“Every system is perfectly designed to get the results it gets.”
In May 2018 a nine-month child was accidently given an overdose of furosemide by his mother. An appreciative enquiry approach was used to process map this incident. Conversations with everyone involved revealed opportunities to prevent the error, as the risks of using different formulations of the same medicine in primary and secondary care were already well known. This raised the question about current methods for capturing and learning from medicines harms. The purpose of this paper is to demonstrate, using the improvement science lens, that there is already information in the system available to help commissioners and providers minimise medication errors. Instead of introducing new methods of capturing medication errors or technical strategies to improve safety, which have balancing effects on the system, rather what is needed is a shift to strengthening human factors and relationship based learning. Primary care networks are an ideal opportunity to test system learning from significant events through multi-disciplinary reflective learning with commissioners so that the World Health Organization (WHO) challenge to every health system to halve severe and avoidable harm caused by medicines by 2022, can be achieved.