ORIGINAL ARTICLE
Implementation of an Advanced Nursing Directive for Children With Right Lower Quadrant Pain Identifying Those Requiring Further Investigation and Improving Flow Metrics Graham C. Thompson, MD, FRCPC,* Simon Kwong, MSc,† Niklas Bobrovitz, MSc,‡ Stephanie Yang, MD,§ Erin Kate Pols, BScN,* Antonia S. Stang, MD, MBA, MSc,*|| Robin C. Eccles, MD, FRCSC,¶ and Veronica Kazoleas, MA# Objectives: Advanced nursing directives (ANDs) empower nursing staff to provide advanced levels of care before physician assessment. The objectives of this study were (1) to determine whether an AND for right lower quadrant (RLQ) pain could identify children who required any further investigation to diagnose appendicitis and (2) to determine whether children meeting AND criteria had better emergency department (ED) flow metrics compared with those who did not meet the criteria. Methods: Health records of children aged 3 to 17 years presenting to the ED with abdominal pain who were managed using the departmental AND for RLQ pain were reviewed. Primary outcomes included (1) the proportion of patients requiring further investigation to diagnose appendicitis and (2) the time interval from triage to blood draw. Secondary outcomes included additional ED flow metrics, perforation rate, and negative appendectomy rate. Results: An AND was completed for 210 children. Those who met the AND criteria were more likely to undergo further investigation to rule out appendicitis than those who did not meet the criteria (92/137 [67.2%] vs 32/73 [43.8%]; odds ratio [OR], 2.62; 95% confidence interval [CI], 1.40–4.90). Time to blood draw was significantly lower for those children meeting the AND criteria (74 vs 162 minutes, P < 0.001) as was time to hospital admission (271 vs 395 minutes, P = 0.008) and appendectomy (498 vs 602 minutes, P = 0.015). The negative appendectomy rate was 8.6% (5/58) for children meeting the AND criteria and 9.1% (2/22) for those not meeting the criteria (OR, 0.94; 95% CI, 0.14–10.67); the perforation rate was 29.3% (17/58) and 4.5% (1/22), respectively (OR, 8.17; 95% CI, 1.17–380.86). Conclusions: Children presenting to the ED with RLQ pain who meet the AND criteria are more likely to require further investigation to rule out appendicitis and have better department flow metrics than those who do not meet the criteria. Our results provide further evidence of the utility of ANDs in the ED. Key Words: appendicitis, advanced nursing directive, flow metrics (Pediatr Emer Care 2016;32: 352–356)
management strategy determination.1–4 Clinical prediction rules provide an objective measure to guide management.5–7 In addition, the use of departmental clinical pathways, protocols, and guidelines, also known as evidence-based clinical algorithms (EBCAs), has increased in the emergency department (ED) setting.8–10 Providing standardized care based on best evidence, EBCAs have been shown to improve outcomes and standard process metrics within the ED.11–13 Most CPRs are derived and validated for physician use. However, newer strategies include the integration of CPRs into EBCAs for multidisciplinary environments, such as the ED.14–17 Incorporating previously validated CPRs into nursing practice has developed a new generation of EBCAs—advanced nursing directives (ANDs).18,19 Advanced nursing directives serve as screening tools empowering nurses to provide advanced levels of care before physician assessment. Implementation of ANDs in the ED have shown significant improvement in patient satisfaction and efficiency.20–22 Appendicitis is the most common nontraumatic emergency in children.23 In an effort to improve patient care and key department flow metrics for children presenting with right lower quadrant (RLQ) pain, our pediatric ED implemented an AND as well as a comprehensive care map (Appendix 1 and 2, Supplemental Digital Contents 1 and 2: http://links.lww.com/PEC/A82, http://links.lww.com/PEC/A83). Using criteria modified from the Alvarado score,24 our appendicitis AND aims to improve early identification of children with appendicitis and reduce the time to key investigations while limiting the number of unnecessary painful procedures. The objectives of this study were to determine whether children who met the criteria for an AND for RLQ pain had better department flow metrics compared with those who did not meet the criteria and to determine whether a positive AND screen could identify children who required any further investigation/intervention to diagnose and manage appendicitis.
T
he last 2 decades have seen a substantial increase in the development of clinical prediction rules (CPRs) to assist health care professionals in risk stratification, outcome prediction, and
METHODS Study Design
From the *Department of Pediatrics, University of Calgary, Calgary, Alberta, Canada; †Division of Medicine, University College London, London; ‡Department of Primary Care Health Sciences, University of Oxford, Oxford, United Kingdom; Departments of §Medicine, ||Community Health Sciences, and ¶Surgery, University of Calgary; and #Data Integration, Management, and Reporting, Alberta Health Services, Calgary, Alberta, Canada. Disclosure: The authors declare no conflict of interest. Reprints: Graham C. Thompson, MD, FRCPC, Alberta Children's Hospital, 2888 Shaganappi Trail NW, Calgary, Alberta T3B 6A8, Canada (e‐mail:
[email protected]). Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's Web site (www.pec-online.com). Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved. ISSN: 0749-5161
352
www.pec-online.com
This article is a postimplementation cohort study.
Study Setting The Alberta Children's Hospital pediatric ED is the tertiary referral center for Southern Alberta, Eastern British Columbia and Western Saskatchewan, a catchment area of 1.8 million. Care is provided for approximately 70,000 visits yearly.
Study Population Children aged 3 to 17 years presenting to the ED with abdominal pain were potential study candidates. After performing Pediatric Emergency Care • Volume 32, Number 6, June 2016
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
Pediatric Emergency Care • Volume 32, Number 6, June 2016
Advanced Nursing Directive for RLQ Pain Children
Data Abstraction Health records from the ED are a hybrid of hard copy and electronic sources. Data elements that were abstracted from ED charts were entered into a custom electronic database, which was subsequently merged with available electronic health records. Demographics, AND criteria, vital signs, investigations, interventions, ED diagnosis, ED disposition, hospital interventions, pathology results, hospital diagnosis, and hospital disposition were extracted by trained research assistants under the direct supervision of senior staff from the Data Integration Measurement and Reporting of the Alberta Health Services.
Outcomes
FIGURE 1. Study flow through health record screening and analysis.
an initial history and assessment either at triage or on arrival to the bedside, ED staff nurses identified those children who complained of any RLQ pain or demonstrated any RLQ pain on assessment. Those children who were identified as complaining of, or demonstrating any, RLQ pain were subsequently evaluated and managed by ED nursing staff according to the departmental AND for RLQ pain and were therefore eligible for the study. Those children with abdominal pain in whom the AND form was not completed and those with previous appendectomy were excluded.
The main outcomes of interest included (1) the time interval from triage to blood draw and (2) the proportion of patients requiring further investigation/intervention to diagnose and manage appendicitis. Time from triage to blood draw was chosen as a process measure because the results of blood tests, particularly the white blood cell (WBC) and neutrophil count, are key elements in the Alvarado score and Pediatric Appendicitis Score; optimizing time to laboratory results has the potential to improve time to disposition planning. Further investigation/intervention was chosen as a measure of accuracy in identifying the appropriate population and was defined as completion of any of the following: advanced imaging studies (US; computed tomography [CT]), surgical consultation, or admission to hospital. Secondary outcomes included further time intervals (triage to US, admission, and appendectomy) and key outcome measures in appendicitis—the perforation rate (PR) and the negative appendectomy rate (NAR). Perforation was defined as a pathology report with any mention of perforation or abscess formation. Negative appendectomy was defined as a pathology report indicating no significant pathological abnormality, no inflammation, or no evidence of acute appendicitis.
The Care Map and AND
Analysis
The Alberta Children's Hospital appendicitis committee included representation from emergency medicine, emergency nursing, surgery, radiology, infectious disease, anesthesia, and pharmacy services. Based on a critical assessment of the literature and expert opinion, a comprehensive ED management plan including our AND and care map (Appendix 1, 2) was developed. Assessment components of the AND were modified from the historical and physical examination elements of the Alvarado score.23 Modifications included an increase in reported or measured temperature to 38°C and removal of weighted scoring for individual criteria. The laboratory elements of the Alvarado score were not included because AND assessment occurs at triage or initial bedside assessment (ie, before obtaining laboratory investigations). Intervention components of the AND include initiation of intravascular access, obtaining blood and urine for investigation, and initiating a normal saline bolus (to assist bladder filling for anticipated ultrasound [US]). Implementation strategies included a series of training sessions for physicians and nurses to ensure familiarity with pathway processes, including formal “lunch-and-learn” sessions, division meetings, and rounds. After a 3-month adaptation period, we initiated a 6-month pathway evaluation process. Health records for all children presenting to the ED with abdominal pain were reviewed to identify those that had been assessed by the nursing staff using the AND. To reduce the incidence of missed cases, health records of all children who underwent appendectomy during the evaluation period were reviewed to identify those that had been assessed using the AND.
Demographic data were evaluated using descriptive statistics with t tests for continuous data and χ2 for categorical data. Odds ratios with 95% confidence intervals were used to describe the difference in investigations and interventions performed. Wilcoxon rank sum defined differences between median times for departmental flow metrics. Overall test characteristics for the AND in identifying children requiring further evaluation were calculated including sensitivity, specificity, positive and negative predictive values, and accuracy. Data were analyzed using STATA 12.1 (StataCorp, College Station, Tex) TABLE 1. Demographics of Children Managed by an AND for RLQ Pain AND Criteria AND Criteria Met (n = 137) Not Met (n = 73) Age, mean (SD), y 11.5 (3.8) Male, n (%) 49 (35.8) CTAS, n (%) 2 29 (21.2) 3 106 (77.4) 4 2 (1.5) Transfer from another 23 (16.8) facility, n (%)
11.9 (4.2) 35 (48.0) 4 (5.5) 62 (84.9) 7 (9.6) 6 (8.2)
P 0.7842 0.086 0.001
0.087
CTAS, Canadian Triage Acuity Score.
© 2015 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
www.pec-online.com
353
Pediatric Emergency Care • Volume 32, Number 6, June 2016
Thompson et al
TABLE 2. Investigations, Management, and Outcomes of Children Managed by an AND for RLQ Pain AND Criteria Met (n = 137) WBC count, n (%) Ultrasound, n (%) CT, n (%) Admission for appendectomy, n (%) Any further intervention, n (%) Appendectomy, n (%) NAR PR
113 (82.5) 58 (42.34) 2 (1.5) 58 (42.3) 92 (67.2) 58 (42.3) 5 (8.6) 17 (29.3)
AND Criteria Not Met (n = 73) 43 (58.9) 26 (35.62) 2 (2.7) 22 (30.1) 32 (43.8) 22 (30.1) 2 (9.1) 1 (4.5)
OR (95% CI) 3.28 (1.65–6.56) 1.33 (0.71–2.50) 0.53 (0.38–7.42) 1.70 (0.90–3.28) 2.62 (1.40–4.90) 0.94 (0.14–10.67) 8.17 (1.17–380.86)
Bold values indicate statistically significant results.
Cases where flow metric analysis indicated an event occurred before the index triage date and time (ie, blood test performed in external physician office, urgent care) were excluded from the flow metric analysis.
Ethics Protocol review from members of the University of Calgary Conjoint Health Research Ethics Board provided approval of this study as a quality assessment and improvement initiative.
RESULTS A total of 518 health records were reviewed. An AND for RLQ pain was initiated in 240 children during the 6-month quality improvement assessment period. Thirty forms were present on the health record but were not completed and therefore were excluded. Data from the 210 remaining health records were analyzed (Fig. 1). Demographics of the patient population are reported in Table 1. Children who met AND criteria had blood tests done more often (odds ratio [OR], 3.28; 95% CI, 1.65–6.56) and were more likely to undergo further investigation to rule out appendicitis (OR, 2.62; 95% CI, 1.40–4.90) than those who did not meet the criteria (Table 2). Including all children in each cohort, the number of negative appendectomies within each group was not significantly different (5/137 [3.6%] vs 2/73 [2.7%]); for those who underwent appendectomy, the NAR was 8.6% (5/58) for children meeting the criteria and 9.1% (2/22) for those not meeting the criteria (OR, 0.94; 95% CI, 0.14–10.67). Including all children in each cohort, the number of perforations was significantly higher in those meeting the criteria (17/137 [12.4%] vs 1/73 [1.4%]); for those who underwent appendectomy, the PR was 29.3% (17/58) and 4.5% (1/22), respectively (OR, 8.17; 95% CI, 1.17–380.86). Table 3 outlines the differences in department flow metrics between children who met AND criteria and those who did not,
with significant differences in time from triage to blood draw, admission, and appendectomy. For children meeting the AND criteria, changes in primary outcome measures over the duration of the quality improvement initiative are outlined in Table 4. Table 5 reports the test characteristics of the AND in predicting the need for further investigation in the ED.
DISCUSSION The results of our quality improvement initiative demonstrate that children meeting the criteria for our RLQ pain AND have significantly better hospital flow metrics (triage to blood draw, admission, and appendectomy) than those not meeting AND criteria. In addition, we have shown that children who met AND criteria underwent further advanced investigations and interventions for suspected appendicitis significantly more often than children who did not meet AND criteria. To our knowledge, this is the first study demonstrating the use of an AND for children presenting to the ED with abdominal pain and its impact on patient care. The WBC count and neutrophil count, biomarkers suggestive of infection, are required to calculate clinical scores including the Alvarado score24 and the Pediatric Appendicitis Score25 and are often one of the earliest steps in the evaluation of a child at risk for appendicitis. In our study population, time to blood sample acquisition in children who met AND criteria was less than half that of those who did not meet the criteria, a reduction of 1.5 hours. Assuming that earlier sample acquisition translates to earlier results, implementation of the AND may lead to much earlier calculation of the Alvarado score or Pediatric Appendicitis Score, thus optimizing time to clinical decisions regarding the need for further evaluation through imaging and surgical consultation. The balancing measure for improved flow attributed to the AND is the ability of the screening tool to identify which children actually go on to further investigations/interventions. Improving ED flow at the cost of increased unnecessary painful procedures
TABLE 3. Department Flow Metrics of Children Managed by an AND for RLQ Pain AND Criteria Met Triage to blood draw Triage to ultrasonography Triage to hospital admission Triage to appendectomy
n
Time
113 58 56 56
74 231 271 498
AND Criteria Not Met
25-75 Percentiles
n
Time
25-75 Percentiles
38–133 154–372 190–389 341–661
43 26 22 22
162 228 395 602
89–226 156–400 249–587 482–931
P*
