Presentation Details
| Using ChatGPT to Bridge Technical Gaps and Accelerate REDCap Development for Perioperative QI Madison Hochrein1, Jordan Lee1, Rebecca M.Gerlach2. 1University of New Mexico School of Medicine, Albuquerque, NM, USA.2Department of Anesthesiology and Critical Care, University of New Mexico Hospital, Albuquerque, NM, USA |
Abstract
BACKGROUND: Building surveys in REDCap can be challenging for researchers who are not familiar with field types, branching logic, or data dictionaries. These details can pose barriers for creating quality improvement projects. ChatGPT can act as a real-time assistant that gives step-by-step build instructions from plain language prompts. We used ChatGPT to help a first-time REDCap user build the data entry tool for a respiratory prehabilitation quality improvement project.
PURPOSE: ChatGPT was prompted to give step-by-step instructions for building each survey question in REDCap. It instructed the user which field type to choose, how to name variables, how to enter choices, and when to use branching logic. A first-time REDCap user followed these directions to build the instrument, which was later used to enter data from paper surveys collected in a Pre-Anesthesia Clinic respiratory prehabilitation project.
RESULTS: With this approach, the new REDCap user created a working data entry instrument without formal REDCap training. Build time was shorter, fields were organized clearly, and the tool behaved as expected during data entry. The user could ask follow-up questions in real time, which helped with troubleshooting and learning. The REDCap instrument was used to enter real study data and worked reliably for further analysis.
CONCLUSIONS: ChatGPT can lower the learning curve for REDCap, help nonexpert users build surveys more efficiently, and support timely perioperative quality improvement work. This approach may allow more learners and clinicians to take part in survey-based QI projects without needing the pre-requisite of having advanced technical skills.
No part of this publication may be reproduced, distributed, or transmitted in any form or by any means, including photocopying, recording, or other electronic or mechanical methods, without the prior written permission of the author.
PURPOSE: ChatGPT was prompted to give step-by-step instructions for building each survey question in REDCap. It instructed the user which field type to choose, how to name variables, how to enter choices, and when to use branching logic. A first-time REDCap user followed these directions to build the instrument, which was later used to enter data from paper surveys collected in a Pre-Anesthesia Clinic respiratory prehabilitation project.
RESULTS: With this approach, the new REDCap user created a working data entry instrument without formal REDCap training. Build time was shorter, fields were organized clearly, and the tool behaved as expected during data entry. The user could ask follow-up questions in real time, which helped with troubleshooting and learning. The REDCap instrument was used to enter real study data and worked reliably for further analysis.
CONCLUSIONS: ChatGPT can lower the learning curve for REDCap, help nonexpert users build surveys more efficiently, and support timely perioperative quality improvement work. This approach may allow more learners and clinicians to take part in survey-based QI projects without needing the pre-requisite of having advanced technical skills.
No part of this publication may be reproduced, distributed, or transmitted in any form or by any means, including photocopying, recording, or other electronic or mechanical methods, without the prior written permission of the author.
Content Locked. Log into a registered attendee account to access this presentation.