Healthcare Decision-Support by Hybrid Simulation – Mobile Stroke Units

Problem




To understand both the medical and economic impact of new healthcare technologies, they must be evaluated before the design and development phase begins. Therefore, decision support system approached through hybrid simulation is applied in the case of Mobile Stroke Units (MSUs).


Stroke causes severe disability, produces high costs for care and rehab, and its incidences are increasing due to an ageing population. Thrombosis causes most strokes and if possible, should be treated with thrombolysis (not in case of haemorrhagic strokes and after 4.5 hours have been elapsed). Currently, the process of transportation and internal hospital administration causes the patient to lose valuable time. 


MSUs have been suggested as a possible improvement. MSUs start with diagnostics and therapy steps at a stroke occurrence location. The purpose is to reduce the alarm-to-therapy-decision-time in order to prevent severe disabilities of people and high costs.


Objectives




The goal is to assess the medical and economic impact of MSUs in comparison to the current situation and optimize the geographic distribution. Main metrics are the impact on the alarm-to-therapy-decision-time, costs, and other relevant output parameters. A metropolitan area (Berlin) and a rural region in Germany are used for comparison since there are few hospitals, and they are different structures.

MSU
Figure 1: Illustration of a Mobile Stroke Unit.


MSU Simulation

Figure 2: MSU scenario animation screenshot

Implementation in AnyLogic




The project has been realized with AnyLogic; that is well suited to develop models using all necessary simulation paradigms and enables the developer to add own Java code to model unique and custom problem solutions. AnyLogic also allows the creation of illustrative animations in order to enhance communication with domain experts.


Figure 2 shows a screenshot of the MSU scenario animation. The MSUs are located on predefined positions and can be sent by the dispatcher when a stroke case occurs. If no free vehicle is available, a usual rescue service will be sent to bring the patient to the next hospital. Wrong decisions and unnecessary MSU missions (that do occur in reality) are modeled as well as cost values for the various actions and interventions. Stroke patients are monitored until 10 years after affection to enable long term cost analyses.


Excel tables with varied parameters are used as input files for different years and are loaded during simulation runs. To evaluate multiple scenarios of healthcare decision-support a library has been developed. Different modules (e.g., demography, error injector) can be connected in order to add certain characteristics to a dedicated scenario. A positioning tool allows defining home locations for MSUs and hospitals on a map by clicking on the area before running a scenario. In the selected papers below, further modeling and implementation features can be found.


Modeling Approach




To perform a simulation and particularly within the scope of healthcare decision-support, macro-simulation as well as micro-simulation approaches are necessary to increase model accuracy. This is where one can benefit from multi-method and hybrid simulation paradigms. System dynamics is used for modeling at a distant perspective, while discrete-event and agent-based techniques are appropriate for detailed modeling at individual levels. Patients are represented by agents, and medical workflows are embedded in system dynamics models which represent aspects as demography, economy, and epidemiology. The impact of medical technologies is represented by quantitative parameters.

MSU Models

Figure 3: Agent-Based, Discrete-Event and System Dynamics Model examples


Outcomes




One significant result from this case study is that MSUs do not automatically lead to more patients with thrombolysis as a treatment. However, those who are treated with thrombolysis receive care earlier, resulting in reduced probability for severe disabilities. This is a clear medical benefit. Research also shows that a broad distribution (e.g., uniform distribution) of MSUs on a map leads to better results in contrast to centralization on few locations (e.g., station, hospital). Furthermore, the simulation has shown that, in rural regions in Germany with only few hospitals, such vehicles are not profitable, as there are not many affections per year, and most of the people live close to an urban center with a hospital able to treat stroke. The above case study results may differ in countries with less specialized hospitals.


Conclusions




AnyLogic allows developing detailed models for healthcare decision-support. The configurable MSU simulation model helped the modelers to answer important questions about the impact of medical and health economics. Regulatory agencies, companies, researchers and other decision-makers can use the results to optimize an MSU roll-out in any country and to improve the stroke diagnostics and treatment in the future. Long-term monitoring of stroke patients allows the comparison of saved costs to the additional costs of MSUs, leading to a basis for an investment decision. This work has been conducted by Prospective HTA (funded by the German Government) in conjunction with doctors, engineers, and health economists, from both industry and academia.


The simulation of large-scaled complex systems (e.g., in healthcare, automotive, industry, and energy) is a major field of work at the University of Erlangen-Nuremberg Computer Networks and Communication Systems Group. Please contact model developers, if you have further questions.


Selected Publications




  1. Djanatliev A. and German R. “Prospective Healthcare Decision-Making by Combined System Dynamics, Discrete-Event and Agent-Based Simulation”. Proceedings of the 2013 Winter Simulation Conference, Editors: R. Pasupathy, S.H. Kim, A. Tolk, R. Hill, M. Kuhl. Washington D.C./USA. December 8th – 11th, 2013, p. 270-281.
  2. Djanatliev A., Kolominsky-Rabas P., Hofmann B. M., Aisenbrey A., German R. “System Dynamics and Agent-Based Simulation for Prospective Health Technology Assessments”. Simulation and Modeling Methodologies, Technologies and Applications - Advances in Intelligent Systems and Computing. Edited by Obaidat, Mohammad; Filipe, Joaquim; Kacprzyk, Janusz; Pina, Nuno (Eds.). Volume 256, 2014, pp 85-96.
  3. Djanatliev A., German R. “Large Scale Healthcare Modeling by Hybrid Simulation Techniques using AnyLogic”. Proceedings of the 6th International ICST Conference on Simulation Tools and Techniques. Cannes/French Riviera. March 5th – 7th, 2013, p. 248-257.

More Case Studies

  • Modeling Operations at Pharmaceutical Distribution Warehouses
    Cardinal Health, a billion dollar pharmaceutical distribution and logistics firm, manages multiple products from brand name pharmaceuticals and generic drugs to over the counter drugs, health & beauty items and their own private label. They face a multitude of typical distribution warehouse challenges that are further complicated by the nature of pharmaceutical products. Brian Heath, Director of Advanced Analytics at Cardinal Health, and an experienced user of AnyLogic software, employed agent based modeling to solve various business problems, saving Cardinal Health over $3 Million annually.
  • Simulation of Maternity Ward Operations
    This model simulates the maternity ward in a hospital currently under construction. The purpose of the model is to support discussions related to which resources, capacity, and work methods are required on the new ward. The project was carried out for Karolinska University Hospital in the Stockholm County, Sweden.
  • Evaluating Hospital Inpatient Care Capacity
    Stockholm County, Sweden was in the process of building a new, highly specialized hospital. The Health Administration of the county questioned whether they would get an acceptable level of care production with the current investments and reasoning concerning various operational and strategic issues. To find the answers, they used simulation modeling in AnyLogic.
  • Handling Total Care Need for Dialysis Patients
    The County of Stockholm (Sweden), like any country or region, experiences a continuous need to handle the healthcare necessities of various patient groups. Each group can be seen as a subpopulation, with its own distinctive traits, characteristics, and challenges. The discussed simulation project focused on the dialysis patients, a group who needs to visit caregiving facilities frequently.
  • Disaster Response Applications Using Agent-Based Modeling
    In an effort to find practical operational solutions for response to an unexpected crisis or natural disaster, Battelle, world’s largest, non-profit, independent R&D organization, needed to test the effectiveness of a 48 hour shelter-in-place order for an Improvised Nuclear Device scenario. The goal was to reduce radiation dosages received during an uncoordinated mass evacuation, by comparing immediate evacuation and shelter-in-place order.
  • Evaluating Healthcare Policies to Reduce Rates of Cesarean Delivery
    The challenge of reducing the cesarean delivery rate has been recognized by numerous researchers for years. For the first time, in research conducted for the Washington State, Alan Mills, FSA MAAA ND, a research actuary, and his colleagues reproduced this part of the United States healthcare system in a simulation model to allow the stakeholders, including health agencies, insurers, clinicians, and legislators, to test their assumptions on the model to find the right solutions.
  • Shaping Healthcare Policy Using Simulation
    An initiative by the Department of Mechanical and Industrial Engineering at the University of Toronto, the Centre for Research in Healthcare Engineering (CRHE), was in response to the immediate and compelling desire for efficiency and quality improvements in the Canadian healthcare system.
  • An Agent-Based Explanation for SPMI Living Situation Changes
    Over the past 60 years, the number of Severely and Persistently Mentally Ill (SPMI) patients in the US living in the community increased. Yet a growing minority of people with severe illness are worse off because they are homeless or incarcerated. In this case study, IBM Global Research and Otsuka Pharmaceuticals used an agent-based approach to model these remarkable swings.
  • Outpatient Appointment Scheduling Using Discrete Event Simulation Modeling
    Indiana University Health Arnett Hospital, consisting of a full-service acute care hospital and a multispecialty clinic, faced poor statistics because the number of no-show patients (those who don’t show up for their scheduled appointments) rose dramatically to 30%. This was primarily connected to the fact that clinic schedules were driven by individual preferences of the medical staff, which led to increased variations in scheduling rules. To eliminate the problem, the client wanted to develop a scheduling methodology that would benefit the clinic, doctors, and patients.
  • Modeling of a Pharmaceutical Product Launch
    One of the huge pharmaceutical companies employed Bayser Consulting for development of product launch strategy. Simulation modeling was applied for reconstruction of interactions between the company, doctors and patients.