We believe narrative reporting is worth introducing into your laboratory practice. We have only done this within our discipline but there may be implications for the other laboratory disciplines as well as other clinical supporting disciplines such as radiology.
Laboratory professionals inside Wales and further afield in the UK and worldwide we hope will be interested in adopting narrative reporting. It does rely on the versatility of the Laboratory Information Management System being able to place detailed clinical comment in such a position to disrupt thinking in the reader and to stop and think.
If we are not careful, a busy clinician will read a pathology report and make a rapid response. This is the brain working in the fast reactive mode Kahneman describes. By relegating antibiotic sensitivities to the bottom of a laboratory report and introducing a comment, we hope to cause this disruption in thinking that will cause the clinician to move to slow, reflective thinking and review what might be the best option for the patient. Our aim is always to keep antibiotic use for clear symptoms and signs of infection and not using antibiotics simply because we have reported a set of sensitivities, which in the past may have contributed to the rise in antimicrobial resistance.
Within 3 months of introducing narrative reporting, the quality of the clinical information on the request cards improved. This becomes a virtuous circle. As the request information improved, so our responses can be tailored to respond to the clinical questions being posed.
As we demonstrated in our published paper in the Journal of Infection Prevention, while we were not directly aware that our urine numbers had declined until we had been narrative reporting for a year, retrospective analysis showed that a change point had occurred within 3 months of the introduction of the new format reports. This was then replicated when we tested a second time for a period of 18 months for Betsi Cadwalader University Health Board in North Wales, who were receiving 15,000 urine samples per month and we were able to accurately predict a decline of 2,000 samples per month at the end of the first year. This has continued, as we discussed in the “Why” tab.