Introduction

Chronic obstructive pulmonary disease (COPD) is a progressive inflammatory lung disease characterized by airflow obstruction and symptoms of dyspnea, cough, and sputum production.1 Airway inflammation in COPD is predominantly neutrophil-mediated; however, an estimated 20–40% of patients also have elevated eosinophilic inflammation.2–4 Evidence suggests that patients with eosinophilic phenotypes are most likely to respond favorably to treatment with anti-inflammatory inhaled corticosteroids (ICS), presenting an opportunity for personalized medicine.5,6

Peripheral blood eosinophil counts (BEC) have been shown to correlate with sputum eosinophil counts and are increasingly used as a more convenient surrogate for eosinophilic lung inflammation.7 BEC predicts the magnitude of the effect of ICS in the prevention of COPD exacerbations, with BEC ≥300 cells/μL predictive of the greatest likelihood of treatment benefit.5,8 For patients with BEC in the range of 100–300 cells/μL, treatment response may be variable. In contrast, ICS have little or no effect in patients with BEC <100 cells/μL and may increase the risk of pneumonia.8,9 Therefore, BEC has emerged as a biomarker for responsiveness to ICS treatment in patients with COPD.10,11

The 2025 Global Initiative for Chronic Obstructive Lung Disease (GOLD) report recommends pharmacological management of stable COPD based on assessment of patient symptoms, history of exacerbations, and BEC.12 Bronchodilator therapy with a long-acting muscarinic antagonist (LAMA), long-acting beta2-agonist (LABA), or a combination of LAMA+LABA is recommended as initial maintenance therapy in patients with zero or one moderate exacerbation (not leading to hospital admission) per year (GOLD group A/B). For patients with two or more moderate exacerbations or at least one exacerbation leading to hospitalization (GOLD group E), dual therapy with LAMA+LABA is recommended. Triple therapy (LAMA+LABA+ICS) may be considered as an initial treatment for GOLD group E patients with BEC ≥300 cells/μL. Follow-up treatment is recommended based on the presence of dyspnea or exacerbations while on current maintenance therapy. Escalation to triple therapy is suggested only for patients with continued exacerbations on bronchodilator monotherapy with BEC ≥300 cells/μL or dual bronchodilator therapy with BEC ≥100 cells/μL.

Although treatment algorithms have changed over time, since 2011, GOLD has consistently recommended reserving triple therapy for patients with frequent exacerbations. Despite these evidence-based recommendations, one in five patients starting triple therapy in clinical practice is maintenance-naive with a low exacerbation history.13 The benefits of ICS should be balanced against the potential risks of ICS-associated adverse effects, including pneumonia, oral candidiasis, skin bruising, and unfavorable changes in the lung microbiome.14,15 We hypothesized that the exacerbation reduction benefits of triple therapy versus LAMA/LABA observed in randomized controlled trials (RCTs) would not translate to routine clinical practice in patients with BEC ≤300 cells/μL. Therefore, this study aimed to compare COPD exacerbations, healthcare resource utilization (HRU), and costs in patients initiating dual bronchodilator therapy with tiotropium/olodaterol (TIO/OLO) or triple therapy with fluticasone furoate/umeclidinium/vilanterol (FF/UMEC/VI), with BEC ≤300 cells/μL.

Methods Study Design and Data Source

This retrospective, non-interventional cohort study used medical and pharmacy claims data from the Optum Research Database (ORD). Medical claims included diagnosis and procedure codes recorded using the International Classification of Diseases, Ninth (ICD-9-CM) and Tenth (ICD-10-CM) Revisions, Clinical Modification codes, Current Procedural Terminology codes, Healthcare Common Procedure Coding System codes, site of service codes, provider specialty codes, revenue codes (for facilities), paid amounts, and other information. This claims database is fully de-identified and compliant with the Health Insurance Portability and Accountability Act. This study used aggregated, de-identified insurance claims data and, therefore, required no review by an institutional review board or informed consent procedures. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines to ensure thorough reporting of the cohort study design, participant selection, variable measurements, and statistical analysis. This study was registered at ClinicalTrials.gov (identifier: NCT05127304). The results for the overall study population have been reported elsewhere.16

Study Population

Patients with COPD were identified based on the presence of at least one facility claim with a diagnosis of COPD in the primary position or at least two professional claims with a diagnosis code of COPD in any position on separate dates of service from 01 June 2014 through 31 December 2019 (study period). The study population included patients with at least 30 consecutive days supply of TIO/OLO or FF/UMEC/VI initiated during the identification period of 01 June 2015 to 30 November 2019, allowing for a minimum of 12 months of data prior to and 30 days after treatment initiation. The index date was defined as the fill date of the first TIO/OLO or FF/UMEC/VI pharmacy claim with a minimum of 30 days of consecutive treatment during the identification period. Continuous health plan enrollment with medical and pharmacy coverage for 12 months prior to and including the index date (baseline period) and for at least 30 days following the index date was required. Included patients were at least 40 years of age on the index date. Individuals with pharmacy claims for both TIO/OLO and FF/UMEC/VI on the index date, and those who switched to another maintenance therapy (any non-index LAMA, LABA, or ICS) within 30-days following the index date were excluded. Patients with two or more baseline medical claims on separate dates of service with a diagnosis of asthma, cystic fibrosis, lung cancer, or interstitial lung disease, or any medical claim for lung volume reduction procedure were also excluded. Patients with 7 or more days of overlapping supply of LAMA and LABA, with or without ICS, in the six months prior to the index date were excluded to ensure that the included patients were new users at the class level. Individuals were also excluded due to unknown age, gender, business line, or unknown/other geographic region.

The subgroup of patients with at least one blood eosinophil laboratory result during the 6 months prior to the index date through 15 days after the index date was retained for this analysis.

Measures

The 12-month period prior to and including the index date (index date – 364 days through index date) was used to assess baseline patient characteristics. Baseline measures included patient demographics, clinical characteristics, exacerbation history, respiratory medication use, HRU, and costs. Baseline respiratory medication use excluded claims on the index date. Patients without any baseline claims for LAMA, LABA, or ICS were considered maintenance-naive.

BEC results from 6 months prior to the index date through 15 days after the index date were assessed and categorized as ≤300 cells/μL or >300 cells/μL, utilizing the result closest to the index date. The highest value was retained for multiple eosinophil laboratory results on the same service date.

Moderate COPD exacerbation was defined as an ambulatory (office or outpatient) visit with a diagnosis code for COPD plus a pharmacy claim for an oral corticosteroid and/or COPD-guideline-recommended antibiotic within seven days of the visit. Severe exacerbation was defined as an inpatient or emergency department (ED) visit with 1) a primary diagnosis of COPD, 2) a primary diagnosis of acute respiratory failure plus a COPD diagnosis, or 3) a primary diagnosis of acute respiratory failure plus an inpatient or ED visit with a COPD diagnosis in any position within ± seven days. Exacerbations occurring within 14 days of each other were considered a single exacerbation episode and were classified according to the highest severity contributing event. Patients were categorized into GOLD group A/B (zero or one baseline moderate exacerbation [not leading to hospital admission]) or GOLD group E (two or more baseline moderate exacerbations or at least one exacerbation leading to hospitalization).

HRU measures included ambulatory (physician office and hospital outpatient), ED, inpatient, other medical visits, and pharmacy claims. Other medical visits included services provided at independent laboratories, assisted living facilities, and by home health providers. COPD and/or pneumonia-related HRU measures were restricted to medical claims with a diagnosis of COPD, pneumonia, or acute bronchitis/bronchiolitis and pharmacy claims for COPD-related treatment, including guideline-recommended antibiotics.

HRU costs were computed as the sum of health plan and patient-paid (copay and/or coinsurance) costs. Total costs were presented as combined medical and pharmacy costs. Costs were adjusted to 2020 US dollars using the medical care component of the Consumer Price Index.17

Outcomes were assessed over a variable follow-up period, with a minimum of 30 days and maximum of 12 months. Patients were followed up until the earliest of treatment switch, discontinuation, health plan disenrollment, end of the study period (31 December 2019), or 1 year of follow-up. Discontinuation was defined as a gap in therapy of ≥60 days following the runout of days’ supply, and the discontinuation date was defined as the runout prior to the gap in therapy. Patients with <60 days of observation time following the runout of their last index prescription claim were flagged as censored because the discontinuation status of patients was unobservable, and their observation period ended on the runout date. An index therapy switch was defined as a pharmacy fill for ≥30 consecutive days of non-index ICS-, LABA-, or LAMA-containing medication (switch date = the first date of the new treatment regimen). The outcomes assessed during the follow-up period included COPD exacerbations, HRU, and costs.

Statistical Analysis

Patients were matched in a 1:1 ratio of TIO/OLO to FF/UMEC/VI initiators without replacement, using nearest-neighbor matching with a caliper of ±0.01. Covariates in the logistic regression model used to compute propensity scores included patient demographics (age, sex, geographic region, insurance type), proxies for COPD severity (baseline moderate and severe exacerbation counts, GOLD group), baseline COPD medication use (maintenance-naive status, count of controller and rescue medications, nebulized respiratory medication, and oxygen therapy), baseline HRU, and costs (all-cause, COPD-related, and pneumonia-related total costs). Patients not matched were excluded from the post-match analysis.

Baseline characteristics before and after propensity score matching were compared between TIO/OLO and FF/UMEC/VI initiators using standardized differences. A standardized difference of ≤10% for a measure was considered an acceptable threshold for balance between the treatment cohorts.

Kaplan-Meier analyses were conducted to examine the time to the first COPD exacerbation (any or severe) and pneumonia diagnosis. Cumulative incidence was estimated as the complement of the Kaplan-Meier survival function. Follow-up counts of COPD exacerbations, HRU, and costs were compared between cohorts and presented as annualized population averages with Wald 95% confidence intervals, using Taylor expansion to estimate the standard error. Annualized population averages were computed as the [(sum of outcome events or costs for each cohort)/(sum of follow-up time in years for each cohort)].

COPD and/or pneumonia-related total, medical, and pharmacy costs were also modeled using generalized linear modeling with a gamma distribution and log link to adjust for residual imbalance after propensity score matching. Statistical tests between cohorts included variance adjustments for clustering due to matching.

Results

From the overall population of TIO/OLO (n=6681) and FF/UMEC/VI (n=9657) initiators who met the study inclusion criteria, 1649 (24.7%) and 2218 (23.0%) had a baseline blood eosinophil laboratory result, respectively (Figure 1). Among these, 1372 (83.2%) TIO/OLO initiators and 1796 (81.0%) FF/UMEC/VI initiators had BEC ≤300 cells/μL. Prior to matching, baseline characteristics between TIO/OLO initiators and FF/UMEC/VI initiators were imbalanced for age, insurance coverage, anxiety, allergic rhinitis, sleep apnea, exacerbation history, prior COPD medication use, and costs (Table 1). After matching, 1098 patients from each cohort were retained. TIO/OLO and FF/UMEC/VI initiators were well balanced on baseline characteristics, except for index year, prior respiratory medication use, allergic rhinitis, and anxiety. Proxies of disease severity, including baseline COPD exacerbations, pneumonia-related events, as well as all-cause and disease-related HRU and cost measures, were well balanced between the study cohorts. Approximately 55% of the patients in each cohort were naive to maintenance therapy and 68% had a GOLD A/B exacerbation history. The mean follow-up durations (days) were similar between cohorts (mean [SD]: 122.6 [112.7] and 129.3 [106.4] for TIO/OLO and FF/UMEC/VI, respectively). However, the reason for the end of follow-up differed between cohorts, with 6.9% of TIO/OLO initiators and 23.6% of FF/UMEC/VI initiators censored due to end of the study period (31 December 2019). Approximately, 9.4% of the TIO/OLO cohort and 5.7% of the FF/UMEC/VI cohort had a complete 12-months of follow-up.

Table 1 Baseline Characteristics of TIO/OLO and FF/UMEC/VI Cohorts with BEC ≤300 Cells/µL Before and After Propensity Score Matching

Figure 1 Patient Disposition.

Abbreviations: COPD, chronic obstructive pulmonary disease; TIO/OLO, tiotropium/olodaterol; FF/UMEC/VI, fluticasone furoate/umeclidinium/vilanterol; BEC, blood eosinophil count; TT, triple therapy; LAMA, long-acting muscarinic antagonist; LABA, long-acting beta2-agonist.

Notes: Bold text indicates study sample and blood eosinophil count cohorts, before and after propensity score matching.

COPD Exacerbations and Pneumonia

Kaplan-Meier analysis censor-adjusted proportions of patients experiencing moderate or severe exacerbation at one year were 48.4% for TIO/OLO initiators and 51.4% for FF/UMEC/VI initiators. The incidence of severe exacerbation at one year was 18.0% for TIO/OLO initiators and 18.6% for FF/UMEC/VI initiators. The time to the first moderate or severe exacerbation (p=0.830) and severe exacerbation (p=0.988) was not significantly different between the TIO/OLO and FF/UMEC/VI initiators (Figures 2 and 3). The follow-up counts (95% CI) of annualized COPD exacerbations were also not significantly different between TIO/OLO and FF/UMEC/VI initiators for moderate or severe exacerbations (1.05 [0.92–1.18] vs 0.99 [0.88–1.11], p=0.535) and severe exacerbations (0.29 [0.23–0.36] vs 0.28 [0.22–0.34], p=0.821).

Figure 2 Time to First Moderate or Severe Exacerbation in the Matched Population with Eosinophil Count ≤300 cells/μL.

Abbreviations: TIO/OLO, tiotropium/olodaterol; FF/UMEC/VI, fluticasone furoate/umeclidinium/vilanterol.

Figure 3 Time to First Severe Exacerbation in the Matched Population with Eosinophil Count ≤300 cells/μL.

Abbreviations: TIO/OLO, tiotropium/olodaterol; FF/UMEC/VI, fluticasone furoate/umeclidinium/vilanterol.

The censor-adjusted findings from the Kaplan-Meier analysis indicated the proportions of patients with a pneumonia diagnosis at one year were 15.6% of TIO/OLO initiators and 17.8% of FF/UMEC/VI initiators, with no significant difference in time to first pneumonia diagnosis (p= 0.772) (Supplementary Figure 1).

Healthcare Resource Utilization

Follow-up annualized COPD and/or pneumonia-related HRU measures were not significantly different between cohorts, except for fewer ED visits among TIO/OLO initiators versus FF/UMEC/VI (0.59 [0.46–0.71] vs 0.83 [0.67–0.98], p=0.018) (Table 2).

Table 2 Population Annualized Average Counts of Follow-up COPD and/or Pneumonia-Related Healthcare Resource Utilization in Matched TIO/OLO and FF/UMEC/VI Cohorts with BEC ≤300 Cells/µL

Costs

The follow-up population annualized average COPD and/or pneumonia-related total (medical and pharmacy) and medical costs were not significantly different between the TIO/OLO and FF/UMEC/VI initiators (Table 3). The total COPD and/or pneumonia-related costs for the TIO/OLO and FF/UMEC/VI initiators were $18,097 and $20,130, respectively (p=0.326). Pharmacy costs were 28.6% lower for TIO/OLO initiators than for FF/UMEC/VI initiators ($4692 vs $6573, p<0.001). The ED visit costs were also lower for TIO/OLO than for FF/UMEC/VI initiators ($370 vs $538, p=0.034). Costs adjusted for baseline characteristics with residual imbalance after propensity score matching using a generalized linear model were consistent with the population annualized costs (Supplementary Table 1).

Table 3 Population Annualized Average Costs of Follow-up COPD and/or Pneumonia-Related Healthcare Resource Utilization in Matched TIO/OLO and FF/UMEC/VI Cohorts with BEC ≤300 Cells/µL

Discussion

This retrospective, non-interventional cohort study assessed clinical and economic outcomes following the initiation of dual bronchodilator therapy with TIO/OLO compared to triple therapy with FF/UMEC/VI in patients with COPD and BEC ≤300 cells/μL. In accordance with GOLD treatment recommendations, the findings suggest that triple therapy did not reduce the risk of COPD exacerbation compared with LAMA/LABA therapy. This result is consistent with that of another real-world study comparing single-inhaler triple therapy with LAMA/LABA, which found no significant difference in the incidence of moderate or severe COPD exacerbations in patients with BEC ≤300 cells/μL, even among a subset of GOLD E patients.18,19

Evidence from RCTs of triple therapy supports the role of BEC in the identification of patients who are most likely to benefit from the addition of ICS to LAMA/LABA. AIRWISE was a pragmatic clinical trial that randomized patients not controlled on LABA, LAMA, or LABA/ICS to either TIO/OLO or triple therapy. A post-hoc subgroup analysis found that the risk of a first moderate or severe COPD exacerbation was similar between TIO/OLO and triple therapy for patients with BEC <300 cells/μL (hazard ratio [95% CI]: 1.02 [0.68–1.53]).20 The IMPACT and ETHOS trials, which randomized patients to single-inhaler triple therapy or LAMA/LABA, found that the magnitude of the benefit of triple therapy on exacerbation reduction was greater with increased BEC. Data modeling from the IMPACT trial revealed a continuous relationship between BEC and exacerbation risk reduction above BEC of 100 cells/μL.21 Post-hoc analysis of the ETHOS trial found that the benefit of triple therapy in reducing the rate of moderate/severe exacerbations compared to LAMA/LABA was greatest in patients with BEC >300 cells/μL.22 However, the generalizability of RCT findings of TT to routine clinical practice may be limited by the study design and selected patient population. The IMPACT and ETHOS trials primarily enrolled patients with a history of exacerbation in the previous year and use of maintenance medications at screening.23,24 Notably, most patients included in our study (approximately 68%) had GOLD A/B exacerbation history, and over half had no evidence of maintenance inhaler therapy in the 12 months prior to the initiation of TIO/OLO or FF/UMEC/VI. A study that examined the eligibility of patients with COPD seen in routine clinical practice in England (n=79,810) for inclusion in RCTs estimated that only 8.9% and 2.0% of the patients would have been eligible for the IMPACT and ETHOS trials, respectively.25 The inclusion criteria for COPD exacerbations and the use of prior maintenance medication led to the exclusion of the greatest proportion of patients, highlighting the lack of generalizability of these trials.

In this study, HRU and costs were generally similar among initiators of TIO/OLO and FF/UMEC/VI. However, ED visits and pharmacy costs related to COPD (including maintenance medications, short-acting bronchodilators, oral corticosteroids, and antibiotics) were significantly lower among TIO/OLO users.

Approximately 17% and 19% of patients initiating TIO/OLO and FF/UMEC/VI, respectively, had a baseline BEC >300 cells/μL. This finding suggests that most patients with COPD who initiate dual bronchodilator or triple therapy do not have high eosinophil counts. Furthermore, less than half (47.2%) of the FF/UMEC/VI initiators with BEC >300 cells/μL also had GOLD E exacerbation history, which is the population most likely to benefit from TT.

The limitations of this study include those inherent to the use of administrative claims data for retrospective observational research. Claims are subject to miscoding, omissions, and lack some clinically important data such as lung function measures, symptom burden, and behavioral characteristics. Propensity score matching was employed to mitigate the risks of confounding by indication, and resulted in well-balanced baseline characteristics of TIO/OLO and FF/UMEC/VI initiators with BEC ≤300 cells/μL. However, several baseline measures remained imbalanced after matching. The proportion of patients with comorbid allergic rhinitis and anxiety differed between cohorts, with standardized differences just above the a priori defined threshold of 10%. Prior maintenance medication use also differed, with a greater number of FF/UMEC/VI initiators with baseline use of LABA/ICS, reflecting expected treatment patterns. Multivariable analyses adjusted for baseline measures that remained imbalanced after matching were conducted for select cost outcomes, with consistent results. Our findings are also limited due to relatively short follow-up duration of approximately 4 months on average, however this reflects real-world treatment patterns (ie, treatment switching and discontinuation), our study design, and data limitations (eg, censoring due to end of data cut).

Additionally, the limited sample size and differences in baseline characteristics between cohorts with BEC >300 cells/μL resulted in inadequate balance after propensity score matching; therefore, we were unable to assess the outcomes in this population. Given the differing dates of market availability, TIO/OLO users who initiated therapy in the years prior to the approval of FF/UMEC/VI were included in this study, resulting in an imbalance in index year. A greater proportion of FF/UMEC/VI patients had index dates in 2019, which resulted in more frequent censoring of follow-up due to the end of the data period. Furthermore, the generalizability of our findings is limited by the study population, which may not represent the broader COPD patient population without commercial or Medicare Advantage insurance coverage. Finally, COPD exacerbations were identified using a claims-based algorithm, which may have resulted in misclassifications. However, the algorithm used in this study was similar to that used in other COPD studies using claims data, including validated algorithms with high performance characteristics.26

Conclusion

Among patients with COPD initiating FF/UMEC/VI, eight in ten had BEC ≤300 cells/μL and less than 10% had BEC >300 cells/μL with GOLD E exacerbation history. TIO/OLO and FF/UMEC/VI initiators with BEC ≤300 cells/μL experienced similar rates of COPD exacerbations and COPD and/or pneumonia-related HRU. TIO/OLO initiators incurred lower ED and pharmacy costs related to COPD and/or pneumonia compared with FF/UMEC/VI initiators. These findings support practice recommendations for the assessment of BEC when considering initiating triple therapy and for reserving triple therapy for frequent exacerbators and patients with BEC ≥300 cells/μL.

Data Sharing Statement

The data used were licensed from Optum and are not publicly available.

Ethics Approval and Informed Consent

As this was a retrospective study, informed consent was not required. All data were de-identified, and therefore, the study did not necessitate ethics committee approval and was fully compliant with HIPAA regulations.

Acknowledgments

The authors wish to thank Felix Cao and Yiyu Fang of Optum for dataset programming support; Swetha R. Palli, MS, of Boehringer Ingelheim at the time of the study, for conceptualization; and Marjan Zakeri, MD, PhD, of Boehringer Ingelheim, for editorial support and formatting assistance. Partial findings from this study were presented at the American Thoracic Society International Conference, May 19-34, 2023, in Washington, DC, USA.

Author Contributions

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Funding

This study was supported and funded by Boehringer Ingelheim Pharmaceuticals, Inc. The authors did not receive payment related to the development of the manuscript. The authors meet criteria for authorship as recommended by the ICMJE. Boehringer Ingelheim Pharmaceuticals, Inc. was given the opportunity to review the manuscript for medical and scientific accuracy as well as intellectual property considerations.

Disclosure

SS has received honoraria for consulting and speaking from Astra Zeneca, Boehringer Ingelheim, Chiesi, Nuvaira, Pulmonx, Glaxo Smith Kline and Sanofi-Regeneron. His institution has received research grants from Chiesi and Astra Zeneca. LGSB was an employee of Optum at the time of the study. EKB and AS are employees of Optum. LGSB and BC are employees of Boehringer Ingelheim Pharmaceuticals, Inc. The authors report no other conflicts of interest in this work.

References

1. Celli B, Fabbri L, Criner G, et al. Definition and nomenclature of chronic obstructive pulmonary disease: time for its revision. Am J Respir Crit Care Med. 2022;206(11):1317–1325. doi:10.1164/rccm.202204-0671PP

2. Higham A, Beech A, Wolosianka S, et al. Type 2 inflammation in eosinophilic chronic obstructive pulmonary disease. Allergy. 2021;76(6):1861–1864. doi:10.1111/all.14661

3. Brightling C, Greening N. Airway inflammation in COPD: progress to precision medicine. Eur Respir J. 2019;54(2):1900651. doi:10.1183/13993003.00651-2019

4. Oshagbemi OA, Burden AM, Braeken DCW, et al. Stability of blood eosinophils in patients with chronic obstructive pulmonary disease and in control subjects, and the impact of sex, age, smoking, and baseline counts. Am J Respir Crit Care Med. 2017;195(10):1402–1404. doi:10.1164/rccm.201701-0009LE

5. Pascoe S, Locantore N, Dransfield MT, Barnes NC, Pavord ID. Blood eosinophil counts, exacerbations, and response to the addition of inhaled fluticasone furoate to vilanterol in patients with chronic obstructive pulmonary disease: a secondary analysis of data from two parallel randomised controlled trials. Lancet Respir Med. 2015;3(6):435–442. doi:10.1016/S2213-2600(15)00106-X

6. Siddiqui SH, Guasconi A, Vestbo J, et al. Blood eosinophils: a biomarker of response to extrafine beclomethasone/formoterol in chronic obstructive pulmonary disease. Am J Respir Crit Care Med. 2015;192(4):523–525. doi:10.1164/rccm.201502-0235LE

7. Negewo NA, McDonald VM, Baines KJ, et al. Peripheral blood eosinophils: a surrogate marker for airway eosinophilia in stable COPD. Int J Chronic Obstr Pulm Dis. 2016;11:1495–1504. doi:10.2147/COPD.S100338

8. Stockley RA, Halpin DMG, Celli BR, Singh D. Chronic obstructive pulmonary disease biomarkers and their interpretation. Am J Respir Crit Care Med. 2018;199(10):1195–1204. doi:10.1164/rccm.201810-1860SO

9. Martinez-Garcia MA, Faner R, Oscullo G, et al. Inhaled steroids, circulating eosinophils, chronic airway infection, and pneumonia risk in chronic obstructive pulmonary disease. A network analysis. Am J Respir Crit Care Med. 2020;201(9):1078–1085. doi:10.1164/rccm.201908-1550OC

10. Singh D. Blood eosinophil counts in chronic obstructive pulmonary disease: a biomarker of inhaled corticosteroid effects. Tuberc Respir Dis. 2020;83(3):185–194. doi:10.4046/trd.2020.0026

11. David B, Bafadhel M, Koenderman L, Soyza AD. Eosinophilic inflammation in COPD: from an inflammatory marker to a treatable trait. Thorax. 2021;76(2):188–195. doi:10.1136/thoraxjnl-2020-215167

12. Global Initiative for Chronic Obstructive Lung Disease (GOLD). Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease (2025 report). Available from: https://goldcopd.org/2025-gold-report/. Accessed December 2024.

13. Bhatt SP, Blauer-Peterson C, Buysman EK, Bengtson LGS, Palli SR. Trends and characteristics of global initiative for chronic obstructive lung disease guidelines-discordant prescribing of triple therapy among patients with COPD. Chronic Obstr Pulm Dis. 2022;9(2):135–153.

14. Quint JK, Ariel A, Barnes PJ. Rational use of inhaled corticosteroids for the treatment of COPD. NPJ Prim Care Respir Med. 2023;33(1):27. doi:10.1038/s41533-023-00347-6

15. Keir HR, Contoli M, Chalmers JD. Inhaled corticosteroids and the lung microbiome in COPD. Biomedicines. 2021;9(10):1312. doi:10.3390/biomedicines9101312

16. Sethi S, Palli SR, Bengtson LGS, et al. Clinical and economic outcomes in patients with chronic obstructive pulmonary disease initiating maintenance therapy with tiotropium bromide/olodaterol or fluticasone furoate/umeclidinium/vilanterol. J Manag Care Spéc Pharm. 2023;29(7):791–806. doi:10.18553/jmcp.2023.22373

17. U.S. Department of Labor, Bureau of Labor Statistics. Consumer price index. medical care. Series ID: CUUR0000SAM. Washington, DC: U.S. Dept. of Labor, Bureau of Labor Statistics; 2022. Available from: http://data.bls.gov/cgi-bin/surveymost?cu. Accessed January26, 2022.

18. Suissa S, Dell’Aniello S, Ernst P. Single-inhaler triple versus dual bronchodilator therapy in COPD: real-world comparative effectiveness and safety. Int J Chronic Obstr Pulm Dis. 2022;17:1975–1986. doi:10.2147/COPD.S378486

19. Suissa S. Single-inhaler triple versus dual bronchodilator therapy for GOLD group E and other exacerbating patients with COPD: real-world comparative effectiveness and safety. Eur Respir J. 2023;62(3):2300883. doi:10.1183/13993003.00883-2023

20. Mina BA, Averill F, Shaikh A, et al. Exacerbation risk in COPD patients based on eosinophil count: post hoc analysis of the randomized pragmatic trial, AIR-WISE, comparing tiotropium/olodaterol with triple therapy. Chest. 2022;162(4):A1899–A1900. doi:10.1016/j.chest.2022.08.1581

21. Pascoe S, Barnes N, Brusselle G, et al. Blood eosinophils and treatment response with triple and dual combination therapy in chronic obstructive pulmonary disease: analysis of the IMPACT trial. Lancet Respir Med. 2019;7(9):745–756. doi:10.1016/S2213-2600(19)30190-0

22. Bafadhel M, Rabe KF, Martinez FJ, et al. Benefits of budesonide/glycopyrronium/formoterol fumarate dihydrate on COPD exacerbations, lung function, symptoms, and quality of life across blood eosinophil ranges: a post-hoc analysis of data from ETHOS. Int J Chronic Obstr Pulm Dis. 2022;17:3061–3073. doi:10.2147/COPD.S374670

23. Lipson DA, Barnhart F, Brealey N, et al. Once-daily single-inhaler triple versus dual therapy in patients with COPD. N Engl J Med. 2018;378(18):1671–1680. doi:10.1056/NEJMoa1713901

24. Rabe KF, Martinez FJ, Ferguson GT, et al. Triple inhaled therapy at two glucocorticoid doses in moderate-to-very-severe COPD. N Engl J Med. 2020;383(1):35–48. doi:10.1056/NEJMoa1916046

25. Whittaker HR, Torkpour A, Quint J. Eligibility of patients with chronic obstructive pulmonary disease for inclusion in randomised control trials investigating triple therapy: a study using routinely collected data. Respir Res. 2024;25(1):43. doi:10.1186/s12931-024-02672-x

26. Mapel DW, Roberts MH, Sama S, et al. Development and validation of a healthcare utilization-based algorithm to identify acute exacerbations of chronic obstructive pulmonary disease. Int J Chronic Obstr Pulm Dis. 2021;16:1687–1698. doi:10.2147/COPD.S302241