In the United States, morbidity and mortality from asthma are declining but racial and ethnic differences in asthma burden remain unresolved.1 Compared to White patients, individuals of underrepresented backgrounds have a higher prevalence of moderate-to-severe persistent asthma, difficult-to-treat asthma, and emergency room utilization for asthma care.2, 3 Asthma biologics that target type-2 airway inflammation lessen disease burden, reduce asthma exacerbation risk, and improve quality-of-life metrics. Therefore, asthmatic patients with more severe disease, such as those of underrepresented race or ethnicity, can especially benefit from biologics. However, as with other diseases, patients of underrepresented backgrounds incur inequitable prescription practices.4 Herein, we explored racial and ethnic differences in asthma biologic prescriptions among US adults.
We conducted a retrospective cohort study utilizing the claims-based database TriNetX Diamond Network™ (TriNetX, LLC, Cambridge, MA). Diamond Network provides real-time access to >210 million patients covering 1.8 million provider sites and 99% of US commercial and public health plans. We used International Code of Disease (ICD)-10, RxNorm, and Current Procedural Terminology codes to identify adult patients (≥18 years) with asthma who received medium-to-high dose inhaled corticosteroids (budesonide, fluticasone, mometasone) and long-acting beta-agonists (formoterol, salmeterol, vilanterol) between 03/15/2017 and 08/15/2018. Patients with chronic obstructive pulmonary disease, bronchiectasis, cystic fibrosis, interstitial lung diseases, and previous exposure to asthma biologics were excluded. Meeting the aforementioned criteria at any time point between 03/15/2017 and 08/15/2018 defined the index event. Asthma patients were grouped into non-Hispanic White, non-Hispanic Black, non-Hispanic Asian, and Hispanic cohorts. The Hispanic cohort included patients identifying as Hispanic or Latino, irrespective of racial identifications. Individuals who died on or before 04/15/2020 were excluded from the analysis. Using 1:1 propensity score matching (PSM), the cohorts were matched for baseline demographics (age, sex); comorbidities known to worsen asthma control or influence biologic prescriptions (allergic and vasomotor rhinitis, nasal polyposis, atopic dermatitis, idiopathic urticaria, other urticaria, urticaria unspecified, body mass index, tobacco use or nicotine dependence, generalized anxiety disorder, major depressive disorder, obstructive sleep apnea); systemic corticosteroid use (dexamethasone, methylprednisolone, prednisone); medications that may influence asthma control (long-acting muscarinic antagonists, leukotriene receptor antagonists, angiotensin-converting enzyme inhibitors, beta-blockers, aspirin and non-steroidal anti-inflammatory agents). These comorbidities and medication utilizations were considered present if individuals had the corresponding ICD-10 or RxNorm code within one year before the index event. TriNetX generates a propensity score for each patient and uses a “greedy nearest neighbor matching” with a caliper of 0.1 pooled standard deviations to identify matched subsets. The matched cohorts were followed for 20 months from the index event and the proportion of patients initiating a specific biologic class (anti-IL-5 agents [benralizumab, mepolizumab, reslizumab], dupilumab, omalizumab) or any asthma biologic across the different racial/ethnic groups were contrasted using the White cohort as a reference. Significance was assessed using Z-test to compare proportions. Further, to account for patients exiting the cohort before the end of the 20-month observation period (i.e., lost to follow-up), we performed additional survival analysis.
We identified 95,363 White patients, 15,435 Black patients, 12,645 Hispanic patients, and 1,673 Asian patients with moderate-to-severe asthma (Table E1). After PSM, 15,432 patients remained in each of the Black and White asthma cohorts, and the two cohorts were well-balanced for baseline demographics, comorbidities, and medication utilization with a standardized mean difference <0.1. As compared to White patients, a higher proportion of Black patients were started on an asthma biologic (Figure 1). While there were no significant differences in the proportions of patients initiated on anti-IL-5 or dupilumab therapies, more Black asthma patients were prescribed omalizumab (Figure 1). There were 12,642 patients in the Hispanic and White asthma cohorts after PSM. While there was no significant difference in the proportion of patients starting any biologic, proportionally fewer Hispanic patients received anti-IL-5 therapies (Figure 2). Within the Hispanic cohort, 5,303 co-identified as White and the proportion of these patients starting a biologic was similar to non-Hispanic White patients (0.72% versus 0.75%; P =.82 after PSM). Because TriNetX obfuscates counts ≤10, the number of Hispanic patients co-identifying as Black (N=282) or Asian (N=54) was too small to permit comparative analyses. The remainder of the Hispanic cohort lacked racial co-identification. There were 1,672 patients in the Asian and White asthma cohorts after PSM with equal proportions prescribed any biologic (1.44% versus 1.20%; P =.0.54). The number of Asian patients was too small for analysis of prescription differences in the specific biologic classes. Additional survival analysis of study cohorts revealed similar findings (Table E2).
Table E1.
Baseline characteristics of the study cohorts before propensity score matching*
| Code | Characteristic | White (n = 95363) | Black (n = 15435) | Hispanic (n = 12645) | Hispanic White (n = 5303) | Asian (n = 1673) |
|---|---|---|---|---|---|---|
| AI | Age at Index, y | 52.4 (16.2) | 49.1 (15.2) | 49.5 (15.9) | 49.8 (16.1) | 51.1 (16.6) |
| F | Female | 65908 (69.1) | 11742 (76.1) | 9130 (72.2) | 3875 (73.1) | 1061 (63.4) |
| J30 | Vasomotor and allergic rhinitis | 31441 (33) | 4568 (29.6) | 3585 (28.4) | 1612 (30.4) | 585 (35) |
| J33 | Nasal polyp | 1226 (1.3) | 159 (1) | 127 (1) | 63 (1.2) | 22 (1.3) |
| J32 | Chronic sinusitis | 10298 (10.8) | 1294 (8.4) | 1021 (8.1) | 484 (9.1) | 132 (7.9) |
| K21 | Gastroesophageal reflux disease | 25287 (26.5) | 3788 (24.5) | 2767 (21.9) | 1196 (22.6) | 325 (19.4) |
| Z68.3 | Body mass index 30–39 | 8432 (8.8) | 1817 (11.8) | 1328 (10.5) | 551 (10.4) | 98 (5.9) |
| Z68.4 | Body mass index 40 or greater | 5019 (5.3) | 1530 (9.9) | 727 (5.7) | 315 (5.9) | 20 (1.2) |
| L20 | Atopic dermatitis | 1242 (1.3) | 313 (2) | 173 (1.4) | 74 (1.4) | 38 (2.3) |
| F17 | Nicotine dependence | 6560 (6.9) | 1521 (9.9) | 776 (6.1) | 286 (5.4) | 41 (2.5) |
| Z72.0 | Tobacco use | 24934 (26.1) | 4079 (26.4) | 1963 (15.5) | 940 (17.7) | 472 (28.2) |
| J38.3 | Other diseases of vocal cords | 366 (0.4) | 47 (0.3) | 31 (0.2) | 19 (0.4) | 10 (0.6) |
| L50.8 | Other urticaria | 262 (0.3) | 51 (0.3) | 36 (0.3) | 17 (0.3) | 10 (0.6) |
| L50.9 | Urticaria, unspecified | 1774 (1.9) | 255 (1.7) | 172 (1.4) | 87 (1.6) | 24 (1.4) |
| L50.1 | Idiopathic urticaria | 243 (0.3) | 42 (0.3) | 36 (0.3) | 17 (0.3) | 10 (0.6) |
| G47.33 | Obstructive sleep apnea | 11121 (11.7) | 1980 (12.8) | 1140 (9) | 465 (8.8) | 92 (5.5) |
| F41.1 | Generalized anxiety disorder | 4341 (4.6) | 395 (2.6) | 431 (3.4) | 180 (3.4) | 34 (2) |
| F33 | Major depressive disorder | 4124 (4.3) | 568 (3.7) | 653 (5.2) | 290 (5.5) | 23 (1.4) |
| 1303098 | Aclidinium | 132 (0.1) | 25 (0.2) | 19 (0.2) | 10 (0.2) | 10 (0.6) |
| 69120 | Tiotropium | 2824 (3) | 509 (3.3) | 372 (2.9) | 155 (2.9) | 35 (2.1) |
| 1487514 | Umeclidinium | 1079 (1.1) | 152 (1) | 129 (1) | 49 (0.9) | 16 (1) |
| 1191 | Aspirin | 4621 (4.8) | 1036 (6.7) | 895 (7.1) | 301 (5.7) | 80 (4.8) |
| MS102 | NSAIDs | 27414 (28.7) | 6088 (39.4) | 4860 (38.4) | 1933 (36.5) | 350 (20.9) |
| 88249 | Montelukast | 31608 (33.1) | 5225 (33.9) | 4796 (37.9) | 1983 (37.4) | 535 (32) |
| CV100 | Beta blockers | 14019 (14.7) | 2677 (17.3) | 1626 (12.9) | 646 (12.2) | 184 (11) |
| CV800 | ACEIs | 13129 (13.8) | 2579 (16.7) | 1888 (14.9) | 737 (13.9) | 178 (10.6) |
| 3264 | Dexamethasone | 5553 (5.8) | 876 (5.7) | 778 (6.2) | 323 (6.1) | 71 (4.2) |
| 6902 | Methylprednisolone | 17577 (18.4) | 2608 (16.9) | 2207 (17.5) | 939 (17.7) | 201 (12) |
| 8640 | Prednisone | 35908 (37.7) | 5400 (35) | 4365 (34.5) | 1828 (34.5) | 525 (31.4) |
Values are expressed as mean (SD) for continuous variables and number (%) for categorical variables. Comparisons after propensity score matching are available from the authors. TriNetX obfuscates counts ≤10 to ensure anonymity of patients’ data. Abbreviations: NSAIDs, non-steroidal anti-inflammatory drugs, ACEIs, angiotensin-converting enzyme inhibitors.
Figure 1.
Biologic use among Black and White patients with asthma. More Black patients were prescribed biologics, particularly omalizumab, when compared to White patients. Results shown are after 1:1 propensity score matching for baseline demographics and medical comorbidities. Patients may have been prescribed >1 biologic during the 20-month observation period.
Figure 2.
Biologic use among Hispanic and White patients with asthma. The proportion of patients who were started on an asthma biologic prescription was equivalent between Hispanic and White patients, but fewer Hispanic patients received anti-IL-5 biologics. Results shown are after 1:1 propensity score matching for baseline demographics and medical comorbidities. Patients may have been prescribed >1 biologics during the 20-month observation period.
Table E2.
Survival analysis
| Log-Rank test | ||||
|---|---|---|---|---|
|
|
||||
| Group comparison | Event (outcome) | X2 | P value | HR (95% CI) |
| Black vs. White | Any biologic | 7.28 | .007 | 1.41 (1.10, 1.81) |
| Anti-IL-5 | 0.61 | .433 | 1.17 (0.79, 1.75) | |
| Dupilumab | 1.60 | .206 | 1.47 (0.81, 2.68) | |
| Omalizumab | 5.36 | .021 | 1.53 (1.06, 2.19) | |
| Hispanic vs. White | Any biologic | 1.04 | .308 | 0.86 (0.65, 1.15) |
| Anti-IL-5 | 4.48 | .034 | 0.65 (0.43, 0.97) | |
| Dupilumab | 0.11 | .745 | 0.88 (0.39, 1.95) | |
| Omalizumab | 0.97 | .326 | 1.25 (0.80, 1.95) | |
| Hispanic White vs. White | Any biologic | 0.02 | .894 | 0.97 (0.62, 1.51) |
| Asian vs. White | Any biologic | 0.50 | .481 | 1.24 (0.68, 2.24) |
Results shown are after 1:1 propensity score matching. In this analysis, patients are removed from the analysis (censored) after the last fact in their record. TriNetX calculates the hazard ratio (HR) and its associated confidence interval (CI) using the R’s Survival package v3.2–3 and validates results by comparing to output from SAS version 9.4. Kaplan-Meier estimated probabilities for initiating a specific biologic agent for asthma treatment for the different racial and ethnic groups over a 20-month observation period are available from the authors upon request.
To our knowledge, this is the first matched-cohort study that utilized real-world claims data to examine racial and ethnic disparities in biologic utilization among US adults with asthma. Our study included publicly and commercially insured patients, and we balanced our cohorts for baseline comorbidities and medication utilization. A previous cross-sectional study found that Asian race was associated with higher asthma biologic use and that Black patients had statistically non-significant higher odds of biologic use compared to White patients.5 Herein, we noted higher biologic use among Black patients with asthma, specifically higher use of omalizumab. This was unexpected as Black patients have been shown to be under-prescribed controller therapies for asthma and generally have less access to biologic therapies as seen with other diseases.6, 7 Further, asthma biologics are underutilized among publicly insured individuals, with more Black individuals being publicly insured.5, 8 This higher utilization of asthma biologics in Black individuals may be explained by Black patients suffering higher asthma morbidity and mortality relative to White patients.1–3 In contrast, Asian patients have lower asthma morbidity relative to White patients, so the reasons underlying the equivalent biologic use among Asian patients need further exploration.1 While omalizumab prescription was higher among Black patients with asthma, the prescription of anti-IL-5 agents and dupilumab was comparable to White patients. While we did not characterize asthma phenotype in this study, Black individuals have higher prevalence of allergic asthma, possibly explaining the higher utilization of omalizumab.9 However, as the eosinophilic asthma phenotype is also more common among Black patients relative to White patients, the lack of prescription differences in anti-IL-5 therapies warrants judicious interpretation.9
Hispanic patients with asthma have comparable asthma attack rates to White patients, but are more likely to seek emergency treatment for asthma.1 Therefore, the relatively lower prescription of anti-IL-5 therapy and the equivalent biologic prescription overall among White and Hispanic patients noted in our study should be cautiously interpreted, especially as more Hispanic individuals are uninsured or publicly insured, a factor we could not account for in our study.4
Our study was limited by reliance on asthma medications to define asthma severity and by the TriNetX platform’s inability to control for payer type. Since more Hispanic and Black patients are publicly insured this could influence the observed rate of biologic use among these cohorts. Moreover, we could not examine biologic use among racial subgroups of Hispanic patients or biracial individuals. Lastly, by controlling for baseline medications and comorbidities, we perhaps selected patients who already had adequate access to care. Nevertheless, our data did not reveal broad racial or ethnic disparities in the prescription of asthma biologics by US physicians. Further research is needed to explore interactions of asthma morbidity and phenotype within more well-defined racial and ethnic groups.
Clinical Implications.
Patients of underrepresented backgrounds are susceptible to disparate management of chronic diseases. Retrospective analysis of claims-based data regarding asthma management showed Black patients are prescribed more anti-IgE biologics and Hispanic patients are prescribed less anti-IL-5 biologics compared to White patients.
Funding:
The project was supported by the National Center for Advancing Translational Sciences, National Institutes of Health, through Grant UL1 TR002014. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.
Footnotes
Conflicts of Interests: T.A.-S. has patents pending for MicroRNAs as Predictors of Response to Anti-IgE Therapies in Chronic Spontaneous Urticaria as well as for MicroRNAs in Methods of Treatment using Omalizumab and Ligelizumab. All other authors have no conflicts of interest to disclose.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Pate CA, Zahran HS, Qin X, Johnson C, Hummelman E, Malilay J. Asthma Surveillance - United States, 2006–2018. MMWR Surveill Summ 2021; 70:1–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Guilbert T, Zeiger RS, Haselkorn T, Iqbal A, Alvarez C, Mink DR, et al. Racial Disparities in Asthma-Related Health Outcomes in Children with Severe/Difficult-to-Treat Asthma. J Allergy Clin Immunol Pract 2019; 7:568–77. [DOI] [PubMed] [Google Scholar]
- 3.Erickson SE, Iribarren C, Tolstykh IV, Blanc PD, Eisner MD. Effect of race on asthma management and outcomes in a large, integrated managed care organization. Arch Intern Med 2007; 167:1846–52. [DOI] [PubMed] [Google Scholar]
- 4.Riley WJ. Health disparities: gaps in access, quality and affordability of medical care. Trans Am Clin Climatol Assoc 2012; 123:167–72; discussion 72–4. [PMC free article] [PubMed] [Google Scholar]
- 5.Inselman JW, Jeffery MM, Maddux JT, Shah ND, Rank MA. Trends and Disparities in Asthma Biologic Use in the United States. J Allergy Clin Immunol Pract 2020; 8:549–54.e1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Hasegawa K, Ahn J, Brown MA, Press VG, Gabriel S, Herrera V, et al. Underuse of guideline-recommended long-term asthma management in children hospitalized to the intensive care unit: a multicenter observational study. Ann Allergy Asthma Immunol 2015; 115:10–6.e1. [DOI] [PubMed] [Google Scholar]
- 7.Sylvestre S, Craig T, Ajewole O, Craig S, Kaur S, Al-Shaikhly T. Racial and Ethnic Disparities in the Research and Care of Hereditary Angioedema Patients in the United States. J Allergy Clin Immunol Pract 2021; 9:4441–9 e2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Robin A. Cohen MEM, Cha Amy E., Terlizzi Emily P.. Health Insurance Coverage: Early Release of Estimates From the National Health Interview Survey, January–June 2021. cdc.gov, 2021. [Google Scholar]
- 9.Lester LA, Rich SS, Blumenthal MN, Togias A, Murphy S, Malveaux F, et al. Ethnic differences in asthma and associated phenotypes: collaborative study on the genetics of asthma. J Allergy Clin Immunol 2001; 108:357–62. [DOI] [PubMed] [Google Scholar]