Measure What Matters: Rethinking Surrogate Metrics of Success in Radiation Oncology.

PURPOSE: Census-based surrogates such as patients on treatment, fractions delivered, and course length have traditionally been used to assess radiation oncology departmental performance. These measures may no longer capture the complexities of modern radiation therapy, given the increased adoption of hypofractionation and stereotactic techniques. This study analyzes longitudinal trends in clinical operations, including treatment modalities, fractionation patterns, and financial performance, to identify value-based departmental performance metrics.

METHODS AND MATERIALS: A retrospective analysis was conducted at an (National Cancer Institute) NCI-designated cancer center spanning a 10-year period (FY15-24). Operational and financial data were extracted from electronic medical records and billing systems, including consultations, treatment starts, total fractions, fractions per patient, and annual collections. Descriptive statistics were used to assess associations between respective operational surrogates and financial outcome.

RESULTS

Consultations increased by 77.8%, treatment starts by 43.6%, and average fractions per patient decreased by 12.0%. Utilization of advanced modalities (intensity modulated radiation therapy, stereotactic body radiation therapy, (stereotactic radiosurgery) SRS) grew substantially, with stereotactic courses rising from 39 courses in FY15 (4%) to 291 courses by FY24 (22%). Mean IMRT course length decreased from 24.6 to 13.8 fractions. Conversion rate from consultations to new treatment starts dropped from 92% to 71%. Fractions per patient (r = -0.42) and other census-based surrogates demonstrated weak or inverse associations with collections. Collections were strongly correlated with consultations (r = 0.97) but demonstrated a comparatively weaker regression slope ($7238/consult) compared with new treatment starts (r = 0.90, R2 = 0.82, 15,500/start). Professional work relative value units also demonstrated strong correlations with collections but showed heterogenous alignment with clinical activity, reflecting sensitivity to coding structure rather than discrete care delivery.

CONCLUSIONS

Legacy tabulation-based metrics no longer reflect operational and financial value in the contemporary era. New treatment starts represent the most reliable surrogate, capturing clinical throughput, financial performance, and physician workload. Benchmarking based on new treatment initiation may improve performance assessment and appropriate resource allocation in alignment with value-based care.