BACKGROUND AND OBJECTIVE: Reduction of critical asthma management time can reduce intensive care utilization. The goal of this study was to determine whether a Critical Asthma Standardized Clinical Assessment and Management Plan (SCAMP) can decrease length of critical asthma management time.
METHODS: This retrospective study compared critical asthma management times in children managed before and after implementation of a Critical Asthma SCAMP. The SCAMP used an asthma severity score management scheme to guide stepwise escalation and weaning of therapies. The SCAMP guided therapy until continuous albuterol nebulization (CAN) was weaned to intermittent albuterol every 2 hours (q2h). Because the SCAMP was part of a quality improvement initiative in which all patients received a standardized therapy, informed consent was waived. The study was conducted in Medicine ICU and Intermediate Care Units in a tertiary care freestanding children’s hospital. Children ≥2 years of age who had CAN initiated in the emergency department and were admitted to the Division of Medicine Critical Care with status asthmaticus were included. The time to q2h dosing from initiation of CAN was compared between the baseline and SCAMP cohorts. Adverse events were compared. The Mann-Whitney test was used for analysis; P values <.05 were considered statistically significant.
RESULTS: There were 150 baseline and 123 SCAMP patients eligible for analysis. There was a decrease in median time to q2h dosing after the SCAMP (baseline, 21.6 hours [interquartile range, 3.2–32.3 hours]; SCAMP, 14.2 hours [interquartile range, 9.0–23.1 hours]; P < .01). There were no differences in adverse events or readmissions.
CONCLUSIONS: A Critical Asthma SCAMP was effective in decreasing time on continuous albuterol.
Asthma continues to be a national and international health priority.1,2 A pediatric study in 2011 showed that although the rate of admission for asthma decreased by one-half during a 15-year period, the rate of critical asthma care tripled and hospital costs increased from $6.6 million to $9.5 million for a single state.3 Two reviews of the Pediatric Health Information System found marked regional variations in the choice of critical asthma medical interventions when children did not improve with standard therapy (inhalation of β-agonists and steroids).4,5 These investigators suggested that a clinical asthma score which directs asthma interventions may decrease clinical variation in care and improve cost of care.
Critical asthma is considered to be present when a patient requires continuous albuterol nebulization or more advanced therapies for respiratory distress after standard asthma treatment in the emergency department (ED) has failed to result in improvement. Ten thousand children per year will require critical asthma care,6 with estimates of the average length of stay (LOS) ranging from 66.2 ± 8.7 hours7 to 116 ± 125 hours.8,9 LOS can be decreased with the use of “standard” pediatric inpatient asthma pathways (LOS, 4.2–2.7 days),10 but there are still concerns regarding the total cost of care and readmission rates.11 Despite these concerns, there are no “standard” asthma therapies for life-threatening asthma.5
Standardized Clinical Assessment and Management Plans (SCAMPs) have been shown to be effective in decreasing variation in clinical practice in children.12,13 They are iterative, data-backed, consensus-based care pathways aimed at improving patient care, reducing unnecessary resource utilization, and lessening clinician practice variability. In contrast to the implementation of traditional clinical practice guidelines, the SCAMP process involves data collection on pathway compliance and measures specific outcomes. This process thus allows for evaluation of the pathway and uses these data to improve upon the pathway in subsequent iterations. In addition, if an area in the algorithm is consistently associated with nonadherence, this pathway is compared with those with adherence to understand if the alternative pathway leads to improved outcomes. Oversight of the production of SCAMPs is provided by the Institute for Relevant Clinical Data Analytics, which is a nonprofit, tax-exempt organization that provides the education and resources for the development, implementation, and analysis of SCAMPs at its member institutions.
The Critical Asthma SCAMP was constructed as a quality improvement project to standardize asthma therapy in children who had continuous albuterol nebulization (CAN) initiated, with helium-oxygen–driven CAN (Heliox-CAN) as the adjunct therapy if the patient clinically did not improve or worsened. The hospital-wide asthma severity score (HASS) was used to direct the step-wise escalation or de-escalation of therapy until intermittent albuterol every 2 hours (q2h) occurred, which is the time point at which a patient can be transitioned to the inpatient floor at our institution. This particular SCAMP was designed to foster autonomy of the nursing and respiratory therapists by instituting conditional orders for management.
The primary aim of the present study was to determine whether the implementation of a SCAMP that used an HASS to trigger escalation and weaning of asthma therapies resulted in a decreased time of critical care management. A secondary aim was to determine whether the Critical Asthma SCAMP could be implemented without leading to worse patient outcomes or increased adverse events.
This single-site retrospective study included all patients (≥2 years of age) with a diagnosis of status asthmaticus in whom CAN was initiated in the ED and were admitted to the Boston Children’s Hospital (BCH) Intermediate Care Unit (InCU) or Medicine ICU (MICU) before any other ICU asthma therapies (intravenous terbutaline and noninvasive or invasive positive pressure ventilation [NIPPV and intermittent mechanical ventilation, respectively]). The InCU and MICU are subdivisions of the Division of Medicine Critical Care but geographically located adjacent to each other. The InCU is staffed 60% of the time by pediatric hospitalists, but an intensivist is immediately available if needed. The remainder of the InCU staffing, the entire MICU staffing, and all times from 5:00 pm to 8:00 am are staffed by pediatric intensivists.
The Critical Asthma SCAMP (Fig 1) standardized the administration of CAN, intravenous steroids, intermittent ipratropium inhalation, and Heliox-CAN as an adjunct therapy in children with critical status asthmaticus. The initiation of CAN occurred independently of the SCAMP per the clinical discretion of the ED attending physician. Before initiation of the SCAMP, all children ≥2 years of age admitted to the InCU or MICU on CAN were treated per the discretion of the Division of Medicine Critical Care attending physician. After implementation of the SCAMP, all children >2 years of age on CAN were managed via the SCAMP. There was no change in the physician, nursing, or respiratory therapy staffing ratio between the baseline and SCAMP periods, and no other algorithms were used to manage children with critical status asthmaticus.
HASS was used to direct the escalation and de-escalation of critical asthma therapies until albuterol was weaned to intermittent dosing q2h. The HASS is currently a nonvalidated asthma severity score created for local use (S. McBride, MD, K. McCarthy, CPN, J Wong, MD, V Chiang, MD, unpublished observations) (Table 1). The primary end point was defined as the time from initiation of CAN to the time when the patient reached q2h albuterol. In children who may have reverted to more frequent albuterol dosing than q2h or back to CAN due to worsening course, the final time that the child successfully achieved q2h dosing was used to assess the primary outcome. The time to q2h dosing was compared between baseline and SCAMP patients.
The study time periods for the baseline period ranged from May 2011 to March 2012 and from June 2012 to March 2013 for the SCAMP cohort. The time period between March 2012 and June 2012 was a pilot period to correct logistical issues of implementation, but the SCAMP was not modified during this time period. All patients meeting inclusion criteria during the time periods were eligible for analysis. Patients were excluded if CAN was initiated outside BCH because the ED treatment at BCH for asthma is also standard, and we could not account for variation in treatment at referring facilities. Information from the medical record pertinent to medical therapy for critical asthma was extracted, compiled, and de-identified for analysis.
Informed consent was waived because this study was a retrospective medical record review. The study was approved by the institutional review board of BCH.
Patient demographic variables including age, sex, BMI, and race were recorded. The vital signs on presentation in the ED were recorded; they included heart rate, respiratory rate, systolic and diastolic blood pressures, pulse oximetry saturation, and percentage of inspired oxygen. The initial venous blood gas results (pH, partial pressures of carbon dioxide and oxygen) in the ED were recorded. The first HASS in the ED and neighboring initial HASS before the start of CAN (±30 minutes) were recorded.
The time of initiation of CAN, time of intermittent albuterol nebulization q2h, time of admission, and time of discharge were recorded. The number of patients who escalated to Heliox-CAN per the algorithm was recorded. Successful weaning from continuous albuterol to intermittent albuterol was defined as not requiring restart of CAN at any point during the hospital course after the first weaning to intermittent albuterol. Failure of successful weaning was recorded. Escalation to any other ICU asthma therapies (terbutaline, NIPPV, or intubation) beyond CAN or Heliox-CAN was also recorded.
To assess whether implementation of the SCAMP was associated with any adverse events, we recorded results of chest radiograph and electrocardiograph (ECG) testing. ST-segment elevation was defined if formal cardiology interpretation of the ECG was of ST-segment elevation. The incidence of adverse effects such as intubation, respiratory or cardiac arrest, and pneumothorax and pneumomediastinum were recorded. Readmissions to the InCU or MICU after discharge to the inpatient floor within 24 hours, readmission to the hospital within 7 days, and any ED visits within 72 hours of discharge were recorded as markers of failure of rapid weaning.
A planned analysis was performed at 9 months from SCAMP initiation to assess outcomes and balancing measures. Demographic and ED presentation characteristics were compared between the baseline and SCAMP cohorts with χ2 tests for categorical variables and Wilcoxon rank sum tests for continuous variables. Similarly, time to q2h and LOS (hours) was compared between the 2 groups by using the Mann-Whitney test.
The following comparisons were made between the 2 groups by using Fisher’s exact test: the rates of escalation to Heliox-CAN; restart of CAN after weaning to intermittent albuterol; and escalation to IV terbutaline, NIPPV (continuous positive airway pressure or bilevel positive airway pressure), or intermittent mechanical ventilation. Rates of ancillary testing were compared with χ2 tests, and rates of adverse events and readmission were compared by using Fisher’s exact tests.
Medical record data were extracted and entered into a Microsoft Excel 2007 spreadsheet (Microsoft Inc, Redmond, WA). Statistical analysis was performed by using SAS version 9.3 (SAS Institute, Inc, Cary, NC). In all tests, statistical significance was achieved if a 2-sided P value was <.05.
In total, 150 baseline and 123 SCAMP patients with a diagnosis of asthma who were initiated on CAN before any ICU therapy were available for analysis. There was no statistical difference between the groups in terms of demographic characteristics except for race (P = .04) (Table 2). The median ages of patients in the baseline and SCAMP groups were 6.2 years and 7.8 years, respectively (P = .29). There were no clinically significant differences in initial vital signs, HASS, or venous blood gas results on clinical presentation between the 2 groups. The first HASS in the ED and the initial HASS before the start of CAN were not statistically different, with a median score of 9 in both cohorts.
There was a significant increase in the use of Heliox-CAN after SCAMP implementation (14.6% baseline vs 46.4% SCAMP; P < .001). There was a 34% decrease in median time to q2h dosing (21.6 hours [interquartile range (IQR), 13.5 to 32.3 hours] vs 14.2 hours [IQR, 9.0 to 23.1 hours]; P < .01), but there was no difference in LOS between the baseline and SCAMP groups (Table 3). In the patients who received Heliox-CAN, there was a significant decrease in median time to q2h dosing: 40.3 hours (IQR, 30.5 to 75.5 hours) for baseline and 20.2 hours (IQR, 15.8 to 33.5) for SCAMP; P < .01. There was no difference in patients requiring restart of CAN after weaning to q2h dosing (14.8% vs 8.8%; P = .14). Similarly, there was no increase in the need for ICU therapy or individual subtype of ICU therapy between the 2 groups (16.0% vs 12.0%; P = .34) (Table 4).
The SCAMP algorithm was adhered to in full by the treating team in 84 (68.3%) of 123 patients. In 39 (31.7%) patients, there was at least 1 diversion from the recommended pathway. In review of patients with diversions, there was no obvious commonality among diversions to allow separate analysis. Patients with SCAMP full adherence were older than those with diversions (median age, 8.3 vs 6.6 years; P = .04). There was no difference in adherence according to median initial HASS (9 in both groups; P = .51). There was no difference in median time to q2h dosing between SCAMP full compliance patients versus diversion patients (16.2 vs 13.6 hours; P = .89). In diversions, 17 patients (43.6%) were weaned from therapy when the SCAMP algorithm recommended continuing therapy because the care team believed that patients were improved, 8 (20.5%) were not weaned when the SCAMP recommended weaning because the care team believed patients still required CAN, 6 (15.4%) escalated therapy when the SCAMP recommended no change due to concerns regarding worsening, and 4 (10.3%) did not escalate therapy when the SCAMP recommended escalation because the care team felt the patients did not require escalation. Four patients did not have a documented reason for diversion.
There was no difference in the initial utilization of chest radiographs between the groups (87.3% baseline vs 85.6% SCAMP; P = .67). Similarly, among those patients with an initial chest radiograph, there was no difference in the number of follow-up chest radiographs between the groups (29.0% baseline and 29.9% SCAMP; P = .88). Fewer initial ECGs were obtained in the SCAMP group (30.7% baseline and 19.2% SCAMP; P = .03). However, there was no difference in the use of follow-up ECGs between the groups (18.0% baseline and 10.4% SCAMP; P = .61).
Only 1 patient required intubation, and this patient was part of the SCAMP group (P = .45). No patient experienced respiratory arrest or cardiac arrest, needed extracorporeal membrane oxygenation, or required isoflurane inhalation. There were no deaths. One patient developed a pneumothorax, and this patient was in the SCAMP group (P = .45); 4 patients developed pneumomediastinum (2 baseline and 2 SCAMP; P = .99). No patients required a chest tube placement. There were no differences in ST-segment changes found on the initial ECG (19 [41.3%] of 46 baseline and 7 [29.2%] of 24 SCAMP; P = .44) or subsequent ECG (2 [7.4%] of 27 baseline and 1 [5.9%] of 17 SCAMP; P = .99) between the groups. There were also no differences in prolonged QTc found on the initial ECG (6 [13.0%] of 46 baseline and 3 [12.5%] of 24 SCAMP; P = .99) or subsequent ECG (3 [7.5%] of 40 baseline and 1 [4.8%] of 21 SCAMP; P = .99) between the groups. Use of Heliox-CAN did not affect these findings (all, P > .05), and there were no findings of ST changes or prolonged QTc in patients who received Heliox-CAN in either group. One patient developed supraventricular tachycardia in the baseline group (P = .99), and 1 patient with a history of ectopic atrial tachycardia developed this condition in the SCAMP group. No patient required electrical cardioversion.
There was no difference in the number of patients who required readmission to either the InCU or MICU within 24 hours of transfer to the inpatient floor (1 baseline vs 3 SCAMP; P = .33). Only 1 patient required an ED visit within 1 week of discharge (0 baseline vs 1 SCAMP; P = .45), and no patient required readmission to the hospital within 7 days.
This study showed that the SCAMP significantly decreased by 34% the time it took to wean patients to intermittent albuterol from the start of continuous albuterol. The SCAMP improved clinical outcomes, and it was also safe. Despite the SCAMP having parameters for escalation of care to ICU therapies if the HASS increased despite use of CAN or Heliox-CAN, there was no increase in the need for ICU therapy compared with the baseline group. Similarly, despite the SCAMP dictating weaning to intermittent nebulization from CAN when the HASS decreased, there was no increase in the need for restarting CAN after the initial weaning to q2h dosing.
The 2007 National Asthma Education and Expert Panel provided extensive recommendations for standardization of outpatient asthma and management of acute exacerbation. However, the panel had limited evidence and guidance for adjunct therapies for critical asthma, including Heliox-CAN, which was categorized as Level B evidence. In addition, indications for initiation of Heliox-CAN were not detailed.14 Heliox-CAN is a controversial therapy that has been shown to be effective15–19 and ineffective7,20 in the treatment of status asthmaticus. In this SCAMP, Heliox-CAN was standardized to both severity and time of application. The SCAMP did increase the use of Heliox-CAN compared with previous research, indicating that compliance to the SCAMP occurred by the treating teams. Although the methods of the SCAMP and this study do not prove that Heliox-CAN is an effective stand-alone therapy, it does show that when Heliox-CAN is part of a bundle of care for critical asthma, this bundle is effective in decreasing time on continuous albuterol.
The present study had several limitations. The SCAMP algorithm was followed fully in only 68.3% of patients. It is possible that the diversions themselves were appropriate clinical decisions, which drove improvement in the other patients, and had the algorithm been followed, these patients would have not had similar outcomes as those who did not deviate. There was no pattern in the deviations, but the majority occurred with decision to wean, and this approach may have further improved the entire cohort’s time to q2hs dosing. In addition, the HASS was developed at BCH and has not been formally validated in children with critical asthma as with other asthma scores.21–23 Despite this limitation, the scoring system did trend the clinical condition of the patient and was used in all patients. In addition, due to the SCAMP being a bundle of care, it is unclear which part of the bundle was the most efficacious in improving outcomes. Seasonality differences before and after the study period could have affected the results; however, we elected to use data that immediately preceded the implemented protocol so as to not have a full season gap between before and after groups. Finally, the SCAMP was implemented as a quality improvement project and, as a result, was not a blinded study; thus, confounders (eg, treatment biases, quality improvement oversight) could have driven the observed improvement. Despite decreasing time on continuous albuterol, it did not decrease hospital LOS. Lack of statistical significance for LOS is most likely due to a power issue because of smaller relative changes and more variance around LOS. In addition, with reaching 2 hours, the quality improvement oversight ended and asthma management reverted to standard clinical management by clinical examination (predominantly auscultation) rather than by HASS. Perhaps an extension of the SCAMP to hospital discharge would affect hospital LOS.
Based on SCAMP methods,24 a revision of the SCAMP is ongoing and will be implemented to assess if patients who had incomplete compliance with the critical asthma bundle had an impact on outcomes and to consider if themes on diversions should be adopted in a new iteration. Lessons learned from further analysis of the data from this SCAMP can lead to future modification of the SCAMP algorithm to further improve outcomes.
A Critical Asthma SCAMP that provided a structured bundle of care, with a HASS-guided management plan for escalation and weaning of CAN and Heliox-CAN, reduced time on continuous albuterol. The approach of bundled standardized care can be a model for improving outcomes and may lead to earlier results when double-blind, randomized controlled trials might not be feasible.
The authors express their gratitude to the nursing and respiratory staff in the InCUs and MICUs, Medicine Critical Care Division SCAMPs Committee, and the Institute for Relevant Clinical Data Analytics. They specifically thank Patricia Mantell, RN, Jason M. Thorton, RN, Daria Donelly, RTT, Danielle Dwyer, MD, Denise Anderson, RN, Heather Kennedy, RN, Melissa Whalen, MPH, Jessica Sang, MPH, Joshua Barlett, BA, Estella Kanevsky, MPH, Dorothy Miller, MPH, Caitlyn McCarthy, MPH, and Suvidha Dabas, MPH.
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
FUNDING: Supported by departmental institutional funds.
POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.
- Bratton SL,
- Newth CJ,
- Zuppa AF,
- et al
- Carroll CL,
- Stoltz P,
- Raykov N,
- Smith SR,
- Zucker AR
- Chen KH,
- Chen CC,
- Liu HE,
- Tzeng PC,
- Glasziou PP
- Angoff GH,
- Kane DA,
- Giddins N,
- et al
- Farias M,
- Jenkins K,
- Lock J,
- et al
- Kim IK,
- Phrampus E,
- Venkataraman S,
- et al
- Rodrigo GJ,
- Castro-Rodriguez JA
- Copyright © 2017 by the American Academy of Pediatrics