Feasibility study of 'MiADE' point of care natural language processing
Research type
Research Study
Full title
Non-randomised feasibility study of point of care natural language processing using the MiADE system to assist structured clinical documentation
IRAS ID
322887
Contact name
Anoop Shah
Contact email
Sponsor organisation
University College London Joint Research Office (part of Research Support Centre)
Clinicaltrials.gov Identifier
Z6364106/2023/03/03 health research, Data protection registration
Duration of Study in the UK
0 years, 7 months, 29 days
Research summary
MiADE is a computer program designed to make it easier for clinicians to enter information in electronic health records in a structured, coded form. MiADE analyses free text notes entered by clinicians at the point of care. MiADE extracts important items of information from the text such as diagnoses, medication and allergies. Structured items extracted from the text are presented to the clinician for validation, and can be amended or corrected before being saved in the patient's record. MiADE is designed to make it easier for clinicians to enter structured data, and hence improve the completeness and quality of data in electronic health records.
This study aims to evaluate MiADE in order to assess its performance and utility, and to inform future improvements and developments. The evaluation consists of two before-and-after studies with qualitative and quantitative assessment.
Study 1 will be conducted in outpatient clinics, in which the number of structured entries that a clinician records per consultation will be compared before and after switching on the MiADE system. Feedback will be gathered from clinicians by surveys, and by interviewing a subset of clinicians. A sample of patient consultations will also be observed, and patients will be interviewed to find out how the system affects patients' experience of their consultation.
Study 2 will take place within the inpatient setting. Inpatient clinical teams will be recruited and surveyed, and the recording of structured entries will be compared before and after MiADE system is switched on. A sample of clinicians will also be interviewed.The findings from this study will be used to support further improvements to MiADE and the roll out of systems to improve structured data entry in electronic health records.
REC name
South Central - Hampshire A Research Ethics Committee
REC reference
23/SC/0221
Date of REC Opinion
2 Aug 2023
REC opinion
Further Information Favourable Opinion