OBJECTIVES:

To characterize the baseline level of emergency preparedness among families of technology-dependent children admitted to our PICU and to determine if an ICU-based planning intervention can sustainably improve families’ disaster preparedness.

METHODS:

A single-arm, survey-based study used to assess participants’ preparedness to handle a 72-hour home power outage on the basis of a novel 8-point checklist. Parents of patients in the study completed the survey questions when their child was admitted to the PICU, discharged, or transferred from the PICU, after at least 2 weeks at home, and after at least 6 months at home. This study included a cohort of 50 children younger than18 years old who required daily use of at least 1 piece of qualifying electronic medical equipment at home and their custodial parents. The checklist surveyed was designed with the goals of maximizing care capacity at home during a power outage and planning for evacuation. Counseling and other resources were provided on the basis of item completion at admission assessment.

RESULTS:

Patients’ families completed a median of 3 items (range: 0–8; N = 50) at admission, 4 items (range: 1–8; n = 45) at discharge, and 7 items (range: 3–8; n = 37) at the 2-week follow-up and retained 7 items (range: 5–8; n = 29) at the 6-month follow-up. Completion rates were significantly higher at each follow-up time point compared with baseline (P < .001).

CONCLUSIONS:

Families of technology-dependent children admitted to our PICU have significant disaster-preparedness needs, which can be addressed with an inpatient intervention.

Estimates of the prevalence of children with special health care needs (CSHCN) includes up to 20% of households.1  Children with medical complexity, a subset of CSHCN, make up a much smaller portion of the population but account for a larger portion of the health care cost, most of which occurs during inpatient admission. Coller et al2  found that complications associated with the technology used by this population were frequently related to potentially preventable admissions. This population is underrepresented in disaster-preparedness literature.3  All children are known to be more vulnerable than adults in times of emergencies or disasters because of physiologic, psychological, and developmental differences.4  Families of children with medical complexity often rely on medical technology dependent on electricity, which may be unavailable in the event of an emergency, heightening the vulnerability of those children.

Data reveal that families of children who are technology dependent are insufficiently prepared to manage their children’s health care needs during prolonged power failures. After the 2003 New York City blackout, 11 children dependent on home respiratory support were admitted to a local PICU solely because of respiratory-device failure.5  Similarly, after the 2011 earthquake in eastern Japan, 55% of all pediatric patients in the area who were technology dependent and 89% of patients requiring a ventilator were admitted to a hospital.6  These hospitalizations represent a preventable surge of admissions after disaster events.

Studies conducted in the outpatient or emergency department setting have identified specific gaps in disaster preparedness among families of children who are technology dependent. Baker and Baker1  studied families of CSHCN and found that they often lacked an emergency kit and communication plan and that this population’s preparedness level was also significantly lower than that found in similar studies of the general population. A subsequent study revealed a similar result for similar cohorts in different geographic locations.7  A third study revealed that an educational intervention led to improvement in preparedness when assessed 30 days later.8  Sakashita et al9  found that only 19 of 35 families of children using a ventilator, oxygen concentrator, or feeding pump could identify whether that machine had a back-up battery and that only 22 of 50 families of children reliant on electricity-powered technology had a prolonged power failure plan.

Hospital admission provides a unique opportunity to identify and address preparedness gaps. Technology-dependent children admitted to a PICU likely represent a particularly vulnerable population. We conducted a single-arm study to characterize the baseline level of emergency preparedness among families of technology-dependent children admitted to our PICU and to determine if an ICU-based planning intervention could sustainably improve families’ disaster preparedness.

Study subjects included a cohort of 50 children admitted to an urban PICU and their custodial parents. A power calculation was completed before the start of the study by using the results of the Sakashita et al9  trial as the estimate for baseline preparedness in this population because that study population was the most closely matched in the literature. To show an increase in preparedness from ∼50% of subjects in that trial to 75% in this trial (an increased considered by the authors of this trial to be a significant difference), at least 38 subjects would need to be enrolled. Eligibility criteria included child age <18 years and daily use of at least 1 piece of qualifying electronic medical equipment at home before admission. Qualifying equipment was defined as a device without which the patient would likely require medical intervention within 24 hours. For this reason, exclusion criteria included exclusive use of mobility devices or respiratory adjunct devices such as nebulizers or humidifiers. Non–English speaking parents were also excluded. Verbal consent was obtained from the custodial parent before enrollment, and when appropriate, verbal assent was obtained from the child if over the age of 7 years and able to answer questions at the time of the initial interview. Of note, of families approached, only 1 family declined to participate; no children declined to assent.

A quantitative and qualitative survey was designed to characterize the child’s technology dependence and to assess the extent to which families would be prepared to care for their technology-dependent child during a 72-hour power failure. The first author acted as the sole surveyor throughout the study period. Study subjects were selected on the basis of daily review of the PICU census during periods when the first author was not responsible for caring for patients in the unit, although families were informed that the surveyor was in the rotation of hospital staff who cared for patients in that unit. Families of patients who met study criteria and were expected to have a PICU length of stay >48 hours were approached by the surveyor after introduction by a member of the care team as soon as practically possible after admission. Completion of the phase 1 survey took 15 to 30 minutes; subsequent phases took <10 minutes. Families were informed of the time commitment and the 3 subsequent surveys at the first visit. Phone numbers for follow-up were obtained at that time directly from the families. They were also given a written information sheet about the project, which contained this information. Surveys were completed verbally by the person who identified as being primarily responsible for caring for the child. If both parents were present and wished to complete the questionnaire together, then the recorded answer represented a consensus between the 2. Responses to open-ended questions were transcribed verbatim by the surveyor.

Preparedness was assessed by using an 8-item checklist used to evaluate strategies to maximize care capacity at home as well as plans for evacuation if needed. Checklist items were chosen on the basis of a review of disaster-preparedness literature and discussions with parents of technology-dependent children who were members of our hospital’s Family Advisory Council. Please see Supplemental Table 4 for a list of the checklist items, the rationale for including each (with references to the articles in which each has been reported previously in the literature), the requirement for completion of each item, and interventions if the item was not completed at the baseline interview. Face validity of the survey tool was assessed before enrollment of subjects by reviewing the questions with the parents of 2 patients in the PICU who were technology dependent. The parents were asked to read the questions and restate each item in their own words. Minor adjustments to item language were made in response.

Preparedness was assessed at baseline (as soon as practicable after the child’s admission), at PICU discharge, and 2 weeks and 6 months after hospital discharge. After collection of baseline data, a targeted educational intervention was performed, specifically addressing each deficiency identified by the preparedness checklist; interventions provided for each specific item are listed in Supplemental Table 4. Follow-up preparedness assessments were conducted at PICU discharge, and after ≥2 weeks and ≥6 months spent at home.

Initial and discharge surveys were completed in person, whereas the 2-week and 6-month follow-up surveys were completed over the phone. Up to 3 attempts were made to contact the caregiver, and only those caregivers who completed the 2-week follow-up were contacted at 6 months. The study was approved by our institutional review board.

The study intervention was the provision of information and materials to parents of children who were technology dependent to improve their ability to care for their children in the case of a prolonged power failure on the basis of any deficiencies noted in the baseline assessment. The materials and education were provided by a study investigator immediately after the baseline survey and preparedness assessment. The education and materials were designed to address 8 preparedness items selected from a review of disaster-preparedness literature (see Supplemental Table 4 for specific references). Checklist items are described in detail in Supplemental Table 4.

The median number of checklist items complete at each time point was compared by using the Friedman test for nonparametric data with multiple repeated samples. Comparisons of the overall checklist at each individual time point were then performed by using Wilcoxon rank tests. Completion of individual items at specific time points was assessed by using the McNemar test for paired samples.

Fifty baseline assessments were completed between February 2016 and December 2016. The flow of subjects through the study is shown in Fig 1. Children in the study used a median of 4 (range: 1–5) qualifying devices daily, with a median daily duration of use of 15.5 hours (range: 1–24 hours). Twenty subjects reported continuous use of at least 1 device. Subject demographics are detailed in Table 1.

FIGURE 1

Flow of subjects through the study.

FIGURE 1

Flow of subjects through the study.

TABLE 1

Subject Demographics

DemographicValue (N = 50)
Child’s age, y, median (range) 3.5 (0.4–17) 
Total length of stay in the PICU, d, median (range) 3 (1–35) 
No. devices used at home, median (range) 4 (1–5) 
 Individual devices, n (%)  
  Feeding pump 46 (92) 
  Suction 43 (86) 
  Oxygen monitor 37 (74) 
  Oxygen concentrator 22 (44) 
  Mechanical vent 14 (24) 
  BiPAP 10 (20) 
  CPAP 2 (4) 
Device-use duration, median (range) 2 y (3 d–15 y) 
Hours per d for most-used device, median (range) 15.5 (1–24) 
DemographicValue (N = 50)
Child’s age, y, median (range) 3.5 (0.4–17) 
Total length of stay in the PICU, d, median (range) 3 (1–35) 
No. devices used at home, median (range) 4 (1–5) 
 Individual devices, n (%)  
  Feeding pump 46 (92) 
  Suction 43 (86) 
  Oxygen monitor 37 (74) 
  Oxygen concentrator 22 (44) 
  Mechanical vent 14 (24) 
  BiPAP 10 (20) 
  CPAP 2 (4) 
Device-use duration, median (range) 2 y (3 d–15 y) 
Hours per d for most-used device, median (range) 15.5 (1–24) 

BiPAP, bilevel positive airway pressure; CPAP, continuous positive airway pressure.

Twenty-three of 50 parents (46%) reported that they had called 9-1-1 or taken their child to a hospital simply because of a past power outage or that they would expect to do so if they experienced a prolonged power failure in the future.

Thirteen parents reported they had experienced a power failure lasting at least 6 hours since the child had been at home using electronic medical equipment, and 11 of those parents used an alternate power source or left home during that time. Of the 2 remaining subjects in this category, 1 was preparing to leave when the power came back on at ∼6 hours. Twenty-four of the 50 subjects surveyed reported having access to a generator or other alternate power source.

The 8-item preparedness assessment was evaluated on admission (N = 50), at discharge (n = 45), 2 weeks after discharge (n = 37), and 6 months after discharge (n = 29). Overall, subjects completed a median of 3 items (range: 0–8) at admission, 4 items (range: 1–8) at discharge, and 7 items (range: 3–8) at the 2-week follow-up and retained 7 items (range: 5–8) at the 6-month follow-up (Fig 2). Completion rates were significantly higher at each follow-up time point compared with baseline (P < .001). This finding was unchanged when only the 29 families with complete data sets were included. Supplemental Figure 3 reveals the median and interquartile range at each phase for the subgroup of subjects who completed all phases; these numbers were not significantly different when compared with the entire cohort.

FIGURE 2

Checklist-item completion for all subjects available at each interview phase. Paired checklist-item completion at each phase for 29 subjects managed through 6 months can be found in Supplemental Fig 3.

FIGURE 2

Checklist-item completion for all subjects available at each interview phase. Paired checklist-item completion at each phase for 29 subjects managed through 6 months can be found in Supplemental Fig 3.

The impact of the intervention on each individual checkpoint item was also assessed. Significant improvement was seen in the proportion of families who had successfully completed 5 of the 8 checklist items at discharge compared with admission. At 6 months, significant improvement was seen in 7 of the 8 items (Tables 2 and 3).

TABLE 2

Proportion of Subjects Who Completed Each Checklist Item at Each Interview Phase for All Available Subjects

Checklist ItemAdmission (N = 50)Discharge (n = 45)2-wk Follow-up (n = 37)6-mo Follow-up (n = 29)
n (%)n (%)Pn (%)Pn (%)P
Phone list 15 (30) 19 (42) .041 31 (84) <.001 24 (83) <.001 
Emergency form 25 (50) 28 (62) .041 35 (95) <.001 26 (90) <.001 
3 days of supplies 38 (76) 37 (82) .248 36 (97) .008 25 (86) .289 
Go bag 20 (40) 20 (44) .248 27 (73) <.001 26 (90) <.001 
Machine backup 31 (62) 35 (78) .023 31 (84) .023 27 (93) .026 
Evacuation plan 11 (22) 15 (33) .041 20 (54) <.001 22 (76) <.001 
Neighborhood resources 22 (44) 23 (51) .134 24 (65) .003 19 (66) .016 
Utility company 19 (38) 28 (62) .003 31 (84) <.001 24 (83) .002 
Checklist ItemAdmission (N = 50)Discharge (n = 45)2-wk Follow-up (n = 37)6-mo Follow-up (n = 29)
n (%)n (%)Pn (%)Pn (%)P
Phone list 15 (30) 19 (42) .041 31 (84) <.001 24 (83) <.001 
Emergency form 25 (50) 28 (62) .041 35 (95) <.001 26 (90) <.001 
3 days of supplies 38 (76) 37 (82) .248 36 (97) .008 25 (86) .289 
Go bag 20 (40) 20 (44) .248 27 (73) <.001 26 (90) <.001 
Machine backup 31 (62) 35 (78) .023 31 (84) .023 27 (93) .026 
Evacuation plan 11 (22) 15 (33) .041 20 (54) <.001 22 (76) <.001 
Neighborhood resources 22 (44) 23 (51) .134 24 (65) .003 19 (66) .016 
Utility company 19 (38) 28 (62) .003 31 (84) <.001 24 (83) .002 

P values indicate results of the McNemar test, used to compare a paired sample of only subjects when complete data were available for a given interview phase.

TABLE 3

Proportion of Subjects Available for All 4 Study Phases Who Completed Each Checklist Item at Each Phase

Checklist ItemAdmission (n = 29)Discharge (n = 29)2-wk Follow-up (n = 29)6-mo Follow-up (n = 29)
n (%)n (%)Pn (%)Pn (%)P
Phone list 7 (24) 11 (38) .134 23 (79) <.001 24 (83) <.001 
Emergency form 9 (31) 14 (48) .074 27 (93) <.001 26 (90) <.001 
3 days of supplies 21 (72) 24 (83) .248 28 (96) .023 25 (86) .289 
Go bag 10 (34) 13 (45) .248 21 (72) .003 26 (90) <.001 
Machine backup 18 (62) 23 (79) .074 24 (83) .041 27 (93) .026 
Evacuation plan 5 (17) 10 (34) .074 15 (52) .004 22 (76) <.001 
Neighborhood resources 9 (31) 12 (41) .248 18 (62) .008 19 (66) .016 
Utility company 11 (38) 21 (72) .004 25 (86) <.001 24 (83) .002 
Checklist ItemAdmission (n = 29)Discharge (n = 29)2-wk Follow-up (n = 29)6-mo Follow-up (n = 29)
n (%)n (%)Pn (%)Pn (%)P
Phone list 7 (24) 11 (38) .134 23 (79) <.001 24 (83) <.001 
Emergency form 9 (31) 14 (48) .074 27 (93) <.001 26 (90) <.001 
3 days of supplies 21 (72) 24 (83) .248 28 (96) .023 25 (86) .289 
Go bag 10 (34) 13 (45) .248 21 (72) .003 26 (90) <.001 
Machine backup 18 (62) 23 (79) .074 24 (83) .041 27 (93) .026 
Evacuation plan 5 (17) 10 (34) .074 15 (52) .004 22 (76) <.001 
Neighborhood resources 9 (31) 12 (41) .248 18 (62) .008 19 (66) .016 
Utility company 11 (38) 21 (72) .004 25 (86) <.001 24 (83) .002 

P values indicate results of the McNemar test.

During the 18-month study period, 19 children from the original cohort of 50 were readmitted to our PICU at least once (32 total readmissions); 1 has remained in a subacute care facility, and 1 has died. The subgroup that required readmission represents a large portion of the cohort. They were not resurveyed on readmission, although additional resources were provided if they were requested by the family. In the subgroup analysis, those 19 families had a median completion of 3 items (range: 1–6) on admission, 4 items (range: 1–6) on initial discharge, 6 items (range: 3–8) 2 weeks after discharge from initial admission, and 7 items58  6 months after discharge from initial admission. Despite repeat exposure to our PICU, the results from this group did not reveal a significant difference from the analysis of the entire cohort.

We found that a simple inpatient educational intervention can significantly and lastingly improve the level of disaster preparedness for families of children who are technology dependent. The American Thoracic Society has recommended that emergency preparedness be part of discharge planning for children on home ventilator support.10  We agree that an inpatient admission offers a valuable opportunity to identify and address gaps in family’s disaster preparedness. Inpatient intervention also allowed for introduction of resources at a time when the subject is a captive audience with access to medical care and care management teams.

Although we conducted this intervention in a PICU, all of the information and education we provided could, and should, be provided to eligible families during all hospital admissions. We suggest that the 8-item checklist we developed for this study serve as a tool for assessing and improving family’s preparedness. The use of the checklist in the PICU setting allowed for a cohort of patients who had dependence on more machines for a greater portion of the day than previously reported,9  but the intervention would likely benefit all levels of technology dependence.

In our cohort, drawn largely from New York City, a high proportion of families did not have access to alternate power sources such as generators. For this population, ensuring that families are registered with the local utility company as having an individual who requires life-sustaining equipment in the home and ensuring that families are fully familiar with the use of back-up batteries are items of particular importance. Hospitals serving different regions of the country will likely encounter different needs specific to their populations.

The reported level of disaster preparedness significantly increased for the study cohort, with a lasting effect that carried through 6 months. The biggest gains were seen in completion of the written phone list, the emergency health information form, and the utility company registration. In this study, the checklist was administered by a PICU fellow who was not on the care team at the time, but it could be accomplished by a variety of care planning team members with the help of the primary care team. Further work is needed to determine how best to incorporate this education and provide families with needed materials as part of routine discharge planning. This approach will allow for the maximum potential impact, with the least burden on the medical care team.

This study was limited primarily by reliance on parent report of item completion. Further limitations include the exclusion of non–English speaking subjects, incomplete follow-up, and lack of conclusive evidence that completion of the items on our checklist will meaningfully impact family behaviors during an actual disaster. Although a significant number of families did not complete all 4 phases, Table 3 and Supplemental Fig 3 reveal that the 29 families who did complete them did not have significantly different results from baseline when compared with the entire cohort. There is also possible concern about power differential on follow-up given that the surveyor was also sometimes responsible for the health care of the subjects. Each family was told that the survey was not a test and that their answers would not impact when their child would be able to leave the ICU. The surveyor was in training at the time, and all final decisions regarding discharge were made by the supervising attending physician on service, who was not involved with the study.

Families of technology-dependent children admitted to our PICU have significant disaster-preparedness needs, which can be addressed with an inpatient intervention. Further research is needed to determine if planning for a large disaster can help families become more resilient in the face of smaller emergencies at home and if this can decrease the readmission rate. A comprehensive, team-based approach that includes both inpatient and outpatient interventions will likely have the greatest effect on reducing the vulnerability of this population.

Dr Gillen conceptualized and designed the study and drafted the initial manuscript; Dr Morris assisted in the study design and reviewed and revised the manuscript; and both authors approved the final manuscript as submitted.

FUNDING: No external funding.

1
Baker
LR
,
Baker
MD
.
Disaster preparedness among families of children with special health care needs
.
Disaster Med Public Health Prep
.
2010
;
4
(
3
):
240
245
2
Coller
RJ
,
Nelson
BB
,
Sklansky
DJ
, et al
.
Preventing hospitalizations in children with medical complexity: a systematic review
.
Pediatrics
.
2014
;
134
(
6
).
3
Murray
JS
.
Disaster preparedness for children with special healthcare needs and disabilities
.
J Spec Pediatr Nurs
.
2011
;
16
(
3
):
226
232
4
Disaster Preparedness Advisory Council
;
Committee on Pediatric Emergency Medicine
.
Ensuring the health of children in disasters
.
Pediatrics
.
2015
;
136
(
5
).
5
Prezant
DJ
,
Clair
J
,
Belyaev
S
, et al
.
Effects of the August 2003 blackout on the New York City healthcare delivery system: a lesson for disaster preparedness
.
Crit Care Med
.
2005
;
33
(
suppl 1
):
S96
S101
6
Nakayama
T
,
Tanaka
S
,
Uematsu
M
, et al
.
Effect of a blackout in pediatric patients with home medical devices during the 2011 eastern Japan earthquake
.
Brain Dev
.
2014
;
36
(
2
):
143
147
7
Baker
LR
,
Cormier
LA
.
Disaster preparedness and families of children with special needs: a geographic comparison
.
J Community Health
.
2013
;
38
(
1
):
106
112
8
Baker
MD
,
Baker
LR
,
Flagg
LA
.
Preparing families of children with special health care needs for disasters: an education intervention
.
Soc Work Health Care
.
2012
;
51
(
5
):
417
429
9
Sakashita
K
,
Matthews
WJ
,
Yamamoto
LG
.
Disaster preparedness for technology and electricity-dependent children and youth with special health care needs
.
Clin Pediatr (Phila)
.
2013
;
52
(
6
):
549
556
10
Sterni
LM
,
Collaco
JM
,
Baker
CD
, et al
;
ATS Pediatric Chronic Home Ventilation Workgroup
.
An official American Thoracic Society clinical practice guideline: pediatric chronic home invasive ventilation
.
Am J Respir Crit Care Med
.
2016
;
193
(
8
):
e16
e35
11
American Academy of Pediatrics Committee on Pediatric Emergency Medicine and Council on Clinical Information Technology
American College of Emergency Physicians
;
Pediatric Emergency Medicine Committee
.
Policy statement–emergency information forms and emergency preparedness for children with special health care needs
.
Pediatrics
.
2010
;
125
(
4
):
829
837
13
Redlener
I
,
Berman
DA
.
National preparedness planning: the historical context and current state of the U.S. public’s readiness, 1940-2005
.
J Int Aff
.
2006
;
59
(
2
):
87
103

Competing Interests

POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.

FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.

Supplementary data