Asthma is a major health problem affecting 17. Quality and Access to Care Among a Cohort of Inner-city Adults With Asthma*

Transcription

1 Quality and Access to Care Among a Cohort of Inner-city Adults With Asthma* Who Gets Guideline Concordant Care? Ethan A. Halm, MD, MPH; Juan P. Wisnivesky, MD, MPH; and Howard Leventhal, PhD Study objectives: Asthma morbidity is highest among inner-city populations. This study measured whether quality and access to care over time was concordant with National Asthma Education and Prevention Program (NAEPP) guidelines. It also identified factors associated with NAEPP guideline-concordant care. Design: A prospective, observational cohort study Setting: An urban academic medical center. Patients: A consecutive cohort of 198 inner-city adults hospitalized for asthma. Measurements: Detailed information about sociodemographics, asthma history, access to care, history of the current exacerbation, prescription and use of inhaled corticosteroids (ICS) and -agonists, and other elements of NAEPP-concordant care (spacers, metered-dose inhaler [MDI] technique, peak flow meters, and action plans) was collected during the index admission and 1 month and 6 months after discharge. Results: In this predominantly low-income, nonwhite cohort, while 92% of patients had insurance and 80% had a usual source of care, 73% reported delays in seeking care. ICS were prescribed for 77% of patients prior to hospital admission, 83% at 1 month, and 67% at 6 months. Adherence with other NAEPP recommendations were 89% for receipt of MDI instruction, 68% for spacers, 80% for peak flow meters, 31% for written action plans for worsening, and 22% for written plans for attacks. In multivariate analysis, greater asthma severity and having a usual source of care increased the odds of receiving ICS, spacers, and peak flow meters. Care by a specialist increased the odds of receiving action plans. Patients who spoke mostly Spanish were less likely to be given spacers or action plans. Conclusion: Baseline problems with quality and access to care persisted over time. Better systems of care are needed to ensure that high-risk patients receive an appropriate step-up in the quality of ongoing asthma care. (CHEST 2005; 128: ) Key words: access; adherence; asthma; guidelines; management; quality Abbreviations: ED emergency department; ICS inhaled corticosteroids; MDI metered-dose inhaler; NAEPP National Asthma Education and Prevention Program *From the Division of General Internal Medicine, Department of Medicine (Drs. Halm and Wisnivisky), Mount Sinai School of Medicine, New York, NY; and Institute for Health, Health Care Policy and Aging Research (Dr. Levanthal), Rutgers University, New Brunswick, NJ. This study was funded by the Agency for Healthcare Research and Quality (RO1 HS09973) and the United Hospital Fund (010608B). Dr. Halm was also supported by the Robert Wood Johnson Foundation Generalist Physician Faculty Scholars Program and Dr. Wisnivesky by the Agency for Healthcare Research and Quality (K08 HS O1A1). Asthma is a major health problem affecting 17 million Americans at a cost of $11 billion per year. 1,2 Minority inner-city populations have disproportionately higher rates of asthma incidence, morbidity, and mortality. 1,3,4 Rates of hospitalizations and emergency department (ED) visits are greatest in the Northeast, especially in cities such as New York. 1,4 East Harlem, in New York City, is one of the most severely affected communities in the United Manuscript received November 9, 2004; revision accepted April 20, Reproduction of this article is prohibited without written permission from the American College of Chest Physicians ( org/misc/reprints.shtml). Correspondence to: Ethan A. Halm, MD, MPH, Division of General Internal Medicine, Mount Sinai School of Medicine, Box 1087, One Gustave L. Levy Place, New York, NY 10029; ethan.halm@mountsinai.org CHEST / 128 / 4/ OCTOBER,

2 States, with asthma hospitalization and death rates that are several times higher than the national average. 5 7 There is overwhelming evidence that inhaled corticosteroids (ICS) reduce symptoms, functional limitations, and health-care utilization due to asthma. 8 For over a decade, there has also been consensus about best practice as outlined in the federal National Asthma Education and Prevention Program (NAEPP) guidelines. 9,10 Numerous government agencies, health plans, and delivery systems have worked to disseminate and implement these guidelines. Despite the confluence of evidence, consensus, and health policy focus, the quality of asthma care in adults is often suboptimal The quality problems of underuse of ICS and overuse of short-acting -agonists also appears to be worse in nonwhites. 13,18,20 Compliance with NAEPP guideline recommendations for self-management education and trigger avoidance was also lower in African Americans. 18,20 Most studies 13 15,17 19 that have measured the quality of care of asthma care have primarily focused on ambulatory populations and were cross-sectional in nature. The few studies 11,12,16 that focused on those sick enough to require emergency treatment or hospitalization have also reported gaps in quality, although these were also cross-sectional designs. There is scant information about whether the quality and comprehensiveness of asthma care improves following an exacerbation serious enough to merit hospitalization, especially in communities with the highest asthma morbidity. The main goals of this study were as follows: (1) to comprehensively measure the quality and access to care in a cohort of inner-city adults hospitalized for asthma, and assess if care is concordant with NAEPP guidelines; (2) to determine whether provision of guideline-concordant care improves over time; and (3) to identify the socioeconomic, clinical, and access to care factors that are independently associated with receipt of NAEPP guideline-concordant care. We hypothesized that hospitalization might be a sentinel event that would provide additional motivation to patients and physicians to intensify treatment and self-management education. Study Participants Materials and Methods We prospectively identified an inception cohort of all adults hospitalized for asthma at Mount Sinai Hospital, a 1,100-bed academic health center in New York City, during a consecutive 12-month period (September 2001 through September 2002). Mount Sinai Hospital is the largest hospital serving East Harlem. We screened computerized hospital admission logs to identify all adults with a primary or secondary admission diagnosis of asthma (493, 493.X, and 493.XX). The study was approved by the local Institutional Review Board. Inclusion and Exclusion Criteria Eligible patients were 18 years old, spoke English or Spanish, were competent to give informed consent, and had asthma as the primary reason for hospital admission (confirmed by chart review). Exclusions were primary COPD, other lung disease, or home oxygen therapy; and hospital admission and discharge on the same weekend when study personnel were not available. Data Collection and Measurement Trained research staff conducted an interviewer-administered survey in English or Spanish during the index admission and 1 month and 6 months after discharge. Domains of interest included the following: sociodemographics, asthma history, access to care, history of the current exacerbation, and process of care (prescription and use of ICS and short-acting -agonists, spacers, and peak flow meters, and action plans, among others). We defined guideline-concordant care as the processes of care outlined in the NAEPP including: prescription of ICS, meter dose inhaler (MDI) instruction, spacers, peak flow meter monitoring, and action plans. 10 All patients in the study met NAEPP indications for ICS as well as all of the above elements of care. We distinguished action plan advice that was written vs oral, as well as those for worsening symptoms vs frank attacks. We considered use of nine or more puffs per day of a short-acting -agonist to be an indicator of poor control and potential overuse, a definition used previously. 13,21 Statistical Analysis Mean SDs are presented for normal data, and medians are presented for nonnormal data. We used 2 tests, Wilcoxon rank-sum tests, t tests, and Cochrane-Mantel-Haenzel trend tests (as appropriate) to examine changes in quality measures over time and univariate predictors of NAEPP guideline-concordant care. We used logistic regression to identify multivariate predictors of guideline-concordant care. We chose to examine each of the six quality indicators as separate dependent variables (rather than constructing a global quality score) because they represent considerably different behaviors and the results were simpler to interpret. Because many of the independent variables were highly correlated, we first identified the most robust independent predictors of guideline-concordant care within each domain (sociodemographics, asthma history, psychosocial, and access to care). The best candidate factors significant at the p 0.2 level for each domain were carried forward into the final multivariate models. Alternate multivariate models that also adjusted for age, sex, and/or other severity measures (age of asthma diagnosis, frequency of oral steroid use, prior hospitalization/ed visits) produced similar results. The final multivariate models for the provision of MDI instruction, spacers, and peak flow meters needed to omit one of the five final covariates (either prior intubation or specialty provider) because there was quasiseparation of the data (100% of patients in these categories received the indicator resulting in a noninformative odds ratio and confidence interval). Omission of this one covariate from the models did not alter the overall findings. All analyses used two-tailed significance levels of p 0.05 and were conducted with statistical software (SAS version 9.0; SAS Institute; Cary, NC) Clinical Investigations

3 Results Patient Participation and Response Rates Of the 384 hospitalizations with a primary or secondary diagnosis of asthma, 335 were confirmed primary asthma admissions among 250 unique patients. Of these, 218 patients met eligibility criteria and 204 patients (94%) consented to participate. Reasons for exclusion were as follows: readmissions among study participants (n 85), active psychiatric problems (n 9), and admitted/discharged on the same weekend (n 23). Of the 204 who consented, 198 patients (97%) completed the baseline interview, 177 patients (87%) completed the 1-month survey, and 170 patients (84%) completed the 6-month survey. Interviews were conducted in Spanish for 28% of patients. Patient Characteristics and Asthma History The characteristics of the cohort are described in Table 1. Consistent with the epidemiology of innercity asthma, this was largely a group of low-income Hispanic and African-American women with high rates of prior intubation, oral steroid use, and asthma-related ED visits and hospitalizations. Access to Care and Delays Overall, 158 patients (80%) had a usual source of asthma care. Among the remaining 40 patients, the reasons for not having a source of care were as follows: not sick/asymptomatic (33%), moving around/new to area (19%), financial/insurance problems (14%), prefer ED care (7%), and do not trust doctors (7%). Most patients (89%) had been going to their usual place of care for at least 1 year. Two thirds of patients were able to name a specific physician as being in charge of their asthma care (79% generalists, 19% pulmonologists, and 2% allergists). Among those with a regular provider, 36% said it was hard to get advice over the telephone and 31% said it was hard to get an urgent appointment when they were sick. One hundred forty-five patients (73%) reported delays in seeking care for their exacerbation, and 76% (150 patients) never called a doctor or visited a clinic during this period of worsening. Among the 48 patients who did seek urgent outpatient care, 77% had a new medication started or regimen increased. Patients were sick for an average of 7.3 days before they first called/visited the doctor or came to the ED. Among patients who delayed seeking care, 48% thought things would get better on their own, 12% were afraid of being hospitalized, and 10% were not sure they were sick. Access to care improved modestly over time. At 6 Table 1 Characteristics of Study Patients (n 198)* Characteristics Data Age (range), yr (18 83) Female gender 78 Ethnicity Hispanic 62 Black 28 White, non-hispanic 3 Other 6 Insurance status Medicaid 53 Medicare 22 Commercial 16 Uninsured 8 Native language English 61 Spanish 39 Income $15,000/yr 64 Education high school 40 Asthma history Age of asthma onset 20 yr 64 Prior intubation 23 Prior oral steroid use 88 Oral steroids (all/most of time past yr) 31 ED visits in past 12 mo Hospitalizations in past 12 mo Hospitalizations in past 5 yr Comorbid conditions Allergic rhinitis 57 Diabetes mellitus 28 Gastroesophogeal reflux disease 24 Depression 36 Anxiety/panic disorders 32 Social habits Current smoker 24 Current problem drinking (alcohol) 12 Inhaled drug use past 12 mo 11 Access to care Usual source of asthma care 80 Regular asthma physician 65 General internist 79 Pulmonologist 19 Allergist 2 *Data are presented as mean SD or %. Marijuana, cocaine, heroin. months, 87% reported a usual source of care (vs 80% at baseline, p 0.10), 77% had a primary asthma physician (vs 65% at baseline, p 0.01), and 69% made routine visits for asthma even when asymptomatic (vs 56%, p 0.003). Underuse of ICS, MDI Instruction, and Spacers We found significant underprescription and underuse of ICS (Table 2). Among the 198 patients with indications for ICS, 77% (152 patients) had ICS prescribed prior to hospitalization. Among these 152 patients, 73% used ICS daily when they were symptomatic, vs 58% when they had no symptoms (p 0.05). Prescription and use of ICS increased CHEST / 128 / 4/ OCTOBER,

4 Table 2 Medication, Peak Flow Monitoring, and Self-Management Behaviors Over Time* Variables Baseline (n 198) 1 Month (n 177) 6 Months (n 170) ICS Currently prescribed 152 (77) 147 (83) 114 (67) Used when symptomatic 111 (73) 122 (83) 91 (80) Used with no symptoms 88 (58) 112 (76) 75 (66) Short-acting -agonist for symptom relief 190 (96) 175 (99) 165 (97) Used albuterol all or most of the time last 2 wk 98 (56) 95 (58) Used nine or more puffs of albuterol in a day last 2 wk 118 (62) 45 (26) 23 (38) Peak flow meter Have peak flow meter 158 (80) 168 (95) 160 (94) Used all or most days 43 (27) 69 (41) 57 (36) Never used peak flow meter to adjust medications 114 (72) 110 (69) Action plans/instructions Written instructions for when asthma worsens 61 (31) 49 (28) 80 (47) Used written instructions all or most of time when worse 29 (48) Oral instructions for when asthma worsens 129 (65) 143 (81) 138 (81) Used oral instructions when asthma was worse 83 (64) *Data are presented as No. (%). Patients who did the behavior in question all or most of the time. modestly in the month after discharge. However, at 6 months, fewer patients (67%) said they were using ICS; among these, only 66% used them regularly when asymptomatic. At baseline, nearly 176 patients (89%) said they had received instructions about proper MDI technique sometime prior to hospitalization. At the 1-month follow-up interview, 61% said that someone discussed MDI technique since their hospital discharge. At baseline, 68% (134 patients) had a spacer. Among those with spacers, only 44% patients used it all or most of the time. Overuse of Short-Acting -Agonists Nearly all patients had a short-acting -agonist inhaler at all three time periods (Table 2). Many patients appeared to overuse their -agonists. Six months after discharge, 48% used their -agonist every day and 38% used nine or more puffs per day in the past 2 weeks. Twenty-three percent used -agonists in the absence of any symptoms at least some of the time. Peak Flow Monitoring At baseline, 80% (158 patients) said they had a peak flow meter for use at home. Among those with peak flow meters, while 84% said it was easy to use, only 27% used it all or most days; 72% never used their peak flow meter to judge if their medicines needed adjustment. At the 6-month interview, nearly all patients reported having a peak flow meter, although only 36% used it all or most days, and 69% never used a flow meter to adjust medications. The main reason for not using a peak flow meter was no symptoms/feeling okay. Self-Management Action Plans At baseline, only 31% (61 patients) had a written action plan about handling worsening symptoms, and 22% (44 patients) had a written plan for attacks. Among those with written plans, 48% reported using them worsening and 55% used them for attacks. Two thirds of patients had been given oral action plans/ instructions for worsening symptoms and attacks (65% for both). Use of this oral advice was higher than the written advice: 64% relied on this oral advice regarding worsening symptoms, and 74% relied on oral advice for attacks. The most common reported advice (written or oral) for dealing with worsening symptoms were, use more albuterol, go to the ED, and call your doctor [in that order]. Higher-level advice such as increasing ICS or starting/increasing oral steroids was not commonly reported. The most frequent advice for managing attacks was, go to the ED, use more albuterol, and call the doctor. One month after hospital discharge, only 28% had written instructions for worsening symptoms (unchanged from baseline), although more had oral instructions (81% vs 65% at baseline, p 0.001). At 6 months, 47% of patients had written and 81% had oral plans for worsening symptoms. Univariate Predictors of NAEPP Guideline Concordant Care Associations between various sociodemographic, clinical, access to care variables, and six different measures of guideline-concordant care are displayed in Table 3. The quality of care was largely independent of age, gender, race, and ethnicity, with very few exceptions. Men were more likely to have action plans. Patients with spacers were older (55.6 years vs 1946 Clinical Investigations

5 Table 3 Univariate Associations Between Sociodemographic, Clinical, and Access-to-Care Variables and Quality-of-Care Indicators* Variables ICS Prescribed (n 152) MDI Instruction (n 176) MDI Spacer (n 134) Peak Flow Meter (n 158) Action Plan for Worsening (n 61) Action Plan for Attacks (n 44) Sociodemographics Gender Female Male Race/ethnicity White Hispanic African American Other Asthma history Prior intubation Yes No Prior oral steroids Yes No Long-term oral steroids All/most of time Some/little/none Childhood asthma Yes No Psychosocial factors Current smoker Yes No Problem drinking Yes No Speaks mostly Spanish Yes No Access to care Usual source of care Yes No Regular asthma provider Yes No Specialty of provider Specialist Generalist Years with provider 1yr yr *Data are presented as % of those who received a specific element of guideline-concordant care, varied by each patient characteristic. p p years, p 0.001). Patients with greater asthma severity (indicated by prior intubation, prior oral steroids, frequency of steroid use, or age of onset) were more likely to receive guideline-concordant care. Similarly, patients with more ED visits in the past 12 months had higher rates of ICS prescribed, MDI instructions, spacers, and peak flow meters (p 0.05). Patients who spoke mostly Spanish were less likely to have spacers or action plans for worsening symptoms, with a trend toward lower rates of peak flow meters and MDI instruction. Patients who had a regular place of care had higher rates of being given ICS, spacers, and peak flow meters. There were similar patterns among those with a regular physician in charge of their asthma or a 1-year relationship with their pro- CHEST / 128 / 4/ OCTOBER,

6 vider. Patients cared for by a specialist (pulmonologist or allergist) had significantly higher rates of being given ICS, spacers, peak flow meters, and action plans. Multivariate Predictors of NAEPP Guideline Concordant Care Table 4 shows the variables that were independently associated with guideline-concordant care. Patients with a history of prior oral steroid use had 3-fold to 14-fold greater odds of been given ICS, spacers, peak flow meters, and instructions on how to use their MDIs. Patients with a usual source of asthma care had sixfold greater odds of getting ICS and threefold greater odds of having a spacer or a peak flow meter. Those cared for by a pulmonary/ allergy specialist had three-times greater odds of receiving action plans. Patients who spoke mostly Spanish were less likely to have spacers or action plans for attacks with a trend toward fewer peak flow meters (p 0.07). Discussion In this study of a consecutive cohort of high-risk, inner-city adults hospitalized for asthma, the quality and access to care was highly variable and suboptimal overall. We found significant underuse of NAEPP-recommended care, including underuse of ICS, peak flow meters, spacers, written action plans, as well as overuse of short acting -agonists. While the multivariate analyses discussed below identified some subgroups that did better, the absolute performance in all groups leaves ample opportunities for improvement. This study was unique in tracking several indicators of quality over time. That these problems persisted over time, despite an asthma hospitalization and subsequent ambulatory visits (both opportunities for implementing guideline-concordant care) suggests fixed deficiencies in the quality of care. The most concerning finding was that underuse of ICS remains a problem, even among high-risk patients. We identified three significant voltage drops that contribute to this underuse problem. At all study intervals, ICS were underprescribed by physicians, underutilized by patients, and used less commonly in the absence of symptoms. Improving performance on these measures is likely to require both systematic interventions directed toward physicians, as well as better patient-oriented strategies for bolstering medication adherence. Consistent with previous reports, 13,19 asthma severity (measured in a variety of ways) was the strongest predictor of receipt of ICS. Our study extends this association to other elements of quality such as MDI instructions, spacers, and peak flow meters. Unfortunately, sicker patients did not receive action plans more often, a cornerstone of NAEPP self-management goals. In contrast to a previous report, 16 suboptimal quality was not primarily an insurance problem but a more subtle blend of other system and patient factors. Having a usual source of care was the most important access variable. Patients cared for by a pulmonology or allergy specialist were more likely to have action plans, something noted in the pediatric literature. 14 Even after controlling for severity, access, and specialty care, patients who spoke mostly Spanish had lowerquality care, something that may partially explain the higher rates of asthma morbidity among Hispanics. 5,6 Efforts to reduce disparities in asthma may need to address larger doctor/patient communication problems among Hispanics. There may be some hidden good news in that the 77% rate of ICS prescriptions in our hospital cohort treated from 2001 to 2002 was greater than the 45% Table 4 Multivariate Predictors of Guideline-Concordant Care* Variables ICS Prescribed Received MDI Instruction Spacer Given Peak Flow Meter Given Action Plan for Worsening Action Plan for Attacks Prior intubation 1.3 ( ) 1.4 ( ) 6.4 ( ) 1.1 ( ) 1.2 ( ) Prior oral steroids 6.6 ( ) 3.1 ( ) 14.2 ( ) 12.8 ( ) 1.8 ( ) 0.9 ( ) Usual source of care 6.2 ( ) 1.2 ( ) 3.8 ( ) 2.7 ( ) 0.6 ( ) 0.5 ( ) Specialty provider 0.9 ( ) 1.8 ( ) 3.5 ( ) 3.0 ( ) Speaks mostly Spanish 0.6 ( ) 0.5 ( ) 0.2 ( ) 0.4 ( ) 0.6 ( ) 0.2 ( ) C statistic *Data are presented as odds ratio (95% confidence interval). Significant at the p 0.05 level. Because of the problem of quasiseparation of data, this variable was not included in the models of MDI instruction, spacer, and peak flow meter given (see Materials and Methods section). This measures model discrimination and is equivalent to the area under the receiver operating characteristic curve, whereby a value of 0.5 indicates chance model performance and 1.0 indicates perfect performance Clinical Investigations

7 rate among hospitalized patients from 1992 to and the 45% rate among ED patients from 1997 to This may suggest broader use of ICS over the past decade. Rates of ICS prescription among outpatients, even among advantaged groups (like managed-care enrollees or nurses) have also been found to be modest (67% and 55%, respectively). 13,19 Patient adherence with daily ICS that others reported has varied from 28 to 77%. 13,22 25 More patients in our inner-city cohort had a peak flow meter (80 to 95%) compared to other multicenter studies 15,20 among ambulatory patients (28% and 38%, respectively). Use of peak flow monitoring (36% at 6 months) in our inpatient cohort was higher than prior outpatient studies: 8% 15 and 28%. 20 Among our patients, the primary reason for not using their peak flow meters was that they were not having symptoms. This suggests a lack of proper education or understanding about the rationale for peak flow monitoring. However, low rates of adherence with peak flow monitoring have even been found among patients in asthma clinical trials, 26,27 suggesting that these are of low saliency by most patients. Use of written action plans (28 to 47%) in our patients was similar to the rates reported in larger national studies 15,16,20 (27 to 52%). One of the goals of asthma self-management is for patients to recognize when they are getting worse so they seek care early to prevent or ameliorate an imminent exacerbation. We found that this happened infrequently, even among patients with insurance and a usual place of care. The fact that three fourths of patients delayed seeking care for the index exacerbation appears to be partly explained by inadequate patient knowledge (not connecting the average 1 week of worsening as the downward slide toward hospitalization) and wishful thinking (half thought things would get better on their own). Providers and health systems may have further contributed to delays because, even among patients with a regular doctor, one third said it was hard to get timely telephone advice or urgent visits. Some limitations of this study are worth addressing. This was a modest-sized cohort who were enrolled based on hospitalizations at a single institution. However, patients received their ambulatory care from a variety of hospital, community health center, and office-based outpatient sites, and the rates of guideline-concordant care we found were largely similar to those reported in large multicenter studies. Because patients had to be hospitalized to be eligible, our data overrepresent those with the worst disease and poorest self-management. We deliberately studied a sick population for study because the importance of guideline-concordant care in these high-risk patients is unequivocal. Ninety percent of the Hispanic patients in our cohort were Puerto Rican, so our findings may not generalize to other Latino subgroups. All of our measures of care were self-reported. Unfortunately, most processes of care we studied are infrequently and inconsistently documented in the record, so patient reports were deemed the best source. Patient reports have been used on by similar asthma quality-of-care studies While self-reported rates of medication compliance can be good, 28 they are likely overestimates. 29 Thus, our data likely underestimate the magnitude of ICS underuse by patients. The fact that we found major deficits in the quality of care among some of the highest-risk patients in the country after nearly a decade of strong evidence, NAEPP consensus guidelines, and intense public policy focus (especially on inner-city asthma) is troubling. Future work is probably needed in two directions. First, there is a need for more detailed understanding of the patient, system, and physician factors that influence physician and patient adherence with NAEPP guidelines in a broader variety of settings. Second, system and patient-oriented interventions to increase the delivery of guideline-concordant care and bolster self-management are badly needed, especially for high-risk, inner-city populations. Identifying effective strategies and resources to implement and sustain such programs will be additionally important though challenging. ACKNOWLEDGMENT: The authors thank Allison Cooperman, MPH; Jason Wang, PhD; Jessica Salazar; Lisa Fitzgerald, BA; Pablo Mora, PhD; and Toni Sturm, MD for their contribution to the project, as well as the patients and their physicians for their goodwill and cooperation. References 1 Mannino DM, Homa DM, Akinbami LJ, et al. Surveillance for asthma: United States, Morb Mortal Wkly Rep CDC Surveill Summ 2002; 51:SS1 S13 2 Weiss KB, Sullivan SD. The health economics of asthma and rhinitis: I. Assessing the economic impact. J Allergy Clin Immunol 2001; 107:3 8 3 Grant EN, Alp H, Weiss KB. The challenge of inner-city asthma. Curr Opin Pulm Med 1999; 5: Homa DM, Mannino DM, Lara M. Asthma mortality in U.S. Hispanics of Mexican, Puerto Rican, and Cuban heritage, Am J Respir Crit Care Med 2000; 161: Carr W, Zeitel L, Weiss K. Variations in asthma hospitalizations and deaths in New York City. Am J Public Health 1992; 82: New York City Community Health Atlas, New York, NY: United Hospital Fund, Stevenson LK. Asthma hospitalization and mortality in New York City, In: Conference proceedings: working together to combat urban asthma. New York, NY: New York Academy of Medicine, Adams N, Bestall J, Jones PW. Budesonide at different doses for chronic asthma. Cochrane Database Syst Rev 2001; 4 9 National Asthma Education and Prevention Program. Expert CHEST / 128 / 4/ OCTOBER,

8 panel report: guidelines for the diagnosis and management of asthma. Bethesda, MD: National Institutes of Health, National Heart, Lung and Blood Institute. Expert panel report 2: Guidelines for the diagnosis and management of asthma. Bethesda, MD: National Institutes of Health, Hartert TV, Togias A, Mellen BG, et al. Underutilization of controller and rescue medications among older adults with asthma requiring hospital care. J Am Geriatr Soc 2000; 48: Hartert TV, Windom HH, Peebles RS Jr, et al. Inadequate outpatient medical therapy for patients with asthma admitted to two urban hospitals. Am J Med 1996; 100: Diette GB, Wu AW, Skinner EA, et al. Treatment patterns among adult patients with asthma: factors associated with overuse of inhaled -agonists and underuse of inhaled corticosteroids. Arch Intern Med 1999; 159: Diette GB, Skinner EA, Nguyen TT, et al. Comparison of quality of care by specialist and generalist physicians as usual source of asthma care for children. Pediatrics 2001; 108: Asthma in America. Research Triangle Park, NC: Glaxo Wellcome, Available at: last. Accessed June 6, Ferris TG, Blumenthal D, Woodruff PG, et al. Insurance and quality of care for adults with acute asthma. J Gen Intern Med 2002; 17: Singh AK, Woodruff PG, Ritz RH, et al. Inhaled corticosteroids for asthma: are ED visits a missed opportunity for prevention? Am J Emerg Med 1999; 17: Legorreta AP, Christian-Herman J, O Connor RD, et al. Compliance with national asthma management guidelines and specialty care: a health maintenance organization experience. Arch Intern Med 1998; 158: Barr RG, Somers SC, Speizer FE, et al. Patient factors and medication guideline adherence among older women with asthma. Arch Intern Med 2002; 162: Krishnan JA, Diette GB, Skinner EA, et al. Race and sex differences in consistency of care with national asthma guidelines in managed care organizations. Arch Intern Med 2001; 161: Vollmer WM, Markson LE, O Connor E, et al. Association of asthma control with health care utilization and quality of life. Am J Respir Crit Care Med 1999; 160: Legorreta AP, Liu X, Zaher CA, et al. Variation in managing asthma: experience at the medical group level in California. Am J Manag Care 2000; 6: Meng YY, Leung KM, Berkbigler D, et al. Compliance with US asthma management guidelines and specialty care: a regional variation or national concern? J Eval Clin Pract 1999; 5: Apter AJ, Reisine ST, Affleck G, et al. Adherence with twice-daily dosing of inhaled steroids: socioeconomic and health-belief differences. Am J Respir Crit Care Med 1998; 157: Mancuso CA, Rincon M, McCulloch CE, et al. Self-efficacy, depressive symptoms, and patients expectations predict outcomes in asthma. Med Care 2001; 39: Gibson PG, Coughlan J, Wilson AJ, et al. Self-management education and regular practitioner review for adults with asthma. Cochrane Database Syst Rev 2000; 2 27 Turner MO, Taylor D, Bennett R, et al. A randomized trial comparing peak expiratory flow and symptom self-management plans for patients with asthma attending a primary care clinic. Am J Respir Crit Care Med 1998; 157: Boudreau DM, Daling JR, Malone KE, et al. A validation study of patient interview data and pharmacy records for antihypertensive, statin, and antidepressant medication use among older women. Am J Epidemiol 2004; 159: Rand CS, Nides M, Cowles MK, et al. Long-term metereddose inhaler adherence in a clinical trial. Am J Respir Crit Care Med 1995; 152: Clinical Investigations