http://informahealthcare.com/jas ISSN: 0277-0903 (print), 1532-4303 (electronic) J Asthma, 2014; 51(5): 474–479 ! 2014 Informa Healthcare USA, Inc. DOI: 10.3109/02770903.2014.891608
UNDERSERVED POPULATIONS
Asthma coalition effects on vulnerable sub groups of children: the most frequent users of health care and the youngest Laurie Lachance, PhD, MPH1, M. Beth Benedict, DrPH, JD2, Linda Jo Doctor, MPH3, Lisa A. Gilmore, MBA, MSW4, Cynthia Kelly, MD, FAAP 5, James Krieger, MD, MPH 6, Marielena Lara, MD, MPH7, John Meurer, MD, MBA8, Amy Friedman Milanovich, MPH1, Elisa Nicholas, MD, MSPH9, Michael Rosenthal, MD10, Peter X. K. Song, PhD11, Shelley C. Stoll, MPH1, Daniel F. Awad, MA1, Margaret K. Wilkin, MPH1, and Noreen M. Clark, PhD1 1
Center for Managing Chronic Disease, University of Michigan, Ann Arbor. MI, USA, 2Centers for Medicare, Medicaid Services, Health & Human Services, Baltimore, MD, USA, 3W.K. Kellogg Foundation, Battle Creek, MI, USA, 4FHI 360, Durham, NC, USA, 5Eastern Virginia Medical School, Children’s Hospital of the King’s Daughters, Norfolk, VA, USA, 6Department of Public Health, Seattle & King County, Seattle, WA, USA, 7RAND Health, Santa Monica, CA, USA, 8Medical College of Wisconsin and Children’s Hospital and Health System, Milwaukee, WI, USA, 9The Children’s Clinic, Serving Children and Their Families, Long Beach, CA, USA, 10Department of Family & Community Medicine, Christiana Care Health Systems, Wilmington, DE, USA, and 11Department of Biostatistics, University of Michigan, Ann Arbor, MI, USA
Abstract
Keywords
Objective: To examine the impact of Allies Against Asthma, community-based coalitions working to improve asthma outcomes, on vulnerable children: those with the most urgent health care use and those of youngest age. Methods: Allies zip codes were matched with comparison communities on demographic factors. Five years of Medicaid data (n ¼ 26,836) for significant health care events: hospitalizations, ED and urgent care facility visits, were analyzed. Longitudinal analyses using generalized estimating equations and proportional hazards models compared Allies and comparison group children. Results: In the two start-up years of Allies, odds of having a significant event were greater for Allies children than for comparison children (p50.05). During the third and fourth years when Allies activities were fully implemented, for frequent health care users at baseline, odds of an asthma event were the same for both Allies and comparison children, yet in the less frequent users, odds of an event were lower in Allies children (p50.0001). In the initial year of Allies efforts, among the youngest, the Allies children had greater odds than comparison children of an event (p50.01), but by the fourth year the Allies group had lower odds (p ¼ 0.02) of an event. Hazard ratios over all years of the study for the youngest Allies children and most frequent baseline users of urgent care were lower than for comparison children (p ¼ 0.01 and p ¼ 0.0004). Conclusion: Mobilizing a coalition of diverse stakeholders focused on policy and system change generated community-wide reductions over the long-term in health care use for vulnerable children.
Management/control, morbidity and mortality, pediatrics, prevention, quality of life
Introduction Children experiencing significant asthma health care events defined as a need for urgent care services (hospitalizations, ED visits, urgent care facility visits), and the youngest patients comprise vulnerable asthma subgroups in the pediatric population [1,2]. Frequent urgent health care utilization and younger age have been shown to be associated with more disease burden for children with asthma, their families and the health care system [3–7]. High rates of urgent care use for asthma are associated with residing in low income and minority communities [8–10]. A variety of ways have been examined in which to improve services, provide selfmanagement support to patients and reduce the demand for Correspondence: Laurie Lachance, PhD, Center for Managing Chronic Disease, University of Michigan, 1415 Washington Heights, Ann Arbor, MI 48109-2029, USA. Tel: +73 47631457. Fax: +73 47639115. E-mail: [email protected]
History Received 15 August 2013 Revised 24 January 2014 Accepted 30 January 2014 Published online 7 March 2014
more expensive asthma care [11–12]. One way discussed in the literature is formation of community-wide asthma coalitions [13]. These are collaborative efforts at the local level to achieve asthma control by bringing together diverse stakeholders from health care providers to grassroots groups to voluntary and government agencies to business and industry who collectively implement strategies to improve asthma outcomes. Currently, there are over 200 asthma coalitions in the U.S [14]. One such effort focused on low income communities: Allies Against Asthma. This national program supported by the Robert Wood Johnson Foundation, has been described in detail in previous publications [15,16]. Evaluation of the Allies coalitions demonstrated that such collective effort can bring about significant policy and system change related to clinical care and management of asthma and that in a cohort of asthma patients followed over one year, significant symptom reduction was evident [15]. Further, using Medicaid data, the Allies study showed that improved asthma outcomes were community
DOI: 10.3109/02770903.2014.891608
wide, that is, an enhanced health status of the population of low income children residing in target communities [16]. It has been shown that Allies coalitions had a broad effect. The study presented here examined Allies coalition effects on a subgroup of low income children and the most vulnerable children with asthma. It addressed the question: were changes in health care use for children in the Allies communities seen among the most frequent users of urgent care services and the youngest children.
Methods All study procedures were approved by the University of Michigan Institutional Review Board. To conduct this study, the Center for Managing Chronic Disease (CMCD) of the University of Michigan entered into collaboration with the Centers for Medicare and Medicaid Services (CMS) of the federal government. CMS provided zip code level data to CMCD and consultation regarding data analysis and variables related to children with asthma covered by Medicaid insurance provisions from 2002 to 2006. Samples Allies coalitions worked in low income neighborhoods in the following six cities across the United States: Hampton Roads, Virginia; Washington, D.C.; Milwaukee, Wisconsin; King County/Seattle, Washington; Long Beach, California; Philadelphia, Pennsylvania. Allies coalitions provided the zip codes for neighborhoods that comprised their focus, that is, the areas of greatest coalition activity. Expert consultants to the CMCD then used a set of criteria to identify matched zip codes in areas where Allies coalitions were not active. For each Allies zip code, one or two demographically similar zip codes (according to Census 2000) were identified to serve as comparison communities. Characteristics for selecting matched comparison groups were percentages of racial/ ethnic groups and median income, as these variables consistently are associated with asthma prevalence. [17] Secondary variables used in matching were total population size, percentage of family households and percentage of population younger than 18 years. Geographic considerations were also made (e.g. rural versus urban settings) in the matching. The non-Allies comparison zip codes were located in Roanoke City, Virginia; Jacksonville, Florida; Everett, Lacey, Olympia and Tacoma, Washington; National City and San Bernardino, California; Baltimore, Maryland; Lorain, Ohio; Muskegon, Detroit, and Flint Michigan; Fort Wayne and Indianapolis, Indiana [18]. The files from CMS for children in the intervention and comparison sites included all of the Medicaid fee-for-service claims and encounter claims from managed care patients submitted in the study years. A person-specific cohort data file was constructed for analysis based on the following inclusion criteria: (1) aged 2–18 on 1 January 2002; (2) not enrolled as disabled; (3) enrolled in 2002 and at least one year follow-up, 2004–2006; and (4) in the baseline year, had at least one health care visit with a principal or secondary diagnosis of asthma (ICD-9 codes 493.XX) to include a hospitalization, a visit to an emergency department, an urgent care facility, or to a physician’s office; or a filled prescription
Asthma coalition outcomes for vulnerable children
475
for at least one of the following asthma medications from HEDIS guidelines [19]: inhaled corticosteroids, inhaled corticosteroid and long acting beta agonist combination, leukotriene modifier, long-acting inhaled beta-2 agonist, mast cell stabilizer, methylxanthine, short-acting inhaled beta-2 agonist. This produced data on 26 836 children for analysis. Data sources Four types of CMS files for each year were examined for inclusion in the study: inpatient (hospitalizations); other services (physician’s visit, emergency department visits (ED) and visits to an urgent care facility), prescription drug record, and person summary (demographic information). Data fields included beginning and ending date of service, diagnosis (asthma principal or secondary), type of service, where service was provided (hospital, ED, urgent care facility), National Drug Code (NDC) for filled prescriptions, and eligibility and payment. Data management and analysis First, the sample was divided by (a) frequency of a significant health care event (with frequent urgent care users defined as those with at least one hospitalization or two emergency department or urgent care facility visits in the baseline year) and (b) age group (ages 2–3, 4–6 and older than 6 years in the baseline year). Then, the annual health care use in each group for hospitalizations, ED and urgent care facility visits were described. Due to the low frequency of these events, a ‘‘significant asthma event’’ was defined as presence or absence of any urgent care visit in a given year and compared using frequency tables where statistical significance was evaluated by Pearson chi-square test. Stratified analyses were conducted to determine differences between Allies and comparison children with the most frequent urgent care utilization and between Allies and comparison children with less frequent use; similar stratified analysis between intervention groups was also conducted by age group. Finally, we conducted longitudinal data analysis that focused on covariates believed to be associated with the outcome of ‘‘significant asthma event.’’ Adjusted odds ratios (OR) comparing the odds of having a ‘‘significant asthma event’’ between the intervention group and the comparison group in each year of follow-up were obtained in a longitudinal generalized estimating equation model through an interaction between year and intervention group. This method allows each OR to be controlled for the other years, including baseline. Baseline age, gender and race/ethnicity were also controlled for in the models (age was excluded in the model comparing age groups). To examine the effect of the Allies interventions over the entire study period, 2002–2006, a recurrent event analysis approach using Cox proportional hazards model [20] was undertaken to analyze the time to ‘‘significant asthma event’’ from previous events. This approach allowed comparison of the hazard of an event between intervention and comparison groups to be conditional on previous asthma events and also to control for demographic and other potentially confounding factors. This analysis was conducted separately in each significant event group and age group to determine if the intervention
476
L. Lachance et al.
J Asthma, 2014; 51(5): 474–479
effect was similar for each stratum. SAS/STAT software, Version 9.2 of the SAS System for Windows was used for analysis [21].
Results The differences in demographic factors between significant event groups and between age groups were few. There were more black children, males and children aged 2–4 with more frequent health care use compared to those with less frequent use. By age, there were more males in the 2–3-year-old group and also more significant asthma events in this group. These findings are consistent with those reported in previous studies [8–10]. Emergency or urgent visits and hospitalizations were relatively rare events. Tables 1 and 2 show proportions with ED or urgent care visits and hospitalizations by utilization status and age group. Ten percent or less of the most frequent urgent care users had a hospitalization in any follow-up year and from 14–31% had an ED or urgent care visit. Around 1–2% of those with less frequent urgent care use and across age groups were hospitalized in follow-up years. Emergency or urgent care visits occurred in 3.8% to 9.4% of subjects across age groups. This was similar for those less frequent urgent care users.
Tables 3 and 4 display adjusted odds of having a significant asthma event in the Allies group versus the comparison group in each year of the study by utilization status (Table 3) and by age (Table 4). In the frequent users of urgent care, in 2003 and 2004, the Allies group had almost 50% greater odds of having a significant event than the comparison group while in years, 2005 and 2006 the odds of having an event were the same in both groups. In the less frequent users, the Allies group had significantly greater odds (25%–40%) of having an asthma event than comparison children in years 2003 and 2004, and significantly lower odds (20%–25%) in 2005 and 2006 (see Table 3) when adjusted for age, gender and race/ethnicity. In all age groups, the Allies group children were more likely to have an asthma event in 2003 and 2004, but less likely in 2005 and 2006, with a 47% reduced odds ratio from 2003 to 2006 in those aged 2-3 (p value ¼ 0.0002), 41% reduced odds ratio from 2003 to 2006 in those aged 4–6 (p value ¼ 0.0012) and 34% reduced odds ratio from 2003 to 2006 in those older than age 6 (p value50.0001) (Table 4) adjusting for an interaction between year and intervention group and for gender and race/ethnicity. Table 5 presents the estimated hazard ratios for the Allies group versus the comparison group, which were obtained by
Table 1. Number and percent of children with at least one urgent health care visit by year and baseline urgent care utilization. Frequent urgent care usersa in baseline year (2002) (n ¼ 1039)
2003, at least one: ED or urgent care Hospitalization 2004, at least one: ED or urgent care Hospitalization 2005, at least one: ED or urgent care Hospitalization 2006, at least one: ED or urgent care Hospitalization
Less or no urgent care use in baseline year (2002) (n ¼ 25 797)
Allies N (%)
Comparison N (%)
Allies N (%)
Comparison N (%)
106 (30.7%) 30 (8.7%)
127 (19.6%) 65 (10.0%)
727 (6.3%) 115 (1.0%)
571 (4.3%) 202 (1.5%)
75 (25.3%) 12 (4.1%)
76 (14.0%) 47 (9.2%)
592 (5.9%) 77 (1.0%)
377 (3.7%) 140 (1.4%)
46 (18.4%) 16 (6.4%)
75 (18.0%) 34 (8.2%)
547 (6.3%) 69 (1.0%)
665 (7.6%) 125 (1.4%)
46 (22.3%) 11 (5.3%)
80 (21.9%) 25 (6.8%)
411 (5.4%) 59 (1.0%)
520 (7.0%) 93 (1.2%)
a
Defined as 2 emergency department or urgent care center visits OR 1 hospitalization for asthma in the baseline year, 2002.
Table 2. Number and percent of children with at least one urgent health care visit by year and baseline age group. Ages 2–3 in 2002 (n ¼ 4829)
2003, at least one: ED or urgent care Hospitalization 2004, at least one: ED or urgent care Hospitalization 2005, at least one: ED or urgent care Hospitalization 2006, at least one: ED or urgent care Hospitalization
Ages 4–6 in 2002 (n ¼ 5610)
Older than age 6 in 2002 (n ¼ 16 397)
Allies N (%)
Comparison N (%)
Allies N (%)
Comparison N (%)
Allies N (%)
Comparison N (%)
209 (9.4%) 36 (1.6%)
143 (5.9%) 68 (2.8%)
162 (6.5%) 31 (1.2%)
150 (5.1%) 62 (2.1%)
462 (6.4%) 78 (1.1%)
405 (4.7%) 137 (1.6%)
145 (7.5%) 15 (1.0%)
88 (4.8%) 43 (2.4%)
144 (6.6%) 26 (1.2%)
110 (4.9%) 38 (1.7%)
378 (6.1%) 48 (1.0%)
255 (3.8%) 106 (1.6%)
118 (7.2%) 17 (1.0%)
111 (7.2%) 28 (1.8%)
126 (6.7%) 17 (1.0%)
170 (8.8%) 40 (2.1%)
349 (6.4%) 51 (1.0%)
459 (8.0%) 91 (1.6%)
82 (5.7%) 18 (1.2%)
101 (7.6%) 18 (1.4%)
97 (5.6%) 14 (1.0%)
140 (8.1%) 28 (1.6%)
278 (6.0%) 38 (1.0%)
359 (7.5%) 72 (1.5%)
Asthma coalition outcomes for vulnerable children
DOI: 10.3109/02770903.2014.891608
477
Table 3. Stratified comparison of adjusted oddsa of having a significant asthma eventb in a given year in the Allies group children versus the comparison group children. Frequent urgent care usersc Allies versus Comparison (n ¼ 1,039) Year
OR
2003 2004 2005 2006
1.503 1.458 1.005 1.090
95% CI (1.126, (1.039, (0.678, (0.731,
2.007) 2.046) 1.491) 1.624)
Less or no urgent care use Allies versus Comparison (n ¼ 25,797) p
OR
0.0057 0.0291 0.9786 0.6719
1.280 1.460 0.785 0.761
95% CI (1.160, (1.290, (0.700, (0.669,
p 50.0001 50.0001 50.0001 50.0001
1.433) 1.653) 0.881) 0.867)
a
Adjusted for age, gender and race/ethnicity. A hospitalization, ED, or urgent care facility visit. c Defined as 2 emergency department or urgent care center visits OR 1 hospitalization for asthma in the baseline year, 2002. b
Table 4. Age stratified comparison of adjusted oddsa of having a significant asthma eventb in a given year between the Allies and comparison children. Allies group versus Comparison group Ages 2–3 in 2002 (n ¼ 4829) a
Year
OR
2003 2004 2005 2006
1.33 1.25 0.92 0.70
95% CI (1.08, (0.97, (0.71, (0.52,
1.63) 1.61) 1.20) 0.95)
Ages 4–6 in 2002 (n ¼ 5610) a
p
OR
0.0077 0.0922 0.5525 0.0198
1.11 1.27 0.68 0.66
95% CI (0.89, (0.99, (0.54, (0.51,
1.37) 1.61) 0.86) 0.86)
Older than age 6 in 2002 (n ¼ 16 397) p
ORa
0.3513 0.0561 0.0012 0.0018
1.23 1.46 0.79 0.81
95% CI (1.08, (1.26, (0.69, (0.70,
1.40) 1.71) 0.91) 0.95)
p 0.0018 50.0001 0.0010 0.0101
a
Adjusted for gender and race/ethnicity. A hospitalization, ED, or urgent care facility visit.
b
Table 5. Stratified comparison of hazard of having a significant asthma eventa over the entire study period from 2002–2006, adjusted for age (only in the frequent users versus less frequent users model), gender and race/ethnicity.
Frequent urgent care user children (n ¼ 1039) Less frequent user children (n ¼ 25 797) Children ages 2–3 Children ages 4–6 Children older than 6 years
b
Hazard ratio for significant asthma eventa (95% confidence interval) Allies group versus Comparison group
p Value
0.88 (0.812, 0.943) 1.03 (0.990, 1.080) 0.903 (0.834–0.979) 0.982 (0.906–1.064) 0.941 (0.896–0.989)
0.0004 0.129 0.0132 0.6572 0.0166
a
A hospitalization, ED, or urgent care facility visit. Defined as 2 emergency department or urgent care center visits OR 1 hospitalization for asthma in the baseline year, 2002.
b
the recent event data analysis of time to an asthma event, adjusting for age (only in the urgent care user models), gender and race/ethnicity. Examination across the entire study period, from 2002–2006, showed that the Allies group children had a lower hazard of having a hospitalization, ED visit, or urgent care facility visit than the comparison group children for the frequent user cohort and for children aged 2–3 or older than 6. Given a hazard ratio of 0.88 in the frequent user cohort, Allies children had approximately 14% longer average time to event than comparison children. In the less frequent users and those aged 4–6, the hazard was equal in both the intervention and comparison group children, after the adjustment for age (in the more urgent care versus less urgent care model), gender, and race/ethnicity (see Table 5).
Discussion The data identifies a clear progression toward more effective asthma control within vulnerable sub-groups of the
population of children with asthma, residing in low income Allies communities. Therefore, the effects of Allies coalitions on significant asthma health care events illustrate a pattern that aligns with the maturation of coalitions as organizations trying to effect change. In the start-up years of functioning (2003–2004), urgent care use actually increased in the Allies communities. It is likely that greater focus of coalition stakeholders to call attention to asthma as a serious community problem, along with the need to introduce changes to enhance services, created more awareness and stimulated care seeking by targeted families. Two years later, when activities of the coalitions’ policy and system changes had time to evolve, children with frequent urgent care use in Allies communities had reduced their frequency of care to the same level of that in comparison communities. By that time point, Allies less frequent urgent care users at baseline experienced significantly fewer health care events. The same pattern was evident in all age groups, however, with greater maturation of the coalition efforts over time, younger children in the Allies
478
L. Lachance et al.
group had the greatest reductions in the odds of need for health care. An obvious interpretation of these data is that community wide change doesn’t happen overnight, even when a wide range of interested stakeholders mobilize to redesign services and education and introduce system and policy change. However, these data do illustrate that given a time period as short as two years, claims for urgent health care use was reduced among vulnerable children. In other words, it was in the later years that significant results were apparent. When viewed across the study years, the hazard ratios for the most frequent urgent care users at baseline and the youngest children were significantly improved in Allies communities. The findings suggest that although the greatest effects coincided with coalitions’ maturation, some impact may have been evident from the outset. There are potential limitations to this study. Average intervention effects over the entire time period were not seen in the less frequent users of urgent services or in the older age groups. In these groups, the odds ratios by year were in favor of the comparison group in the first two years of start up and in favor of the Allies group in the third and fourth years of follow-up. This may have led to a masking of the true hazard over the entire time period. Claims from both the Allies and comparison sites may not have included all Medicaid enrolled children during the study years because not all encounter claims from managed care patients are submitted to CMS nor are all children continuously enrolled in Medicaid. To understand effects, if any, related to these phenomena, in the proportional hazards model, children continuously enrolled in Medicaid were compared to children that may have had a gap in enrollment. This comparison showed very comparable, statistically significant results, which suggested that the assumption of no occurrences of events in the gap time was reasonable in the recurrent event data analysis. Another limitation is that the level of exposure of any given child to the various components of service improved by the Allies coalitions and policy and systems changes initiated is not known. It is known, however, that over time, children who were frequent urgent care users and those who were younger living in the areas where changes were introduced subsequently had fewer significant asthma care events, an indication that whatever the level of exposure the outcome was positive. An important aspect of this study is that it reports community-wide results for children participating in Medicaid across the Allies neighborhoods. Given the large amount of data available to in this study, a longitudinal, cohort analysis could be undertaken. This method provided the strong and clear pattern of intervention effects over time.
Conclusion The findings from this study indicate that mobilizing diverse stakeholders, including consumers, and focusing on policy and system changes can generate significant reductions in health care use for vulnerable children with asthma. The Allies coalitions generated important community benefits. Claims for Health care utilization for the highest users of urgent services declined significantly as did use by the
J Asthma, 2014; 51(5): 474–479
youngest children, two important subgroups in the pediatric population. Community-wide efforts to engage stakeholders in initiating far reaching outcomes through systemic change can greatly benefit poor children across targeted neighborhoods.
Acknowledgements We gratefully acknowledge the guidance and contributions of Allies Advisory board members and consultants Michele Carrick, MSW, Elaine Cassidy, PhD, Mary desVignesKendrick, MD, MPH, Susan Downey, M.Ed., Rachel A. Gonzales-Hanson, Rob Fulwood, PhD, JA Grisso, MPH, MD, Barbara Israel, Ph.D., Talmadge King, MD, Floyd Malveaux, MD, PhD, Robert Mellins, MD, Steve Page, MPA, Guy Parcel, PhD, Stephen Redd, MD, Jeanne Taylor, PhD, Abe Wandersman, PhD, Sandra Wilson, PhD, Albert Yee, MD, Sarah Hearn, Ye Yang, and Charles Brinker. Additionally, we thank Christine Joseph, PhD and her team for matches of Allies zip codes. We also wish to thank the dedicated members of all the Allies coalitions. This work is dedicated to the memory of Dr. Noreen M. Clark, Director of the Center for Managing Chronic Disease, University of Michigan School of Public Health and to the memory of Dr. M. Beth Benedict, retired from the Centers for Medicare, Medicaid Services, Health & Human Services with gratitude for their vision and guidance.
Declaration of interest This study was supported by the Robert Wood Johnson Foundation, with additional support provided by the W.K. Kellogg Foundation. The authors have no further relevant interests to disclose.
References 1. Clement LT, Jones CA, Cole J. Health disparities in the United States: childhood asthma. Am J Med Sci 2008;335:260–265. 2. Federico MJ, Liu AH. Overcoming childhood asthma disparities of the inner-city poor. Pediatr Clin North Am 2003;50:655–675. 3. Bateman ED, Bousquet J, Keech ML, Busse WW, Clark TJ, Pedersen SE. The correlation between asthma control and health status: the GOAL study. Eur Respir J 2007;29:56–62. 4. Sav M, King MA, Whitty JA, Kendall E, McMillan SS, Kelly F, Hunter B, Wheeler AJ. Burden of treatment for chronic illness: a concept analysis and review of the literature. Health Expect [Internet] 2013 [p. 13]. Available from: http://onlinelibrary.wiley. com/doi/10.1111/hex.12046/full [last accessed 5 Nov 2013]. 5. Vollmer WM, Markson LE, O’ Connor E, Sanocki LL, Fitterman L, Berger M, Buist AS. Association of asthma control with healthcare utilization and quality of life. Am J Respir Crit Care Med 1999;160: 1647–1652. 6. Diette GB, Sajjan SG, Skinner EA, Weiss TW, Wu AW, Markson LE. Using the pediatric asthma therapy assessment questionnaire to measure asthma control and healthcare utilization in children. Patient 2009;2:233–241. 7. Gandhi PK, Kenzik KM, Thompson LA, DeWait DA, Revicki DA, Shenkman EA, Huang IC. Exploring factors influencing asthma control and asthma-specific health-related quality of life among children. Respir Res 2013;14:1–10. 8. Lara M, Rosenbaum S, Rachelefsky G, Nicholas W, Morton SC, Emont S, Branch M, et al. Improving childhood asthma outcomes in the United States: a blueprint for policy action. Pediatrics 2002; 109:919–930. 9. Oraka E, Iqbal S, Flanders WD, Brinker K, Garbe P. Racial and ethnic disparities in current asthma and emergency department
Asthma coalition outcomes for vulnerable children
DOI: 10.3109/02770903.2014.891608
10.
11. 12. 13.
14. 15.
16.
visits: findings from the National Health Interview Survey, 2001– 2010. J Asthma 2013;50:488–496. Leong AB, Ramsey CD, Celedon JC. The challenge of asthma in minority populations. In: Gershwin ME, Albertson TE, editors. Bronchial Asthma: a guide for practical understanding and treatment. New York: Springer Publishing; 2012:357–383. Andrews KL, Jones SC, Mullan J. Asthma self-management in adults: a review of current literature. Collegian 2013 [Epub ahead of print]. Hennessy-Harstad E. Asthma and adolescents: review of strategies to improve control. J Sch Nurs 2012;29:39–51. Clark NM, Doctor LJ, Friedman AR, Lachance LL, Houle CR, Geng X, Grisso JA. Community coalitions to control chronic disease: allies against asthma as a model and case study. Health Promot Pract 2006;7:14S–22S. Clark NM, Mitchell HE, Rand CS. Effectiveness of educational and behavioral asthma interventions. Pediatrics 2009;123:S185–S192. Clark NM, Lachance L, Doctor LJ, Gilmore L, Kelly C, Krieger J, Lara M, et al. Policy and system change and community coalitions: outcomes from Allies against Asthma. Am J Public Health 2010; 100:904–912. Clark NM, Lachance LL, Benedict MB, Doctor LJ, Gilmore L, Kelly CS, Krieger J, et al. Improvements in health care use
17. 18.
19.
20. 21.
479
associated with community coalitions: long-term results of the Allies Against Asthma initiative. Am J Public Health 2013;103: 1124–1127. Moorman JE, Akinbami LJ, Bailey CM, Zahran HS, King ME, Johnson CA, Liu X. National surveillance of asthma: United States, 2001-2010. Vital Health Stat 3. 2012;(35):1–67. Lachance LL, Houle CR, Cassidy EF, Bourcier E, Cohn JH, Orians CE, Coughey K, et al. Collaborative design and implementation of a multisite community coalition evaluation. Health Promot Pract 2006;7(2 Suppl):44S–55S. National Quality Measures Clearinghouse (NQMC). Measure summary: use of appropriate medicines for people with asthma. In: National Quality Measures Clearinghouse (NQMC). Rockville (MD): Agency for Healthcare Research and Quality (AHRQ). Available from: http://www.qualitymeasures.ahrq.gov. Prentice RL, Williams BJ, Peterson AV. On the regression analysis of multivariate failure time data. Biometrika 1981;68: 373–379. The analysis for this paper was generated using SAS/STAT software, Version 9.2 of the SAS System for Windows. Copyright ß 2008 SAS Institute Inc. SAS and all other SAS Institute Inc. product or service names are registered trademarks or trademarks of SAS Institute Inc., Cary, NC, USA.
Copyright of Journal of Asthma is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
UNDERSERVED POPULATIONS
Asthma coalition effects on vulnerable sub groups of children: the most frequent users of health care and the youngest Laurie Lachance, PhD, MPH1, M. Beth Benedict, DrPH, JD2, Linda Jo Doctor, MPH3, Lisa A. Gilmore, MBA, MSW4, Cynthia Kelly, MD, FAAP 5, James Krieger, MD, MPH 6, Marielena Lara, MD, MPH7, John Meurer, MD, MBA8, Amy Friedman Milanovich, MPH1, Elisa Nicholas, MD, MSPH9, Michael Rosenthal, MD10, Peter X. K. Song, PhD11, Shelley C. Stoll, MPH1, Daniel F. Awad, MA1, Margaret K. Wilkin, MPH1, and Noreen M. Clark, PhD1 1
Center for Managing Chronic Disease, University of Michigan, Ann Arbor. MI, USA, 2Centers for Medicare, Medicaid Services, Health & Human Services, Baltimore, MD, USA, 3W.K. Kellogg Foundation, Battle Creek, MI, USA, 4FHI 360, Durham, NC, USA, 5Eastern Virginia Medical School, Children’s Hospital of the King’s Daughters, Norfolk, VA, USA, 6Department of Public Health, Seattle & King County, Seattle, WA, USA, 7RAND Health, Santa Monica, CA, USA, 8Medical College of Wisconsin and Children’s Hospital and Health System, Milwaukee, WI, USA, 9The Children’s Clinic, Serving Children and Their Families, Long Beach, CA, USA, 10Department of Family & Community Medicine, Christiana Care Health Systems, Wilmington, DE, USA, and 11Department of Biostatistics, University of Michigan, Ann Arbor, MI, USA
Abstract
Keywords
Objective: To examine the impact of Allies Against Asthma, community-based coalitions working to improve asthma outcomes, on vulnerable children: those with the most urgent health care use and those of youngest age. Methods: Allies zip codes were matched with comparison communities on demographic factors. Five years of Medicaid data (n ¼ 26,836) for significant health care events: hospitalizations, ED and urgent care facility visits, were analyzed. Longitudinal analyses using generalized estimating equations and proportional hazards models compared Allies and comparison group children. Results: In the two start-up years of Allies, odds of having a significant event were greater for Allies children than for comparison children (p50.05). During the third and fourth years when Allies activities were fully implemented, for frequent health care users at baseline, odds of an asthma event were the same for both Allies and comparison children, yet in the less frequent users, odds of an event were lower in Allies children (p50.0001). In the initial year of Allies efforts, among the youngest, the Allies children had greater odds than comparison children of an event (p50.01), but by the fourth year the Allies group had lower odds (p ¼ 0.02) of an event. Hazard ratios over all years of the study for the youngest Allies children and most frequent baseline users of urgent care were lower than for comparison children (p ¼ 0.01 and p ¼ 0.0004). Conclusion: Mobilizing a coalition of diverse stakeholders focused on policy and system change generated community-wide reductions over the long-term in health care use for vulnerable children.
Management/control, morbidity and mortality, pediatrics, prevention, quality of life
Introduction Children experiencing significant asthma health care events defined as a need for urgent care services (hospitalizations, ED visits, urgent care facility visits), and the youngest patients comprise vulnerable asthma subgroups in the pediatric population [1,2]. Frequent urgent health care utilization and younger age have been shown to be associated with more disease burden for children with asthma, their families and the health care system [3–7]. High rates of urgent care use for asthma are associated with residing in low income and minority communities [8–10]. A variety of ways have been examined in which to improve services, provide selfmanagement support to patients and reduce the demand for Correspondence: Laurie Lachance, PhD, Center for Managing Chronic Disease, University of Michigan, 1415 Washington Heights, Ann Arbor, MI 48109-2029, USA. Tel: +73 47631457. Fax: +73 47639115. E-mail: [email protected]
History Received 15 August 2013 Revised 24 January 2014 Accepted 30 January 2014 Published online 7 March 2014
more expensive asthma care [11–12]. One way discussed in the literature is formation of community-wide asthma coalitions [13]. These are collaborative efforts at the local level to achieve asthma control by bringing together diverse stakeholders from health care providers to grassroots groups to voluntary and government agencies to business and industry who collectively implement strategies to improve asthma outcomes. Currently, there are over 200 asthma coalitions in the U.S [14]. One such effort focused on low income communities: Allies Against Asthma. This national program supported by the Robert Wood Johnson Foundation, has been described in detail in previous publications [15,16]. Evaluation of the Allies coalitions demonstrated that such collective effort can bring about significant policy and system change related to clinical care and management of asthma and that in a cohort of asthma patients followed over one year, significant symptom reduction was evident [15]. Further, using Medicaid data, the Allies study showed that improved asthma outcomes were community
DOI: 10.3109/02770903.2014.891608
wide, that is, an enhanced health status of the population of low income children residing in target communities [16]. It has been shown that Allies coalitions had a broad effect. The study presented here examined Allies coalition effects on a subgroup of low income children and the most vulnerable children with asthma. It addressed the question: were changes in health care use for children in the Allies communities seen among the most frequent users of urgent care services and the youngest children.
Methods All study procedures were approved by the University of Michigan Institutional Review Board. To conduct this study, the Center for Managing Chronic Disease (CMCD) of the University of Michigan entered into collaboration with the Centers for Medicare and Medicaid Services (CMS) of the federal government. CMS provided zip code level data to CMCD and consultation regarding data analysis and variables related to children with asthma covered by Medicaid insurance provisions from 2002 to 2006. Samples Allies coalitions worked in low income neighborhoods in the following six cities across the United States: Hampton Roads, Virginia; Washington, D.C.; Milwaukee, Wisconsin; King County/Seattle, Washington; Long Beach, California; Philadelphia, Pennsylvania. Allies coalitions provided the zip codes for neighborhoods that comprised their focus, that is, the areas of greatest coalition activity. Expert consultants to the CMCD then used a set of criteria to identify matched zip codes in areas where Allies coalitions were not active. For each Allies zip code, one or two demographically similar zip codes (according to Census 2000) were identified to serve as comparison communities. Characteristics for selecting matched comparison groups were percentages of racial/ ethnic groups and median income, as these variables consistently are associated with asthma prevalence. [17] Secondary variables used in matching were total population size, percentage of family households and percentage of population younger than 18 years. Geographic considerations were also made (e.g. rural versus urban settings) in the matching. The non-Allies comparison zip codes were located in Roanoke City, Virginia; Jacksonville, Florida; Everett, Lacey, Olympia and Tacoma, Washington; National City and San Bernardino, California; Baltimore, Maryland; Lorain, Ohio; Muskegon, Detroit, and Flint Michigan; Fort Wayne and Indianapolis, Indiana [18]. The files from CMS for children in the intervention and comparison sites included all of the Medicaid fee-for-service claims and encounter claims from managed care patients submitted in the study years. A person-specific cohort data file was constructed for analysis based on the following inclusion criteria: (1) aged 2–18 on 1 January 2002; (2) not enrolled as disabled; (3) enrolled in 2002 and at least one year follow-up, 2004–2006; and (4) in the baseline year, had at least one health care visit with a principal or secondary diagnosis of asthma (ICD-9 codes 493.XX) to include a hospitalization, a visit to an emergency department, an urgent care facility, or to a physician’s office; or a filled prescription
Asthma coalition outcomes for vulnerable children
475
for at least one of the following asthma medications from HEDIS guidelines [19]: inhaled corticosteroids, inhaled corticosteroid and long acting beta agonist combination, leukotriene modifier, long-acting inhaled beta-2 agonist, mast cell stabilizer, methylxanthine, short-acting inhaled beta-2 agonist. This produced data on 26 836 children for analysis. Data sources Four types of CMS files for each year were examined for inclusion in the study: inpatient (hospitalizations); other services (physician’s visit, emergency department visits (ED) and visits to an urgent care facility), prescription drug record, and person summary (demographic information). Data fields included beginning and ending date of service, diagnosis (asthma principal or secondary), type of service, where service was provided (hospital, ED, urgent care facility), National Drug Code (NDC) for filled prescriptions, and eligibility and payment. Data management and analysis First, the sample was divided by (a) frequency of a significant health care event (with frequent urgent care users defined as those with at least one hospitalization or two emergency department or urgent care facility visits in the baseline year) and (b) age group (ages 2–3, 4–6 and older than 6 years in the baseline year). Then, the annual health care use in each group for hospitalizations, ED and urgent care facility visits were described. Due to the low frequency of these events, a ‘‘significant asthma event’’ was defined as presence or absence of any urgent care visit in a given year and compared using frequency tables where statistical significance was evaluated by Pearson chi-square test. Stratified analyses were conducted to determine differences between Allies and comparison children with the most frequent urgent care utilization and between Allies and comparison children with less frequent use; similar stratified analysis between intervention groups was also conducted by age group. Finally, we conducted longitudinal data analysis that focused on covariates believed to be associated with the outcome of ‘‘significant asthma event.’’ Adjusted odds ratios (OR) comparing the odds of having a ‘‘significant asthma event’’ between the intervention group and the comparison group in each year of follow-up were obtained in a longitudinal generalized estimating equation model through an interaction between year and intervention group. This method allows each OR to be controlled for the other years, including baseline. Baseline age, gender and race/ethnicity were also controlled for in the models (age was excluded in the model comparing age groups). To examine the effect of the Allies interventions over the entire study period, 2002–2006, a recurrent event analysis approach using Cox proportional hazards model [20] was undertaken to analyze the time to ‘‘significant asthma event’’ from previous events. This approach allowed comparison of the hazard of an event between intervention and comparison groups to be conditional on previous asthma events and also to control for demographic and other potentially confounding factors. This analysis was conducted separately in each significant event group and age group to determine if the intervention
476
L. Lachance et al.
J Asthma, 2014; 51(5): 474–479
effect was similar for each stratum. SAS/STAT software, Version 9.2 of the SAS System for Windows was used for analysis [21].
Results The differences in demographic factors between significant event groups and between age groups were few. There were more black children, males and children aged 2–4 with more frequent health care use compared to those with less frequent use. By age, there were more males in the 2–3-year-old group and also more significant asthma events in this group. These findings are consistent with those reported in previous studies [8–10]. Emergency or urgent visits and hospitalizations were relatively rare events. Tables 1 and 2 show proportions with ED or urgent care visits and hospitalizations by utilization status and age group. Ten percent or less of the most frequent urgent care users had a hospitalization in any follow-up year and from 14–31% had an ED or urgent care visit. Around 1–2% of those with less frequent urgent care use and across age groups were hospitalized in follow-up years. Emergency or urgent care visits occurred in 3.8% to 9.4% of subjects across age groups. This was similar for those less frequent urgent care users.
Tables 3 and 4 display adjusted odds of having a significant asthma event in the Allies group versus the comparison group in each year of the study by utilization status (Table 3) and by age (Table 4). In the frequent users of urgent care, in 2003 and 2004, the Allies group had almost 50% greater odds of having a significant event than the comparison group while in years, 2005 and 2006 the odds of having an event were the same in both groups. In the less frequent users, the Allies group had significantly greater odds (25%–40%) of having an asthma event than comparison children in years 2003 and 2004, and significantly lower odds (20%–25%) in 2005 and 2006 (see Table 3) when adjusted for age, gender and race/ethnicity. In all age groups, the Allies group children were more likely to have an asthma event in 2003 and 2004, but less likely in 2005 and 2006, with a 47% reduced odds ratio from 2003 to 2006 in those aged 2-3 (p value ¼ 0.0002), 41% reduced odds ratio from 2003 to 2006 in those aged 4–6 (p value ¼ 0.0012) and 34% reduced odds ratio from 2003 to 2006 in those older than age 6 (p value50.0001) (Table 4) adjusting for an interaction between year and intervention group and for gender and race/ethnicity. Table 5 presents the estimated hazard ratios for the Allies group versus the comparison group, which were obtained by
Table 1. Number and percent of children with at least one urgent health care visit by year and baseline urgent care utilization. Frequent urgent care usersa in baseline year (2002) (n ¼ 1039)
2003, at least one: ED or urgent care Hospitalization 2004, at least one: ED or urgent care Hospitalization 2005, at least one: ED or urgent care Hospitalization 2006, at least one: ED or urgent care Hospitalization
Less or no urgent care use in baseline year (2002) (n ¼ 25 797)
Allies N (%)
Comparison N (%)
Allies N (%)
Comparison N (%)
106 (30.7%) 30 (8.7%)
127 (19.6%) 65 (10.0%)
727 (6.3%) 115 (1.0%)
571 (4.3%) 202 (1.5%)
75 (25.3%) 12 (4.1%)
76 (14.0%) 47 (9.2%)
592 (5.9%) 77 (1.0%)
377 (3.7%) 140 (1.4%)
46 (18.4%) 16 (6.4%)
75 (18.0%) 34 (8.2%)
547 (6.3%) 69 (1.0%)
665 (7.6%) 125 (1.4%)
46 (22.3%) 11 (5.3%)
80 (21.9%) 25 (6.8%)
411 (5.4%) 59 (1.0%)
520 (7.0%) 93 (1.2%)
a
Defined as 2 emergency department or urgent care center visits OR 1 hospitalization for asthma in the baseline year, 2002.
Table 2. Number and percent of children with at least one urgent health care visit by year and baseline age group. Ages 2–3 in 2002 (n ¼ 4829)
2003, at least one: ED or urgent care Hospitalization 2004, at least one: ED or urgent care Hospitalization 2005, at least one: ED or urgent care Hospitalization 2006, at least one: ED or urgent care Hospitalization
Ages 4–6 in 2002 (n ¼ 5610)
Older than age 6 in 2002 (n ¼ 16 397)
Allies N (%)
Comparison N (%)
Allies N (%)
Comparison N (%)
Allies N (%)
Comparison N (%)
209 (9.4%) 36 (1.6%)
143 (5.9%) 68 (2.8%)
162 (6.5%) 31 (1.2%)
150 (5.1%) 62 (2.1%)
462 (6.4%) 78 (1.1%)
405 (4.7%) 137 (1.6%)
145 (7.5%) 15 (1.0%)
88 (4.8%) 43 (2.4%)
144 (6.6%) 26 (1.2%)
110 (4.9%) 38 (1.7%)
378 (6.1%) 48 (1.0%)
255 (3.8%) 106 (1.6%)
118 (7.2%) 17 (1.0%)
111 (7.2%) 28 (1.8%)
126 (6.7%) 17 (1.0%)
170 (8.8%) 40 (2.1%)
349 (6.4%) 51 (1.0%)
459 (8.0%) 91 (1.6%)
82 (5.7%) 18 (1.2%)
101 (7.6%) 18 (1.4%)
97 (5.6%) 14 (1.0%)
140 (8.1%) 28 (1.6%)
278 (6.0%) 38 (1.0%)
359 (7.5%) 72 (1.5%)
Asthma coalition outcomes for vulnerable children
DOI: 10.3109/02770903.2014.891608
477
Table 3. Stratified comparison of adjusted oddsa of having a significant asthma eventb in a given year in the Allies group children versus the comparison group children. Frequent urgent care usersc Allies versus Comparison (n ¼ 1,039) Year
OR
2003 2004 2005 2006
1.503 1.458 1.005 1.090
95% CI (1.126, (1.039, (0.678, (0.731,
2.007) 2.046) 1.491) 1.624)
Less or no urgent care use Allies versus Comparison (n ¼ 25,797) p
OR
0.0057 0.0291 0.9786 0.6719
1.280 1.460 0.785 0.761
95% CI (1.160, (1.290, (0.700, (0.669,
p 50.0001 50.0001 50.0001 50.0001
1.433) 1.653) 0.881) 0.867)
a
Adjusted for age, gender and race/ethnicity. A hospitalization, ED, or urgent care facility visit. c Defined as 2 emergency department or urgent care center visits OR 1 hospitalization for asthma in the baseline year, 2002. b
Table 4. Age stratified comparison of adjusted oddsa of having a significant asthma eventb in a given year between the Allies and comparison children. Allies group versus Comparison group Ages 2–3 in 2002 (n ¼ 4829) a
Year
OR
2003 2004 2005 2006
1.33 1.25 0.92 0.70
95% CI (1.08, (0.97, (0.71, (0.52,
1.63) 1.61) 1.20) 0.95)
Ages 4–6 in 2002 (n ¼ 5610) a
p
OR
0.0077 0.0922 0.5525 0.0198
1.11 1.27 0.68 0.66
95% CI (0.89, (0.99, (0.54, (0.51,
1.37) 1.61) 0.86) 0.86)
Older than age 6 in 2002 (n ¼ 16 397) p
ORa
0.3513 0.0561 0.0012 0.0018
1.23 1.46 0.79 0.81
95% CI (1.08, (1.26, (0.69, (0.70,
1.40) 1.71) 0.91) 0.95)
p 0.0018 50.0001 0.0010 0.0101
a
Adjusted for gender and race/ethnicity. A hospitalization, ED, or urgent care facility visit.
b
Table 5. Stratified comparison of hazard of having a significant asthma eventa over the entire study period from 2002–2006, adjusted for age (only in the frequent users versus less frequent users model), gender and race/ethnicity.
Frequent urgent care user children (n ¼ 1039) Less frequent user children (n ¼ 25 797) Children ages 2–3 Children ages 4–6 Children older than 6 years
b
Hazard ratio for significant asthma eventa (95% confidence interval) Allies group versus Comparison group
p Value
0.88 (0.812, 0.943) 1.03 (0.990, 1.080) 0.903 (0.834–0.979) 0.982 (0.906–1.064) 0.941 (0.896–0.989)
0.0004 0.129 0.0132 0.6572 0.0166
a
A hospitalization, ED, or urgent care facility visit. Defined as 2 emergency department or urgent care center visits OR 1 hospitalization for asthma in the baseline year, 2002.
b
the recent event data analysis of time to an asthma event, adjusting for age (only in the urgent care user models), gender and race/ethnicity. Examination across the entire study period, from 2002–2006, showed that the Allies group children had a lower hazard of having a hospitalization, ED visit, or urgent care facility visit than the comparison group children for the frequent user cohort and for children aged 2–3 or older than 6. Given a hazard ratio of 0.88 in the frequent user cohort, Allies children had approximately 14% longer average time to event than comparison children. In the less frequent users and those aged 4–6, the hazard was equal in both the intervention and comparison group children, after the adjustment for age (in the more urgent care versus less urgent care model), gender, and race/ethnicity (see Table 5).
Discussion The data identifies a clear progression toward more effective asthma control within vulnerable sub-groups of the
population of children with asthma, residing in low income Allies communities. Therefore, the effects of Allies coalitions on significant asthma health care events illustrate a pattern that aligns with the maturation of coalitions as organizations trying to effect change. In the start-up years of functioning (2003–2004), urgent care use actually increased in the Allies communities. It is likely that greater focus of coalition stakeholders to call attention to asthma as a serious community problem, along with the need to introduce changes to enhance services, created more awareness and stimulated care seeking by targeted families. Two years later, when activities of the coalitions’ policy and system changes had time to evolve, children with frequent urgent care use in Allies communities had reduced their frequency of care to the same level of that in comparison communities. By that time point, Allies less frequent urgent care users at baseline experienced significantly fewer health care events. The same pattern was evident in all age groups, however, with greater maturation of the coalition efforts over time, younger children in the Allies
478
L. Lachance et al.
group had the greatest reductions in the odds of need for health care. An obvious interpretation of these data is that community wide change doesn’t happen overnight, even when a wide range of interested stakeholders mobilize to redesign services and education and introduce system and policy change. However, these data do illustrate that given a time period as short as two years, claims for urgent health care use was reduced among vulnerable children. In other words, it was in the later years that significant results were apparent. When viewed across the study years, the hazard ratios for the most frequent urgent care users at baseline and the youngest children were significantly improved in Allies communities. The findings suggest that although the greatest effects coincided with coalitions’ maturation, some impact may have been evident from the outset. There are potential limitations to this study. Average intervention effects over the entire time period were not seen in the less frequent users of urgent services or in the older age groups. In these groups, the odds ratios by year were in favor of the comparison group in the first two years of start up and in favor of the Allies group in the third and fourth years of follow-up. This may have led to a masking of the true hazard over the entire time period. Claims from both the Allies and comparison sites may not have included all Medicaid enrolled children during the study years because not all encounter claims from managed care patients are submitted to CMS nor are all children continuously enrolled in Medicaid. To understand effects, if any, related to these phenomena, in the proportional hazards model, children continuously enrolled in Medicaid were compared to children that may have had a gap in enrollment. This comparison showed very comparable, statistically significant results, which suggested that the assumption of no occurrences of events in the gap time was reasonable in the recurrent event data analysis. Another limitation is that the level of exposure of any given child to the various components of service improved by the Allies coalitions and policy and systems changes initiated is not known. It is known, however, that over time, children who were frequent urgent care users and those who were younger living in the areas where changes were introduced subsequently had fewer significant asthma care events, an indication that whatever the level of exposure the outcome was positive. An important aspect of this study is that it reports community-wide results for children participating in Medicaid across the Allies neighborhoods. Given the large amount of data available to in this study, a longitudinal, cohort analysis could be undertaken. This method provided the strong and clear pattern of intervention effects over time.
Conclusion The findings from this study indicate that mobilizing diverse stakeholders, including consumers, and focusing on policy and system changes can generate significant reductions in health care use for vulnerable children with asthma. The Allies coalitions generated important community benefits. Claims for Health care utilization for the highest users of urgent services declined significantly as did use by the
J Asthma, 2014; 51(5): 474–479
youngest children, two important subgroups in the pediatric population. Community-wide efforts to engage stakeholders in initiating far reaching outcomes through systemic change can greatly benefit poor children across targeted neighborhoods.
Acknowledgements We gratefully acknowledge the guidance and contributions of Allies Advisory board members and consultants Michele Carrick, MSW, Elaine Cassidy, PhD, Mary desVignesKendrick, MD, MPH, Susan Downey, M.Ed., Rachel A. Gonzales-Hanson, Rob Fulwood, PhD, JA Grisso, MPH, MD, Barbara Israel, Ph.D., Talmadge King, MD, Floyd Malveaux, MD, PhD, Robert Mellins, MD, Steve Page, MPA, Guy Parcel, PhD, Stephen Redd, MD, Jeanne Taylor, PhD, Abe Wandersman, PhD, Sandra Wilson, PhD, Albert Yee, MD, Sarah Hearn, Ye Yang, and Charles Brinker. Additionally, we thank Christine Joseph, PhD and her team for matches of Allies zip codes. We also wish to thank the dedicated members of all the Allies coalitions. This work is dedicated to the memory of Dr. Noreen M. Clark, Director of the Center for Managing Chronic Disease, University of Michigan School of Public Health and to the memory of Dr. M. Beth Benedict, retired from the Centers for Medicare, Medicaid Services, Health & Human Services with gratitude for their vision and guidance.
Declaration of interest This study was supported by the Robert Wood Johnson Foundation, with additional support provided by the W.K. Kellogg Foundation. The authors have no further relevant interests to disclose.
References 1. Clement LT, Jones CA, Cole J. Health disparities in the United States: childhood asthma. Am J Med Sci 2008;335:260–265. 2. Federico MJ, Liu AH. Overcoming childhood asthma disparities of the inner-city poor. Pediatr Clin North Am 2003;50:655–675. 3. Bateman ED, Bousquet J, Keech ML, Busse WW, Clark TJ, Pedersen SE. The correlation between asthma control and health status: the GOAL study. Eur Respir J 2007;29:56–62. 4. Sav M, King MA, Whitty JA, Kendall E, McMillan SS, Kelly F, Hunter B, Wheeler AJ. Burden of treatment for chronic illness: a concept analysis and review of the literature. Health Expect [Internet] 2013 [p. 13]. Available from: http://onlinelibrary.wiley. com/doi/10.1111/hex.12046/full [last accessed 5 Nov 2013]. 5. Vollmer WM, Markson LE, O’ Connor E, Sanocki LL, Fitterman L, Berger M, Buist AS. Association of asthma control with healthcare utilization and quality of life. Am J Respir Crit Care Med 1999;160: 1647–1652. 6. Diette GB, Sajjan SG, Skinner EA, Weiss TW, Wu AW, Markson LE. Using the pediatric asthma therapy assessment questionnaire to measure asthma control and healthcare utilization in children. Patient 2009;2:233–241. 7. Gandhi PK, Kenzik KM, Thompson LA, DeWait DA, Revicki DA, Shenkman EA, Huang IC. Exploring factors influencing asthma control and asthma-specific health-related quality of life among children. Respir Res 2013;14:1–10. 8. Lara M, Rosenbaum S, Rachelefsky G, Nicholas W, Morton SC, Emont S, Branch M, et al. Improving childhood asthma outcomes in the United States: a blueprint for policy action. Pediatrics 2002; 109:919–930. 9. Oraka E, Iqbal S, Flanders WD, Brinker K, Garbe P. Racial and ethnic disparities in current asthma and emergency department
Asthma coalition outcomes for vulnerable children
DOI: 10.3109/02770903.2014.891608
10.
11. 12. 13.
14. 15.
16.
visits: findings from the National Health Interview Survey, 2001– 2010. J Asthma 2013;50:488–496. Leong AB, Ramsey CD, Celedon JC. The challenge of asthma in minority populations. In: Gershwin ME, Albertson TE, editors. Bronchial Asthma: a guide for practical understanding and treatment. New York: Springer Publishing; 2012:357–383. Andrews KL, Jones SC, Mullan J. Asthma self-management in adults: a review of current literature. Collegian 2013 [Epub ahead of print]. Hennessy-Harstad E. Asthma and adolescents: review of strategies to improve control. J Sch Nurs 2012;29:39–51. Clark NM, Doctor LJ, Friedman AR, Lachance LL, Houle CR, Geng X, Grisso JA. Community coalitions to control chronic disease: allies against asthma as a model and case study. Health Promot Pract 2006;7:14S–22S. Clark NM, Mitchell HE, Rand CS. Effectiveness of educational and behavioral asthma interventions. Pediatrics 2009;123:S185–S192. Clark NM, Lachance L, Doctor LJ, Gilmore L, Kelly C, Krieger J, Lara M, et al. Policy and system change and community coalitions: outcomes from Allies against Asthma. Am J Public Health 2010; 100:904–912. Clark NM, Lachance LL, Benedict MB, Doctor LJ, Gilmore L, Kelly CS, Krieger J, et al. Improvements in health care use
17. 18.
19.
20. 21.
479
associated with community coalitions: long-term results of the Allies Against Asthma initiative. Am J Public Health 2013;103: 1124–1127. Moorman JE, Akinbami LJ, Bailey CM, Zahran HS, King ME, Johnson CA, Liu X. National surveillance of asthma: United States, 2001-2010. Vital Health Stat 3. 2012;(35):1–67. Lachance LL, Houle CR, Cassidy EF, Bourcier E, Cohn JH, Orians CE, Coughey K, et al. Collaborative design and implementation of a multisite community coalition evaluation. Health Promot Pract 2006;7(2 Suppl):44S–55S. National Quality Measures Clearinghouse (NQMC). Measure summary: use of appropriate medicines for people with asthma. In: National Quality Measures Clearinghouse (NQMC). Rockville (MD): Agency for Healthcare Research and Quality (AHRQ). Available from: http://www.qualitymeasures.ahrq.gov. Prentice RL, Williams BJ, Peterson AV. On the regression analysis of multivariate failure time data. Biometrika 1981;68: 373–379. The analysis for this paper was generated using SAS/STAT software, Version 9.2 of the SAS System for Windows. Copyright ß 2008 SAS Institute Inc. SAS and all other SAS Institute Inc. product or service names are registered trademarks or trademarks of SAS Institute Inc., Cary, NC, USA.
Copyright of Journal of Asthma is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
