Purpose: To describe the application of a big data science framework to develop a pain information model and to discuss the potential for its use in predictive modeling. Design and Method: This is an application of a cross-industry standard process for a data mining adapted framework (the Applied Healthcare Data Science Framework) to build an information model on pain management and its potential for predictive modeling. Data were derived from electronic health records and were composed of approximately 51,000 records of unique adult patients admitted to clinical and surgical units between July 2015 and June 2019. Findings: The application of the Applied Healthcare Data Science Framework steps allowed the development of an information model on pain management, considering pain assessment, interventions, goals, and outcomes. The developed model has the potential to be used for predicting which patients are most likely to be discharged with self-reported pain. Conclusions: Through the application of the framework, it is possible to support health professionals’ decision making on the use of data to improve the effectiveness of pain management. Clinical Relevance: In the long term, the framework is intended to guide data science methodologies to personalize treatments, reduce costs, and improve health outcomes.
Bibliographical noteFunding Information:
This study was financed in part by the Coordination for Improvement of Higher Education Personnel (CAPES; Brazilian Federal Agency for the Support and Evaluation of Graduate Education; Finance Code 001) and the National Council for Scientific and Technological Development (CNPq; 426779/2018‐5). This study was also supported by the Research and Events Incentive Fund of the Hospital de Clínicas de Porto Alegre (FIPE/HCPA).
© 2021 Sigma Theta Tau International
- Big data science
- electronic health records
- nursing informatics
- pain management
PubMed: MeSH publication types
- Journal Article