Amass casualty incident (MCI)refers to any incident in whichemergency medical service
resources, such as personnel and equipment, are overwhelmed by the large number and severity of casualties within a given community.1
For instance, the 2008 Wenchuan earthquake in Sichuan, China was one of the most devastating disasters in the past 10 years and caused more than 370 000 casualties, 69 000 deaths.2
The 2010 Haiti earthquake disaster devastated the city with 222 750 reported deaths, 300 000 people injured, 1.5 million displaced, and more than 3 million affected.3
The Great East Japan earthquake caused more than 12 000 deaths, 15 000 people missing.4
In the 2013 Lushan earthquake in Sichuan, China, 196 were killed and 13 484 were injured5
Compared to the conventional casualty incidents, MCI has three characteristics.6
large number of casualties causes a lag between available resources and medical needs. Secondly, the common shortage of staff, rescue facilities, and logistical supplies further lengthens the lag. Thirdly, rescue circumstances surrounding a given event are far beyond the capacity of medical response. MCI response system as a complex system should be synthetically considered from multiple angles and in multiple levels, and should take into account multiple factors. The medical management of casualties in MCI alone includes four links, which are search and rescue, triage and initial treatment, definitive treatment and evacuation.7
The support system of MCI response involves even more subsystems within the fully integrated emergency management system. Many scholars showed their interest in disaster research, but no review of research methods of MCI response has been publicized so far. The focus of this article is on the review of the research methodology of MCI response. The main task of this paper is to explore the new direction of complexity science applied to the field of MCI response.
Conventional method of MCI response research
Theoretical research approach to MCI response
It is essential for MCI response study to be based on theoretical research. A theoretical research approach means formal logical deduction of conclusions from a set of initial assumptions. If the axioms are true and the rules are logically sound, the conclusions are true as well. There are different forms of theoretical approaches. One can draw new ideas by derivation, summarization, review and analysis. The theories, assumptions, conceptual or analytical frameworks need to be further validated. The practices for disaster response can be summarized and analyzed, and scientific issues stemmed from such practices can be further explored. It is usually presented in the form of review or commentary.There are both distinctive advantages and disadvantages in the current theoretical research method. The main advantage of this method is that the deduction process can incorporate empirical researches and case studies. The limitation is that the relationship among the variables cannot be clarified or supported accurately by actual data and the scientific rules of MCI response may not be explored in-depth. In the realm of theoretical research, the representative theoretical model is Haddon matrix and the representative theory is surge theory. Haddon matrix is an important theoretical framework to explore emergency response mechanisms in three periods, before, during, and after an accident, as well as three dimensions, host, factors (or carrier/media) and the environment of an accident. It is a basic method for data processing. However, it is insufficient in identifying cause-and-effect relationship between the externalphenomena and internal mechanisms of accidents.8Haddon matrix has been used to analyze medical emergency relief on theJuly 7, 2005London bombings.9In recent years, surge theory has been proposed based on the theoretical core of surge and surge capacity. In the aspect of surge capacity, Hick et al proposed the CO-S-TR conceptual model, which is divided into three sub-systems of incident management, logisticsmanagement and casualty management. Each of these sub-systems consists of four sub-elements.10Furthermore, Hick et al proposed the refined surge capacity model involving conventional, contingency and crisis capacities.11Currently, surge theory has been gradually demonstrating its application prospects.12,13
Empirical approach to MCI response
Empirical approach is a basic research method of MCI response. Previous empirical researches on MCI response are usually based on typical case studies, and are effective in exploring scientific research themes. Empirical research can basically describe incidents and may also find scientific rules. The only problem is that case studies are often limited in quantity and are insufficient in research data. Such studies can not fully derive scientific rules. Yet case studies on MCI response have been conducted widely regardless of many of its shortcomings.
Case study as a most influential empirical approach has been used to conduct research in many typical aspects of MCI. Data collection on emergency medical MCI response generally has to be retrospective. For example, health needs, casualty statistics, environmental health, mental health, casualty of rescue personnel, long-term effects and community health needs during the incident can be accessed by on-site investigation in the attacks of September 11, 2001.14In the retrospective survey of the 2005 July London bombings,the data of casualty statistics, outcome, triage, patient flow and resources have been observed from pre-hospital and in-hospital information. This survey has shown that wounded mortality was not related to over-triage and was only associated with the rapid and advanced emergency management system.15An epidemiological investigation and preliminary analysis of injury and death cases in the Typhoon Rananim 2004 in Wenling City has shown the main types and characteristics of injuries and fatalities caused by typhoon. Further, the relationships between the occurrence of typhoon-related injuries and wind speed and rainfall amount during typhoonhave been explored. Injury rates during typhoon were not associated with the amount of rainfall. It has been noted that the peak time of death or injury was before or during the landfall of typhoon.16After the 2008 Wenchuan earthquake in Sichuan, China, a systematic retrospective analysis has been conducted by Chinese scholars to investigate the command system, deployment of medical rescue force, medical treatment, public health response for infectious disease control, and health counterpart assistance for emergency medical rescue in the disaster. It has been proposed thatcrucial factors for reducing mortality and morbidity and promoting overall effectiveness of rescue efforts after a major earthquake involve the establishment of a national disaster medical response system, an active and effective commanding system, successful coordination between rescue forces and government agencies, effective treatments, a moderate, timely and correct public health response system, and long-term psychological supports.17In the retrospective survey of the Boston Marathon attack two blasts ripped through the crowd that was gathered along the approach to the finish line, killing 3 people and injuring more than 260 at 2:50 p.m. on April 15, nearly 3 hours after the first runner completed the Boston Marathon. It has been proposed that the remarkably low mortality rate (1%) of the attack was attributable to six factors: a large number of police, security and EMS personnel deployed, less than full capacity of the city’s operating rooms and other clinical services required, a full complement of care providers on site at each facility, seven trauma centers and multiple world-class hospitals in Boston, wise distribution of casualties among the area’s trauma centers and the swift extrication of victims in the absence of structural collapse.18
Evidence-based science in MCI response
Knowledge-based on systematically collected data from MCI response studies might help planners avoid common pitfalls of disaster management, thereby improving MCI response planning.19Evidence-based science isthe process of implementing evidence into practice. The importance of evidence-based disaster planning is highlighted recently. Systematic review is the classical method with comprehensive collection ofrelevant studieswhich were comprehensively analyzed and evaluated, and necessary meta-analysis can be used to draw general conclusions.20,21The advantage of evidence-based science is to minimize bias and to approach scientific facts as close as we can by usingdata collection. On the other hand, the limitation of the existing literature related to emergency and disasters is a lack of impact and extrapolation power of conclusions drawn from systematically collected evidence.22The evidence-based science might be associated with selection bias which may occur when the number of studies enrolled is small. In addition, many research reports are not published in peer-reviewed journals but rather in reports published by government agencies or academic institutions. The methodological and report quality of included reviews has not been well evaluated. For instance, two cohort, two case-control and four cross-sectional studies were included in a systematic review to access and identify risk factors related to deaths and injuries in earthquakes. The methodological quality of the included cohort and case-control studies was assessed, and the potential risk factors of earthquake related deaths and injuries were systematically enumerated.It had drawn a main conclusionthatearthquake casualties are determined by a number of factors including demographic characteristics, architectural vulnerability factors,earthquakes andgeographical factors. The conclusion of this review may be limited by the potential selection bias of the included studies and the regional characteristics of the included populations. It may decrease the extrapolation of research results and need more data to support the conclusion of risk factors of seismic casualty inChina quake zone.23Evidence-based health policy-making approach is essential for MCI response. SUPPORT Tools for evidence-informed health Policymaking (STP) as a method for producing high quality evidence in decision making have been usedforcomprehensive intervention for complex policy problems.24
Mathematical modeling approach to MCI response
Since the occurrence of MCI is infrequent, non-replicable and the experiments of MCI is dangerous,mathematical modeling approach is the recent trend in MCI responseresearch. The simulated MCI response exercisesmainly include desktop analog simulation, functional simulation, field simulation and so on. The first two simulations can produce significant distortions due to their lack of dynamic changes. Field simulation is of high cost, and moreover it is impossible to simulate all possible scenarios and hard to predict and analyze results under its assumptions. Math- ematical modelingcan remedyits shortcoming to a certain extent and has been widely accepted as an effective method for MCI management. A benefit of this methodology is that conditions in simulated systems can be modified and manipulated for research purpose, which can not be done to existing systems in reality. But it is difficult to give a brief description of complex interactions among elements in MCI response system. To illustrate complex processes through mathematical modeling, the simplified models used have to remove many important factors that happen in realityand thus are significantly different from the actual situation. Therefore, studyconclusions are limited in their application. Hupert has constructed a simulation model of mass casualty trauma care to examine the relationship between the rate of overtriage and the critical mortality rate of survivors. Model variables include triage performance, treatment capability, treatment time, and time-dependent mortality of critically injured patients. Study has shown that the ratio of critical patients to treatment capability has a more crucial impact on critical mortality than overtriage level or time-dependent mortality assumption in all scenarios.25Lerner et al evaluated the effectiveness of SALT (sort, assess, life-saving interventions, treatment and/or transport) mass casualty triage in a virtual MCI and found that SALT triage was accurate and quick in assessment during a simulated incident. The accuracy rate was higher than those published for other triage systems of similar speed.26In view of the confirmed cases of H7N9 virus infections in eastern China, Chinese scientists developed a diffusion model to spatiotemporally characterize the impacts of bird migration and poultry distribution on the geographic spread of H7N9 infection. With reference to the reported infection cases, the demonstrated risk mapping results will provide guidance in the active surveillance and control of human H7N9 infection by taking intensive intervention in poultry markets.27Previously, Steven studied the impact of public health interventions on the transmission dynamics of the etiological agent of SARS in Hong Kong with mathematical models and simulation techniques.28In applications of seismic modeling of making decisions, Chinese scholars conducted a study on an intelligent decision support system for prediction of catastrophic earthquake casualties. The study on predictionmethods for earthquake casualty is based on the combination of quantitative and qualitative approaches to earthquake information database. The research on earthquake prediction and disease spectrum of casualties has been conducted to provide an objective basis for earthquake medical response. The system can automatically get the information ofpopulation, culture, geography, disease spectrum, weather and road traffic in disaster areas, quickly predictearthquake casualty based on seismic intensity, and generate disaster assessment reports using the techniques of geographic information systems and intelligent decision to form a detailed plan of earthquake medical relief. For instance, in the 2010 Yushu earthquake, the system had successfully predicted the profile of casualties, and according to the intelligent positioning function of the system, fast and convenient pathways have been provided for rescue teams to rush to the quake zone.29Recently, an outstanding academic study from Chinese scholars has summarized previous major disasters to propose a triangular framework for public security science. The triangular framework of the public safety emergency response system included emergencies, acceptors, response and management.The national emergency platform system in China based on this triangular framework has played a key role in emergency response to major emergencies such as Wenchuan and Yushu earthquakes.30
Experimental methods in MCI response
In the study of MCI response, it is difficult to draw conclusions by observation alone and needs further experimental studies. For MCI research, observational experiments with direct results or with direct access to data are almost unavailable. Therefore, MCI simulation experiments gradually show their importance in providing available data for analysis. Based on the principle of similarity, a simulation model can be used to replace the study object. MCI can be “moved” into lab with simulation technologies and processing results of incidents based on simulated MCI can be obtained, assessed and amended.31
At present, policy simulation (PS) is an important type of simulation which is a method for the development of policy strategy and options and the assessment of policy effects via modeling using mathematical methods and computer simulation.33
PS as a large software is an effective tool to explore various policy scenarios. PS has been used in earlier policy simulations of disaster prevention and mitigation.34
Scenario approach to MCI response
As time and space condition is different for different catastrophic events. Scenarios are effective in dealing with uncertainties. Dynamic decision-making in real-time based on specific MCI scenario is named “scenario-response” management mode of emergency. “Scenario-response” has been regarded as a more effective emergency management model for unconventional emergency response. Scenario analysis is a multivariate analysis method. With the probabilities of the occurrence of various scenarios presetted, the potential impact of multiple factors can be simultaneously studied and viable response options can begained
by rigorous reasoning and detailed analysis.35
With the absence of historical data and the uncertainty of future environment, scenario analysis use representative scenes to simulate more actual scenarios and explore the potential threats of disasters. It can be used to propose response strategies to achieve a target state. The disadvantage of scenario approach is that till now there is no unanimously approved classification of scenario due to its extremely complex nature. For unpredictable unconventional emergencies, scenario analysis, a study method for “decision-making” based on the analysis of “scenario”, is gradually being developed. Scenario analysis is a prediction method based on the possible impact of disasters. One study conducted a scenario analysis of the emergency medical services (EMS)for large-scale multi-location emergencies.36
Currently, scenario planning construction is one of the most active areas of emergency research. For instance, U.S. Department of Homeland Security has implemented a major research project called “National Planning Scenarios” which has reviewed the high-consequence emergencies of all types occurred in USA and other countries in recent years. Potential high-consequence emergencies in future are being analyzed systematically and the related risk is being assessed. Systematic summarization has been conducted considering initial source, damage severity, scope affected, complexity degree and the long-term potential impact of high-consequence emergencies. Collectively, this project proposed 8 groups of important scenario with common characteristics.37
Unfortunately, China has not yet conducted the scenario planning construction of high-consequence emergencies of all types.
MCI response is a complex system
The complex type of MCI belongs to unconventional
MCI has been divided into conventional and unconventional MCI.38
Unconventional MCI, the more complex type of MCI, refers to emergencies with insufficient premonition, obvious complex characteristics, potential secondary derivative hazards and severe damage, which cannot be dealt with routine supervisor modes,39
such asWenchuan earthquake, Haidi earthquake, Japan earthquake and 911 incident. Many factors involve in the study of medical emergency response to this complex type of MCI.40
As the research object, MCI response system is of high complexity degree, with complex logic relations among various factors. The target of MCI response is rarely single and linear, but rather it pursues an overall optimal effect by comprehensively coordinating multiple objectives. For MCI response, constraints exist in many aspects, such as information, time, resource, technical methods, manpower, transportation capacity and infrastructure condition. Therefore, MCI response system is a complex system. Theoretical and methodological research on MCI response should adapt to such complexity.
Complexity characteristics of MCI response
MCI is generally large and complex. The complexity characteristics of MCI response can be fully reflected in the medical emergency relief for 2008 Wenchuan earthquake in China
Complexity characteristics of MCI response are as follows. First, MCI response is multivariate. For example, in the medical emergency relief in the 2008 Wenchuan earthquake, in addition to the earthquake, the course of MCI is often affected by multiple factors such as local climatic condition, geographical environment and road traffic conditions, etc. Secondly, MCI response has multiple agents. Wenchuan earthquake response was an integrated rescue involving government-led coordination of military and civilian rescue forces. In addition to the professional rescue, MCI response also involves non-professional rescue. Thirdly, it is consisted of multiple stages. Wenchuan earthquake was not only a large-scale earthquake, but also required a multi-stage rescue. The rescue focus after the disaster turned from the initial search and rescue into medical treatment and evacuation. Medical focus changed from medical treatment alone in the earlier stage to a focus on both medical treatment and epidemic prevention in the late stage of MCI. Fourthly, it is a multi-target response. Medical rescue for Wenchuan earthquake is a comprehensive rescue, the objective of decision making in MCI response is complex, taking into account of various aspects including casualty, loss, environmental impacts, rescue time and costs, etc. Fifthly, the interaction of various factors in MCI response is complex. In the medical emergency rescue for 2008 Wenchuan earthquake, in addition to the overlapping influences of individual factors, there were also non-linear effects of many factors. Moreover, MCI response is of dynamic characteristics. The discrepancy betweenmedical demands and resources, as well as external conditions fluctuate with a significantly dynamic characteristic during the medical rescue for 2008 Wenchuan earthquake. The final characteristic of MCI response is its uncertainty. In the early phase of Wenchuan earthquake relief, the greatest challenge was the uncertainty of time and location of medical needs and the uncertainty of rescue conditions. The external environment is full of unexpected variations. This caused great difficulty in making decisions based on insufficient information in the initial phase of Wenchuan earthquake relief.
Previous research methods did not adapt to the
complexity of MCI
There are many limitations in current research on disaster medical response. The mechanism of occurrence development and evolution of conventional incidents has already been explored. Impact pathways, modes and consequences of conventional incidents are clarified and can be strongly predicted. Thus, the basic strategy for conventional incident is “predict-response” mode. MCI response system is complex. Not only does it involve more links and elements than conventional incident response, but also the logical relations among factors in MCI are much more complex. Furthermore, emergency decision making during MCI shows obvious dynamic characteristics, and the evolution course of MCI response is a dynamic one. The classical decision theory can not deal with such complexity. Previous mathematical research simplifies MCI response system into mathematical models by using mathematical formulas to describe the interactions among variables and the complex situation in the system. Mainstream theories based on “simplicity” face major challenges in explaining the response to catastrophic events. The models are poor in fitting real situations, because of the conflicts between linear mathematical models used and nonlinear relations in actual scenarios and the infeasibility of many mathematical expressions, such as equations, due to their high requirements for input data. Traditional mathematical models are simplified, removing many important conditions, and thus are very different from actual situations, so conclusions gained in these models cannot be generalized. A novel method in researching the complex systems of MCI response should be required.
What is complexity science
Complexity science, as an interdisciplinary science, has attracted scholars in various areas. Before a further discussion in complexity science, we need to clarify the concept of system science first. System science is different from other disciplines in that it studies objective world by investigating the relationship between the whole and the components within and studies all levels within this objective world. Knowing parts of the system is not equal to the understanding of the whole system.42
Under internal and external forces and constraints, the system evolves following certain rules. System science emphasizes on overall comprehension. But in solving specific problems, it generally focuses on the study of various relationships in the model system and the method used is mainly analysis. Complexity science evolves from system science and studies complex systems.43
Complexity science shows that the whole, a higher level, has the ability to control, coordinate and select over the parts, the lower level, in complex systems. Complexity science, beyond reductionism, builds up scientific thinking modes for complexity. Currently, the theoretical tools of complexity science mainly include decision-making techniques under uncertainty conditions, comprehensive integration technology, computational intelligence, non-linear science, mathematical logic and computer simulation.44
Meta-synthesis proposed by Qian Xuesen who was a representative Chinese scholar in the field of complex systems, pointed out that the currently advocated system theory was neither holism nor reductionism, but a dialectical unity of both, and the only way to solve all complex issues was through integrative methods with a combination of quantity and quality. The proposed “fromqualitative to quantitative meta-synthesis method”, as well as its practical form “from qualitative to quantitative system of hall for workshop of meta synthesis”, is essentially an integration of statistical data and information, expert systems, and computer systems, which constitutes a combination of highly intelligent human-machine systems.45
What can be solved by complexity science
Science of complexity is trying to solve issues that all general scientific disciplines are unable to deal with. Complexity science is the new stage of the development of system science. Using non-reductionism method, complexity science studies the complexity of the mechanism and the evolution of complex systems in order to understand various aspects of the system, including integrity and mutation, hierarchy and structure, evolutionary purpose, coherence and openness. Complexity science clarifies “widgets” in the system, identifies rules of their interaction, opens access to macro world through understanding micro ones, integrates macro and micro worlds. On the basis of complexity science, combined with computer simulation, many systems that cannot use traditional mathematical models can be studied in depth.
Basic research strategies of complex science
In general, there are two strategies for complexity science research.44
The first is “emerge” and the other is “control”. The first strategy is the emergence-based bottom-up design method. Using computer simulation methods, simulated individuals within the system interact and evolve in virtual environments, causing the overall complexity of the system to “emerge” from bottom-up. The second strategy is the control-based top-down design method. “Control” methods are all ready for practical use after the appearance of artificial intelligence, such as the application in the expert system of practice and various systems of logic simulation.
ACP method as the new perspective of applying complexity science to MCI study
The concept of ACP
ACP method combines artificial systems, computational experiments and parallel execution. It is a novel approach to dynamic process in conducting complex research on complex systems.46
Artificial system method for managing complex systems
The modeling of complex systems is not aimed at approximating actual complex systems, but rather to construct artificial objects. It uses the knowledge of the simple to construct models in the greatest extent, which is the modeling principle of simple consistency. Agent programming is the main approach of artificial system modeling. Using the proxy method of artificial systems, establishment of the virtual system derived from reality and beyond reality will allow different parameters and interaction between parameters into digital signals that can be recognized and treated for signal processing, analysis, numerical integration and application.
Computational experiments for managing complex systems
For the study of complex systems, in most cases neither a sufficiently precise model of a system nor a short-term model to forecast the system is available. It is impossible or difficult to analyze and forecast the behavior of complex systems, and experimental studies of complex systems are also impossible or difficult to conduct. “Equivalent description” and “computational experiment” are used for complex systems studies to solve the difficulties in experimental issues and the analyses and assessment of system behaviors.47
By considering the impact of various factors on a system in the process of computational experiment, the evolutional trend of the whole complex process can be directly simulated with experimental data which can be unattainable in conventional experiments. And system reaction process can be speculated under human intervention. Many characteristics can also be easier to manipulate and repeat and various precise controls of “testing” can be used.
Parallel execution for managing complex systems
A parallel system means the common system composed of a natural reality system and the corresponding one or more virtual or ideal artificial systems. Parallel system approach is a further development and natural extension of control systems and computer-based modeling and simulation, resulting from recent advances in computing technology. Parallel system methods might provide effective tools for the control and management of complex systems that can not be modeled precisely or experimented repeatedly. The basic framework of a parallel system applied to complex systems primarily includes the actual system and manual systems, through the interaction of the two to complete the actual system of management and control, related experiments and evaluation of behavior, and decision-making.48
Parallel control is unique in that it changes the non-dominant position of artificial systems, making their roles from passive to active, static to dynamic, offline to online, and subordinate status to equal status, and it carries out comparative analysis of the results of the parallel implementation of actual complex systems and artificial systems.
Construction of artificial systems for “equivalent” description of complex systems uses “computing experiment” and “parallel” control to solve the difficulties in restoration, testing, analysis and process evaluation in complex systems. Construction of artificial MCI response systems to conduct virtual MCI experiments may be an alternative and assist in the scientific research of MCI in reality in the future. It can achieve goals which are difficult to be achieved in practical experiments and observe phenomena and laws which are difficult to be observed in the experiments of real response systems. And complex experimental conditions can be set for full intervention studies.
In all, the main problem in approaches to MCI response currently is that the adopted research methods are widely based on reductionism, which is inadequate for the research on MCI response. Prospective studies on MCI are needed to verify its applicability, generalizability and validity.As MCI research is still in an initial stage, we should learn experiences and lessons from other fields of emergency management to enhance methodology research. Complexity science research method applied to MCI response has only just begun and no consensus is yet reached, it needs MCI response researchers around the world to develop together. For complex system of MCI response, consideration of complex issues with a holistic perspective is necessary. It is worth noting that no research method is able to solve all problems. In this paper, research methods mentioned are just part of the list, not comprehensive, and selection of appropriate research methods needs to be based on specific research issues.
Bozeman WP. Mass casualty incident triage. Ann Emerg Med 2003; 41: 582-583.
Zhang L, Liu X, Li Y, Liu Y, Liu Z, Lin J, et al. Emergency medical rescue efforts after a major earthquake: lessons from the 2008 Wenchuan earthquake. Lancet 2012; 379: 853-861.
Shannon D, Megan C, Thomas K. Mortality following the Haitian earthquake of 2010: a stratified cluster survey. Popul Health Metr 2013; 11: 5.
Nagata T, Halamka J, Kennochi H, Himeno S, Himeno A, Hashizume M. Using a cloud-based electronic health record during disaster response: a case study in Fukushima, March 2011. Prehosp Disaster Med 2013; 28: 1-5.
Li TS, Chai JK. Implementation of telemedicine services in the earthquake disaster relief: the best medical experts provide direct medical service to the affected people (in Chinese). Chin Crit Care Med 2013; 25: 262-264.
Shen WF, Xu SW, Zhang M, Ma YF, Gan JX, Jiang GY. Discussion on strategies of medical response to catastrophic surge (in Chinese). Chin J Hospital Admin 2008; 24: 394-396.
Schnitzer JJ, Briggs SM. Earthquake relief -- the U.S. medical response in Bam, Iran. N Engl J Med 2004; 350: 1174-1176.
Baker SP, G Li. Epidemiologic approaches to injury and violence. Epidemiol Rev 2012; 34: 1-3.
Arnold JL. The 2005 London bombings and the Haddon matrix. Prehosp Disaster Med 2005; 20: 278-281.
Hick JL, Koenig KL, Barbisch D, Bey TA. Surge capacity concepts for health care facilities: the CO-S-TR model for initial incident assessment. Disaster Med Public Health Prep 2008; 2 Suppl 1: S51-S57.
Hick JL, Barbera JA, Kelen GD. Refining surge capacity: conventional, contingency, and crisis capacity. Disaster Med Public Health Prep 2009; 3 (2 Suppl): S59-S67.
Abir M, Davis MM, Sankar P, Wong AC, Wang SC. Design of a model to predict surge capacity bottlenecks for burn mass casualties at a large academic medical center. Prehosp Disaster Med 2013; 28: 23-32.
Watson SK, Rudge JW, Coker R. Health systems’ “surge capacity”: state of the art and priorities for future research. Milbank Q 2013; 91: 78-122.
From the centers for disease control and prevention. Rapid assessment of injuries among survivors of the terrorist attack on the World Trade Center--New York City, September 2001. JAMA 2002; 287: 835-838.
Aylwin CJ, König TC, Brennan NW, Shirley PJ, Davies G, Walsh MS, et al. Reduction in critical mortality in urban mass casualty incidents: analysis of triage, surge, and resource use after the London bombings on July 7, 2005. Lancet 2006; 368: 2219-2225.
Gong Z, Chai C, Tu C, Lin J, Gao Y, Qiu Y. Injuries after a typhoon in China. N Engl J Med 2007; 356: 196-197.
Zhang L, Liu X, Li Y, Liu Y, Liu Z, Lin J, et al. Emergency medical rescue efforts after a major earthquake: lessons from the 2008 Wenchuan earthquake. Lancet 2012; 379: 853-861.
Kellermann AL, Peleg K. Lessons from Boston. N Engl J Med 2013; 368: 1956-1957.
Auf der Heide E. The importance of evidence-based disaster planning. Ann Emerg Med 2006; 47: 34-49.
Navin DM, Sacco WJ. Science and evidence-based considerations for fulfilling the SALT triage framework. Disaster Med Public Health Prep 2010; 4: 10-12.
Bayntun C. A health system approach to all-hazards disaster management: a systematic review. PLoS Curr 2012; 4: e50081cad5861d.
Debacker M, Hubloue I, Dhondt E, Rockenschaub G, Rüter A, Codreanu T, et al. Utstein-style template for uniform data reporting of acute medical response in disasters. PLoS Curr 2012; 4: e4f6cf3e8df15a.
Zou K, Liu GJ, Li YP, Liang DU, He J. Risk factors for death and injuries in earthquakes: A systematic review (in Chinese). Chin J Evid Med 2008; 8: 477-482.
John NL, Andrew DO, Simon L, Atle F. SUPPORT Tools for evidence-informed health Policymaking (STP) 3: Setting priorities for supporting evidence-informed policymaking. Health Res Policy Syst 2009; 7 Suppl 1: S3.
Hupert N, Hollingsworth E, Xiong W. Is overtriage associated with increased mortality? Insights from a simulation model of mass casualty trauma care. Disaster Med Public Health Prep 2007; 1 (1 Suppl): S14-S24.
Lerner EB, Schwartz RB, Coule PL, Pirrallo RG. Use of SALT triage in a simulated mass-casualty incident. Prehosp Emerg Care 2010; 14: 21-25.
Shi BY, Xia S, Yang GJ, Zhou XN, Liu JM. Inferring the potential risks of H7N9 infection by spatiotemporally characterizing bird migration and poultry distribution in eastern China. Infect Dis Poverty 2013; 2: 8.
Riley S, Fraser C, Donnelly CA, Ghani AC, Abu-Raddad LJ, Hedley AJ, et al. Transmission dynamics of the etiological agent of SARS in Hong Kong: impact of public health interventions. Science 2003; 300: 1961-1966.
Gu F. Disaster intelligent decision system for armed police hospital 10 minutes to predict earthquake casualty (in Chinese). (accessed March 27, 2013, at http://news.enorth.com.cn/system/2013/03/27/010788827.shtml)
Fan WC, Liu Y, Weng WG. "Triangle" framework and the “4+1” methodology of public safety technology (in Chinese). Sci Technol Rev 2009; 6: 3.
Zhu JM, Huang J, Liu DG, Han JY. Randomized algorithm model for medical supplies in for vehicle routing large-scale emergencies (in Chinese). Operations Res Manag Sci 2010; 1: 9-14.
Franc-Law JM, Bullard MJ, Della Corte F. Accuracy of computer simulation to predict patient flow during mass-casualty incidents. Prehosp Disaster Med 2008; 23: 354-360.
Chen A, Li ML. New train of thought on several theories and practice in modern emergency management (in Chinese). Bull Chin Acad Sci 2008; 23: 531-537.
Tian W, Li MN, Zhang LL, Ma YQ. Policy intervention of public health service in China (in Chinese). Chin Gener Prac 2010; 13: 1093-1095.
Liu TR. Studies on scenes construction of emergency planning—part I of emergency planning technology based on scene-task-ability (in Chinese). J Safety Sci Technol 2012; 8: 5-12.
Jia HZ, Ordonez FDessouky MM. Solution approaches for facility location of medical supplies for large-scale emergencies (in Chinese). Comp Indust Engin 2007; 52: 257-276.
Jiang H, Hou JS. Study on the construction of scenario planning for major emergencies (in Chinese). China Emerg Manag 2012; 4: 18-23.
Benin-Gorena1 O, Blumenfelda A. Internalization of medical protocols during medical responses to conventional and unconventional mass-casualty incidents. Prehosp Dis Med 2005; 20 Supplement S1: 85-86.
Qing Y, Fan Y, Yu YL. Multi-agents computational experiment on twice-linked model of unconventional emergencies evolution mechanism (in Chinese). J Comput 2013; 8: 1035-1042.
Shu QL. Study on scenario evolvement and alternative generation of scenario-response decision-making in unconventional emergencies (in Chinese). J Univ Sci Technol China 2012; 42: 936-941.
Hua GW, Yu LA, Wang SY. Characterization of unconventional emergency and emergency decision-making (in Chinese). J Xidian Univ (Soc Sci Ed) 2011; 13: 33-36.
Qi LL. Differences and similarities between systems science and complexity science (in Chinese). J Systems Sci 2008; 16: 31-35.
Xu Y. Complexity science theory and crisis management of nuclear accidents and radioactive emergencies (in Chinese). Soft Sci Health 2007; 21: 294-296.
Cao Z, Zhang XP, Cao XD, Yin X, Wang D. A discussion on methodologies for research into complex systems (in Chinese). Caai Transs Intel Sys 2009; 4: 76-80.
Dai RW, Li YD. Researches on hall for workshop of metasynthetic engineering and system complexity (in Chinese). Compl Syst Complex Sci 2004; 1: 1-24.
Wang FY, Qiu XG, Zeng DJ, Cao ZD, Fan ZC. A computational experimental platform for emergency response based on parallel systems (in Chinese). J Syst Sci Complex 2010; 7: 1-10.
Wang FY. Computational theory and method on complex system (in Chinese). China Basic Sci 2004; 6: 3-10.
Guo G, Chen B, Qiu XG. Artificial environment construction technology of parallel system (in Chinese). J Syst Simul 2011; 23: 1686-1690.
(Received December 21, 2013)