Identifying Fraudulent Responses in a Study Exploring Delivery Options for Pregnancies Impacted by Gestational...

Area of Science:

• Perinatal medicine
• Research methodology
• Data integrity

Background:

• Current literature lacks clarity on gestational diabetes mellitus delivery timing and safety.
• Web-based surveys are increasingly used in medical research, but methods for handling fraudulent responses are underdeveloped.
• An incident of fraudulent survey activity shifted research focus to addressing data integrity.

Purpose of the Study:

• To highlight a systematic approach for identifying and handling fraudulent responses in a web-based survey.
• To address the implications of fraudulent data in clinical perinatal research.
• To emphasize the importance of robust methods for maintaining research integrity.

Main Methods:

• Cross-sectional web-based surveys were distributed to pregnant patients and perinatal health care providers in Canada.
• Fraudulent responses were identified using criteria such as identical timestamps, similar wording, and invalid email addresses.
• Systematic guidelines were developed in consultation with IT services and the research ethics board to filter data.

Main Results:

• An influx of nearly 2500 responses was recorded, with a significant portion flagged as fraudulent.
• Identified fraudulent responses included identical entries, minor wording variations with similar timestamps, and fake email addresses.
• The incident underscored the vulnerability of web-based surveys to fraudulent activity.

Conclusions:

• Implementing systematic guidelines is crucial for preserving research integrity and extracting valid data.
• Addressing fraudulent responses is essential to maintain the credibility of research findings and evidence-based practices.
• Further research is needed on optimal methods for detecting and mitigating fraud in web-based surveys.