A decision-theoretic approach to identifying future high-cost patients.

Area of Science:

• Health economics
• Healthcare management
• Medical informatics

Background:

• Very-high-cost (VHC) patients represent a significant financial burden on healthcare systems.
• Accurate allocation of funds for VHC patients is essential for equitable reimbursement.
• Existing methods for VHC patient identification and fund allocation require refinement.

Purpose of the Study:

• To develop and evaluate a novel method for allocating funding for very-high-cost (VHC) patients among hospitals.
• To assess the effectiveness of a Bayesian decision rule for classifying VHC patients using diagnostic cost groups (DCGs).
• To compare the proposed allocation method with previous year's allocation for VHC patients.

Main Methods:

• Diagnostic cost groups (DCGs) were employed for risk adjustment in a cohort of 253,013 veterans.
• A Bayesian decision rule was developed to classify patients as VHC (defined as >$75,000) or not VHC.
• The method utilized FY 2003 DCGs for FY 2004 VHC fund allocation and compared it with FY 2002 DCGs for FY 2003 allocation.

Main Results:

• The developed decision rule identified DCG 17 as the optimal cutoff for predicting next year's VHC patients.
• Comparing allocation methods, the previous year's distribution most closely approximated the actual VHC patient distribution.
• The decision-theoretic approach offered insights into the economic impact of VHC patient classification.

Conclusions:

• The decision-theoretic approach provides valuable insights into the financial implications of classifying patients as VHC.
• Further research is necessary to refine methods for identifying future VHC patients.
• Fair reimbursement for healthcare systems treating VHC patients requires improved predictive identification strategies.