The general aim of the Laboratory of Clinical Epidemiology is to contribute to the improvement of health care in different medical fields.
The are two main guiding principles: to help physicians in using the available knowledge and resources at their best, and to contribute to the growth of applied knowledge for clinical practice.
The laboratory operates in the field of intensive care medicine. As the main area of activity the laboratory coordinates one of the first and largest groups of collaborative research in the world, the GiViTI (Italian Group for theEvaluation of Interventions in ICU).
Quality of Care in the Intensive Care Units
The main purpose of these research projects is the assessment and improvement of the quality of care in Italian Intensive Care Units (ICUs). It is a multi-annual project promoted on behalf of GiViTI, a collaborative network composed by more than half of the Italian ICUs and coordinated by the laboratory. The main focus is the Margherita Project, whose aim is the continuous evaluation of the quality of care. It is based on a free software developed by the laboratory and distributed to all the ICUs adhering to the GiViTI group. The software has been created with a modular structure, which easily enables the integration of the basic data collection (the “core” of Margherita) with the data collection for specific research projects (the Margherita “petals”). In January 2011 Margherita became an international project. Thanks to funding from the European Union and other contracts, the laboratory has, in fact, been able to develop new software and to distribute the project to eight countries: Slovenia, Hungary, Poland, Cyprus, Israel, Afghanistan, Sudan, and Switzerland. http://www.giviti.marionegri.it/
Appropriateness of the Intensive Care Units
ICU is a high technology environment that requires a large number of highly specialized personnel. Hence, the cost of these units is extremely significant and special attention to not wasting resources is mandatory. In this field, the laboratory launched a study to assess the level of appropriateness of the use of ICU beds in two Italian regions: Toscana and Piemonte. Such an evaluation is based on the understanding that the level of care provided by an ICU should correspond to the level of care it can theoretically provide, given the available resources. In this framework, patients are classified as requiring high-, low-, or ordinary-care, and beds are independently classified as high- or low-level. The appropriateness evaluation protocol adopted verifies the agreement between these two separate classifications. Ideally, an ICU bed equipped to provide a certain level of assistance should be used only for the corresponding level of complexity of the problem and of the healthcare need. When this occurs, it can be considered an appropriate healthcare assistance level. Again, ideally, there should be clear boundaries between levels: a higher healthcare assistance level should not be used to treat a problem of a smaller size, just as a low or medium assistance level should not be employed to treat a problem of greater need. When this occurs, it is considered an inappropriate use of resources. An inappropriate service is therefore one that could be provided at a lower assistance level (the inappropriateness, in this case, can be referred to as a waste of high-level resources), or one that should be provided at a higher assistance level (inappropriateness in terms of suitability of the ICU). The StART study’s aim is to assess these phenomenons and identify their causes in order to improve the appropriateness of ICU admissions.
The Reconstruction of Clinical Reasoning in Medical Practice and Education
This area represents the main concern of the Unit of Clinical Knowledge Engineering, whose objective is the valorization of clinical reasoning in solving complex clinical problems. The diagnosis of pulmonary embolism still represents a relevant clinical challenge due to the complexity of the patient's clinical presentation and the variability of diagnostic resources among centres. In this regard, we are conducting an Italian multicenter study, involving mainly emergency units, with the aim of prospectively validating the diagnostic software BayPAD (Bayes Pulmonary embolism Assisted Diagnosis). Such a tool, relying on a probabilistic model covering 72 clinical variables without the need to input all the contemplated observations, would overcome the main reasons that prevented ordinary clinical guidelines from being largely accepted. Moreover, the results of the retrospective validation of the system have been obtained. The unit started a project for the creation of a software assisting the physician in tracing back the basis of his clinical decisions before the description provided by clinical reports, among those that are typical of particular medical specialty. The software has a double target: to create specific applications based on probabilistic models representing complex clinical decision problems, and to involve physicians in their construction. The latter is achievable given the strong analogy between the causal structure of the exploited models (bayesian networks) and the pathophysiological structure of medical knowledge. This system could be adopted within medical training projects, with a special attention to e-learning programs.
An Electronic Health Record to Promote Research in Intensive Care Medicine
The main aim of this project is to develop an electronic health record (EHR) that allows the assessment of indicators of the process of care in the ICU. A multidisciplinary team of intensivists, ICU nurses, epidemiologists, statisticians, and IT specialists had the responsibility of planning the HER, which is now already shared by 30 Italian ICUs. This made it possible to launch the first analysis of the process that has as its goal the improvement of the practice of weaning from the ventilator.
Antibiotic Pharmacokinetics in Critically Ill Patients
AbioKin is a multicenter prospective observational study aimed at studying the pharmacokinetic properties of some of the most used antimicrobial drugs in Intensive Care Units (ICUs) and is carried out in ICUs adopting the electronic clinical record MargheritaTre. In the first phase of the project, plasma samples will be withdrawn from critically ill patients undergoing antibiotic therapy with at least one of the investigated molecules. In the second phase we will develop pharmacokinetics models describing the evolution of the plasma concentrations (measured in the collected samples) as a function of the patients' clinical conditions (extracted from MargheritaTre). The third phase will consist in the creation of a simulator to be integrated in MargheritaTre. This tool will allow the plasma concentrations of antimicrobial drugs, as a function of patients' conditions, to be computed and predicted in order to support clinicians in designing personalized therapies meeting the optimal therapeutic targets.
International Consensus on Cardiopulmonary Resuscitation.