The evidence base of what works in chronic care management programs is underdeveloped. To fill the gap, we pooled and reanalyzed data from ten randomized clinical trials of heart failure care management programs to discern how program delivery methods contribute to patient outcomes. We found that patients enrolled in programs using multi-disciplinary teams and in programs using in-person communication had significantly fewer hospital readmissions and readmission days than routine care patients had. Our study offers policymakers and health plan administrators important guideposts for developing an evidence base on which to build effective policy and programmatic initiatives for chronic care management.