METHODS

The toolkit was distributed to all 145 upstate NY hospitals via mail in August 2008. The toolkit was distributed to either the medical director or nurse manager contact determined to have responsibility for pediatric emergency preparedness planning; either title may fulfill this role, depending on the hospital organizational structure. Of the 145 upstate NY hospitals, 102 were designated by NYSDOH as perinatal centers (7 Regional, 13 Level 3, 18 Level 2, and 64 Level 1); 43 were not designated as perinatal centers. The hospitals in upstate New York included 101 with less than 250 beds and 44 with 250 beds or more; 84 of these hospitals had 1 or more pediatric beds (pediatric medicine/surgery, pediatric intensive care unit [PICU], and/or neonatal ICU [NICU]), and 61 did not. These hospitals represented a mix of rural, suburban, and urban facilities, as well as a combination of academic and nonacademic institutions.

A cross-sectional structured survey of these hospitals was conducted to evaluate awareness and utilization of the previously distributed toolkit and to identify whether sufficient emergency preparedness plans were in place for pediatric and obstetric patients as well as unaccompanied or displaced children and adolescents. Displaced children and adolescents were defined as those up to age 18 years who have been separated from their caregiver following an emergency event and who are temporarily under the care of the hospital until they can be reunited with their caregiver. The survey was granted exempt status by the NYSDOH Institutional Review Board because it did not meet the definition of human subjects' research due to its focus on program evaluation of the toolkit.

This report focuses only on results from the 10 primary questions related to pediatric preparedness, 7 of which required a “Yes/No” response. The other three questions contained one or more subquestions; two of these contained qualitative data in their subquestions (ie, “If No, provide reason” and “If Other, describe”). Questions centered on the hospital contact's awareness and review of the toolkit, staff pediatric clinical expertise, and appointment of pediatric physician and nurse coordinators. Hospital staff with expertise were defined on the survey as those who had the following criteria, per the discretion of the facility: a license issued by the NYS Department of Education; specialty skills, consisting of additional training and certification (in the specified discipline); and experience, consisting of clinical experience in hospital or clinic (in the specified discipline). If pediatric clinical coordinators had not been appointed, a reason was requested, using a drop-down menu with four answer options: “Cost,” “Implementation Not Started,” “No Person Available,” or “Other,” with opportunity for explanation. Hospitals were also queried about having an emergency management plan (EMP) to evaluate and treat pediatric patients, plans for housing pediatric patients in-place, whether staff and equipment were available to meet pediatric patients' medical needs, and whether the facility had a plan in place to track the movement of unaccompanied or displaced children and adolescents to another medical facility or supervised shelter for future reuniting with family (tracking plan).

The toolkit survey was electronically distributed via the NYS Health Emergency Response Data System (HERDS) in September 2009 to the hospital contact on record as having been sent the toolkit, with a requested respond-by date. HERDS was selected as the survey mechanism of delivery for two reasons: HERDS was thought to be most effective, since its information was viewed as an authoritative, credible source by hospitals; and hospitals were accustomed to logging-in to complete surveys and reports via this secure, electronic mechanism. At approximately half-way through the response period, a reminder was sent via HERDS for hospitals to complete and submit the toolkit survey. The HERDS system allowed only one response per facility, avoiding the issue of receipt and analysis of multiple responses.

Results for closed-ended questions were calculated and presented as both number and percentage of “Yes” responders. Two open-ended subquestions' qualitative results for “Other Responses” were reported but did not enable statistical analysis. Survey results were calculated aggregately, by hospital size, and by presence of one or more pediatric beds. A χ² analysis enabled detection of statistically significant associated variables as well as nonassociated variables. An odds ratio (OR) was calculated for each statistically significant association to show the difference in likelihood of occurrence between comparison groups.

RESULTS

Of the 145 surveyed hospitals, most facilities (116 [80%]) submitted fully completed surveys, whereas 4 (3%) facilities partially completed the survey, and 25 (17%) facilities had not started survey completion. Only submitted, fully completed surveys (116) were included in the data analysis (Table 1). No significant association was found between presence of one or more pediatric beds and survey responder status or when comparing responders to nonresponders in terms of hospital size (≥250 beds vs <250 beds). Hospital size was significantly associated with presence of one or more pediatric beds; hospitals with 250 beds or more were greater than 10 times more likely to have one or more pediatric beds than were hospitals with fewer than 250 beds (P <. 0001) (Table 2).

TABLE 1 Toolkit Survey Aggregate Responses for Pediatric-Related Questions

TABLE 2 χ² Analysis and Odds Ratios for Significant Associations

Nearly all (106 [91%]) of the responding facilities confirmed that the person listed on the survey was aware of the toolkit, and the majority of these responders (100 [86%]) confirmed that this person had reviewed the toolkit. The χ² analysis showed no significant association between hospital size and either toolkit awareness or review.

Overall, most facilities reported having physician and registered nurse (RN) pediatric expertise (109 [94%] and 105 [91%], respectively). Hospitals with staff physician and RN pediatric expertise were significantly more likely to have at least one pediatric bed (P =. 0061 and P =. 0023, respectively). The χ² analysis showed no significant association between hospitals that had staff physicians with pediatric expertise and those that had an EMP that included how to evaluate and treat pediatric patients.

More than one-half (75 [65%]) of responding facilities affirmed having an EMP in place that included an element regarding evaluation and treatment of pediatric patients. Slightly more (90 [78%]) of the responding facilities reported having an emergency plan with elements for pediatric patient housing in-place, should stabilization or transportation be unavailable. The majority of responding hospitals (93 [80%]) had plans that included elements on staff availability to meet pediatric patient medical needs, and most (98 [84%]) hospitals had emergency plans that addressed adequate in-house equipment for pediatric patients. Hospital size was significantly associated with having a plan to house pediatric patients in-place (P =. 0438) but not with any other pediatric EMP elements. Only slightly more than one-half (41 [62%]) of hospitals with one or more pediatric beds and merely 17 (34%) of hospitals without pediatric beds had a tracking plan in place for unaccompanied or displaced children.

A key point of guidance within the toolkit was the appointment of both a pediatric physician coordinator and a pediatric nurse coordinator (pediatric clinical coordinator). While most (109 [94%]) facilities reported physician pediatric expertise, just more than one-half (70 [60%]) of these facilities had appointed a pediatric physician coordinator, representing 80% of the hospitals with one or more pediatric beds and only 34% of the hospitals without pediatric beds. Similarly, less than three-quarters (47 [71%]) of hospitals with one or more pediatric beds had appointed a pediatric nurse coordinator and only a fraction (10 [20%]) of hospitals without pediatric beds had done the same. Hospital size was determined to have significant association with appointment of a pediatric clinical coordinator; hospitals with 250 or more beds were more than four times as likely to have assigned a pediatric physician coordinator and more than five times as likely to have assigned a pediatric nurse coordinator than those with fewer than 250 beds. Further, hospitals with one or more pediatric beds were nearly 8 times more likely to have appointed a pediatric physician coordinator and nearly 10 times more likely to have appointed a pediatric nurse coordinator than those without pediatric beds.

Appointment of pediatric clinical coordinators greatly impacted presence of all analyzed pediatric preparedness elements. Hospitals with an appointed pediatric physician coordinator were 4 times as likely to have an emergency plan that included how to evaluate and treat pediatric patients, 6.5 times as likely to have a tracking plan in place for unaccompanied or displaced children, more than 15 times as likely to have a plan to house pediatric patients in-place, and more than 7 times as likely to have in-house equipment to meet pediatric patient medical needs than those with no assigned pediatric physician coordinator. Similarly, hospitals with an appointed pediatric nurse coordinator were 3 times as likely to have an emergency plan that included how to evaluate and treat pediatric patients, more than 5 times as likely to have a tracking plan in place for unaccompanied or displaced children, 11.5 times as likely to have a plan to house pediatric patients in-place, and more than 6 times as likely to have in-house equipment to meet pediatric patient medical needs than were hospitals with no assigned pediatric nurse coordinator.

The most common (22 [44%]) reason reported for not having appointed a pediatric physician coordinator was “Implementation Not Started,” followed by “Other” (20 [40%]), “No Person Available” (7 [14%]), and “Cost” (1 [2%]) (Table 3). Follow-up explanations to responses of “Other” included the following: “in-house physicians provide this function”; “critical access hospital”; “no in-house pediatric unit/wing”; “limited pediatric experience available, inpatient admissions over 17 years of age”; “ED [emergency department staff] are PALS [pediatric advanced life support] certified”; and “did not know we needed.” Explanations of responses of “Other” for no appointment of a pediatric nurse coordinator were similar.

TABLE 3 Reasons for Not Appointing Pediatric Clinical Coordinators

The χ² analysis found statistically significant associations related to presence of one or more pediatric beds and all queried emergency plan elements, including presence of an EMP that included how to evaluate and treat pediatric patients, an emergency plan to house pediatric patients in-place, in-house equipment to meet pediatric patient medical needs, and a tracking plan for unaccompanied or displaced children (P <. 05). Further, χ² analysis showed even stronger association between appointment of either a pediatric physician or nurse coordinator and all evaluated emergency plan elements (P <. 01). In contrast, staff pediatric physician expertise was not significantly associated with having appointed a pediatric physician coordinator, having an emergency plan that tells how to evaluate and treat pediatric patients, or having a tracking plan for unaccompanied or displaced children.

DISCUSSION

Emergency preparedness training and planning for the triage, admission and/or transfer, and treatment of pediatric patients in the event of a public health disaster is an essential public health issue that every hospital ought to be prepared for, regardless of whether or not it typically treats children. Following the Katrina disaster, children's health suffered and many were displaced from their families, resulting in a significant pediatric preparedness focus.Reference Ferrer, Balasuriya, Iverson and Upperman9Reference Fox and Trimm10 To be better prepared for children when the United States faces the next disaster, President Obama and Congress charged the National Commission on Children and Disasters with the first-ever comprehensive review focused on children's preparedness needs.12 The American Academy of Pediatrics (AAP), American College of Emergency Physicians (ACEP), and Emergency Nurses Association (ENA), in their Joint Policy Statement–Guidelines for Care of Children in the Emergency Department, stress that a physician coordinator and a nurse coordinator be appointed for pediatric emergency care.13 Research has found that, despite the vulnerability of children, both prehospital emergency medical service (EMS) agencies and hospitals have significant deficiencies in pediatric EMPs.Reference Shirm, Liggin and Dick3Reference Ferrer, Balasuriya, Iverson and Upperman9 It is especially essential for hospitals that do not normally admit pediatric patients to be prepared for an influx of the pediatric population so that they adequately plan for staff, equipment, and protocols for the next mass casualty event. Results analysis from the toolkit survey revealed the overall need for improvement in terms of hospitals incorporating the pediatric patient group into their EMPs, particularly for those that are smaller and/or without pediatric beds.

A problematic, overall finding of this project was that although 91% of responding facilities confirmed awareness of the toolkit and most (86%) facilities confirmed that the person receiving the toolkit had reviewed it, utilization of some key toolkit recommendations had not occurred within the 13 months since the toolkit distribution. Just more than one-half (60%) of the facilities had appointed a pediatric physician coordinator and just less than one-half (49%) had appointed a pediatric nurse coordinator; both of these percentages are lower than the number of responding facilities claiming respective clinical expertise. The χ² analysis demonstrated that presence of one or more pediatric beds, and even more-so, assignment of a pediatric clinical coordinator (vs toolkit review or even a staff physician with pediatric expertise) were significantly associated with having an EMP in place that included pediatric evaluation, treatment, staff, medical equipment, and tracking, making assignment of the pediatric clinical coordinator roles perhaps the greatest imperative for a facility striving to improve pediatric preparedness. Hospitals with at least 250 beds were far more likely to have had appointed either a pediatric physician coordinator or pediatric nurse coordinator, and to have one or more pediatric beds than were smaller hospitals of fewer than 250 beds. These findings indicate that facilities that most need improvement in pediatric preparedness have not yet been reached, and focus should be placed on smaller facilities that do not typically see pediatric patients.

To attempt to answer the question of how hospital appointment of pediatric clinical coordinators can be encouraged and what barriers may exist, it is helpful to examine the hospitals' reasons for not having appointed clinicians to these positions. The responses of “Implementation Not Started,” “No Person Available,” and “Cost” could indicate the barriers of lack of time, resources, and/or perceived need. Evaluation of “Other” responses showed an apparent lack of awareness by some that it was necessary to have pediatric clinical coordinators in place, seen in the literal responses “Didn't know it was necessary” and “ED staff are PALS certified.” It must be noted that some facilities reported having in-house physicians that fulfill this role. Formalizing this role by aligning the responsibilities of the pediatric clinical coordinators under this title might aid preparedness efforts, as survey results demonstrated that appointment of clinical coordinators was significantly associated with presence of all evaluated emergency plan elements. Other responses stated that either pediatric patients or pediatric services are not found at that facility, which alluded to the perception that only a hospital that regularly admitted pediatric patients would need these positions in place. Similar responses were received regarding reasons for not having appointed a pediatric nurse coordinator.

Some qualitative toolkit survey responses to the reason for not having clinical coordinators for the pediatric population point to a perception among some hospitals without pediatric capacity that pediatric capabilities are “not necessary” for hospitals that do not already have these resources. This perception has been cited previously in the literature; it has been found that general hospitals did not anticipate a surge of pediatric victims in an emergency but thought they would be transported to a children's hospital.Reference Ferrer, Balasuriya, Iverson and Upperman9 There was an evident lack of awareness among respondents that, during a disaster, pediatric patients may arrive at hospitals that do not typically handle these patients, and further, that dependent on the type and degree of the disaster there may not be a viable option for patient transfer. It is known from past events that in an emergency people will go to the nearest or easiest-to-reach facility, whether or not needed services (such as pediatric intensive care) are normally offered there. Each facility is unique and hence needs a plan that is tailored to its own needs, resources, and circumstances, taking into account that the transfer of pediatric patients may not be feasible.

One potential channel for addressing the barrier of lack of pediatric preparedness resources is the NY Regional Resource Center (RRC) structure, which coordinates regionalized groups of hospitals to potentially pool resources (eg, emergency medical equipment, supplies, training, expertise); this concept of regionalization has been cited in the literature and recommended by the AAP and the American College of Critical Care Medicine (ACCCM).Reference Barbera, Yeatts and Macintyre14Reference Gamble, Hanners, Lackey and Beaudin15 Upstate New York is organized into eight geographic regions defined by counties, each with a trauma center acting as the RRC. Each RRC coordinates preparedness activities with the acute care hospitals and other health care partners within their region. The NYS RRC structure is implemented in a manner similar to other states (for example, California's structure is coordinated by mutual aid and administrative regions, whereas Indiana is divided by Department of Homeland Security Districts). Many health care organizations have developed similar coalitions for the purpose of strengthening health care systems through collaborative emergency planning.

Planning for adequate staffing to handle pediatric patients in the event of a disaster is a critical component for a hospital EMP, as is training that staff for diagnosis and treatment of pediatric patients. Implementing these positions has proven to have positive impact outside of this toolkit survey research—a 2003 survey of US emergency departments' pediatric preparedness based on these AAP/ACEP/ENA guidelines demonstrated that facilities with physician and nursing pediatric coordinators scored a higher pediatric-preparedness score than hospitals without these clinical coordinators.Reference Gausche-Hill, Schmitz and Lewis4 Additional training may be helpful for developing existing staff with pediatric expertise into pediatric clinical coordinators. Based on survey findings, additional training should be weighted toward hospitals without pediatric beds for maximum impact.

Statistically significant association was found between appointment of either a pediatric physician or nurse coordinator and each of the pediatric emergency plan elements about which the survey inquired. No association was found between having physicians with pediatric expertise and having a tracking plan for displaced children and adolescents, demonstrating that appointment of at least one pediatric clinical coordinator is more indicative of a facility having a comprehensive pediatric emergency plan in place. Appointing pediatric physician and nurse coordinators should be a priority for not only hospitals with pediatric beds but for every hospital as an essential step toward achieving pediatric preparedness.

LIMITATIONS

To our knowledge, this is the first known project evaluating hospital review and utilization of guidance communicated in a previously distributed pediatric emergency preparedness toolkit. Neither facility size nor presence of pediatric beds contributed to systematic bias in terms of survey response vs nonresponse, making the results likely a representative sample of upstate NY hospitals. However, this project has several limitations that should be noted. Survey data shared similarities with other pediatric preparedness surveys of hospitals in the literature and had a strong response rate of 80%; however, data were not available for the 25 nonresponding hospitals, and data were not analyzed on the 4 facilities that only partially completed the survey. Because it was a survey from NYSDOH, viewed by hospitals as a regulatory agency, this could have introduced response bias if perceived as an activities assessment. The data are self-reported and, therefore, there is no way to know that survey completers responded accurately or if the responses represented the true state of pediatric preparedness. Methods for evaluating preparedness have been shown to produce disparate results; accurate measurement of preparedness is challenging and subject to bias, with no universally accepted tool.Reference Kaji, Langford and Lewis16 While this survey assessed utilization of toolkit guidance, the survey could not possibly assess the effects of utilizing toolkit guidance; this survey was a first step to assess if toolkit guidance was incorporated. A future evaluation would be required to determine the effect of utilization. Finally, there was one year between toolkit distribution and survey distribution, so there may have been staff turnover since the initial toolkit distribution. At the same time, the length of time since toolkit receipt should have provided the facilities ample time for review and utilization of pediatric preparedness guidelines.

CONCLUSIONS

This project presents the results of a survey of upstate NY hospitals' familiarity with and utilization of the NY State Pediatric and Obstetric Emergency Preparedness Toolkit and application of key guidance in developing a hospital disaster plan that includes the pediatric population. Only one-half of the responding facilities indicated that they have a tracking plan in place for unaccompanied or displaced children and adolescents. While most responding facilities had physicians and nurses with pediatric expertise, a smaller percentage had appointed pediatric clinical coordinators or incorporated pediatric patients into facility EMPs. Hospitals with a large number of beds were more likely to have assigned a pediatric clinical coordinator. Hospitals with a pediatric clinical coordinator were more likely to have developed a pediatric EMP that includes evaluation, treatment, medical staff, medical equipment, housing in-place, and tracking of the pediatric patient population.

This project demonstrates a need for awareness that pediatric patients may arrive during a disaster at hospitals that do not typically handle pediatric patients and that there may be no option for transfer. Further effort should strive to increase hospitals' awareness that a disaster may bring atypical patients to any hospital, with particular focus on hospitals with fewer than 250 beds and those without pediatric beds. Tactics to improve awareness may include messages to hospitals via a state's health emergency response data system, incorporation of the awareness message into health emergency preparedness training exercises and drills, and development of disaster education materials for health care professionals and first responders focused on pediatric patients. Hospitals should be encouraged to proactively plan for the management of pediatric patients so that they may effectively serve the children of their communities during a disaster.

There is much opportunity for improvement in incorporating the pediatric patient group into facility EMPs. Appointment of a pediatric physician and/or RN coordinator may help facilities develop an EMP for this population. Utilization of the toolkit and its cited resources can save each facility time as it develops its unique plan. This project could be viewed as a benchmark for future work that measures pediatric preparedness by drills, exercises, or knowledge questionnaires. Further research should examine existing barriers, as well as identify best practices, for incorporation of the pediatric population into a hospital's disaster plan so that upstate NY hospitals are better prepared for the children who may arrive at their facility during a disaster.