Feedback from Incident Reporting Systems in Healthcare

Project summary

Aim: To investigate potential mechanisms of safety feedback from incident reporting systems for NHS organisations based upon published research and relevant expertise in international health care and high risk industry.

Background and overview: In the UK, the department of health report An Organisation with a Memory (2000) highlighted the need for the National Health Service to develop systems that allow organisations to learn from failure and respond effectively to preventable patient safety incidents. Considerable subsequent emphasis has been placed upon establishing incident reporting systems at Trust and National levels, based upon models from safety management in high risk industries. Progress has been made in methods for collecting incident data and investigating root causes of incidents. There is less understanding, however, of how to effectively close the safety feedback loop, use information from reporting to respond to safety issues and develop solutions that effectively improve the safety of care delivered to patients. Furthermore, lack of feedback from incident reporting and the perception that reports disappear into a “black hole” is often cited as a barrier to reporting amongst health care professionals.

This project aimed to identify effective forms of feedback from incident reporting systems based upon systematic review of relevant research literature and consultation with an international expert panel on patient safety event reporting and safety management in high risk industries. The results from the review, along with input from a national work shop event for NHS risk managers, were used to define models of best practice for safety feedback from risk management systems. The resulting framework for Safety Action and Information Feedback from Incident Reporting (SAIFIR) was employed by a team at Coventry University in analysis of further case studies and a national survey of existing safety feedback practices in NHS Trusts.

Results: The 23 international health care reporting systems analysed in the review represented voluntary, anonymous and multi-institutional level systems, amongst other types. The specific feedback processes incorporated in each system were recorded. There was considerable heterogeneity in the mechanisms and modes of feedback used across local contexts. The mechanisms identified were classified according to whether they predominantly formed information or action outputs:

• Information feedback: In the form of safety newsletters or publications, to raise awareness of current operational risks amongst front line staff and to encourage reporting by promoting an open reporting culture.
• Action feedback: To address specific identified vulnerabilities and introduce corrective modifications in work systems, processes, equipment, training and organisational policy (Figure A).

• Figure A: A generic safety action feedback loop, forming a continuous cycle of detection, safety issue analysis, development and implementation of process changes, and evaluation.

From review of exemplary systems and subject-matter expertise, a generic framework called SAIFIR was developed, in order to provide a model for key feedback processes in NHS risk management systems. SAIFIR incorporated five key information and action feedback modes mapped to stages within the safety issue management process for reporting systems (Figure B).

 

• Figure B: A framework for Safety Action and Information Feedback from Incident Reporting (SAIFIR) incorporating five modes of feedback employed by systems with comprehensive information and action outputs.

Although all reporting systems purported to result in the introduction of beneficial changes to improve safety, only 70% incorporated rapid feedback capabilities for implementing an immediate response to a reported incident. Over 90% of reporting systems employed newsletters to disseminate information to front line personnel regarding operational risks, demonstrating a reliance upon this type of feedback in existing systems. 39% of the systems reviewed fed back information concerning incident outcome and safety issue progress to the original reporter(s). Drawing upon analysis of experience within various high-risk domains it was possible to identify a series of requirements for the design of effective feedback systems, which formed evaluative criteria for subsequent case study work. For example, the expert panel review highlighted the importance of individual dialogue with reporters in order to: a) demonstrate that actions are taken on the basis of reports (and in so doing stimulate future reporting to the system), and b) gain the input of local front line expertise concerning the causes of failures and how these might be addressed with practical and workable safety solutions.

Conclusions

• Effective safety feedback involves timely, visible and repeatable corrective action and quality improvement processes, in addition to publishing data on incident rates.
• Multiple means of feeding back actions and safety information should be employed to improve clinical processes, promote safety culture and awareness, and encourage reporting.

• Further research is needed into the relative effectiveness of different feedback mechanisms.

This research was undertaken in collaboration with the Coventry University and the University of Warwick.