Health Care Systems Engineering for Scientists and Practitioners

Falls are critical accidents occurring in hospitals and nursing homes. They can reduce the quality of life of care-receivers (patients and elderly persons) and deteriorate the professional duties of care-givers (nurses and care workers). As the initial approach to preventing such falls, micro-electro-mechanical systems (MEMS) have been applied around the care-receiver’s bed. For the present study, a system was developed using a tool kit with an AVR microprocessor (Arduino Uno), seven force-sensing resistors, and two human-detection sensors. This system can detect and predict dangerous motions that may lead to a fall. A predicted motion is notified visually and aurally to care-givers and/or care-receivers. The developed system is a technical product, and thus an organizational measurement is also required for preventing falls. As the second approach, the nursing process of care-givers was visualized using process modeling. Two diagrams, process content and state transition diagrams, were generated from daily nursing logs taken at night in both a hospital and a long-term care facility. Although a process content diagram can be used to visualize the nursing process of care-givers, a fall is a non-process type accident. A state transition diagram can thus express the nursing process in a care-receiver centered manner. Whereas care-givers take care of care-receivers early in the night, non-process type accidents may occur after this early period until the following morning. The developed sensing-device based system is useful after the early period of care at night for the prevention of falls.

In this research we schedule the admission of patients trying to smooth the utilization of diagnosis resources such as CT-scans and of surgical resources such as operating rooms. We also want to allocate a bed to each patient if possible in an adequate care unit taking into account the bed capacity of each care unit. We are faced with a bi-objective problem. The first objective minimizes the peak of resource demands and the second one minimizes the penalty cost due to inappropriate patient assignment to care units. We use a lexicographic method to solve this multi-objective model. The latter has been experimented and the results are discussed.

Once elective patients have been referred to surgery, they are is put in a waiting list until they can be scheduled according to their particular condition and the time they have spent in the list. Waiting lists are often managed in a rather informal and unstructured manner, leading to a lack of robustness and homogeneity in the assessing of the relative priority of patients. This study points out these drawbacks and proposes two multi-criteria decision making models, Group Analytic Hierarchy Process (AHP) and Analytic Network Process (ANP), in an attempt to formalize the prioritization process and mitigate the limitations of the prioritization systems observed in practice. Our numerical study confirms that the proposed models, which consider various perspectives in determining patients’ priorities, show a remarkable robustness.

Emergency departments in France are often saturated, “victims” of a vast number of patients seeking medical care, and of logistical problems linked to the lack of beds for patients waiting to be admitted. For over 10 years now, our ward has emergency nurses specially designated to facilitate patient orientation after emergency care. Over the years, the role and responsibilities of these nurses has evolved and improved. Their expertise is recognized at the highest level of the hospital administration. More recently in other emergency wards it has also become common practice, strongly backed by the Ministry of Health. These nurses are commonly known as Bed Managers. We would like to share our experience with you.

We present a novel approach for increasing the operating room efficiency by significantly reducing the turnover time in elective surgical case scheduling. Reduced turnover time typically leads to increased surgeon utilization and increased patient throughput. Our main contribution is an optimization model for generating operating room schedules in a surgery department, where the pre-procedure of a surgical case is allowed to overlap with the procedure and post-procedure activities of an ongoing surgical case assigned to the same operating room. In addition, we present a computational experiment, where we create and compare 1-day schedules for a surgery department consisting of three operating rooms. The results of the experiment clearly show the potential of the idea of allowing overlapping surgical activities. This feature allows for significantly improved schedules compared to when no overlaps are allowed. The experiment also verifies the correctness of the optimization model.

Observing China’s healthcare systems, this paper establishes a two-tiered healthcare network which plays a basic role in analyzing and designing multi-tiered healthcare systems in many cities of China. Note that the two-tiered healthcare network can express the health service delivery between different levels of hospitals under necessary cooperation. This paper proposes a mean-field theory for performance analysis of the complex two-tiered healthcare network, where the mean-field theory is developed by two key techniques: A time-inhomogeneous queue and a nonlinear Markov process. Based on this, we give some highlight on the healthcare service delivery system management, and the methodology of this paper is promising for being able to consider more general large-scale healthcare systems.

This paper addresses the problem of scheduling Magnetic Resonance Imaging Examinations and has a twofold aim: investigating, via simulation, the long-term effect of the implementation of a First Come First Served-like heuristic; assessing the effectiveness, in terms of due date fulfilment, of the implementation of three managerial actions: cross training radiologists; reserving capacity to urgent patients, reducing the number of inappropriate prescriptions. By means of a factorial analysis, we demonstrate that cross-training radiologists and reducing the amount of inappropriate prescriptions, allow better matching the patient’s due-dates, than reserving capacity to urgent patients.

Improving quality within healthcare systems is one of the biggest challenges policy makers are facing today. In this paper, we focus on planning for health policy makers with the two main purposes of improving patients’ health conditions and helping in strategic planning. These two objectives are considered together thanks to the volume-outcome association, a relation that associates higher volume of activity to better results, especially for surgical operations. The managerial use of the volume-outcome association, which has not been investigated so far, can allow us to determine the optimal ward dimension. We modeled the requirement for improved patients’ health conditions in different ways, analytically examining their implications. We used data from the Italian Piano Nazionale Valutazione Esiti (PNE) to compare the proposed solutions with a real provincial configuration on a limited subset of pathologies.

The admission of new patients to Home Care (HC) services is a relevant problem, which requires managing waiting lists while respecting a maximum admissible period between patient’s request and service beginning. In the literature, this problem is only marginally addressed when compared to other HC management issues, e.g., visit scheduling and patient assignment to personnel. The goal of our work is to develop a discrete event simulation framework for analysing admission policies and waiting list management strategies in HC services. In this paper, we propose and test a first model for the admission of patients to a HC rehabilitation service, where the care service must start within a fixed period from the acute event. We refer to the specific case of a provider operating according to the regulation of the health system of Milan, Italy, which is however general enough to derive general considerations. Results show the effectiveness of the methodology for this application, its versatility, and a quantitative evaluation of some directions of improvement in the system.

We study a problem of locating ambulances so that their coverage for timely response is maximized. While ambulance location problems have been extensively studied, the model proposed in this paper presents two novel features. First, our model explicitly takes a dispatching policy into account, which is motivated by the fact that a dispatching policy is a key component for ambulance operations. Second, instead of a probabilistic model commonly found in the literature, we take a stochastic programming approach to incorporate temporal variations in call arrivals. The advantage of our algorithm is demonstrated by comparing performances of our algorithm with other location models.

With rising healthcare costs, resources have to be used as efficiently as possible to maintain adequate patient care. In hospitals this holds especially true for the availability of beds for stationary patients. We propose a mixed-integer model where we cluster clinical departments to level out associated bed requirements over time, hereby incorporating seasonal effects of individual departments. This will reduce the probability of not being able to admit patients or having to search for available beds within other parts of the hospital as well as optimally allocate ward capacity to departments. To optimally cluster departments we consider medical prerequisites as well as staffing constraints. Medically, certain combinations of departments will not be allowed. For example, immunocompromised and infectious patients should not be sharing the same ward as well as children should not be sharing wards with adults. In terms of staffing constraints, we consider additional qualification and controlling cost of nursing staff due to increased requirements for clusters with multiple departments. Furthermore, next to allocating adequate ward capacity to each cluster of clinical departments we also consider relative distances within clusters as well as absolute distances between clusters and relevant infrastructure. This assures both, minimal walking distance between patients for physicians as well as proximity to important cluster-specific facilities such as the OR or radiology department. Nummerical studies show the improvement in terms of costs and required beds.

The objective of this paper is to investigate the optimal staffing ratio under various primary clinic settings. Specifically, by using simulations, we investigate the effects of workload shift and identify the proper ratio between medical assistants (MAs) and physicians (MDs) to achieve effective and efficient service level. The results articulate that the optimal staffing ratio is achieved when the workloads of physicians and MAs are balanced.

Outbreak of diseases can lead to a great number of deaths. Therefore, the establishment of an effective disease prevention and control plan, which can decrease the number of deaths, is of paramount importance. This paper critically reviews literatures focusing on disease prevention and control plans and proposes the directions for the future research. Because an effective disease prevention and control plan should be established based on the type of the disease, selected papers are first classified according to the diseases. Then, according to the characteristics of the disease, different methods are used to classify papers studying the same disease. Through the classification and analysis of the selected papers, a series of insights are derived and several future research directions are proposed.

In this paper, we propose a mixed integer goal-programming model to support the master surgical scheduling process. In order to comply with the process stakeholders, the objective function comprises four criteria, namely (1) the respect of the patients’ priorities, (2) the balancing of the utilisation of the operating rooms, (3) the balancing of the utilisation of the post-surgical units and (4) the maximisation of the number of scheduled surgeries. The model is tested on real data coming from a leading Italian hospital. The results of the preliminary numerical experiments show how the model allows obtaining reasonable trade-off among the considered objectives.

Emergency department overcrowding has been reported over decades around the globe and the phenomenon is observed to be worsening in recent years. The overcrowding issue will hinder critically-ill patients from accessing timely and adequate medical services, and may result in unnecessary deaths of emergency patients. Furthermore, it may lead to patient dissatisfaction due to the many hours of waiting for consultation. While most studies suggest that there is a mismatch between demand and supply for emergency care and this is the primary factor for the phenomenon, reducing system inefficiency is a possible way to relieve the overcrowding situation when the demand and supply are not adjustable. In this paper, we study the impacts of missing patients, referring to the patients who are not present at the time that they are called for consultation. We conduct a real case study and a simulation study of an emergency department in Hong Kong. We found that even if there is only a small proportion of missing patients and their missing time is short, there is a significant increase in patient waiting time. We suggest that emergency departments should consider to adopt information technology to reduce the inefficiency due to missing patients.

This paper introduces a tool to assess the impact of organizational strategies that are intended to allocate inpatient beds amongst emergent and elective flows inside a hospital. The tool, based on a System Dynamics model, is able to reproduce the entire system and the relationship between the various flows. In the absence of corrective strategies, an exogenous increase in the rate of arrivals at the Emergency Department (e.g., in winter) can trigger a reinforcing loop increasing elective waits and further overcrowding emergency rooms. The model can be used to discover the best strategies aimed at managing bed capacity between emergent and elective flows. Some preliminary results are given in the context of a public hospital located in Genova (Italy).

Efficient use of medical resources is essential to satisfying the surge demand experienced in the aftermath of a disaster. In hospitals, inefficiency exists in the use of bed resources when less urgent patients pre-occupy the beds so that there is no bed for more urgent patients arriving later. We develop a decision model to minimize such inefficiency in the event of a disaster through admission control of patients entering a hospital. We use Markov Decision Process (MDP) to make admission decisions for a finite horizon. It models the time-dependent arrival of disaster victims and their time-dependent survival probabilities. We numerically solve the MDP model using a discretization technique. The results of experiments conducted using virtual patient arrival data, in which the efficiency of our MDP solution was compared with that of other operating schemes, indicate that our proposed MDP model can improve the efficiency of current operations.

Computers are omnipresent within hospitals. All possible measures are taken to avoid failure. It is the responsibility of the IT department to ensure the security of data. On August 28, 2014, Saint-Joseph/Saint-Luc Hospital was confronted with a major computer system failure. The emergency management plan was launched; it is based on a “paper kit” to ensure continuity and the traceability of cares for all units.