RESEARCH IN PEDIATRIC EDUCATION

Asthma Treatment Decisions by Pediatric Residents

Do Not Consistently Conform to Guidelines or Improve

With Level of TrainingSande O. Okelo, MD, PhD; George K. Siberry, MD, MPH; Barry S. Solomon, MD, MPH;Andrew L. Bilderback, MS; Michiyo Yamazaki, MHS, PhD; Theresa Hetzler, MD;Cynthia L. Ferrell, MD, MSEd; Nui Dhepyasuwan, MEd; Janet R. Serwint, MD; for theCORNET Investigators

From the Division of Pediatric Pulmonology, The David Geffen School of Medicine at UCLA, Los Angeles, Calif (Dr Okelo); Pediatric,Adolescent, and Maternal AIDS (PAMA) Branch, Eunice Kennedy Shriver National Institute of Child Health and Human Development, NationalInstitutes of Health, Bethesda, Md (Dr Siberry); Division of General Pediatrics & Adolescent Medicine, The Johns Hopkins University School ofMedicine, Baltimore,Md (Drs Solomon andSerwint); Department ofMedicine, The JohnsHopkinsUniversity School of Medicine, Baltimore,Md(Mr Bilderback); Department of Family, Population andReproductive Health, The JohnsHopkins Bloomberg School of Public Health, Baltimore,Md (Dr Yamazaki); Department of Pediatrics, New York Medical College, Valhalla, NY (Dr Hetzler); Department of Pediatrics, Oregon Health &Science University, Portland, Ore (Dr Ferrell); and Academic Pediatric Association, McLean, Va (Ms Dhepyasuwan)The authors declare that they have no conflict of interest.Address correspondence to Sande O. Okelo, MD, PhD, Division of Pediatric Pulmonology, Department of Pediatrics, David GeffenSchool of Medicine at UCLA, 10833 Le Conte Ave, MDCC 3878, Los Angeles, CA 90095 (e-mail: sokelo@mednet.ucla.edu).Received for publication November 9, 2012; accepted December 17, 2013.

ABSTRACT

AC

OBJECTIVE: To compare asthma treatment decisions by pedi-atric residents to current asthma guidelines and to learn whethertreatment decisions vary by postgraduate year in training.METHODS: We conducted a Web-based survey of residentsfrom 10 training programs through the Continuity ResearchNetwork of the Academic Pediatric Association (CORNET).Surveys included 6 vignettes of patients receiving low-doseinhaled steroids with guideline- and non-guideline-based indi-cators of asthma status and 1 stable patient on high-intensitymedication.RESULTS: There were 369 resident respondents (65% responserate), 26% postgraduate year (PGY) 1, 38% PGY2, and 36%PGY3þ. Seventy-five percent of each resident group reportedseeing fewer than 1 asthma patient per continuity clinic session.A majority of residents made appropriate treatment recommen-dations in 2 of 4 vignettes of guideline-based indicators ofasthma status: first, 97% overall stepping up treatment formild persistent asthma; and second, 52% overall steppingdown treatment for a patient with well-controlled asthma onhigh-intensity medications. Inconsistent with guideline recom-mendations, 82% of residents overall did not step down treat-

CADEMIC PEDIATRICSopyright ª 2014 by Academic Pediatric Association

287

ment for a patient with well-controlled asthma receiving low-intensity therapy; 75% of residents did not step up treatmentfor a patient with a recent hospitalization for asthma. Of the 3vignettes evaluating non-guideline-based indicators of asthmastatus, a majority of residents (60%) stepped up treatment forparental reports of worse asthma, while a minority did so fora parental report of being bothered by their child’s asthma(27%) or when wheezing was reported at physical examination(43%). Therewere no statistically significant differences for anyof the comparisons by year in training.CONCLUSIONS: Pediatric residents’ management of asthma isconsistent with national guidelines in some cases but not inothers. There were no differences in the outpatient asthma man-agement decisions between residents by years in training.Educational efforts should be focused on strategies to facilitatepediatric resident adherence to national asthma guideline rec-ommendations for outpatient asthma management.

KEYWORDS: asthma; CORNET; decision making; pediatricresident education; survey; treatment; vignettes

ACADEMIC PEDIATRICS 2014;14:287–293

1

WHAT’S NEW

Pediatric residents infrequently manage asthma in theoutpatient setting. They vary little by level of trainingin making asthma treatment decisions. Efforts shouldfocus on strategies to facilitate improvement in residentasthma management.

DESPITE THE AVAILABILITY of National Institutes ofHealth asthma guidelines since 1991, many children

continue to experience poor asthma care, particularlyunderuse of anti-inflammatory medications. Suboptimalcare may stem from the persistence of barriers precludingphysician adherence to guidelines (eg, lack of awarenessof specific recommendations).2,3 Furthermore, physiciansmay use clinical criteria not incorporated into asthmaguideline algorithms, considering factors other than asthmacontrol in their treatment decisions.4,5 Last, suboptimalasthma care may also be the result of inadequatepreparation of providers during their training. Examinations

Volume 14, Number 3May–June 2014

288 OKELO ET AL ACADEMIC PEDIATRICS

of asthma care quality have tended to focus on providerscurrently in practice and not pediatric residents.6

The assessment and management practices of pediatricresidents are important to consider because many residentscare for inner-city, poor, and/or minority children who areat the highest risk for poor asthma care and poor asthmacontrol.7 Evaluating resident practices is essential fordetermining whether we are training residents effectivelyand emphasizing the correct information for their futureroles in the pediatric workforce. Ideally, residents areexposed to the most up-to-date asthma information andtrained to practice using an evidence-based approach. Ifit is evident that residents display poor asthma assessmentand management behaviors, then interventions to improveasthma care should target training environments.

There is little information about the quality of asthmacare delivered by pediatric residents and what their knowl-edge base is regarding the assessment and treatment ofasthma. Ozuah and Reznik examined pediatric residents’abilities to classify asthma severity correctly using stan-dardized patients and found no difference by year intraining.8 However, this study did not assess asthma man-agement behaviors, which would provide a deeper under-standing of how those in training provide asthma care.Furthermore, an implicit assumption in residency trainingis that those with more experience (eg, senior residents)have more expertise in disease management than more ju-nior residents (eg, interns). However, this volume–compe-tence relationship has not been examined in asthma care.

We sought to further study these issues through the Con-tinuity Research Network of the Academic Pediatric Asso-ciation (CORNET) network. CORNET is a nationalpractice-based research network of pediatric resident con-tinuity clinic practices.9 This network is endorsed by theAcademic Pediatric Association, and its research goalsinclude health care issues of minority and underserved pa-tients, health care disparities, and resident education (eg,the continuity practice of residents). The evaluation ofpractice behaviors is important because pediatric residentsconstitute our future workforce, and one of the purposes ofthe continuity experience is to prepare residents to serve asprimary care providers. The CORNET network providesan opportunity to learn about asthma treatment from adiverse and representative spectrum of residents (eg, by ge-ography, residency size, and setting). Compared to those inprivate practice, CORNET practices provide health care toa larger proportion of low-income, African American, andMedicaid-insured patients.9

The objectives of this study among a national sample ofpediatric residents were to compare asthma treatment deci-sions by pediatric residents to current asthma guidelines; tolearn whether non-guideline-based clinical criteria areinfluential in treatment decision making; and to determinewhether asthma management practices vary by postgrad-uate year of training. Our goal was to study a large repre-sentative sample of pediatric residents to enhance thegeneralizability of our findings. We hypothesized that post-graduate year (PGY) 3 residents would be more likely torecommend appropriate treatment than PGY1 residents.

SUBJECTS AND METHODS

PROCEDURES

A cross-sectional survey of pediatric residents was con-ducted betweenMay 2008 and July 2008. Participation waslimited to those who were residents at the time the studystarted (May 2008). An e-mail with a Web-based link toan online survey was sent to prospective residents whosecontinuity clinic practices were enrolled in the CORNET.Completion of the survey served as consent to participatein the study. Survey responses were anonymous. Institu-tional review board approval was obtained from each clin-ical site that participated.

STUDY POPULATION

Categorical pediatric residents or residents in combinedmedicine–pediatric programs were eligible if their continu-ity site was enrolled in CORNETand the resident agreed toparticipate in the study. At the time of this study, CORNETincluded 77 pediatric training programs in 95 enrolled clin-ical practice sites, representing all regions of the UnitedStates. An invitation to participate was e-mailed to the con-tinuity clinic directors of all 77 CORNET programs. De-mographic and practice characteristics, including the yearinto residency training (PGY1, PGY2, and PGY3þ),were collected.

SURVEY

The survey was designed to evaluate asthma assessmentand management behaviors through clinical vignettes rep-resenting ambulatory experiences specific to asthma. Stan-dardized vignettes of patients aged between 5 and 10 yearsreturning for a 3-month follow-up clinical visit were usedas a means of evaluating pediatric resident treatment prac-tices. Seven vignettes were presented in the same order forall respondents, each with a specific clinical factor(Table 1):1. No risk factors and receiving a low-intensity regimen of

fluticasone 44 mg (2 puffs twice daily). This was createdto provide a comparison for the influence of other risk fac-tors on resident recommendations to step up treatment.

2. Wheeze (presence of faint wheeze at physical examina-tion): “Good air movement, with faint wheeze.”

3. Acute care: “Patient was hospitalized for asthma 6months ago.”

4. Bother (parental report of being bothered by the child’sasthma).

5. Poor asthma control (wheeze and albuterol use 4 to 5days per week): “Wheezing 4 to 5 days/week; albuteroluse 4 to 5 days/week.”

6. Direction (subjective parental report of child’s asthmabeing “worse” compared to the last clinic visit 3 monthsearlier): “Symptoms worse since last visit.”

7. No risk factors and receiving a high-intensity regimenof fluticasone 220 mg (2 puffs twice daily); long-acting inhaled b-agonist and leukotriene modifier.This was created to provide a comparison for the influ-ence of other risk factors on resident recommendationsto step down treatment.

Table 1. Concept Map of Vignettes

Vignette No. Treatment* Acute Care Bother Control Direction Wheeze Anticipated Treatment Decision†

1‡ Low intensity No No Sx once/2 wk Unchanged No Step down2 Low intensity No No Sx once/2 wk Unchanged Yes Step up3 Low intensity Yes No Sx once/2 wk Unchanged No Step up4 Low intensity No Yes Sx once/2 wk Unchanged No Step up5 Low intensity No No Sx 4–5 times/wk Unchanged No Step up6 Low intensity No No Sx once/2 wk Worse No Step up7§ High intensity No No Sx once/2 wk Unchanged No Step down

Italic areas indicate where there were variations by vignette (eg, wheezing 1 time in the past 2 weeks vs wheezing 4 to 5 times per week).

Sx ¼ symptoms of asthma.

*Low-intensity treatment was 2 puffs twice a day of fluticasone 44 mg and as-needed albuterol prn. High-intensity treatment was 2 puffs twice

a day of fluticasone 220 mg with long-acting inhaled b agonist, leukotriene modifier, and as-needed albuterol.

†These are the anticipated treatment recommendations of the resident respondents, which are based on asthma guideline recommenda-

tions (vignettes 1, 3, 5, and 7) or on focus group feedback from patients and physicians regarding information deemed important to determine

appropriate asthma treatment (vignettes 2, 4, and 6).

‡Vignette provided as a baseline comparison for stepping up treatment. Analyses compared vignettes 2, 3, 4, 5, and 6 to vignette 1.

§Vignette 7 was created to evaluate decisions for stepping down treatment.

ACADEMIC PEDIATRICS ASTHMA TREATMENT DECISIONS 289

The clinical factors included in each vignette weredeveloped from previous work by the first author in 2003at Johns Hopkins University and Howard University10,11

and involved recruiting focus groups of patients, primarycare physicians, and asthma specialists to learn whatfactors are important to incorporate into asthmatreatment decision making.5,10,11 From these focusgroups emerged some clinical factors reflective ofexisting asthma guidelines (acute asthma care; poorasthma control) but also some clinical factors notincluded in existing asthma guidelines (direction, bother,risk; the focus groups were held before the 2007 EPR-3guidelines, which is when the concept of risk was intro-duced into the guidelines). Case vignettes of patientswith asthma were constructed to incorporate each of theclinical factors that emerged from the focus groups. (Theonline Appendix contains all vignettes used in thisstudy.) A survey of practicing pediatricians, family practi-tioners, and pulmonologists revealed that these conceptswere influential in treatment decision making.5,12 Thecase vignettes were reviewed by local physicians forclarity and content.

Three of the clinical factors (poor asthma control, acutecare, wheeze) directly map onto guideline concepts ofimpairment and risk, while 2 other clinical factors werenot guideline based (direction, bother). Patients weretreated with fluticasone (44 mg 2 puffs twice daily) exceptfor 1 vignette in which fluticasone 220 mg with long-actinginhaled b-agonist 2 puffs twice daily and a leukotrienemodifier was used (vignette 7). For each vignette, residentswere asked if they would step up medications, step downmedications, or leave medications unchanged (Figure;Table 1). No restrictions were placed on resident respon-dents regarding sources of information used to answerthe survey.

Vignette 1 represented a baseline comparison forstepping up treatment (well-controlled symptoms, low-intensity therapy, and no risk factors), while vignette7 represented a reference comparison for steppingdown treatment (similar patient without risk factors onhigh-intensity therapy). On the basis of current asthmaguideline recommendations, we expected residents to

step up treatment when confronted with vignettes 3and 5, and to step down treatment when confrontedwith the vignettes without any risk factors (vignettes 1and 7).To demonstrate a 20% difference in treatment recom-

mendations between residents of 2 different years intraining, with a ¼ 0.05 and power ¼ 0.80, we estimatedthat we needed a sample size of 91 resident respondents,per class or a total of 273 residents.

ANALYSIS

Means and proportions were calculated for descriptivecharacteristics. A chi-square test for trend was used todetermine whether descriptive characteristics and treat-ment recommendations varied significantly by residentyear in training. We then used logistic regression toexamine the effect of each clinical factor on the oddsof residents stepping up or stepping down treatment.The responses of the PGY1 group to vignettes 1 and 7served as the reference groups (to step up and stepdown treatment, respectively) for the logistic regressionanalyses. Variability in multiple responses from a singlerespondent was taken into account to estimate the oddsusing robust variance estimation. P values of < .05were considered statistically significant. Analyses wereperformed by Stata 11 software (StataCorp, College Sta-tion, Tex).

RESULTS

RESPONDENT CHARACTERISTICS

Of the 568 surveys sent to pediatric residents in the 10participating programs, we received 367 complete re-sponses (overall 65% response rate). Three hundredtwenty-seven (89%) were categorical pediatric residents,and 40 (11%) respondents were internal medicine–pediat-ric residents. Year in training was unrelated to demo-graphic and practice characteristics except for self-ratedasthma experience (Table 2), where PGY1 residents wereless likely than PGY2 or PGY3þ residents to report mod-erate or extensive asthma experience (P < .001). Weobserved a trend suggesting that PGY2 (21.4%) and

Figure. Vignette representing floor for comparison of studied factors. Bold and underlined areas highlight where there were variations by

vignette (eg, wheezing 1 time in the past 2 weeks vs wheezing 4 to 5 times per week).

290 OKELO ET AL ACADEMIC PEDIATRICS

PGY3þ (25.2%) residents were more likely to report hav-ing at least 1 asthma patient in every continuity clinic ses-sion compared to PGY1 (12.5%) residents (P ¼ .06)(Table 3).

TREATMENT DECISIONS BASED ON GUIDELINE CRITERIA

Amajority of PGY1, PGY2, and PGY3þ residents indi-cated that they would step up treatment for poor asthmacontrol (frequent symptoms and albuterol use) (97%,96%, and 98%, respectively; P ¼ .7). Of the remaining vi-gnettes where guidelines suggested stepping up treatment,a minority of residents did so, regardless of year in training.Specifically, the recommendations to stepping up treatmentamong the PGY1, PGY2, and PGY3þ residents amongthese remaining vignettes included wheeze at physical ex-amination (45%, 41%, 44%) and recent hospitalization(25%, 25%, 26%; P > .5 for both comparisons).

In terms of scenarios in which guidelines suggested res-idents should step down treatment, a slight majority of res-idents overall (52%) recommended this type of treatmentfor the high-intensity medication vignette without anyother clinical factors (53%, 56%, 48%), while 18% of res-idents overall did so for the low-intensity medicationvignette (18%, 13%, 22%; P > .5 for all comparisons).

TREATMENT DECISIONS BASED ON NON-GUIDELINE-BASED

CRITERIA

Amajority of PGY1, PGY2, and PGY3þ residents indi-cated that they would step up treatment when parents re-ported that the child’s asthma was worse than at the lastvisit (63%, 58%, and 59%, respectively, P ¼ .6). For theother non-guideline-based vignette, a minority of residentsrecommended stepping up treatment based on parentalreport of being bothered by the child’s asthma (28%,26%, 27%).

LOGISTIC REGRESSION

Logistic regression models revealed that there were nosignificant differences in treatment recommendations byyear in training of the resident respondent. In particular,second- and third-year residents were as likely as first-year residents to step up treatment for a patient with arecent hospitalization, a parental report of being botheredby the child’s asthma, a parental report of worse asthma,wheezing at physical examination, and uncontrolled symp-toms (P > .05 for the odds of PGY3þ vs PGY1 and forPGY2 vs PGY1 to increase treatment). Similarly, second-and third-year residents were as likely as first-yearresidents to step down treatment for a patient with

Table 2. CORNET Respondents’ Demographic Characteristics by Year in Training

Characteristic Variable

% of CORNET Respondents

P ValueAll (N ¼ 367) PGY1 (n ¼ 96) PGY2 (n ¼ 140) PGY3þ (n ¼ 131)

Gender Female 76.2 72.9 79.3 75.4 NSAge, y, mean (SD) 29.7 (2.8) 29.0 (2.7) 29.4 (2.4) 30.6 (3.0) <.001Race/ethnicity White 69.5 67.4 70.8 69.8 NS

Black 7.2 7.4 5.1 9.3Asian 16.6 16.8 16.8 16.3Hispanic 2.8 3.2 2.9 2.3Other 3.9 5.3 4.4 2.3

Area of practice Inner city 52.5 45.3 52.1 58.1 NSUrban 32.7 37.9 31.4 30.2Suburban 12.4 13.7 14.3 9.3Rural 1.4 1.1 1.4 1.6Other 1.1 2.1 0.7 0.8

Self-rated asthma experience Limited 38.4 66.7 38.6 17.6 <.001Moderate 55.3 32.3 57.1 70.2Extensive 6.3 1.0 4.3 12.2

Region of United States Northeast 33.0 35.4 32.9 31.3 NSSouth 36.0 36.5 32.9 38.9Midwest 26.2 19.8 28.6 28.2West 4.9 8.3 5.7 1.5

CORNET ¼ Continuity Research Network of the Academic Pediatric Association; PGY ¼ postgraduate year.

ACADEMIC PEDIATRICS ASTHMA TREATMENT DECISIONS 291

well-controlled asthma and high-intensity treatment orlow-intensity treatment (P > .05 for the odds of PGY3þvs PGY1 and for PGY2 vs PGY1 to increase treatment).

DISCUSSION

In this vignette-based survey of a national sample ofpediatric residents, we observed the following: 1) a major-ity of residents recommended treatment in accordancewithasthma guidelines for 2 of the 4 guideline-based vignettes(97% stepping up treatment for mild persistent symptoms;52% stepping down treatment for well-controlled symp-toms); 2) a majority of residents recommended treatmentfor only 1 of 3 focus group based (nonguideline) indicatorsof asthma morbidity (60% stepping up for worse asthma);3) there were no differences by year in training for deci-sions to adjust asthma therapy; and 4) residents weremore inclined to step up treatment than to step down treat-ment. These findings suggest that pediatric residentsmay only sometimes incorporate guideline- and non-guideline-based criteria into their treatment decision mak-ing for asthma.

Stepping down treatment of patients with well-controlled asthma is another concept that we tested in 2 vi-gnettes. Interestingly, a majority of residents appropriatelysuggested stepping down treatment for a patient with

Table 3. Physician Practice Characteristics

Characteristic Variable All (N ¼ 367)

Clinic setting Academic-based 89.4Community hospital with residents 10.6

Weekly volumeof asthma patients

$1 per session 20.4

<1 per session 79.6

CORNET ¼ Continuity Research Network of the Academic Pediatric

well-controlled asthma receiving a high-intensity medica-tion regimen, but not for a similar patient receiving alow-intensity medication regimen. We believe that this dif-ferential response may suggest discomfort with the high-intensity medication regimen rather than a recognitionthat medications should be decreased for patients whoare doing well. Given the widespread problems of subopti-mal asthma care (eg, undertreatment), it may be that themessage to step down treatment for those doing well hasnot been equally received. Perhaps the patient populationand concerns regarding follow-up may also modify thestep-down approach. Standards for National Committeefor Quality Assurance Medical Home Qualification thatfocus on keeping patients with persistent asthma on inhaledcorticosteroids without emphasizing stepping down ther-apy may also contribute to this. More education is neededin this area.The lack of difference in treatment recommendations

between residents of different years in training was anunanticipated finding. There may be multiple differentreasons for this finding. First, there was a relativelysimilar volume of asthma patients by postgraduate yearcared for in the residents’ continuity clinic: 80% of res-idents reported caring for fewer than 1 asthma patientper clinic session, while 45% reported that they see nomore than 1 asthma patient in 4 clinic sessions. For

% of CORNET Respondents

P ValuePGY1 (n ¼ 96) PGY2 (n ¼ 140) PGY3þ (n ¼ 131)

87.5 92.1 87.8 NS12.5 7.9 12.212.5 21.4 25.2 .060

87.5 78.6 74.8

Association; PGY ¼ postgraduate year.

292 OKELO ET AL ACADEMIC PEDIATRICS

residents with 1 ambulatory clinic session per week, thisequates to approximately 11 asthma encounters per year.So although we conducted our study between May andJuly, during which time the PGY1 residents wouldhave had almost a full year of postgraduate training, attheir reported level of ambulatory clinic exposure, withsuch low volumes of exposure, PGY2 or PGY3þ resi-dents may not be more sophisticated in evaluatingasthma than their PGY1 counterparts. Other possible ex-planations include the following: 1) residents may notreceive appropriate supervision during continuity clinicsessions as a result of a poor knowledge base or theteaching skills of their preceptor; 2) the duration oftime spent in training is not sufficient to attain a knowl-edge base or competency level sufficient to practiceguideline-consistent asthma care; or 3) better tools/aidsare needed to facilitate the training of residents inproviding asthma care. We were not able to examineall of these factors in this study.

There are several limitations to this study. First, wehave assessed treatment decisions using vignettesrather than examining actual clinical practice. However,use of clinical vignettes is an established method ofmeasuring physician behavior.13,14 We do not know thefull extent of information or prior training used by theparticipants to respond to each vignette. Nor do wehave information on what exposure residents have inmanaging outpatient asthma beyond their reportedcontinuity clinic exposure. However, we have sought tounderstand how specific and different clinical factorsinfluence the final step of the decision-making process(ie, making a treatment recommendation), which maybe better addressed with a standardized vignette method-ology allowing comparison of residents at different stagesof training and work in different settings (eg, differentcase mix, patient populations). We also acknowledgethat in the acute-care vignette, the patient had been hospi-talized 6 months earlier, with an intervening office visit 3months after the hospitalization. Therefore, the condi-tions of this vignette are different than the others, whichhad no intervening office visit. We acknowledge that res-idents who did not step up treatment for this vignette maynot be inconsistent with asthma guidelines depending onthe child’s status at the intervening visit. Our results maynot be representative of nonrespondents at CORNET sitesor of pediatric residents at sites that are not members ofCORNET. However, we did include a national sampleof pediatric residents—the largest we know of to date.Participation by CORNET sites is elective and is basedon interest in the topic and feasibility of participation.The number of participating sites is consistent withother studies conducted within CORNET.7,9 We alsoobserved that resident treatment recommendations wereconsistent with practicing physicians who haveparticipated in similar studies.5,12 Last, we do not knowhow these residents’ clinic preceptors would respond toour vignettes, nor do we know what the clinicpreceptors taught residents about asthma guidelines andthe practice of asthma.

CONCLUSIONS

The findings from this study demonstrate that pediatricresidents’ management of asthma is consistent with na-tional guidelines in some cases but not in others. Addition-ally, there were no meaningful differences in the outpatientasthmamanagement between residents of different years intraining. Future efforts should be focused on developingstrategies to facilitate pediatric residence adherence tomanagement strategies that conform to national guidelinerecommendations and to adaptation of other asthma-related parameters for outpatient asthma management.

ACKNOWLEDGMENTS

We appreciate the efforts of our site coinvestigators: Baystate Medical

Center: Matthew Sadof, MD; Children’s Memorial Hospital, North-

western University: Sandra M. Sanguino, MD, MPH; DeVos Children’s

Hospital: William Stratbucker, MD; College/Westchester: Theresa Het-

zler, MD; North Carolina Children’s Hospital: Michael Steiner, MD;

University of Texas at Houston: Michelle S. Barratt, MD, MPH, and

Lisa de Ybarrondo, MD; University of Texas Health Sciences at San

Antonio: Pamela Wood, MD; University of Utah: Wendy Hobson,

MD; and Wilmington Hospital Health Center: Shirley Klein, MD. Sup-

ported by NHLBI K23HL089410. The findings and conclusions in this

report are those of the authors and do not necessarily represent the views

of the National Institutes of Health or the Department of Health and Hu-

man Services.

SUPPLEMENTARY DATA

Supplementary data related to this article can be foundonline at http://dx.doi.org/10.1016/j.acap.2013.12.008.

REFERENCES

1. Crocker D, Brown C, Moolenaar R, et al. Racial and ethnic disparities

in asthmamedication usage and health-care utilization. Data From the

National Asthma Survey. Chest. 2009;136:1063–1071.

2. Cabana MD, Ebel BE, Cooper-Patrick L, et al. Barriers pediatricians

face when using asthma practice guidelines. Arch Pediatr Adolesc

Med. 2000;154:685–693.

3. Cabana MD, Rand CS, Becher OJ, Rubin HR. Reasons for pediatri-

cian nonadherence to asthma guidelines. Arch Pediatr Adolesc Med.

2001;155:1057–1062.

4. Demissie S, Riekert KA, Eakin MN, et al. How do perceptions of

asthma control and severity relate to indicators of asthma status and

treatment recommendations by pediatricians? Pediatr Allergy Immu-

nol Pulmonol. 2012;25:17–23.

5. Okelo SO, Patino CM, Riekert KA, et al. Patient factors used by pedi-

atricians to assign asthma treatment. Pediatrics. 2008;122:e195–e201.

6. Yee AB, Fagnano M, Halterman JS. Preventive asthma care delivery

in the primary care office: missed opportunities for children with

persistent asthma symptoms. Acad Pediatr. 2013;13:98–104.

7. Krugman SD, Racine A, Dabrow S, et al. Measuring primary care of

children in pediatric resident continuity practices: a Continuity

Research Network study. Pediatrics. 2007;120:e262–e271.

8. Ozuah PO, Reznik M. Using unannounced standardized patients to

assess residents’ competency in asthma severity classification. Ambul

Pediatr. 2008;8:139–142.

9. Serwint JR, Thoma KA, Dabrow SM, et al. Comparing patients seen

in pediatric resident continuity clinics and national ambulatory med-

ical care survey practices: a study from the continuity research

network. Pediatrics. 2006;118:e849–e858.

10. Patino CM, Riekert KA, Quartey RI. Howard-Hopkins Center for

Reducing Asthma Disparities. Development of the Asthma Control

and Communication Instrument (ACCI). Am J Respir Crit Care

Med. 2005;A254.

ACADEMIC PEDIATRICS ASTHMA TREATMENT DECISIONS 293

11. Patino CM, Okelo SO, Rand CS, et al. The Asthma Control and

Communication Instrument: a clinical tool developed for ethnically

diverse populations. J Allergy Clin Immunol. 2008;122:936–943.

12. Diette GB, Patino CM, Merriman B, et al. Patient factors that physi-

cians use to assign asthma treatment. Ann Intern Med. 2007;167:

1360–1366.

13. Dresselhaus TR, Peabody JW, Luck J, Bertenthal D. An evaluation of

vignettes for predicting variation in the quality of preventive care. J

Gen Intern Med. 2004;19:1013–1018.

14. Peabody JW, Luck J, Glassman P, et al. Measuring the quality of

physician practice by using clinical vignettes: a prospective validation

study. Ann Intern Med. 2004;141:771–780.