BACKGROUND AND OBJECTIVE: Bronchiolitis is the most common reason for hospital admission in patients aged <1 year. Admissions have been increasing with hypoxia frequently cited as the determinant. Home oxygen (O2) has been shown to be feasible, although safety data are lacking. The objective of this study was to evaluate the impact of a home O2 clinical care protocol on admission rates in patients with bronchiolitis from the pediatric emergency department.
METHODS: We performed a retrospective chart review of patients with bronchiolitis who presented to a children’s hospital pediatric emergency department (altitude 1600 m) between 2005 and 2009. Patients between the ages of 1 and 18 months were included in the analysis. Patients requiring baseline O2 were excluded. We calculated the percentage of patients discharged on O2 and their readmission rates. We reviewed charts of patients who were admitted after home O2 for adverse outcomes. We also compared rates of admission before and after initiation of the protocol.
RESULTS: In this study, 4194 illnesses were analyzed; 2383 (57%) were discharged on room air, 649 (15%) were discharged on O2, and 1162 (28%) were admitted. Of those discharged on room air, 4% were subsequently admitted, and 6% of those discharged on O2 were admitted. There were no ICU admissions or need for advanced airway management in those patients discharged on O2. Our overall admission rates for bronchiolitis dropped from a rate of 40% to 31%.
CONCLUSIONS: Home O2 is an effective way to decrease hospital admissions in a select group of patients with bronchiolitis.
- CI —
- confidence interval
- O2 —
- PCP —
- primary care provider
- PED —
- pediatric emergency department
- RA —
- room air
What’s Known on This Subject:
Bronchiolitis is the most common cause for hospital admission in patients aged <1 year. Hypoxia is a common reason for admission. Despite a multitude of studies looking at various treatment strategies, no clear benefit has been found.
What This Study Adds:
With oxygen therapy being the main therapeutic option, home oxygen offers a novel way to manage bronchiolitis. This study shows that home oxygen is a safe and effective way to decrease hospital admissions in a select group of patients.
Bronchiolitis is the most common reason for hospital admission in children <1 year of age.1 It is estimated that there are >150 000 hospitalizations a year for bronchiolitis. The primary reasons for admission include respiratory distress, poor feeding, and hypoxia or the need for supplemental oxygen (O2). Practitioners are influenced by small changes in pulse-oximetry data in the decision to admit patients with bronchiolitis.2 With the increased use of pulse-oximetry, patients also remain in the hospital for supplemental O2 for longer periods of time3 after other parameters such as work of breathing and feeding have returned to normal.4 The American Academy of Pediatrics recommends that O2 therapy be initiated judiciously when O2 saturations levels fall below 90% and that the intensity of monitoring O2 saturation levels be reduced as the infant improves.5 With an estimated cost of hospitalization for bronchiolitis being ∼$500 million per year in the United States alone,6 the utility of home O2 for bronchiolitis should be assessed.7–9
Although several studies have focused on a predetermined pulse-oximetry cutoff that would be safe for discharge from the hospital,10–12 only 2 have looked at discharging patients home on O2. These 2 studies focused on the feasibility of discharging a low-risk population on home O2 from a tertiary care pediatric emergency department (PED)13 and from the inpatient setting.14 Although there was high caregiver and primary care physician (PCP) satisfaction, rates of unscheduled returns and subsequent admissions were not calculated given the small sample size.13
Before the winter season of 2005–2006, a home O2 protocol was introduced at our institution for the use of home O2 in patients with uncomplicated bronchiolitis and hypoxia. Uncomplicated bronchiolitis was defined as a lower respiratory tract illness associated with wheezing and/or crackles in an otherwise healthy, well-appearing patient without evidence of underlying cardiopulmonary disease. Hypoxia was defined as a room air (RA) pulse oximetry of <88%. We sought to characterize rates of home O2 use and rates of adverse outcomes in patients discharged from the hospital on O2 and to determine the impact this protocol had on hospital admission rates for bronchiolitis.
We performed a retrospective chart review of all visits of patients with bronchiolitis who presented to an academic tertiary care children’s hospital PED as well as 4 of the satellite PEDs and urgent care centers in our system (altitude 1600 m) between November 1 and April 30 inclusive for each year from 2005 to 2009. Patients between the ages of 1 and 18 months who were diagnosed with bronchiolitis (International Classification of Diseases, Ninth Revision codes: 466.1. 466.11, 466.19, 480.1 documented in the electronic health record) were included in the analysis. Patients with an underlying cardiopulmonary disease requiring baseline O2 were excluded. For patients with multiple visits, a discrete illness was defined as new symptoms, and if unknown, a minimum of 2 weeks between visits. All charts were analyzed to determine initial disposition (discharge home on RA, discharge home on O2, or admit/transfer). For patients who had extended observation in the PED, defined as ≥24 hours, their disposition was classified as an admission. Patients discharged on home O2 were identified through an order in the chart written for home O2 or a plan in the note stating that the patient was discharged from the hospital on O2. A systematic chart review was performed for all patients who were discharged on RA or home O2 to determine if they were subsequently admitted. For patients who were discharged on O2, we reviewed their charts to determine reason for admission if subsequently admitted and if they had any adverse outcomes including admission to the ICU and/or need for advanced airway management defined as the use of continuous positive airway pressure, biphasic positive airway pressure, or endotracheal intubation. Although we did not have access to all patients’ health records, it is the policy of the state health department to contact the facility if an unexpected death occurred.
The home O2 protocol recommends that patients between the ages of 3 and 18 months (minimum of 48 weeks corrected for preterm birth) who have no underlying chronic cardiopulmonary disease be observed in the PED or urgent care center for a minimum of 8 hours on continuous pulse oximetry with vital signs recorded every 2 hours. If the patients O2 saturations are ≥90% on 0.5 L/minute nasal cannula ≤O2 while awake, asleep, and feeding; the patient is able to maintain hydration; there are no signs of deteriorating respiratory status; and the attending and caregiver are comfortable with discharge home, they are discharged from the hospital on O2 with a 24-hour follow-up arranged with either the PCP or the PED (if the PCP is unavailable). The patient is given a small portable O2 tank, and a home health care company is contacted by our respiratory therapists to deliver additional O2 tanks and supplies to the families’ homes.
We found that patients between the ages of 1 and 3 months with bronchiolitis were also being discharged from the hospital on O2. Although the home O2 protocol is for patients aged 3 to 18 months, we included the 1- to 3-month-old patients in our analysis.
Demographic and clinical variables from the PED including age, gender, insurance, lowest RA O2 saturation, vital signs, medications given, disposition (discharge from the hospital on RA, discharge from the hospital on O2, or admit/transfer), amount of O2 at discharge from the PED if applicable, subsequent need for admission, and reasons for subsequent admission and adverse outcomes in patients discharged initially on home O2 were abstracted from the electronic health record onto a standardized data collection sheet. The primary outcome measures were the percentage of patients discharged from the PED on O2, the rates of subsequent admission, and rates of adverse outcomes for those discharged on O2. Secondary outcome measures were reasons for subsequent admission and cost savings.
Cost data for inpatient stays, PED, and observation visits were obtained from the facility’s finance department and averaged over the study period. Authorization to publish cost data were given by the chief financial officer. We also attempted to contact the local home oxygen company to assess the cost of home oxygen.
Categorical variables were evaluated with the χ2 test or Fisher exact test. Continuous variables were analyzed by using either a Student t test or Wilcoxon rank sum test. Relative risk and confidence intervals (CI) of subsequent admissions were calculated between patients who were discharged from the hospital on O2 and patients who were not discharged on O2 during their illness. Demographic variables, including age, gender, and insurance status, were compared among these groups to ensure equal distribution. Data were analyzed by using SAS 9.2 (SAS Institute Inc, Cary, NC). This study was approved and informed consent was waived by the Colorado Multiple Institutional Review Board.
The study included 3983 patients with 4194 discrete illnesses that met the inclusion criteria. Of these, 2383 patient illnesses (57%) were initially discharged from the hospital on RA, 649 (15%) were discharged on O2, and 1162 (28%) were admitted or transferred to another facility for admission (Fig 1). Of the 1162 who were initially admitted or transferred, 247 (21%) were not admitted on oxygen. Reasons for admission in these patients included, but were not limited to, age <3 months, dehydration, and need for frequent deep suctioning. The overall mean age of our study population was 7.6 ± 4.8 months, 59% were boys, and 56% had Medicaid. For those patients discharged on O2, the mean age was 8.9 ± 4.4 months, 60% were boys, and 52% had Medicaid (Table 1). For patients who were discharged on O2 or on RA, those who were subsequently admitted were younger than those who were not (Table 1).
The mean lowest RA O2 saturation recorded for those patients discharged from the hospital on O2 was 84% with an SD of 3%. The mean discharge amount of supplemental O2 was 0.39 L/minute with an SD of 0.1 L/minute (Table 2). Two patients exceeded the suggested supplemental O2 of 0.5 L/minute, 1 being discharged on 0.75 L/minute and the other on 1 L/minute. Both were during the first bronchiolitis season after the initiation of the home O2 protocol. Neither of these patients was subsequently admitted.
Of the 2383 patients who were discharged from the hospital on RA, 90 (4%; 95% CI 3.1–4.6) were subsequently admitted. Of the 649 patients who were discharged on O2, 38 (6%; 95% CI 4.3–7.9) were subsequently admitted (P = .03; Fig 1). The most common reasons for subsequent admission were increased O2 requirement or increased work of breathing (Table 3). There were no adverse outcomes, including ICU admission or need for advanced airway management, defined as the use of continuous positive airway pressure, biphasic positive airway pressure, or endotracheal intubation, in those patients initially discharged on O2. There were 2 patients initially seen in the PED and discharged on O2 who were subsequently admitted for reasons other than bronchiolitis. One patient had a ventricular-peritoneal shunt and was admitted with abdominal pain. On admission, he was no longer hypoxic, and his O2 was discontinued. The other was admitted with pyelonephritis and vomiting oral antibiotics. She required O2 for 1 day in the hospital, with no change in respiratory status or increase in the amount of O2, and was discharged on RA. Both of these patients were excluded from analysis because their readmission was not secondary to bronchiolitis. There was no statistical difference in the mean lowest RA saturation for patients discharged on O2 who were subsequently admitted (85%) and those who were not (84%; P = .58). There was a statistically significant difference in vital signs (temperature, heart rate, and respiratory rate) as well as an increased use of racemic epinephrine in those patients initially discharged on RA who were not admitted compared with those who were discharged on RA and subsequently admitted. However, this difference was not seen in those initially discharged on O2 (Table 2). Our overall admission rate for bronchiolitis dropped from a historical rate of 40% (L.B., unpublished data) to 31% over the past 4 seasons (Table 4), yielding an absolute reduction of 9%.
Thirty-five patients between the ages of 1 and 3 months were discharged on O2. However, during the study period, only 3 of the 35 were subsequently admitted, making it difficult to draw any conclusions from this subgroup of patients. The 3 who were readmitted did not require ICU admission or advanced airway management.
For PED patients observed and then discharged on O2, the mean PED length of stay was 10 hours (SD 4.8 hours). The mean length of stay for a patient with bronchiolitis admitted to the hospital in our institution is 2 days. The mean cost of a 2-day inpatient hospital stay for uncomplicated bronchiolitis over the study period was $2606 (US dollars). Of note, 40% of these inpatients are sent home on oxygen from the inpatient setting. Over the same time period, the mean cost of the combination of a PED visit, observation for a minimum of 8 hours, and 1 follow-up visit was $1344. This results in a cost savings of $1262 per patient. We attempted to contact the main home oxygen provider to assess the cost of home oxygen, but they did not respond to our inquiries. We do know, however, that our Medicaid reimbursement for up to a month of home O2 is $90.
This is the first study demonstrating that a home O2 protocol in a select group of patients with uncomplicated bronchiolitis from the PED can be both successful and sustainable. With increasing PED overcrowding and boarding of inpatients, the development and evaluation of outpatient management strategies, such as home O2, are imperative.
Recent literature has highlighted the increasing hospital burden of bronchiolitis admissions in otherwise healthy infants. Although hospital admissions have drastically increased,15,16 there has been no change in mortality rates from bronchiolitis.17 The most likely reasons for this increased rate of admission for bronchiolitis are dependence on pulse-oximetry and lack of a proven therapy. Despite a multitude of studies looking at various treatment strategies,18–23 no clear benefit has been found. Given the lack of effective therapies, there is considerable variability in the management and treatment of bronchiolitis.24 Providers have been shown to vary markedly in their recommendation for admission on the basis of minor differences in pulse-oximetry values.2 Patients also remain in the hospital for supplemental O2 after other discharge parameters such as oral intake and work of breathing have been reached.4 With O2 therapy being the main therapeutic option for these patients, home O2 offers a novel way to manage bronchiolitis.
Our historical bronchiolitis admission rate of 40% is similar to other rates reported in the literature. Mansbach et al and Corneli et al reported admission rates of 40% and 43%, respectively.12,18 Our admission rate dropped from 40% to 31% over the study period with our home O2 protocol without any other significant changes in bronchiolitis care at our institution. This absolute reduction of 9% corresponds to a relative reduction of 25%. Although there was statistical significance (P = .03) between those patients who were subsequently admitted after being discharged from the hospital on RA, and those who were subsequently admitted after being discharged on O2 (Table 4), we do not feel that this change is clinically significant nor does it outweigh the benefit of reducing overall hospital admission. Treatment options for bronchiolitis have shown little impact on admission rates and length of stay, but we were able to have a significant effect on our admission rate by successfully managing approximately 15% of all bronchiolitis patients and 39% of all bronchiolitis patients with hypoxia (611 of 1564 patients) with home oxygen.
In 2006, bronchiolitis admissions were estimated to cost more than $500 million annually with a mean cost of $3799 per hospitalization, highlighting the need for initiatives to safely reduce bronchiolitis hospitalizations and thereby decrease health care costs.6 Some studies report a mean length of stay of 3 days compared with our mean inpatient length of stay of 2 days.6,24 If other institutions were to establish similar home O2 protocols, especially those with a longer average length of stay, the savings could be substantial. In our previous work, we showed a trend toward fewer missed caregiver workdays in the group that was sent home from the PED on O2.13 We were not able to collect these data for this study. We were also unable to estimate the “cost” of the physician time. The total cost savings of the home oxygen program over traditional inpatient hospitalization is likely substantial, but this study design does not allow us to calculate the savings precisely.
Apnea is an important concern in the management of young infants with bronchiolitis. It has been shown that age and previous apnea are the 2 most likely predictors of apnea.25–27 In a study of 691 patients admitted with bronchiolitis, Willwerth et al found 19 to have apnea while admitted. The patients who experienced apnea were <1 month old, preterm, and <48 weeks postconception, or had a witnessed apneic episode at home.27 Per our home O2 protocol, these patients would not be considered low risk and therefore would not qualify for home O2. In this study, we had no patients who were discharged on O2 who returned with apnea requiring advanced airway management. Although we had 35 patients aged 1 to 3 months who were discharged on O2 and no adverse outcomes were seen in these patients, it is not appropriate to draw conclusions from this group of patients because their age is outside the protocol limits and the number of patients in this group is small.
Our study has several limitations. It is retrospective in nature, and thus we were unable to identify risk factors in patients who required subsequent admission. Although there was a statistically significant difference in vital signs as well as in the use of racemic epinephrine in those patients initially discharged on RA who were or were not subsequently admitted, this difference was not seen in those initially discharged on O2 (Table 2). It is therefore difficult to determine the true significance of this finding. We only have access to data from our own hospital and its network locations. Thus, our admission rates may be an underestimate if patients were admitted at nonnetwork locations. This study was performed at an institution located at an altitude of 1600 m in an area where home O2 is prevalent for a variety of pediatric conditions, including bronchopulmonary dysplasia and congenital heart disease, which may have affected the ease of establishing a home O2 protocol and may limit its generalizability. We have respiratory therapists available in the PED 24 hours a day, 7 days a week. They perform home O2 teaching with the family and arrange for O2 to be delivered to their home. We also have support from the PCPs in the community who have made themselves available to follow-up with patients within 24 hours of discharge. They are comfortable caring for their patients on home O2, including weaning O2 in an outpatient setting. Finally, the PED attending physicians have become comfortable discharging patients on home O2 because it has become routine in our setting and facilitates patients’ disposition. It is also possible that children with oxygen saturations <90% at altitude differ in severity from those with similar oxygen levels at sea level, and therefore the results may not be directly generalizable. Most attending physicians at our institution, however, do not place well-appearing patients with bronchiolitis on oxygen unless their oxygen saturation is <88%. Because it is impossible to know what our patients clinical status would be at sea level, we recommend that this study be repeated in a well-controlled setting at sea level.
Home O2 is an effective way to decrease hospital admissions for patients with uncomplicated bronchiolitis.
We thank Michael D. Rannie, RN, MS, for his participation in the concept and study design development and his thoughtful review of the manuscript.
- Accepted November 2, 2011.
- Address correspondence to Sarah Halstead, MD, Department of Pediatrics, Section of Pediatric Emergency Medicine, University of Colorado at Denver, 13123 East 16th Ave, B251, Aurora, CO 80045. E-mail:
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
FUNDING: No external funding.
- Unger S,
- Cunningham S
- American Academy of Pediatrics Subcommittee on Diagnosis and Management of Bronchiolitis,
- Diagnosis and management of bronchiolitis.
- Pelletier AJ,
- Mansbach JM,
- Camargo CA Jr
- Schroeder AR,
- Marmor A,
- Newman TB
- Zorc JJ,
- Hall CB
- Harrison AM,
- Boeing NM,
- Domachowske JB,
- Piedmonte MR,
- Kanter RK
- Mansbach JM,
- Clark S,
- Christopher NC,
- et al
- Bajaj L,
- Turner CG,
- Bothner J
- Tie SW,
- Hall GL,
- Peter S,
- et al
- Langley JM,
- LeBlanc JC,
- Smith B,
- Wang EE
- Shay DK,
- Holman RC,
- Roosevelt GE,
- Clarke MJ,
- Anderson LJ
- Corneli HM,
- Zorc JJ,
- Mahajan P,
- et al.,
- Bronchiolitis Study Group of the Pediatric Emergency Care Applied Research Network (PECARN)
- Patel H,
- Platt RW,
- Pekeles GS,
- Ducharme FM
- Christakis DA,
- Cowan CA,
- Garrison MM,
- Molteni R,
- Marcuse E,
- Zerr DM
- Copyright © 2012 by the American Academy of Pediatrics