A Statewide Assessment of Pediatric Emergency Care Surge Capabilities

This was a cross-sectional descriptive study to assess pediatric surge capacity and capability during standard and disaster operations in Massachusetts. Pediatric surge capacity refers to the ability to care for a high number of pediatric patients beyond standard operating capacity. Pediatric surge capability is the ability to manage pediatric patients requiring unusual or very specialized medical care.²⁶  Data for hospital-level pediatric licensed inpatient and ICU capacity was obtained from the Massachusetts Department of Public Health (MA DPH) licensing records in May 2021. The MA DPH licensing records are maintained by the Bureau of Health Care Safety and Quality and updated in real-time as license changes are approved with monthly data quality checks. Data on standard operations and surge capacity and capability were obtained via survey. The survey was sent to all Massachusetts acute care hospitals with any type of inpatient licensed beds (pediatric and/or adult). The Boston Children’s Hospital and MA DPH Institutional Review Boards deemed the study protocol exempt from review as the unit of interest was the medical center or hospital.

Using MA DPH records, we summarized hospital characteristics including annual pediatric ED volume, transfer center availability, and hospital trauma accreditation. Massachusetts state-wide population data were obtained using 2020 United States Census. We classified annual pediatric ED volume as follows: low volume <1800 patients per year, medium volume 1800 to 4999 patients per year, medium to high volume 5000 to 9999 patients per year, and high volume >10 000 patients per year.²⁷  Hospital trauma accreditation was defined and evaluated by the American College of Surgeons based on trauma resources present for adult and/or pediatric patients. Level I can provide care for every aspect of injury-from prevention through rehabilitation. Level II can initiate definitive care for all injured patients. Level III can provide prompt assessments, resuscitation, surgery, intensive care, and stabilization of injured patients. Level IV can provide advanced trauma life support before transport to a higher level trauma center. Level V is able to provide initial evaluation, stabilization and diagnostic capabilities, and prepare patients to transfer to a higher level of care.^28,29  We assessed pediatric inpatient bed capacity. In Massachusetts, licensed hospital beds are defined as the number of beds a facility is certified to maintain. We calculated hospital-level pediatric licensed inpatient beds and ICU beds in each Massachusetts hospital as of May 3, 2021.

Three pediatric emergency medicine physicians, including 1 with pediatric disaster expertise, (S.C., J.L., A.L.B.) in collaboration with the MA DPH Emergency Medical Services for Children office designed the survey. We designed the survey to be completed by the hospitals’ emergency management directors with assistance from physician and nursing leadership (See Hospital Survey in Supplemental Information). The survey assessed hospital standard operations and plans for pediatric surge situations.

The survey draft was reviewed by 4 pediatric emergency medicine physicians, 1 general emergency medicine physician, and regional public health officials and hospital emergency managers who are part of Massachusetts health and medical coordinating coalitions. We piloted the survey with 3 Massachusetts hospitals to ensure appropriate understanding of each question and to assess respondent burden. We edited the survey for clarity and brevity based on this feedback.

The final survey assessed surge capacity, clinical therapies, and provider availability for pediatric patients. Surge capacity was assessed by collecting information about additional pediatric bed availability through the addition of pediatric beds and/or conversion of adult beds for pediatric use. Specific clinical therapies for children at different ages during both standard and surge conditions were assessed, including extracorporeal membrane oxygenation, high frequency oscillation ventilation, standard mechanical ventilation, high flow nasal cannula, continuous nebulization, hemodialysis, and peritoneal dialysis.^30–32  The survey assessed the availability of clinicians who take care of pediatric inpatients in standard and disaster operations, including emergency medicine doctors, family medicine doctors, general pediatricians, pediatric emergency medicine clinicians, pediatric hospitalists, pediatric intensivists, advanced practice providers, anesthesiologists capable of managing pediatric airways, certified pediatric nurses, certified pediatric pharmacists, certified pediatric respiratory therapists, child life specialists, and NICU staff. The survey also assessed the presence of medical and surgical subspecialty availability for pediatric patients in both standard and disaster scenarios as listed in Table 4.

Lastly, the survey collected information about pediatric disaster policies, including the presence of a pediatric surge plan, coordination with other hospitals during surge events, procedures to manage a missing or abducted child, a family reunification plan, and a plan for a safe area for unaccompanied children.

Using ArcGIS Survey123 version 3.14, we emailed the finalized survey instrument to the emergency management director of every Massachusetts hospital with licensed inpatient beds and emergency departments (ED) from May to July 2021. Emergency management directors were instructed to complete the survey and they were advised to include physician and nursing leadership if needed to obtain the most accurate information. Weekly reminders were sent to nonresponders and regional health care coalition directors. Phone calls were made to nonresponders from July to August 2021. Regional health care coalition leaders were included on all emails starting in June 2021 and encouraged to contact their region’s nonresponding hospitals during their standing biweekly meetings. For survey nonresponders, we assessed their current licensed capacity, overall pediatric volume, and trauma accreditation via MA DPH data.

Our primary outcome was pediatric inpatient surge capacity, which we defined as the total available acute care hospital pediatric beds during standard operations (from the MA DPH data) plus the additional pediatric beds that could become available during surge conditions (from the survey). Our secondary outcomes included availability of pediatric clinical therapies as well as availability of specialists who routinely care for children and subspecialists who could care for children, during both standard operations and surge conditions. Additional outcomes included the presence of disaster plans and policies as described above.

We used frequencies and proportions to describe hospital-level categorical variables. We summarized hospital characteristics, hospital-level pediatric inpatient capacity, critical care therapeutic capabilities, and surgical and medical subspecialty availability under standard and surge situations. We also described hospital-level disaster policies and procedures. To assess the correlation between population and bed capacity, we calculated a Spearman correlation between the number of available pediatric beds and total population at the zip code level. To evaluate the odds of therapy availability and population, we first estimated a logistic regression model at the zip code level with any therapy during standard operations as the dependent variable and population (eg, zip code level population divided by 1000 to express effect estimates as odds changes per 1000 persons) as the independent variable. This model was repeated with any therapy availability during surge situations as the dependent variable.

To test if therapy availability differed by region, (eastern versus western Massachusetts), we estimated a logistic regression model at the zip code level with therapy availability as the dependent variable and region as the independent variable, with each observation (ie, zip code) weighted according to its population. Eastern Massachusetts is defined as all zip codes in Health and Medical Coordinating Coalitions regions 3, 4AB, 4C, and 5. Western Massachusetts is defined as all zip codes in Health and Medical Coordinating Coalitions regions 1 and 2.³³  This model was repeated with any therapy availability during surge situations as the dependent variable. All analyses were conducted using SAS version 9.4 software (SAS Institute, Cary, NC). We created geographical maps utilizing ArcGIS Pro version 2.9.0 software overlaying our findings with the general population distribution of Massachusetts based on 2020 census data.

Of 64 Massachusetts hospitals with licensed inpatient services, 58 (91%) completed the survey. The most common role for respondents was emergency management staff (78%, n = 45; Table 1). The majority of hospitals had medium (29%, n = 16) or medium-high (29%, n = 16) annual pediatric ED volume with no trauma designation (74%, n = 43). (Table 1) There were 6 hospitals that did not complete the survey, all of which had similar hospital characteristics to the survey respondents. (Supplemental Table 5)

During standard operations, 19% (n = 2159) of all inpatient beds in Massachusetts are available for patients less than 18 years old. These beds are available across 35 hospitals licensed for inpatient pediatric units and 29 hospitals that can use licensed inpatient adult non-ICU beds for patients less than 18 years old. Of the 50 hospitals that have an adult ICU, 30% (n = 15) will also admit ICU patients less than 18 years old.

During a surge, 8 hospitals (14% of all respondents) indicated that they could add 171 (2%) pediatric beds by either converting available adult inpatient beds or adding additional beds. (Supplemental Fig 3) Thus, during a surge, Massachusetts could provide 124 additional beds per 1 million children, expanding overall pediatric capacity from 2159 beds to 2310 beds. The details of pediatric inpatient bed capacity based on ICU status are shown in Table 2 and the geographic distribution is shown in Fig 1. We found a weak but statistically significant correlation (Spearman = 0.24, P < .01) indicating that as the total population within a Massachusetts zip code increases, the number of pediatric beds increases.

For those hospital units able to convert an adult bed for a pediatric patient during a disaster (n= 15 non-ICU hospital units and 18 ICU hospital units), the median minimum age they would admit was 15 years (range 0–18). Eleven of the non-ICU hospital units and 12 of the ICU hospital units indicated they would be able to convert adult inpatient beds for pediatric patients in a disaster, but they could not specify the number of beds, so these were not included in the surge bed tabulation. (Supplemental Table 6)

During standard operations, respiratory therapies most commonly available to pediatric patients included high flow nasal cannula (36% of respondent hospitals, n = 21), continuous nebulization (33%, n = 19), mechanical ventilation (28%, n = 16), and high frequency oscillation ventilation (12%, n = 7). Similarly, respiratory therapies were the most commonly available adult therapy that could be converted for a pediatric patient, with high flow nasal cannula being most common (69%, n = 40). (Supplemental Table 7)

Among all Massachusetts zip codes, for every 1000 person increase in population, the odds of being able to provide any therapy increases by 5% (95% confidence interval [CI] 1.03–1.08) during standard operations and increases by 7% (95% CI 1.05–1.10) during a surge event. When comparing eastern Massachusetts to western Massachusetts, during standard operations and a surge event, zip codes in the eastern Massachusetts have a 1% (95% CI 1.01–1.02) and 54% (95% CI 1.53–1.55), increased odds of being able to provide at least 1 therapy relative to zip codes in western Massachusetts, respectively. The geographic distribution of select clinical care therapies with their lowest possible age for administration during standard operations and surge situations is illustrated in Fig 2.

During standard operations, the most common specialty physicians available to care for children were general emergency medicine physicians (98%, n = 57) with 52% (n = 30) of respondent hospitals staffing nurse practitioners who can see pediatric patients. Overall, most hospitals did not have certified pediatric health care staff including nurses, pharmacists, respiratory therapists, or child life specialists. (Supplemental Table 8)

During standard operations, the most common surgical specialist available to care for children was general surgery (59%, n = 34) followed by orthopedics (48%, n = 28). The most common medical subspecialist able to care for children was radiology (69%, n = 40) followed by infectious disease (45%, n = 26). During a surge situation, orthopedics would additionally provide care (an additional 28% of hospitals, n = 16), making them overall the most common surgical subspecialty service that could additionally care for pediatric patients in a surge situation (total 76%, n = 44), and pulmonology (an additional 31%, n = 18) was the most common medical subspecialty service that could additionally care for pediatric surge patients. However, with the exception of general surgery and orthopedics, the majority of hospitals lack surgical specialists who could care for pediatric patients under any circumstance. (Table 3)

The majority of hospitals that responded are able to coordinate with others in the event of a surge (91%, n = 53). Most hospitals have a protocol for a missing or abducted pediatric patient (57%, n = 33). Otherwise, most hospitals do not have other pediatric-specific disaster policies such as a reunification protocol (45%), safe area for unaccompanied minors (36%), or a pediatric surge plan (22%). (Supplemental Fig 4)

We found that inpatient pediatric capacity during standard operations is 19% of all inpatient beds in Massachusetts. Although inpatient capacity could increase by 171 beds (2%) during a surge, Massachusetts would not meet the ASPR benchmark for a surge capacity increase of 20%. During both standard and disaster operations, some hospitals could provide respiratory therapies to children, but most hospitals lack surgical subspecialists for children under any circumstance. Moreover, the available beds and services are primarily located in high population areas, whereas services are lacking in rural areas.

Although previous studies have investigated pediatric capacity, our study is the first to assess state-wide pediatric clinical therapy capabilities and subspecialist availability during standard operations and surge situations across all hospital types, including community hospitals. The methodology in our study for assessing pediatric capacity and capabilities can serve as a model for other states to assess their surge capacity and capabilities and can easily be widened for regional assessments. The next step would be to expand this statewide assessment to the entire New England region to gain a more comprehensive understanding of capacity and capabilities since patients may cross state borders to access care.

The 2020 US census reports 72.8 million children reside in the United States, almost one quarter of the population.³⁴  This is in the setting of increasing surge events, especially active shooter events targeting children.³⁵  Although adults and children can both be affected during a surge event, children are often a significant proportion of victims, with some natural disasters reporting up to 43% of victims being less than 18 years old.^5–10  Similar to prior literature, most hospitals in our study did not have a pediatric surge plan. Consistent with our study, pediatric inpatient capacity has been shown to be insufficient for disaster situations when pediatric volume surges.^{15,25,36–39}  This is compounded by the fact that capacity strain, defined as excessive demand on a hospital’s resources and/or abilities, can result in adverse outcomes for patients, including a five-fold increase in mortality.^17–24 

In a surge, victims will head to the nearest hospital for care. Community hospitals will need to stabilize and continue to care for pediatric patients longer than they normally would. Our study shows the willingness of adult units in Massachusetts to care for pediatric patients during a surge, however they preferentially would admit older adolescents. Our findings are consistent with another multistate study where general hospitals preferentially admit patients aged 16 years and above.⁴⁰  There are no formal federal recommendations on the age range that pediatric surge beds should cover. However, past events have shown that children can represent a large percentage of victims. In the 2013 Haiyan typhoon, 21% of victims were less than 5 years old.¹⁰  In the 2009 H1N1 influenza pandemic, 31% of all hospitalizations were for children less than 17 years old.⁴¹  Although state-wide coordination efforts could attempt to coordinate older children to be cared for in community hospitals with children’s hospitals concentrating on younger children, this may not be feasible in all states. Therefore, adult units should develop contingency plans to expand their capabilities to more age groups as needed.

Adult units should develop contingency plans to expand their capabilities to pediatric age groups as needed. There are a variety of ways to help increase their capacity and capabilities. The coronavirus disease 2019 pandemic illustrated a novel means to expand capacity. Many hospitals expanded their adult capacity by converting their pediatric inpatient capacity and capability for adult care. Helpful lessons from their experience include the need for ongoing preparatory and just-in-time training for caring for different age populations, developing protocols for additional age groups, health care provider mental health support, and plans for collaborative teams such as team nursing or mixed physician provider teams. These studies not only show the ability to pivot staff and space as needed but also provide a helpful roadmap for converting adult resources for pediatric patients.^42–46 

Additional strategies include improving pediatric readiness and the use of real-time dashboards. The National Pediatric Readiness Project (NPRP) is a multiagency collaborative effort with the goal to improve pediatric emergency care in all ED settings, especially since the majority of pediatric ED visits occur in general EDs.^47–49  Whereas the National Pediatric Readiness Project has primarily been focused on EDs, the need for readiness to care for the acutely deteriorating child is now expanding to inpatient units and clinics.⁵⁰  One study has shown improved mortality rates in facilities that have higher pediatric readiness scores.⁵¹  By working to improve their overall pediatric readiness, hospitals will both improve their current pediatric care overall and be more prepared for a pediatric surge. Moreover, a real-time capacity and capability dashboard would allow for regional coordination of surge capacity and capabilities by expediting transfers. Information obtained from a survey such as ours can be used as a basis to populate these dashboards.

This study had some limitations. First, our survey results are a snapshot of pediatric capacity and capabilities that could change as pediatric units and hospitals open and close. Additionally, as a survey, our data represents the best estimate of additional pediatric bed availability by emergency management staff. These numbers are limited by possibly inaccurate estimates (both over and under), real-time staffing availability for these physical beds, and limitations in the setting of a surge affecting both pediatric and adult patients. Although the exact numbers may not be precise, the survey is designed as a tool that can be used to update information in the future and can serve as a useful tool for other areas that have not done a similar assessment before. Second, our study did not focus on space or equipment since previous studies have addressed the availability of equipment, usage of alternate care spaces, and altering standards of care.^52,53  Third, we are unable to obtain the exact pediatric population data for each zip code during the study period, so total population was used as a proxy estimate for our study analysis. Whereas pediatric patients comprise an estimated 20% of all Massachusetts residents, it is difficult to know how this would translate to actual pediatric bed utilization in a disaster. Lastly, this study assessed 1 state, which may limit generalizability. However, this is the first study to assess both capacity and capabilities for an entire state and the methodology can serve as a model for other states.

There is limited pediatric inpatient surge capacity in Massachusetts, and we would be unable to meet the recommended 20% increase in inpatient beds in a surge. During both standard operations and surge situations, several hospitals could provide respiratory therapies to children, but most hospitals lack surgical subspecialists for children during any circumstance as well as a pediatric surge plan. Moreover, available services are clustered in high population areas, leaving large rural areas with less services. Further surge planning and improvement in overall pediatric readiness is needed to provide improved access to care for pediatric patients in all areas.

We thank Aaron Gettinger, MS, NREMT, Tammy Goodhue, Ben Palmere, BA, NRP, TP-C, Jeff Doyle EMT and Nicole Daniels, MS from the Office of Preparedness and Emergency Management at the Massachusetts Department of Public Health and our colleague, Lise Nigrovic MD MPH, for their support and assistance of this project.

MA

DPH, Massachusetts Department of Public Health