on the following prespecified subgroups: age, sex, race/ethnicity, presumed etiology of heart failure (i.e., ischemic vs. non-ischemic), duration of diagnosed heart failure (i.e., \<1 vs. \>1 year), reason for hospital admission (i.e., primary vs. secondary discharge diagnosis for heart failure), left ventricular ejection fraction category (\<40% vs. 41-49%), B-type natriuretic peptide (BNP) level (i.e., \< vs. \> the median value), TSAT (i.e., \<15% vs. 15-19%), estimated glomerular filtration rate (eGFR) level (i.e., \< vs. \> the median), hemoglobin level (i.e., \< vs. \> the median), anemia status, and KPNC service area.

The primary analysis will focus on the implementation outcome of the administration of IV iron during the index hospitalization to determine how the provider-facing notification influences provider behavior and care processes. The secondary analysis will evaluate effectiveness outcomes, including the composite of all-cause death and heart failure-related hospitalization, to determine whether any observed changes in the implementation process outcome also translate into improved patient clinical outcomes during the 12 months after randomization.

• % of patients who experience the composite outcome of all-cause death and heart failure-related hospitalization [Time frame: From randomization to the end of 12 months after randomization]