Evolution of ICS in the United States
- Top of page
- Evolution of ICS in the United States
- Evidence regarding the system's implementation
- Factors influencing ICS implementation
The United States has a chronic response problem. Disasters are too often seemingly ad hoc or piecemeal in nature. Throughout the nation's history, response has been typified by a host of difficulties related to interorganizational communication, coordination, collaboration, and leadership, among other challenges (see for example: Auf der Heide, 1989, p. 157; National Research Council, 2006, p. 141; Tierney, Lindell, & Perry, 2001, p. 211). According to National Research Council (2006) problems commonly associated with response to large-scale incidents include
… failure to recognize the magnitude and seriousness of an event; delayed and insufficient responses; confusion regarding authorities and responsibilities, often resulting in major ‘turf battles’; resource shortages and misdirection of existing resources; poor organizational, inter-organizational, and public communications; failures in intergovernmental coordination; failures in leadership and vision; inequities in the provision of disaster assistance … (p. 141).
As a result, of these problems, response efforts have often been characterized as inefficient, wasteful, and/or ineffective (see for example: Waugh, 2006). Despite a variety of national level efforts to minimize the effect of, if not eliminate, common response issues, the problem has endured. After the 11 September 2001 terrorist attacks revealed a response once again fraught with problems (see for example: National Commission on Terrorist Attacks upon the United States, 2004), the federal government initiated a series of sweeping policy changes to address response deficiencies.
Key among these initiatives was Homeland Security Presidential Directive 5 that required the development of a National Incident Management System (NIMS) to provide ‘a consistent nationwide approach for federal, state, and local governments to work effectively and efficiently together to prepare for, respond to, and recover from domestic incidents, regardless of cause, size, or complexity’ (Bush, 2003, p. 1). When the newly formed Department of Homeland Security rolled out NIMS in 2004, a mandate to implement the ICS in the on-scene response to all incidents was included within its Command and Management Component.
The view of the Department of Homeland Security is that ICS is ‘a management system designed to enable effective and efficient domestic incident management by integrating a combination of facilities, equipment, personnel, procedures, and communications operating within a common organizational structure’ (Department of Homeland Security, 2004, p. 14). ICS is based on a variety of principles including hierarchical modular organization, span of control, establishment and transfer of command, chain of command, unity of command, and accountability, among others (Department of Homeland Security, 2004, 2008). The system aims to provide an organizational structure through which activities related to the functional areas of command, operations, planning, logistics, and finance and administration can be carried out in concert by all responding entities to any incident. The organizational structure of ICS includes a variety of units, positions, processes, and tools (Department of Homeland Security, 2008). As of 2004, all organizations in the United States are expected to utilize ICS to structure their interorganizational efforts for all response efforts regardless of the incident's cause, size, or complexity. It is believed that utilization of ICS will eliminate age-old response problems – at least those related to on-scene operations – and the following scenario will result:
When agencies involved in a major emergency use ICS … there are few, if any, differences in operations. In essence, they are ‘one’ organization, and can be managed as such. Instead of several command posts operating independently, the total operation can be directed from only one location. Instead of preparing several sets of plans (with no guarantee of coordination among them) only one set need be prepared to inform all participants. In place of several logistical and communications processes, only one system of collective and integrated procedures is used (Auf der Heide, 1989, p. 157).
Prior to the mandate to use ICS within NIMS, there was reason to believe that the system would be both used and useful in addressing the United States' response problem. After all, the system had been developed to address the very same range of interorganizational issues noted in the response to the 11 September 2001 terrorist attacks.
Response efforts to wildfires in California in the late 1960s evidenced numerous interorganizational response challenges. An interagency group representing all levels of government was convened in 1972 to solve these problems. The project they engaged in was titled Firefighting Resources of California Organized for Potential Emergencies – better known by its acronym FIRESCOPE. The outcome of FIRESCOPE was the Wildfire ICS
The Wildfire ICS came to be used by and popular among wildland firefighters within California and beyond in the decades that followed its conception. The Wildfire ICS was also adapted in some form or fashion by the fire service more broadly (Nicholson, 2006; Perry, 2003; Wenger, Quarantelli, & Dynes, 1990), and, eventually by other organizations such as the United States Coast Guard, Occupational Health and Safety Administration and the Environmental Protection Agency (Christen et al., 2001; Harrald, 2006; Hannestad, 2005; Molino, 2006). The system was credited as a key facilitator of effective response to a number of events including the Murrah Federal Building bombing in Oklahoma City in 1995 (Cook, 2009). Its use became so widespread that at least one scholar referred to ICS as ‘the de facto standard for firefighting and emergency management’ (Harrald, 2006, p. 263).
The system came to be perceived positively by a range of practitioners involved in response in the years leading up to the 11 September 2001 terrorist attacks (Buck, Trainor, & Aguirre, 2006; Tierney et al., 2001; Waugh, 2009a, 2009b; Wenger et al., 1990). ICS is viewed by practitioners as having the potential to bring order out of chaos, alleviate communications issues, clarify leadership issues, reduce duplication of effort, reduce response-related expenditures, facilitate the integration of human and other resources into operations, and potentially increase the safety of responders (Anderson, Compton, & Mason, 2004; Perry, 2003, p. 407). The system is believed to be applicable to the management of virtually any situation including ‘… urban fires, high rise fires and other agencies responsible for different types of incidents such as search and rescue, earthquakes, rock concerts, and the 1984 Summer Olympics in Los Angeles’ (Federal Emergency Management Agency, 1987, p. 3). These benefits, and others, have led certain responding organizations to enthusiastically support the system (Christen et al., 2001; Hannestad, 2005; Harrald, 2006; Waugh, 2009b).
The narrative to this point could lead a reader to believe that the evolution of ICS has been nice, neat, linear, and natural – the increasing use and eventual incorporation of it in the NIMS mandate propelled by the successful implementation of the system in practice and its popularity among practitioners. Similar narratives have been shared before countless times (see for example: Bigley & Roberts, 2001; Christen, Maniscalco, Vickery, & Winslow, 2001; Lam, Lin, Tasi, & Chiu, 2010; Perry, 2003; Stambler & Barbera, 2011; Tsai & Chi, 2012). Yet, the narrative to this point is incomplete.
While ICS did come to be used widely and enjoyed popularity in the years leading up to the 9/11, there has been consistent controversy regarding whether one system is suitable for response to all fires within the firefighting community – particularly between wildland and urban firefighters. A single system has not been used in fire – ‘many … have evolved over the past forty years, not in a linear format but rather more concurrently’ (Molino, 2006: 1). For example, while believing that response challenges in wildland and urban firefighting were similar, Chief Alan Brunacini of the Phoenix Fire Department, believed that the FIRESCOPE model was designed primarily for the management of large-scale incidents involving multiple agencies as opposed to the day-to-day urban fires with which his department contended (Angle, Gala, Harlow, Lombardo, & Craig, 2001). He developed an alternative system – the Fire Ground Command System – in the late 1970s and early 1980s (Angle et al., 2001; Coleman, 1997; Green, 2002).
The Fire Ground Command System is based on the same principles as the FIRESCOPE model (Coleman, 1997); however, as Coleman (1997) described it, the Brunacini version is ‘less formal and more laid back’ (p. 8). Brunacini designed his version of ICS specifically for small-scale incidents with a limited amount of impact that occur over a short period of time (Green, 2002; Molino, 2006). It uses a simplified organizational structure and is less hierarchical (Green, 2002, p. 1). The Fire Ground Command System came to be quite popular; and, it, like the FIRESCOPE model, came to be adapted and used in firefighting in the 1980s and early 1990s (Angle et al., 2001; Federal Emergency Management Agency, 2004; Molino, 2006).
The National Fire Protection Association (NFPA) – the standard setting body for the fire service – recognized that more than one version of the ICS was being used for various reasons. When NFPA 1561: Standard on Fire Department Incident Management Systems was released in 1990, it required that fire departments use an incident management system and specified elements that the system must include, but the standard did not require a particular model of the system be used (National Fire Protection Association, 1990). The very same year that NFPA 1561 was released, the National Fire Service Incident Management Consortium convened ‘recognizing the continuing challenges occurring in the fire service in applying a common approach to incident command …’ (Federal Emergency Management Agency, 2004, p. 2). The Consortium's mission was to try to merge the Fire Ground Command System and FIRESCOPE ICS into a single system that both fire and rescue personnel could utilize. The result of the Consortium's work was Model Procedures Guide for Structural Firefighting, released in 1993. The National Fire Academy adopted the changes suggested in the Guide and the FIRESCOPE model was adjusted to reflect the Model Procedures (Federal Emergency Management Agency, 2004); yet, the controversy regarding whether a single incident management system could be equally appropriate for managing wildland and urban firefighting continued (see for example: Brunacini, 2008; Green, 2002).
In addition to concerns about the applicability of a single management system to all fires, firefighters have also noted other concerns including, for example, the appropriateness of the system for rural areas and volunteer and small fire departments (see for example: Clayton, 1990; Dyer, 1989; Henderson, 1991), diminishing usefulness of ICS when other responding organizations are not also using the system (see for example: Kaskey, 1989; Ullman, 1998) and obstacles to implementation related to disagreements over who should be in command of an incident (see for example: Podlubny, 1992). Concerns about ICS have not been limited just to the fire service. There are concerns among practitioners that the system may not be as flexible and scalable as it needs to be and may not be applicable to all kinds of incidents, particularly those that are not localized, such as pandemics (Waugh, 2009a). Public health agencies have complained about conflicts between ICS' unity of command and their more collegial decision processes and military after action reports following fire responses have expressed concerns about ICS operations not fitting their battle rhythm, i.e., the tempo of decision-making and operations (Waugh, 2009a). Law enforcement agencies also have expressed concerns that the top-down command structure is inconsistent with the autonomy normally granted to police officers on the street (Waugh, 2009a). The cultures of fire service and law enforcement agencies are quite different and command structures reflect those differences. The Emergency Management Accreditation Program, the organization that sets standards for, and accredits state and local emergency management programmes in the United States (now including Ottawa, Canada; the Centers for Disease Control and Prevention; and one US Army Corps of Engineers regional organization) has long supported the use of formal incident management systems, but only offers ICS and NIMS as examples (see Standard 2.11 at http://www.emaponline.org). Even Craig Fugate, the current Federal Emergency Management Agency Administrator, expressed concerns about ‘ICS zealots’ during a plenary session at the annual Natural Hazards Workshop at the University of Colorado in 2006. Some people, according to Fugate (2006), are unable or unwilling to modify the system when events suggest that it is not working or that another organizational structure might work better.
Practitioners are not alone; disaster scholars too have voiced issues with ICS and command and control models over the course of its evolution. In fact, critiques have been common within the disaster research community (see for example: Drabek, 2003; Dynes, 1983, 1993, 2000; Neal & Phillips, 1995; Schroeder, Wamsley, & Ward, 2001; Tierney, Lindell, & Perry, 2001; Perrow, 2005; Walker, Harrald, Ducey, & Lacey, 1994; Waugh & Streib, 2006; Wenger et al., 1990; Wise & Nader, 2002). Scholars have questioned whether such systems are needed (see for example: Drabek & McEntire, 2002; Dynes, 1994), flexible enough (see for example: Drabek & McEntire, 2002; Waugh, 2009a, 172), appropriate for response operations (see for example: Drabek, 1983, 1985; Drabek & McEntire, 2002; Neal & Phillips, 1995; Waugh, 2009a; Waugh & Streib, 2006; Wenger et al., 1990), information sharing (see for example: Comfort, 2002, 2007), and collaborative leadership (see for example: Waugh, 2009a, p. 168); and, whether use of such a system is actually counterproductive relative to its intent (see for example: Jensen, 2009; 2010, 2011; McEntire, 2001: 8). In fact, the National Research Council (2006, p. 142) went so far as to state that ‘empirical research … finds essentially no support for the command-and-control model either as a heuristic device for conceptualizing the disaster management process or as a strategy employed in actual disasters’.
The evolution of ICS is not as simple as most narratives might lead one to believe. There has been an ongoing debate about the applicability of one on-scene incident management system to all incidents, situations, and organizations among scholars and practitioners – even firefighters. This debate suggests that the assumptions upon which the ICS mandate is based may be inaccurate. Yet, a history of competing command systems and questions regarding the system does not mean that ICS is not used in practice.
Evidence regarding the system's implementation
- Top of page
- Evolution of ICS in the United States
- Evidence regarding the system's implementation
- Factors influencing ICS implementation
Examination of the evidence regarding ICS implementation in the United States suggests that ICS is being used and has been for decades, but there are caveats to this observation. While ICS has been used as designed, it has also, concurrently, been ignored, partially used, underused, misused, and used in organizations and jurisdictions in ways that are not necessarily compatible with the way the system is used in other organizations and jurisdictions. This variation in implementation seems to characterize the fire service's experience with the system pre- and post-NIMS mandate as well as the experience of other organizations and jurisdictions.
Most of the evidence regarding the implementation of ICS in the United States comes from firefighters. The National Fire Academy Executive Fire Officer Program requires participants to complete an applied research project on a topic of their interest as a capstone to their training. Many participants have chosen to focus on ICS for the research. Their papers reveal a history of fire departments, agencies, and jurisdictions not using ICS at all (see for example: Burton, 1989; Brent, 1992; Byrnes, 2005; Funai, 2013; Juratovac, 2004; Miller, 1991; Otto, 1997; Ross, 1994; Wieland, 1995), underusing it, inconsistently using it, or misusing the system (see for example: Johnson, 2002; Kernan, 1995; Kitterman, 2009; Podlubny, 1992; Price, 2000; Spencer, 2010) as well as a history of modifying one of the dominant models of the system (see for example: Coleman, 1991; Cooper, 2001; Obier, 1990; Ullman, 1998; Weston, 1997; Williams, 1989). Other scholarly, peer-reviewed works reporting the results of empirical research on ICS also suggest that the system is not being implemented consistently day-to-day (Jensen, 2009, 2011; Jensen & Yoon, 2011) or in response to large-scale incidents (Branum, Dietz, & Black, 2010; Buck et al., 2006; Jensen, 2008; Lutz & Lindell, 2008; Moynihan, 2008, 2009a). And, there is evidence to suggest that as the system is being modified county-to-county within and across states, although the extent to which the modifications being made are compatible county-to-county is unclear (Jensen & Youngs, forthcoming). The literature seems to suggest a wide range of variables influence whether, and how, ICS is implemented.
- Top of page
- Evolution of ICS in the United States
- Evidence regarding the system's implementation
- Factors influencing ICS implementation
There are good reasons to believe ICS will be a consistently effective organizing mechanism for the management of on-scene response. ICS is currently the required system for on-scene incident management in the United States. It has been mandated for use by all responding organizations for nearly a decade. There is significant support of the system within the emergency management community. The system has a clear underlying theory (i.e., if it is implemented, effective response will result); and, to support making the theory real in practice, a series of structures, processes, and positions are specified as well as their relationships to one another. The system can, and has been, used successfully to facilitate command and control, communication, and coordination.
There are also good reasons to want it to ‘work’. The United States has an enduring response problem that the use of the system is purported to correct. A variety of attempts have been made to address response issues over time and have not worked as hoped. Significant time, energy, and funding have been invested in seeing the NIMS mandate, and its subcomponent ICS, implemented, and implemented effectively.
None of the existing empirical research on ICS has stated that the theory underlying ICS is a flawed or that the desire to see ICS work is unfounded. Instead, research has suggested ICS has not been, and is not currently, consistently used or equally effective in all events and contexts in the United States. A lack of use, misuse, underuse, partial use has been observed as has a tendency to modify the system from its design. A critical issue to address through continued dialogue in the emergency management, academic, and policy communities is why organizations and jurisdictions perceive the need to adjust the system when the system itself is premised on the need for a common approach to incident response. Nevertheless, it does not appear from an examination of the system's evolution and evidence regarding its implementation that the system is consistently implemented, implemented similarly, or equally suited to all events, contrary to the assumptions upon which it is based.
Review of the small body of empirical research published in scholarly, peer-reviewed publications suggests that myriad variables influence the system's implementation and effectiveness. Variables related to those who participate in the system including each individual responder, each organization, and leaders within the system appear to be important as do variables related to the response network overall, the local area, the hazard event, and how the system is implemented in response. Many of the variables seem to need to be in place before an incident occurs; and, many of the variables seem to need to be in alignment during the response itself.
All of the variables may have to be aligned for ICS to be implemented as it ought and achieve the ends for which it was designed; alternatively, only some may be preconditions for implementation and/or effectiveness. Rather than being preconditions, perhaps some of the myriad variables are simply influential. Regardless, the suggestion that the systems use and/or effectiveness is vulnerable to any variables – much less a wide range of variables – is concerning.
Unfortunately, it is impossible to ensure that all of the variables are in alignment pre- and per-incident at any given time much less over time and across the United States. The presence or absence of variables related to the hazard event and local area are completely beyond control while others, like those related to individual participants, organizations participating, and the response network, can be influenced, although still not controlled. For instance, training and practice can be provided to all individuals and organizations pre-incident and efforts can be made to generate buy-in and commitment, but whether the individuals and organizations who respond to a particular incident have actually trained, practiced, and developed relationships with one another based on trust, or, whether they are each committed to implementing the system cannot be guaranteed.
Based on the review of the system's evolution, evidence of its use, and the variables influencing its use and effectiveness, it seems safe to say that the assumption that mandating ICS for use in on-scene response will solve common response problems is also wrong. While the incident command system can, and has been, implemented successfully, this review would suggest it cannot be expected to be everywhere, all the time.
It is important to keep in mind that there has been very little empirical, scholarly, peer-reviewed research on ICS within or outside of the United States and more is urgently needed. The landscape of literature related to ICS is broad. Dotting the landscape is the aforementioned discussion of command and control models by disaster researchers as well as literature specifically addressing ICS in professional magazine articles, National Fire Academy Executive Fire Officer papers, graduate student papers, theses, and dissertations, book chapters, and scholarly journal articles. (See Table 1 for an outline of the major features of this landscape and sample citations.) Very little of what has been written reports on empirical research and even less has been peer-reviewed and published in scholarly outlets.
Table 1. The Landscape of ICS Literature
|Literature related to command and control||See for example: Comfort, 2002; 2007; Drabek, 2003; 2007; Drabek and McEntire, 2002; Dynes, 1983; 1993; 1994; 2000; Neal and Phillips, 1995; Schroeder, Wamsley, and Ward, 2001; Tierney et al., 2001; Perrow, 2005; Walker et al., 1994; Waugh and Streib, 2006; Waugh, 1993; 2009a; Wise and Nader, 2002|
|Articles in professional magazines||See for example: Bennett, 2011; Cardwell and Cooney, 2000; Decker, 2011; Dudfield, 2008; Goldfarb, 1997; Larson, 1998; Parker, 2005|
|National Fire Academy Fire Executive Officer papers||See for example: Bardwell, 2005; Berk, 2001; Brent, 1992; Cole, 2001; Nash, 1994; Juratovac, 2004; Webb, 1990|
|Naval post graduate student papers||See for example: Favero, 1999; Templeton, 2005|
|Student theses and dissertations based on empirical research||See for example: Arney, 1993; Benson, 2004; Chian-Cheng, 2005; Cone, 2007; Domorod, 1991; English, 1992; Fakhoury, 2009; Freeman, 2005; Hancock, 2010; Huang, 2004; Huang, J., 2004; Mason, 2006; O'Neill, 2008; Schoen, 2008; Su, 2004; Szabo, 1993; Williams, 2005|
|Scholarly, peer-reviewed journal articles NOT based on empirical research||See for example: Adams et al., 2010; Andrew and Kendra, 2012; Arnold et al., 2005; Autry and Moss, 2006; Fishbane, Kist, and Schieber, 2012; Gyorfi et al. 2008; Hansen, 2007; Hayes, 2012; Jacoby, 2005; Madigan and Dacre, 2009; Moynihan, 2007; 2008; Nicholson, 2006; Phonburee et al., 2010; Stumph, 2001; Wenger, Quarantelli, and Dynes, 1990; Zane and Prestipino, 2004|
|*Scholarly, peer-reviewed journal articles based on empirical research on specific aspects of ICS in the United States||See for example: Arnold, Paturas, and Rodoplu, 2005; Branum, Dietz, and Black, 2010; Crichton, Lauche, and Flin, 2005; Granillo et al., 2010; McLennan et al., 2006; Stambler and Barbera, 2011; Thomas et al., 2004|
|*Scholarly, peer-reviewed journal articles based on empirical research on ICS overall in the United States||Bigley and Roberts, 2001; Buck, Trainor, and Aguirre, 2006; Jensen and Yoon, 2011; Lutz and Lindell, 2008; Moynihan, 2008; 2009a; b|
|Scholarly, peer-reviewed journal articles OUTSIDE the United States||Arnold et al., 2001; Djalai et al., 2012; Lam et al., 2010; Nja and Rake, 2008;Tsai and Chi, 2012; Tsai et al., 2004; Wang et al., 2008; Yarmohammadian et al., 2011|
The authors of this article were only able to identify 14 scholarly, peer-reviewed journal articles reporting the findings of empirical research on ICS in the United States. Some of these articles focused only on one aspect of the system and others were more general and many evidenced methodological issues (Jensen, forthcoming). The reader should not draw from our discussion any sweeping conclusions about the veracity of the assumptions upon which the system is based or whether the mandate to use ICS will ultimately lead to the elimination, or reduction, of common response shortcomings.
Going forward, it is critical that more research on ICS implementation and effectiveness be conducted pre- and per-response and in a variety of organizational, geographic, national, and event settings before we can confidently assert conclusions regarding the system's potential. The discussion in this paper has been offered to both instigate research and provide a foundation for its conduct. Building on the prior research on ICS and exploring the variables it has suggested are influential and would allow disaster researchers to amass a significant body of knowledge related to ICS usefulness in a relatively short period of time.
It will fall to future researchers to investigate a range of topics related to ICS. For instance, more needs to be known about how the system is implemented day-to-day and in incidents of varying size, scope and complexity (e.g., extent of use, appropriateness of use given the situation) and by whom. Moreover, research needs to explore the extent to which ICS is effective in addressing common response problems. Yet, how effectiveness would actually be assessed will need to be carefully addressed including whether a qualitative or quantitative approach will be used and whether the data gathered will be subjective or objective.
Based on the existing ICS research, there is reason to believe that all of the variables discussed are relevant to understanding ICS implementation and effectiveness. Future research will need to first establish whether the variables identified from the prior work on ICS are actually related to these two dependent variables and then if they are predictive of them. Conducting such an exploration will require some degree of operationalization and measurement of the variables identified and discussed in this paper. How these variables will be operationalized using a qualitative or quantitative approach will also need to be carefully addressed by future research.
There seems to be an implication within the literature that the variables discussed are not simply or solely related to the system's usefulness. Rather, the literature indicates that the manner in which the system is used in a given response may be an intervening variable that explains the extent to which the system is effective. If relationships are discovered between the independent variables identified in this paper and ICS usefulness, then research must examine whether the variables are (1) directly related to the system's effectiveness, (2) directly related to how the system is implemented, and (3) whether implementation is an intervening variable related to effectiveness of the system. Where direct relationships are found between the variables identified in this paper and ICS implementation and/or effectiveness, then the strength of those relationships and their relative influence should be examined. The extent to which the variables identified in this paper are themselves interrelated will also need to be investigated.
It will be difficult, if not impossible, for researchers to explore the relationship of all of the variables we identified in this paper in one single incident response study. Yet, there is considerable value in future research doing less than a comprehensive study of all of the variable groups and the variables within them (e.g., taking one or more group of variables and exploring their relationship to ICS implementation or effectiveness) particularly if the researcher(s) involved link their research findings to the prior work on the topic and explicitly state the variable relationships they do/do not discover. There is a significant need for research in this area and it must begin somewhere.
While there is much work to be done if we are to understand when ICS is used and useful, it is a work worth doing. Response efforts have time and again revealed similar problems in the United States (and around the world for that matter) and a solution to those problems is desirable. The review of the system's evolution, evidence related to its use and the plethora of factors that the ICS research suggests influence its implementation and effectiveness suggest ICS is unlikely to be the solution. Yet, empirical research is needed to demonstrate the potential of the system.
Should some, or all, of the variables identified in this paper be found to be related to ICS implementation or effectiveness, it would be cause for concern. The ICS mandate is based on the assumptions that the system will be used and useful in all places and contexts, but if there are variables that lead to a lack of use, partial use, or misuse of the system and/or variables that lead to the system being less than effective, then it may be wise to revisit the mandate to use ICS, revise the system, or – at the very least – revise ICS related training to include reference to the variables that are related to ICS in practice and what can be done to limit the negative influence they might have.