Research chat 16: online feedback as a safety valve

Our 16th research chat was a special extended edition (45 minutes) as part of the Care Opinion 2022 autumn online conference on 10 November 2022.

In this chat I spoke with Alex Gillespie and Tom Reader of the Department of Psychological and Behavioural Science at LSE about their recently published research:

Online patient feedback as a safety valve: An automated language analysis of unnoticed and unresolved safety incidents

This research analysed over 146,000 stories on Care Opinion using an automated machine-learning approach. Key findings were:

• automated analysis can reliably detect patient safety issues reported by patients
• online patient safety concerns are associated with hospital level mortality
• staff reported patient safety concerns are not associated with hospital level mortality

The paper concludes: "Augmenting staff reports with patient reports may be especially valuable when there is uncertainty about staff reporting. In such contexts patients reporting freely online (e.g., anonymously, without consequence) may act as a safety valve, revealing safety incidents that have been unnoticed or unresolved."

I hope you enjoy the discussion!

Watch the session

Past research chats

Recorded video from previous research chats

Future research chats

Find out about future scheduled chats, and how to register