Identification of Outcome Relevant Indicators in Routine Data

Overview

The availability of electronic documentation systems in patient care means that large amounts of clinical routine data are available from which conclusions can be drawn for improving patient care. Compared to conventional research approaches, a data science-oriented approach offers the possibility of identifying patterns in routine data ("pattern recognition") that are relevant for patient-centered outcomes. Numerous projects and sub-projects can be evaluated from this data set.

The patterns that are relevant for patient-centered outcomes can be combinations of different parameters (e.g. vital signs, laboratory values, previous illnesses), which in themselves do not necessarily have a pathogenic effect, but in a specific combination may have a high relevance for the patient-centered outcome. This project pursues as research goal the anesthesiological and intensive care risk reduction. To this end, the existing data sets of routine care are to be used to identify outcome-relevant patterns in order to derive recommendations for improving treatment in line with the patient's wishes. Standard Operating Procedures (SOPs) and Quality Indicators (QIs) in combination with the data of routine clinical care will be used as a basis. The approach outlined is closely linked to the development of quality-based treatment structures. In order to be able to offer medical treatment at a high level, associated processes must be known and operationalized, i.e. measurable. QIs (quality indicators) are an established instrument for measuring individual dimensions of treatment quality, and our clinic is a leading participant in this process at both national and international level (see Spies et al. Guidelines for Delirium, Analgesia and Sedation). The mapping of quality-based treatment structures as SOPs (Standard Operating Procedures) is also essential in this context (see Spies et al. SOPs in Anesthesiology and Pain Therapy, Thieme Verlag). By applying data science-based methods, this study pursues the overall goal of supporting the transfer of evidence-based findings in the form of QIs and SOPs into patient care. Numerous projects and sub-projects can be evaluated from this data set.

Conditions