Insurance status and asthma-related health care utilization in patients with severe asthma

Insurance status and asthma-related health care utilization in patients with severe asthma Anju T. Peters, MD*; Julie C. Klemens, MD*; Tmirah Haselkorn, PhD†; Scott T. Weiss, MD‡; Leslie C. Grammer, MD*; June H. Lee, MD§; and Hubert Chen, MD, MPH储; for the TENOR Study Group

Background: Medicaid insurance has been associated with worse asthma outcomes, but the degree to which demographic factors contribute to this relationship has not been well explored. Objective: To evaluate whether insurance status is independently associated with health care utilization (HCU) and asthma control when demographic differences are taken into account. Methods: We used baseline data from adults with severe asthma in the Epidemiology and Natural History of Asthma: Outcomes and Treatment Regimens study. HCU was defined as hospitalization or emergency department visit for asthma in the past 3 months. Asthma control was evaluated using the Asthma Therapy Assessment Questionnaire. Multiple logistic regression was used to compare HCU and asthma control in patients with Medicaid vs those with private health insurance. Results: Of 1,315 patients analyzed, 130 (9.9%) had Medicaid insurance and 1,185 (90.1%) had private insurance. Medicaid insurance was associated with younger age, female sex, race other than white, obesity, active smoking, lower education level, and unemployment. In unadjusted analyses, Medicaid patients had significantly higher HCU (odds ratio [OR], 3.08; 95% confidence interval [CI], 2.11– 4.50) and poorer asthma control (OR, 2.56; 95% CI, 1.84 –3.57) compared with patients with private insurance. After adjusting for demographic differences, insurance status was no longer associated with HCU (OR, 1.43; 95% CI, 0.92–2.23), and the strength of its association with asthma control was reduced (OR, 1.67; 95% CI, 1.17–2.40). Conclusions: Medicaid insurance is not associated with increased HCU in patients with severe asthma once demographic factors have been taken into account but remains modestly associated with poorer asthma control. Ann Allergy Asthma Immunol. 2008;100:301–307.

INTRODUCTION Asthma morbidity and mortality have substantially increased during the past few decades and have resulted in increased economic, social, and health burdens.1 Asthma is a leading cause of hospitalizations, emergency department (ED) visits, and missed days from school or work.2,3 The direct and indirect cost of asthma in the United States is estimated to be between $5.8 billion and $10.7 billion annually.4 Visits to the ED and hospitalizations represent a large proportion of the costs, accounting for 26% to 50% of the total costs of treatment for asthma.3– 6 Furthermore, most of the cost is consumed by a small proportion of the asthmatic population, with 80% of resources used by 20% of the individuals.4 Studies have shown that the Medicaid population is more likely to use the ED or be hospitalized for their asthma care than individuals with private insurance,7–9 suggesting that insurance status may affect quality Affiliations: * Feinberg School of Medicine, Northwestern University, Chicago, Illinois; † EpiMetrix Inc, Sunnyvale, California; ‡ Channing Laboratory, Brigham & Women’s Hospital and Harvard Medical School, Boston, Massachusetts; § Genentech Inc, South San Francisco, California; 储 University of California, San Francisco, California. Disclosures: Authors have nothing to disclose. Drs Peters and Klemens contributed equally as first authors to the analytic design, interpretation of data, and drafting of the final manuscript. Received for publication August 6, 2007; Received in revised form September 5, 2007; Accepted for publication September 5, 2007.

VOLUME 100, APRIL, 2008

of asthma care received and subsequent need for acute asthma care.10 –14 The relationship between insurance status and asthmarelated health care utilization (HCU) is complex and not well understood. Certain studies suggest that Medicaid participants may be receiving suboptimal treatment because of various factors, resulting in increased acute care visits. For example, studies have shown that Medicaidinsured children are frequently prescribed fewer inhaled corticosteroids10,12 and reported increased use of shortacting ␤-agonists,12 barriers to obtaining asthma medications,12,13 and difficulty with scheduling outpatient follow-up care.15 Other studies have suggested that demographic variables such as race/ethnicity, age, sex, education, obesity, and smoking status could confound the relationship between insurance status and HCU.9,16 –19 We hypothesize that higher HCU observed among Medicaid participants is the result of demographic differences not insurance status. To test this hypothesis, we used baseline data from the Epidemiology and Natural History of Asthma: Outcomes and Treatment Regimens (TENOR) study, a large, multicenter, prospective cohort study of patients with severe or difficult-to-treat asthma. The demographic diversity of study participants, high HCU, and observational design of the TENOR study make it a particularly valuable cohort in which to study the effects of insurance status on asthma-related outcomes.

301

METHODS Study Design and Participants TENOR was a 3-year (2001–2004), multicenter, prospective, observational study of patients diagnosed as having either severe or difficult-to-treat asthma. A detailed description of the study methods has been previously published.20 In brief, participants were recruited from pulmonary and allergy practices across the United States from diverse health care settings. To be eligible for the TENOR study, individuals had to be diagnosed as having either severe or difficult-to-treat asthma by their treating physician and meet prespecified criteria for high HCU or high medication use within the last 12 months. Participants with a history of smoking of 30 pack-years or more, cystic fibrosis, severe cardiovascular disease, cancer, severe psychiatric disorder, or other systemic disease associated with an expected life span of less that 2 to 3 years were excluded. Informed consent was obtained from each study participant. The study protocol was approved by a central institutional review board and, if necessary, the institutional review board at each site. Demographic and clinical information was obtained via study coordinator– conducted interviews at each site using standardized case report forms. Key variables collected included age, sex, race/ethnicity, education, smoking status, employment, health care insurance coverage, asthma history, medication use, comorbid conditions, and HCU in the past 3 months. To evaluate the level of asthma control, participants completed the self-administered Asthma Therapy Assessment Questionnaire (ATAQ). For this analysis, we used baseline data for 1,315 individuals 18 years or older who were classified as having severe asthma by physician evaluation. Predictor Variables Participants were categorized as having either Medicaid or private insurance based on their response at the time of their baseline visit. Private insurance was defined as having 1 of the following types of insurance plans: commercial or preferred provider organization, health care maintenance organization, or self-pay. Individuals who indicated other types of insurance plans or with missing data were excluded from this analysis. For this analysis, we also considered a number of potential confounders known to be associated with both insurance status and asthma outcomes. These confounders included age, sex, race/ethnicity, body mass index (BMI), smoking status, and education level. Income and employment status were excluded from multivariate analyses for reasons discussed later. Outcome Measures The primary outcome measure for this analysis was HCU, defined as ED visits and/or hospitalization for asthma in the previous 3 months, based on participant-elicited survey responses at the time of the baseline visit. Unscheduled office visits, steroid bursts, and medication use were not included as outcomes, since they were used as part of the inclusion criteria for the TENOR study. Secondary outcome measures

302

included number of asthma control problems, as evaluated using the ATAQ control index (range, 0 – 4 control problems), and participant responses to individual ATAQ items relating to daytime activities, nighttime awakenings, frequency of rescue inhaler use, and overall self-perceived asthma control. Cross-sectional and longitudinal validity of the ATAQ has been demonstrated in multiple studies.21–23 Statistical Analysis Descriptive statistical analyses were generated for demographic variables at baseline by insurance type (Medicaid vs private insurance). The t test and the ␹2 test were used to test for significant differences between Medicaid and privately insured patients for continuous and categorical characteristics, respectively. Similarly, ␹2 tests were used to test bivariate relationships between insurance status and HCU, insurance status and asthma control, and HCU and asthma control. In the case of the ATAQ control index, the Mantel-Haenszel test was used to test for trends. Multivariate analyses were performed controlling for age, sex, race/ethnicity, BMI, education level, and smoking status as covariates. Separate logistic regression models were used for HCU and the ATAQ control index. Patients who marked “unsure” as a response for items pertaining to “missed activity,” “asthma control,” and “woken by asthma” were excluded. Because the ATAQ control index is represented by a 5-level ordinal variable, a proportional odds model (ordinal logistic regression) was used to model the relationship between insurance status and asthma control. Using this model, odds ratios (ORs) reflect the increased odds of having an ATAQ score above any given threshold on the ordinal scale. To develop the most parsimonious model, we specified all covariates as dichotomous, with the exception of age, which was treated as continuous. Employment status and income were not included in the final models because of colinearity with insurance status. RESULTS Patient Characteristics Of 4,755 patients enrolled in the TENOR study, 1,315 adults with physician-evaluated severe asthma were included in the current analysis. Baseline demographics for the analytic cohort by insurance status are presented in Table 1. Overall, approximately 10% of the individuals had Medicaid insurance and 90% were covered by private insurance (including managed care organizations). Patients with Medicaid were more likely to be younger, female, a race other than white, less educated, unemployed, and current smokers and to have a higher BMI compared with those with private insurance. Bivariate Comparisons In unadjusted analyses, Medicaid patients demonstrated higher HCU and poorer asthma control than those with private insurance. Specifically, 41% of Medicaid patients visited the ED or required hospitalization for asthma, compared with only 18% of patients with private insurance (P ⬍ .001; Table

ANNALS OF ALLERGY, ASTHMA & IMMUNOLOGY

Table 1. Baseline Characteristics of 1,311 TENOR Patients Included in the Analysis Insurance status Characteristics Age, y Mean ⫾ SD Median (range) Sex, No. (%) Male Female Race/ethnicity, No. (%) White Race other than white BMIc Mean ⫾ SD Median (range) Smoking status, No. (%) Current smoker Past or nonsmoker Education level, No. (%) High school or less More than high school Employment status, No. (%) Employed Unemployedd

Medicaid (n ⴝ 130)

Non-Medicaida (n ⴝ 1,185)

40.9 ⫾ 13.6 40.0 (18–100)

46.8 ⫾ 12.5 48.0 (18–89)

P valueb

⬍.001 ⬍.001

22 (17) 108 (83)

369 (31) 816 (69)

61 (47) 69 (53)

964 (81) 221 (19)

33.8 ⫾ 9.9 33.0 (16.8–72.0)

30.7 ⫾ 7.9 29.1 (16.4–68.3)

⬍.001

⬍.001 ⬍.001

14 (11) 116 (89)

41 (3) 1,144 (97)

91 (70) 39 (30)

279 (24) 906 (76)

40 (31) 90 (69)

869 (73) 315 (27)

⬍.001 ⬍.001

Abbreviations: BMI, body mass index; TENOR, Epidemiology and Natural History of Asthma: Outcomes and Treatment Regimens. Health maintenance organization or preferred provider organization or commercial insurance. b Reported P values are based on a 2-sided t test for continuous variables and a ␹2 test for categorical variables. c Missing data for 6 patients (1 Medicaid, 5 non-Medicaid). d Includes retired, homemaker, unemployed, or disabled. a

2). Furthermore, Medicaid patients demonstrated greater asthma control problems than those with private insurance (P ⬍ .001; Table 3). Overall, only 37% of asthmatic patients with Medicaid considered their asthma to be well controlled, compared with 57% among those with private insurance (P ⬍ .001). Asthmatic patients insured by Medicaid were also more likely to miss daytime activities, report nighttime waking, and use a rescue inhaler in the past 4 weeks.

Multivariate Analyses Results of multiple logistic regression models for HCU and asthma control are given in Table 4. After adjustment for age, sex, race/ethnicity, obesity, education level, and smoking status, the observed association between insurance status and HCU was no longer statistically significant (adjusted OR, 1.43; 95% confidence interval [CI], 0.92–2.23). In contrast,

Table 2. Insurance Status and Asthma-Related Health Care Utilization No. (%) of patients by insurance status Outcome measure Hospitalization for asthmab Yes No ED visit for asthmab Yes No Hospitalization or ED visit for asthmab Yes No

P value

Medicaid (n ⴝ 130)

Non-Medicaida (n ⴝ 1,185)

23 (18) 107 (82)

84 (7) 1099 (93)

⬍.001

52 (40) 78 (60)

203 (17) 980 (83)

⬍.001

53 (41) 77 (59)

216 (18) 967 (82)

⬍.001

Abbreviation: ED, emergency department. a Health maintenance organization or preferred provider organization or commercial insurance. b Health care utilization in the past 3 months (assessed at baseline visit).

VOLUME 100, APRIL, 2008

303

Table 3. Insurance Status and Asthma Control No. (%) of patients by insurance status Outcome measure

Medicaid (n ⴝ 130)

Non-Medicaida (n ⴝ 1,185)

3 (2) 10 (8) 38 (30) 46 (36) 31 (24)

159 (14) 210 (18) 351 (30) 296 (25) 150 (13)

78 (61) 46 (36) 4 (3)

429 (37) 738 (63) 7 (1)

93 (73) 34 (27) 1 (1)

581 (50) 567 (48) 25 (2)

47 (37) 62 (48) 19 (15)

668 (57) 414 (35) 93 (8)

P valueb ⬍.001

ATAQ control index No problems 1 problem 2 problems 3 problems ⱖ4 problems Missed daytime activities due to asthma in the past 4 weeks Yes No Unsure Nighttime awakening due to asthma in the past 4 weeks Yes No Unsure Asthma well controlled in the past 4 weeks Yes No Unsure

⬍.001

⬍.001

⬍.001

Abbreviation: ATAQ, Asthma Therapy Evaluation Questionnaire. Health maintenance organization or preferred provider organization or commercial insurance. b Reported P values are based on the Fisher exact test and Mantel-Haenszel ␹2 test for trend (for ATAQ control index). a

Table 4. Unadjusted and Adjusted Logistic Regression Models for Health Care Utilization and Asthma Control OR (95% CI)

Unadjusted model Medicaid vs non-Medicaid Adjusted model with covariates Medicaid vs non-Medicaid Age, per 10-year increase Female vs male Race other than white vs white BMI ⬎30 vs ⱕ30 Lower vs higher education level Current smoker vs nonsmoker

Hospitalization or ED visit (n ⴝ 1,309)a

ATAQ control index (n ⴝ 1,290)b

3.08 (2.11–4.50)

2.56 (1.84–3.57)

1.43 (0.92–2.23) 0.72 (0.64–0.81) 1.62 (1.15–2.27) 2.08 (1.51–2.86) 1.39 (1.04–1.85) 1.56 (1.14–2.15) 2.31 (1.28–4.15)

1.67 (1.17–2.40) 0.73 (0.68–0.80) 2.01 (1.61–2.50) 1.20 (0.94–1.53) 1.47 (1.21–1.80) 1.10 (0.87–1.38) 1.85 (1.12–3.06)

Abbreviations: ATAQ, Asthma Therapy Evaluation Questionnaire; BMI, body mass index; CI, confidence interval; ED, emergency department; OR, odds ratio. a Higher OR indicates increased odds of hospitalization or ED visit. b Higher OR indicates increased odds of being less controlled (higher ATAQ score) based on a proportional odds model (ordinal logistic regression).

the relationship between insurance status and asthma control remained significant, but the strength of association was substantially reduced. Specifically, patients covered by Medicaid were 1.7 times more likely (adjusted OR, 1.67; 95% CI, 1.17–2.40) than privately insured patients to report higher ATAQ scores, indicating worse asthma control. Although insurance status was not independently associated with HCU in our multivariate analysis, statistically significant associations were observed for several of the demographic covariates considered in the same model, including age, sex, race/ethnicity, BMI, education, and smoking status

304

(Table 3). Unlike HCU, asthma control was not independently associated with either race/ethnicity or education level once insurance status had been taken into account. HCU and Asthma Control Overall, a greater number of asthma control problems, as measured by the ATAQ control index, was associated with increased likelihood of HCU in the past 3 months (Table 5). Patients with 2 to 4 control problems at baseline were more likely to be seen in the ED or hospitalized for asthma than patients with only 0 to 1 control problem (27% vs 5%; P ⬍

ANNALS OF ALLERGY, ASTHMA & IMMUNOLOGY

Table 5. Asthma Control and Health Care Utilizationa Hospitalization or ED visit, No. (%)b ATAQ control index All patients 0–1 problem 2–4 problems Medicaid patients 0–1 problem 2–4 problems Non-Medicaid patients 0–1 problem 2–4 problems

P value Yes

No

19/382 (5) 245/912 (27)

363/382 (95) 667/912 (73)

⬍.001

4/13 (31) 48/115 (42)

9/13 (69) 67/115 (58)

.45

15/369 (4) 197/797 (25)

354/369 (96) 600/797 (75)

⬍.001

Abbreviations: ATAQ, Asthma Therapy Evaluation Questionnaire; ED, emergency department. Health care utilization in the past 3 months (assessed at baseline visit). b Values expressed as number (percentage) responding “yes” vs “no” within each stratum. a

.001). When stratified by insurance status, a similar relationship was observed for those with private insurance (25% vs 4%; P ⬍ .001) but not for those insured by Medicaid (42% vs 31%; P ⫽ .45). DISCUSSION The association between insurance status and asthma outcomes has been previously reported; however, the extent to which specific demographic variables may explain the relationship has not been fully explored. In this analysis, we took advantage of health registry data from a well-described cohort of patients with severe asthma to evaluate the relationship between insurance status and 2 key asthma outcomes: HCU and asthma control. Our results show that although Medicaid-insured patients demonstrated significantly higher HCU and worse asthma control than non-Medicaid patients, the strength of these relationships was mitigated substantially after controlling for demographic factors known to negatively affect asthma outcomes. These findings suggest that demographic differences play a major role in determining the relationship between insurance status and asthma outcomes, but also that the degree to which such factors influence outcomes may depend on the outcome measure. Previous studies that have attempted to address this question have focused primarily on children.8,9 In a study by Finkelstein et al,8 children insured by Medicaid were 1.4 times more likely to receive care in the ED and 1.3 times more likely to be hospitalized for asthma than non-Medicaid children in the same managed care setting. The effect of race and sociodemographic factors remains unclear, because only age, sex, and ␤-agonist dispensing rate were taken into account. In another pediatric study, Ortega et al9 found that after adjusting for frequency of asthma-related primary care visits, primary provider practice type, use of asthma specialist, age, sex, medication use, and symptoms, Medicaid children still required more frequent ED visits for asthma than privately insured children. Race/ethnicity was not found to modify the relationship between insurance status and HCU, but among black children, those insured by Medicaid were less likely to

VOLUME 100, APRIL, 2008

make use of routine primary care services than privately insured children. The present analysis adds to the current literature by studying the relationship between insurance status and health outcomes in an adult population with severe asthma. In addition to studying HCU, we also considered the effect of insurance status on asthma control, an outcome increasingly emphasized in recent treatment guidelines. After adjusting for demographic confounding, we found that the relationship between Medicaid insurance and HCU was no longer statistically significant, whereas the relationship between Medicaid and asthma control remained statistically significant, although the strength of association was substantially reduced. Several possible explanations for these observations exist. Insurance status may influence HCU and asthma control through different pathways. In the case of HCU, race and socioeconomic status have been particularly well studied and are known to be associated with increased ED visits and hospitalization for asthma.24,25 In the Medicaid population, where patients were more likely to be races other than white, less educated, and unemployed, such demographic factors are likely to play a major role in determining HCU. In the case of asthma control, demographic differences also appear important; however, a number of other nondemographic factors may come into play. As a composite patient-reported outcome, the ATAQ control index incorporates daytime activities, nighttime awakenings, frequency of rescue inhaler use, and overall self-perceived asthma control. Factors related to insurance status may have a more direct effect on such outcomes as the result of access to care, prescription coverage, or ancillary support programs, such as asthma education. In addition, there may be a number of latent factors that influence an individual’s perception of his or her asthma control. Specifically, psychosocial stressors associated with Medicaid insurance have been shown to negatively affect quality of life and asthma control.26 –28 Depression, for example, has been shown to be more common and more severe in

305

the Medicaid population compared with those privately insured.29,30 Although asthma control and HCU are often closely related, this relationship may vary, depending on the circumstance. In our analysis, we examined the relationship between asthma control and HCU and found that higher ATAQ scores (2 or more problems) were associated with an increased likelihood of either an ED visit or hospitalization for asthma in the past 3 months. This relationship has also been demonstrated by Vollmer et al,22,23 as well as in the TENOR cohort.31 Interesting, we found that when these results were stratified by insurance status, the relationship between asthma control and HCU was no longer significant among patients insured by Medicaid. One reason could be the strong demographic differences known to be associated with HCU in the Medicaid population (Table 1). Although we can only speculate as to the exact cause, the weak relationship between asthma control and HCU in Medicaid patients helps to explain the observed difference between the 2 outcomes with respect to insurance status. The total number of Medicaid participants in the TENOR study was relatively small, however, thereby limiting the statistical power of our stratified analyses. Because of the inherent design of the TENOR study, the present analysis has other potential limitations. The TENOR study was originally intended to focus on those individuals with severe or difficult-to-treat asthma, and therefore our results may not be applicable to populations with milder disease. As a result of the TENOR enrollment criteria, the study population has been enriched for patients at risk for high utilization based on recent unscheduled office visits, systemic steroid use, and the need for multiple controller medications. We were therefore unable to evaluate these events as outcomes. Individuals who participated in the TENOR study may have also been more motivated to comply with care and may have received closer medical attention as a consequence of participating in the study. Furthermore, all participants in TENOR received regular care from either an allergist or pulmonologist. Such selection biases may have blunted our ability to evaluate the effect of insurance-related differences in adherence or access to care that might otherwise play a greater role in the general population. Finally, we acknowledge that although the use of observational data can be powerful for establishing associations, or lack thereof, causality should not be implied. Despite these limitations, our results are encouraging in that they appear to suggest that the observed disparity in asthma outcomes associated with Medicaid insurance can be minimized when racial and socioeconomic barriers are removed and patients receive regular care from an asthma specialist. Future studies will be required to understand exactly how demographic factors, either individually or collectively, can mediate the effect of insurance status on different asthma outcomes.

306

ACKNOWLEDGMENTS The TENOR Study is sponsored by Genentech Inc and Novartis Pharmaceutical Corp. For a complete list of TENOR study group members, please contact Genentech Inc. We thank Dave Miller and Lawrence Rasouliyan from ICON Clinical Research (San Francisco, California) for their statistical support. REFERENCES 1. Mannino DM, Homa DM, Pertowski CA, et al. Surveillance for asthma–United States, 1960 –1995. MMWR CDC Surveill Summ. 1998; 2447(1):1–27. 2. Adams PF, Marano MA. Current estimates from the National Health Interview Survey, 1994. Vital Health Stat 10. 1995;193(pt 1):1–260. 3. Cisternas MG, Blanc PD, Yen IH, et al. A comprehensive study of the direct and indirect costs of adult asthma. J Allergy Clin Immunol. 2003;111(6):1212– 8. 4. Smith DH, Malone DC, Lawson KA, Okamoto LJ, Battista C, Saunders WB. A national estimate of the economic costs of asthma. Am J Respir Crit Care Med. 1997;156(3 pt 1):787–93. 5. Gergen PJ. Understanding the economic burden of asthma. J Allergy Clin Immunol. 2001;107(5 Suppl):S445–S448. 6. Weiss KB, Sullivan SD, Lyttle CS. Trends in the cost of illness for asthma in the United States, 1985–1994. J Allergy Clin Immunol. 2000; 106(3):493–9. 7. Ferris TG, Blumenthal D, Woodruff PG, Clark S, Camargo CA. Insurance and quality of care for adults with acute asthma. J Gen Intern Med. 2002;17(12):905–13. 8. Finkelstein JA, Barton MB, Donahue JG, Algatt-Bergstrom P, Markson LE, Platt R. Comparing asthma care for Medicaid and non-Medicaid children in a health maintenance organization. Arch Pediatr Adolesc Med. 2000;154(6):563– 8. 9. Ortega AN, Belanger KD, Paltiel AD, Horwitz SM, Bracken MB, Leaderer BP. Use of health services by insurance status among children with asthma. Med Care. 2001;39(10):1065–74. 10. Apter AJ, Van Hoof TJ, Sherwin TE, Casey BA, Petrillo MK, Meehan TP. Assessing the quality of asthma care provided to Medicaid patients enrolled in managed care organizations in Connecticut. Ann Allergy Asthma Immunol. 2001;86(2):211– 8. 11. Emerman CL, Cydulka RK, Rimm A. Changes in asthma claims in a Medicaid population from 1991–1994. Am J Emerg Med. 1999;17(6): 526 –31. 12. Shireman TI, Heaton PC, Gay WE, Cluxton RJ, Moomaw CJ. Relationship between asthma drug therapy patterns and healthcare utilization. Ann Pharmacother. 2002;36(4):557– 64. 13. Warman KL, Jacobs AM, Silver EJ. If we prescribe it, will it come? access to asthma equipment for Medicaid-insured children and adults in the Bronx, NY. Arch Pediatr Adolesc Med. 2002;156(7):673–7. 14. Wissow LS, Gittelsohn AM, Szklo M, Starfield B, Mussman M. Poverty, race, and hospitalization for childhood asthma. Am J Public Health. 1988;78(7):777– 82. 15. Fredrickson DD, Molgaard CA, Dismuke SE, Schukman JS, Walling A. Understanding frequent emergency room use by Medicaid-insured children with asthma: a combined quantitative and qualitative study. J Am Board Fam Pract. 2004;17(2):96 –100. 16. Boudreaux ED, Emond SD, Clark S, Camargo CA Jr. Race/ethnicity and asthma among children presenting to the emergency department: differences in disease severity and management. Pediatrics. 2003;111(5 pt 1):e615– e621. 17. Ford ES. The epidemiology of obesity and asthma. J Allergy Clin Immunol. 2005;115(5):897–909. 18. Lieu TA, Lozano P, Finkelstein JA, et al. Racial/ethnic variation in asthma status and management practices among children in managed Medicaid. Pediatrics. 2002;109(5):857– 65. 19. Schatz M, Camargo CA Jr. The relationship of sex to asthma prevalence, health care utilization, and medications in a large managed care orga-

ANNALS OF ALLERGY, ASTHMA & IMMUNOLOGY

nization. Ann Allergy Asthma Immunol. 2003;91(6):553– 8. 20. Dolan CM, Fraher KE, Bleecker ER, et al. Design and baseline characteristics of the epidemiology and natural history of asthma: Outcomes and Treatment Regimens (TENOR) study: a large cohort of patients with severe or difficult-to-treat asthma. Ann Allergy Asthma Immunol. 2004; 92(1):32–9. 21. Rowe BH, Oxman AD. Performance of an asthma quality of life questionnaire in an outpatient setting. Am Rev Respir Dis. 1993;148(3): 675– 81. 22. 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(5 pt 1):1647–52. 23. Vollmer WM, Markson LE, O’Connor E, Frazier EA, Berger M, Buist AS. Association of asthma control with health care utilization: a prospective evaluation. Am J Respir Crit Care Med. 2002;165(2):195–9. 24. Weitzman M, Byrd RS, Auinger P. Black and white middle class children who have private health insurance in the United States. Pediatrics. 1999;104(1 pt 2):151–7. 25. Zoratti EM, Havstad S, Rodriguez J, Robens-Paradise Y, Lafata JE, McCarthy B. Health service use by African Americans and Caucasians with asthma in a managed care setting. Am J Respir Crit Care Med. 1998;158(2):371–7. 26. Calfee CS, Katz PP, Yelin EH, Iribarren C, Eisner MD. The influence of perceived control of asthma on health outcomes. Chest. 2006;130(5): 1312– 8. 27. Wisnivesky JP, Leventhal H, Halm EA. Predictors of asthma-related

VOLUME 100, APRIL, 2008

28.

29.

30. 31.

health care utilization and quality of life among inner-city patients with asthma. J Allergy Clin Immunol. 2005;116(3):636 – 42. Yen IH, Yelin EH, Katz P, Eisner MD, Blanc PD. Perceived neighborhood problems and quality of life, physical functioning, and depressive symptoms among adults with asthma. Am J Public Health. 2006;96(5): 873–9. Buck JA, Miller K. Mental Health and Substance Abuse Services in Medicaid, 1995. Rockville, MD: Center for Mental Health Services, Substance Abuse and Mental Health Services Administration; 2002. DHHS publication (SMA) 02-3713. Lesser IM, Leuchter AF, Trivedi MH, et al. Characteristics of insured and noninsured outpatients with depression in STAR(*)D. Psychiatr Serv. 2005;56(8):995–1004. Sullivan SD, Wenzel SE, Bresnahan BW, et al. Association of control and risk of severe asthma-related events in severe or difficult-to-treat asthma patients. Allergy. 2007;62(6):655– 60.

Requests for reprints should be addressed to: Hubert Chen, MD, MPH 350 Parnassus Ave Suite 609 Box 0924 San Francisco, CA 94143-0924 E-mail: [email protected]

307