Validation of primary and hospital care resource use and cost estimates in OMOP-mapped vs source healthcare administrative datasets

Journal article

Objectives

To assess whether estimates of primary and hospital care resource use and costs from Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM) mapped data are consistent with source healthcare administrative datasets and to discuss implications for OMOP-based health economic analyses.

Methods

We conducted a retrospective cohort study (2010–2018) of women aged >50 years with a fragility fracture. Analyses were performed in parallel using source Clinical Practice Research Datalink Aurum linked to Hospital Episode Statistics data and two OMOP-based strategies: OMOP-standard (standardised concepts only) and OMOP-source (also including source variables). Primary and hospital care utilisation and costs were estimated over two years. Mean differences were summarised with bootstrapped 95% confidence intervals.

Results

Cohorts were comparable (23,106 source; 22,900 OMOP), with similar characteristics and follow-up. Resource utilisation was consistent across databases (<1% difference). Total cost differences were 2.2% (OMOP-standard) and 1.7% (OMOP-source, not statistically significant). Primary care, inpatient and outpatient cost differences were ≤3% among costed records.  Emergency care costs were ~21% lower in both OMOP strategies, reflecting a different costing method rather than mapping fidelity (0.1% when the same method was applied to source data). Discrepancies were mainly due to costing assumptions, and less so to incomplete vocabularies and structural constraints.

Conclusions

OMOP-mapped data produced resource use estimates highly consistent and aggregate cost estimates concordant with source data. Remaining cost discrepancies were concentrated in specific settings and reduced when source variableswere used. Refining mapping and costing approaches will improve accuracy and support federated economic research using OMOP-CDM.

Authors: Fabiano, G; Njuki, NJOKI; Delmestri, A; Pinedo Villanueva, R

• Publication status: Accepted
• Peer review status: Peer reviewed
• Publisher: Elsevier
• Journal: Value in Health More from this journal
• Acceptance date: 2026-11-02
• EISSN: 1524-4733
• ISSN: 1098-3015
• Language: English
• Keywords: federated data networks; healthcare resource use and costs; real-world data; administrative healthcare data; OMOP Common Data Model