Objective: Global Trigger Tool (GTT) has been proposed as a low-cost method to detect adverse events (AEs). The validity of the methodology has been questioned because of moderate interrater agreement. Continuous training has been suggested as a means to improve consistency over time. We present the main findings of the implementation of the Italian version of the GTT and evaluate efforts to improve the interrater reliability over time. Methods:The Italian version of the GTT was developed and implemented at the San Bonifacio Hospital, a 270-bed secondary care acute hospital in Verona, Italy. Ten clinical records randomly selected every 2 weeks were reviewed from 2009 to 2014. Two-stage interrater reliability assessment between team members was conducted on 2 subsamples of 50 clinical records before and after the implementation of specific review rules and staff training.Results: Among 1320 medical records reviewed, a total of 366 AEs were found with at least 1 AE on 20.2% of all discharges, 27.7 AEs/100 admissions, and 30.6 AEs/1000 patient-days. Adverse events with harm score E and F were respectively 58.2% (n = 213) and 38.8% (n = 142). First round interrater reliability was comparable with other international studies. The interrater agreement improved significantly after intervention (κ interrater I = 0.52, κ interrater II = 0.80, P < 0.001).Conclusions: Despite the improvements in the interrater consistency, overall results did not show any significant trend in AEs over time. Future studies may be directed to apply and adapt the GTT methodology to more specific settings to explore how to improve its sensitivity.

Introduction: The role of the emergency medical dispatch centre (EMDC) is essential to ensure coordinated and safe prehospital care. The aim of this study was to implement an incident report (IR) system in prehospital emergency care management with a view to detecting errors occurring in this setting and guiding the implementation of safety improvement initiatives. Methods: An ad hoc IR form for the prehospital setting was developed and implemented within the EMDC of Verona. The form included six phases (from the emergency call to hospital admission) with the relevant list of potential error modes (30 items). This descriptive observational study considered the results from 268 consecutive days between February and November 2010. Results: During the study period, 161 error modes were detected. The majority of these errors occurred in the resource allocation and timing phase (34.2%) and in the dispatch phase (31.0%). Most of the errors were due to human factors (77.6%), and almost half of them were classified as either moderate (27.9%) or severe (19.9%). These results guided the implementation of specific corrective actions, such as the adoption of a more efficient Medical Priority Dispatch System and the development of educational initiatives targeted at both EMDC staff and the population. Conclusions: Despite the intrinsic limits of IR methodology, results suggest how the implementation of an IR system dedicated to the emergency prehospital setting can act as a major driver for the development of a "learning organization" and improve both efficacy and safety of first aid care.

Purpose According to literature, interruptions during drug administration lead to a significant proportion of medication errors. Evidence on the effectiveness of interventions to reduce interruption is still limited. The purpose of this paper is to explore main reasons for interruptions during drug administration rounds in a geriatric ward of an Italian secondary hospital and test the effectiveness of a combined intervention. Design/methodology/approach This is a pre and post-intervention observational study based on direct observation. All nurse staff (24) participated to the study that lead to observe a total of 44 drug dispensing rounds with 945 drugs administered to 491 patients in T0 and 994 drugs to 506 patients in T1. Findings A significant reduction of raw number of interruptions (mean per round from 17.31 in T0 to 9.09 in T1, p<0.01), interruptions/patient rate (from 0.78 in T0 to 0.40 in T1, p<0.01) and interruptions/drugs rate (from 0.44 in T0 to 0.22 in T1, p<0.01) were observed. Needs for further improvements were elicited (e.g. a greater involvement of support staff). Practical implications Nurse staff should be adequately trained on the risks related to interruptions during drug administration since routine activity is at high risk of distractions due to its repetitive and skill-based nature. Originality/value A strong involvement of both MB and leadership, together with the frontline staff, helped to raise staff motivation and guide a bottom-up approach, able to identify tailored interventions and serve concurrently as training instrument tool.