RESEARCH ARTICLE

Choosing Wisely Campaign: Report Cardand Achievable Benchmarks of Care forChildren’s HospitalsMario Reyes, MD,a,b Evan Paulus, MS,a Carla Hronek, RN, PhD,c Veronica Etinger, MD,a,b Matt Hall, PhD,c Joyee Vachani, MD, MEd,d,e Jennifer Lusk, MD,f

Christopher Emerson, MS,a Patty Huddleson, RN, BSN,f Ricardo A. Quinonez, MDd,e

A B S T R A C T OBJECTIVES: In 2013, the Society of Hospital Medicine (SHM) released 5 pediatric recommendations for the ChoosingWisely Campaign (CWC). Our goals were to develop a report card on the basis of those recommendations, calculateachievable benchmarks of care (ABCs), and analyze performance among hospitals participating in the Pediatric HealthInformation System.

METHODS: Children hospitalized between January 2013 and September 2015 from 32 Pediatric Health InformationSystem hospitals were studied. The quality metrics in the report card included the use of chest radiograph (CXR) inasthma and bronchiolitis, bronchodilators in bronchiolitis, systemic corticosteroids in lower respiratory tract infections(LRTI), and acid suppression therapy in gastroesophageal reflux (GER). ABCs were calculated for each metric.

RESULTS: Calculated ABCs were 22.3% of patients with asthma and 19.8% of patients with bronchiolitis having a CXR, 17.9%of patients with bronchiolitis receiving bronchodilators, 5.5% of patients with LRTIs treated with systemic corticosteroids,and 32.2% of patients with GER treated with acid suppressors. We found variation among hospitals in the use of CXR inasthma (median: 34.7%, interquartile range [IQR]: 28.5%–45.9%), CXR in bronchiolitis (median: 34.4%, IQR: 27.9%–49%),bronchodilators in bronchiolitis (median: 55.4%, IQR: 32.3%–64.9%), and acid suppressors in GER (median: 59.4%, IQR:49.9%–71.2%). Less variation was noted in the use of systemic corticosteroids in LRTIs (median: 13.5%, IQR: 11.1%–17.9%).

CONCLUSIONS: A novel report card was developed on the basis of the SHM-CWC pediatric recommendations,including ABCs. We found variance in practices among institutions and gaps between hospital performances andABCs. These findings represent a roadmap for improvement.

aNicklaus Children’sHospital, Miami, Florida;

bDepartment ofPediatrics, HerbertWertheim College of

Medicine, FloridaInternational University,

Miami, Florida;cChildren’s HospitalAssociation, Lenexa,

Kansas; dTexas Children’sHospital, Houston, Texas;

eDepartment ofPediatrics, Baylor College

of Medicine, Houston,Texas; and fChildren’s

Hospital of OrangeCounty, Orange, California

www.hospitalpediatrics.orgDOI:https://doi.org/10.1542/hpeds.2017-0029Copyright © 2017 by the American Academy of Pediatrics

Address correspondence to Mario Reyes, MD, Division of Pediatric Hospital Medicine, Department of Pediatrics, Nicklaus Children’sHospital, 3100 SW 62nd Avenue, Miami, FL 33155. E-mail: mario.reyes@mch.com

HOSPITAL PEDIATRICS (ISSN Numbers: Print, 2154-1663; Online, 2154-1671).

FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.

FUNDING: No external funding.

POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.

Dr Reyes and Mr Paulus conceptualized and designed the report card and the study, made a substantial contribution to the analysis andinterpretation of data, and drafted, critically reviewed, and revised the manuscript; Ms Hronek made a substantial contribution to the designof the report card, coordinated data collection and analysis and drafted, critically reviewed, and revised the manuscript; Dr Etingerparticipated in data analysis and interpretation, made significant contributions to the initial draft, and reviewed and revised the manuscriptfor final submission; Mr Hall participated in the conceptualization and design of the report card, coordinated acquisition of the study data,conducted the statistical analysis, and drafted, critically reviewed, and revised the manuscript; Drs Vachani and Lusk participated indata analysis and interpretation, made significant contributions to the initial draft, and reviewed and revised the manuscript; Mr Emersonparticipated in the conceptualization and development of the report card and reviewed the manuscript; Ms Huddleson assisted in datacollection and analysis and in the design of the report card; Dr Quinonez participated in the conceptualization and design of the initial studydraft, made a substantial contribution to the analysis and interpretation of data, and critically drafted, reviewed, and revised themanuscript; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

HOSPITAL PEDIATRICS Volume 7, Issue 11, November 2017 633

by guest on May 20, 2018http://hosppeds.aappublications.org/Downloaded from

In 2012, the American Board of InternalMedicine Foundation launched the ChoosingWisely Campaign (CWC) with the goal ofavoiding wasteful or unnecessary medicaltests, treatments, and procedures. Thecampaign encouraged medical societiesto develop a list of 5 therapies ortests that physicians and patientsshould question. In 2013, the Societyof Hospital Medicine (SHM) PediatricCommittee published a list of 5 pediatricrecommendations and the methodology forits development.1 These recommendationsinclude the following:

1. do not order chest radiographs (CXRs)for children with uncomplicated asthmaor bronchiolitis;

2. do not routinely use bronchodilators inchildren with bronchiolitis;

3. do not use systemic corticosteroids inchildren ,2 years of age with anuncomplicated lower respiratory tractinfection (LRTI);

4. do not treat gastroesophageal reflux(GER) in infants routinely with acidsuppression therapy; and

5. do not use continuous pulse oximetryroutinely in children with acuterespiratory illness unless they are onsupplemental oxygen.

The authors encouraged the use of theSHM-CWC pediatrics list in qualityimprovement (QI) efforts aimed at improvingthe care of hospitalized children. Indeed,since the publication of theserecommendations, at least 1 institution hasundertaken and published a QI projectinspired by them.2 However, a comprehensiveexamination regarding variation acrosshospitals as to the adherence to the SHM-CWC pediatric recommendations is lacking.

Hospital dashboards are automatedsystems that report on a discrete numberof key metrics. They are easy to useand displayed graphically, which allowsstakeholders to have an up-to-date glanceat institutional performance on a varietyof metrics.3 Institution members of theChildren’s Hospital Association (CHA) whosubmit data to the Pediatric HealthInformation System (PHIS) have Web-basedaccess to several clinical report cards. The

existing report cards measure institutionaland group practice performance in the careof asthma, bronchiolitis, appendicitis,tonsillectomy, diabetic ketoacidosis, hospitalacquired conditions, Agency for HealthcareResearch & Quality Pediatric QualityIndicators, and in emergency departments.4

The primary goal of this article is todescribe the development of a new reportcard based on the SHM-CWC pediatricrecommendations from PHIS administrativedata. This dashboard will help institutionsand group practices to measure andbenchmark individual performance onadherence to these recommendations.

However, the recommendations are notexpected to achieve 100% compliance.An additional purpose of this article isproviding realistic goals for benchmarkingby calculating achievable benchmarks ofcare (ABCs) by using a previously publishedmethod.5–7

By compiling the performance of thetertiary children’s hospitals included in thisstudy, we expect to have a glimpse at thecompliance with the 2013 SHM-CWCpediatric recommendations in thesecenters.

METHODS

From 2014 to 2016, a multicenter workinggroup from 4 children’s hospital membersof CHA and analytic support specialistsworked on a collaborative project. The PHISdatabase (CHA, Lenexa, KS) was queried todevelop a report card for the SHM-CWCpediatric recommendations. This databaseincludes clinical and daily billing data from48 tertiary care children’s hospitals andaccounts for ∼20% of all US pediatrichospitalizations every year. Participatinghospitals submit de-identifying data. A jointeffort between CHA and participatinghospitals ensures the quality of the datasubmitted, as has been previouslydescribed.8 These data are subjected to aroutine quality check with each submissionand within each report. In total, 32 PHIShospitals met the strict quality standardsfor inclusion of submitted data. The rest ofthe hospitals did not meet the strict dataquality requirements from PHIS and wereexcluded from this study.

This research was exempt from approval bythe institutional review board of NicklausChildren’s Hospital.

Development of the Report Card

The development of a hospital-specificreport card required multiple conferencecalls among the authors and included thefollowing steps:

1. assembly of content (all authors) anddesign of the report card (developmentexperts);

2. selection and definition of measures,population inclusion, and exclusioncriteria;

3. selection of ABCs as a method forestablishing targets in the report card;

4. selection of the established CHA reportcard format for consistency inpresentation and ease of use; and

5. education regarding content andpurpose provided to users throughWebcasts and newsletters.

SHM-CWC PediatricRecommendations Translated IntoQuality Metrics

Four of the 5 SHM-CWC pediatricrecommendations were able to betranslated into 5 quality metrics. The firstrecommendation (avoiding the use of CXR inasthma and bronchiolitis) was convertedinto 2 separate metrics, 1 for eachcondition. The fifth SHM-CWC pediatricrecommendation, limiting the use ofcontinuous pulse oximetry for acuterespiratory illness unless the patient isreceiving supplemental oxygen, could notbe translated into a metric by using PHISdata because of inconsistent reporting ofthe use of these resources. Consequently,the report card has 5 metrics:

1. percentage of patients withuncomplicated asthma receiving CXR;

2. percentage of patients withuncomplicated bronchiolitis receivingCXR;

3. percentage of patients withuncomplicated bronchiolitis receivingbronchodilators;

4. percentage of patients with LRTIreceiving systemic corticosteroids; and

634 REYES et al

by guest on May 20, 2018http://hosppeds.aappublications.org/Downloaded from

5. percentage of patients withuncomplicated GER receiving acidsuppressor therapy.

These metrics have the study populationselected as the denominator and thenumber of patients who received thecontraindicated therapy or intervention asa numerator. Thus, the percentagesdisplayed on the report card representnonadherence to the recommendations.

Study Population

Children were eligible for inclusion if theywere hospitalized (inpatient or observationstatus) between January 1, 2013 andSeptember 31, 2015. The SHM-CWC pediatric

recommendations refer to “uncomplicated”conditions. Some of the original

recommendations do not specify age

groups.1 We defined different age groups

for each condition as well as inclusion

and exclusion criteria aimed at

identifying only uncomplicated cases,

as follows:

• Asthma: 3 to 17 years. Considering thatthe diagnosis of asthma in infants and

children younger than 3 years old is

challenging and it may overlap with other

causes of wheezing. The upper limit of

17 years of age will capture most children

with this condition.

• Bronchiolitis: $1 month to ,1 year.This age group was selected to excludenewborn infants who tend to have a moresevere disease and patients older than1-year-old who may have other causes ofrecurrent wheezing that may requireinhaled bronchodilators and CXR.

• LRTI: $1 month to ,2 years. This agegroup closely matches the age group ofthe SHM-CWC pediatric recommendationon avoiding systemic corticosteroids.Newborn infants are excluded consideringthe likelihood of more severe disease.This group includes mostly cases ofbronchiolitis and pneumonia who do notbenefit from this intervention.

TABLE 1 SHM-CWC Pediatric Recommendations in 2013

Condition Inclusion Criteria Exclusion Criteria Metric, %

Asthma Age: 3–17 y ICU admission, APR-DRG SOI, and 4 ICD-9 codes,including

Receiving CXR

ICD-9 codes: asthma PneumoniaLOS #5 d MalnutritionPatient type: inpatient and observation Neurologically impaired children

CLDCongenital lung anomaliesCongenital heart defectsRespiratory failure and/or acidosisOrofacial anomaliesPresence of tracheostomy tube

Bronchiolitis Age: $1 m and ,1y ICD-9 codes: bronchiolitis ICU admission APR-DRG SOI, and 4 ICD-9 codes,including

Receiving CXR

LOS #5 d Asthma Receiving bronchodilatorPatient type: inpatient and observation Malnutrition

Neurologically impaired childrenCLDCongenital lung anomaliesCongenital heart defectsRespiratory failure and/or acidosisOrofacial anomaliesPresence of tracheostomy tube

LRTI Age: $1 m and ,2 y ICU admission APR-DRG SOI, and 4 ICD-9 codes,including

Receiving systemic corticosteroid

ICD-9 codes: bronchiolitis, pneumonia, other LRTI AsthmaMalnutrition

Patient type: inpatient and observation Neurologically impaired childrenCLDCongenital lung anomaliesCongenital heart defectsRespiratory failure and/or acidosisOrofacial anomaliesPresence of tracheostomy tube

GER Age: #1 y ICU Admission, APR-DRG SOI, 4 ICD-9 codes, including Receiving acid suppression therapyICD-9 codes: esophageal reflux Malnutrition and/or failure to thrivePatient type: inpatient and observation Apnea and/or ALTE

EsophagitisPeptic ulcer diseaseGastritis and/or duodenitis

CHA-PHIS report card: definitions of uncomplicated conditions and metrics. A complete list of inclusion and exclusion codes is available on request by contactinganalytics@childrenshospitals.org or see online Supplemental Materials. ALTE, apparent life threatening event; CLD, chronic lung disease; SOI, severity of illness.

HOSPITAL PEDIATRICS Volume 7, Issue 11, November 2017 635

by guest on May 20, 2018http://hosppeds.aappublications.org/Downloaded from

• GER: #1 year. This age group matchesthe age group of infants, for whomthe specific SHM-CWC pediatricrecommendation is intended. In this agegroup, the presence of physiologic,uncomplicated GER is more likely, and theroutine use of acid suppressors is notrecommended.

Inclusion and Exclusion Criteria

As inclusion criteria, the working groupdeveloped comprehensive lists ofInternational Classification of Diseases,Ninth Revision (ICD-9) codes9 that identifybronchiolitis, asthma, LRTI and GER.Exclusion criteria comprised a detailed listof ICD-9 codes9 corresponding to multipleconditions and comorbidities that maycomplicate the course or treatment ofthese conditions. Additional exclusioncriteria also included were length of stay(LOS) of more than 5 days, hospitalizationswith charges for intensive care services,and extreme severity of illness (level 4)from the All Patient Refined DiagnosisRelated Group (3M Corp, Wallingford, CT).This way, the cases selected for the studyhave a short LOS, no complicationsrequiring intensive care, and no otherchronic or comorbid conditions.

Additionally, to minimize misclassification,for each category of respiratory conditions,we excluded cases with any of the otherconditions if they were listed as acomorbidity because they may requiredifferent management. For example,patients with an ICD-9 code indicatingasthma are excluded from the population ofpatients with bronchiolitis and LRTI, and viceversa.

The resulting cohorts represent uniformpopulations of uncomplicated conditions, asexplicitly targeted in the SHM-CWC pediatricrecommendations.

The interventions and therapies notroutinely recommended are CXR,bronchodilators, systemic corticosteroids,and acid suppressor therapy. Developmentexperts define these interventions through amapping process in which PHIS hospitals’data are translated from their charge codesinto clinical translation classification codes.These codes are a meaningful tool to

make hospital billing comparable amonginstitutions, and they are updated yearly.

Table 1 shows the age groups for eachcondition, a summary of the inclusion andexclusion criteria, and the 5 metricsselected. Readers can obtain the completelists of inclusion and exclusion criteria andthe most current group of clinicaltranslation classification codes identifyinginterventions not recommended in theSHM-CWC pediatric recommendations bycontacting Carla Hronek (carla.hronek@childrenshospitals.org). The detailed ICD-9lists of inclusion and exclusion criteria foreach condition and the lists of codes forprocedures and interventions are availablein the Supplemental Fig 4, which details theexclusion list for respiratory conditions).

Statistical Analysis

We compiled the combined hospitalperformances on each of the quality metricsincluded in the report card. Categoricalvariables are summarized with countsand percentages, whereas medians andinterquartile ranges (IQRs) summarizedcontinuous variables.

By using a previously published method,5–7

ABCs for the 5 SHM-CWC pediatricmetrics included in this study wereestimated. ABCs are calculated by firstranking the hospitals on the basis of theirperformance on a specific metric. Then, thebest-performing hospitals that comprise10% of the total patient population areselected. Finally, calculated ABCs representthe average from those top-performingcenters.5

Results are presented in tables andgraphics. All statistical analyses wereperformed by using SAS version 9.3 (SASInstitute Inc, Cary, NC).

RESULTS

Table 2 shows the demographic informationand the APR-DRG distribution of the cohortsfor each condition included in the SHM-CWCpediatric recommendations. The studyincluded 32 487 cases of asthma,23 531 cases of bronchiolitis, 5963 casesof LRTI, and 2946 cases of GER. A slightpredominance of boys in all categorieswas noted: 61.9% in asthma, 59.6% inbronchiolitis, 55% in LRTI, and 53.6% in GER.

TABLE 2 Cohort Demographics

Asthma Bronchiolitis LRTI GER

N 32 487 23 531 5963 2946

Age, y

,1 — 23 531 (100) 2858 (47.9) 2719 (92.3)

1–4 9520 (29.3) — 3105 (52.1) 227 (7.7)

5–9 15 088 (46.4) — — —

10–18 7879 (24.3) — — —

Sex

Male 20 124 (61.9) 14 036 (59.6) 3281 (55) 1578 (53.6)

Female 12 363 (38.1) 9495 (40.4) 2681 (45) 1368 (46.4)

Race

Non-Hispanic white 8081 (24.9) 10 055 (42.7) 2209 (37) 1582 (53.7)

Non-Hispanic African American 13 980 (43) 4400 (18.7) 987 (16.6) 464 (15.8)

Hispanic 7218 (22.2) 6202 (26.4) 1835 (30.8) 580 (19.7)

Asian American 679 (2.1) 516 (2.2) 241 (4) 34 (1.2)

Other 2529 (7.8) 2358 (10) 691 (11.6) 286 (9.7)

APR-DRG SOI

Minor 24 829 (76.4) 16 644 (70.7) 3141 (52.7) 1697 (57.6)

Moderate 7385 (22.7) 6636 (28.2) 2562 (43) 957 (32.5)

Major 273 (0.8) 251 (1.1) 260 (4.4) 292 (9.9)

Q1 of 2013 to Q3 of 2015. N 5 patients from 32 hospitals. Source: PHIS database.10 SOI: severity of illness;—, not applicable.

636 REYES et al

by guest on May 20, 2018http://hosppeds.aappublications.org/Downloaded from

We also observed a higher prevalence ofnon-Hispanic African American patients withasthma (43%). Inversely, non-Hispanic whitepatients were more prevalent in thebronchiolitis, LRTI, and GER groups, mostlyin the GER cohort (53.7%). The “minor”(severity level 1) APR-DGR severity of illnessfor each condition represents more thanhalf of the patients in each of the diagnosticgroups, ranging from 52.7% in LRTI to 76.4%in asthma.

Table 3 shows the calculated ABCs foreach of study populations, along with thenumber of hospitals used to createthese benchmarks. These calculated ABCsare 22.3% of patients with asthma and19.8% of patients with bronchiolitis havinga CXR performed, 17.9% of patients withbronchiolitis receiving bronchodilators,5.5% of patients with LRTI treated withsystemic corticosteroids, and 32.2% ofpatients with GER treated with acidsuppressor therapy. According to themethodology described above, only 3 to4 hospitals were included in the estimationof the ABCs, representing ∼10% of thepopulation for each cohort.

Figure 1 represents the variation inperformance among the participatinghospitals for the 5 metrics included in thestudy in box plots. Each measure, exceptsteroid use in LRTIs, shows significantvariability among institutions, as depicted

by the whiskers on each box plot: CXR inasthma (median: 34.7%, IQR: 28.5%–45.9%),CXR use in bronchiolitis (median: 34.4%,IQR: 27.9%–49%), bronchodilators usein bronchiolitis (median: 55.4%, IQR:32.3%–64.9%), and use of acid suppressortherapy in GER (median: 59.4%, IQR:49.9%–71.2%). The use of systemiccorticosteroids in LRTI shows less variance(median: 13.5%, IQR: 11.1%–17.9%).Additionally, we observed a clear gapbetween hospital performances and ABCs,as represented by the black circles, in allconditions.

Figure 2 shows the performance of eachhospital, benchmarked against the ABCsfor each of the SHM-CWC pediatricrecommendations in a heat map. Ingeneral, more hospitals performed furtherfrom the ABC in the use of relievers forbronchiolitis and the use of acid suppressortherapy in GER, represented by the darkestgray boxes (.20 percentage points frombenchmark) and medium gray boxes(11–20 percentage points from benchmark).On the other hand, participating hospitalswere closer to or met the ABC regarding theavoidance of steroids for LRTI, representedby the more frequent lighter gray (0–10

percentage points from benchmark) andclear boxes (meet benchmark). Fewhospitals were closer to or met benchmarks(ABCs) on most metrics, and none were ableto meet benchmarks or to reach ,10%variability from the benchmark on all5 metrics.

Figure 3 displays bar graphics that showthe variation in performance on the5 metrics among the 32 hospitals includedin the study. The shorter bars, closer to theABC line, represent the best-performingcenters, whereas the tallest barscorrespond to institutions performingfurther away from the benchmark. Theranges of variation from the best-performing hospital (lower percentage ineach metric) to the worst-performing center(higher percentage) are as follows: CXR inasthma (6.8%–71.2%); CXR in bronchiolitis(7.5%–69.8%); relievers in bronchiolitis(14.3%–82.6%); systemic corticosteroids inLRTI (4.8%–28%); and acid suppressortherapy in GER (26.2%–100%).

This report card is displayed in thestandard format used by CHA in otherclinical report cards. This format allows fora visual and numeric comparison of a giveninstitution or physician group with both the

FIGURE 1 SHM-CWC pediatric recommendations. Variation among 32 children’s hospitals thatare CHA and/or PHIS members from 2013 Q1 to 2015 Q3 is shown. In a box plot, thecentral rectangle spans the 25th percentile to the 75th percentile. The line inside therectangle displays the median. The horizontal lines extending from the box indicatethe 25th percentile 2 (1.5 3 IQR) and the 75th percentile 1 (1.5 3 IQR). Source: PHISdatabase.10

TABLE 3 ABCs for Conditions Included inthe SHM-CWC PediatricRecommendations

Condition andTreatment

No. of Hospitalsin ABCa

ABC,%

Asthma

Received CXR 4 22.3

Bronchiolitis

Received CXR 4 19.8

Received reliever 3 17.9

LRTI

Received systemicsteroid

4 5.5

GER

Received acidsuppressor

4 32.2

Q1 of 2013 to Q3 of 2015. Source: PHIS database.10a Number of sample hospitals that contributed tothe calculation of the ABC by measure.

HOSPITAL PEDIATRICS Volume 7, Issue 11, November 2017 637

by guest on May 20, 2018http://hosppeds.aappublications.org/Downloaded from

peer group of participating hospitals andwith the ABC for each metric. The reportcard allows customization by selectingspecific timeframes and provider groups forbenchmarking purposes. CHA- and PHIS-participating institutions can generate theircustomized, landscape version of the reportcard by accessing www.childrenshospitals.org.10

DISCUSSION

The SHM-CWC pediatric recommendationsreport card is the latest available for PHISmembers. This is the first clinical reportcard that includes calculated ABCs asbenchmarks of excellence and has a focuson overuse. Institutions can use these datato identify areas in which they aresignificant outliers for local QI projects.

The aggregated results presented hererepresent the first available data setof the performance of children’shospital members of CHA after therecommendations were released.

As previously described, the use of ABCsis an original method intended to helpindividual clinician groups and institutionsbenchmark their performance. ABCsrepresent (1) a measurable level ofexcellence, (2) attainable goals, and (3)a benchmark derived from data in anobjective and reproducible manner.5–7 Thecalculated ABCs in our study compare wellwith the ABCs published by Parikh et al in20145 in 4 of our 5 metrics. The 2 studiesused the same database (PHIS), but their

study included data from the calendar year2012, whereas our study included largercohorts corresponding to 11 quarters(January 2013–September 2015).Furthermore, we used more restrictivecriteria to define uncomplicated conditions,adding extreme APR-DRG, intensive careservices, and any LOS of more than5 days as exclusion criteria, in addition toincluding multiple chronic conditions orcomorbidities. Despite some differences indefining uncomplicated cases, the finding ofsimilar ABCs obtained in 2 different periodsmay suggest that no significant changeswere present in practices among theincluded children’s hospitals since 2012,when the calculation of ABCs took place forthe first time.

Peer-reviewed, society-endorsed guidelinesfor the conditions included in this study arewidely available.11–13 Clinical guidelines aredesigned to provide guidance while stillpreserving individual physician autonomy,modification on the basis of the clinicalscenario, and accommodation for patient/family preferences. Although full adherenceto clinical guidelines is not expected,reasonable and attainable goals ofcompliance are. However, as shown in Figs1–3, significant variation in care still existsamong the tertiary centers included inthis study, as well as a clear gap betweenABCs and actual performance by most of theinstitutions. These facts suggest an overuseof resources with an associated increasein costs related to hospital care. Thesefindings have been previously reported byothers.5,14–17

Recent evidence suggests some progress inthe care of hospitalized children withbronchiolitis since the publication ofAmerican Academy of Pediatrics guidelinesfor this common inpatient condition.18–21

However, as we reaffirm in this study, muchunwarranted variation is still the norm,specifically in the excessive use ofbronchodilators. Improvement will requireconcerted local efforts aimed at specificconditions. Recent evidence demonstratesthat it is possible to change the cultureand create meaningful and sustainableimprovement through local efforts18 ormulticenter QI collaboratives,21 both in

FIGURE 2 Children’s hospitals’ performance on 5 quality measures on the SHM-CWC pediatricrecommendations from 2013 Q1 to 2015 Q3. Source: PHIS database.10

638 REYES et al

by guest on May 20, 2018http://hosppeds.aappublications.org/Downloaded from

tertiary children hospitals and communitysettings. Similar collaborative efforts todecrease unnecessary care in children withother lower respiratory tract illnesses,including asthma, are lacking. Thesignificant improvements noted throughsuch efforts addressing bronchiolitis, andthe evidence of room for improvementdemonstrated through our scorecard, couldserve as the impetus for such work.

This work adds a newly calculated anachievable benchmark for the use of acidsuppressor therapy in children withuncomplicated GER (32.2%), as shownin Table 2. Several studies, includingsystematic reviews, have described theineffectiveness of acid-reducing medicationsin the treatment of children with GER.22,23

Despite this, since the year 2000, a threefoldincrease in diagnosis (3.4%–12.3%) and a

concomitant doubling of the use of protonpump inhibitors (31.5%–62.6%) haveoccurred in infants.24 A substantial growingbody of evidence suggests that this practicemay also lead to significant harm.25 Ourstudy suggests that wide variation stillexists in the use of acid suppressor therapyin children with GER across institutions(Figs 1–3). The fact that most institutionsare far from meeting this benchmark of

FIGURE 3 Variation in Metrics from the SHM-CWC pediatric recommendations from 32 children’s hospitals in the United States from 2013 Q1 to2015 Q3. The interrupted horizontal line indicates ABCs. Best-performing hospitals (shorter bars) are closer to the ABC line. Source: PHISdatabase.10 A, CXR for asthma. B, CXR for bronchiolitis. C, Relievers for bronchiolitis. D, Acid suppression for GER. E, Steroids for LRTIs.

HOSPITAL PEDIATRICS Volume 7, Issue 11, November 2017 639

by guest on May 20, 2018http://hosppeds.aappublications.org/Downloaded from

excellence, coupled with the strong evidenceof ineffectiveness and potential harm of thistherapy, makes this topic ripe for QI effortsat local and national levels.

The SHM-CWC pediatric recommendationsreport card could be specifically tailoredby each institution member of the CHAconsortium and has been readily availableto use for internal QI prioritization since2016. As shown in Figs 2 and 3, mosthospitals tended to underperform on theuse of relievers in bronchiolitis andprescription of acid suppressor therapy inGER. On the other hand, the use of steroidsin LRTI represents the indicator with thelowest performance gap from thebenchmark. Thus, prioritization of QI effortscould involve decreasing bronchodilatoruse in children with bronchiolitis andprescribing acid suppressor therapy ininfants with uncomplicated, physiologic GER.

This report card represents an initial, validtool to assess ongoing performance on themetrics selected for this study and theircomparison with “best practices.” Theinclusion of attainable benchmarks, newlyand previously described for these specificconditions, makes the report card amuch more meaningful instrument forbenchmarking, by providing realistic goalson the basis of peer performance.

This study has several limitations, includingthose limitations associated withadministrative and billing databaseresearch. Variation in documentation andcoding practices may result in the presenceof misidentified children in the cohortsand may hinder the risk adjustment formeaningful benchmarking. Additionally, theextensive list of inclusion and exclusioncriteria based on chronic and/orcomorbid conditions may exclude someuncomplicated cases for which therecommendations were intended.

Future versions of this report card based ondata from the more granular InternationalClassification of Diseases, Tenth Revisionclassification may be more revealing. Also,this study does not include informationon whether a hospital implemented anychanges or interventions in an attempt tofollow the SHM-CWC pediatricrecommendations.

Another limitation is the lack of data onhospital care provided at institutions thatare not members of CHA. It is estimated thatthe majority (∼75%) of pediatrichospitalizations in the United States takeplace in hospitals that are not children’shospitals.20 Both the ABCs and theaggregated results may be different inthat population.

The study does not include directions orsuggestions on how to change practicesthrough particular interventions or QIefforts. One final limitation is that the reportcard is available only to PHIS memberinstitutions. However, by using themethodology and definitions described hereand the information provided in the onlinesupplement, other institutions couldreproduce the report card and use thecalculated ABCs as a benchmarkingreference.

CONCLUSIONS

A report card for the purpose of assessingand benchmarking 5 metrics derived fromthe SHM-CWC pediatric recommendations iscurrently available for CHA and/or PHISmember institutions. In this study, wedescribe the development of the report cardand the performance of these institutionsfrom the first quarter (Q1) of 2013 to thethird quarter (Q3) of 2015, along withcalculated ABCs for each metric. In thisstudy, we primarily demonstrate variation inpractices and provide an initial pathway forchange. In the future, individual institutionsand collaborative projects can use thesedashboards to track their performance andas a roadmap for improvement. QI effortsderived from the report card findings maylead to decreased variation in the care thathospitalized children receive for thesecommon inpatient pediatric illnesses.

Acknowledgments

We acknowledge the Research Institute andthe Medical Library from Nicklaus Children’sHospital in Miami, Florida, for theirassistance and involvement in thepreparation and submission of this article.

REFERENCES

1. Quinonez RA, Garber MD, Schroeder AR,et al. Choosing wisely in pediatric

hospital medicine: five opportunities forimproved healthcare value. J Hosp Med.2013;8(9):479–485

2. Schondelmeyer AC, Simmons JM, StatileAM, et al. Using quality improvement toreduce continuous pulse oximetry use inchildren with wheezing. Pediatrics. 2015;135(4). Available at: www.pediatrics.org/cgi/content/full/135/4/e1044

3. Serb C. Effective dashboards: what tomeasure and how to show it. HospHealth Netw. 2011;85(6):8 p following40, 2

4. Children’s Hospital Association. Reportcards. Available at: https://www.childrenshospitals.org/-/media/files/groups/phis/conferences/2015-spring/hronek–utlizing-phis-clinical-report-cards.pdf?la5en&hash521029365952270ae82e898c5dc1a8a3af15f972f. Accessed

5. Parikh K, Hall M, Mittal V, et al.Establishing benchmarks for thehospitalized care of children withasthma, bronchiolitis, and pneumonia.Pediatrics. 2014;134(3):555–562

6. Kiefe CI, Weissman NW, Allison JJ,Farmer R, Weaver M, Williams OD.Identifying achievable benchmarks ofcare: concepts and methodology. Int JQual Health Care. 1998;10(5):443–447

7. Weissman NW, Allison JJ, Kiefe CI, et al.Achievable benchmarks of care: theABCs of benchmarking. J Eval Clin Pract.1999;5(3):269–281

8. Mongelluzzo J, Mohamad Z, Ten Have TR,Shah SS. Corticosteroids and mortalityin children with bacterial meningitis.JAMA. 2008;299(17):2048–2055

9. Centers for Disease Control andPrevention. Classification of diseases,functioning, and disability. 2013.Available at: www.cdc.gov/nchs/icd.htm.Accessed April 1, 2017

10. Children’s Hospital Association. Pediatrichealth information system database.Available at: https://www.childrenshospitals.org/Programs-and-Services/Data-Analytics-and-Research/Pediatric-Analytic-Solutions/Pediatric-Health-Information-System. AccessedSeptember 14, 2017

640 REYES et al

by guest on May 20, 2018http://hosppeds.aappublications.org/Downloaded from

11. Bradley JS, Byington CL, Shah SS, et al;Pediatric Infectious Diseases Society andthe Infectious Diseases Society ofAmerica. The management ofcommunity-acquired pneumonia ininfants and children older than3 months of age: clinical practiceguidelines by the Pediatric InfectiousDiseases Society and the InfectiousDiseases Society of America. Clin InfectDis. 2011;53(7):e25–e76

12. National Asthma Education andPrevention Program; National Heart,Lung, and Blood Institute. Expert panelreport III: guidelines for the diagnosisand management of asthma. 2007.Available at: https://www.nhlbi.nih.gov/files/docs/guidelines/asthgdln.pdf.Accessed April 1, 2017

13. Ralston S, Lieberthal A, Meissner H, et al.Clinical practice guideline: the diagnosis,management, and prevention ofbronchiolitis AAP gateway. Pediatrics.2014;134(5):e1474–e1502

14. Macias CG, Mansbach JM, Fisher ES, et al.Variability in inpatient management ofchildren hospitalized with bronchiolitis.Acad Pediatr. 2015;15(1):69–76

15. Christakis DA, Cowan CA, Garrison MM,Molteni R, Marcuse E, Zerr DM. Variationin inpatient diagnostic testing and

management of bronchiolitis. Pediatrics.2005;115(4):878–884

16. Bratton SL, Odetola FO, McCollegan J,Cabana MD, Levy FH, Keenan HT. Regionalvariation in ICU care for pediatric patientswith asthma. J Pediatr. 2005;147(3):355–361

17. Florin TA, Byczkowski T, Ruddy RM, ZorcJJ, Test M, Shah SS. Variation in themanagement of infants hospitalized forbronchiolitis persists after the2006 American Academy of Pediatricsbronchiolitis guidelines. J Pediatr. 2014;165(4):786–792.e1

18. Mittal V, Hall M, Morse R, et al. Impact ofinpatient bronchiolitis clinical practiceguideline implementation on testing andtreatment. J Pediatr. 2014;165(3):570–576.e3

19. Parikh K, Hall M, Teach SJ. Bronchiolitismanagement before and after the AAPguidelines. Pediatrics. 2014;133(1).Available at: www.pediatrics.org/cgi/content/full/133/1/e1

20. McCulloh RJ, Smitherman SE, Koehn KL,Alverson BK. Assessing the impact ofnational guidelines on the managementof children hospitalized for acutebronchiolitis. Pediatr Pulmonol. 2014;49(7):688–694

21. Ralston SL, Garber MD, Rice-Conboy E,et al; Value in Inpatient Pediatrics

Network Quality Collaborative forImproving Hospital Compliance with AAPBronchiolitis Guideline (BQIP). AMulticenter collaborative to reduceunnecessary care in inpatientbronchiolitis. Pediatrics. 2016;137(1):e20150851

22. van der Pol RJ, Smits MJ, van Wijk MP,Omari TI, Tabbers MM, Benninga MA.Efficacy of proton-pump inhibitors inchildren with gastroesophageal refluxdisease: a systematic review. Pediatrics.2011;127(5):925–935

23. Higginbotham TW. Effectiveness andsafety of proton pump inhibitors ininfantile gastroesophageal refluxdisease. Ann Pharmacother. 2010;44(3):572–576

24. Nelson SP, Kothari S, Wu EQ, Beaulieu N,McHale JM, Dabbous OH. Pediatricgastroesophageal reflux disease andacid-related conditions: trends inincidence of diagnosis and acidsuppression therapy. J Med Econ. 2009;12(4):348–355

25. Chung EY, Yardley J. Are there risksassociated with empiric acidsuppression treatment of infants andchildren suspected of havinggastroesophageal reflux disease? HospPediatr. 2013;3(1):16–23

HOSPITAL PEDIATRICS Volume 7, Issue 11, November 2017 641

by guest on May 20, 2018http://hosppeds.aappublications.org/Downloaded from

DOI: 10.1542/hpeds.2017-0029 originally published online October 24, 2017; 2017;7;633Hospital Pediatrics 

QuinonezVachani, Jennifer Lusk, Christopher Emerson, Patty Huddleson and Ricardo A. Mario Reyes, Evan Paulus, Carla Hronek, Veronica Etinger, Matt Hall, Joyee

for Children's HospitalsChoosing Wisely Campaign: Report Card and Achievable Benchmarks of Care

ServicesUpdated Information &

http://hosppeds.aappublications.org/content/7/11/633including high resolution figures, can be found at:

Referenceshttp://hosppeds.aappublications.org/content/7/11/633#BIBLThis article cites 18 articles, 6 of which you can access for free at:

Subspecialty Collections

provement_subhttp://classic.hosppeds.aappublications.org/cgi/collection/quality_imQuality Improvementedicine_subhttp://classic.hosppeds.aappublications.org/cgi/collection/hospital_mHospital Medicineion:practice_management_subhttp://classic.hosppeds.aappublications.org/cgi/collection/administratAdministration/Practice Managementfollowing collection(s): This article, along with others on similar topics, appears in the

Permissions & Licensing

mlhttp://classic.hosppeds.aappublications.org/site/misc/Permissions.xhtin its entirety can be found online at: Information about reproducing this article in parts (figures, tables) or

Reprintshttp://classic.hosppeds.aappublications.org/site/misc/reprints.xhtmlInformation about ordering reprints can be found online:

by guest on May 20, 2018http://hosppeds.aappublications.org/Downloaded from

DOI: 10.1542/hpeds.2017-0029 originally published online October 24, 2017; 2017;7;633Hospital Pediatrics 

QuinonezVachani, Jennifer Lusk, Christopher Emerson, Patty Huddleson and Ricardo A. Mario Reyes, Evan Paulus, Carla Hronek, Veronica Etinger, Matt Hall, Joyee

for Children's HospitalsChoosing Wisely Campaign: Report Card and Achievable Benchmarks of Care

http://hosppeds.aappublications.org/content/7/11/633located on the World Wide Web at:

The online version of this article, along with updated information and services, is

http://hosppeds.aappublications.org/content/suppl/2017/10/13/hpeds.2017-0029.DCSupplementalData Supplement at:

Pediatrics. All rights reserved. Print ISSN: 2154-1663. Boulevard, Elk Grove Village, Illinois, 60007. Copyright © 2017 by the American Academy of published, and trademarked by the American Academy of Pediatrics, 141 Northwest Pointpublication, it has been published continuously since 2012. Hospital Pediatrics is owned, Hospital Pediatrics is the official journal of the American Academy of Pediatrics. A monthly

by guest on May 20, 2018http://hosppeds.aappublications.org/Downloaded from