Facebook Pixel Code

Information for health-care providers

Implementing our Choosing Wisely recommendations provided an opportunities for meaningful discussion, education, and quality initiatives and has generated measurable improvements. Outside of SickKids, our recommendations have been disseminated nationally and internationally to help educate the greater paediatric provider population through presentations and publications, and being featured by Canadian Paediatric Society, Pediatric Academic Societies and Choosing Wisely Canada. Our methods for implementation have also been shared with other paediatric institutions to help them conceptualize and implement recommendations of their own.

Although the implementation strategies for our Choosing Wisely recommendations varied, we have identified a number of enablers for success. These include:

  • assigning a physician champion with multidisciplinary team involvement;
  • clear support of hospital leaders;
  • implementing educational initiatives to support the recommendations;
  • availability of data to measure impact;
  • ability to implement systematic changes to the process of ordering tests.

It is essential to ensure that high-quality outcomes derived through Choosing Wisely are achieved without any unintended consequences. Throughout implementation, we measure potential negative outcomes resulting from our recommendations and no significant changes have been noted to date.

Learn more about some of our ongoing implementation activities by scrolling down the page or clicking the icons below:



Recommendation: Don't routinely order nasopharyngeal (NP) testing for typical respiratory viruses unless results are likely to impact management. 

Respiratory viral infections frequently occur in children and are a common reason to seek medical care. The diagnosis is made clinically and usually does not require confirmatory testing. Nasopharyngeal (NP) testing is uncomfortable for children and the results frequently do not impact their medical management. Therefore, NP testing should only be considered in high-risk patients where results will influence treatment decisions such as the need for antibiotics, performance of additional tests, or hospitalization. Reducing routine respiratory viral testing promotes high-value care and allows for more effective allocation of health care resources.​

Background: In 2014, almost 6,000 NP swabs were completed on children at SickKids. Of the 2,600 NP swabs ordered from the Emergency Department (ED) alone, approximately 63% of these swabs were completed on children who were discharged home and these results were generally never followed up by a provider or relayed to the family.

Implementation Plan: A multi-divisional expert panel reviewed published guidelines and formulated the Choosing Wisely Respiratory Virus Pathway for the ED and General Paediatrics.  In January 2016, the Pathway was launched using a multi-faceted intervention strategy including force functions in our health information system, an educational campaign, and audit and feedback. By reducing unnecessary NP swab testing, resources were allocated towards two new tests that promote high-quality care. The first was a Point of Care Rapid Influenza test that provides immediate results for children presenting with flu-like symptoms where oseltamivir treatment may be indicated or where an immediate positive influenza result will impact further management decisions like antibiotics, imaging, or even admission. The second test was a new and improved Respiratory Virus Multiplex PCR test that can detect 16 respiratory viruses in children who will benefit from the test, like those admitted to our ICUs.

Results: Since 2016, the hospital has reduced the use of nasopharyngeal (NP) swab tests for typical respiratory viruses.  Significant gains have been made in the ED with a reduction in NP swab testing of over 80%. With the introduction of the new rapid influenza test in the ED, viral respiratory testing has still decreased greater than 50% and this reduction has been sustained.  These gains have spread to the General Paediatrics units where NP testing has decreased by 35%.

To learn more, read "Less is more: new review suggests paediatricians should think twice before routinely testing for respiratory viruses"

Return to top

Recommendation: Don’t routinely perform a voiding cystourethrogram (VCUG) in infants after a first febrile urinary tract infection.  

Previous guidelines recommended routine VCUG after first febrile urinary tract infections (UTI) in young children. However, a more recent body of research has not supported this practice. In addition VCUGs are uncomfortable, and expose children to ionizing radiation. Thus, most recent guidelines suggest that VCUG should be considered after febrile UTI only in select circumstances: for example, when the renal ultrasound is abnormal suggestive of higher grade reflux, or scarring; in atypical circumstances; or with recurrent UTI.

Background: From January 2014 to November 2015, an audit examined how frequently infants and children admitted to the General Pediatrics Inpatient Unit at SickKids for a UTI received a VCUG within three months of admission.  There were 149 admissions for UTI to the General Pediatrics Inpatient Unit during this period.  Within this patient population, the audit found that 29/149 (19.4%) received a VCUG. Only 2/29 (6.8%) VCUGs performed were not in accordance with guideline recommendations. This represents a 1.3% rate of inappropriate VCUG performance for all UTI admissions to the General Pediatric Inpatient Unit. For the 27 infants and children who had a VCUG appropriately performed, reasons included recurrent UTI, abnormal renal ultrasound, atypical/severe infection (sepsis, meningitis), genitourinary abnormality, and neurogenic bladder.​

Implementation Plan: Given the very low rate of inappropriate VCUG performance SickKids Choosing Wisely did not embark on a major change plan. As part of the Choosing Wisely campaign, the team has used education directed at clinicians to sustain current performance. 

Return to top

Recommendation: Don’t use continuous pulse oximetry routinely in children hospitalized with acute respiratory illness unless they are on supplemental oxygen. 

When children are in a stable phase of their acute respiratory hospitalization and do not require supplemental oxygen, observational research suggests that the use of continuous oxygen saturation monitoring leads to over-diagnosis and overtreatment of hypoxemia resulting in longer hospital stay.

Background:  Random audits on the Paediatric Medicine Inpatient Unit began in February 2015 to examine what proportion of patients with bronchiolitis, asthma, and pneumonia who are not on supplemental oxygen are on continuous monitoring.  These audits occurred on specific days after morning medical patient rounds.  ​In 2015, when the initiative was first launched, it was found that for the month of February, 10/18 (56%) children and for the month of March 13/17 (76%) children audited who were not on supplemental oxygen were on continuous oxygen saturation monitoring.  This audit was repeated close to two years later, and in October of 2017, it was found that 16/22 (72%) of children audited who were not on supplemental oxygen were on continuous oxygen saturation monitoring. This demonstrated persistent use of continuous oxygen saturation monitoring despite education efforts between audits.

Implementation Plan: Following the 2017 audit, a number of implementation strategies were developed including the development of a comprehensive education and awareness package targeted at medical (staff, fellows, residents), nursing and respiratory therapy staff; point-of-care initiatives in the form of visual cues such as sticker, notices, logos, posters, reminders placed directly on each communication board within patient rooms, as well at communal areas including workstations, pager chargers etc.; and audit and feedback occurring on the inpatient unit monthly.

Results: Prior to the initiative, the General Paediatrics wards were monitoring appropriately 60% of the time; after the first four weeks of interventions this increased to 78%.

Return to top

Recommendation: Don't automatically give IVIG as first-line treatment for children with newly diagnosed, typical ITP 

Management choices for children with newly diagnosed, typical immune thrombocytopenia (ITP) include observation (when the bleeding is mild), prednisone, or intravenous immunoglobulin (IVIG). Each option has risks and benefits; ideally these can be discussed with families and their preferences accounted for. There is no evidence of a relationship between any of these initial therapies and the subsequent development of severe bleeding. Prior to choosing IVIG, consideration should be given to its expense, its requirement for a day-hospital or overnight admission, and its side effect profile that frequently includes aseptic meningitis.

Background:​ From 2007 to 2009, a quality improvement (QI) team developed a cross-sectional study of management of newly diagnosed, typical ITP at SickKids to assess the IVIG treatment rate.

Implementation Plan​​: In 2013, the working group implemented a QI bundle to achieve their aim. The bundle included developing a patient information sheet; developing an evidence-informed, consensus-based protocol; and promoting shared decision-making via stakeholder engagement and education.  Strategies include regular communication with stakeholders (Emergency Medicine, Haematology, Paediatric Medicine); electronic availability of the protocol; and accessibility of the Paediatric Medicine day hospital to care for this population as an alternative to inpatient admission. Since 2017, refinements to the protocol have resulted in an increased number of patients being cared for via an outpatient pathway.

Results: The rate of IVIG use has decreased from 88% to 20% by not automatically giving IVIG as first-line treatment for children with newly diagnosed, typical ITP and sustained to date.

Resources: CPS Practice Point

To learn more, read "SickKids study leads to new practice point for assessing management options for newly diagnosed, typical immune thrombocytopenia"

Return to top

Recommendation: ​​​Don't use routine radiography in children who present with acute ankle injuries and meet criteria for a low-risk examination. 

In North America, approximately 2 million children present to emergency departments annually with ankle injuries; about 12% demonstrate fractures on plain films. A paediatric clinical decision rule (Low-Risk Ankle Rule) has been validated and demonstrated a safe reduction in unnecessary radiographs by up to 60%. Implementing this rule reduces unnecessary radiation exposure and saves health care resources.

Background: Baseline audits completed in the ED found that 90% of children between the ages of three to 16 years of age received radiographs for acute ankle injuries.

Implementation Plan: Following the audit, a number of implementation strategies were developed to enforce use of the Low-Risk Ankle Rule including education dissemination, visual cues such as posters, modified diagnostic imaging requisition and a clinical decision rule in our health information system.

Results: Since the initial audit routine radiography for children with low-risk acute ankle injuries has decrease from 30% to 53% in the ED.

To learn more, read "Growth plate fractures in children’s ankle injuries are much rarer than previously thought"

Return to top

Recommendation: ​​​Don't routinely continue antibiotics for surgical site infection prevention after the patient has left the operating room. 

Surgical antibiotic prophylaxis (SAP) given immediately prior to surgery has been shown to significantly decrease the risk of surgical site infection (SSI). However, prolonged use has not demonstrated benefit over the single dose of antibiotics preoperatively. As such, prolonged SAP results in unnecessary antibiotic exposure which may contribute to the development of resistant organisms.  This does not apply to situations where there is infection that requires antibiotic treatment.  ​

Background: In 2016, the World Health Organization (WHO) published comprehensive recommendations on strategies to prevent SSIs. As part of this, a systematic review was conducted comparing single dose SAP (no postoperative dose) with any prolonged postoperative prophylaxis. There was moderate quality of evidence from a high number of randomized control trials (44 studies) that prolonged SAP postoperatively had no benefit in reducing SSI when compared to a single preoperative dose. As such, the WHO recommends against any SAP administration after completion of the operation for the purpose of preventing SSI.  

An audit of postoperative antibiotic use was conducted in 2015 to assess compliance with our current SickKids guidelines, which are consistent with the WHO recommendations. Overall adherence to the SickKids guidelines was 70%, with the highest adherence being 100% and the lowest 0% depending on the surgical division. Some patients or prescribers continued antibiotics because of the presence of indwelling drains/catheters, a practice that is currently not recommended.

Implementation Plan: In 2019, SickKids targeted SAP after surgery by including the measurement of appropriateness on the hospital’s Key Performance Indicators. Appropriate use improved after meetings with specific surgeons and surgical groups; increasing awareness of existing guidelines with Choosing Wisely initiatives; and modification of guidelines to align with evidence-informed practices, consensus guidelines, and protocols.  Monthly audits of approximately 10% of surgical cases from seven surgical services were reviewed to assess adherence to SickKids formulary guidelines. Through this audit and feedback strategy as well as education, stakeholder engagement and order-set revisions in Epic, adherence has improved.

Results: The average appropriateness increased from the baseline of <75% to a year to date average of 82% with several months exceeding our 85% target. The ongoing focus on surgical antibiotic prophylaxis has sustained with the help of Choosing Wisely.

Return to top

Recommendation: ​​​Don't routinely order a CT abdomen/pelvis for paediatric trauma patients deemed low risk by established decision rules for clinically significant intra-abdominal injury. 

CT imaging among paediatric trauma patients is often overused leading to increased risk from ionizing radiation, increased need for procedural sedation, and increased cost to the health care system. For paediatric trauma, abdominal/pelvic CT imaging is commonly ordered because of a concerning mechanism of injury, which has not on its own been demonstrated to correlate with intra-abdominal injury. Rates of CT abdominal/pelvic imaging can be safely reduced by utilizing evidence-based clinical decision rules, of which two exist to guide diagnostic imaging in this patient population. 

These two distinct clinical decision rules, developed through two separate clinical research networks (PECARN - Pediatric Emergency Care Applied Research Network, and PedSRC - Pediatric Surgery Research Collaborative​), have both demonstrated abdominal/pelvic CT scan rates of 45-46% for all pediatric trauma activation patients of which 17-25% of children fall into a very low risk criteria, where CT imaging was not indicated. 

Background: At SickKids, a team has identified the problem of overuse of CT imaging as a screen for abdominal, head and cervical spine injuries, particularly in low-risk patients through concerns brought forth from informal observations, email concerns, chart reviews and M&M reports. At our institution many CT scans are obtained on the basis of mechanism of injury alone, which has not held up as a valid predictor of clinically important injury in pediatric trauma.

A preliminary audit of paediatric trauma registry patients was performed to determine appropriateness of CT imaging for head, spine, chest, and abdomen/pelvis over a six year period (2010-2016).   On average, 41.5% of paediatric trauma patients received abdominal/pelvic CT scans, of which 27.9% had an injury severity score (ISS) ≤​8, suggesting many of these children may not have merited abdominal/pelvic imaging.  In 2016-2017 rates went up significantly with 55.1% of children having an abdominal/pelvic imaging and 32.0% had an ISS ≤8.  The team will perform a retrospective chart review to fully understand the indications for imaging and the rates of positive studies.

Implementation Plan: Reduction in unnecessary CT imaging rates were achieved through implementation of a standardized evidence-based clinical guideline within the Hospital, which was developed and disseminated amongst all trauma providers.  Additionally, modifications to the diagnostic imaging requisitions specifically for trauma, forcing ordering clinicians to choose an indication were developed.  Further, Choosing Wisely educational outreach sessions specifically for trauma providers were executed, with close collaboration amongst Diagnostic Imaging, General Surgery and Emergency Medicine.  Rates of abdominal/pelvic imaging are collected prospectively with chart reviews done to determine if and when indications are followed. 

Results: To date, mean abdominal/pelvic imaging of very low risk paediatric trauma patients has dropped from 26.8% to 8.6%, with a statistically and clinically significant absolute reduction of 18.2%.

Return to top

Recommendation: Don’t routinely send blood cultures in well-appearing children who are at low risk for bacteremia.   

Fever in children is a very common reason to seek medical attention and blood cultures are often drawn with bloodwork irrespective of the patient’s clinical presentation. Current rates of bacteremia in healthy vaccinated children are extremely low and contamination is common resulting in repeat visits, additional testing, and even hospital admissions. Blood cultures should only be sent when there is a clinical suspicion for bacteremia.

Background: The ED is the highest user of blood cultures in the hospital with 8% of all children visiting the ED getting a blood culture. Practice patterns often promote overuse of ordering blood cultures in efforts to prevent a child from needing a second poke or to try to reduce diagnostic uncertainty. However, the true positive rate is 4.3% with a contamination rate of 2% resulting in repeat ED visits and even hospital admissions due to contaminated blood cultures.  Importantly, blood cultures are commonly drawn for certain focal infections such as pneumonia, skin and soft tissue infections (SSTIs), asthma, bronchiolitis as well as UTIs with evidence to show that the rates of positive blood cultures are low and do not change clinical management and outcomes.   

Implementation Plan: A number of activities have been implemented to increase understanding of when blood cultures should be sent.  This includes audit of blood cultures to better understand which patients are getting blood cultures; education to target awareness of known focal conditions that do not benefit from blood cultures; requiring a physician order for blood cultures to be processed; removing blood cultures from the Epic Quick List of investigations so that ordering requires a conscious decision. Future work will be on physician audit and feedback as well as developing a clinical decision rule for low risk children with fever who do not need blood cultures.

Results: A retrospective chart review showed that 65% of blood cultures were drawn in low-risk patients, with over half of these patients being discharged home. Only 0.4% of these patients had positive blood cultures, none of which altered management. The majority of blood cultures drawn  in low-risk patients (30%) were in patients with a diagnosis of fever; 10% were drawn in patients with a diagnosis of pneumonia where current practice guidelines do not support the use of routine blood cultures.  

Return to top

Recommendation: Don’t routinely continue broad spectrum antibiotics, such as vancomycin and meropenem after 48 hour cultures return, unless there is a clear clinical or microbiological indication. 

Antibiotics are usually started when there is concern for possible infection. After two days, if culture results do not demonstrate a resistant organism, antibiotics should either be changed to a narrow-spectrum agent or discontinued if there is no longer evidence of infection. Minimizing patient exposure to broad spectrum antibiotics will help to decrease development of resistance to these agents and preserve their efficacy for future use.

Background: Targeted drug utilization evaluations of vancomycin and meropenem have shown that duration of treatment is a key factor contributing to inappropriate use. Sixty-three per cent of empiric courses of vancomycin were used for more than three days. At baseline, only 9% were found to be appropriate for definitive treatment of a proven pathogen. Additionally, 69% of patients started on meropenem empirically continued on treatment for an average of nine days, without documented evidence of infection requiring meropenem.

Implementation Plan: Starting in June 2017, prolonged treatment (>72 hours) with vancomycin and meropenem became actively restricted by the Antimicrobial Stewardship Program (ASP) and the Infectious Disease (ID) service. This was done in accordance with the SickKids Restricted Drugs Policy. When either of these medications is ordered, the default stop date is set at 72 hours from the initial dose. Approval must be obtained to continue the medication beyond 72 hours. Each new order automatically generates a review by either the ASP or ID teams at 48 hours into therapy to review appropriateness and recommending continuing or changing therapy.  The team tracks the percentage of patients that receive > 72 hours of therapy with vancomycin or meropenem, and addresses individual units and services as needed.

Results: Comparing the six months prior to implementation of the stop date, to six months after implementation, the number of courses of meropenem that were >72 hours decreased from 65% to 55%. Prolonged courses of vancomycin decreased from 41% to 35%.

Return to top

Recommendation: ​​​Don't empirically start antibiotics for children over three months of age with low risk of urinary tract infection (UTI) without evidence of nitrites or significant pyuria on urine dipstick.  Do stop antibiotics if the urine culture is negative. 

Urinary tract infections (UTIs) are a common infection in children and a leading cause for acute care visits in pediatrics. The diagnosis is often made on the basis of clinical symptoms, pyuria on dipstick analysis and confirmed by a positive urine culture. Since urine culture results are not immediately available, clinicians often empirically prescribe antibiotics to patients for suspected UTIs. However, since UTI symptoms are often nonspecific and urinalysis has varying sensitivity and specificity, children over three months of age that are low risk should not receive empiric antibiotics without evidence of nitrites or significant pyuria on urine dipstick. Empiric antibiotics should be discontinued if final urine culture results are negative.​

Background: During a three month period in 2016 at SickKids, 184 low-risk patients were diagnosed with a UTI in the ED.  Approximately 47% of these patients received antibiotics despite negative urine cultures and none of these patients received notification to stop.  This practice led to 652 unnecessary antibiotic days for these patients.

Implementation Plan: To promote high-value care and reduce unnecessary exposure to antibiotics, a multifaceted implementation strategy was developed involving key stakeholders in Emergency Medicine, Infectious Disease, Microbiology and Urology. In order to improve the accuracy of an empiric UTI diagnosis, a guided, evidence based algorithm was created based on urinalysis results to determine whether an empiric antibiotic treatment is recommended while pending urine culture results.  Additionally, a standard antibiotic discontinuation protocol for patients with negative urine cultures is being implemented in the ED. A strategy to standardize antibiotic prescription duration will also be instituted along with a targeted educational campaign to promote the initiative throughout the hospital.

Results: The introduction of the Choosing Wisely UTI algorithm has resulted in a 40% decrease in inappropriate UTI diagnosis and 770 fewer unnecessary antibiotic days.

Return to top