Implications of Uncertainty in Hospital Information System Data for Performance Evaluation: A Robust Data Envelopment Analysis Approach

Background: Hospital efficiency evaluation relies heavily on data obtained from Health Information Systems. Along with deterministic data, these systems contain indicators that may be affected by measurement limitations, data recording quality, and information fluctuations, resulting in different levels of uncertainty. However, existing approaches do not explicitly consider the coexistence of deterministic and uncertain data.

Methods: Considering the nature of data generated during hospital information management processes and differences in the reliability of indicators, a generalized robust DEA framework was developed to simultaneously model deterministic and uncertain data. Three robust models were developed to incorporate uncertainty in inputs, outputs, and both inputs and outputs. The performance of the proposed models was evaluated through a hospital case study.

Results: The results showed that uncertainty in HIS data can lead to instability in efficiency values and reduce the reliability of performance evaluation results. Compared with the conventional DEA model, the proposed models provided more stable efficiency scores, more consistent rankings, and reduced the occurrence of unrealistic efficiency values.

Conclusion: By distinguishing between deterministic and uncertain data, the proposed framework provides a more realistic and reliable evaluation of hospital performance within Health Information Systems.

Key Message: Applying this framework in Health Information Systems can reduce the effects of data uncertainty on performance evaluation and support managerial decision-making, particularly in resource allocation and hospital performance improvement.