The performance of medical record review as an instrument for measuring and improving patient safety

The general aim of this thesis was to evaluate the performance of medical record review (MRR) as an instrument to measure and improve patient safety in clinical care. This method consisted of the Harvard Medical Practice Study (HMPS) trigger tool, although slightly adapted and only applied to patients who died during their stay in our hospital. We calculated the predictive value of the triggers and their reproducibility. Furthermore, we assessed the inter- and intrarater reproducibility for the decision on the presence of an adverse event (AE), the potential preventability and the contribution to death of a patient. After that, we assessed the external reproducibility of MRR regarding this second stage. Additionally, we investigated if an automatic trigger tool could be helpful in the selection of cases with a (potentially preventable) AE. Finally, we evaluated the potential improvement of the trigger system by adding lab values. This thesis investigated the current MRR process in our centre and possibilities for improvement. It would also be of interest to expand the research and evaluate if and how this process affects the improvement of care. In short, medical record review can be used to evaluate the trend of AEs in a hospital. Due to its low reproducibility it is less useful for the evaluation of individual cases. To improve the reproducibility of preventability, it is advisable to define a uniform definition for preventability. Also, an evaluation of the costs versus the effect is necessary