Combining institutional and administrative data to assess hospital costs for patients receiving ventricular assist devices

OBJECTIVES: The aim of this study was to describe patient level costing methods and develop a database of healthcare resource use and cost in patients with AHF receiving ventricular assist device (VAD) therapy. METHODS: Patient level micro-costing was used to identify documented activity in the years preceding and following VAD implantation, and preceding heart transplant for a cohort of seventy-seven consecutive patients listed for heart transplantation (2009-12). Clinician interviews verified activity, established time resource required for each activity, and added additional undocumented activities. Costs were sourced from the general ledger, salary, stock price, pharmacy formulary data, and from national medical benefits and prostheses lists. Linked administrative data analyses of activity external to the implanting institution, used National Weighted Activity Units (NWAU), 2014 efficient price, and admission complexity cost weights and were compared with micro-costed data for the implanting admission. RESULTS: The database produced includes patient level activity and costs associated with the seventy-seven patients across thirteen resource areas including hospital activity external to the implanting center. The median cost of the implanting admission using linked administrative data was $246,839 (interquartile range [IQR] $246,839-$271,743), versus $270,716 (IQR $211,740-$378,482) for the institutional micro-costing (p = .08). CONCLUSIONS: Linked administrative data provides a useful alternative for imputing costs external to the implanting center, and combined with institutional data can illuminate both the pathways to transplant referral and the hospital activity generated by patients experiencing the terminal phases of heart failure in the year before transplant, cf-VAD implant, or death.