RESEARCH ARTICLE

A Quality Improvement Intervention to ImproveInpatient Pediatric Asthma Controller AccuracyAlexander H. Hogan, MD, MS,a Deepa Rastogi, MBBS, MS,b Michael L. Rinke, MD, PhDb

A B S T R A C T OBJECTIVES: Our objective was to investigate if a rigorous quality improvement (QI) interventioncould increase accuracy of pediatric asthma controller medications on discharge from an inpatienthospitalization.

METHODS: Our interprofessional QI team developed interventions such as improvingdocumentation and creating standardized language to ensure patients were discharged on anappropriate asthma controller medication and improve assessment of asthma symptom control.Each week of 2015–2016, the first 5 patients discharged with status asthmaticus from the pediatricwards were reviewed for documentation of the 6 asthma control questions and accuracy of thedischarge controller therapy. Correct discharge medication was defined as being prescribed theage-appropriate medication and dose on the basis of baseline controller therapy, compliance withbaseline medication, and responses to asthma control assessment. The weekly proportion of controlquestions that were accessed and correct controller medications that were prescribed were analyzedby using Nelson rules and interrupted time series.

RESULTS: A total of 240 preintervention and 252 postintervention charts were reviewed. Theprimary outcome of the median proportion of patients discharged on appropriate controller therapyimproved from 60% in preintervention data to 80% in the postintervention period. The processmeasure of proportion of asthma control questions that were assessed improved from 43% in thepreintervention period to 98% by the final months of the intervention period. Both of these changeswere statistically significant as per Nelson’s rules and interrupted time series analyses (P 5 .02 andP , .001, respectively, for postintervention break).

CONCLUSIONS: An interdisciplinary QI team successfully improved the accuracy of asthmacontroller therapy on discharge and the inpatient assessment of asthma control questions.

aDepartment ofPediatrics, Connecticut

Children’s Medical Center,Hartford, Connecticut;

and bChildren’s Hospitalat Montefiore, Bronx,New York, New York

www.hospitalpediatrics.orgDOI:https://doi.org/10.1542/hpeds.2017-0184Copyright © 2018 by the American Academy of Pediatrics

Address correspondence to Alexander H. Hogan, MD, MS, Department of Pediatrics, Connecticut Children’s Medical Center,282 Washington St, Hartford, CT 06106. E-mail: ahogan@connecticutchildrens.org

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 Hogan designed the study, conducted all analyses, and drafted the initial manuscript; Dr Rastogi aided in study design, conductedvalidating chart review, and critically reviewed the manuscript; Dr Rinke mentored throughout the project, aided in study design anddata analysis, and reviewed and revised the manuscript; and all authors approved the final manuscript as submitted and agree to beaccountable for all aspects of the work.

HOSPITAL PEDIATRICS Volume 8, Issue 3, March 2018 127

by guest on September 2, 2020www.aappublications.org/newsDownloaded from

Asthma is the most common chronicpediatric condition, affecting 7 millionchildren in the United States and causing190 000 hospitalizations annually. Thismorbidity can be reduced throughappropriate preventative or “controller”medications, especially inhaledcorticosteroids (ICS).1–4 The current asthmamanagement guidelines from the NationalHeart, Lung and Blood Institute’s NationalAsthma Education and Prevention Program(NAEPP) recommend that severity beclassified for all children with asthma(intermittent, mild persistent, moderatepersistent, severe persistent), and thenappropriate controller medications shouldbe prescribed, with attention to “steppingup” or “stepping down” medication dosageson the basis of disease control.4 Adherenceto the NAEPP guidelines increases ICS useand decreases asthma-specific emergencydepartment visits and hospitalization rates.5

Unfortunately, these guidelines are often notfollowed and controller medications areoften not appropriately prescribed.6 Much ofthis research has been conducted in theambulatory setting,7–11 and it is unclear howit translates to the pediatric inpatient arenawhere the sickest asthmatic patients areadmitted.

In the inpatient setting, researchers ofquality improvement (QI) studies havepredominately evaluated the effect ofclinical pathways on asthma processmeasures and outcomes.12–14 Compliancewith Joint Commission–mandatedmeasures such as the percent of patientsreceiving albuterol and systemic steroidsare not associated with improved outcomes,likely because compliance is generally highat baseline.15 In terms of ICS prescriptions,the authors of a systematic review foundthat only half of the studies in which theimpact of inpatient asthma protocolswas assessed contained reports of theproportions of patients on any ICS atdischarge. The studies in which ICSprescriptions were reported had widevariability, with the mean percent ofpatients discharged on any medicationranging from 0.54% to 92% inpreintervention data.16 Although somestudies were focused on improving ICS useon discharge,17,18 assessing the accuracy of

these ICS prescriptions has been hamperedby the lack of baseline asthma severity andcontrol assessments.19 Methods to quantifyappropriate controller prescription at timeof discharge by inpatient providers in anenvironment in which ICS prescribing ratesare already high are also not known.

Our purpose with this project was toinvestigate if a rigorous QI interventioncould improve compliance with NAEPPguideline–based classification of asthmacontrol in the inpatient setting, and therebyimprove patient discharge on theappropriate controller medication, not justany controller medication. Our specific aimwith the project was to discharge 75% ofpatients 5 to 18 years of age admitted instatus asthmaticus on an appropriatecontroller medicine as defined by theNAEPP guidelines by 1 year of projectimplementation. An interdisciplinary QI teamcreated a series of interventions centeredaround improving assessments of asthmacontrol in an inpatient setting.

METHODSSetting

Our hospital is a 132-bed, urban, academicchildren’s hospital serving the Bronx, NewYork, with ∼1100 asthma admissionsannually. Patients are primarily of minorityrace and ethnicities (40% African American;30% Hispanic), and 73% are on publicinsurance. The 3 inpatient wards are staffedby teams consisting of a mix of providers,including medical students, physicianassistants, and residents who rotate every2 to 4 weeks, with teams typically consistingof an attending physician, 3 residents, a

physician assistant, and 2 medical students.Seventy-eight percent of asthma patientswere cared for by attending physicians fromthe pediatric hospital medicine division,with the remainder cared for by theadolescent medicine division (19%) andpediatric pulmonary division (3%). Ourhospital had a baseline rate of 88% ofpatients being discharged on any asthmacontroller medication.

Planning the Intervention

An interdisciplinary QI team was assembled,with representatives from key stakeholders:residents, chief residents, physicianassistants, respiratory therapists, nursingleadership, QI leaders, and attendingphysicians from pulmonary, adolescent, andpediatric hospital medicine divisions. Usingthe Model for Improvement and plan-do-study-act cycles,20 the QI team identifiedinterventions to improve documentation ofasthma control questions and standardizecommunications between providers (Fig 1).This team met monthly with multipleinformal contacts in between meetings. Atthe conclusion of the project, a control planwas designed to sustain the gains of theinterventions. There was no outside fundingor other standard institutional and/ordivisional support for this project.

The Intervention

The QI team created a bundle ofinterventions initially focused on improvingdocumentation of inpatients’ asthma controlquestions. This area was targeted because,as per the NAEPP guidelines, withoutknowledge of asthma symptom control andmedication compliance, it is impossible to

Improved documentation of

Use standardized languagein documentation

Weekly feedback

Food incentive forsuccesses

Smartphone application

Simplified flowchart

Resource on computers

Smartphrase and formwith control questions

Increase appropriatecontroller prescriptions for

children admitted withstatus asthmaticus to 75%

by February 2017

asthma severity

Specific Aim Key Drivers Interventions

step of therapy6 control questionsmedication reconciliation

FIGURE 1 Key driver diagram.

128 HOGAN et al

by guest on September 2, 2020www.aappublications.org/newsDownloaded from

discharge a patient on the correctcontroller medication. The symptomquestions from the NAEPP guidelines arelisted in Table 1. These interventions canbe broadly categorized as visual aids,technological aids, and incentives. For avisual aid, we created a simplified flowchartsummarizing the NAEPP control questions,corresponding asthma severity andrecommended controller therapy withage-appropriate dosing (Supplemental Fig 4).To improve access to the control questionsvia technological aids, we encouragedproviders to download a free iPhoneapplication (STAT Asthma NHLBI Guidelines;Austin Physician Productivity, LLC, Austin, TX)and had control questions available on thedesktop of all provider computers to aidin standardizing language. Both of thesetools also aided providers in using theasthma symptom control assessment tochoose the correct controller medication.QI team members oriented medical teamsto the current interventions in personfor the first 6 months of the project.

After a month of education andencouragement around providerassessment of asthma control, the teamidentified provider buy-in as a barrier toparticipation and awareness. Therefore, wecreated a food-based incentive in whichif all patients were discharged on theappropriate controller medication andall patients had all control questionsdocumented in a given week, then the QIteam would bake specialty cookies for thesuccessful team. As part of this incentiveeffort, providers were also shown theirteam’s data via weekly feedback e-mails.Weekly feedback emails highlightedcurrent interventions, progress of theproject thus far, and gave feedback on

the causes of failures from the previousweek’s data.

The final intervention, again withtechnological aids being used, occurredhalfway through the project period when anew electronic medical record (EMR) wasintroduced. The original EMR allowed onlyfor documentation in free text form. Thenew EMR allowed multiple methods fordocumentation of asthma control: free text,a documentation form with “yes/no”clickable dialogues, and a text macrowith selectable text, commonly called a“smartphrase.” Both the documentationform and smartphrase contained theappropriate asthma control questions,which aided in standardizingdocumentation. The smartphrase waseasier to adapt to resident feedback, wasmade accessible via expected phrases(ie, “AsthmaControlQuestions”) and alsomemorable phrases (ie, “Cookie” and“GetThatCookie”), and anecdotally had moreuptake than the clickable form. A monthafter the EMR introduction, we adapted theweekly reminder e-mails to be focusedon encouraging providers to use thesmartphrase because those using the toolwere documenting more questions andgetting more children on the correctmedications.

Methods of Evaluation

All patients 5 to 18 years old who weredischarged from the children’s hospital withan All Patients Refined Diagnosis RelatedGroup code for asthma were considered forinclusion in the study. To approximatelycoincide with the calendar year, thepreintervention period was defined fromJanuary 15, 2015 to January 14, 2016, andthe postintervention period was definedfrom January 15, 2016 to February 1, 2017.The postintervention period was slightlylonger to correspond with the end ofa resident team work cycle. The first5 patients who met inclusion criteria andwere discharged from the inpatient wardseach week were identified by All PatientsRefined Diagnosis Related Group code byusing Looking Glass Clinical Analytics(Streamline Health, Atlanta, GA), and theproject leader (A.H.H.) reviewed theircharts. In the preintervention period, this

review was done retrospectively, and inthe postintervention period, the reviewwas done in real time to give immediatefeedback to providers. Chart reviewincluded all inpatient provider notes for thatadmission, including discharge summaries,to maximize the capture of asthma controlquestion documentation both in thepreintervention and postintervention data.

The primary outcome was the percentageof patients discharged on the correctcontroller medication, as documented in thedischarge summary, on the basis of theirresponses to the NAEPP control questions.To our knowledge, there are nostandardized tools to adjudicate correctcontroller medication prescription outsideof the NAEPP guidelines themselves. Wefound that many adjudication situationswere clear, such as a patient documentedas noncompliant on their controllermedication who is restarted on their homemedication, or a patient newly started on acontroller medication who had persistentsymptoms. However, some adjudicationsituations were less clear, as when few orno control questions were documented anda patient was discharged on their homemedication. For these reasons, we defineda patient as “discharged on the correctcontroller medication” when they had anappropriate change or maintenance incontroller medication on the basis ofdocumentation of control questions in anyprovider note. To evaluate external validityof these adjudications, a local asthmaexpert who is the director of the PediatricAsthma Center (D.R.) conducted a chartreview of 25 charts before the projectbegan, and assessed agreement withadjudications and gave feedback on anydiscrepant charts.

The process measure of the proportion ofpatients with asthma control assessed wasexamined similarly by using chart review ofevery provider note and tracking whetherall 6 of the control questions (symptomfrequency, nighttime awakenings, albuteroluse for symptom control, interference withnormal activity, asthma exacerbationsrequiring orally systemic corticosteroids,and compliance with current controllertherapy, if prescribed) were documented.

TABLE 1 Asthma Control Questions

Symptom Assessment

No. asthma symptoms per wk?

No. nighttime awakenings per mo?

Albuterol use for symptom control per wk?

Degree of interference with normal activity

Asthma exacerbations requiring oral systemiccorticosteroids in the last y?

Compliance with current controller therapy?

HOSPITAL PEDIATRICS Volume 8, Issue 3, March 2018 129

by guest on September 2, 2020www.aappublications.org/newsDownloaded from

Balancing measures included length ofstay, the proportion of patients dischargedafter 1 PM, and the proportion of patientsdischarged on any ICS. At completion of theintervention period, a control plan wasimplemented with assessment frequencydecreased from weekly to monthly, witha plan to increase reminder e-mails ifdocumentation of asthma control questionsdropped below 80%. The Albert EinsteinCollege of Medicine Institutional ReviewBoard approved this study, and it wasdeemed exempt from written consent.

Statistical Analysis

Demographics and balancing measureswere assessed by x2 for proportions andWilcoxon rank sum for nonnormallydistributed data. To track the effect of ourinterventions and give real-time feedback toprovider teams, run charts were created byplotting the median proportion of 5 patientsreviewed each week who were dischargedon the appropriate controller medication orthe median proportion of control questionsthat were documented. Medians werecalculated for the preintervention periodand new medians were calculated in thepostintervention period when a run of 8or more weeks were greater than theprevious mean per Nelson’s rules.21 Wealso conducted an interrupted timeseries analysis using segmented logisticregression to compare pre- versuspostintervention means for both outcomes.22

The utility of standardized documentationwas analyzed by x2. Data were analyzedby using Stata 14.1 (StataCorp, CollegeStation, TX).

RESULTS

There were 240 charts reviewed inthe preintervention and 252 in thepostintervention periods. The number ofcharts was unbalanced because of2 additional weeks in the interventionperiod and because the intervention periodhad more weeks during the summermonths during which the full 5 chartswere available to review. There was 85%agreement between the primary reviewerand the asthma expert on 25 charts, with aCohen’s k of 0.65, which is statisticallysignificant and substantial.23 There waslikely additional improvement after these

25 charts because feedback was given tothe primary reviewer (A.H.H.) after chartreview to increase agreement. Comparingthe pre- versus postintervention periods,there were no significant differencesin age or race of the patients; however,there were significantly more girls andpatients identifying as Hispanic in thepostintervention group (Table 2). Pre-and postintervention groups were notsignificantly different with regards tochronic asthma severity classification, theproportion that were already on an ICS,the proportion admitted to the ICU, or theproportion discharged from the hospitalmedicine, adolescent, or pulmonary teams(Table 2).

The primary outcome of median proportionof patients on appropriate controllertherapy on discharge was 60% in thepreintervention data. There was asignificant shift in the data after March2016, 12 weeks postintervention, leading toa new median of 80% of patients dischargedon the correct medication in thepostintervention period (Fig 2). The processmeasure of proportion of patients with

asthma control severity assessed in thepreintervention period was 43%. Controlassessment improved to a median of 83%,with a shift in the data immediately afterinterventions began. A second shift occurredafter the new EMR implementation and thechange in reminder emails, focusing onusing smartphrases, to 98% of patientsreceiving symptom control assessments(Fig 3). Balancing measures of length of stay(1.67 vs 1.71 days; P 5 .45), proportion ofpatients discharged after 1 PM (66% vs 73%;P 5 .13), and proportion of patientsdischarged on any ICS (85% vs 84%; P 5 .6)were not significantly different comparingthe pre- and postintervention periods.

Time series analysis for the primaryoutcome was statistically significant in theupward break (ie, a change in the absoluteproportion of patients discharged on thecorrect medication in the postinterventionperiod; P 5 .02). The trend was alsostatistically significant (ie, the change inproportion correct per week; P 5 .01) afterthe intervention. The proportion of NAEPPcontrol questions assessed was alsostatistically significant for the upward break

TABLE 2 Pre- and Postintervention Cohort Characteristics

Preintervention Postintervention P

N 240 252 —

Age in y, median (IQR) 8 (6–12) 8 (6–12) .76

Girls, n (%) 126 (53) 103 (41) .01

Hispanic ethnicity, n (%) 44 (21) 68 (30) .04

Race, n (%) .58

African American 104 (48) 111 (46)

White 10 (5) 7 (3)

Other 104 (48) 121 (51)

Severity on admission, n (%) .1

Intermittent 70 (29) 70 (28)

Mild persistent 87 (36) 72 (29)

Moderate persistent 59 (25) 86 (34)

Severe persistent 24 (10) 24 (10)

ICS prescribed before admission, n (%) 153 (64) 166 (66) .76

Admitted to PICU,a n (%) 46 (19) 49 (19) .94

Discharging team, n (%) .42

Hospital medicine 188 (78) 209 (83)

Adolescent medicine 45 (19) 38 (15)

Pulmonary medicine 7 (3) 5 (2)

IQR, interquartile range; —, not applicable.a Admitted directly to PICU from emergency department or outside hospital.

130 HOGAN et al

by guest on September 2, 2020www.aappublications.org/newsDownloaded from

(P , .001) and trend (P 5 .01) in thepostintervention period.

Use of standardized documentationtemplates increased from 40% pre-EMRrollout to 60% post-EMR rollout (P , .001).Those using either the smartphrase ordocumentation form in the post-EMR periodwere significantly more likely to dischargepatients on the correct medication (24% vs77%; P 5 .02) and document all asthmacontrol questions (8% vs 91%; P , .001).

Most patients adjudicated as dischargedon the correct medication were restartedon the controller therapies they werenoncompliant with (48% preintervention,43% postintervention), newly started on acontroller medication (32% preintervention,20% postintervention), or were steppedup on their current controller therapy(12% preintervention, 17% postintervention)(Table 3). Conversely, most patients(78% preintervention, 58% postintervention)

adjudicated as discharged on an incorrectcontroller medication had inadequatemedical documentation to justify themedication (Table 4).

DISCUSSION

An interdisciplinary QI team successfullyincreased the median proportion of patientsdischarged on the correct asthma controllermedication from 60% to 80%, surpassing the75% aim of our project. By engaging withfrontline providers and incorporating theirfeedback, we were able to createinterventions that directly affected our keydrivers and made a significant change in ourasthma care. To our knowledge, this is thefirst project to be focused on andsuccessfully and consistently improvedischarging of pediatric patients on thecorrect controller medication, not just on anycontroller medication.

Multiple interventions and driverscontributed to this project’s success. Ourmain driver was improving assessmentof asthma control through documentationof the NAEPP control questions. Althoughpocket cards have been shown to bemoderately effective in improving ICSprescriptions,10 our frontline providersrejected having another item on their badgeor in their pocket. Our solution used atool already in most of their pockets (asmartphone) and facilitated understandingvia a flowchart in the resident workroomsand on every computer. Our second mostimportant driver was standardizingcommunication around asthma care. Beforethese interventions, triggers for a patient’scurrent exacerbation and their exposures,such as secondhand smoke, animals, andcarpets, were almost always documented.Unfortunately, the documentation ofNAEPP asthma control questions (the keycomponent to decide the appropriatecontroller prescription) was present in only43% of the patients. By making thesequestions accessible on every computer,we had moderate improvement. The EMRintervention, which allowed for easy toremember smartphrases, markedlyimproved documentation of controlquestions and thereby the percentage ofchildren receiving the correct controllermedication at discharge. Although some of

FIGURE 2 Run chart of primary outcome. The proportion of patients discharged from thehospital on correct medication of the 5 charts reviewed each week is shown.

FIGURE 3 Run chart of primary process measure. The proportion of asthma control questionsdocumented each week of the 5 charts reviewed is shown.

HOSPITAL PEDIATRICS Volume 8, Issue 3, March 2018 131

by guest on September 2, 2020www.aappublications.org/newsDownloaded from

these interventions were institution specific(ie, those without smartphrases in theirEMR could not use this intervention), mostwere either low cost (food incentives, e-mailreminders) or are provided here asSupplemental Fig 4. The next steps of thisproject are to sustain the improvements wehave made locally and disseminate ourwork to other institutions throughpresentations and this publication.

Our findings are consistent with otherstudies that have revealed that guidelinecompliance can be improved. The authors ofa recent systemic review reported that theauthors of 15 studies used decision supportand found up to a 34% increase inprescription of any ICS medicines after

interventions. Multicomponent interventionsusing combinations of feedback, decisionsupport, and other techniquesdemonstrated significant increases inprescribing of ICS by 25% to 49% but thesuccess varied between the studies.10

Whereas the authors of 3 studies foundbetween 25% and 49% increases in ICSprescriptions, the authors of another4 studies did not find any difference inprescription rates.10 Because our baselinefrequency of ICS prescription was already85%, with 95% of persistent asthmaticson any controller, it was not our goal toincrease rates but rather to improveaccuracy of those and other prescriptionsfor controller medications. As a balancing

measure, we tracked ICS prescribing ratesbecause we did not want to inadvertentlydecrease our high ICS rates by accidentallydiscouraging ICS in favor of alternatecontrollers, such as leukotriene receptorantagonists. We did not find a significantdifference in ICS rates (P 5 .61), orprescription of any controller (P 5 .95).

Assessing accuracy of controllermedications is difficult because of a lackof standardized metrics. The NationalCommittee for Quality Assurance identifiesappropriate medications for children withasthma as those with persistent asthmaprescribed any controller medication.24

Using this metric, we found that 98% of ourpatients were “appropriate” before the startof the project, which makes this metric ofquestionable utility for our institution. Intheir review of quality metrics, Nkoy et al19

determined that there is evidence level“A” for use of “proportion of patientsdischarged with proper controllermedications, according to their chronicasthma severity classification.”10 However,the metric was rejected by the study’sauthors because “data on chronic asthmaseverity…were rarely available in themedical record and therefore were notfeasible to obtain.”19 We have shown that aninterdisciplinary QI team can increase theasthma severity data needed to make thisintegral quality metric feasible to evaluate.

There are limitations to our project. Our QIteam did not include a family member, a keystakeholder. This was an oversight, and weencourage all QI teams to include patientand family representatives on QI projects.We made the adjudication of controllermedication accuracy feasible; thedetermination of the primary outcome wasmade by 1 reviewer who was not blinded tothe study’s time period. Although most ofthe determinations (85% preintervention,76% postintervention) were not subjective(eg, a patient who was not compliant withICS was restarted on the same medication),there were a proportion of determinationsthat could have been subjective (15%preintervention, 23% postintervention).Tables 3 and 4 enumerate the subjective andnot subjective determinations used in ouranalysis. We attempted to mitigate concerns

TABLE 3 Description of the Causes of Patients Being Classified as “Discharged on CorrectController Therapy”

Preintervention Postintervention

N 127 212

Noncompliant with controller; restarted homecontroller,a n (%)

61 (48) 90 (43)

Some documentation 58 (95) 25 (28)

All questions documented 3 (5) 65 (72)

Poorly controlled on no controller; controllerstarted,a n (%)

40 (32) 42 (20)

Some documentation 38 (95) 9 (21)

All questions documented 2 (5) 33 (79)

Poorly controlled, compliant with controller;therapy stepped up,a n (%)

15 (12) 35 (17)

Some documentation 14 (93) 10 (29)

All questions documented 1 (7) 25 (71)

Well controlled apart from current illness;medication correctly not advanced,b n (%)

4 (3) 35 (17)

Some documentation 1 (25) 3 (9)

All questions documented 3 (75) 32 (91)

Home therapy increased in last 2 mo; controllercorrectly not advanced,a n (%)

4 (3) 4 (2)

Some documentation 4 (100) 1 (25)

All questions documented — 3 (75)

Noncompliant, but controller advanced anywaybecause of severe presentation,b n (%)

3 (2) 3 (1)

Some documentation 2 (67) 1 (33)

All questions documented 1 (33) 2 (67)

Poorly controlled, attempted to increase therapybut parents refused,a n (%)

0 (0) 3 (1)

Some documentation — 1 (33)

All questions documented — 2 (67)

—, not applicable.a Considered a “nonsubjective” assessment.b Considered a “subjective” assessment.

132 HOGAN et al

by guest on September 2, 2020www.aappublications.org/newsDownloaded from

of adjudication subjectivity by calculating aninterrater reliability between the primaryrater and an asthma expert. Anotherlimitation is that our primary outcomewas not a patient outcome, such asreadmissions or recurrent needs for steroiduse. These outcomes were beyond the scopeof this inpatient-focused study, and thelow rate of some of these patient outcomesnecessitates multicenter studies to evaluatethem. In addition, we did not have accessto the insurance claims data that wouldmake the evaluation of readmission at anyhospital or prescription fill rates possible.However, given the overwhelming evidenceof the benefits of ICS and guidelineadherence, we believe our primary outcomeis justified.2–4,25–27 Finally, because this is apre-post outcome study, we cannot accountfor temporal trends that may haveinfluenced the change in our outcomes,although our interrupted time-seriesanalysis tested for this and revealed theinterventions themselves had a significanteffect on our primary outcome.

CONCLUSIONS

An interdisciplinary QI team cansuccessfully improve the accuracy ofasthma controller therapy for children withacute asthma exacerbations on dischargeby improving assessment of asthma control.As documentation can be improved to make

accuracy determinations feasible,standardized tools must be developed toassure the validity of this metric for qualitytracking.

Acknowledgments

The authors acknowledge all the residents,chief residents, physician assistants, andattending physicians who participated inthis project, as well as the leadership of theChildren’s Hospital at Montefiore.

REFERENCES

1. Calpin C, Macarthur C, Stephens D,Feldman W, Parkin PC. Effectiveness ofprophylactic inhaled steroids inchildhood asthma: a systemic review ofthe literature. J Allergy Clin Immunol.1997;100(4):452–457

2. Rachelefsky G. Inhaled corticosteroidsand asthma control in children:assessing impairment and risk.Pediatrics. 2009;123(1):353–366

3. Suissa S, Ernst P. Inhaledcorticosteroids: impact on asthmamorbidity and mortality. J Allergy ClinImmunol. 2001;107(6):937–944

4. National Heart, Lung and Blood Institute;National Asthma Education andPrevention Program. Expert panel report3: guidelines for the diagnosis andmanagement of asthma. 2007. Available

at: https://www.nhlbi.nih.gov/files/docs/guidelines/asthgdln.pdf. AccessedJanuary 4, 2018

5. Cloutier MM, Hall CB, Wakefield DB,Bailit H, . Use of asthma guidelines byprimary care providers to reducehospitalizations and emergencydepartment visits in poor, minority,urban children. J Pediatr. 2005;146(5):591–597

6. Ting S. Multicolored simplified asthmaguideline reminder (MSAGR) for betteradherence to national/global asthmaguidelines. Ann Allergy Asthma Immunol.2002;88(3):326–330

7. Rastogi D, Shetty A, Neugebauer R,Harijith A. National Heart, Lung, andBlood Institute guidelines and asthmamanagement practices among inner-citypediatric primary care providers. Chest.2006;129(3):619–623

8. Bunik M, Federico MJ, Beaty B, Rannie M,Olin JT, Kempe A. Quality improvementfor asthma care within a hospital-basedteaching clinic. Acad Pediatr. 2011;11(1):58–65

9. Cabana MD, Bruckman D, Meister K,Bradley JF, Clark N. Documentation ofasthma severity in pediatric outpatientclinics. Clin Pediatr (Phila). 2003;42(2):121–125

10. Okelo SO, Butz AM, Sharma R, et al.Interventions to modify health careprovider adherence to asthmaguidelines: a systematic review.Pediatrics. 2013;132(3):517–534

11. Wisnivesky JP, Lorenzo J, Lyn-Cook R,et al. Barriers to adherence toasthma management guidelines amonginner-city primary care providers. AnnAllergy Asthma Immunol. 2008;101(3):264–270

12. Gold J, Reyes-Gastelum D, Turner J,Davies HD; Pediatric Asthma QualityTeam. A quality improvement study usingfishbone analysis and an electronicmedical records intervention to improvecare for children with asthma. Am J MedQual. 2014;29(1):70–77

13. Morse RB, Hall M, Fieldston ES, et al.Hospital-level compliance with asthma

TABLE 4 Description of the Causes of Patients Being Classified as “Discharged on IncorrectController Therapy”

Preintervention Postintervention

N 113 40

Inadequate control question documentation tojustify discharge medication,a,b n (%)

88 (78) 23 (58)

No documentation 21 (24) 0 (0)

Some documentation 67 (76) 23 (100)

Poorly controlled, compliant with controller; failedto step up therapy,b,c n (%)

15 (13) 13 (33)

Some documentation 14 (93) 3 (23)

All questions documented 1 (7) 10 (77)

Poorly controlled, compliant with controller;stepped up controller more than 2 steps,c,d n (%)

10 (9) 4 (10)

Some documentation 9 (90) 1 (25)

All questions documented 1 (10) 3 (75)

a By definition, 0 patients had all questions documented.b Considered a “nonsubjective” assessment.c Patients had at least some control questions.d Considered a “subjective” assessment.

HOSPITAL PEDIATRICS Volume 8, Issue 3, March 2018 133

by guest on September 2, 2020www.aappublications.org/newsDownloaded from

care quality measures at children’shospitals and subsequent asthma-related outcomes. JAMA. 2011;306(13):1454–1460

14. Rutman L, Atkins RC, Migita R, et al.Modification of an established pediatricasthma pathway improves evidence-based, efficient care. Pediatrics. 2016;138(6):e20161248

15. Fassl BA, Nkoy FL, Stone BL, et al. TheJoint Commission Children’s AsthmaCare quality measures and asthmareadmissions. Pediatrics. 2012;130(3):482–491

16. Dexheimer JW, Borycki EM, Chiu KW,Johnson KB, Aronsky D. A systematicreview of the implementation andimpact of asthma protocols. BMC MedInform Decis Mak. 2014;14:82

17. Edwards E, Fox K. A retrospective studyevaluating the effectiveness of anasthma clinical pathway in pediatricinpatient practice. J Pediatr PharmacolTher. 2008;13(4):233–241

18. Kelly CS, Andersen CL, Pestian JP, et al.Improved outcomes for hospitalizedasthmatic children using a clinical

pathway. Ann Allergy Asthma Immunol.2000;84(5):509–516

19. Nkoy FL, Fassl BA, Simon TD, et al. Qualityof care for children hospitalized withasthma. Pediatrics. 2008;122(5):1055–1063

20. Langley GJ, Nolan KM, Nolan TW, NormanCL, Provost LP. The Improvement Guide: APractical Approach to EnhancingOrganizational Performance. SanFrancisco, CA: Jossey-Bass; 1996

21. Nelson L. Technical aids. J Qual Technol.1984;16(4):238–239

22. Wagner AK, Soumerai SB, Zhang F, Ross-Degnan D. Segmented regressionanalysis of interrupted time seriesstudies in medication use research.J Clin Pharm Ther. 2002;27(4):299–309

23. Landis JR, Koch GG. The measurement ofobserver agreement for categoricaldata. Biometrics. 1977;33(1):159–174

24. National Quality MeasuresClearinghouse. Use of appropriatemedications for people with asthma:percentage of patients 5 to 64 years ofage during the measurement year who

were identified as having persistentasthma and who were appropriatelydispensed medication during themeasurement year. Available at: https://qualitymeasures.ahrq.gov/summaries/summary/48818/use-of-appropriate-medications-for-people-with-asthma-percentage-of-patients-5-to-64-years-of-age-during-the-measurement-year-who-were-identified-as-having-persistent-asthma-and-who-were-appropriately-dispensed-medication-during-the-measurement-year?q5Education.Accessed January 4, 2018

25. Allen DB. Systemic effects of inhaledcorticosteroids in children. Curr OpinPediatr. 2004;16(4):440–444

26. Global Initiative for Asthma. Globalstrategy for asthma management andprevention. 2016. Available at: http://ginasthma.org/gina-reports/. AccessedAugust 10, 2016

27. Blais L, Ernst P, Boivin JF, Suissa S.Inhaled corticosteroids and theprevention of readmission to hospitalfor asthma. Am J Respir Crit Care Med.1998;158(1):126–132

134 HOGAN et al

by guest on September 2, 2020www.aappublications.org/newsDownloaded from

DOI: 10.1542/hpeds.2017-0184 originally published online February 13, 2018; 2018;8;127Hospital Pediatrics 

Alexander H. Hogan, Deepa Rastogi and Michael L. RinkeController Accuracy

A Quality Improvement Intervention to Improve Inpatient Pediatric Asthma

ServicesUpdated Information &

http://hosppeds.aappublications.org/content/8/3/127including high resolution figures, can be found at:

Supplementary Material

2017-0184.DCSupplementalhttp://hosppeds.aappublications.org/content/suppl/2018/02/09/hpeds.Supplementary material can be found at:

Referenceshttp://hosppeds.aappublications.org/content/8/3/127#BIBLThis article cites 23 articles, 5 of which you can access for free at:

Subspecialty Collections

rovement_subhttp://www.hosppeds.aappublications.org/cgi/collection/quality_impQuality Improvementhttp://www.hosppeds.aappublications.org/cgi/collection/asthma_subAsthmaunology_subhttp://www.hosppeds.aappublications.org/cgi/collection/allergy:immAllergy/Immunologyon:practice_management_subhttp://www.hosppeds.aappublications.org/cgi/collection/administratiAdministration/Practice Managementfollowing collection(s): This article, along with others on similar topics, appears in the

Permissions & Licensing

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

by guest on September 2, 2020www.aappublications.org/newsDownloaded from

DOI: 10.1542/hpeds.2017-0184 originally published online February 13, 2018; 2018;8;127Hospital Pediatrics 

Alexander H. Hogan, Deepa Rastogi and Michael L. RinkeController Accuracy

A Quality Improvement Intervention to Improve Inpatient Pediatric Asthma

http://hosppeds.aappublications.org/content/8/3/127located 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/2018/02/09/hpeds.2017-0184.DCSupplementalData Supplement at:

Print ISSN: 1073-0397. Illinois, 60143. Copyright © 2018 by the American Academy of Pediatrics. All rights reserved. published, and trademarked by the American Academy of Pediatrics, 345 Park Avenue, Itasca,publication, it has been published continuously since 1948. Hospital Pediatrics is owned, Hospital Pediatrics is the official journal of the American Academy of Pediatrics. A monthly

by guest on September 2, 2020www.aappublications.org/newsDownloaded from