The coauthors, Mary Beth Allen, MBA, PhD and Mehdi Mirsaeidi, MD, MPH, are my former colleagues from the University of Louisville and both have significant experience and expertise regarding infectious disease. Dr. Allen is a healthcare research consultant and former senior program manager for infectious disease at the University of Louisville. Dr. Mirsaeidi is currently an Associate Professor at the University of Miami Miller School of Medicine—Michael Mills.

There are many dimensions to an emergent pandemic such as COVID‐19. Leaders all over the world are grappling with the complexity of this disease as they consider the multifaceted burden it places on communities, healthcare delivery systems, financial security, and the role of government. Effective policy must address each and every dimension of this pandemic and do so early and decisively while also earning public confidence in order to ensure cooperation and compliance. While some policy makers are adept at the formation, implementation, and assessment of public policy, many rely on a myriad of resources for support through complex problems. Among these resources are the support of experts representing various dimensions of the problem situation, historical data from previous pandemics, observational data from other impacted regions, and the use of systems dynamics tools that predict the spread and impact of disease over time. System dynamics describes the process of representing a complex system with interrelated parts that interact in a nonlinear and unpredictable method within a system and predicting those interactions and outcomes. The process of mapping a system is described as system thinking, and this step alone can force its users to easily understand and predict outcomes in a complex system using mental modeling. The next step applies the map into a mathematical model through a system of algorithms to calculate the interactions and make predictions regarding future interactions and outcomes over time. System dynamics has many applications for both routine and unexpected problem situations and represents an important decision support tool that helps make determinations about the effectiveness of potential policy interventions prior to implementation. Modeling as a resource not only has many implications for the current pandemic, but also may play an important role in the dynamic field of medical physics.

Modeling has played a key role in the management of the COVID‐19 pandemic in many settings. By predicting the epidemiology of this disease, modeling has allowed policy makers, healthcare delivery systems, and other stakeholders to make short‐ and long‐term policy that addresses the emerging needs of systems on the local, state level, and national level. However, effective modeling is dependent upon accurate data along with the integration of valid assumptions. For this situation specifically, the rapid accumulation of data being generated worldwide has the potential to support better modeling, and making this timely data available for immediate use through open source publishing could be a huge benefit. With so many unknowns still associated with this particular coronavirus being a novel virus, any accumulation of data is key as so many questions remain regarding diagnostics, risk factors, disease presentation, transmission patterns, and treatment.

Access to timely data is also key as enacting policy early in a pandemic is an important component of its impact and outcomes. When it was clear that the virus had arrived to the US, it was incumbent on national and state leaders to act swiftly and decisively. Because limitations in the ability to administer adequate testing and isolate early cases left policymakers with few options, widespread social distancing became state and national policy that led to forced closures of major sectors of the economy. These policies had an immeasurable impact on the economy forcing immediate and record job loss, market volatility, and uncertainty regarding the future of many industries. While concerns for the economic system that provides financial security for so many remains an important consideration, American economists with few outliers have supported lockdowns as a means mitigating the loss of life and other burdens of disease that can have a lasting impact on the economy. Modeling using accurate data can not only have an impact on predicting the epidemiology of disease, but it can also be used to address other dimensions of this pandemic that may be just as critical to the health and safety of Americans.

Still, the vast majority of models have focused specifically on the epidemiology of disease. Several models predicting the epidemiology in the US have been released during different time periods in this pandemic that have driven national policy. The increased availability of data, changes in state policy and local responses, and access to new technology will continue to impact the outcomes of these models, which depend on timely information from worldwide sources. Among the earliest models released, a March 17 report from epidemiologist Dr. Neil Ferguson published by the Imperial College of London under the umbrella of the World Health Organization predicted that mortality for COVID‐19 at 2.2 million people in the US. This model relies heavily on data from the 1918 H1N1 pandemic, assuming a R₀ of 2.4 with symptomatic individuals 50% more contagious than asymptomatic individuals, an incubation period of 5.1 days, immunity following infection, and no widespread and enforced social distancing policy enacted (https://www.imperial.ac.uk/media/imperial‐college/medicine/sph/ide/gida‐fellowships/Imperial‐College‐COVID19‐NPI‐modelling‐16‐03‐2020.pdf). In contrast, models released by the US Coronavirus Task Force have provided much lower numbers, but each of these models has only calculated mortality as an outcome. The first model was released on March 30 and anticipated 200 000 deaths in the US. The only assumption that was shared as a basis for these predictions was strong and effective policy interventions nearing perfection in terms of adherence and reaching goals (https://www.washingtonpost.com/national/coronavirus‐deaths‐warning‐america/2020/03/30/522221ce‐72a6‐11ea‐87da‐77a8136c1a6d_story.html). Several updates to this initial model have been reported, the first of which was released on April 1 noting a much lower projected mortality of 90 000 (https://www.latimes.com/politics/story/2020‐04‐08/new‐data‐suggests‐u‐s‐deaths‐may‐be‐lower‐than‐feared), mortality at 60 415 through August 4, 2020 reported on April 8 (https://covid19.healthdata.org/united‐states‐of‐america), and rising to a range between 80 000 and 90 000 deaths with daily mortality of 3000 through June 1, (https://www.nytimes.com/2020/05/04/us/coronavirus‐updates.html). A May 4, 2020 updated model increased anticipated deaths to 134 000 in the US as a result of the relaxation of state‐level social distancing policies (https://www.cnn.com/2020/05/04/health/us‐coronavirus‐monday/index.html). This number was most recently increased to 147 000 by early August on May 12, 2020 (https://www.forbes.com/sites/mattperez/2020/05/12/coronavirus‐model‐used‐by‐white‐house‐now‐projecting‐147000‐us‐deaths‐by‐early‐august/#6c259abcb0f1). Among the major assumptions in these latest models is the relaxation of social distancing and the reopening of major segment of the economy as is occurring in most every state.

The utility of models extends well beyond generating an accurate prediction, as it is well understood that all models have their limitations. Models test the effectiveness of policy or calculate the rate of change in a variable, and they do so to provide support for policy. Among the limitations of these models used on the federal level is that the myriad of embedded assumptions, variables, and mathematical process used to build the model have not been made available. This has huge implications for the validity of the model, while also acting to exclude collaboration from experts within the academic and scientific community. The sharing of resources and information promotes the creation of better research, processes, policy, and outcomes. Also, the majority of these models only measure the outcome of mortality, as other key outcomes such geographical spread, number of cases, and hospitalizations have important implications for state and community level policy and resource planning. Furthermore, the timeliness of the data used as inputs for these models are unknown, but there is no question that immediately available open source could be an asset for this rapidly changing problem situation.

Using a systems dynamics approach to modeling and understanding the rapid spread of an infectious disease, such as COVID‐19, researchers from the University of Alicante, Spain created an open access model that is sufficiently complex, adaptable, and potentially valuable to policymakers all over the world. The publication provides an open access model published on the STELLA® platform (iseesystems), a widely adopted platform that allows users to customize the model to any area, and evaluate the behavior of the disease using a wide variety of projections, measured data, and scenarios. Although this publication is not currently peer reviewed, users of this technology will immediately appreciate the complexity and sophistication of the model and its potential to yield results that are meaningful and a source of significant utility. Additionally, since all components of the model may be viewed and are publicly available, the time to develop consensus and wide political support could be accelerated. As many regions and communities adopt the model, the data has the potential to be aggregated and applied as important policy support for major transitions in public policy (https://www.medrxiv.org/content/10.1101/2020.03.30.20047043v1).

Among the variables of the model, key inputs include the total population, the number infected, the number of contacts per day (R₀), the rate to symptomatic presentation, the asymptomatic rate, the rate of patients hospitalized, and the rate of ICU patients hospitalized. The R₀ is not a single number, but a graphical function over time, which is likely much more realistic and allows for more accurate predictions that calculate the impact of social distancing policy over time, as well as the impact of relaxing those measures. With accurate information as inputs, this model is capable of predicting not only mortality, but also other key outputs such as peak curves, case count, and hospitalization. The article uses official data from the Spanish Ministry of Health as input for calibration (https://www.mscbs.gob.es/en/profesionales/saludPublica/ccayes/alertasActual/nCov‐China/home.htm).

Although this model has important utility it is likely less complex than models used by major governments, international organizations, and prestigious academic institutions. For this reason, it is well suited for regional and community level adoption, which in the US remains the governmental institution involved in creating and enforcing pandemic response policy. Like all models, the utility will improve as the variables used as inputs become more accurately measured or defined. In the US, this will depend on the continued availability of diagnostic testing in both the outpatient and hospital settings along with accurate antibody testing. Also, public health functions and continued surveillance such as contact tracing might result in much more accurate numbers, including better estimates of R₀ over time. The authors are careful to remind readers that models are inherently wrong and exist less as a means of predicting outcomes, but more as a utility for policy makers. The authors also emphasize their open access format as an important strength, which makes this resource available to potentially anyone as an adaptable resource that can be applied to a variety of settings. Although some familiarity with STELLA software is required to adapt the model, this should not be a major challenge for those comfortable with other modeling programs. The authors are currently developing a video tutorial as a resource to support the adaption of this model in order to address any specific barriers to this software. Also, the authors of this paper feel strongly about the importance of social distancing policy relative to the COVID‐19 pandemic, encouraging that this policy be maintained over time. This recommendation stands despite skeptics who erroneously attribute this policy to undue economic hardship. Sweden became an interesting test case for an alternative to forced closure of the private business, and the outcomes were a case mortality that was remarkably higher than similar populations (https://www.spiked‐online.com/2020/04/22/there‐is‐no‐empirical‐evidence‐for‐these‐lockdowns/). But, the evidence that social distancing policy had reduced transmission in many settings continues to accumulate. A study conducted at the University of Kentucky found that the aggressive Healthy At Home initiative, which has resulted in Kentucky ranking among the highest in the US in terms of Coronavirus response, is believed to have saved as many as 2000 lives in the commonwealth so far, with less than 200 deaths having been reported at that time (https://www.kentucky.com/opinion/linda‐blackford/article242367161.html).

The need for modeling will continue to be an important resource as new data continues to emerge and social distancing policies relax across the world. Also, the experiences from different settings will become data points in an algorithm. Communities in Italy and Spain were among the earliest impacted regions in Europe, but the late to arrive the UK has currently surpassed these early hot spots in terms of case count and mortality despite having more data. Meanwhile, Germany over prepared with early lock downs, initiated widespread testing, and an excess of hospital capacity shared with its neighbor by taking Italian patients. In the US, New York was the epicenter for the American burden with communities in Florida and Illinois on the rise. Also, state leadership of lockdown policy has resulted in a patchwork of lockdowns targeting different industries with different timelines for reopening, which will provide rich data for analysis. Creating a mechanism to share this data in real time through open source publishing could make this analysis available sooner to support the evaluation of current policy rather than historic analysis.

Medical science depends on the proliferation, dissemination, and collaboration of knowledge. Increasing the availability of timely data though open source publishing has the potential to escalate this process, which has the potential to benefit everyone. This process begins with individual sectors of the medical community making available to one another and related sectors valuable and emerging research that is relevant to the practice and the evolution of a discipline. Also, despite the specialization and fragmentation of medicine, major developments from one discipline can have a resounding impact on many other fields of practice. Furthermore, resources developed in related fields and from industry have found adoption in the field of medical physics. System dynamics modeling is no exception as it has been adopted specifically in the economics of medical physics, and has been noted to predict the supply and demand of radiation oncology physicists with accuracy (https://aapm.onlinelibrary.wiley.com/doi/10.1120/jacmp.v11i2.3005). In regards to the COVID‐19 pandemic, it is safe to state that no sector of the economy, much less no sector of the medical economy, will remain insulated from the impact likely for years to come. As modeling may predict the distribution and outcomes of disease in care settings where we practice, we will be able to adapt these models in a variety of ways as our roles have evolved substantially within hospitals, universities, and our communities. The unique disease curve for each of us will not be congruent, and we too can use modeling to make calculations regarding the needs for policy interventions relative to our unique care delivery systems. This might be especially critical for our medically fragile patients who will depend on us to provide for their safety. For this reason alone, we have a responsibility to anticipate challenges to the healthcare system and communicate recommended interventions to our elected leaders, hospital administrators, and other important community level leaders. By acting early, we can be not only supporting our patients, but also acting in support of our leaders and our communities. We are also obligated to protect ourselves as essential healthcare workers and our employees who act in support of essential care delivery. Identifying our own risks and exposures, creating policy to undermine that risk, and calculating the optimal timing of changes in policy are all important interventions we can make on the community level that can be supported by modeling. Furthermore, as the current crisis concludes, perhaps the adoption of modeling skills can serve our field in the future. Systems dynamics can be used to model daily practice and its many disruptions, which could include minor changes in reimbursement or payer mix or the impact of new technology as is common in our profession. If we gain the ability to adapt systems dynamics, there is no end to the applications of this technology as well as the daily application of systems thinking, which describes a mental model that emulates the technology by analyzing and predicting the interaction of interrelated parts of a system over time. Healthcare delivery is dynamic, but the technology rich and constantly evolving field of medical physics is arguably among the most dynamic in all of healthcare, which implies that systems dynamics may have an important role in the field of medical physics. As we remain collaborative with other fields, most notably though the proliferation of open access formats becoming more common in other fields of medicine, basic science, social science, and industry, it is likely that systems dynamics may represent one of several opportunities for medical physics to advance.

合著者是玛丽·贝思·艾伦（Mary Beth Allen）MBA博士和医学博士（Mehdi Mirsaeidi），医学博士（MPH），我是路易斯维尔大学的前任同事，在传染病领域都拥有丰富的经验和专业知识。艾伦博士是一名医疗研究顾问，曾任路易斯维尔大学传染病高级计划经理。Mirsaeidi博士目前是迈阿密米勒大学医学院的副教授，迈克尔·米尔斯。

紧急大流行有很多方面，例如COVID-19。全世界的领导人正在考虑这种疾病的复杂性，因为他们考虑到该疾病给社区，医疗服务系统，财务安全以及政府的角色带来了多方面的负担。有效的政策必须解决这一大流行的各个方面，并尽早果断地采取行动，同时还要赢得公众的信任，以确保合作与合规。尽管某些决策者擅长公共政策的形成，实施和评估，但许多决策者依靠大量资源来解决复杂问题。这些资源包括代表问题情况各个方面的专家的支持，以前大流行的历史数据，其他受影响地区的观测数据，以及使用系统动力学工具来预测疾病随时间的传播和影响。系统动力学描述了用相互关联的部分表示复杂系统的过程，这些部分在系统内以非线性且不可预测的方法进行交互，并预测了这些交互作用和结果。映射系统的过程称为系统思考，仅此步骤便可以迫使其用户使用心智模型轻松理解和预测复杂系统中的结果。下一步，通过算法系统将地图应用到数学模型中，以计算交互作用，并对未来的交互作用和结果做出预测。系统动力学对于常规和意外问题情况都有许多应用程序，并且代表重要的决策支持工具，可帮助在实施之前确定潜在的政策干预措施的有效性。作为一种资源进行建模，不仅对当前的大流行有很多影响，而且可能在医学物理学的动态领域中发挥重要作用。

在许多情况下，建模在COVID-19大流行的管理中发挥了关键作用。通过预测这种疾病的流行病学，建模使政策制定者，医疗保健提供系统和其他利益相关者可以制定短期和长期策略，以解决地方，州和国家级系统的新兴需求。但是，有效的建模取决于准确的数据以及有效假设的整合。特别是对于这种情况，在全球范围内快速生成的数据积累有可能支持更好的建模，并且使这些及时的数据可通过开源发布立即使用可能会带来巨大的好处。由于这种特殊的冠状病毒仍然是新型病毒，因此还有许多未知数，

及时获得数据也是关键，因为在大流行早期制定政策是其影响和结果的重要组成部分。当很明显该病毒已经到达美国时，国家和州领导人就有责任迅速果断地采取行动。由于管理足够的测试和隔离早期病例的能力受到限制，决策者几乎没有选择余地，因此广泛的社会隔离成为州和国家的政策，导致经济的主要部门被迫关闭。这些政策对经济产生了不可估量的影响，迫使立即和创纪录的工作流失，市场动荡以及许多行业未来的不确定性。尽管对为许多人提供金融安全的经济体系的担忧仍然是重要的考虑因素，很少有离群值的美国经济学家支持封锁，以减轻可能对经济产生持久影响的生命损失和其他疾病负担。使用准确的数据进行建模不仅会影响疾病的流行病学预测，而且还可用于解决这一流行病的其他方面，这些方面可能对美国人的健康和安全至关重要。

尽管如此，绝大多数模型还是专门针对疾病的流行病学。在该流行病的不同时期发布了几种预测美国流行病学的模型，这些模型推动了国家政策的制定。数据可用性的提高，国家政策的变化和地方对策以及对新技术的获取将继续影响这些模型的结果，这些模型的结果取决于来自全球的及时信息。在最早发布的模型中，伦敦帝国学院在世界卫生组织的保护下于3月17日发表的流行病学家尼尔·弗格森（Neil Ferguson）博士的报告预测，美国的COVID-19死亡率为220万人。该模型严重依赖1918年H1N1大流行的数据，假设R ₀2.4的有症状个体比无症状个体传染性高50％，潜伏期为5.1天，感染后具有免疫力，并且没有制定广泛而强制的社会疏远政策（https://www.imperial.ac.uk/media/imperial- College / medical / sph / ide / gida-fellowships / Imperial-College-COVID19-NPI-modelling-16-03-2020.pdf）。相比之下，美国冠状病毒特别工作组发布的模型提供的数字要低得多，但是这些模型中的每一个都仅计算出死亡率作为结果。第一个模型于3月30日发布，在美国预计有20万人死亡。作为这些预测基础的唯一假设是，在遵守和实现目标方面，有力而有效的政策干预措施已接近完美（https：//www.washingtonpost。com / national / coronavirus-deaths-warning-america / 2020/03/30 / 522221ce-72a6‐11ea‐87da‐77a8136c1a6d_story.html）。已经报告了对该初始模型的一些更新，其中第一个更新于4月1日发布，指出预计的死亡率要低得多（https://www.latimes.com/politics/story/2020-04-08/new数据建议的死亡率可能低于恐惧水平），4月8日报告的死亡率为2020年8月4日的60415（https://covid19.healthdata.org/united-states-美国），到6月1日为止的死亡人数上升到8万至9万，每天死亡3000（https://www.nytimes.com/2020/05/04/us/coronavirus-updates.html ）。2020年5月4日更新的模型由于放松了州一级的社会隔离政策而使美国的预期死亡人数增加到134 000（https：//www.cnn。com / 2020/05/04 / health / us-coronavirus-monday / index.html）。到2020年5月12日，这个数字最近增加到147 000（https://www.forbes.com/sites/mattperez/2020/05/12/coronavirus-model-used-white-house-现在正计划在8月初/＃6c259abcb0f1死亡。在这些最新模型中，主要的假设之一是放宽了与社会隔离的距离，并且在大多数州都发生了经济主要部分的重新开放。

模型的效用远远超出了生成准确的预测的范围，因为众所周知，所有模型都有其局限性。模型可以测试政策的有效性或计算变量的变化率，从而为政策提供支持。在联邦一级使用的这些模型的局限性在于，尚无用于构建模型的大量嵌入式假设，变量和数学过程。这对模型的有效性具有巨大的影响，同时也排除了学术界和科学界专家的协作。资源和信息的共享促进了更好的研究，流程，政策和成果的创建。此外，这些模型中的大多数仅衡量死亡率的结果，其他重要结果，例如地域分布，病例数和住院情况，对州和社区一级的政策和资源规划都具有重要意义。此外，用作这些模型的输入数据的及时性是未知的，但是毫无疑问的是，对于这种迅速变化的问题情况，立即可用的开源可能是一项资产。

西班牙阿利坎特大学的研究人员使用系统动力学方法来建模和了解传染病的快速传播，例如COVID-19，西班牙阿利坎特大学的研究人员创建了一个开放访问模型，该模型足够复杂，适应性强，对整个决策者都有潜在价值世界。该出版物提供了在STELLA®平台（iseesystems）上发布的开放访问模型，该平台是一种广泛采用的平台，它使用户可以针对任何区域定制模型，并使用各种预测，测量数据和评估疾病的行为。场景。尽管当前尚未对该出版物进行同行评审，但该技术的用户将立即意识到该模型的复杂性和复杂性及其产生有意义的结果和巨大实用性的潜力。另外，由于模型的所有组成部分都可以查看并且可以公开获得，因此可以加快形成共识和广泛政治支持的时间。由于许多地区和社区都采用该模型，因此数据有可能被汇总并用作公共政策重大转变的重要政策支持（https://www.medrxiv.org/content/10.1101/2020.03.30.20047043v1）。

在模型的变量中，关键输入包括总人口，感染人数，每天的接触人数（R ₀），有症状的出现率，无症状的发生率，住院患者的发生率以及ICU发生率患者住院。R ₀不是一个单一的数字，而是一个随时间变化的图形函数，它可能更现实，并且可以进行更准确的预测，以计算随着时间的推移社会疏远政策的影响以及放宽这些措施的影响。以准确的信息作为输入，此模型不仅可以预测死亡率，而且还可以预测其他关键输出，例如峰值曲线，病例数和住院情况。本文将西班牙卫生部的官方数据用作校准的输入（https://www.mscbs.gob.es/en/profesionales/saludPublica/ccayes/alertasActual/nCov‐China/home.htm）。

尽管此模型具有重要的实用性，但它可能不如主要政府，国际组织和著名的学术机构所使用的模型复杂。因此，它非常适合区域和社区一级的采用，在美国，它仍然是参与制定和执行大流行应对政策的政府机构。与所有模型一样，该实用程序将随着用作输入的变量变得更准确地测量或定义而得到改善。在美国，这取决于门诊和医院环境中诊断测试的持续可用性以及准确的抗体测试。此外，公共卫生职能和持续监视（例如接触者追踪）可能会导致数字更加准确，包括对R _0的更好估算随着时间的推移。作者谨在此提醒读者，模型天生就是错误的，其存在较少作为预测结果的手段，而更多地是作为决策者的工具。作者还强调了他们的开放式访问格式是一种重要的优势，这使得该资源可以作为可能适用于多种设置的适应性资源供任何人使用。尽管需要熟悉STELLA软件才能适应模型，但对于那些熟悉其他建模程序的人来说，这应该不是主要挑战。作者目前正在开发视频教程作为资源，以支持该模型的改编，以解决该软件的任何特定障碍。此外，本文的作者强烈地感到，与COVID-19大流行相比，社会疏远政策的重要性，鼓励随着时间的推移保持这一政策。尽管怀疑者错误地将此政策归因于过度的经济困难，但该建议仍然有效。瑞典成为强制关闭私人企业的替代方法的一个有趣的测试案例，其结果是案例死亡率显着高于类似人群（https://www.spiked-online.com/2020/04/22/这些锁定没有经验证据。但是，越来越多的证据表明，社会疏远政策已经减少了许多地方的传播。肯塔基大学进行的一项研究发现，积极进取的“健康在家”计划已使肯塔基州在冠状病毒反应方面名列美国最​​高之列，迄今已挽救了多达2000条生命。 ，

随着新数据的不断涌现以及社会隔离政策在全球范围内的放松，对建模的需求将继续成为重要资源。同样，来自不同设置的体验将成为算法中的数据点。意大利和西班牙的社区是欧洲受灾最早的地区之一，尽管数据较多，但到达英国较晚的人数和死亡率目前已超过这些早期热点。同时，德国已做好充分准备，尽早锁定，开始进行广泛的测试，并通过接收意大利患者与邻国共享过多的医院容量。在美国，纽约是佛罗里达州和伊利诺伊州居民数量上升的美国负担的中心。也，国家锁定政策的领导层导致针对不同行业，具有不同重开时间表的锁定拼凑而成，这将为分析提供丰富的数据。建立一种通过开源发布实时共享这些数据的机制，可以使这种分析更早地用于支持当前政策的评估，而不是历史分析。

医学取决于知识的扩散，传播和协作。尽管开源发布可以提高及时数据的可用性，但有可能使这一过程升级，并有可能使所有人受益。这一过程始于医学界的各个部门，彼此相关的部门以及与该学科的实践和发展相关的有价值的新兴研究。同样，尽管医学专门化和分散化，但一门学科的重大发展可能对许多其他实践领域产生巨大影响。此外，在相关领域和工业中开发的资源已经在医学物理学领域得到采用。系统动力学建模也不例外，因为它已在医学物理学经济学中被特别采用，并且已被注意到可以准确地预测放射肿瘤学物理学家的供需（https://aapm.onlinelibrary.wiley.com/doi/ 10.1120 / jacmp.v11i2.3005）。关于COVID-19大流行，可以肯定地说，任何经济领域，尤其是医疗经济领域，都不会在未来几年受到影响。由于建模可以预测我们所护理的医疗机构中疾病的分布和结局，因此随着我们在医院，大学和社区中的角色已发生重大变化，我们将能够以各种方式适应这些模型。对于我们每个人来说，独特的疾病曲线不会完全一致，而且我们也可以使用模型来计算与我们独特的护理提供系统相关的政策干预需求。这对于我们将依靠我们提供安全保障的医学上脆弱的患者尤其重要。仅仅出于这个原因，我们有责任向预期的卫生保健系统的挑战和沟通推荐干预措施，以我们的民选领导人，医院管理人员，以及其他重要的社区级领导。通过尽早采取行动，我们不仅可以为患者提供支持，还可以为我们的领导者和社区提供支持。我们还有义务保护自己作为基本的医疗保健工作者以及我们为支持基本医疗保健提供服务而行动的员工。识别我们自己的风险和风险，制定破坏该风险的政策，以及计算政策变更的最佳时机都是我们可以在社区级别上进行的重要干预，并且可以通过建模来支持。此外，随着当前危机的结束，也许采用建模技能可以在将来为我们的领域服务。系统动力学可用于模拟日常操作及其许多干扰，其中可能包括报销或付款人组合的微小变化，或者是本行业中常见的新技术的影响。如果我们具有适应系统动态的能力，那么该技术的应用以及系统思维的日常应用将无止境，它描述了一种通过分析和预测部件的相关部分之间的相互作用来模拟该技术的思维模型。系统随着时间的推移。医疗保健服务是动态的，但是，医学物理学领域技术的不断发展和不断发展，可以说是所有医疗领域中最动态的，这意味着系统动力学可能在医学物理学领域中发挥着重要作用。随着我们与其他领域保持合作，最显着的是，尽管开放访问格式的扩散在医学，基础科学，社会科学和工业的其他领域变得越来越普遍，但系统动力学可能代表了医学物理学的众多机遇之一推进。