Purpose: Literature on early mobilization and exercise training in patients recovering from acute heart failure (AHF) is promising. However, there lacks uniformity in the time of initiation, exercise prescription, safety criteria, and termination criteria. Thus, the aim was to develop a mobilization algorithm for patients recovering from AHF.

Methods: A modified web-based Delphi process was undertaken involving 15 panelists from across the globe. In Round 1, new variables, modifications to suggested variables, and agreement-disagreement within the panelists were obtained. In Round 2, agreement on a 5-point Likert scale was obtained. In Round 3, the algorithm was compiled, and excluded statements were discussed via web-based calls. The new variables were grouped into themes via an inductive process. The level of agreement and rating for each statement were analyzed using descriptive statistics, including frequency and percentages. We used Kappa statistics to examine the level of agreement between the panelists for each criterion. The consensus criterion was defined a priori as statements with a mean rating of ≥4 on the 5-point Likert scale by ≥70% of the panelists and Kendall’s coefficient of concordance (W) of ≥0.3 between panelists.

Results: From Round 1, 54 new variables were obtained. More than two-thirds (118/170, 69%) of statements reached consensus and reported a fair level of agreement between panelists (W ≥ 0.3). The final algorithm with all its criteria received a 100% (13/13) consensus.

Conclusion: Physician-referral criteria, physiotherapy assessment process, exercise prescription, safety criteria, and termination criteria were formulated for early mobilization and exercise training for patients recovering from AHF.