A Data-Based Model Predictive Decision Support System for Inventory Management in Hospitals.

This paper presents experimental results from the application of a data-based model predictive decision support system to drug inventory management in the pharmacy of a mid-size hospital in Spain. The underlying objective is to improve the efficiency of their inventory policy by exploiting pharmacy historical data. To this end, the pharmacy staff was aided by a decision support system that provided them with quantities needed for the satisfaction of clinical needs and the risk of stockout in case no order is placed for different time horizons. With this information in mind, the pharmacy service takes the final order decisions. The results obtained during a test period of four months are provided and compared with those of a previous model predictive control approach, which was implemented in the same hospital in the past, and with the usual policy of the pharmacy department.

[Clinical simulation for quality improvement in the care of postpartum haemorrhage]. / Simulación clínica para la mejora de la calidad en la atención a la hemorragia posparto.

OBJECTIVE: To analyse and redesign the care process for patients with severe post-partum haemorrhage using simulation and a human factors approach. METHODS: The project was developed between June 2016 and May 2017. The working group was created following the participatory leadership method and included professionals with knowledge and position to influence the organisation. The existing process and clinical protocols were reviewed. An in situ simulation was used to observe team performance in the hospital recovery room. Information was expanded through an open and structured interview with professionals selected for their experience in the subject matter. Task analysis was used to document the process, and a failure mode and effects analysis was made to assess system vulnerabilities. Results were mapped using a flow chart. RESULTS: The analysis identified six groups of people working on different tasks, their activities and sequence of action, the importance of naming an explicit coordinator, the way in which information is disseminated and transformed, and the stages where it is necessary to share information and make key clinical decisions. The existing clinical protocols and the aids established in order to use the available resources were integrated, including blood draws and haemostatic agents, as well as an administration guide. CONCLUSIONS: The analysis of the patient care process in post-partum haemorrhage using in situ simulation with a human factors perspective, including an analysis of the interaction between professionals and the system where they work, established a detailed and personalised map of the components that determine how work flows through the organisation.

Learning curve patterns generated by a training method for laparoscopic small bowel anastomosis.

BACKGROUND: The identification of developmental curve patterns generated by a simulation-based educational method and the variables that can accelerate the learning process will result in cost-effective training. This study describes the learning curves of a simulation-based instructional design (ID) that uses ex vivo animal models to teach laparoscopic latero-lateral small bowel anastomosis. METHODS: Twenty general surgery residents were evaluated on their performance of laparoscopic latero-lateral jejuno-jejunal anastomoses (JJA) and gastro-jejunal anastomoses (GJA), using swine small bowel and stomach on an endotrainer. The ID included the following steps: (1) provision of references and videos demonstrating the surgical technique, (2) creation of an engaging context for learning, (3) critical review of the literature and video on the procedures, (4) demonstration of the critical steps, (5) hands-on practice, (6) in-action instructor's feedback, (7) quality assessment, (8) debriefing at the end of the session, and (9) deliberate and repetitive practice. Time was recorded from the beginning to the completion of the procedure, along with the presence or absence of anastomotic leaks. RESULTS: The participants needed to perform 23.8 ± 6.96 GJA (12-35) and 24.2 ± 6.96 JJA (9-43) to attain proficiency. The starting point of the learning curve was higher for the GJA than for the JJA, although the slope and plateau were parallel. Further, four types of learning curves were identified: (1) exponential, (2) rapid, (3) slow, and (4) no tendency. The type of pattern could be predicted after procedure number 8. CONCLUSIONS: These findings may help to identify the learning curve of a trainee early in the developmental process, estimate the number of sessions required to reach a performance goal, determine a trainee's readiness to practice the procedure on patients, and identify the subjects who lack the innate technical abilities. It may help motivated individuals to become reflective and self-regulated learners. Moreover, the standardization of the ID may help to measure the effectiveness of learning strategies and make comparisons with other educational strategies.

[Bioethics is dead. Long live medical ethics!]. / La Bioética ha Muerto. ¡Viva la Ética Médica!

The purpose of this paper is to show a paradigmatic crisis in academic bioethics. Since an important part of bioethicists began to relativize the ethical prohibition of killing an innocent human being, one way or another they began to ally with the death industry: the business of abortion, and then that of euthanasia. The thesis of this paper is that by crossing that Rubicon bioethics has been corrupted and has lost its connection to the ethical, political and legal discourse. One can only hope that it will revive from its ashes if it retakes the ″taboo″ of the sacredness of human life, something for which medical ethics could provide invaluable help, because it still keeps the notion that ″a doctor should not kill″, although in an excessively ″discreet″ and somehow ″ashamed″ way. However, conscientious doctors know more about ethics than most bioethicists.

Chance-constrained model predictive control applied to inventory management in hospitalary pharmacy.

This extended abstract addresses the preliminary results of applying uncertainty handling strategies and advanced control techniques to the inventary management of hospitality pharmacy. Inventory management is one of the main tasks that a pharmacy department has to carry out in a hospital. It is a complex problem because it requires to establish a tradeoff between contradictory optimization criteria. The final goal of the proposed research is to update the inventory management system of hospitals such that it is possible to reduce the average inventory while maintaining preestablished clinical guarantees.