Using a Text-Mining Approach to Evaluate the Quality of Nursing Records

Stud Health Technol Inform. 2016:225:813-4.

Abstract

Nursing records in Taiwan have been computerized, but their quality has rarely been discussed. Therefore, this study employed a text-mining approach and a cross-sectional retrospective research design to evaluate the quality of electronic nursing records at a medical center in Northern Taiwan. SAS Text Miner software Version 13.2 was employed to analyze unstructured nursing event records. The results show that SAS Text Miner is suitable for developing a textmining model for validating nursing records. The sensitivity of SAS Text Miner was approximately 0.94, and the specificity and accuracy were 0.99. Thus, SAS Text Miner software is an effective tool for auditing unstructured electronic nursing records.

MeSH terms

  • Data Accuracy*
  • Data Mining / methods*
  • Electronic Health Records / classification*
  • Electronic Health Records / statistics & numerical data
  • Natural Language Processing
  • Nursing Records / classification*
  • Nursing Records / statistics & numerical data
  • Quality Assurance, Health Care / methods*
  • Reproducibility of Results
  • Sensitivity and Specificity
  • Software
  • Taiwan