Cheney, Joyce and Barber, Scott and Altamirano, Luis and Cheney
MedicoCirujano, Marise and Williams, Chris and Jackson, Mary and Yates,
Patsy and O'Rourke, Peter and Wainwright, Claire (2005) A Clinical Pathway
for Bronchiolitis is Effective in Reducing Readmission Rates . Journal of
Pediatrics 147(5):pp. 622-626.
Copyright 2005 Elsevier.
Accessed from: http://eprints.qut.edu.au/archive/00003671
A Clinical Pathway for Bronchiolitis is Effective in Reducing Readmission Rates
Joyce Cheney BA, , Scott Barber BBus, ADCLT, Luis Altamirano, MedicoCirujano,
Marise Cheney BNurs, MPH, Chris Williams BSc (Nurs), Mary Jackson RN, Patsy
Yates MSocSc, PhD, Peter O'Rourke BSc, PhD and Claire Wainwright MBBS, MD
From the Department of Respiratory Medicine, Royal Children's Hospital, the School
of Nursing, Queensland University of Technology, and the School for Population
Health, University of Queensland, Brisbane, Queensland, Australia
To examine the use of a clinical pathway in the management of infants hospitalized
with acute viral bronchiolitis.
A clinical pathway with specific management and discharge criteria for the care of
infants with bronchiolitis was developed from pathways used in tertiary care pediatric
institutions in Australia. Two hundred and twenty-nine infants admitted to hospital
with acute viral bronchiolitis and prospectively managed using a pathway protocol
were compared with a retrospective analysis of 207 infants managed without a
pathway in 3 regional and 1 tertiary care hospital.
Readmission to hospital was significantly lower in the pathway group (P = .001), as
was administration of supplemental fluids (P = .001) and use of steroids (P = .005).
There were no differences between groups in demographic factors or clinical severity.
The pathway had no overall effect on length of stay or time in oxygen.
A clinical pathway specifying local practice guidelines and discharge criteria can
reduce the risk of readmission to hospital, the use of inappropriate therapies, and help
with discharge planning.
Abbreviations: ICU, Intensive care unit; LOS, Length of stay; RSV, Respiratory
syncytial virus; TIO, Time spent in oxygen
Clinical pathways, developed to provide a systematic and organized method of
managing the care of patients with a variety of conditions, have been used
extensively, often without formal evaluation of their effectiveness.1 and 2
Acute viral bronchiolitis results in hospital admission in about 1 in 100 usually
healthy infants.3 Clinical pathways for bronchiolitis have been established in pediatric
tertiary care centers in Australia. Two studies were unable to demonstrate any
improvement in the management of oxygen therapy or hydration therapy after the
introduction of clinical pathways; however, they did find a reduction in the use of
antibiotics.4 and 5 Clinical practice guidelines for bronchiolitis have also been
assessed and, according to Perlstein et al,6 use of guidelines has produced dramatic
decreases in the number of admissions and the mean length of stay (LOS). Subsequent
authors3 have not been able to identify a similar level of success, although one recent
large study7 found a significant decrease in resource use after introduction of clinical
guidelines. A recent review of current management provides specific assessment and
admission guidelines to promote best practice.8
A clinical pathway was developed and used to promote consistency of nursing
management during a prospective clinical trial of epinephrine at 1 tertiary care and 3
regional hospitals in southeast Queensland.9 The epinephrine study identified oxygen
saturation on admission, need for supplemental oxygen, and need for intravenous
therapy as the major factors affecting LOS in hospital. The intervention of the
epinephrine study had no effect on outcomes, and there was no difference in outcomes
between the patients who took part in the study and those patients managed on a
clinical pathway who did not take part. This lack of difference in outcomes afforded a
unique opportunity to examine the use of the clinical pathway in the management of
infants with acute viral bronchiolitis.
Selection of Patients
This study combined a prospective analysis of 229 patients admitted with
bronchiolitis treated using a pathway protocol (pathway group) with a retrospective
analysis of 207 randomly selected historical control patients managed without a
pathway. For inclusion in the study, the infants had to be under 12 months of age (or
12 months corrected age if premature) with their first wheezing episode requiring
admission. Infants with heart disease or significant respiratory problems, such as
cystic fibrosis, and infants who had received corticosteroids within 24 hours of
presentation or used bronchodilators within the previous 4 hours, were excluded. The
pathway included specific admission assessment information, clinical management
guidelines, and discharge criteria (Table I).
Clinical pathway admission assessment, clinical management and discharge criteria
Initial respiratory assessment recorded
Oxygen saturation in room air
Respiratory rate and effort
Skin color and tone
Presence/absence of wheezing and crackles on chest auscultation
Level of hydration
Medical and social history recorded
Previous medical problems
Infectious disease screening
Family history, eg, atopy
Social history, esp. smoking status of parents
Discharge support eg. LMO/GP
Guidelines for oxygen administration and supplemental fluids
Administer supplemental oxygen if:
Oxygen saturation < 94%
A combination of: significant respiratory distress, respiratory rate > 60/min,
Wean or cease supplemental oxygen if:
Oxygen saturation consistently above 93%
Infant stable for more than 4 hours and tolerating oral feeds
Intravenous fluids administered if:
Respiratory rate > 60/min
Oral feeding inadequate
Intravenous fluids ceased if:
Infant able to tolerate oral feeds
Respiratory status stable in room air for > 10 hours
Nil or mild chest recessions
Oxygen saturation in room air > 93%
Discharge plan for the parents complete
Letter for the GP complete
Setting and Time Frame
The setting for this study was 1 tertiary care hospital (Royal Children's Hospital,
Brisbane) and 3 regional hospitals (Gold Coast Hospital, Redcliffe Hospital, and
Caboolture Hospital), all located in southeast Queensland.
The historical control group was admitted between May 1998 and August 1999, and
the infants in the prospective study who were managed on a pathway were admitted
between May 2000 and August 2001.
Baseline clinical data, and data on parental smoking, admission to the pediatric
intensive care unit (ICU), and use of drug therapy were collected from the clinical
notes for the historical control group and from the clinical pathway records in the
pathway group. Data on all pathology tests ordered and chest X-ray films were
obtained from the hospital databases and clinical records.
For the pathway group, infants who required supplemental oxygen, had weaned to air,
and then required further supplemental oxygen, were identified. The lack of detailed
documentation about supplemental oxygen requirements in the control group
prohibited further investigation of these infants. The possible predictive factors that
could account for this repeated oxygen requirement were examined and included
hospital site, age, sex, respiratory syncytial virus (RSV) status, severity at admission
(as determined by oxygen saturation, respiratory rate, and presence of recession),
length of time of initial supplemental oxygen requirement, need for supplemental
fluids, and smoking status of parents.
Study outcomes included LOS, time spent in oxygen (TIO), use of supplemental
fluids, readmission to hospital within 2 weeks of discharge, use of pathology,
radiological investigations, and medications—specifically antibiotics and steroids—
and predictors for return to oxygen therapy.
Local institutional ethics committees from the 4 centers approved the study.
Sample Size Estimate
On the basis of a mean LOS of 82 hours and standard deviation of 67 hours from the
epinephrine study, it was estimated that a sample size of 200 subjects in each group
was required to detect a difference of 12 hours in length of stay (a clinically relevant
difference) between groups with a power of 80% and an α = 5%.
We compared the infants in the historical control group and the infants managed using
a clinical pathway with respect to demographic data, baseline clinical measures, and
study outcomes. Comparisons between different hospitals, as well as between the 2
groups, were made to determine whether there was any interhospital variation in
practice. The χ2 test for significance was used to determine any statistically
significant differences between the 2 sets of discrete data, and a 1-way analysis of
variance test was used for the continuous data. A P value < .05 was considered
statistically significant. Logistic regression was used to screen risk factors for return
Baseline Demographic and Clinical Data
The baseline characteristics of the prospective and control groups were similar with
regard to age, sex, RSV infection, and admitting hospital. The 2 groups had similar
severity of illness at admission, as determined by presence/absence of recession,
respiratory rate, requirement for supplemental oxygen, oxygen saturations at
admission, and admission to ICU (Table II).
Demographic/clinical data at admission (Given as mean [SD]) Control patients (n =
207) Pathway patients (n = 229) P value
Age (months) 4.0 ± 3.0 4.3 ± 2.9 .40
Sex (male/female) 132 (63.8%)/75 (36.2%) 141 (61.6%)/88 (38.4%) .64
Nasopharyngeal aspirate positive 139 (72.4%) 175 (76.4%) .35
RSV + 127 (66.1%) 155 (67.7%) .74
Admission numbers (% of total) .72
RCH 114 (55.1%) 114 (49.8%)
GCH 40 (19.3%) 50 (21.8%)
Caboolture 27 (13.0%) 31 (13.5%)
Redcliffe 26 (12.6%) 34 (14.8%)
Chest recession present (n) 172 (90.5%) 211 (92.1%) .56
Respiratory rate (per min) 50.7 (±12.0) 51.8 (±13.6) .41
O2 saturations on arrival (before oxygen) 93.1% (± 5.3) 93.4% (±4.0) .48
Supplemental O2 requirement 120 (58.0%) 148 (64.6%) .15
Admission to ICU 8 (3.9%) 7 (3.1%) .64
RCH, Royal Children's Hospital; GCH, Gold Coast Hospital.
There was no significant difference overall between the 2 groups for LOS or TIO. The
mean LOS in hours for infants in the control group was 86 hours (75.3-96.6, 95% CI),
and 88.6 hours (78.8-98.3, 95% CI) for the pathway group. The length of stay was
22.0 hours longer in the tertiary care hospital than in the regional hospitals (P = .006),
but this difference was reduced to 11.1 hours (P = .064) after adjustment for use of
supplemental oxygen and supplemental fluids. Fifty-eight percent of infants in the
control group required supplemental oxygen for a mean time of 72.6 hours (59.9-85.4,
95% CI), whereas 65% of the pathway group required supplemental oxygen for a
mean time of 79.9 hours (69.1-90.8, 95% CI). There was a significant difference,
however, in a subgroup of both cohorts. Infants in the pathway group requiring
oxygen but not supplemental fluids had a longer LOS (P = .006) and a longer TIO (P
= .003) than infants in the control group requiring oxygen only.
Fifteen infants (7.2%) in the control group required readmission within 2 weeks after
discharge, with 9 of these infants also being administered supplemental oxygen during
their initial hospital stay. This number compares with only 2 infants (0.9%) in the
pathway group (P = .001) requiring readmission. Both these infants needed
Supplemental fluids were administered to 33.8% of the control group compared with
19.2% of infants managed on a pathway (P = .001); however, more patients required
supplemental fluids in the tertiary care hospital than regional hospitals (28.9% vs
16.3%, P = .006) There was greater use of steroids in the control group compared
with the pathway group (P = .005), although the use of antibiotics was not
significantly different between the 2 groups.
There was a marked difference between the 2 groups in the collection of data on the
smoking status of the parents. In the control group, the parental smoking status of
only 16% of the infants was collected. In the pathway group, however, data on
smoking status were obtained for every infant (Table III).
Outcome measures (mean - 95% confidence interval) Outcome measures Control
patients (n = 207) Pathway patients (n = 229) P value
LOS (hours) 86.0 (75.3-96.6) 88. 6 (78.8-98.3) .72
Nil O2 or supp fluids 40.2 (33.1-47.3) 34.4 (29.9-38.9) .18
O2 only 81.5 (70.5-92.6) 107.0 (94.9-119.1) .006
O2 and supp fluids 158.0 (132.5-183.5) 147.6 (115.7-179.4) .61
TIO (hours) 72.6 (59.9-85.4) 79.9 (69.1-90.8) .39
O2 only 45.0 (35.0-55.1) 69.8 (58.7-80.9) .003
O2 and supp fluids 101.1 (79.2-123.1) 106.4 (80.7-132.1) .77
Number requiring supplemental O2 (%) 120 (58%) 148 (64.6%) .15
Number requiring supplemental fluids (%) 70 (33.8%) 44 (19.2%) .001
Readmission within 2 weeks 15 (7.2%) 2 (0.9%) .001
Parental smoking status known (%) 33 (16%) 229 (100%) .029
There was no significant difference between the 2 groups for the number of chest X-
ray films or pathology tests including blood cultures, arterial blood gases, full blood
counts, urea and electrolytes and liver function tests, lumbar punctures, methicillin-
resistant Staphylococcus aureus screening, and urine culture and sensitivity tests. A
higher proportion of patients in the control group (4%) had blood taken for Bordetella
pertussis compared with the pathway group (0.9%) (χ2 = 6.32; P = .13). In contrast, a
higher proportion of patients on the pathway (11%) had feces culture and sensitivity
testing done, compared with the control group (4%) (χ2 = 7.23; P = .007).
Forty infants (27%) on the clinical pathway required supplemental oxygen again after
being weaned to room air. Boys were 3 times more likely to return to oxygen (OR =
3.30; 95% CI, 1.39-7.81) after adjusting for oxygen saturations on admission, in spite
of the fact that there was no significant difference between the oxygen saturation
levels on admission of boys and girls for the entire cohort. Eight hours of monitoring
after weaning from oxygen resulted in only 14 (9.5%) requiring a return to oxygen
therapy. For every 1% increase in admission oxygen saturation levels, the infant's
likelihood of returning to oxygen decreased by 16% (OR = 0.84; 95% CI, 0.75-0.95).
Of the infants who did require further supplemental oxygen, 80% had supplemental
oxygen restarted between 6 am and 6 pm. Time to return to oxygen was not affected
by sex, age at admission, or oxygen saturation levels on admission.
One of the major benefits of using this clinical pathway for infants with bronchiolitis
appears to be in discharge planning and reducing the chances of a required
readmission through the achievement of specified clinical outcomes before discharge.
This benefit aligns with current findings by Meuthing et al7 that emphasize the
importance of having specific guidelines, including admission and discharge criteria,
which direct clinical practice.
Another advantage to be gained from use of a clinical pathway may be a more
accurate assessment of the infant in relation to their need for supplemental fluids.
Predetermined guidelines to direct clinical practice may lead to the elimination of
some variability in practice and the avoidance of unnecessary invasive procedures for
the patient. It may also be valuable as a cost-cutting measure, both in the up-front
costs associated with nasogastric and intravenous feeds, and in terms of a reduction in
the nursing time required in setting up and monitoring these procedures.
The areas in which specific recommendations were identified in the clinical pathway
were the areas where there were statistically significant differences in outcomes
between groups. This finding is evidenced by the reduction in the administration of
supplemental fluids and steroids and the consistent collection of data on parental
smoking status. The reduction in the use of steroid therapy that we found occurred
while the patient LOS was unaltered and readmission rates to hospital dropped
suggests that the change in drug prescribing practice had no detectable negative effect
on short-term patient outcomes. Wang et al10 describe the variations in interventions
relating to both steroids and antibiotics, suggesting that variation can occur when the
most appropriate management is not clear. Areas of the pathway that did not mention
specific management guidelines such as the administration of antibiotics, and the
requirements for pathology testing and chest X-ray films, saw no significant change in
practice between the 2 groups.
The lack of significant difference between the groups overall in terms of LOS or TIO
suggests that either the particular clinical pathway used was not able to affect these
outcomes, or that use of any clinical pathway may not significantly change these
outcome measures for bronchiolitis. The longer LOS for children in the pathway
group who required supplemental oxygen is likely to be due to the guidelines given
for oxygen administration. The pathway was designed to be reasonable and acceptable
to health care providers at a number of centers, and the actual cutoff for an acceptable
level of oxygen saturation for weaning had not previously been examined. The value
chosen is likely to have an effect on both LOS and possible readmission and should
be formally tested. The subgroups of infants from the tertiary care hospital who were
hospitalized and in supplemental oxygen for longer periods of time would be expected
because this was the center to which the sicker infants were transferred because it was
the only one with a pediatric ICU. Boys and infants with lower oxygen saturation
levels on admission were the most “at-risk” group for requiring further supplemental
oxygen after being previously weaned to room air. In addition, the fact that
significantly more infants returned to oxygen therapy during the day may lead us back
to the cornerstone of bronchiolitis management, namely minimal handling and
supportive nursing care.
Although the use of a historical control group is not ideal, prospectively randomizing
patients to pathway care or not pathway care would be extremely difficult from the
point of view of nursing practice and adherence with pathway use. This study was
conducted in a variety of hospital settings that allowed larger numbers and also
ensured that the study participants were likely to be representative of most infants
hospitalized in either tertiary care or regional hospitals. Seasonal variation occurs with
bronchiolitis leading to significantly greater numbers presenting during the winter
months. Genotype variations in the causative organisms for bronchiolitis can mean
that 1 year might have milder cases than a subsequent year, theoretically resulting in
shorter LOS or reduced TIO. By collecting data over 2 winter periods, the possibility
of such an occurrence was minimized. In addition, severity data suggested similar
groups were studied.
Clinical pathways provide a systematic way of managing the care of infants with
acute viral bronchiolitis. This pathway significantly decreased the likelihood of
readmission almost certainly because it provided specific discharge criteria. Other
tangible positive outcomes from using this pathway with specified management
criteria were decreased use of supplemental intravenous fluids and steroid therapy and
the collection of more accurate baseline data. Although this clinical pathway is no
longer used in the tertiary hospital, it is still being used in the 3 regional centers
probably because the specific criteria developed at the local level provided guidelines
in hospitals that do not have large numbers of experienced pediatric staff. This study
demonstrates the importance of clear specific guidelines to effect change in clinical
practice. The challenge is to ensure that these guidelines are evidence based and
achieve the best possible outcomes for patients.
Special thanks to the nursing and medical staff and the medical records staff at the 4
participating hospitals who assisted with the study.
1 C.B. Forrest, S.A. Shipman, D. Dougherty and M.R. Miller, Outcomes research in
pediatric settings: recent trends and future directions, Pediatrics 111 (2003), pp. 171–
2 T.S. Kwan-Gett, P. Lozano, K. Mullin and E. Marcuse, One-year experience with
an inpatient asthma clinical pathway, Arch Pediatr Adolesc Med 151 (1997), pp. 684–
689. Abstract-MEDLINE | Abstract-EMBASE
3 N.M. Kini, J.M. Robbins, M.S. Kirschbaum, S.J. Frisbee and U.R. Kotagal,
Inpatient care for uncomplicated bronchiolitis: comparison with Milliman and
Robertson guidelines, Arch Pediatr Adolesc Med 155 (2001), pp. 1323–1331.
4 S. Peter and M. Fazakerley, Clinical effectiveness of an integrated care pathway for
infants with bronchiolitis, Pediatr Nurs 16 (2004), pp. 30–35. Abstract-MEDLINE
5 S.D. Wilson, B.B. Dahl and R.D. Wells, An evidence-based clinical pathway for
bronchiolitis safely reduces antibiotic overuse, Am J Med Qual 17 (2002), pp. 195–
6 P.H. Perlstein, U.R. Kotagal, C. Bolling, R. Steele, P.J. Schoettker and H.D.
Atherton et al., Evaluation of an evidence-based guideline for bronchiolitis, Pediatrics
104 (1999), pp. 1334–1352.
7 S. Muething, P.J. Schoettker, W.E. Gerhardt, H.D. Atherton, M.T. Britto and U.R.
Kotagal, Decreasing overuse of therapies in the treatment of bronchiolitis by
incorporating evidence at the point of care, J Pediatr 144 (2004), pp. 703–710.
SummaryPlus | Full Text + Links | PDF (360 K)
8 D.A. Fitzgerald and H.A. Kilham, Bronchiolitis: assessment and evidence-based
management, MJA 180 (2004), pp. 399–404. Abstract-MEDLINE | Abstract-
9 C. Wainwright, L. Altamirano, M. Cheney, J. Cheney, S. Barber and D. Price et al.,
A multicenter, randomized, double-blind, controlled trial of nebulized epinephrine in
infants with acute bronchiolitis, N Eng J Med 349 (2003), pp. 27–35. Abstract-
EMBASE | Abstract-Elsevier BIOBASE | Full Text via CrossRef
10 E.E.L. Wang, B.J. Law, F.D. Boucher, D. Stephens, J.L. Robinson and S. Dobson
et al., Pediatric Investigators Collaborative Network on Infections in Canada
(PICNIC) study of admission and management variation in patients hospitalized with
respiratory syncytial viral lower respiratory tract infection, J Pediatr 129 (1996), pp.
390–395. SummaryPlus | Full Text + Links | PDF (525 K)
Supported by a grant from the Queensland Nursing Council, Queensland, Australia.
The views expressed do not necessarily represent the views of the Council or the
members, executive officer, or staff of the Council. Support was also given by the
Royal Children's Hospital Foundation, Brisbane, for the duration of this study.
Reprint requests: Joyce Cheney, Department of Respiratory Medicine, Royal
Children's Hospital, Herston Rd., Brisbane QLD 4029, Australia.