Face validity of progressive rank criteria to identify high-utilizer sickle cell patients in an adult sickle cell disease medical home

Abstract

Abstract Background Community-based case management within a sickle cell disease (SCD) medical home offers the potential to reduce costly emergency and inpatient utilization. However, implementation often depends on the ability to identify and prioritize patients most likely to benefit from intensive services. While hospital systems may favor revenue-generating interventions, comprehensive SCD care programs prioritize approaches that reduce fragmented care and address complex psychosocial needs. To support the sustainability of our adult SCD medical home, we developed and evaluated a reproducible, face-valid strategy to identify the highest-utilizing patients annually for targeted case management. Methods Between 2018 and 2024, we implemented a progressive ranking strategy to identify the top 50 high-utilizer patients each year from a panel of over 300 adults enrolled in the SCD medical home. Using a full dataset of annual patient-level administrative data, we applied a single-step Excel Custom Sort to rank all patients in descending order using five utilization criteria:Total annual healthcare chargesNumber of inpatient discharges30-day hospital readmission rateTotal number of emergency department (ED) visitsRate of ED returns within 3 days This composite sort allowed us to rank patients based on multiple dimensions of healthcare utilization simultaneously, ensuring that both cost-heavy and volume-heavy users were captured. After sorting, the top 50 patients on the list were selected as high utilizers for intensive case management. For reporting purposes, patients were subsequently categorized into cost tiers:Red: >$199,999/yearYellow: $100,000–$199,999/yearGreen: <$100,000/year Results The ranking algorithm successfully identified 50 high-utilizer patients in each year from 2018 to 2024. While many patients fell into the Red tier, a significant proportion were in the yellow tier, indicating that high service utilization did not always correspond to the highest cost group. This reinforced the value of incorporating multiple utilization metrics into the ranking approach. Despite an overall downward trend in average annual charges, the methodology continued to yield cohorts with consistent and complex utilization profiles. Most selected patients had overlapping high need across both acute care and readmission dimensions, highlighting the multifactorial nature of SCD-related healthcare utilization. The approach proved adaptable to year-over-year shifts in healthcare costs and utilization patterns, while preserving clarity, reproducibility, and operational efficiency. Conclusion This progressive, multi-criteria ranking method provides a pragmatic and scalable approach to identifying high-utilizer patients in a sickle cell medical home setting. By leveraging simultaneous sorting across five utilization domains, it maintains clinical relevance and avoids overreliance on cost alone. The method demonstrated strong face validity across seven years of implementation and offers a flexible foundation for targeting high-touch interventions. With further validation, this approach may be generalizable to other chronic disease populations with complex care needs.