ARTICLE |
a Asthma Center, Connecticut Children's Medical Center, Hartford, Connecticut
b Department of Pediatrics, University of Connecticut Health Center, Farmington, Connecticut
c ConnectiCare, Farmington, Connecticut
d Department of Epidemiology and Population Health, Albert Einstein College of Medicine, Bronx, New York
| ABSTRACT |
|---|
|
|
|---|
METHODS. A trial of a disease management program (Easy Breathing II) involving 20 private pediatric practices in the greater Hartford, Connecticut area was conducted between January 1, 2001, and December 31, 2003. Demographic data on participating practitioners and patients were obtained from questionnaires. Medical services utilization data from claims were obtained from ConnectiCare, a regional managed care organization.
RESULTS. Of the 16750 children enrolled in Easy Breathing II, 2458 were enrolled in ConnectiCare and 490 had asthma. Inhaled corticosteroid use increased in the community overall during the study period. After enrollment in Easy Breathing II, with adjustment for age, gender, ethnicity, asthma severity, season, and calendar year, children with persistent asthma experienced an additional 47% increase in inhaled corticosteroid use, a 56% reduction in outpatient visits, and a 91% decrease in emergency department visits for treatment of asthma. Adherence to national asthma guidelines for prescribing inhaled corticosteroids was 95%. Seventeen of the 20 practices are still using Easy Breathing, 5 years after program implementation.
CONCLUSIONS. Pediatric primary care clinicians in private practice settings can implement an asthma management program patterned after the national asthma guidelines. When implemented, this program is successful in reducing medical services utilization for children with asthma. Just as differences in patterns of medical services utilization exist in private practices, compared with urban clinics, the impact of disease management on medical services utilization differs in private practices, compared with urban clinics.
Key Words: Easy Breathing disease management children National Asthma Education and Prevention Program medical services utilization claims data
Abbreviations: NAEPP—National Asthma Education and Prevention Program EPR—Expert Panel Report ED—emergency department ICS—inhaled corticosteroid MCO—managed care organization CI—confidence interval
Asthma is the most common chronic disease of children, with prevalence rates of 6% to 13%.1–4 Because asthma affects urban minority children disproportionately, most studies have focused on the epidemiologic features of and risk factors for asthma in children who reside in urban communities and on the effectiveness of interventions designed to improve asthma therapy and to reduce asthma morbidity rates.5,6 However, according to the 2000 US Census Bureau, 50% of the 81 million children who reside in the United States live in nonurban communities, defined as areas with population densities of <1000 people per square mile.7,8 Few studies have examined asthma epidemiologic features and morbidity rates for these nonurban children.9,10 In a cross-sectional study of 19076 children living in nonurban residential communities in the greater Hartford, Connecticut area, we found that risk factors for asthma were similar to those observed in urban communities (defined as census tracts where
20% of the households had incomes below the federal poverty level, with population densities of >1000 people per square mile) and included lower socioeconomic status, nonwhite ethnicity, male gender, and exposure to tobacco smoke, dust, and cockroaches.10 The available information suggests that asthma also represents a significant cause of morbidity for these nonurban children.
National guidelines for asthma diagnosis and management were released beginning in 1991 (National Asthma Education and Prevention Program [NAEPP] Expert Panel Report [EPR])11 and were updated most recently in 199712 and 2002 (NAEPP EPR-2),13 but they have not been adopted extensively by the nonasthma specialty community.14–16 We demonstrated previously that primary care clinicians who work in urban clinics can adhere to the national asthma guidelines.17 In addition, we demonstrated that adherence to a standardized, systematic approach to asthma management, based on the national guidelines, resulted in significant decreases in medical services utilization, including decreases in emergency department (ED) visits for treatment of asthma and hospitalizations, and significant increases in inhaled corticosteroid (ICS) therapy.17 As part of a Translating Research into Practice Initiative18 sponsored by the Agency for Healthcare Research and Quality, in partnership with a private managed care organization (MCO), we examined whether practitioners in private practice office settings could adhere to the national asthma guidelines and, importantly, whether adherence resulted in a reduction in asthma morbidity rates among children with asthma who receive their care in this type of practice setting.
| METHODS |
|---|
|
|
|---|
Private Practices and Clinicians
Twenty-seven private practices located in the greater Hartford, Connecticut, region and enrolled in 1 of 2 independent practice associations with reportedly high (>25%) penetration by a partner MCO (ConnectiCare) were invited to participate in the Easy Breathing II program. ConnectiCare is a regional MCO that serves children and adults in Connecticut. All practices were fee for service, and none received government support or belonged to a health maintenance organization. Eighty-six percent of the clinicians (65 of 76 clinicians) from 74% of the invited practices (20 of 27 practices) accepted the invitation. Demographic information about the practice and the practitioners was obtained, and practices were matched according to number and type of providers and assigned randomly to receive the Easy Breathing II program either in year 1 or in year 3 of the project.
Participants
All children between 6 months and 18 years of age who presented for care for any reason between January 1, 2001, and December 31, 2003, at a participating private primary care practice that had been trained in Easy Breathing II were eligible for enrollment in the program. However, only those who were enrolled and were members of ConnectiCare were included in the study sample reported here. The demographic features of all participants (19076) who were enrolled in Easy Breathing II between January 1, 2001, and December 31, 2003, were reported previously.10 The children reported here represent a subset of that larger study sample.
Data Sources
Medical services utilization claims data (inpatient, outpatient, ED, and pharmacy) for the years 1998 through 2003 were obtained for all children who were enrolled in Easy Breathing II and insured by ConnectiCare and were matched to claims. National Drug Codes were used to identify asthma drugs, which were then categorized as short-acting ß2-receptor agonists, long-acting ß2-receptor agonists, ICS, orally administered corticosteroids, or nonsteroidal antiinflammatory drugs (including leukotriene modifiers). Participant demographic features, diagnosis (asthma or no asthma), asthma severity and treatment (where appropriate), and information regarding exposure to potential asthma triggers were collected from participants with the Easy Breathing II instruments.
Statistical Analyses
Data from the Easy Breathing II survey were analyzed by using Pearson
2 analyses and multivariate logistic regression analyses to compare demographic features, asthma risk factors, and exposures for children with asthma versus those without asthma. Multivariate analyses controlled for gender, ethnicity, age, and asthma severity (where appropriate). To determine the generalizability of the results to all children enrolled in Easy Breathing II, the demographic features of the participants enrolled in Easy Breathing II who were members of ConnectiCare were compared with the demographic features of the participants in Easy Breathing II who were not members of ConnectiCare.
ConnectiCare claims data were used to compare the number of hospitalizations, ED visits, and office visits, both overall and for asthma. Pharmacy claims for asthma medications were compared before and after enrollment in Easy Breathing II. Claims data were available from January 1, 1998, through December 31, 2003. Therefore, 3 years of historical data before the start of the Easy Breathing II program and 3 years of data after the start of the program were examined. Each child contributed to the analysis for each month in which they were eligible for coverage by ConnectiCare. Beginning in 2001, children were enrolled continuously into Easy Breathing II; therefore, at any time beginning in 2001, there were children who were enrolled in Easy Breathing II, children who were not yet enrolled in Easy Breathing II, and children who were never enrolled in Easy Breathing II.
For the analyses, utilization data between January 1998 and December 2003 for all children for the entire time period before enrollment were pooled and data for all children for the entire time period after enrollment were pooled. Relative rates of medical services utilization (events per child-months) before and after enrollment in the Easy Breathing II program were examined. Therefore, the analyses compared each child's entire history of utilization between 1998 and 2003 before and after enrollment in Easy Breathing II, as well as comparing utilization at a given time for all enrolled children and those not yet enrolled. Therefore, in this analysis, the primary efficacy comparisons are both historical (using each child's utilization data before and after enrollment) and contemporaneous (using the utilization data at a given time for all children enrolled and all children yet to be enrolled in Easy Breathing II).
Generalized estimating equations were used to fit multivariate, marginal, binary and Poisson regression models to account for the fact that each child had multiple data points in the data set. Gender, ethnicity, asthma severity, and season were controlled in the analyses. The effect of longer-term secular trends (eg, changes in ICS use in the community at large) and the effect of the aging of the cohort were also controlled. Some ConnectiCare members do not receive pharmacy coverage through ConnectiCare, and those children were excluded from the pharmacy utilization analyses. Data were analyzed using SAS 9.1 (SAS Institute, Cary, NC).
| RESULTS |
|---|
|
|
|---|
Between January 1, 2001, and December 31, 2003, 16750 children were enrolled in Easy Breathing II. Of these, 3004 indicated that their primary insurance carrier was ConnectiCare, and 2458 (82%) were matched with claims data. These 2458 children represent the study population (Table 1). Matched children in Easy Breathing II were slightly older (P = .0005), with an average age at survey of 8.14 ± 5.03 years, than were unmatched ConnectiCare children, with an average age at survey of 7.32 ± 5.30 years.
|
We next compared the enrolled ConnectiCare children with the other children who were enrolled in Easy Breathing II in these 20 practices and who were not insured by ConnectiCare. There was no difference in gender, ethnicity, proportion with asthma, or asthma severity; however, ConnectiCare children were slightly older (P < .0001), with an average age at survey of 8.14 ± 5.03 years, whereas non-ConnectiCare children had an average age at survey of 7.58 ± 5.02 years.
Asthma in the Study Population
Of the 2458 children who were enrolled in ConnectiCare and had claims data available for review, 490 (20%) had a physician-confirmed diagnosis of asthma. When the demographic features of these children (Table 1) were compared with the demographic features of all of the children with asthma in Easy Breathing II, there was no difference in gender (P = .36), ethnicity (P = .10), or asthma severity (P = .57). However, there were fewer ConnectiCare children with asthma who were <5 years of age (25% vs 31%; P = .006). For ConnectiCare children, male patients were 1.49 times (95% Cl: 1.21–1.82 times) more likely to have asthma, compared with female patients, and older children (
5 years of age) were 1.52 times (95% CI: 1.21–1.91 times) more likely to have asthma. There was no difference in self-reported ethnicity between ConnectiCare-enrolled children with and without asthma (P = .21).
Risk factors for asthma and increased asthma severity for the ConnectiCare children who were enrolled in Easy Breathing II are shown in Table 2. In the multivariate analysis, Hispanic ethnicity, male gender, age of >5 years, family history of asthma, and exposure to tobacco smoke and dust were risk factors for asthma prevalence, whereas self-reported exposure to dust was the only significant risk factor for persistent disease.
|
|
|
|
| DISCUSSION |
|---|
|
|
|---|
Multiple studies have demonstrated that use of ICS therapy decreases asthma exacerbations.17,23,24 Despite the publication of national guidelines for asthma therapy, including the recently updated recommendations using an evidence-based approach (NAEPP EPR-2), ICS therapy continues to be underutilized by primary care clinicians.14,15,25–27 We demonstrate that primary care clinicians in both urban and private practice settings can use the NAEPP EPR-2 guidelines and, when used, the guidelines are effective in reducing asthma morbidity. The program was successful in private practices of varying sizes, with different types of clinicians and different patient mixtures. These practices had all used quality-improvement projects to improve practice efficiency. None of the practices, however, had used a disease management program to improve patient care. It is noteworthy that, 5 years after initiation of the program, 17 of the 20 practices are still using Easy Breathing.
Why was the Easy Breathing II program adopted successfully by pediatric primary care clinicians when other programs have not been embraced so thoroughly? We think that the program's success originates from the simplicity of the program itself. Easy Breathing is not an all-encompassing, comprehensive, asthma management program. Rather, the program focuses on 4 major aspects of the NAEPP guidelines, namely, asthma diagnosis, determination of asthma severity, use of ICS and other antiinflammatory treatments appropriate for the asthma severity, and development of a written asthma treatment plan. All required program elements are contained on 2 pieces of paper (survey/provider assessment and asthma treatment plan). Although other elements, such as patient education, patient adherence techniques, spirometry, and allergy skin testing, are offered, clinicians choose when and how they wish to use these additional elements. Over time, requests from practitioners for these additional elements have increased.
Because the program has 4 major components, it is not possible to determine which of these components is responsible for the decrease in medical services utilization. We demonstrated in this study and a previous study,17 however, that the program's success is not solely attributable to prescribing of ICS. ICS prescriptions for children not enrolled in Easy Breathing increased over the study period, but those children did not benefit in the same way that children in Easy Breathing II benefited. We also know that, before implementation of Easy Breathing II, <5% of children with asthma cared for in these private practices received a standardized set of written instructions on how to manage their asthma. Use of a written asthma treatment plan has been implicated in reducing asthma hospitalization and ED visit rates.28
This study has several limitations. Although the practices were assigned randomly to receive Easy Breathing either in year 1 or in year 3, children within the practices were not assigned randomly. Some clinicians targeted children with asthma, whereas others enrolled children during routine physical examinations or at specific times during the day (eg, all 10 AM and 2 PM appointments). Other priorities and time constraints were the reasons most often given for not enrolling all of the children in the practice in the program. Therefore, it is likely that children with asthma who used more medical services were more likely to be sampled. The only difference, however, that we were able to measure regarding who was enrolled in the program was an oversampling of young children. As has been observed in other epidemiologic studies, children with asthma were more likely to report tobacco smoke, dust, and cockroach exposure and to be male, young, and from a minority ethnic group.5,28,29
Practices from 1 of 2 independent practice associations with high rates of penetration by ConnectiCare were invited to participate. We had predicted that we would be able to enroll >5000 ConnectiCare children during the study period. The practices themselves, however, reported lower penetration rates, and their a priori reported penetration rate was consistent with the 14.7% of enrolled children who reported coverage by ConnectiCare and who were matched subsequently to ConnectiCare claims data.
Use of paid claims might have underestimated the number of prescriptions written and dispensed by the primary care clinicians. However, there should not have been any difference in the frequency of free medicine dispensing between children enrolled in Easy Breathing II and children with asthma before their enrollment in the program. Any difference would most likely be conservative, in that it would underestimate the increase in ICS use after enrollment in Easy Breathing II. Underuse of the prescribed therapy was observed, as evidenced by the small number of prescriptions filled by patients. In conclusion, pediatric primary care clinicians in private practices can adhere to the NAEPP guidelines, and adherence to these guidelines is associated with reductions in urgent care outpatient visits and ED visits for treatment of asthma and with modest cost savings.
| ACKNOWLEDGMENTS |
|---|
We thank the providers and office staff members of the 20 primary care practices for their willingness to participate in the Easy Breathing II Program, including Central Pediatrics and Adolescent Medicine; Children's Medical Group, Hamden; Connecticut Valley Pediatrics; Leonard Forner, MD; John J. Fote, MD, John H. Lavalette, MD, Jennifer L. Schwab, MD, and Jennifer L. Nichols, APRN; Kids Station Pediatrics; Sobhy Ghabrial, MD; Healthwise Medical Associates; Naseem Deen, MD; Wanda Merced, MD, Barbara Renshaw, APRN; Vernon Pediatrics; Pedicorp; Pediatric Associates of Marlborough; Prohealth Physicians MSO; Children's Medical Group; Sherry Banack, MD, and Candra Smith-Slatas, MD, PhD; Vandana Sachetti, MD; Shoreline Pediatrics and Adolescent Medicine; Michael S. Levine, MD, and Leah Galjan Post, MD; Robert R. Toscano, MD; and UConn Health Partners. We also thank Alicia Ignatowicz, MPH, for data entry and program support, Jennifer Adkins for data management, and Olga Früster for administrative assistance.
| FOOTNOTES |
|---|
Address correspondence to Michelle M. Cloutier, MD, Asthma Center, Connecticut Children's Medical Center, 282 Washington St, Hartford, CT 06106. E-mail: mclouti{at}ccmckids.org
The authors have indicated they have no financial relationships relevant to this article to disclose.
| REFERENCES |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||