Acute bronchiolitis (AB) is one of the most important pathologies in paediatrics.1 It is the most common lower respiratory tract infection in the first year of life, with an annual incidence of 10%–30%.2,3 In Spain it is the leading cause of hospitalisation in children under one year of age.4 The main aetiological agent is the respiratory syncytial virus (RSV), with a variable frequency of around 65%–85%.5,6 There is no other acute disease that has such an important effect on the health of infants and that produces such a high social and economic impact as AB.7

The diagnosis must be clinical, based on the findings from the medical history and physical examination. Routine performance of diagnostic tests is not recommended.8–12

As for treatment, there is great variability in clinical practice in every care setting, and treatments of unproven efficacy are often used indiscriminately.13–16 For this reason, more and more authors are suggesting that the time has come to recognise the limitations of therapeutic interventions and to focus efforts on doing more for the patient with less medication.17 Numerous meta-analyses have clarified the efficacy of the available therapeutic options.18–20 These have given rise to new clinical practice guidelines (CPGs) which discourage the use of bronchodilator treatments.10–12

The difficulties that arise in care delivery when trying to apply the best evidence included in CPGs to everyday practice are well known.21 In the case of AB, the recommendations of CPGs have had a limited impact.22–24 Consequently, some centres have adopted quality improvement strategies. There is a wide range of approaches and published results, and no evidence to identify which strategies are superior.25–28 In most cases it is a matter of implementing local protocols, with the intention of standardising management and putting into practice the recommendations formulated in CPGs.29 Acute bronchiolitis is a special case because the CPGs do not recommend specific interventions, but are rather designed to prevent excessive treatment and avoid unnecessary diagnostic procedures. Unfortunately, less research has been directed towards evaluating strategies for reducing the use of non-recommended resources than towards evaluating strategies for increasing the use of effective resources.30

The objective of this study is to analyse the impact of a quality improvement strategy, maintained over a period of five years, directed at care of patients with AB in the Emergency Department (ED), with the aim of bringing clinical practice closer to the available scientific evidence by decreasing the use of bronchodilator medication.

This is a retrospective descriptive study conducted in the ED of a specialised paediatric hospital which treats between 65 000 and 70 000 episodes per year. Medical staff participating in ED care during the study period included 8 paediatricians working full-time in the ED and 48 doctors from other paediatric specialties, as well as 120 paediatric residents and 50 family and community medicine residents.

Results

A total of 1767 children under the age of 2 years with AB were included in the study (Fig. 1). The annual reports of the ED indicate that 1017 episodes of AB were managed in 2012, 1126 in 2014, 1143 in 2016 and 1162 in 2018. The patients included in the study represented 44.1% of all AB cases managed in 2012, 35.8% in 2014, 37.5% in 2016 and 41.9% in 2018.

Figure 1.

Patients included in the study. ED, emergency department; AB, acute bronchiolitis.

(0.18MB).

Table 2 presents the ages and clinical characteristics (respiratory rate, oxygen saturation and severity) of the patients included in each of the study periods.

Table 2.

Demographic and clinical characteristics of the patients with acute bronchiolitis.

	December 2012	December 2014	December 2016	December 2018	P value
n	449	403	429	487
Median age in months (IQR)	5.7 (3.1–9.2)	5.9 (3.3–9.0)	6.5 (3.6–10.5)	5.4 (3.0–8.9)	.167
Median respiratory rate, bpm (IQR)	48 (40–56)	48 (41–55)	48 (40–54)	48 (42–56)	.903
Median oxygen saturation, % (IQR)	97 (96–99)	97 (96–99)	97 (96–99)	97 (95–99)	.853
Mild AB, % (N)	68.3 (307)	67.5 (272)	64.8 (278)	67.9 (331)	.760
Moderate AB, % (N)	27.2 (122)	27.8 (112)	31.7 (135)	28.1 (137)
Severe AB, % (N)	4.5 (20)	4.7 (19)	3.5 (16)	3.9 (19)

AB, acute bronchiolitis; IQR, interquartile range; bpm, breaths per minute.

The described intervention achieved a 84.8% reduction in the use of salbutamol in the ED (95% confidence interval [CI], 79.1%–89.0%; P < .001) and a 98.4% reduction in the use of epinephrine (95% CI, 88.6%–99.8%; P < .001). When it came to the prescription of nebulised salbutamol on discharge, we found a decrease of 92.1% (95% CI, 88.0%–94.8%; P < .001) (Table 3).

Table 3.

Bronchodilator treatment used in the emergency department.

	December 2012	December 2014	December 2016	December 2018	P value
n	449	403	429	487
Nebulised epinephrine, n (%)	58 (12.9) (95% CI, 10.1%–16.3%)	39 (9.7) (95% CI, 7.2%–13%)	8 (1.8) (95% CI, 0.9%–3.6%)	1 (0.2) (95% CI, 0%–1.1%)	<.001
Nebulised salbutamol, n (%)	230 (51.2) (95% CI, 46.6%–55.8%)	137 (34.0) (95% CI, 29.5%–38.7%)	85 (19.8) (95% CI, 16.3%–23.8%)	38 (7.8) (IC 95%: 5.7%–10.5%)	<.001
Salbutamol on discharge, n (%)	234 (65.7) (95% CI, 60.6%–70.4%)	140 (45.4) (95% CI, 39.9%–51.0%)	81 (23.1) (95% CI, 19.0%–27.8%)	25 (5.2) (95% CI, 3.4%–7.8%)	<.001

CI, confidence interval.

Table 4 summarises the results regarding the effect of the strategy by age (younger or older than 12 months) and severity.

Table 4.

Proportion of use of salbutamol in the emergency department by age and severity.

	Proportion of salbutamol used, % (95% CI), N/N
	December 2012	December 2014	December 2016	December 2018	P value
Younger than 12 months
Mild bronchiolitis	39.5% (33.7%–45.6%). 102/258	23.9% (18.9%–29.7%). 56/234	5.4% (3.1%–9.2%). 12/221	2.4% (1.1%–4.8%). 7/292	<.001
Moderate bronchiolitis	61.4% (51.6%–70.2%). 62/101	42.4% (33.1%–52.2%). 42/99	24.5% (17.3%–33.5%). 26/106	11.7% (7.1%–18.7%). 14/119	<.001
Severe bronchiolitis	61.1% (38.6%–79.6%). 11/18	29.4% (13.2%–53.1%). 5/17	50.0% (26.7%–72.2%). 7/14	11.1% (3.1%–32.7%). 2/18	.003
Older than 12 months
Mild bronchiolitis	65.3% (51.3%–77.0%). 32/49	55.3% (39.7%–69.8%). 21/38	24.6% (15.2%–37.3%). 14/57	17.9% (8.9%–32.6%). 7/39	<.001
Moderate bronchiolitis	100%. 21/21	84.6% (57.7%–95.6%). 11/13	82.8% (65.4%–92.4%). 24/29	38.9% (20.3%–61.3%). 7/18	.002
Severe bronchiolitis	100%. 2/2	100%. 2/2	100%. 2/2	100%. 1/1	–

CI, confidence interval.

As regards patient outcomes, the median length of stay in the ED decreased from 59 min (IQR, 24–99) in 2012 to 25 min (IQR, 15–78) in 2018 (P < .001). In 2012, the rate of hospital admission was 20.7% (95% CI, 17.2%–24,7%), the readmission rate was 16.3% without hospitalisation (95% CI, 12.8%–20.5%) and 4.5% with hospitalisation (95% CI, 2.8%–7.1%). In 2018, the rate of hospitalization was 16.8% (95% CI, 13.7%–20.4%) (P = .045) and the readmission rate was 14.8% without hospitalisation (95% CI, 11.7%–18.6%) and 4.0% with hospitalisation (95% CI, 2.4%–6.3%) (Table 5). The overall rate of hospitalisation documented in the annual report was 4.96% in 2012 (95% CI, 4.86%–5.20%), 5.02% in 2014 (95% CI, 4.87%–5.22%), 4.78% in 2016 (95% CI, 4.63%–4.95%) and 4.83% in 2018 (95% CI, 4.67%–4.99%) (P = .186).

Table 5.

Patient outcomes.

	December 2012	December 2014	December 2016	December 2018	P value
Median care time in Emergency Department, min (IQR)	59 (24–99)	60 (24–114)	26 (15–85)	25 (15–78)	<.001
Hospitalisation rate, n (%)	93 (20.7) (95% CI, 17.2%–24.7%)	95 (23.6) (95% CI, 19.6%–27.9%)	79 (18.5) (95% CI, 15.0%–22.3%)	82 (16.8) (95% CI, 13.7%–20.4%)	.045
Rate of readmission within 72 h with hospitalisation, n (%)	16 (4.5) (95% CI, 2.8%–7.1%)	20 (6.5) (95% CI, 4.2%–9.8%)	12 (3.9) (95% CI, 2.3%–6.8%)	16 (4.0) (95% CI, 2.4%–6.3%)	.372
Rate of readmission within 72 h without hospitalisation, n (%)	58 (16.3) (95% CI, 12.8%–20.5%)	40 (13.0) (95% CI, 9.7%–17.2%)	48 (13.7) (95% CI, 10.5%–17.7%)	60 (14.8) (95% CI, 11.7%—18.6%)	.940

CI, confidence interval; IQR, interquartile range.

Table 6 shows the multivariate analysis of the main outcomes adjusted for age and severity.

Table 6.

Multivariate analysis of the main outcomes adjusted for age and severity of symptoms.

	2012	2014	2016	2018
Nebulised epinephrine	Reference	OR 0.63 (95% CI, 0.38–1.06)	OR 0.10 (95% CI, 0.04–0.22)	OR 0.01 (95% CI, 0.00–0.06)
Nebulised salbutamol	Reference	OR 0.37 (95% CI, 0.26–0.52)	OR 0.11 (95% CI, 0.07–0.16)	OR 0.04 (95% CI, 0.03–0.07)
Salbutamol on discharge	Reference	OR 0.31 (95% CI, 0.21–0.46)	OR 0.07 (95% CI, 0.05–0.11)	OR 0.02 (95% CI, 0.01–0.03)
Hospitalisation rate	Reference	OR 1.25 (95% CI, 0.83–1.88)	OR 0.83 (95% CI, 0.55–1.27)	OR 0.41 (95% CI, 0.27–0.63)

CI, confidence interval; OR, odds ratio.

Maintaining an improvement strategy for 5 years, with progressively restrictive indications for bronchodilator treatment, succeeded in reducing the use of epinephrine and salbutamol below the levels previously published in studies conducted in the ED setting, without increasing length of stay in the ED, the hospitalisation rate or the readmission rate.

We ought to highlight the homogeneity of the patients across the seasons under study in terms of age (similar to that in other studies33,34), clinical parameters, such as respiratory rate and oxygen saturation, and severity.

Prior to the intervention, half the patients received salbutamol at the ED, and salbutamol was prescribed to 2 out of every 3 patients on being discharged home, similarly to what has been reported in other studies.27,33 We managed to decrease its use drastically (<8%), bringing it below the achievable benchmark of care, which Ralston et al, after a systematic review of published quality improvement strategies in hospitalised patients with AB, set at 16%,29 and our own initial target. The maintenance of this strategy over time and more restrictive recommendations succeeded in bringing the use of salbutamol below that threshold, which was not the case when we made the first analysis of the impact of this intervention in 2016.35 There are similar quality improvement strategies described in the literature, but few of them focus on the ED setting. Akenroye et al, with a local protocol introduced in 2011, managed to lower the use of salbutamol from 54% to 47% after a 2-year intervention; they achieved greater success by reducing diagnostic testing.27 Montejo et al describe a coordinated strategy of an ED with primary care centres, which, in the case of the ED, managed to bring the use of salbutamol down to 9.1%, although the use of epinephrine remained unchanged (9%).28 In our study, treatment with salbutamol was reduced both in children younger than 12 months and in the 12- to 24-month age group. If we take account only of patients aged less than 12 months, as is the case in most recent studies, the use of salbutamol in the ED was even lower (5.3%) and the decrease was independent of severity. In the 12- to 24-month age group, a reduction was observed only in mild or moderate bronchiolitis, which was consistent with the results described by Tyler et al.36

Nebulised epinephrine was used primarily in younger patients. The initial situation was similar to that of other EDs.33 In 2012, the CPGs mentioned the possibility of performing a therapeutic test when the evidence was considered equivocal or insufficient.8,9 A reduction in the use of epinephrine towards its total elimination, in accordance with the increasingly restrictive recommendations contained in the protocol, has not been described in any other study.

Healthcare overload, increased waiting times and the decreased availability of observation beds that occurs during the annual epidemic of AB caused by RSV make the length of stay in the ED a particularly important factor. More than half the patients who attended the ED with AB in the last year of the study were discharged in less than 30 min. The decrease in the use of bronchodilators makes it possible to send patients home more quickly, producing a saving not only of material resources but also of human resources.

For the strategy to be successful it is important to show that the hospitalisation and 72-h readmission rates did not increase with the reduction in administration of drugs. In this study, the hospitalisation rate for patients treated for AB not only did not increase, but decreased and stood below the levels described in other studies.37,38 This may be related to an increasing assurance on the part of all the professionals working in the ED in treating this disease, with a reduction in the side effects associated with the medications not being administered,39 as well as with a decrease in the oxygen saturation threshold contained in the protocol to determine the need of hospital admission, which has been gradually reduced from 94% to 91%. This decrease in the hospitalisation rate of patients with AB is not perceptible in the overall rate of admission of the ED.

Given the results of the intervention described, and taking all the available evidence into consideration, perhaps the wisest course is to forget about treatment with bronchodilators once and for all and concentrate efforts on standardising and coordinating care delivery across all levels of care. We must be very cautious in interpreting the modest positive results shown in some studies which continue to analyse various therapeutic interventions together or separately in the interminable search for that subgroup of patients with AB that can benefit from them. The only question to be resolved is the role of high-flow oxygen therapy compared to conventional oxygen therapy outside intensive care units, introduced by many hospitals in recent years despite the lack of solid evidence supporting its use.40 Finally, it is essential to devote the necessary time to explain to parents the expected course of the disease, the absence of effective treatments, the general recommendations and the signs of worsening that they must watch out for at home.

The results of the quality improvement strategy implemented have the following limitations:

• -

  The study was conducted in a single centre. Although it might be possible to extrapolate the intervention and the resources underpinning it to other centres, other factors specific to the hospital could have assisted its success.

• -

  There could be a long-term trend of reduction in the use of bronchodilators for management of AB that acted independently of the strategy implemented and that could not be assessed due to the lack of a control group. However, there is an extensive literature on the limited effect of CPG recommendations in the absence of other interventions.22,35

• -

  Given that the impact of the intervention was only measured in the month corresponding to the peak incidence of the epidemic season, the study did not allow us to evaluate the variability of prescription that may exist in the other months.

• -

  Although most of the study variables are objective and easily measurable, in the severity scale objective parameters coexist with subjective ones, and therefore mistakes may be made in interpreting them despite the provided clinical teaching sessions.

• -

  With regard to patient outcomes, we cannot guarantee that they did not visit other hospitals and eventually admitted. If this did occur, there is no reason why it should have happened more frequently in one season than in another.

• -

  Although the study is powerful enough to estimate the decrease in the use of bronchodilators, it was not designed to estimate the positive effects on the outcomes that were observed, so these must be interpreted with caution.

The future in AB demands that we continue moving forward in reducing the variability of clinical practice and the inappropriate use of therapeutic resources. Given the modest effectiveness of CPGs in changing the usual practice of health care professionals, local quality improvement strategies need to be designed and implemented. To be successful, such interventions must be continued over time, periodically updated, endowed with adequate resources and capable of producing measurable results. In our case, the findings described have served as a starting point for setting in motion the “Doing Nothing Strategy in Bronchiolitis” project, in coordination with all the EDs and primary care centres in the Community of Madrid. Eliminating the use of salbutamol and epinephrine in AB patients is possible.

This research did not receive any external funding.

The authors have no conflicts of interest to declare.