Progression from GOLD A/B to GOLD E: a claims analysis of patients with COPD newly initiating inhaled therapy

Abstract

Background

Chronic obstructive pulmonary disease (COPD) is a progressive disease associated with substantial morbidity and mortality. Acute COPD exacerbations are a primary driver of significant burden and contribute to disease progression.

Methods

This retrospective, observational cohort study used the Optum Clinformatics^® Data Mart database to identify patients with COPD who were classified as Global Initiative for Chronic Obstructive Lung Disease (GOLD) A/B0 or A/B1 based on exacerbation history (i.e., they had either 0 [GOLD A/B0] or 1 [GOLD A/B1] moderate exacerbation and 0 severe exacerbations in a 12-month baseline period). Patients were required to be aged ≥ 40 years and to have newly initiated inhaled maintenance therapy for COPD from January 2016 to June 2023. The rates of and time to progression to GOLD E (defined in the claims data as experiencing 2 moderate exacerbations within a 12-month period or 1 severe exacerbation) were estimated using the Kaplain-Meier method. Predictors of progression to GOLD E were analyzed using multivariable Cox proportional hazard models.

Results

Of the 156,462 included patients, the largest proportion of patients (46.6%) were initiated on long-acting beta-agonists/inhaled corticosteroids. The majority of patients progressed to GOLD E over 5 years. The risk of progressing to GOLD E was approximately 3 times higher in the GOLD A/B1 versus GOLD A/B0 group (hazard ratio [HR] 2.92; 95% CI 2.84–3.00; P < 0.001). The strongest predictor of progressing to GOLD E was history of having a moderate exacerbation. Other independent predictors included older age, having Medicare versus commercial insurance, and the presence of Elixhauser comorbidities.

Conclusions

Despite use of inhaled maintenance treatments for COPD, most patients still progressed to a frequent or severe exacerbator phenotype. New therapies are needed to modify the disease trajectory in COPD.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12890-025-03898-1.

Introduction

Chronic obstructive pulmonary disease (COPD) is a leading cause of morbidity and mortality worldwide and is responsible for a substantial and increasing clinical and economic burden [1]. A significant portion of the burden of COPD is due to exacerbations, which are sudden periods of worsening of symptoms that can be triggered by infections or environmental pollutants [1–3]. Exacerbation severity is typically classified by the degree of healthcare utilization required, with moderate exacerbations treated in an outpatient setting and severe exacerbations requiring hospitalization or an emergency department (ED) visit [1]. Exacerbations contribute significantly to disease progression and result in an increased risk of cardiovascular events and mortality [1, 4–6]. A history of exacerbations has been identified as one of the strongest risk factors for future exacerbations [7] and is therefore a defining factor in the management of COPD.

Current assessment tools for COPD recognize the impact of exacerbations on disease progression, incorporating exacerbation history into treatment algorithms [1]. Specifically, the 2025 Global Initiative for Chronic Obstructive Lung Disease (GOLD) Report recommends initial maintenance pharmacological treatment for patients depending on their history of exacerbations and symptoms [1]. Previous iterations of the GOLD Report initially moved from a spirometric grading system to one with a combination of symptoms based on the modified Medical Research Council (mMRC) dyspnea questionnaire or the COPD Assessment Test (CAT) score, the severity of airflow obstruction, and the frequency of previous exacerbations [1]. More recent reports removed the severity of airflow obstruction from the assessment scheme [1]. In 2023, the GOLD Report was updated to combine historical categories C and D (history of previous exacerbations, delineated by symptom burden) into one category, GOLD E for patients at high risk of exacerbations [1]. Here, patients are classified into 1 of 3 groups; GOLD A, GOLD B, or GOLD E [1]. The GOLD A and GOLD B classification encompasses patients with 0 or 1 moderate exacerbations in the prior year [1]. The delineation between GOLD A and GOLD B is determined by symptom burden, as measured by the mMRC dyspnea questionnaire (GOLD A, mMRC 0–1; GOLD B, mMRC ≥ 2) or the CAT score (GOLD A, CAT < 10; GOLD B, CAT ≥ 10) [1]. GOLD E does not delineate based on symptom burden and includes patients with a history of frequent or severe exacerbations, defined as ≥ 2 moderate exacerbations or ≥ 1 severe exacerbation in the prior year [1].

The current literature regarding COPD exacerbations focuses on burden and risk factors for future exacerbations. The progression to a frequent/severe exacerbator phenotype (i.e., from GOLD A/B to GOLD E) in a real-world United States (US) population is unknown. Understanding the rates of progression despite initiation of inhaled maintenance therapy can help assess clinical burden and unmet needs. Additionally, identifying key predictors of exacerbations may inform the development of recommendations for early interventions in specific patient subgroups. As just one exacerbation has detrimental effects on disease progression and mortality risk [2, 4, 6], management of COPD should focus on preventing the first sentinel exacerbation and progression to a frequent/severe exacerbator phenotype, in addition to preventing subsequent exacerbations.

The Optum Clinformatics^® DataMart (CDM) database has been extensively used for research related to COPD, including patient characteristics and outcomes based on exacerbation history [8–10]. Here, we used the Optum CDM database to identify and describe patients with COPD in the US who initiated inhaled maintenance therapy in routine clinical practice settings. The objective of this study was to provide real-world evidence of the frequency and predictors of progression from GOLD A/B to GOLD E while patients remained on treatment after newly initiating inhaled maintenance therapy.

Methods

Data sources

This retrospective, observational cohort study used deidentified inpatient, outpatient, and pharmacy claims from the Optum CDM dataset, which contains US claims data derived from 84 million patients with commercial and Medicare Advantage insurance plans. Patient claims data from January 2016 to June 2023 were used for this analysis. As all analyses were conducted on anonymized retrospective patient data, institutional review board approval was not sought. The analyzed dataset was licensed by the study sponsor from Optum^®, with permission obtained from Optum to access and use these non-publicly available data.

Study population and design

This study included Optum CDM data from patients aged ≥ 40 years with a COPD diagnosis who initiated new inhaled maintenance therapy during the observation period (January 2016 to June 2023; Fig. 1). The Optum CDM data included paid claims for all healthcare services reimbursed by patients’ health plans. Claims were submitted by healthcare providers and subsequently verified and adjudicated by the health insurance plans. These data contained information on patients’ demographics (including age, sex, race, and ethnicity), plan enrollment (allowing for assessment of continuous insurance coverage), diagnostic codes, and the dates for each healthcare encounter. The dataset also included information on dispensed prescription drugs, such as the National Drug Code (NDC) and days of supply. Patient-level enrollment files were merged with healthcare utilization files to create an analytic dataset that included baseline characteristics and outcomes following each patient’s index date. The index date was defined as the date of the patient’s first prescription claim for an inhaled maintenance therapy after a clean 12-month period without maintenance therapy.

Fig. 1.

Study design. COPD, chronic obstructive pulmonary disease; GOLD, Global Initiative for Chronic Obstructive Lung Disease

A COPD diagnosis defined as ≥ 1 International Classification of Diseases, Tenth Revision (ICD-10) diagnosis code for COPD (Supplemental Table 1) in any position was required during the 12-month baseline (pre-index) period. Inhaled maintenance therapies included long-acting beta-agonists (LABA), long-acting muscarinic antagonists (LAMA), LABA/LAMA, LABA/inhaled corticosteroids (LABA/ICS), or LABA/LAMA/ICS (single inhaler or multiple inhalers). Patients were excluded from the analysis if they had a diagnosis of asthma, interstitial lung disease, cystic fibrosis, and/or lung cancer during the 12-month baseline period to reduce potential confounding from coexisting conditions that may influence treatment patterns and outcomes. Patients were also excluded if they had previous inhaled maintenance therapy for COPD during the baseline period. The baseline period was defined as the 12 months prior to each patient’s index date.

Patients’ exacerbation histories were assessed in the 12-month baseline period prior to initiation of inhaled maintenance therapy. A moderate exacerbation was defined as an event treated in the office or outpatient setting with a COPD diagnosis code and a pharmacy claim for an oral corticosteroid and/or antibiotic (Supplemental Table 2) within 7 days of the visit. A severe exacerbation was defined as an event treated in an inpatient admission or ED visit, defined as either: (1) a COPD diagnosis code in the primary position; (2) acute respiratory failure diagnosis code in the primary position with a COPD diagnosis code in any position; or (3) acute respiratory failure diagnosis code in the primary position and an additional inpatient admission or ED visit within 7 days with a COPD diagnosis code in any position. If exacerbation events were identified within 14 days of each other, they were analyzed as a single exacerbation event, classified according to the event with the highest severity. Patients classified as GOLD A or GOLD B pre-index, with evidence of 0 or 1 moderate exacerbations and 0 severe exacerbations during the 12-month baseline period, were retained for this analysis. Patients were excluded if they had ≥ 2 moderate exacerbations or ≥ 1 severe exacerbation in the baseline period (i.e., patients who were classified as GOLD E). As symptom data were unavailable in healthcare claims databases, GOLD A and GOLD B could not be delineated separately and are reported together throughout this analysis (i.e., GOLD A/B). While the GOLD categories are historically used only for initial patient assessment, they were used here to identify and evaluate future exacerbations in patients who had not historically experienced frequent or severe exacerbations.

Patients were further sub-classified as either GOLD A/B0 if they had 0 moderate and 0 severe exacerbations in the baseline period prior to the index date or GOLD A/B1 if they had 1 moderate and 0 severe exacerbations in the baseline period prior to the index date. Patients were required to have ≥ 12 months of prior continuous insurance eligibility before the index date. Patients were excluded from the analysis if they had a diagnosis of asthma, interstitial lung disease, cystic fibrosis, and/or lung cancer during the 12-month baseline period. Patients were followed while on treatment, allowing for medication supply gaps < 90 days at index and during patients’ follow-up. In all time-to-event analyses, patients were censored at the first instance of any of the following: treatment discontinuation (> 90 days), end of continuous insurance eligibility, death, or at the end of the study period (June 30, 2023).

Patient characteristics and comorbidity profile

Baseline demographics and comorbidities were collected from the 12-month baseline period prior to the index date. Patient baseline demographics included age, sex, race/ethnicity, region, and type of health insurance plan at the time of study entry (i.e., the index date). Comorbidities were assessed using the Elixhauser Comorbidity Index, which is composed of 30 separate comorbidities and uses ICD-10 codes from administrative data [11, 12]. Patient variables were stratified by class of maintenance therapy (LABA, LAMA, LABA/LAMA, LABA/ICS, or LABA/LAMA/ICS) and exacerbation history (GOLD A/B0 or GOLD A/B1).

Progression to GOLD E

To examine progression from GOLD A/B to GOLD E, patients were followed from the index date while they remained on treatment until the end of the study period, discontinuation of health insurance coverage, discontinuation of treatment, or death, whichever occurred first. Progression to GOLD E was defined as experiencing ≥ 2 moderate exacerbations within a 12-month period or ≥ 1 severe exacerbation. For the occurrence of ≥ 2 moderate exacerbations, 12-month sliding windows post-index throughout follow-up were used, requiring that both moderate exacerbations occurred within the same 12-month window. The first moderate exacerbation could occur prior to the index date, as long as the second occurred after the index date, at which point the patient was considered to have progressed to GOLD E.

Statistical analyses

Patient clinical and demographic characteristics were summarized using means and standard deviations (SDs) for continuous variables and counts and proportions for categorical variables. Kaplan-Meier (KM) estimates were used to evaluate the proportion of patients who progressed from GOLD A/B to GOLD E, accounting for censoring due to treatment discontinuation, loss of insurance coverage, death, or end of the study window. Median (interquartile range [IQR]) was used to express time to reaching GOLD E; however, only medians are shown where < 75% of patients transition to GOLD E during follow-up, as it was not possible to estimate the upper value of the IQR in these instances.

To evaluate the association between baseline exacerbation status (0 vs. 1 moderation exacerbation) and progression from GOLD A/B to GOLD E, multivariable Cox proportional hazard models were used. Stepwise variable selection using backward elimination with an alpha level < 0.05 was applied to identify baseline demographic and clinical characteristics for inclusion in the final model that assessed the predictors of progression to GOLD E. Baseline exacerbation status served as the main effect, and the final model was adjusted for age, sex, race, insurance type, geographic region, Elixhauser comorbidities (modeled as a total score), and selected COPD-related comorbidities such as ischemic heart disease, pneumonia and tobacco use. Cox proportional hazard model results were reported as hazard ratios (HRs) with 95% confidence intervals (CIs).

All data processing and analyses were conducted within the Optum server environment using SQL for data extraction and cohort construction, and R for statistical analysis. Data were assumed to be missing at random, and patients with missing data were excluded from the Cox proportional hazard regression analyses.

Results

Patient disposition, demographics, and clinical characteristics

A total of 156,462 patients met the study inclusion criteria (Fig. 2). Of these patients, 112,324 (71.8%) did not have a moderate exacerbation during the baseline period (GOLD A/B0), and 44,138 (28.2%) had 1 moderate exacerbation during the baseline period (GOLD A/B1).

Fig. 2.

Patient flow for GOLD A/B analysis. COPD, chronic obstructive pulmonary disease; GOLD, Global Initiative for Chronic Obstructive Lung Disease; ICS, inhaled corticosteroids; LABA, long-acting beta-agonist; LAMA, long-acting muscarinic antagonist

Overall, 50.3% of patients were female, and most patients (73.8%) were White with a mean (SD) age of 70.6 (9.75) years (Table 1; demographics and clinical characteristics according to exacerbation history can be found in Supplemental Table 3). Most patients (85.2%) had Medicare Advantage insurance. The most common Elixhauser comorbidities during the 12-month baseline period included uncomplicated hypertension (75.4%), peripheral vascular disorders (30.6%), cardiac arrythmias (28.1%), uncomplicated diabetes (27.9%), depression (26.8%), and complicated diabetes (23.6%). The mean (SD) Elixhauser Index was 5.41 (3.16). Tobacco use was recorded in 63.8% of the population. The largest proportion of patients initiated LABA/ICS (46.6%), followed by LABA/LAMA (19.3%), LAMA (19.1%), LABA/LAMA/ICS triple therapy in a single inhaler (TTS; 12.2%), LABA/LAMA/ICS triple therapy in multiple inhalers (TTM; 2.0%), and LABA (0.9%).

Table 1.

Baseline demographic and clinical characteristics of patients classified as GOLD A/B with COPD by treatment type

	LABA​ (n = 1354)	LABA/ICS​ (n = 72,850)	LABA/LAMA​ (n = 30,256)	LAMA​ (n = 29,822)	TTM​ (n = 3090)	TTS​ (n = 19,090)	Overall (N = 156,462)
Demographics a
Age
Mean (SD)	73.5 (9.96)	70.3 (10.1)	70.7 (9.41)	71.0 (9.53)	70.4 (9.51)	71.0 (9.23)	70.6 (9.75)
Median [min, max]	74.0 [41.0, 90.0]	71.0 [40.0, 90.0]	71.0 [40.0, 90.0]	71.0 [40.0, 90.0]	70.0 [40.0, 90.0]	71.0 [40.0, 90.0]	71.0 [40.0, 90.0]
Age group, years
40–64	258 (19.1)	19,250 (26.4)	7300 (24.1)	7073 (23.7)	773 (25.0)	4243 (22.2)	38,897 (24.9)
65–74	454 (33.5)	28,068 (38.5)	12,190 (40.3)	11,994 (40.2)	1262 (40.8)	7979 (41.8)	61,947 (39.6)
75+	642 (47.4)	25,532 (35.0)	10,766 (35.6)	10,755 (36.1)	1055 (34.1)	6868 (36.0)	55,618 (35.5)
Sex
Female	717 (53.0)	38,016 (52.2)	14,512 (48.0)	14,953 (50.1)	1417 (45.9)	9038 (47.3)	78,653 (50.3)
Male	637 (47.0)	34,829 (47.8)	15,743 (52.0)	14,867 (49.9)	1673 (54.1)	10,051 (52.7)	77,800 (49.7)
Unknown	0 (0)	5 (0.0)	1 (0.0)	2 (0.0)	0 (0)	1 (0.0)	9 (0.0)
Race/ethnicity
White​	1052 (77.7)	52,605 (72.2)	22,854 (75.5)	22,487 (75.4)	2321 (75.1)	14,227 (74.5)	115,546 (73.8)
Asian​	14 (1.0)	1290 (1.8)	506 (1.7)	522 (1.8)	55 (1.8)	290 (1.5)	2677 (1.7)
African American​	121 (8.9)	8335 (11.4)	3455 (11.4)	3394 (11.4)	366 (11.8)	2105 (11.0)	17,776 (11.4)
Hispanic	95 (7.0)	6338 (8.7)	1892 (6.3)	1785 (6.0)	200 (6.5)	1190 (6.2)	11,500 (7.4)
Unknown​	72 (5.3)	4282 (5.9)	1549 (5.1)	1634 (5.5)	148 (4.8)	1278 (6.7)	8963 (5.7)
Type of insurance
Commercial​	142 (10.5)	11,459 (15.7)	4521 (14.9)	4404 (14.8)	435 (14.1)	2234 (11.7)	23,195 (14.8)
Medicare​	1212 (89.5)	61,391 (84.3)	25,735 (85.1)	25,418 (85.2)	2655 (85.9)	16,856 (88.3)	133,267 (85.2)
Region
Midwest	285 (21.0)	14,726 (20.2)	5958 (19.7)	6712 (22.5)	594 (19.2)	3321 (17.4)	31,596 (20.2)
Northeast	177 (13.1)	7816 (10.7)	3395 (11.2)	3854 (12.9)	447 (14.5)	1949 (10.2)	17,638 (11.3)
South	530 (39.1)	34,710 (47.6)	15,029 (49.7)	10,787 (36.2)	1331 (43.1)	10,730 (56.2)	73,117 (46.7)
West	359 (26.5)	15,425 (21.2)	5813 (19.2)	8395 (28.2)	715 (23.1)	3051 (16.0)	33,758 (21.6)
Other	2 (0.1)	115 (0.2)	42 (0.1)	51 (0.2)	2 (0.1)	35 (0.2)	247 (0.2)
Missing	1 (0.1)	58 (0.1)	19 (0.1)	23 (0.1)	1 (0.0)	4 (0.0)	106 (0.1)
Prior 12-month exacerbation history
GOLD A/B0	1041 (76.9)	50,399 (69.2)	22,717 (75.1)	22,573 (75.7)	2214 (71.7)	13,380 (70.1)	112,324 (71.8)
GOLD A/B1	313 (23.1)	22,451 (30.8)	7539 (24.9)	7249 (24.3)	876 (28.3)	5710 (29.9)	44,138 (28.2)
Elixhauser Comorbidity Index, mean (SD)	5.97 (3.51)	5.43 (3.21)	5.31 (3.04)	5.34 (3.16)	6.02 (3.54)	5.43 (3.09)	5.41 (3.16)
Select comorbidities b , n (%)
Hypertension, uncomplicated	1028 (75.9)	54,977 (75.5)	22,867 (75.6)	22,070 (74.0)	2372 (76.8)	14,641 (76.7)	117,955 (75.4)
Tobacco use everc	791 (58.4)	43,095 (59.2)	20,863 (69.0)	20,280 (68.0)	2156 (69.8)	12,582 (65.9)	99,767 (63.8)
Dyspnea	749 (55.3)	32,963 (45.2)	16,365 (54.1)	14,374 (48.2)	1723 (55.8)	10,080 (52.8)	76,254 (48.7)
Peripheral vascular disorders	446 (32.9)	21,695 (29.8)	9376 (31.0)	9260 (31.1)	1015 (32.8)	6096 (31.9)	47,888 (30.6)
Coronary artery disease	433 (32.0)	20,577 (28.2)	9824 (32.5)	9185 (30.8)	1069 (34.6)	6228 (32.6)	47,316 (30.2)
Cardiac arrhythmias	465 (34.3)	19,915 (27.3)	8374 (27.7)	8765 (29.4)	1068 (34.6)	5370 (28.1)	43,957 (28.1)
Diabetes, uncomplicated	386 (28.5)	21,001 (28.8)	8220 (27.2)	7853 (26.3)	958 (31.0)	5185 (27.2)	43,603 (27.9)
Depression	418 (30.9)	20,308 (27.9)	7469 (24.7)	7980 (26.8)	808 (26.1)	4941 (25.9)	41,924 (26.8)
Diabetes, complicated	330 (24.4)	17,871 (24.5)	6853 (22.7)	6509 (21.8)	790 (25.6)	4508 (23.6)	36,861 (23.6)
Hypertension, complicated	360 (26.6)	16,607 (22.8)	6682 (22.1)	6467 (21.7)	895 (29.0)	4518 (23.7)	35,529 (22.7)
Obesity	271 (20.0)	16,843 (23.1)	6625 (21.9)	5903 (19.8)	726 (23.5)	4420 (23.2)	34,788 (22.2)
Renal failure	341 (25.2)	15,846 (21.8)	6293 (20.8)	6106 (20.5)	734 (23.8)	4046 (21.2)	33,366 (21.3)
Congestive heart failure	374 (27.6)	14,854 (20.4)	6032 (19.9)	6089 (20.4)	934 (30.2)	4094 (21.4)	32,377 (20.7)
Hypothyroidism	332 (24.5)	15,391 (21.1)	5917 (19.6)	5770 (19.3)	578 (18.7)	3817 (20.0)	31,805 (20.3)
Fluid and electrolyte disorders	313 (23.1)	12,873 (17.7)	4849 (16.0)	5280 (17.7)	763 (24.7)	3043 (15.9)	27,121 (17.3)
Valvular disease	244 (18.0)	11,907 (16.3)	5466 (18.1)	5217 (17.5)	638 (20.6)	3257 (17.1)	26,729 (17.1)
Ischemic heart disease	235 (17.4)	11,649 (16.0)	5267 (17.4)	4981 (16.7)	663 (21.5)	3199 (16.8)	25,994 (16.6)
Obstructive sleep apnea	218 (16.1)	11,273 (15.5)	5562 (18.4)	4628 (15.5)	574 (18.6)	3452 (18.1)	25,707 (16.4)
Pneumonia	238 (17.6)	8901 (12.2)	2938 (9.7)	3171 (10.6)	574 (18.6)	2039 (10.7)	17,861 (11.4)

COPD, chronic obstructive pulmonary disease; ICS, inhaled corticosteroids; LABA, long-acting beta-agonist, LAMA, long-acting muscarinic antagonist; TTM, triple therapy in multiple inhalers; TTS, triple therapy in a single inhaler

^aData are n (%), unless otherwise stated

^bSelect comorbidities include Elixhauser conditions and non-Elixhauser conditions with a prevalence of 10% or higher

^cTobacco use data are only presented on a dichotomous scale, as current smoking status and pack-year data are not available

Progression to GOLD E overall, by exacerbation history, and by index inhaled maintenance treatment

Of the 156,462 patients classified as GOLD A/B, 45,079 patients progressed to GOLD E over the entire observation period of up to 7 years after initiating inhaled maintenance therapy (mean [SD] follow-up on treatment was 232 [359] days). Accounting for censoring, 23.0%, 35.4%, and 57.8% of the patients at risk progressed to GOLD E over 1, 2, and 5 years post-index, respectively (Fig. 3a). The median time to GOLD E was 1363 days (3.73 years).

Fig. 3.

Progression to GOLD E following treatment initiation^a (a) overall and (b) based on exacerbation history.^b GOLD, Global Initiative for Chronic Obstructive Lung Disease. ^aPatients were censored when they stopped treatment, discontinued enrollment, died, or reached the end of the study window (June 30, 2023). ^bP values are based on log-rank tests and indicate statistical difference in GOLD E rates between patients with vs. those without moderate exacerbations during their baseline period

Progression to GOLD E according to exacerbation history is shown in Fig. 3b. Of the 45,079 patients who progressed to GOLD E, 26,644 progressed from GOLD A/B0 and 18,435 progressed from GOLD A/B1. In the GOLD A/B0 group, 16.6%, 29.6%, and 52.9% of patients progressed to GOLD E over 1, 2, and 5 years, respectively, and the median time to GOLD E was 1655 days. In the GOLD A/B1 group, 41.1%, 51.6%, and 71.4% of patients progressed to GOLD E over 1, 2, and 5 years, respectively. The median (IQR) time to GOLD E among patients in the GOLD A/B1 group was 679 (154–2149) days. The risk of progressing to GOLD E was approximately 3 times higher in the GOLD A/B1 vs. GOLD A/B0 group (hazard ratio [HR] 2.92; 95% CI 2.84–3.00; P < 0.001).

Pathways of progression from GOLD A/B0 and GOLD A/B1 to GOLD E are shown in Fig. 4. For the patients classified as GOLD A/B0 who progressed to GOLD E, 56.7% progressed by having 1 severe exacerbation, while 43.3% progressed by having 2 moderate exacerbations. For the patients classified as GOLD A/B1 who progressed to GOLD E, the majority (71.6%) progressed to GOLD E by having another moderate exacerbation within 12 months of the first moderate exacerbation and 28.4% progressed by having 1 severe exacerbation.

Fig. 4.

Patient flow for progression from GOLD A/B to GOLD E. GOLD, Global Initiative for Chronic Obstructive Lung Disease

GOLD A/B to E progression according to treatment is shown in Fig. 5. Among patients who progressed to GOLD E, the median (IQR) time to reach GOLD E was 851 (269-2088) days in patients on TTM therapy; 1101 days in those on TTS therapy, and 1336 days in patients on LABA/ICS therapy. Patients on LABA/LAMA and LAMA therapies reached GOLD E at a median of 1490 and 1462 days, respectively. Patients on LABA therapy reached GOLD E at a median of 1328 days.

Fig. 5.

GOLD A/B to E progression by treatment. GOLD, Global Initiative for Chronic Obstructive Lung Disease; ICS, inhaled corticosteroids; LABA, long-acting beta-agonist, LAMA, long-acting muscarinic antagonist; TTM, triple therapy in multiple inhalers; TTS, triple therapy in a single inhaler

Predictors of progressing to GOLD E

Predictors of progressing to GOLD E are shown in Table 2. In the overall population, the strongest predictor was a history of having 1 moderate exacerbation vs. no moderate exacerbations during the baseline period (HR [95% CI]: 2.85 [2.77–2.93]; P < 0.001). Patients who initiated ICS-containing treatments vs. those who did not were slightly more likely to progress to GOLD E (HR [95% CI]: 1.07 [1.04–1.10]; P < 0.001). Moreover, patients aged ≥ 75 years were more likely than younger patients (aged 40–64 years) to progress to GOLD E (HR [95% CI]: 1.12 [1.07–1.17]; P < 0.001) and having Medicare vs. commercial insurance was a significant predictor (HR [95% CI]: 1.17 [1.11–1.23]; P < 0.001). An increased number of Elixhauser comorbidities was also a predictor of progressing to GOLD E (HR [95% CI]: 1.05 [1.05–1.06]; P < 0.001), as were a history of pneumonia (HR [95% CI]: 1.29 [1.24–1.34]; P < 0.001) and tobacco use (HR [95% CI]: 1.10 [1.07–1.14]; P < 0.001). The baseline demographics predictive of progressing to GOLD E were generally consistent across treatment types, except for race. Hispanic and Asian patients had lower rates of progression in the overall sample, but results were mixed when categorized by treatment.

Table 2.

Predictors of progressing to GOLD E^{a, b}

	LABA (n = 1351) Number of events = 197		LABA/ICS (n = 72,672) Number of events = 8745		LABA/LAMA (n = 30,194) Number of events = 4209		LAMA (n = 29,746) Number of events = 3858		TTM (n = 3087) Number of events = 604		TTS (n = 19,050) Number of events = 2450		Overallb (N = 156,100) Number of events = 20,063
	HR (CI)	P-value	HR (CI)	P-value	HR (CI)	P-value	HR (CI)	P-value	HR (CI)	P-value	HR (CI)	P-value	HR (CI)	P-value
Treatment group
Without ICS (reference)
With ICS													1.07 (1.04–1.1)	< 0.001
Exacerbation group
GOLD A/B0 (reference)
GOLD A/B1	2.94 (2.20–3.94)	< 0.001	2.78 (2.66–2.90)	< 0.001	2.92 (2.74–3.10)	< 0.001	3.01 (2.82–3.21)	< 0.001	2.19 (1.86–2.59)	< 0.001	3.03 (2.80–3.28)	< 0.001	2.85 (2.77–2.93)	< 0.001
Sex
Female (reference)
Male	1.19 (0.90–1.58)	0.229	0.99 (0.95–1.03)	0.679	0.94 (0.88-1)	0.035	0.92 (0.86–0.98)	0.012	0.97 (0.82–1.14)	0.682	0.93 (0.86–1.01)	0.068	0.96 (0.94–0.99)	0.006
Race
White (reference)
Asian	0.92 (0.13–6.71)	0.937	0.61 (0.49–0.76)	< 0.001	0.80 (0.61–1.06)	0.123	0.91 (0.69–1.19)	0.488	1.29 (0.74–2.25)	0.377	0.96 (0.66–1.39)	0.824	0.77 (0.67–0.87)	< 0.001
African American	1.63 (1.03–2.58)	0.037	1.06 (0.99–1.13)	0.093	1.05 (0.95–1.16)	0.323	0.90 (0.81–1.01)	0.068	0.99 (0.76–1.29)	0.924	0.92 (0.80–1.05)	0.213	1.01 (0.97–1.06)	0.580
Hispanic	0.66 (0.33–1.30)	0.230	0.82 (0.75–0.90)	< 0.001	0.81 (0.70–0.94)	0.005	0.93 (0.8–1.07)	0.310	0.93 (0.63–1.35)	0.688	0.69 (0.56–0.85)	0.001	0.82 (0.77–0.87)	< 0.001
Unknown	1.17 (0.63–2.18)	0.611	1.03 (0.94–1.13)	0.553	1.04 (0.89–1.2)	0.643	1.06 (0.92–1.23)	0.418	1.23 (0.83–1.82)	0.306	1.03 (0.87–1.22)	0.710	1.04 (0.97–1.11)	0.251
Insurance
Commercial (reference)
Medicare	0.87 (0.48–1.55)	0.628	1.21 (1.12–1.30)	< 0.001	1.06 (0.95–1.18)	0.295	1.23 (1.09–1.38)	< 0.001	1.08 (0.81–1.43)	0.615	1.16 (0.99–1.35)	0.068	1.17 (1.11–1.23)	< 0.001
Region
Midwest (reference)
Northeast	1.19 (0.72–1.96)	0.508	0.99 (0.92–1.07)	0.823	0.92 (0.82–1.03)	0.150	0.90 (0.81–1.01)	0.067	0.91 (0.68–1.20)	0.486	0.88 (0.75–1.03)	0.112	0.95 (0.90-1.00)	0.039
South	1.00 (0.67–1.51)	0.993	1.00 (0.95–1.06)	0.968	1.02 (0.94–1.10)	0.678	1.03 (0.95–1.12)	0.520	1.19 (0.96–1.48)	0.119	1.01 (0.90–1.12)	0.918	1.01 (0.98–1.05)	0.518
West	1.06 (0.69–1.63)	0.798	0.91 (0.85–0.97)	0.003	0.89 (0.81–0.98)	0.018	0.86 (0.79–0.94)	0.001	0.85 (0.66–1.09)	0.202	0.87 (0.76-1.00)	0.052	0.89 (0.86–0.93)	< 0.001
Age group
40–64 (reference)
65–74	1.44 (0.88–2.36)	0.143	0.99 (0.93–1.05)	0.732	1.03 (0.94–1.14)	0.471	0.97 (0.88–1.07)	0.595	1.08 (0.85–1.37)	0.543	0.98 (0.87–1.11)	0.790	1.00 (0.96–1.04)	0.980
75+	1.61 (0.99–2.61)	0.055	1.09 (1.02–1.17)	0.007	1.10 (0.99–1.21)	0.064	1.09 (0.99–1.20)	0.094	1.49 (1.16–1.90)	0.002	1.18 (1.04–1.33)	0.010	1.12 (1.07–1.17)	< 0.001
Number of Elixhauser comorbidities c	1.06 (1.02–1.11)	0.006	1.05 (1.05–1.06)	< 0.001	1.05 (1.04–1.06)	< 0.001	1.05 (1.04–1.06)	< 0.001	1.04 (1.01–1.06)	0.004	1.05 (1.04–1.06)	< 0.001	1.05 (1.05–1.06)	< 0.001
Other comorbidities d
Pneumonia	1.00 (0.69–1.44)	0.988	1.34 (1.26–1.42)	< 0.001	1.24 (1.13–1.36)	< 0.001	1.26 (1.15–1.38)	< 0.001	1.40 (1.14–1.71)	0.001	1.17 (1.04–1.32)	0.008	1.29 (1.24–1.34)	< 0.001
Tobacco use	1.02 (0.76–1.37)	0.875	1.16 (1.11–1.21)	< 0.001	1.03 (0.97–1.11)	0.329	1.07 (0.99–1.14)	0.070	1.02 (0.85–1.21)	0.865	1.07 (0.98–1.17)	0.133	1.10 (1.07–1.14)	< 0.001

CI, confidence interval; GOLD, Global Initiative for Chronic Obstructive Lung Disease; HR, hazard ratio; ICS, inhaled corticosteroids; LABA, long-acting beta-agonist, LAMA, long-acting muscarinic antagonist; TTM, triple therapy in multiple inhalers; TTS, triple therapy in a single inhaler

^aUsing Cox proportional hazards model with stepwise variable selection

^bEffective sample size after excluding patients with missing covariate values

^cElixhauser comorbidities are modelled as a score, rather than individually

^dOther comorbidities are not included in the Elixhauser Comorbidity Index

Discussion

These data offer a new perspective on individuals with COPD that persist as GOLD A/B without progression to GOLD E. To the best of our knowledge, no other claims-based studies have been identified that evaluate and characterize the eventual progression of these patients to GOLD E. Studies outside of the US have utilized the GOLD categories to assess clinical outcomes but primarily focus on the exacerbation burden experienced by patients, not necessarily their journey from low to high exacerbator status [13]. Taken together with the current literature, this analysis’s evidence of eventual progression to frequent/severe exacerbations could guide clinicians to proceed with early exacerbation prevention for patients with COPD.

This real-world database analysis of patients with COPD demonstrates high rates of progression to a frequent or severe exacerbator phenotype, despite having initiated inhaled maintenance therapy. The majority of patients progressed to a frequent or severe exacerbator phenotype over 5 years. Patients with a history of 1 moderate exacerbation (GOLD A/B1) were nearly 3 times more likely to progress to GOLD E compared with those without a history of moderate exacerbations (GOLD A/B0). This finding is supported by existing literature on predictors of exacerbations [7, 14, 15]. Importantly, this analysis demonstrated that patients without an exacerbation history also had high rates (approximately 53% over 5 years) of progression to GOLD E. As exacerbations are associated with disease progression, mortality, and increased cardiovascular risk [1, 4–6], prevention of these sentinel events should be a treatment goal for every patient, regardless of exacerbation history.

In the overall population, key predictors of progressing to GOLD E other than exacerbation history included initiation of ICS-containing treatments, the presence of multiple comorbidities at baseline, older age, and having Medicare insurance. The group of patients initiating ICS-containing treatments may capture the specific patients at risk for exacerbating (confounding by indication), but due to the nature of claims data, other risk factors for exacerbations are not identifiable. Existing literature supports the presence of multiple comorbidities and older age as a risk factor for future exacerbations [7, 16], which is confirmed by this analysis. Moreover, in a systematic review of studies related to COPD and its comorbidities, the most prevalent comorbidities were hypertension (17–64.7%), coronary artery disease (19.9–47.8%), and diabetes mellitus (10.2–45%) [17], which is in line with the results of this claims-based study (hypertension, 22.7–75.4%; coronary artery disease, 30.2%; diabetes, 23.6–27.9%). Many of the published studies indicated that such comorbid conditions, especially that of cardiovascular disease, contribute to increased mortality, increased frequency and severity of exacerbations, and/or decreased quality of life compared with patients with COPD who did not have these comorbidities [17]. In particular, our study highlights a stronger effect of pneumonia and congestive heart failure relative to other comorbidities on the risk of transitioning to GOLD E. This knowledge could help tailor individual treatment needs that may warrant a multidisciplinary approach, such as that for both cardiac and pulmonary care, with additional consideration of cardiac side effects associated with COPD treatment.

Interestingly, in this database study, almost 15% of patients classified as GOLD A/B initiated inhaled maintenance treatment with triple therapy, which is not recommended by the current GOLD Report [1]. The choice to use triple therapy in a patient classified as GOLD A/B may be explained by the convenience of maximum inhaled therapy in one device and lack of treatment options aimed at preventing disease progression. Despite the fact that ICS therapy is used to decrease the risk of future exacerbations [1], our data suggest that ICS-containing treatment is a predictor of progression.

Our analysis supports that despite the availability of numerous treatments, the progressive nature of COPD dictates that new treatment options for COPD to prevent exacerbations and/or reduce the severity of exacerbations are needed. The US Food and Drug Administration recently approved two novel therapies for COPD in 2024, ensifentrine and dupilumab [18, 19]. Ensifentrine, a selective dual inhibitor of phosphodiesterase (PDE3) and PDE4, was approved for the maintenance treatment of COPD [18]. In a recently published analysis of the pooled ENHANCE-1 and ENHANCE-2 studies, patients treated with ensifentrine experienced a 41% reduction in the rate of moderate or severe exacerbations (rate ratio [RR] [95% CI]: 0.59 [0.43–0.80]; P < 0.001), as well as a delayed transition from GOLD B to GOLD E vs. placebo [20]. Importantly, these studies were conducted in patients without regard to exacerbation history (approximately 20% of patients had a history of an exacerbation in the 15 months prior to study enrollment). Dupilumab, a monoclonal antibody and interleukin-4 receptor alpha antagonist, was approved as add-on maintenance treatment for adult patients with inadequately controlled COPD and an eosinophilic phenotype [19]. In the BOREAS and NOTUS trials, dupilumab resulted in a reduced rate of moderate/severe exacerbations over 52 weeks (BOREAS: 30% reduction; RR [95% CI]: 0.70 [0.58–0.86]; P < 0.001; NOTUS: 34% reduction; RR [95% CI]: 0.66 [0.54–0.82]; P < 0.001) [21, 22].

There are a number of strengths in this study. A primary strength of this study was its reliance on a geographically diverse sample of the US population, which increases the generalizability of our findings. This analysis also used current real-world data on patients with COPD and captures a wide range of inhaled treatments for COPD. Additionally, to our knowledge, this is the first analysis in the US that utilizes GOLD categories to evaluate progression, which allows for interpretation by clinicians who are familiar with GOLD categories. The reliance on closed claims data, along with the requirement of continuous insurance eligibility, increases our confidence that all relevant care covered by patients’ healthcare plans was captured.

There are several limitations associated with the use of claims data. Since symptom data such as cough, sputum, and dyspnea are either not available or are often under-coded, delineation between GOLD A and GOLD B categories to further elucidate risk of progression to GOLD E is not possible. Eosinophil count is also not available, which may inform the decision to include ICS in the maintenance therapy [1]. Claims data also lack important clinical measures such as lung function (e.g., forced expiratory volume in one second [FEV1] from spirometry), which is an important predictor of disease progression. Future analyses should explore the impact of clinical features such as symptoms, eosinophil count, and pulmonary function on rates of and progression to GOLD E. Moreover, data confirming a filled prescription does not necessarily confirm that the medication was taken as prescribed, limiting interpretation of treatment initiation or duration. Additionally, to be included in the database, patients had to have insurance coverage and any healthcare not covered by patients’ insurance was not captured in the analyses. Furthermore, as this analysis encompassed data from January 2016 to June 2023, the results are not necessarily reflective of more recent years or post–June 2023 treatment patterns. Lastly, multivariable regression models estimated the independent effect of each predictor; however, as in all observational studies, unobservable confounding and measurement errors remain possible.

Conclusions

In this analysis, the majority of patients classified as GOLD A/B progressed to GOLD E within 5 years, despite initiation of inhaled maintenance therapies. This suggests that patients with COPD are at high risk for exacerbations, regardless of exacerbation history, and that patients with GOLD A/B COPD should be closely monitored to avoid sentinel exacerbation events. New therapies are needed that may further alleviate the burden of COPD on patients, their families, and the US healthcare system.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

CDM

Clinformatics^® DataMart

COPD

chronic obstructive pulmonary disease

ED

emergency department

GOLD

Global Initiative for Chronic Obstructive Lung Disease

HR

hazard ratio

ICD-10

International Classification of Diseases, Tenth Revision

ICS

inhaled corticosteroid

IQR

interquartile range

LABA

long-acting beta-agonist

LAMA

long-acting muscarinic antagonist

PDE3

phosphodiesterase 3

PDE4

phosphodiesterase 4

RR

rate ratio

TTS

triple therapy in a single inhaler

TTM

triple therapy in multiple inhalers

Author contributions

TS contributed to the conceptualization, methodology, interpretation of results, and the critical review and editing of the manuscript. SS contributed to the interpretation of results and critical review and editing of the manuscript. EW contributed to the review and interpretation of results and the drafting and critical review of the manuscript. CL, KA, and AD contributed to the study design, interpretation of results, and critical review and editing of the manuscript. YP conducted the statistical analyses and critically reviewed the manuscript. VT contributed to the study design, statistical methodology, interpretation of results, and critical review and editing of the manuscript.

Funding

This study was funded by Verona Pharma plc. The authors have received no other financial and/or material support for this research or the creation of this work apart from that disclosed.

Data availability

Data from Optum Clinformatics^® Data Mart database are outside of Verona Pharma’s data sharing policy and are unavailable for sharing.

Declarations

Ethical approval and consent to participate

Only deidentified patient data from Optum Clinformatics^® Data Mart database were used in this study and may be considered exempt. The study was not submitted to an Institutional Review Board approval for review.

Consent for publication

Not applicable.

Clinical trial number

Not applicable.

Competing interests

TS reports research grants from AstraZeneca, Sanofi, GlaxoSmithKline and receives personal fees from Verona Pharma, GlaxoSmithKline, and Apogee Therapeutics for advisory work. SS has received research grants (to his institution) from Chiesi, AstraZeneca, and Sanofi-Regeneron; has received royalties from Wolters Kluver Health; and has received consulting fees or honoraria from AstraZeneca, Boehringer Ingelheim, Chiesi, GlaxoSmithKline, Nuvaira, and Pulmonx. EW has served on a scientific advisory board for Verona Pharma; EW also receives research grant funding from the U.S. Department of Veterans Affairs (outside the current work). CL, KA, and AD are employees of and may own stock/stock options in Verona Pharma plc. YP is an employee of Stratevi, a research consulting firm that received funding from Verona Pharma. VT reports grant funding from the National Institutes of Health/National Heart, Lung, and Blood Institute, Grifols and Fisher & Paykel Healthcare, as well as consulting fees from ThermoFisher and Sanofi/Regeneron all outside the scope of this work.