Operations Research for Health Care最新文献

Long wait times for health care services is a known challenge in most health care systems. This is partially due to limited capacity and increased demand, but also due to sub-optimal scheduling policies. In this paper, we consider a health system in which patients are prioritized based on their acuity level. We assume that there is a wait time target for each acuity level to ensure that patients of lower acuity do not wait for an unreasonable amount of time while higher acuity patients are being served. We apply a robust optimization (RO) approach to schedule patients over a multi-period finite horizon considering the wait targets. First, we present a deterministic mixed-integer programming model which considers patient priorities, available capacity, and wait time targets for each priority level. We then investigate the robust counterpart of the model by considering uncertainty in demand and employing the notion of budget of uncertainty. Finally, we numerically compare the proposed robust model with the deterministic method. Our results demonstrate that the proposed robust approach provides solutions with higher service levels and lower wait times. Our results also provide insights on how expanding capacity and choosing the level of uncertainty affect the performance of the system.

UNAIDS’ 90–90–90 goal for 2020 is for 90% of HIV-infected people to know their status, 90% of infected individuals to receive antiretroviral therapy (ART), and 90% of those on ART to achieve viral suppression. To achieve these ambitious goals, effective care delivery programs are needed. In this paper we present a case study showing how HIV care can be improved by viewing the patient care process as a production process and applying methods of process improvement analysis. We examine the continuum of HIV care at a hospital-based HIV clinic in Kingston, Jamaica. We perform qualitative analysis to identify key programmatic, personnel, and clinical areas for process improvement. We then perform quantitative analysis. We develop a stochastic model of the care process which we use to evaluate the effects of potential process improvements on the number of patients who receive ART and the number who achieve viral suppression. We also develop a model for optimal investment of a fixed budget among interventions aimed at improving the care cascade and we use the model to determine the optimal investment among three interventions that the clinic could invest in. By viewing the patient care process as a production process and applying qualitative and quantitative process improvement analysis, our case study illustrates how clinics can identify the best ways to maximize clinical outcomes. Our methods are generalizable to other HIV care clinics as well as to clinics that provide care for other chronic conditions (e.g., diabetes, hepatitis B, or opioid use disorder).

One of the main elements affecting the performance of the organ transplantation system is the set of organ allocation boundaries that limits the number of organs shared among regions. To overcome these boundary limits, members of the Organ Procurement Transplant Network - OPTN - (including, transplant hospitals, Organ Procurement Organizations (OPO), medical/scientific members, among others) can propose a variance to the current allocation system to allocate organs differently than the OPTN policies. Over the years, several variances have been enacted by different members for various organs. In this study, we focus on the analysis of sharing variances which allow allocating organs within participating members before offering them at other levels. This type of variance has been successfully implemented in the past. For example, Florida and Tennessee created the Statewide Sharing program whereby kidneys are made available within-state donor service areas before they are made available for regional or national allocation. This program removed geographic disparities within those two states and resulted in better performance of the system in the states. Given these success stories, we propose a multi-period optimization model that can be used to determine the best policy for a local sharing program for any given OPO. We use liver allocation for the GALL OPO (i.e., LifeLink of Georgia) in the state of Georgia (USA) as a test case; however, our approach could be used for a variety of organs in any OPO.

Healthcare supply chain systems around the world face an immediate pressure to transform in the face of unprecedented demand and high Customer Service Level (CSL). And the expiration date of a pharmaceutical product is also a widespread issue in a customer’s purchase decision. In view of this, studies are essential to understand operations in healthcare systems and to offer decision support tools that improve public health, patient safety and strategic decision-making in the healthcare systems. This paper proposes a game theory approach for addressing inventory strategies for managing the flow of a pharmaceutical raw-material incorporating different quality characteristics in a two echelon hospital and pharmaceutical company supply chain. This research also considers a more realistic situation where the deteriorating rate of a finished product gradually increases as the expiration date approaches. We design a procedure for deciding an optimal strategy to achieve the target CSL of the hospital with the help of game theory payoff matrix. Numerical example is presented to illustrate the solution procedure and the sensitivity analysis of some key parameters is provided to demonstrate the proposed model.

This paper proposes an integrated approach to merge patient prioritization and patient scheduling to improve access to services in an elective (i.e., non-urgent) context. In particular, we assume that patients are included on a waiting list for a given surgery, and that every patient on the list has received a “utility score”, which is a proxy for the relative urgency with regards to the other patients on the list. A mathematical model is formulated to solve the patient scheduling problem, i.e., the simultaneous assignment of surgery sessions to surgeons and patients to surgeons, in such a way that the total utility is maximized along with other practical requirements. The model has been applied to a testbed of randomly generated instances, inspired by the context of the Urology Department at a University Hospital in Quebec City. Experiments have been conducted to analyze both the short- and medium-term behaviors of the proposed approach. The numerical results confirm that the use of an objective function designed to maximize utility does not deteriorate the efficiency of the resulting schedules in terms of the number of surgeries performed. They also show that, as expected, higher utility patients are scheduled first, and their waiting time before surgery are shorter than those of lower utility. However, this approach may lead to longer, and even unacceptable waiting times for low utility patients. To mitigate such an undesirable effect, a dynamic utility updating approach is proposed to progressively increase the utility of patients according to their time spent on the waiting list. This approach seems to adequately balance the advantages of scheduling patients based on their utility and the risk of causing too much delay for low priority patients.

Health posts aim to elevate the level of public health by providing primary health services. The effectiveness of a health post network depends on factors such as the number and location of health posts. The service quality in such networks can be improved by controlling the amount of congestion at the facilities. An important policy for decreasing operational cost of these networks is to consider a workforce mix of flexible and dedicated servers. This paper integrates the network design with workforce cross-training in the presence of congestion, where the queuing system at each health post is modeled by a set of multi-class M/G/$m$ queues offering multiple service types. The problem is formulated as an integer nonlinear programming model, and a linearization method is used to solve it. A hypothetical case study illustrates how the model can be used, and interesting managerial insights are presented.

The goal of this study is to determine which provider-level properties (e.g., provider types, patient contact factors) of healthcare workers (HCW) have the greatest impact on the transmission of healthcare associated infections (HAIs). This study focused on Carbapenem-resistant Enterobacteriaceae (CRE) acquisition for patients who stayed in a long-term acute care hospital (LTACH) in central Virginia during July and August 2014. We used both patient data (e.g., bed movement, screening results for CRE) and provider activity data documented through the electronic medical record. We created a network of patients for each HCW and performed Poisson regression analysis including the network measures. A total of 204 providers saw at least one of the nine positive patients who stayed in the LTACH over the study period. From the Poisson regression, provider types, total number of patients each provider saw, LTACH workdays, average number of patients per day during LTACH workdays, and the provider’s network were associated with the frequency of case contact. Our study demonstrated that in addition to patient data, provider activity logs that show provider-level properties can be used to assess the role of healthcare workers in transmitting HAIs and highlight risk mitigation opportunities.

This paper investigates the current patient transportation between care units in a large hospital to determine possible solutions to reduce total completion times of demands. The goal is to avoid major changes in the current staff schedules. Historical data of the service calls is available and an in-depth analysis is conducted to identify popular routes and current assignment of demands to patient transport employees. We present a mixed-integer model to determine the best distribution of the employees throughout the most popular routes of the hospital to minimize costs. Experiments are conducted on real data from CHU de Québec-Université Laval, HEJ, in the province of Québec, Canada. Results obtained from assigning specific employees to routes instead of the current method, which consists at assigning employees to all of the hospital are compared and show that there is a gain in doing so.

Non-clinical hospital services to support clinical activities, such as the sterilization service and clinical engineering, are an important technology asset in healthcare, and require constant improvement aimed to reduce economic burden and increase quality. The selection of the most effective healthcare service to adopt in a healthcare facility is a multi-criteria decision problem that classical Health Technology Assessment, being mostly focused on medicines, vaccines and medical devices, cannot easily address.

Here we present a methodology based on Multi-Criteria Decision Analysis allowing a full assessment of non-clinical hospital services and supporting the selection of the most suitable solution in a certain environment.

The methodology involves two different panels of experts: the first one includes international professionals and is aimed at selecting the assessment criteria that are relevant to the target service; the second one is a local panel whose members know the needs and peculiarities of the specific healthcare facility. This approach allows the final decision makers to take into account changes and constraints of their environment, but examining criteria that are internationally recognized as of interest. The proposed methodology, tested in a real context of an Italian Local Health Authority, is versatile and can be applied in any context, even out of the healthcare domain, especially if data in the literature are not sufficient to allow comparisons with similar services in different settings.