The Benefits of Designing a Stratification System for New York City Pediatric Intensive Care Units for Use in Regional Surge Capacity Planning and Management


Accurate assessment of New York City (NYC) pediatric intensive care unit (PICU) resources and the ability to surge them during a disaster has been recognized as an important citywide emergency preparedness activity. However, while NYC hospitals with PICUs may be expected to surge in a disaster, few of them have detailed surge capacity plans. This will likely make it difficult for them to realize their full surge capacity both on individual and regional levels. If the pediatric resources that each NYC PICU hospital has can be identified prior to a disaster, this information can be used to both determine appropriate surge capacity goals for each PICU hospital and the additional resources needed to reach those goals. City agencies can then focus citywide planning efforts on making these resources available and more easily anticipate what a hospital will need during a disaster. Communication of this hospital information both prior to and during a surge situation will be aided by a stratification system familiar to both city planners and hospitals. The goal of this project was to design a NYC PICU surge stratification system that would aid physicians, hospitals and city agencies in regional surge capacity planning for critical pediatric patients. This goal was demonstrated through two objectives. The first identified major factors to consider when designing a stratification system. The second devised a preliminary system of PICU stratification based on clinical criteria and resources.


The importance of being able to surge in hospital beds as a part of disaster preparedness planning has long been recognized in New York City (NYC), as well as in the United States as a whole [13]. However, while significant efforts have been made in surge capacity planning for adult patients in medical facilities, fewer resources have been devoted to similar planning for the pediatric population. A 2003 recommendation by the US Department of Health and Human Services suggested that hospital surge capacity plans aim to accommodate an additional 500 inpatients (adults and children) or casualties per one million population [1]. These guidelines, while often referred to as general guidelines for surge capacity, were actually set with a bioterrorism-type event in mind, such as one with a chemical or infectious etiology. They were not based on previously collected data. They may not be equally applicable to all disaster events. They fail to distinguish between the surge expectations for lower acuity inpatient beds versus intensive care beds. They also provide a goal for surge capacity for a large geographical area without addressing how surge goals for individual hospitals should be determined.

Surge capacity, a common component of disaster planning in health care is generally considered to be the ability to manage a sudden increase in the number of patients presenting to a facility that might severely challenge or exceed the current capacity of that institution [4]. In addition to referring to an individual facility, the term can apply to a smaller unit, such as a pediatric intensive care unit (PICU), or to a larger one, such as a particular geographical region or entire health care system. In a 2007 study done by the New York City Department of Health and Mental Hygiene, a definition of surge capacity was established that addressed the need to uphold the level of care normally provided by each particular inpatient unit [5]. This definition required that the additional beds created be appropriately equipped to provide a level of care appropriate for the level of patient acuity, with the corresponding additional staff, supplies and equipment [5].

Hospitals in New York City have been encouraged by multiple health care organizations to do a certain amount of pediatric surge capacity planning [6]. However, according to a 2007 survey done by the New York City Department of Health and Mental Hygiene (NYC DOHMH) as part of the Pediatric Resource and Emergency Preparedness (PREP) Initiative, many pediatric intensive care physicians and emergency preparedness staff in NYC hospitals are still unaware of surge capacity plans that address pediatric patients in their hospitals [5]. This lack of planning is notably the case in the majority of the NYC hospitals with PICUs, and concerning in light of the fact that these hospitals are often expected to have the greatest pediatric resources.

Many hospitals seem to be at a loss as to how to design their plans. There is also a lack of understanding as to the degree to which each hospital should be able to surge [5]. It may be the case that PICU hospitals with a large amount of overall pediatric healthcare providers and equipment, such as those that are solely pediatric hospitals, should be able to surge by a greater percentage than hospitals with fewer pediatric resources. Hospitals that fail to focus exclusively on pediatric patients usually have fewer staff members that are comfortable caring for children, as well as less excess pediatric equipment that could be used to care for a surge of patients. If the pediatric resources in each PICU hospital can be determined prior to a disastrous event, this information can be used to both determine what the goal for each PICU hospital should be in surge capacity planning and what additional resources these hospitals may need to reach that goal (additional staff, beds, pediatric ventilators, etc.).

The development of a PICU stratification system for use in surge planning might help to simplify communications between PICU hospitals and city agencies such as the NYC DOHMH or the New York City Office of Emergency Management (OEM). These agencies play a role in helping PICUs obtain supplemental resources prior to and during a disaster. There also play a significant role in managing surge capacity on a citywide level during disaster. Having the ability to refer to each PICU by a predetermined stratification level should simplify conversations concerning regional surge planning prior to a disaster. In addition, being able to refer to PICUs by stratification level during a disaster while city agencies are attempting to properly distribute critical pediatric patients throughout the NYC region, should be easier and faster than having to discuss the individual clinical services that each PICU offers.

Another problem during a pediatric disaster may be the decreased availability of pediatric transfers, because of the decreased availability of transfer teams and vehicles, obstructive traffic conditions, and decreased bed availability at receiving institutions [7]. In the case of an event that was not citywide, but that affected a large amount of children in one place (such as a public school bombing), a large number of critically ill pediatric patients may unexpectedly present to the nearest PICU. If that PICU was a small unit with limited resources, there may be a tendency to transfer those critically ill children from the smaller PICU to a larger PICU hospital. However, in such a case, it may be better to limit these transfers to the sickest children who required the highest level of care, and for those children that could be managed in lower acuity PICUs to remain in place. It may also be better to distribute some of the less critical children to other small PICUs if they had the resources to handle them. A system of PICU stratification would predetermine the level of care that each PICU should be able to give in accordance with their resources. This may help to limit interhospital transfers to high acuity PICUs to those patients that truly require it for survival.

The goal of this project was to create a general system of NYC PICU stratification that would aid physicians, hospitals and city agencies in regional surge capacity and emergency preparedness planning for critical pediatric patients. To achieve this goal, the following two objectives were employed: (1) to describe the major factors to be considered when designing a PICU surge stratification system for PICUs in NYC and (2) to devise a basic system of PICU surge stratification based on clinical criteria and resources designed to aid both hospitals and city agencies in regional pediatric surge capacity planning.

It is important to note that the stratification system described in this paper was designed to be only one of several tools used in NYC PICU surge capacity planning. In addition to its use, it would be advisable for each NYC PICU hospital to develop its own facility-specific surge capacity plan describing the expansion of bed capacity and accompanying resources.


This project was determined by the Institutional Review Board of the SUNY Downstate Medical Center to be exempt from review. The project did not deal with individual level patient data. In addition, because the project was not designed to result in generalizable scientific information, it was determined not to meet the definition of research.

Definition of Terms

Prior to examining ways to stratify PICUs, surge capacity or other aspects of disaster planning, it was first necessary to define certain terms. PICUs have been defined in several ways by several different organizations. A general definition given by the American College of Critical Care Medicine and the Society of Critical Care Medicine (SCCM) states that a PICU should ideally provide definitive care for a wide range of complex, progressive and rapidly changing medical, surgical, and traumatic disorders occurring in all pediatric patients aside from premature newborns [8]. However, if certain needed services are not available in a particular PICU, that unit should be able to provide stabilization of critically ill children before transfer to a more appropriate facility. The SCCM also indicates that the goal should be to avoid long-distance transfers of patients with less complex or acute disorders if it is possible to care for them in their current location. In addition to this general description, the SCCM also outlines numerous specific recommendations for PICUs and the various services that they should be able to provide [8].

While there does seem to be a general consensus concerning the guidelines set by the SCCM, many of the inpatient units currently providing pediatric critical care in NYC may not meet all of the specific criteria described. There has been some debate over whether units that fail to meet some of the more specific criteria should be allowed to call themselves PICUs and whether they should instead be required to use a different title, such as a Pediatric Special Care Unit. However, much of this debate is over PICUs in the context of normal daily operations. It has been suggested that in a disaster situation in which pediatric critical care resources are stretched to the maximum, that many of the SCCM requirements would be of lesser immediate importance.

The definition of a PICU set forth in a study conducted by the NYC DOHMH in 2007 created a definition of a PICU designed to capture all the inpatient units providing some form of pediatric critical care in NYC for the purposes of determining NYC PICU surge capacity and other aspects of disaster planning. This definition required only that these units see pediatric patients up to at least 17 years old but not commonly past the age of 21, that they be able to provide 24 hour cardiorespiratory monitoring, mechanical ventilation, and that they have an attending specifically dedicated to the unit and available 24 hours a day [8, 9].

The NYC DOHMH PICU definition is designed to apply in situations where the need for PICU care stretches or outstrips the resources available. Such a situation might occur on a local level, such as surge in a particular PICU located near a school where a bomb might have gone off. In such a situation, the hope is that the surge at that particular PICU would be limited, as critical pediatric patients would be appropriately distributed to other PICUs in NYC. However, an event that created an extended surge of critical pediatric patients across NYC (such as an influenza pandemic might) would require a different type of management. Such a situation might require the provision of something called mass critical care (MCC).

MCC, as described in manuscripts resulting from the 2007 Task Force for Mass Critical Care Summit Meeting, is designed to be implemented in the case of a disaster yielding overwhelming numbers of critically ill victims, and consists of a limited number of essential, life-sustaining critical care services for those patients who are most likely to survive [1012]. This type of critical care also has a population-oriented approach to patient care, as opposed to the individual, patient-focused approach commonly used in normal daily operation. It is also controversial and designed to be used in only the most extreme surge situations.

Reference Materials

The majority of reference materials for this paper were obtained via an online search of peer reviewed journals via the PUBMED database using such keywords as pediatric intensive care, critical care, surge capacity, disaster preparedness planning, and mass casualty critical care. The search was focused on pediatric surge capacity and critical care surge capacity. A search for information on stratification, pediatrics, and/or critical care yielded few results, with no evidence of PICU stratification systems for surge or disaster being available. All references obtained had been published between 2002 and 2009. Information was also taken from the official websites of governmental organizations such as the NYC DOHMH, the New York State Department of Health, and the Agency for Healthcare Research and Quality [5, 9, 1315].

Once the stratification criteria to be used in the NYC PICU surge stratification system had been determined, data was taken from the NYC DOHMH New York City Pediatric Intensive Care Directory [9]. Data for this directory had been obtained from the PICU medical directors and emergency preparedness coordinators for each NYC PICU hospital via computerized survey, in-person interview, and hospital site-visit. This information was used to determine the clinical services offered by each NYC PICU hospital and to stratify them appropriately.

The NYC PICU stratification system outlined in this paper was also reviewed in various stages by the medical director and a medical coordinator at the NYC DOHMH Healthcare Emergency Preparedness Program. (This review was provided for educational purposes. Neither the NYC DOHMH or the staff involved officially endorsed the project.) Feedback offered by these individuals, both experts in their field, helped to shape the list of relevant stakeholders, as well as the basic characteristics of the system itself.

Major Factors to Consider in Designing a PICU Surge Stratification System

Prior to developing the NYC PICU surge stratification system, there were several areas that needed to be addressed. First, an examination of existing methods of ICU stratification (for both pediatric and adult patients) for use under normal non-disaster conditions was conducted. While an investigation of pediatric and adult ICU stratification methods for use in surge or disaster situations would have been desirable, none were found during the literature search. However, a NYC hospital-based stratification system of burn care for use in surge situations was described, and is discussed later in this section of the paper. Since the PICU stratification system was designed to stratify PICU hospitals on the basis of clinical services offered, there was some research done on which clinical services might be most desirable in a disaster. Finally, the development of any public health program requires the involvement of appropriate stakeholders in the earliest stages of planning. This topic is also addressed later in this section of the paper.

Systems of Stratification in Critical Care for Use in Normal Non-Surge Conditions

While there was little easily obtained information on ICU stratification systems for use during disasters, there were several examples of systems designed for everyday use.

The most commonly accepted system of stratification in PICU care was delineated by the American College of Critical Care Medicine of the SCCM [8]. These guidelines, originally published in 1993, were updated and re-released in 2004. They are largely designed to be used under normal operating conditions and stratify all PICUs into two levels.

Level 1 PICUs are considered to be the ideal setting for the provision of pediatric critical care and have numerous requirements addressing admission and treatment policies, the 24 hour availability of a pediatric intensivist as well as pediatric anesthesiology, neurosurgery, cardiology and a host of other subspecialties. There are also requirements for the physical facility in terms of numbers of electrical outlets per bed, computerized laboratory reporting and radiological services. These requirements also include training and continuing education on the part of the physicians and nursing staff. The requirements for level 2 PICUs are similar to the ones for level 1 units, but slightly less stringent.

There are other systems of stratification for adult intensive care units (ICUs). The American College of Critical Care Medicine outlines a three level system for the stratification of adult ICUs, again with the level 1 units being required to meet the most stringent criteria [16]. However, this system, unlike the two-level pediatric one, allows for level 3 units, which may be located in community hospitals not associated with academic centers, and which may be able to stabilize critically ill patients, but unable to provide comprehensive critical care. These units are expected to have written policies addressing the transfer of critically ill patients to a level 1 or 2 unit, when the care required outstrips the resources available. The SCCM also suggests that there be a certain level of cooperation between hospitals within a geographical region to assure that an adequate number of level 1, 2 and 3 ICUs are available and that care is appropriately distributed among them [16].

The Veterans Health Administration (VA) has proposed yet another way of classifying adult ICUs [17]. This system divides ICUs into four levels, with level 1 having the most stringent criteria. This system, like the SCCM adult ICU system, allows room for ICUs that, while not having the resources to provide comprehensive care, are able to stabilize critical patients and provide basic life-sustaining and monitoring services. It also allows for the inclusion of level 4 units at small rural hospitals.

Again, these more limited ICUs are expected to have the ability to transfer patients to a higher level ICU when appropriate, and adequate regional distribution of the various levels of units is desirable [17]. Both the SCCM and VA adult ICU stratification systems differ from the pediatric system in that they take into account the smaller ICUs that are still able to provide some level of critical care. These units that fulfill a useful roll under normal operating conditions are likely to be even more important in a disaster situation when ICU beds may be scarce. As a result, it seemed advisable to include these smaller units in the NYC PICU surge stratification system.

Systems of Stratification for Use in Disaster, Surge or MCC Situations

While the systems of ICU stratification described above are largely designed to address critical care under normal operating circumstances, there have been systems, albeit not PICU or ICU focused, created to address the provision of medical care in a disaster situation.

A working group associated with the New York Presbyterian Healthcare System, and funded via the NYC DOHMH Healthcare Emergency Preparedness Program, developed a system to address a potential event involving 400 adult and pediatric victims of a burn disaster on a regional level [18]. This system was particularly desirable due to the low number of actual burn centers currently located in NYC.

Four levels of burn disaster receiving hospitals (BDRHs) were designated, based on their clinical resources. In addition, supplemental training and supplies for burn care were provided to the lower tier hospitals. The plan allowed for burn victims to be sent to local or regional burn centers (tier 1 BDRHs) for initial care until full capacity was reached. At that point, additional burn victims were to be cared for in non-burn center facilities (tier 2 through four BDRHs) for up to 3–5 days until a city, regional or national burn bed became available. The ultimate triage of burn patients in a burn event large enough to require the activation of the four tiered system was to be handled by a centrally located triage team.

This plan was well designed in that it aimed to create additional surge capacity for a particular type of victim requiring specific and labor intensive care through pre-planning. It also made use of facilities previously thought to have resources too limited to be useful by stratifying them by their ability to provide the clinical care and supplementing their knowledge and supplies as needed. The reasoning behind this approach was that early management of burn patients could be managed by trauma centers and that specialty treatment would really only be required later in the burn patients care. The organization of the various facilities into only four levels of stratification also created a form of shorthand that makes it possible to immediately assess the resources available at a particular BDRH.

Finally, the system also made use of planning on the regional, state and national levels. All of these aspects might also be applied to disaster and surge capacity planning for other types of specialized and somewhat limited care, such as that which is required for pediatric intensive care patients.

The importance of coordinated disaster planning over a larger region, such as the five boroughs of NYC, in disaster planning has been repeatedly referred to in the relevant literature [2, 4, 15, 19]. The American Academy of Pediatrics in combination with the SCCM released a statement commenting on the importance of the regionalization of services for critically ill or injured children [19]. Because the resources needed to care for these children can be limited, regionalization can allow for their more appropriate distribution. Additionally, the regionalization of other specialized resources, such as neonatal intensive care, and trauma, has been associated with better patient outcomes [19].

Along with a focus on regionalization, the altering of standards of care in extreme cases, such as those that would exist in a mass critical care situation may often be required [2022]. A 2007 study looked at expected mortality in a hypothetical sudden disaster involving 500 children per million population needing hospitalization, with 30% of them requiring PICU care [20]. The researchers found that mortality decreased by 11% when regional surge distribution was used to spread out the patients between area PICUs. The mortality dropped by 24%, when altered standards of care were used to quadruple pediatric ICU and non-ICU capacity. The concepts of regionalization and altering standards of care to accommodate more patients are worth incorporating into any stratification system designed to be used for such specialized care as PICU or burn treatment in a disaster setting.

Clinical Services that PICUs Should Provide in a Disaster, Surge or MCC Situation

The NYC PICU surge stratification system was conceived of as being based on the clinical services that each NYC PICU was able to provide. As a result, it was necessary to look into which clinical services might be more important than others in a disaster or surge situation. Using the definition of a PICU used in the NYC DOHMH study mentioned earlier, a PICU in a disaster setting is any unit able to provide 24 hour cardiorespiratory monitoring, mechanical ventilation and 24 hour a day coverage by a pediatric intensivist or other attending physician specifically assigned to cover the unit [5, 9]. This definition already limits the clinical services that a PICU should provide. However, in a disaster situation, there is a larger, and yet still limited number of clinical services that a PICU should offer.

The Task Force on Mass Critical Care delineated five services that any ICU should be able to provide in a mass critical care situation [10]. These were mechanical ventilation, IV fluid resuscitation, vasopressor administration, the administration of medications for specific disorders such as antimicrobials and antidotes, and sedation and analgesia. They also indicated that ICUs should be able to perform practices to reduce the adverse consequences of critical illness and injury, such as keeping the head of mechanically ventilated patients at 45 degrees to reduce the risk of pneumonia, or providing thromboembolism prophylaxis to bed-ridden patients [10].

In a mass critical care situation, these would be the only services provided in order to conserve resources to be able to treat and save the largest number of patients. However, in a less severe situation still resulting in PICU surge, while all PICUs would still be required to provide a minimum of the above services, patients requiring further care would ideally be transferred to other PICUs within the region able to provide additional needed services. Some examples might be the need for pediatric surgery, orthopedics, neurosurgery, neurology, nephrology, burn care, otolaryngology, or hematology/oncology services.

Stakeholders in Designing a PICU Surge Stratification System

While the specific stakeholders in designing a PICU stratification system for surge planning may differ by geographical area, the general types of stakeholders should remain the same. In a 2004 article, Hicks suggested that key community stakeholders might include EMS and other patient transportation resources, emergency management and public health agencies, hospitals and health care systems, jurisdictional legal authorities, the American Red Cross, mental health providers and organizations, health professional training institutions and health care professional associations [4]. These stakeholders can be organized into six tiers of health care disaster response ranging from the Tier 1 individual health care facility response to the tier 6 federal response (Table 1) [4, 23].

Table 1 Tiers of health care disaster response

As for the stakeholders in the design and implementation of an NYC PICU surge stratification system, NYC PICU hospitals could constitute the first tier of disaster response. The other potential stakeholders might be categorized into health care disaster response tiers 2 through six as described in the fourth column of Table 1. In the case of the PICU surge stratification system, making the system known to EMS might allow for better distribution of casualties between PICU hospitals [7, 24]. The list of stakeholders mentioned here is far from all-encompassing. In addition, it may not be necessary to have equal involvement of stakeholders from all the response tiers for every disastrous event.

A Sample PICU Surge Stratification System for NYC PICU Hospitals

Considerations Specific to NYC in Designing a Surge Stratification System

There are several NYC-specific considerations when designing a PICU surge stratification system. NYC is a large city with a high population density. It is also divided across several bodies of land separated by water. As a result, while there may be a certain number of available PICU beds, in the case of an unusable bridge or tunnel (due to actual destruction or chaotic traffic conditions) access to many of these beds may be curtailed. Consequently, while more complicated critical patients are usually transferred to large tertiary care hospitals, a significant percentage of which are located in the borough of Manhattan (Fig. 1), this may not be possible in a disaster situation [5]. In addition, while attempts may be made by FDNY EMS or the Office of Emergency Management to distribute patients throughout the five boroughs, the proximity of certain hospitals to the precipitating event may result in a localized surge of critical pediatric patients. In such cases, it may be important to be aware of the pediatric critical care capabilities available even in small community hospitals and to include them in the stratification system.

Fig. 1

The distribution and size of pediatric intensive care units in the five boroughs of New York City

The stratification system proposed here will not suggest actual surge capacity goals for NYC PICU hospitals in terms of numbers of beds or percentage increases in bed capacity. However, specific surge goals should be determined as part of these hospitals overall disaster plans, and the stratification system might be used to aid in this task. For example, PICU hospitals with a higher stratification level, assigned due to a greater degree of pediatric critical care resources, might be required to surge by a greater percentage of beds than hospitals with lower stratification levels. The stratification system will also not take into account hospital size or number of beds, although it is again likely that hospitals with a greater degree of pediatric critical care resources and higher stratification level will be larger facilities with a greater number of staffed PICU beds.

The Clinical Criteria

The system will require that all PICUs included be able to provide the five basic critical care services as outlined by the Task Force of Mass Critical Care [10]. These services are reiterated below.

  1. 1.

    Mechanical Ventilation

  2. 2.

    Vasopressor Administration

  3. 3.

    Medications for Specific Disorders (ex:, antimicrobials, antidotes)

  4. 4.

    Sedation and Analgesia

  5. 5.

    Practices to Reduce Adverse Consequences of Critical Illness and Injury (ex: elevating the head of mechanically ventilated patients by 45 degrees to reduce pneumonia risk)

Additional services beyond those listed above will be required for PICUs at the higher stratification levels. Because of current concerns about a possible influenza pandemic, services needed to care for the resulting patients should be included. In addition, blast injuries from explosions have historically resulted in high numbers of casualties, both domestically and internationally [25, 26]. As a result, the clinical services needed to care for these patients, such as ophthalmology, neurosurgery and orthopedics, should also be considered. For influenza, pulmonology and other clinical services needed to care for critical patients are largely covered by the five basic clinical services listed above. However, in the case of blast injury, in addition to lung damage, ocular, aural, traumatic brain, and orthopedic damage are also common [25, 26].

The Different Levels of Stratification

The stratification system outlined by the American College of Critical Care Medicine uses a three level system to classify adult ICUs [8]. This breadth of classification would seem to be the most appropriate for NYC PICUs during disaster. The current AAP/SCCM 2 tiered system would be too restrictive in that its stringent criteria would exclude many PICUs that regularly deliver the five basic critical care services. The Veterans Health Administration four level system would probably be too broad, in that the fourth level of stratification is designed to include small rural hospitals, which do not exist in the NYC area [17]. In addition to using an SCCM-type three level system, the region-based organizational aspect of the New York Presbyterian/NYC DOHMH system of burn disaster receiving hospitals might also be incorporated [18].

Since the PICU surge stratification system is designed for regional/citywide use, activation of the system (the official change from normal operating conditions to surge or disaster conditions) will have to come from outside the individual PICU hospitals (OEM, or the NYC DOHMH) although individual facilities will be able to contact the appropriate agency to let them know that there may be a growing need for activation.

The system would be designed to work on a regional level and be managed by a centrally located PICU triage and resource monitoring team. This team may be newly created, or might incorporate pre-existing information response systems, such as the NY State Health Emergency Response Data System (HERDS) system, which can be accessed and updated in real time via any computer with internet access [27]. The HERDS system would also be advantageous in that it is designed to handle both surveillance data and health facility asset tracking, as well as to provide a link between regional (NYC) and state health emergency responses.

The five basic critical care services would be the clinical criteria for level 3 NYC PICU hospitals. These hospitals would likely consist of smaller community hospitals with more limited pediatric resources. Level 1 and 2 hospitals would be required to have the minimum services needed to deal with most blast and other traumatic injuries, and would likely be associated with academic centers. Most level 1 hospitals would be within tertiary care centers and able to provide most pediatric clinical services. The three levels of the PICU stratification system, their criteria, and the NYC hospitals that would fall into each level are listed in Table 2. For hospitals to meet the criteria, the subspecialty services must be available to treat all pediatric patients, as opposed to limited age groups (such as only adolescents). In this case, a pediatric patient, will be defined as anyone ranging in age from 0 to at least 17 years, but not routinely more than 21 years at the time of hospital admission [9]. These services must also be potentially available 24 hours a day.

Table 2 The NYC PICU surge stratification system

The criteria included in level 2a would be recommended in the case of a surge event with a large component of trauma patients resulting from a blast. As seen in Table 2, the inclusion of the ability to provide burn treatment in the level 2a criteria kept many hospitals, some being major tertiary care centers, from qualifying for that level. This suggests that the need for hospitals to be able to provide burn treatment in the event of a surge may be a major problem that needs to be addressed by such programs as the aforementioned NY Presbyterian/NYC DOHMH burn disaster response plan. For an event stemming from an infectious disease threat, the level 2 criteria would be of less importance.

In addition, the criteria for levels 1 and 2 tend to center more on the surgical subspecialties than the medical ones, such as nephrology, pulmonology or hematology. This is because surgical subspecialty involvement, if required, would likely require the presence of the subspecialty physician to personally perform an intervention. In the case of the medical subspecialists, many of their interventions and recommendations could be communicated at a distance (via phone or email) and then carried out by another physician at the bedside. As a result, a number of medical subspecialists might be able to influence the treatment of a larger number of critical patients than the same number of surgical subspecialists. Indeed, medical subspecialists might be able to cover patients in more than one facility. This medical information might also be communicated to multiple NYC PICU hospitals via messages from a central source, such as an antibiotic recommendation communicated by the NYC DOHMH in the case of a biological terrorism event.


Considerably less effort has been put into planning for children than for adults in hospital-based disaster preparedness. While there has been some work in terms of individual PICU preparation for disaster and surge planning, there has been little work done on the sort of regional planning that has been recommended by the AAP and other medical societies and organizations [19, 2830]. This paper has attempted to discuss many of the issues and arguments surrounding regional organization in PICU surge capacity planning and to create an example of a basic system of NYC PICU stratification to be used as a tool in that planning.

Some of the areas that were examined in preparation for designing the stratification system were the stratification systems currently in existence. These ranged from those looking at pediatric and adult ICUs under everyday circumstances, to systems designed to be used in disasters, such as the burn disaster response plan [18]. We then selected the clinical services that would be used to stratify PICUs into the various levels. To do this, we chose to focus on the services that would be needed to treat patients both in a medical disaster, such as an influenza pandemic, or a traumatic disaster, such as an explosive blast. There was some discussion of the various stakeholders that should be involved in the design and implementation of the system. Finally, the existing NYC PICU hospitals were stratified into three levels based on the clinical services they offered. This data was obtained from the publicly-available NYC DOHMH directory of NYC PICU resources.

The main idea behind designing a NYC PICU stratification system was that the clinical in individual NYC PICUs might be better communicated to regional emergency planners prior to an event in a straightforward and easy to grasp language. Having the ability to refer to each PICU by stratification level would simplify conversations concerning regional surge planning prior to a disaster. In addition, being able to refer to PICUs by stratification level during a disaster when city agencies are distributing patients throughout NYC, should be easier than having to discuss the individual clinical services that each PICU offers. However, while such a system might make it easier to manage PICU surge both before and during an incident, there are several potential barriers to designing, implementing and maintaining such a system.

A major complication in designing such a system is that it would need to be extensively vetted by disaster planers. Ideally, these planners would have an in-depth knowledge of pediatric critical care. This was an issue that was repeatedly mentioned by the experts from the NYC DOHMH Healthcare Emergency Preparedness Program who reviewed this project during its development. For the purposes of this particular academic project, the decision was made to discuss the stratification system with a limited number of individuals with experience in pediatric disaster planning. This was due largely to both time and logistical constraints. However, if steps were taken to design a PICU stratification system for actual use, this process of working with various experts and stakeholders, and repeatedly making changes would be vital to both getting the project off the ground, and having the system function properly when tested. It is likely that even something as simple as the selection of which planners would be able to first view and comment on the project would have political ramifications. It is also likely that it would be difficult to convince key individuals to invest time and effort into the project without the backing of at least one well known health care organization or NYC agency. This interactive and political process leading to a final compromise, while common when attempting to effectively deal with public health issues, would be difficult and time consuming.

The PICU stratification system discussed here would be based on clinical services offered. However, the services that specific hospitals offer can change over time. Smaller hospitals, in which a subspecialty department consists of a single subspecialist, are vulnerable to functionally loosing that department on short notice if the individual chooses to leave or becomes ill. Other services that may require significant funds to provide (such as pediatric hemodialysis) may be lost if the larger department or hospital finances are compromised. Such changes might render a surge stratification system unreliable if it is not kept up to date. In addition, individual hospitals may gain new clinical services, also possibly requiring a change in the hospitals stratification level. This may happen on an individual hospital basis, or on a broader basis, such as in the case of multiple hospitals gaining the ability to better care for burn victims due to the NYC DOHMH/New York Presbyterian burn disaster response plan. To monitor these changes and to alter the stratification levels appropriately will likely require additional staff hours, which will raise the cost of maintaining the system.

Another potential problem with the design of the NYC PICU surge stratification plan is that it defines the five boroughs of NYC as the region of interest. From an administrative and governmental standpoint, this may make sense, particularly because there are numerous city governmental agencies that deal with disaster planning and response specifically within NYC. However, from a geographical standpoint, it may be more beneficial to widen the physical area that the plan addresses. Healthcare facilities in the Long Island and Westchester areas are in very close proximity to the five boroughs and perhaps should be included as well. To do this would likely require the involvement of additional New York State-based agencies as well as an even wider group of potential stakeholders.


The NYC PICU surge stratification system, the burn disaster response plan and other similar regional endeavors, while only tools to be used in surge capacity planning and management, are the next step in the development of regional disaster preparation for both adult and pediatric patients. To actually implement such plans will require further investigation and development through dealings with local governmental organizations and other stakeholders. This is likely to be a time-consuming and difficult process. However, hopefully through further public health efforts in this area, and with a little perseverance, we can move a little closer to being able to plan for the unthinkable.


  1. 1.

    National Bioterrorism Hospital Preparedness Program Cooperative Agreement Guidance, Health Resources and Services Administration. (2003). PDF document. Retrieved from

  2. 2.

    Kanter, R. K., & Moran, J. R. (2007). Hospital emergency surge capacity: An empiric New York statewide study. Annals of Emergency Medicine, 50(3), 314–319.

    Article  PubMed  Google Scholar 

  3. 3.

    Burkle, F. M., Jr. (2006). Population-based triage management in response to surge-capacity requirements during a large-scale bioevent disaster. Academic Emergency Medicine, 13(11), 1118–1129.

    Article  PubMed  Google Scholar 

  4. 4.

    Hick, J. L., Hanfling, D., Burstein, J. L., & DeAtley, C. (2004). Health care facility and community strategies for patient care surge capacity. Annals of Emergency Medicine, 44, 253–261.

    Article  PubMed  Google Scholar 

  5. 5.

    New York City Department of Health and Mental Hygiene. The Pediatric Resource and Emergency Preparedness Initiative, PICU Phase [PDF document]. Retrieved from New York City Department of Health and Mental Hygiene Online Web site:

  6. 6.

    Markinson, D., & Redlener, I. (2003) Pediatric preparedness for disasters and terrorism a national consensus conference. National Center for Disaster Preparedness 2003 [PDF document]. Retrieved from

  7. 7.

    Raiter, Y., Farfel, A., Lehavi, O., Goren, O. B., et al. (2008). Mass casualty incident management, triage, injury distribution of casualties and rate of arrival of casualties at the hospitals: Lessons from a suicide bomber attack in downtown Tel Aviv. Emergency Medicine Journal, 25, 225–229.

    CAS  Article  PubMed  Google Scholar 

  8. 8.

    Rosenberg, D. I., & Moss, M. M. (2004). The American college of critical care medicine of the society of critical care medicine. Guidelines and levels of care for pediatric intensive care units. Critical Care Medicine, 32(10), 2117–2127.

    Article  PubMed  Google Scholar 

  9. 9.

    New York City Pediatric Intensive Care Directory. (2007). PDF document. Retrieved from New York City Department of Health and Mental Hygiene Online Web site:

  10. 10.

    Rubinson, L., Hick, J. L., Hanfling, D. G., Devereaux, A. V., et al. (2008). Definitive care for the critically ill during a disaster: Medical resources for surge capacity from a task force for mass critical care summit meeting, January 26–27, 2007, Chicago, IL. Chest, 133(5), 18–31.

    Article  Google Scholar 

  11. 11.

    Rubinson, L., Hick, J. L., Curtis, J. R., Branson, R. D., et al. (2008). Definitive care for the critically Ill during a disaster: Medical resources for surge capacity from a task force for mass critical care summit meeting, January 26–27, 2007, Chicago, IL. Chest, 133(5), 32–50.

    Article  Google Scholar 

  12. 12.

    Devereaux, A. V., Dichter, J. R., Christian, M. D., Dubler, N. N., et al. (2008). Definitive care for the critically ill during a disaster: Medical resources for surge capacity from a task force for mass critical care summit meeting, January 26–27, 2007, Chicago, IL. Chest, 133(5), 51–66.

    Article  Google Scholar 

  13. 13.

    Pediatric Disaster Toolkit: Hospital Guidelines for Pediatrics during Disasters (2nd Edition 2006). Retrieved from New York City Department of Health and Mental Hygiene Online Web site:

  14. 14.

    New York State Department of Health Disaster Preparedness Draft Guidelines (February 14, 2002 Edition). New York State Department of Health and the Greater New York Hospital Association. Retrieved from Greater New York Hospital Association Online Web site:

  15. 15.

    Bravata, D. M., McDonald, K., Owens, D. K. Regionalization of Bioterrorism Preparedness and Response. Prepared for: Agency for Healthcare Research and Quality, U.S. Department of Health and Human Services [PDF document]. Retrieved from

  16. 16.

    Haupt, M. T., Bekes, C. E., Brilli, R., Carl, L. C., et al. (2003). Guidelines on critical care services and personnel: Recommendations based on a system of categorization of three levels of care. Critical Care Medicine, 31, 2677–2683.

    Article  PubMed  Google Scholar 

  17. 17.

    Almenoff, P., Sales, A., Rounds, S., et al. (2007). Intensive care services in the veterans health administration. Chest, 132, 1455–1462.

    Article  PubMed  Google Scholar 

  18. 18.

    Yurt, R. W., Lazar, E. J., Leahy, N. E., et al. (2008). Burn disaster response planning: An urban region’s approach. Journal of Burn Care and Research, 29(1), 158–165.

    PubMed  Google Scholar 

  19. 19.

    Consensus Report for Regionalization of Services for Critically Ill or Injured Children. American Academy of Pediatrics, Committee on Pediatric Emergency Medicine and American College of Critical Care Medicine and Society of Critical Care Medicine, Pediatric Section, Task Force on Regionalization of Pediatric Critical Care (2002). Pediatrics, 105, 152–155.

  20. 20.

    Kanter, R. K. (2007). Strategies to improve pediatric disaster surge response: Potential mortality reduction and tradeoffs. Critical Care Medicine, 35(12), 2837–2842.

    Article  PubMed  Google Scholar 

  21. 21.

    Rubinson, L., Nuzzo, J. B., Talmor, D. S., O’Toole, T., Kramer, B. R., & Inglesby, T. V. (2005). Augmentation of hospital critical care capacity after bioterrorist attacks or epidemics: Recommendations of the working group on emergency mass critical care. Critical Care Medicine, 33, 2393–2403.

    Article  PubMed  Google Scholar 

  22. 22.

    Mahoney, E. J., Biffl, W. L., & Cioffi, W. G. (2008). Mass-casualty incidents: How does an ICU prepare? Journal of Intensive Care Medicine, 23(4), 219–235.

    Article  PubMed  Google Scholar 

  23. 23.

    Barbera J., & Macintyre, A. (2004). Medical surge capacity and capability: A system template for managing medical and health resources during large-scale emergencies. Alexandria, Virginia: CNA Corporation. Retrieved from

  24. 24.

    Farmer J. C., & Carlton P. K. Jr. (2006). Providing critical care during a disaster: The interface between disaster response agencies and hospitals. Critical Care Medicine, 34(Suppl.), S56–S59.

    Google Scholar 

  25. 25.

    Ritenour, A. E., & Baskin, T. W. (2008). Primary blast injury: Update on diagnosis and treatment. Critical Care Medicine, 36(7), S311–S317.

    Article  PubMed  Google Scholar 

  26. 26.

    Centers for Disease Control and Prevention, Mass Casualties, Explosions and Blast Injuries: A Primer for Clinicians. Retrieved from Centers for Disease Control and Prevention Online Web site:

  27. 27.

    Gotham, I. J., Sottolano, D. L., Hennessy, M. E., et al. (2007). An integrated information system for all-hazards health preparedness and response: New York State health emergency response data system. Journal of Public Health Management and Practice, 13(5), 486–496.

    PubMed  Google Scholar 

  28. 28.

    Parker, M. (2006). Critical care and disaster management. Critical Care Medicine, 34(Suppl.), S52–S55.

    Google Scholar 

  29. 29.

    Gomersall, C. D. (2006). Expanding ICU facilities in an epidemic: Recommendations based on experience from the SARS epidemic in Hong Kong and Singapore. Intensive Care Medicine, 32(7), 1004–1013.

    Article  PubMed  Google Scholar 

  30. 30.

    Roccaforte, J. D., & Cushman, J. G. (2002). Disaster preparation and management for the intensive care unit. Current Opinions in Critical Care, 8(6), 607–615.

    Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Christiana Campbell.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Campbell, C. The Benefits of Designing a Stratification System for New York City Pediatric Intensive Care Units for Use in Regional Surge Capacity Planning and Management. J Community Health 35, 337–347 (2010).

Download citation


  • Pediatric intensive care
  • Pediatrics
  • Disaster planning
  • Disaster preparedness
  • Surge capacity