Costs and implementation outcomes associated with strategies to increase use of a hospital assisted mobility program: evidence from a hybrid type 3 cluster randomized trial

Abstract

Abstract Background Assessing the cost of implementation strategy bundles and the relationship between costs and implementation outcomes is critical for informing investments in spreading an evidence-based practice. Thus, we compared costs of two strategies with different levels of intensity to support implementation of ‘assiSTed eaRly mobIlity for hospitalizeD older vEterans’ (STRIDE), an evidence-based mobility program that consists of a gait-assessment and supervised walks for hospitalized older adults in the Veterans Health Administration (VHA). Methods We measured STRIDE implementation costs across 35 VA medical centers (VAMC). As part of a type 3 hybrid cluster randomized trial, VAMCs were randomly assigned to receive either (i) standardized technical support (REP) or (ii) standardized technical support and external practice facilitation (enREP). We used an activity-based costing methodology to account for costs for external and site-level implementation activities across a 10-month active implementation period from the perspective of health system leadership. Costs were obtained using survey data and national salary estimates. We describe the association between implementation costs and STRIDE penetration, or the extent to which the program reached patients, and program adoption, based on reaching a pre-specified level of program activity. Results Median REP VAMC ( n = 19 VAMCs) cost was $18,645 and varied significantly (Q1–Q3: $14,089–$28,579) and median enREP VAMC ( n = 16 VAMCs) cost was $20,007 (Q1–Q3: $17,341 - $23,083). 10% of REP costs and 13% of enREP costs were associated with external support. Implementation cost was not strongly correlated with penetration of STRIDE (Spearman r = 0.08). Costs were greater at REP VAMCs that adopted STRIDE. For enREP VAMCs, costs were similar for VAMCs that did and did not adopt STRIDE. Conclusions The enREP intensified facilitation support resulted in an increase in cost compared to REP but was not associated with improved implementation outcomes. Decision-makers planning to scale STRIDE-like hospital mobility programs should anticipate that local site-level costs are the largest driver of cost and should be the principal target for budget planning. Study findings may have implications for how to best design and deploy varying levels of intensity of implementation support to effectively scale hospital mobility programs in real-world care settings. Trial registration ClinicalTrials.gov (identifier NCT04868656). No key protocol deviations or unplanned analysis related to cost occurred in this trial.