Low-Value Diagnostic Imaging Use in the Pediatric Emergency Department in the United States and Canada

Key Points

Question  Are there differences in the use of diagnostic imaging between pediatric emergency departments located in the United States and Canada?

Findings  In this cohort study, overall and low-value diagnostic imaging use rates were lower in pediatric emergency departments in Ontario, Canada, than those in the United States. No differences in post–emergency department adverse outcomes were observed.

Meaning  There may be opportunities to safely reduce imaging rates in pediatric emergency departments in the United States.

Importance  Diagnostic imaging overuse in children evaluated in emergency departments (EDs) is a potential target for reducing low-value care. Variation in practice patterns across Canada and the United States stemming from organization of care, payment structures, and medicolegal environments may lead to differences in imaging overuse between countries.

Objective  To compare overall and low-value use of diagnostic imaging across pediatric ED visits in Ontario, Canada, and the United States.

Design, Setting, and Participants  This study used administrative health databases from 4 pediatric EDs in Ontario and 26 in the United States in calendar years 2006 through 2016. Individuals 18 years and younger who were discharged from the ED, including after visits for diagnoses in which imaging is not routinely recommended (eg, asthma, bronchiolitis, abdominal pain, constipation, concussion, febrile convulsion, seizure, and headache) were included. Data analysis occurred from April 2018 to October 2018.

Exposures  Diagnostic imaging use.

Main Outcome and Measures  Overall and condition-specific low-value imaging use. Three-day and 7-day rates of hospital admission and those admissions resulting in intensive care, surgery, or in-hospital mortality were assessed as balancing measures.

Results  A total of 1 783 752 visits in Ontario and 21 807 332 visits in the United States were analyzed. Compared with visits in the United States, those in Canada had lower overall use of head computed tomography (Canada, 22 942 [1.3%] vs the United States, 753 270 [3.5%]; P < .001), abdomen computed tomography (5626 [0.3%] vs 211 018 [1.0%]; P < .001), chest radiographic imaging (208 843 [11.7%] vs 3 408 540 [15.6%]; P < .001), and abdominal radiographic imaging (77 147 [4.3%] vs 3 607 141 [16.5%]; P < .001). Low-value imaging use was lower in Canada than the United States for multiple indications, including abdominal radiographic images for constipation (absolute difference, 23.7% [95% CI, 23.2%-24.3%]) and abdominal pain (20.6% [95% CI, 20.3%-21.0%]) and head computed tomographic scans for concussion (22.9% [95% CI, 22.3%-23.4%]). Abdominal computed tomographic use for constipation and abdominal pain, although low overall, were approximately 10-fold higher in the United States (0.1% [95% CI, 0.1%-0.2%] vs 1.2% [95% CI, 1.2%-1.2%]) and abdominal pain (0.8% [95% CI, 0.7%-0.9%] vs 7.0% [95% CI, 6.9%-7.1%]). Rates of 3-day and 7-day post-ED adverse outcomes were similar.

Conclusions and Relevance  Low-value imaging rates were lower in pediatric EDs in Ontario compared with the United States, particularly those involving ionizing radiation. Lower use of imaging in Canada was not associated with higher rates of adverse outcomes, suggesting that usage may be safely reduced in the United States.

Minimizing care that provides little benefit (ie, low-value care) has become an important focus for decreasing health care costs and improving the quality of care delivery.¹ A key target for low-value care is emergency department (ED) diagnostic imaging overuse. Choosing Wisely, a campaign targeting overuse, was launched in 2012 and has spread internationally.² Diagnostic imaging overuse in children who present to the ED accounts for 3 of the first 5 of the 2013 American Academy of Pediatrics Choosing Wisely Recommendations³ and 6 of 10 Choosing Wisely Recommendations in emergency medicine in Canada.⁴

Canadian and US populations have similar rates of ED use,⁵ but the systems differ in overall organization of care and payment structures.⁶ Canada provides government-funded universal access to acute care services, while residents in the United States receive care funded through a number of sources, including public coverage, private insurance, and/or out-of-pocket spending. There is an increased perception of medicolegal risk in the United States that could lead to overreliance on potentially low-yield diagnostic imaging.^7,8 Consequently, varying practice patterns may account for differences in health care expenditures and diagnostic imaging use. It is unknown whether campaigns such as Choosing Wisely have affected the rates of low-value diagnostic imaging in pediatric EDs.

The aim of this study was to compare the use of diagnostic imaging in children evaluated in pediatric EDs in the United States and Ontario, Canada’s most populous province, and explore whether observed differences are associated with variation in outcomes. We hypothesized that rates of use would be higher in the United States for a subset of diagnoses in which diagnostic imaging is not routinely recommended without improved outcomes, thus reflecting potential overuse.

This was a retrospective study of ED visits to children’s hospitals for individuals 18 years and younger from calendar years 2006 through 2016, using administrative health databases in Ontario, Canada, and the United States. We excluded patients who died in the ED, those transferred into the ED from another hospital, and those transferred to another facility for ongoing care. Ethics review for this study was obtained from the institutional review board of the Hospital for Sick Children. All data were deidentified in all sources, and consent was therefore waived. Diagnostic imaging and diagnostic codes from which data were collected are described in eTables 1 and 2 of the Supplement.

Data from visits to any of Ontario’s 4 pediatric EDs were abstracted from population-based databases at ICES (formerly known as the Institute for Clinical Evaluative Sciences), an independent, nonprofit research institute whose legal status under Ontario’s health information privacy law allows it to collect and analyze health care and demographic data, without patient consent, for health system evaluation and improvement. These data sets were linked using an unique encoded patient-level identifiers and analyzed at ICES.⁹ Ontario is Canada’s most populous province, with about 14 million residents (approximately 40% of Canada’s population). Databases used included the National Ambulatory Care Reporting System, containing information on discharge diagnosis (International Classification of Diseases, Tenth Revision [ICD-10] codes), and admission-discharge-transfer systems for every ED visit in the province.¹⁰ The Ontario Health Insurance Plan database contains billing data for all publicly funded fee-for-service physician services in Ontario, including diagnostic imaging. The Canadian Institute for Health Information Discharge Abstract Database was used to evaluate clinical information for hospitalized patients. The Ontario Registered Persons Database was used to identify postal code of residence for included individuals.

Data from the United States were abstracted from the Pediatric Health Information System database (from the Children’s Hospital Association). The Pediatric Health Information System database contains clinical and billing data from 51 not-for-profit, tertiary care children’s hospitals in the United States. Participating hospitals are located in 27 states plus the District of Columbia and account for approximately 13% of all pediatric ED visits in the United States (independent analysis conducted by the Children’s Hospital Association). The data collection, validation, and safeguarding procedures are assured through a joint effort between the Children’s Hospital Association and participating hospitals.¹¹ In this study, we included data from the 26 hospitals with complete demographic and billing information during the study period for all ED encounters.

We abstracted the following characteristics of the ED visit: patient age, sex, neighborhood income quintile (derived in the data from Canada using postal code linkage at the level of dissemination area [400-700 residents] by applying 2011 census data and in data from the United States by zip code, linked to zip code tabulation areas from the 2010 census data), straight-line distance from the centroid of the child’s residential postal or zip code to the hospital’s postal or zip code, arrival day of week, and arrival time. Patient complexity was assessed using International Classification of Diseases, Ninth Revision or ICD-10, mapping to complex chronic conditions (CCCs).¹² The presence of a CCC was ascertained using diagnoses associated with the ED visit or present during a prior ED visit or hospitalization to the same institution within the preceding 365-day period.

Imaging data were captured for the diagnostic modalities of chest radiographic imaging, abdominal radiographic imaging, abdominal ultrasonographic imaging, abdominal computed tomography (CT), head CT, and head magnetic resonance imaging (MRI). Rates of use in the ED were collected for all users of the ED in both Ontario and the United States.

To assess the main study objective, potential low-value imaging use, we assessed imaging use in individuals who were discharged from the ED (ie, not admitted to the hospital). We limited the population to those individuals with visits resulting in discharge to focus on a lower-acuity population for whom routine use of imaging is not recommended.¹³ The conditions assessed were determined by 11 study authors (E.C., S.B.F., P.L.A., H.K.S., J.R.M., M.S.-K., E.R.A., R.B.M., S.S.S., A.P., and M.I.N.) by group consensus and included the following primary diagnoses and their associated imaging modalities: asthma (chest radiographic imaging),^14,15 bronchiolitis (chest radiographic imaging),^15,16 abdominal pain (abdominal radiographic imaging, abdominal ultrasonographic imaging, and abdominal CT),¹⁷ constipation (abdominal radiographic imaging, abdominal ultrasonographic imaging, and abdominal CT),^18-20 concussion (head CT and head MRI),^21,22 simple febrile convulsion (head CT),²³ complex febrile convulsion (head CT),²⁴ seizure (head CT and head MRI),^25,26 and headache (head CT and head MRI).²⁷

As balancing measures, we assessed whether patients who were discharged without imaging had deleterious outcomes by examining the rates of hospital admission within the 3-day and 7-day periods after the ED visit, and those admissions resulting in intensive care unit care hospitalization, surgery, or mortality. These outcomes were also evaluated in the full cohort of children, including children with imaging and children without imaging, to allow comparisons that incorporate possible differences in severity of illness or other diagnostic or therapeutic interventions between Ontario and the United States.

To assess the accuracy of imaging use capture across the 2 data sets, we measured imaging use for conditions in which diagnostic imaging is routinely indicated. These quality-control measures included hospitalized children with pneumonia (chest radiographic imaging), bowel obstruction (abdominal radiographic imaging), intussusception (abdominal ultrasonographic imaging), pancreatic or liver laceration (abdominal CT), ventricular shunt malfunction (head CT), and acute stroke (head MRI).

Data were presented as proportions with 95% CIs. Comparisons between Canada and the United States were expressed as absolute differences (with 95% CIs calculated using standard formulas for differences in proportions) for the main analysis (an ED discharge diagnosis in which routine use of imaging is not recommended), for the 3-day and 7-day outcome balancing measures (hospitalization, a need for intensive care unit care, a need for surgery, or in-hospital mortality), and imaging rates for the quality-control measures. Since children with underlying comorbidities may be more likely to receive diagnostic imaging, we performed a sensitivity analysis limited to children without CCCs. Yearly proportions of imaging use were calculated to examine trends across the study period. Analyses were performed with SAS version 9.4 (SAS Institute, Inc), and P values less than .001 were considered statistically significant because of the large sample sizes.

During the 10-year study period, there were 1 875 608 ED visits among children 18 years and younger to pediatric EDs in Ontario and 22 256 888 visits to the study pediatric EDs in the United States (Figure 1). After exclusions, 1 783 752 ED visits (95.1%) from Canada and 21 807 332 ED visits (98.0%) from the United States were included. In Ontario, the mean [SD] age was 6.1 [5.5] years, and 821 984 patients (46.1%) were female. In the United States, the mean [SD] age was 5.6 [5.3] years, and 10 198 965 patients (46.8%) were female.

Compared with those in the United States, children presenting to pediatric EDs in Canada lived closer to the children’s hospital than those in the United States (Canada, 47 432 of 1 783 752 [2.7%]; the United States, 1 792 720 of 21 807 332 [4.2%] more than 80 km from the hospital; P < .001), resided in wealthier neighborhoods (340 155 [19.1%] vs 3 252 381 [14.9%] in the highest quintile; P < .001), were less likely to have a CCC (101 608 [5.7%] vs 1 792 720 [8.2%]; P < .001), and were less likely to be admitted to the hospital (135 189 [7.6%] vs 2 296 996 [10.5%]; P < .001; Table 1). No material differences were observed between Canadian and US ED visits with respect to patient age, sex, arrival day of the week, or arrival time.

Compared with children presenting to pediatric EDs in the United States, those in Ontario had lower overall use of head CT (Canada, 22 942 [1.3%] vs the United States, 753 270 [3.5%]; P < .001), abdomen CT (5626 [0.3%] vs 211 018 [1.0%]; P < .001), chest radiography (208 843 [11.7%] vs 3 408 540 [15.6%]; P < .001), and abdominal radiography (77 147 [4.3%] vs 3 607 141 [16.5%]; P < .001). Use of abdominal ultrasonographic imaging was higher in Canada compared with the United States (69 750 [3.9%] vs 624 518 [2.9%]; P < .001), while MRI head use was 0.5% of ED visits in both countries (9224 in Canada and 105 515 in the United States).

Imaging use for patients who were discharged from the ED with a diagnosis in which routine use of imaging is not recommended across all years is presented in Figure 2. For chest radiography, compared with pediatric EDs in the United States, use was lower in Ontario for both bronchiolitis (absolute difference, 6.8% [95% CI, 6.2%-7.4%]) and asthma (absolute difference, 0.7% [95% CI, 0.3%-1.1%]). For abdominal radiography, use was lower in pediatric EDs in Ontario compared with the United States for both constipation (absolute difference, 23.7% [95% CI, 23.2%-24.3%]) and abdominal pain (absolute difference, 20.6% [95% CI, 20.3%-21.0%]). Use of abdominal CT in patients was uncommon in both countries for the target conditions, but approximately 10-fold lower in Canada than the United States for both constipation (0.1% [95% CI, 0.1%-0.2%] vs 1.2% [95% CI, 1.2%-1.2%]) and abdominal pain (0.8% [95% CI, 0.7%-0.9%] vs 7.0% [95% CI, 6.9%-7.1%]). Use of head CT was lower in Canada for febrile convulsion (both simple [absolute difference, 1.7% (95% CI, 1.4%-1.9%)] and complex [absolute difference, 6.6% (95% CI, 3.2%-9.9%)]), seizure (absolute difference, 9.2% [95% CI, 8.1%-10.3%]), and headache (absolute difference, 6.9% [95% CI, 6.4%-7.4%]) and, most notably, for concussion (absolute difference, 22.9% [95% CI, 22.3%-23.4%]). For nonionizing imaging modalities, use of abdominal ultrasonographic imaging was lower for constipation (5.8% [95% CI, 5.6%-6.1%] vs 6.8% [95% CI, 6.8%-6.9%]) and higher for abdominal pain (25.9% [95% CI, 25.5%-26.3%] vs 22.9% [95% CI, 22.8%-23.1%]) in Ontario than the United States. Use of head MRI for concussion, seizure, and headache did not show clinically meaningful differences across the 2 countries.

Differences in low-value imaging between Ontario and the United States were not categorically sustained over the entire study period for all imaging modalities (Figure 3). In some cases, the difference between the 2 countries was reduced in more recent years. The 11-year difference in chest radiography use for bronchiolitis and asthma was −6.8% (95% CI, −7.4% to −6.2%) and −0.7% (95% CI, −1.1% to −0.3%), respectively. However, in 2016 alone, those differences were −1.0% (95% CI, −2.7% to 0.7%) and 3.0% (95% CI, 1.6%-4.4%), respectively (eTable 3 in the Supplement). For some diagnoses, differences in low-value imaging between Ontario and the United States varied in some imaging modalities but not others. While the use of abdominal ultrasonographic imaging for constipation in the United States compared with Canada was 3.1% lower in 2006 and 4.2% higher in 2016, rates of abdominal CTs and abdominal radiographic imaging for constipation remained higher in the United States across all 11 years.

There were no meaningful differences found in 3-day or 7-day hospitalization rates (or hospitalization resulting in intensive care unit care, operative care, or death) between individuals discharged from a pediatric ED in Ontario vs the United States including subanalyses of individuals without imaging during the index ED visit (Table 2 and eTable 4 in the Supplement), as well as among all children visiting the ED (with and without imaging; eTable 5 in the Supplement). An additional analysis limited to children without CCCs similarly did not yield substantive difference in the results (eTable 6 and eTable 7 in the Supplement).

Use of imaging for the quality-control conditions was higher or similar among patients presenting to Ontario EDs compared with EDs in the United States (eFigure in the Supplement). This included admitted patients with pneumonia who had chest radiographic imaging (95.2% [95% CI, 94.7%-95.7%] vs 84.7% [95% CI, 84.4%-84.9%]; P < .001), patients with bowel obstruction who had abdominal radiographic imaging (96.2% [95% CI, 94.9%-97.5%] vs 90.0% [95% CI, 89.4%-90.5%]; P < .001), patients with intussusception who had abdominal ultrasonographic imaging (95.9% [95% CI, 94.1%-97.8%] vs 74.5% [95% CI, 73.6%-75.4%]; P < .001), patients with liver or pancreas laceration who had abdominal CT (85.7% [95% CI, 76.5%-94.9%] vs 56.9% [95% CI, 55.1%-58.8%]; P < .001), patients with shunt malfunction who had head CT (81.8% [95% CI, 79.3%-84.3%] vs 79.9% [95% CI, 79.2%-80.5%]; P = .15), and patients with acute stroke with head MRI (82.3% [95% CI, 75.3%-89.3%] vs. 75.2% [95 % CI, 73.2%-77.1%]; P = .06).

Overall and condition-specific diagnostic imaging rates for a variety of conditions in which imaging is not routinely recommended were higher for children discharged from pediatric EDs in the United States compared with those located in Ontario, Canada. This was most pronounced for diagnostic imaging associated with ionizing radiation (ie, plain radiographs and CT), gastrointestinal diagnoses (ie, abdominal pain and constipation), and concussion. Furthermore, lower imaging rates in Canada were not associated with future hospitalization, intensive care unit admissions, surgery, or death, suggesting that imaging rates may be safely reduced in the United States.

Reports on population-level use suggest that imaging use is consistently higher in the United States than other high-income countries. Compared with Canada, the United States has a higher rate of use of both CT (153 vs 245 per 1000 population) and MRI (56 vs 118 per 1000 population).²⁸ This study extends these findings to other imaging modalities and discharge diagnoses in pediatric EDs, except for MRI, for which rates were similar in both countries. The findings are consistent with a previous report of increased use of CT from 2003 to 2008 in both pediatric and adult patients presenting to EDs in the United States compared with Ontario.²⁹ The overall proportion of children receiving a CT (4.7% in the United States and 1.4% in Ontario) was similar to the findings of this study.

There are a number of potential explanations for these findings. First, although many health care practitioners in both countries are remunerated in fee-for-service payment models, Canadian physicians practice within a broader system of strict global budgets for hospitals and regional health authorities.³⁰ Such financial restrictions may reduce the overall volume of certain services, such as diagnostic imaging, leading to selective use of imaging in certain situations. Second, there may be differences in physician practice associated with differences in training³¹ and/or national guidelines³² in each country. However, guidelines are often outdated and may be poorly adhered to,^33,34 and their use can cross international borders. Third, there may be heightened perception of medicolegal risk in the United States, which may result in reliance on diagnostic imaging for conditions in which the yield of imaging is low.^35,36 Canadian physicians are named in malpractice suits approximately one-quarter as frequently as their US counterparts.³⁷ The decline observed in both countries for some low-value imaging predated Choosing Wisely, possibly because of other published data on the risk of ionizing radiation in children.^38-40 Fourth, patient and parent expectations may differ between the 2 countries. For example, in a setting of paying for services in the United States, the limiting of imaging may not be viewed favorably by some health care consumers. Fifth, patient populations may differ somewhat in the 2 countries, leading to different clinical practice in pediatric ED settings. For instance, patients in the United States may have higher levels of underlying medical complexity and hence may be more challenging to evaluate without imaging (eg, there were more patients with CCCs in the US pediatric EDs). However, exclusion of patients with CCCs did not meaningfully change the findings of this study. Sixth, some diagnostic imaging may be more readily available in pediatric EDs in the United States, although imaging use was higher for every quality-control indication in Ontario. Lastly, higher rates of uninsurance and underinsurance in the United States compared with universal health care coverage in Canada may lead to more reliance on diagnostic testing in the ED, because these patients may lack reliable follow-up care or alternative sources of health care services.

We captured virtually all pediatric ED visits with minimal exclusions in the most populated region of Canada and in a representative sample of pediatric ED visits in the United States. However, most ED visits (approximately 85%) occur at general (rather than pediatric-specific) EDs.⁴¹ We would not generalize these findings to nonpediatric EDs. There may also be differences in patient demographic factors and practice patterns across US regions and Canadian provinces outside Ontario that may limit national generalizations of these findings. We used 2 different data sources with differing quality control and verification processes, which could have introduced misclassification bias in both data sets. We were unable to completely harmonize data collection (eg, race/ethnicity data are not routinely collected in Canada, triage scores were not available in the United States), which limits the ability to compare these cohorts across all relevant variables. There were differences in some baseline demographic factors (eg, neighborhood income, distance from hospital), but these were relatively small compared with the differences in use for some indications. Unfortunately, it was not possible to combine individual-level data across data sets, because the sharing of such data outside of each country was not permitted to conduct adjusted analyses. We were thus limited in ability to assess the appropriateness of imaging at the patient level. While there are situations where imaging is appropriate for the conditions evaluated (eg, atypical presentations), this is likely true in both countries, and we were unable to discern any differences in adverse outcomes between the countries of patients discharged from the pediatric ED without imaging. Additionally, the condition-specific rates of diagnostic imaging relied on discharge diagnosis coding for the ED encounter, rather than chief complaints or reason for visits. Thus, appropriateness of imaging for individual patients cannot be assessed. All analyses were conducted at the level of the ED encounter and patients with multiple ED presentations with the same discharge diagnosis may be at differential risk of having imaging performed. In some cases, the risk may be higher (eg, if undiagnosed symptoms are not resolving), and in other cases, it may be lower (eg, if they present with an exacerbation of a chronic condition). However, we have no reason to believe that multiple ED visits for individual patients differ between countries. Patients with return visits may have also presented at another hospital, so the report of return admissions are likely underestimates. Patients may have also had other consequences of not having imaging performed in the ED, such as return ED visits. The ED returns were not selected as an outcome, because recidivism is not directly associated with disease progression.⁴² Diagnostic imaging rates may have been underreported owing to bundled charges or billing errors or because some imaging may not be captured (eg, those conducted in the community prior to referral to the pediatric ED). However, the quality-control data suggest that, if anything, imaging was more likely underreported in the United States, which would bias this study toward the null hypothesis. We also did not assess each hospital’s imaging availability (eg, after hours), but the quality-control data did not suggest that there was less access in Ontario than the United States.

In this analysis of visits to pediatric EDs across the United States and Ontario, overall and low-value use of diagnostic imaging is higher in the United States compared with Canada for conditions in which routine use of imaging is not routinely recommended. Lower imaging rates were not associated with increased post-ED adverse outcomes. These findings suggest that there may be an opportunity to safely reduce imaging rates in pediatric EDs in the United States to better align resource use with high-quality care delivery.

Accepted for Publication: March 31, 2019.

Published Online: June 3, 2019. doi:10.1001/jamapediatrics.2019.1439

Correction: This article was corrected on August 5, 2019, to fix the graph in Figure 3F. The y-axis of the graph, which shows percentages of patients receiving head magnetic resonance imaging, was mislabeled 0, 10, 20, 30, and 40; it should have been labeled 0, 1, 2, 3, 4, 5, and 6. The Figure has been corrected online.

Corresponding Author: Eyal Cohen, MD, MSc, Division of Pediatric Medicine, Department of Pediatrics, The Hospital for Sick Children, The University of Toronto, 555 University Ave, Toronto, ON M5G 1X8, Canada (eyal.cohen@sickkids.ca).

Author Contributions: Drs Rodean and Diong had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Cohen, Hall, Freedman, Simon, Samuels-Kalow, Morse, Shah, Peltz, Neuman.

Acquisition, analysis, or interpretation of data: Cohen, Rodean, Diong, Hall, Freedman, Aronson, Marin, Alpern, Morse, Shah, Peltz, Neuman.

Drafting of the manuscript: Cohen, Hall, Simon, Neuman.

Critical revision of the manuscript for important intellectual content: Cohen, Rodean, Diong, Freedman, Aronson, Simon, Marin, Samuels-Kalow, Alpern, Morse, Shah, Peltz, Neuman.

Statistical analysis: Rodean, Diong, Hall, Neuman.

Administrative, technical, or material support: Cohen, Freedman, Neuman.

Supervision: Cohen, Freedman, Neuman.

Conflict of Interest Disclosures: Dr Samuels-Kalow reports grants from the National Center for Advancing Translational Sciences/National Institutes of Health (grant UL 1TR002541), during the conduct of the study. Dr Shah reports grants from the Patient-Centered Outcomes Research Institute, grants from the National Heart, Blood, and Lung Institute, grants from the National Institute of Allergy and Infectious Diseases, and personal fees from Society of Hospital Medicine outside the submitted work. No other disclosures were reported.

Funding/Support: This study was supported by the Agency for Healthcare Research and Quality (grant K08HS026006 [Dr Aronson]) and the Alberta Children’s Hospital Foundation Professorship in Child Health and Wellness (Dr Freedman). In addition, ICES is funded by an annual grant from the Ontario Ministry of Health and Long-Term Care. Parts of the data and/or information compiled were provided by the Canadian Institute for Health Information.

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: The opinions, results, and conclusions reported in this article are those of the authors and are independent from the funding sources. No endorsement by ICES, the Ontario Ministry of Health and Long-Term Care, Canadian Institute for Health Information, or the Agency for Healthcare Research and Quality is intended or should be inferred.

Additional Contributions: The authors wish to thank Cynthia Fleming, BSc, and Cindy Groff, the Hospital for Sick Children, for administrative support in the preparation of the manuscript. They were not compensated by a study funder for their contributions.