Medical Science Data Value Evaluation Model: Mixed Methods Study.

Abstract

Background: Medical science data hold significant value, and open platforms play a crucial role in unlocking this potential. While relevant platforms are being developed, the overall usage of these data values remains limited.

Objective: This study aims to propose a set of practical and effective data value evaluation processes and methods for medical science data open platforms, enabling them to manage and unlock the value of these data.

Methods: Integrating the information system success model, technology acceptance model, and consumer perceived value theory, a set of medical science data value assessment index systems was developed by adopting the literature review and expert survey methods. Data from 10 domestic and international open platforms were collected and empirically analyzed using the entropy-weighted Technique for Order Preference by Similarity to Ideal Solution technique.

Results: Based on the scores of each indicator, the intragroup correlation coefficient was calculated to be 0.489, indicating consistency in the evaluation. The highest information entropy values and weights determined using the entropy weighting method were the number of datasets (0.70, 17.68%), data timeliness (0.77, 13.44%), search comprehensiveness (0.78, 12.92%), and system responsiveness (0.80, 11.55%), respectively. Based on the weighted analysis, the platform with the highest overall score was the National Population Health Sciences Data Center, with a score of 62.32.

Conclusions: The evaluation index system and model developed can be used not only to optimize the platform's data value evaluation processes, but also to enhance the platform's overall data value and encourage users to reuse data.