BACKGROUND AND OBJECTIVES: Despite widespread use of the rotavirus vaccine in the last decade, dehydrating illnesses impact almost 2 billion children worldwide annually. Evidence supports oral rehydration therapy as a first-line treatment of mild to moderate dehydration. Ondansetron has proven to be a safe and effective adjunct in children with vomiting. We implemented a clinical pathway in our pediatric emergency department (ED) in January 2005 to improve care for this common condition. Our objective in this study was to determine the long-term impact of the pathway for acute gastroenteritis (AGE) on the proportion of patients receiving intravenous (IV) fluids and ED length of stay (LOS) for discharged patients.
METHODS: Cases were identified by using International Classification of Diseases, Ninth Revision, Clinical Modification diagnosis codes. We used statistical process control to analyze process and outcome measures for 2 years before and 10 years after pathway implementation.
RESULTS: We included 30 519 patients. We found special cause variation with a downward shift in patients receiving IV fluids after initiation of the pathway and later with addition of ondansetron to the pathway from 48% to 26%. Mean ED LOS for discharged patients with AGE decreased from 247 to 172 minutes. These improvements were sustained over time.
CONCLUSIONS: Implementation of a clinical pathway emphasizing oral rehydration therapy and ondansetron for children with AGE led to decreased IV fluid use and LOS in a pediatric ED. Improvements were sustained over a 10-year period. Our results suggest that quality-improvement interventions for AGE can have long-term impacts on care delivery.
- AGE —
- acute gastroenteritis
- CSW —
- clinical standard work
- ED —
- emergency department
- ICD-9-CM —
- International Classification of Diseases, Ninth Revision, Clinical Modification
- IV —
- LOS —
- length of stay
- ORT —
- oral rehydration therapy
- QI —
- quality improvement
- SPC —
- statistical process control
Dehydration because of acute gastroenteritis (AGE) is one of the most common problems of childhood. Despite widespread use of the rotavirus vaccine in the last decade, dehydrating illnesses impact almost 2 billion children worldwide annually and cause more than 700 000 deaths.1–3 Treatment options include intravenous (IV) fluids, nasogastric fluids, and oral rehydration therapy (ORT). Since the 1980s, researchers have demonstrated that ORT is a safe and effective therapy for mild to moderate dehydration.4–8
The World Health Organization, the American Academy of Pediatrics, and the Centers for Disease Control and Prevention have promoted the use of evidence-based guidelines emphasizing ORT since the mid-1990s.9,10 Our institution first implemented an emergency department (ED)-based clinical pathway for AGE in 2005 to standardize and improve the care of children presenting with dehydration because of simple AGE with vomiting. The pathway emphasizes the use of ORT for mild to moderate dehydration and includes the use of ondansetron for children with vomiting. Our aim in this project was to determine the short- and long-term impacts of implementing this evidence-based pathway for children with mild to moderate dehydration on ED length of stay (LOS), IV fluid use, and 72-hour unplanned returns.
The setting was a tertiary, university-affiliated, 323-bed pediatric hospital with a dedicated pediatric ED (43 000 annual visits). Since 2002, the institution has developed and implemented clinical standard work (CSW) pathways for common conditions. CSW is an evidence-based approach to the management of particular patient populations or diagnoses. Clinical pathways are designed as flowcharts or algorithms to guide provider decision-making and provide education to learners on the evidence behind the recommendations. They are linked to diagnosis-specific electronic order sets. Currently, more than 50 CSW pathways have been implemented. Pathways are formally reviewed on a regular basis to ensure they remain consistent with current medical literature and national guidelines.
A multidisciplinary stakeholder group developed the clinical pathway for children presenting to the ED with AGE. This group included physicians and nurses from the ED and inpatient units as well as pharmacists and information technology specialists.
Pathway development began with a literature review of Embase, PubMed, and national guideline clearinghouses. The team of physician stakeholders rated the quality of the evidence and generated a series of recommendations. When evidence was not available from the literature, recommendations were made on the basis of local expert consensus. The team used these recommendations to develop the content of the pathway and electronic order set. The original AGE pathway and electronic order sets were launched in January 2005 (Fig 1). Subsequent re-evaluation of the literature resulted in minimal change to the pathway over time (Fig 2).11
The implementation of the pathway was led by the ED physicians as well as clinical nursing specialists. Multiple strategies were used to support uptake and adherence. Before implementation, the pathway was discussed at ED provider meetings. E-mail notifications including physician and nurse job aids were sent. Copies of the pathway were placed outside patient rooms and in provider work areas to insure visibility and access.
In March 2006, changes were made to ondansetron access and use. These changes included the following: (1) hospital formulary, the addition of gastroenteritis as an indication for the use oral ondansetron with dosage recommendations; (2) AGE pathway, recommendations for ondansetron use for patients with vomiting in the past 6 hours were added; and (3) medication stocking, ondansetron was located in ED Omnicell for easy accessibility.
This was a quality-improvement (QI) study to assess outcomes related to the implementation of the AGE pathway. We considered process, outcome, and balancing measures. Process measures were selected to reflect adherence to the recommendations of the pathway, specifically the proportion of AGE patients receiving IV fluids. The outcome measure used was LOS for patients discharged from the ED and was selected to determine the efficiency of care provided. The balancing measure of unplanned returns to the ED within 72 hours of discharge, for symptoms related to gastroenteritis, was selected to monitor and identify any unintended consequences of the AGE pathway; an increase in 72-hour returns could signal inappropriate discharge or inadequacy of care provided during the initial ED visit.
We analyzed data for 2 years before and 10 years after the implementation of our AGE pathway. We included children aged 3 months to 18 years who presented to our ED with AGE from January 2003 through April 2015 and who were eligible for the AGE pathway. We defined the population of patients “who were eligible for the AGE pathway” as those having a primary International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) diagnosis code associated with both AGE and vomiting. Patients were excluded from the analysis if they were <3 months old or assigned ICD-9-CM or International Classification of Diseases, 10th Revision discharge diagnosis codes associated with bloody diarrhea or comorbid conditions (eg, medical complexity, renal failure, cardiac disease, neurologic disease, and sepsis) that would have excluded their eligibility for the pathway (Supplemental Tables 2 and 3).
Descriptive statistics were used to compare demographics of the pre- and postpathway groups. Statistical process control (SPC) was used to analyze process, outcome, and balancing measures.12,13 All control charts were created by using the QI Charts 2.0 add-in for Microsoft Excel (Process Improvement Products, Austin, TX).
This study was approved under expedited review by our institution’s Internal Review Board.
During the 12-year study period, 30 519 patients met eligibility criteria. There were no statistically significant differences between the pre- and postpathway groups with regard to age, sex, or race/ethnicity. The postpathway group included a larger percentage of children with commercial insurance (Table 1).
There were challenges associated with the initial implementation of the pathway. First, many referring primary care providers, ED providers (doctors and registered nurses), and some families believed that IV fluids were more efficient and effective than ORT, and there was hesitation to forego IV placement. Sharing initial process metrics with the care team led to increased provider buy-in. Second, ondansetron was a relatively new medication in the pediatric ED. Limited experience with the medication led to pharmacy concerns about appropriate dosing and physician concerns that surgical pathologies may be masked. With time and ongoing provider education, many of these initial concerns dissipated. Despite these challenges, the pathway underwent minimal modification over time (Figs 1 and 2).
We noted special cause variation with a downward shift (8 points below the centerline) in the percentage of patients receiving IV fluids from 48% to 44% after pathway implementation. Special cause variation occurred again, with an additional downward shift of the centerline to 26%, after changes were made related to ondansetron use and accessibility (Figs 3 and 4). The decrease in IV fluid use was sustained for the entirety of the 10-year postpathway period. We also measured oral ondansetron use for patients with AGE in the ED over the same study period. We found an increase from a baseline of 4% to 20% after pathway implementation. The proportion of patients receiving ondansetron in the ED steadily increased after it was more readily available and was sustained at ∼65% by the end of the study period (Supplemental Fig 6).
Similar to the pattern of decreased IV fluid use, we noted special cause variation with a clinically significant decrease in ED LOS for discharged patients by ∼60 minutes immediately after the pathway was implemented. We noted further improvement with an additional 20-minute decrease in ED LOS after oral ondansetron recommendations were added to the pathway.
Unplanned 72-hour returns to the ED for AGE patients did not change over this time frame (Fig 5).
In this study, we used SPC analyses to examine the long-term impact of a clinical pathway for AGE on IV fluid use and efficiency (regarding LOS) in the ED of a single free-standing, tertiary-care, pediatric hospital. We found that pathway implementation, including provider education regarding the use of ORT for patients with mild to moderate dehydration, decreased IV fluid use and ED LOS for discharged patients. Further improvements in IV fluid use and ED LOS were achieved once the hospital formulary and the pathway included recommendations for the use of ondansetron in gastroenteritis and the medication was readily available for use in our ED. Importantly, our analysis included 12 years of data, and our results were sustained for the duration of the 10-year postpathway period.
There is a wealth of published data supporting the use of ORT in children. One randomized controlled trial sought to demonstrate the noninferiority of ORT for successful rehydration of moderately dehydrated children. This study showed that ORT was as effective as IV fluids for rehydration in the ED and demonstrated that time to treatment was shorter for those patients receiving ORT.14 In a retrospective, cross-sectional, provincial study of pediatric AGE patients, the use of a medical directive for ORT was associated with fewer unplanned ED returns.15 Our results similarly demonstrate that the standardization of care with an emphasis on ORT was temporally related to a decreased use of IV fluids and improved ED throughout, without a significant change in unplanned returns.
The adjunctive use of ondansetron in AGE has also been well studied. The benefits of ondansetron include a reduction in the likelihood and frequency of vomiting, an increase in oral intake, and a decrease in the need for IV fluids.16–21 A cost analysis of ondansetron use estimated that routine administration of the medication to eligible children in the United States would save both society and health care payers >$60 million annually.22 Although we did not perform a cost-effectiveness analysis as part of this study, we did demonstrate improved outcomes once ondansetron became readily available.
Despite more than 2 decades of evidence supporting ORT and ondansetron as the standard of care for treatment of mild to moderate dehydration, gaps remain in current clinical practice. A survey comparing practice patterns of United States and Canadian providers found that 76% of Canadian physicians reported initiating ORT in moderately dehydrated children compared with 47% of their US colleagues. American physicians were found to administer larger IV fluid boluses over shorter time periods and repeated boluses more frequently.23 Although we were able to achieve and sustain significant reduction in our IV fluid use, from 44% to 26%, it is important to note that our new baseline is still higher than what is reported in Canadian EDs.24 This might reflect differences in patient, family, and provider expectations between the 2 countries. A survey of US pediatric emergency medicine program directors revealed that they do not routinely use ORT because they believe it takes longer than IV hydration, and they feel parents and referring providers expect IV fluids.25 Our findings support the use of ORT with ondansetron, rather than IV fluids, as an efficient treatment modality for most children with mild to moderate dehydration.
The utility of ED-based clinical pathways for dehydration secondary to AGE has been suggested by researchers in studies from Australia and Canada, but no studies to date have been published from US-pediatric EDs15,26,27 and none have demonstrated the long-term sustainability of QI efforts. There is a gap in the literature regarding how to achieve provider buy-in and maintain adherence or if results can be sustained over long periods of time. This study adds to this knowledge base, demonstrating that the maintenance of a well-established clinical pathway is a successful approach to producing ongoing improvements in provider practice and patient outcomes that are similarly sustained over time.
This study has limitations. First, patients were identified for analysis primarily by using ICD-9-CM discharge diagnosis codes. This method can lead to the misclassification of patients, although we expect any bias to be similar in the pre- and postpathway period. Second, secular trends, including the use of the rotavirus vaccine, may have had an impact on our results. Notably, rotavirus vaccine was licensed in 2008 and in regular use in Washington state by 2010. Although we did not note special cause variation on any of our SPC charts at that time, we do not know if widespread vaccine use changed the makeup of pediatric gastroenteritis patients (potentially resulting in an overall less-sick population of patients) and thereby contributed to the long-term sustainability of our results. Furthermore, although we assume that reduced resource use (ie, fewer IV starts and IV fluid boluses) and improved efficiency (reduced LOS) were cost-effective, we did not have cost data available for analysis. Future studies of the sustainability of QI interventions should include cost measures.
Finally, our institution was an early adopter of care standardization and has implemented clinical pathways since 2002. Over time, our culture has shifted such that providers are familiar with clinical pathway use. This infrastructure and level of provider buy-in may limit the generalizability of this intervention to other institutions.
The implementation of a clinical pathway emphasizing ORT and ondansetron for children with AGE led to sustained decreases in IV fluid use and ED LOS over a 10-year period. Our results suggest QI interventions, such as clinical pathways for AGE, can have long-term impacts on care delivery. Contextual factors, including institutional culture and leadership support for such interventions, are important considerations for implementation success and long-term sustainability.
- Accepted June 15, 2017.
- Address correspondence to Lori Rutman, MD, MPH, Division of Pediatric Emergency Medicine, Seattle Children’s Hospital, 4800 Sand Point Way NE, Seattle, WA 98105. E-mail:
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
FUNDING: Provided by the Marco J. Heidner Charitable Trust.
POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.
- GBD 2015 Mortality and Causes of Death Collaborators
- Ozuah PO,
- Avner JR,
- Stein REK
- Mackenzie A,
- Barnes G
- American Academy of Pediatrics, Provisional Committee on Quality Improvement, Subcommittee on Acute Gastroenteritis
- World Health Organization
- Seattle Children’s Hospital,
- O’Callaghan J,
- Akhter S,
- Austin E, et al
- Perla RJ,
- Provost LP,
- Murray SK
- Provost LP,
- Murray SK
- Spandorfer PR,
- Alessandrini EA,
- Joffe MD,
- Localio R,
- Shaw KN
- Carter B,
- Fedorowicz Z
- Boyd R,
- Busuttil M,
- Stuart P
- Copyright © 2017 by the American Academy of Pediatrics