Abstract
Objectives
To assess real-world survival and healthcare resource utilization (HCRU) in US patients with non-cystic fibrosis bronchiectasis (NCFBE).
Methods
This retrospective analysis, using data from the STATinMED RWD Insights database from Jan 2015–Oct 2022, included adults with NCFBE (from Jan 2015–Oct 2021) and non-NCFBE comparators (from Jan 2015–Aug 2020); baseline characteristics were balanced by inverse probability treatment weighting. Outcomes included survival through end of study. HCRU was assessed over 12 months.
Results
117,718 patients with NCFBE and 306,678 comparators were included. Patients with NCFBE had a 77% higher risk of death than comparators (hazard ratio [HR] 1.77 [95% CI 1.74-1.80]). Risk of death was higher among patients aged ≥65 years (vs 18–34 years; HR 11.03 [95% CI 10.36-11.74]), among Black patients (vs White; HR 1.53 [95% CI 1.50-1.55]), and among patients with comorbid COPD (HR 1.42 [95% CI 1.40-1.44]). Patients with NCFBE incurred higher all-cause and respiratory-related HCRU than comparators for outpatient office, outpatient hospital, emergency department (ED), inpatient and respiratory-related pulmonologist visits (all p < .0001); HCRU increased with exacerbations.
Conclusions
Patients with NCFBE have high mortality burden and incur high HCRU, both of which are further increased with exacerbations. Prevention and delay of exacerbations are key areas for improvement of disease management.
Keywords: Bronchiectasis, exacerbation, survival, healthcare resource utilization
Introduction
Non-cystic fibrosis bronchiectasis (NCFBE) is a heterogeneous chronic respiratory disease characterized by irreversible dilation of the bronchi, recurrent infection, and chronic airway inflammation, which causes a progressive worsening of disease and substantial impact on patient health-related quality of life.1,2 The estimated prevalence of NCFBE in the US ranges from 139 to 213 per 100,000 adults, and prevalence rates per 100,000 adults have risen by approximately 8% per year since 2001.3–5
Patients with NCFBE are at risk of recurrent exacerbations, which are associated with progressive lung damage and significant morbidity and mortality.6,7 Bronchiectasis exacerbations are commonly defined as a deterioration in ≥3 specified symptoms for ≥48 hours causing a change in treatment. 8 Patients with a history of multiple exacerbations are at higher risk for future exacerbations and likely to have lower quality of life, more hospital admissions, and poorer prognosis than those with fewer exacerbations.7,9,10
NCFBE treatment goals are to reduce symptoms, improve quality of life, prevent exacerbations and hospital admissions, preserve lung function, and reduce mortality.3,11,12 Current guidelines recommend management strategies that are focused on alleviating disease symptoms and treatment of infections 13 ; however, these recommendations may not all be supported by strong evidence.11,12 Current management approaches focus on three main components: treatment of underlying conditions, airway clearance, and short- or long-term antimicrobial treatment. 13
There are limited real-world data on mortality in patients with NCFBE, particularly in the US. The available studies were performed in small populations outside of the US and report varying survival rates but generally suggest poor prognosis. 14 Reported 4-year survival ranges from 58% to 91% and declines considerably over time, with factors including advanced age, Pseudomonas aeruginosa infection, comorbidities, lung function, body mass index, hospitalizations, and exacerbations impacting survival.14–16 Additionally, despite NCFBE’s significant economic burden,12,17,18 which is largely driven by hospitalization costs related to exacerbations, 19 there is a paucity of real-world studies evaluating disease-related healthcare resource utilization (HCRU). Understanding the disease burden and HCRU associated with NCFBE would contribute toward improving disease management and optimizing healthcare resource allocation.
This real-world study aimed to assess survival, disease burden, and HCRU in patients with NCFBE, compared with patients without NCFBE, from a large US claims database.
Methods
Study design and data source
This was a retrospective, longitudinal, observational study using deidentified data from the STATinMED Real-World Data Insights database from January 1, 2014, to October 31, 2022 (Figure 1). The STATinMED Real-World Data Insights database is an all-payer medical and pharmacy claims source with commercial, Medicaid, and Medicare health plan data that covers nearly 80% of the US healthcare system (300 million unique patients) and includes mortality data from the Death Master Index that has records for approximately 90% of all US deaths. 20 Data were deidentified, and data use complied with the Health Insurance Portability and Accountability Act for the privacy and security of protected health information. 21
Figure 1.
Study design. The identification periods differed between cases and comparators as the comparator cohort was previously established while the case cohort was able to utilize more recent data. The case and comparator cohorts were matched using IPTW, with a ratio of 1:2.5 patients with NCFBE to comparators, to account for potential bias. The index date was the earliest claim used to identify each patient with NCFBE during the case-identification period or a randomly assigned date for comparator patients during the comparator-identification period. IPTW, inverse probability treatment weighting; NCFBE, non-cystic fibrosis bronchiectasis.
Study population
Included in this study were adult patients (≥18 years) with a NCFBE diagnosis defined as ≥2 outpatient claims with a diagnosis code for NCFBE dated ≥30 days apart, or ≥1 inpatient claim with NCFBE as the primary diagnosis, or ≥1 high-resolution chest computerized tomography scan followed by ≥1 inpatient or ≥2 outpatient claims for NCFBE. 22 Patients were required to have ≥12 months of continuous data capture prior to the index date (baseline period) and after the index date (follow-up period); data capture included both medical and pharmacy benefits. Excluded were patients with ≥1 inpatient or ≥2 outpatient claims ≥30 days apart with a diagnosis code for cystic fibrosis, interstitial fibrosis, pulmonary fibrosis, or sarcoidosis during the baseline period. Patients without a NCFBE diagnosis who met all other inclusion and exclusion criteria (apart from the NCFBE diagnosis criteria) were included as comparators. The index date was the earliest claim used to identify each patient with NCFBE during the case-identification period (January 1, 2015, to October 31, 2021) or a randomly assigned date for non-NCFBE comparators during the comparator-identification period (January 1, 2015, to August 30, 2020).
Inverse probability treatment weighting
Patients with NCFBE were compared with non-NCFBE comparators using IPTW to balance patient characteristics. The propensity score was calculated using a logistic model with patients with NCFBE and comparators included in the model, using the comparator patients as the reference. The baseline characteristics controlled for in the IPTW model were based on clinical rationale and standardized differences (Supplement Table S1). The propensity score acts as a balancing score between patients with NCFBE and comparators. After calculating the propensity score, the distribution of the propensity scores was reviewed. Patients with NCFBE were assigned a weight of 1/(propensity score) and comparators were assigned a weight of 1/(1-propensity score). A ratio of 1:2.5 patients with NCFBE to comparators was used for the IPTW weighting analysis. For patients with a very low propensity score, a very large weight was generated, which can increase the variability of the estimated effect. To address this issue, the weights were stabilized by multiplying the weights for both patients with NCFBE and comparators by a constant, equal to the mean of propensity score for each group (expected value of being in the NCFBE or comparator groups, respectively). This reduced both the variability of the weights and the variance of the effect estimates.
Study endpoints
Demographic characteristics were assessed at index, and clinical characteristics were assessed during the baseline period. Prespecified comorbidities were identified using ICD-9/10-CM codes and selected based on clinician input.
Risk of death in patients with NCFBE was compared with non-NCFBE comparators through end of the study period to minimize immortal time bias. The probability of survival in patients who were alive at the end of the 12-month follow-up period was assessed at 1, 3, and 5 years post follow-up.
Bronchiectasis exacerbations were assessed over the baseline, 12-month, and 24-month follow-up periods. Bronchiectasis exacerbations were those that led to either (1) an inpatient claim with a primary/admitting diagnosis of NCFBE or an outpatient claim with any intravenous (IV) antibiotic procedure code or (2) any pharmacy claim for an oral antibiotic filled for up to a 14-day supply. Oral antibiotics included quinolone (ciprofloxacin), amoxicillin, amoxicillin and clavulanate, sulfamethoxazole and trimethoprim, cephalosporins (including ceftriaxone), tetracycline (doxycycline), and clarithromycin. Azithromycin was not included to minimize bias from capturing its use as long-term anti-inflammatory therapy. Risk factors for exacerbations, by severity and frequency of exacerbations, were reported.
All-cause, respiratory-related, and exacerbation-related HCRU (hospital and ED. visits and length of stay) in the follow-up period were assessed for patients with NCFBE and comparators, patients with NCFBE with or without exacerbations, and patients with NCFBE with exacerbations, by frequency of exacerbations. Disease-specific HCRU was based on the existence of NCFBE - or exacerbation-specific codes available in the claims.
Statistical analyses
Between-group differences were evaluated using t tests for continuous variables, chi-square tests for categorical variables, and Wilcoxon tests as appropriate. Time-to-death and mortality risk were examined using Cox proportional hazards regression models weighted with the propensity scores to adjust for confounders. The proportional hazards assumption was evaluated by visually inspecting the Kaplan-Meier curves by month among the NCFBE cases and comparators. A landmark analysis was conducted for patients surviving within the study (alive at 12 months postindex) to predict the probability of survival at 1, 3, and 5 years beyond the end of the follow-up period.
A logistic regression with a backward stepwise selection at a 0.05 significance level was used to analyze the relationship between NCFBE cases and exacerbations, controlling for all variables in the IPTW. Ordinal logistic regression was used to examine the odds of a NCFBE case having ≥1 exacerbation event.
Generalized linear models (GLMs) were used to examine all-cause, respiratory-related, and exacerbation-related HCRU. For length of stay for inpatient hospitalizations, a GLM with negative binomial distribution was used and log link function was applied. For a number of visits, a GLM with Poisson distribution with a log link was used. The least square means were reported for NCFBE cases and comparators.
Results
Patient population
A total of 117,718 patients with NCFBE and 306,678 non-NCFBE comparators were included in the study (Supplement Figure S1). Sample sizes were 355,353 in the NCFBE cohort and 833,167 in the comparator cohort after IPTW weighting to balance patient characteristics. Patients with NCFBE were older than comparators (68.9 vs 57.8 years) and more likely to have disease-associated comorbidities or symptoms (Quan-CCI [Charlson Comorbidity Index] score 3.3 vs 1.1) (Table 1; Supplement Table S2 for standardized differences). The most common baseline comorbidities or symptoms in the NCFBE group were hypertension (55.0%), cough (44.4%), cardiovascular disease (40.9%), COPD (34.5%) and gastroesophageal reflux disease (30.1%). More patients with NCFBE had anxiety (15.8% vs 9.9%) and depression (8.9% vs 4.3%). After IPTW weighting, mean age of patients with NCFBE was 63.1 years vs 65.2 years for comparators, while Quan-CCI score was 2.3 for patients with NCFBE vs 4.7 for comparators.
Table 1.
Patient sociodemographic and clinical characteristics at baseline, unadjusted and after IPTW weighting.
| Baseline | Unadjusted | After IPTW weighting a | |||
|---|---|---|---|---|---|
| NCFBE cohort (N = 117,718) | Non-NCFBE comparator cohort (N = 306,678) | NCFBE cohort (N = 355,353) | Non-NCFBE comparator cohort (N = 833,167) | ||
| Age, mean (SD) | 68.9 (11.5) | 57.8 (17.5) | 63.1 (27.9) | 65.2 (24.7) | |
| Age, n (%) | 18–34 years | 2312 (2.0) | 43,139 (14.1) | 31,506 (8.9) | 53,478 (6.4) |
| 35–54 years | 9790 (8.3) | 73,676 (24.0) | 61,105 (17.2) | 105,631 (12.7) | |
| 55–64 years | 20,776 (17.6) | 60,185 (19.6) | 66,982 (18.8) | 177,198 (21.3) | |
| ≤64 years | 32,878 (27.9) | 177,000 (57.7) | 159,592 (44.9) | 336,307 (40.4) | |
| ≥65 years | 84,840 (72.1) | 129,678 (42.3) | 195,761 (55.1) | 496,860 (59.6) | |
| Female sex, n (%) | 71,856 (61.0) | 180,395 (58.8) | 201,570 (56.7) | 417,834 (50.2) | |
| Race, n (%) | White | 67,850 (57.6) | 170,263 (55.5) | 193,099 (54.3) | 437,058 (52.5) |
| Black | 7598 (6.5) | 21,965 (7.2) | 26,369 (7.4) | 68,422 (8.2) | |
| Asian | 2532 (2.2) | 5749 (1.9) | 6812 (1.9) | 14,188 (1.7) | |
| Other | 409 (0.3) | 1042 (0.3) | 1379 (0.4) | 1795 (0.2) | |
| Unknown | 39,329 (33.4) | 107,659 (35.1) | 127,694 (35.9) | 311,705 (37.4) | |
| Health insurance, n (%) | Commercial | 22,241 (18.9) | 160,947 (52.5) | 132,902 (37.4) | 226,482 (27.2) |
| Medicare | 78,377 (66.6) | 98,351 (32.1) | 163,594 (46.0) | 442,209 (53.1) | |
| Medicaid | 16,504 (14.0) | 38,299 (12.5) | 52,496 (14.8) | 154,443 (18.5) | |
| Other/unknown | 596 (0.5) | 9081 (3.0) | 6362 (1.8) | 10,034 (1.2) | |
| Geographic region, n (%) | Northeast | 22,171 (18.8) | 63,156 (20.6) | 73,132 (20.6) | 174,848 (21.0) |
| North central | 30,666 (26.1) | 67,755 (22.1) | 81,964 (23.1) | 172,811 (20.7) | |
| South | 42,712 (36.3) | 108,654 (35.4) | 129,096 (36.3) | 346,692 (41.6) | |
| West | 22,137 (18.8) | 66,683 (21.7) | 70,608 (19.9) | 138,357 (16.6) | |
| Other | 32 (0.0) | 430 (0.1) | 553 (0.2) | 459 (0.1) | |
| Quan-CCI score, mean (SD) | 3.3 (2.9) | 1.1 (1.8) | 2.3 (3.9) | 4.7 (8.0) | |
| Quan-CCI comorbidities, n (%) | AIDS | 1038 (0.9) | 1387 (0.5) | 1685 (0.5) | 18,835 (2.3) |
| Cancer | 26,355 (22.4) | 22,509 (7.3) | 53,416 (15.0) | 183,453 (22.0) | |
| Cerebrovascular disease | 10,354 (8.8) | 10,945 (3.6) | 19,180 (5.4) | 156,564 (18.8) | |
| COPD | 40,557 (34.5) | 7522 (2.5) | 49,656 (14.0) | 391,582 (47.0) | |
| CHF | 17,210 (14.6) | 11,875 (3.9) | 34,272 (9.6) | 186,471 (22.4) | |
| Dementia | 2539 (2.2) | 4331 (1.4) | 5148 (1.4) | 45,784 (5.5) | |
| Diabetes | 31,211 (26.5) | 68,896 (22.5) | 63,689 (17.9) | 321,112 (38.5) | |
| Hemiplegia or paraplegia | 1184 (1.0) | 1868 (0.6) | 2591 (0.7) | 20,819 (2.5) | |
| Liver disease | 10,527 (8.9) | 8585 (2.8) | 22,744 (6.4) | 124,168 (14.9) | |
| Metastatic carcinoma | 5271 (4.5) | 1754 (0.6) | 6606 (1.9) | 73,869 (8.9) | |
| Myocardial infarction | 581 (0.5) | 498 (0.2) | 801 (0.2) | 9648 (1.2) | |
| Peptic ulcer disease | 1988 (1.7) | 1796 (0.6) | 3643 (1.0) | 41,820 (5.0) | |
| PVD | 19,472 (16.5) | 13,188 (4.3) | 38,131 (10.7) | 164,847 (19.8) | |
| Renal disease | 3492 (3.0) | 4025 (1.3) | 8876 (2.5) | 36,657 (4.4) | |
| RD | 9622 (8.2) | 3742 (1.2) | 13,695 (3.9) | 129,374 (15.5) | |
| Comorbidities or symptoms with NCFBE claim, n (%) b | Abnormal sputum | 687 (0.6) | 65 (0.0) | 1276 (0.4) | 18,838 (2.3) |
| Acute lower respiratory tract infections | 681 (0.6) | 160 (0.1) | 1314 (0.4) | 1041 (0.1) | |
| Acute upper respiratory tract infections | 7052 (6.0) | 8086 (2.6) | 15,647 (4.4) | 36,653 (4.4) | |
| Alpha-1 antitrypsin deficiency | 447 (0.4) | 33 (0.0) | 1460 (0.4) | 339 (0.0) | |
| Aspergillus | 1547 (1.3) | 22 (0.0) | 1685 (0.5) | 10,680 (1.3) | |
| Aspiration | 2576 (2.2) | 442 (0.1) | 5814 (1.6) | 102,201 (12.3) | |
| Asthma | 27,048 (23.0) | 9950 (3.2) | 40,151 (11.3) | 202,716 (24.3) | |
| CRS | 9452 (8.0) | 4419 (1.4) | 22,010 (6.2) | 23,117 (2.8) | |
| CVID | 5932 (5.0) | 1666 (0.5) | 8297 (2.3) | 38,930 (4.7) | |
| CTD | 8690 (7.4) | 3226 (1.1) | 12,590 (3.5) | 115,067 (13.8) | |
| Cough | 52,226 (44.4) | 15,784 (5.1) | 68,200 (19.2) | 425,437 (51.1) | |
| Dyspnea | 21,779 (18.5) | 5365 (1.7) | 27,990 (7.9) | 215,094 (25.8) | |
| Fatigue or malaise | 26,663 (22.6) | 26,964 (8.8) | 57,265 (16.1) | 321,647 (38.6) | |
| GERD | 35,399 (30.1) | 29,738 (9.7) | 65,873 (18.5) | 358,001 (43.0) | |
| GVHD | 368 (0.3) | 17 (0.0) | 909 (0.3) | 78 (0.0) | |
| IBD | 1535 (1.3) | 1380 (0.4) | 4038 (1.1) | 17,566 (2.1) | |
| Influenza | 239 (0.2) | 80 (0.0) | 607 (0.2) | 510 (0.1) | |
| NTM | 438 (0.4) | 2 (0.0) | 471 (0.1) | 48 (0.0) | |
| Otitis | 2003 (1.7) | 3224 (1.1) | 5563 (1.6) | 9849 (1.2) | |
| Pathogens of interest c | 4083 (3.5) | 256 (0.1) | 4602 (1.3) | 16,207 (1.9) | |
| Pneumonia | 1803 (1.5) | 306 (0.1) | 1899 (0.5) | 125,004 (15.0) | |
| PCD | 18,697 (15.9) | 1278 (0.4) | 46,218 (13.0) | 87,706 (10.5) | |
| Pseudomonas aeruginosa | 930 (0.8) | 157 (0.1) | 1022 (0.3) | 42,510 (5.1) | |
| Pulmonary tuberculosis | 939 (0.8) | 26 (0.0) | 969 (0.3) | 5477 (0.7) | |
| Comorbidities or symptoms separate from NCFBE claim, n (%) d | Allergic rhinitis | 15,751 (13.4) | 13,150 (4.3) | 33,698 (9.5) | 151,968 (18.2) |
| Anxiety | 18,606 (15.8) | 30,303 (9.9) | 43,265 (12.2) | 148,119 (17.8) | |
| Atrial fibrillation | 15,206 (12.9) | 14,300 (4.7) | 32,416 (9.1) | 116,837 (14.0) | |
| Bronchitis | 1545 (1.3) | 623 (0.2) | 2811 (0.8) | 4494 (0.5) | |
| CVD | 48,135 (40.9) | 43,084 (14.0) | 107,974 (30.4) | 439,015 (52.7) | |
| CKD | 16,357 (13.9) | 18,591 (6.1) | 39,777 (11.2) | 151,372 (18.2) | |
| CHF | 17,210 (14.6) | 11,875 (3.9) | 34,272 (9.6) | 186,471 (22.4) | |
| Depression | 10,443 (8.9) | 13,241 (4.3) | 22,831 (6.4) | 99,823 (12.0) | |
| Emphysema | 329 (0.3) | 54 (0.0) | 766 (0.2) | 251 (0.0) | |
| Hemoptysis | 5083 (4.3) | 321 (0.1) | 15,675 (4.4) | 99,872 (12.0) | |
| Hypertension | 64,745 (55.0) | 107,949 (35.2) | 166,197 (46.8) | 458,629 (55.0) | |
| Interstitial lung disease | 7630 (6.5) | 231 (0.1) | 7883 (2.2) | 144,244 (17.3) | |
| Intraoperative and postprocedural complications/disorders of respiratory system, not elsewhere classified | 342 (0.3) | 131 (0.0) | 633 (0.2) | 1664 (0.2) | |
| Lung diseases due to external agents | 79 (0.1) | 13 (0.0) | 197 (0.1) | 81 (0.0) | |
| MRSA | 688 (0.6) | 419 (0.1) | 1615 (0.5) | 3710 (0.4) | |
| Obesity | 13,700 (11.6) | 29,223 (9.5) | 31,043 (8.7) | 182,065 (21.9) | |
| Osteoarthritis | 23,755 (20.2) | 29,889 (9.7) | 51,118 (14.4) | 191,521 (23.0) | |
| Osteoporosis | 14,065 (11.9) | 9382 (3.1) | 29,738 (8.4) | 69,899 (8.4) | |
| Other diseases of the pleura | 2317 (2.0) | 267 (0.1) | 5758 (1.6) | 22,709 (2.7) | |
| Other diseases of the respiratory system | 8090 (6.9) | 974 (0.3) | 12,956 (3.6) | 55,823 (6.7) | |
| Other diseases of upper respiratory tract | 741 (0.6) | 85 (0.0) | 1500 (0.4) | 412 (0.0) | |
| Peptic ulcer disease | 1988 (1.7) | 1796 (0.6) | 3643 (1.0) | 41,820 (5.0) | |
| Pregnancy | 286 (0.2) | 9023 (2.9) | 1597 (0.4) | 10,258 (1.2) | |
| Tobacco dependence | 15,605 (13.3) | 19,116 (6.2) | 40,470 (11.4) | 110,247 (13.2) | |
aThe IPTW model controlled for the following baseline characteristics: age group; sex, race, payer type, region, CCI score, anxiety/depression, asthma, COPD, cough, CHF, CKD, CVD, dyspnea, GERD, hypertension, interstitial lung disease, primary tuberculosis, prior infections (including pneumonia, P aeruginosa, aspergillus, NTM, and pathogens of interest [Haemophilus influenzae and Mycobacterium avium]), and RD/CTD.
bComorbidities that were included on the same claim as that for NCFBE were identified in the NCFBE cohort. Patients in the comparator group with the identified comorbidities are also shown.
cHaemophilus influenzae and Mycobacterium avium.
dComorbidities that were claimed on a separate claim as that for NCFBE were identified in the NCFBE cohort. Patients in the comparator group with the identified comorbidities are also shown.
CCI, Charlson Comorbidity Index; CHF, congestive heart failure; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; CRS, chronic rhinosinusitis; CTD, connective tissue disease; CVD, cardiovascular disease; CVID, common variable immunodeficiency; GERD, gastroesophageal reflux disease; GVHD, graft-vs-host disease; IBD, inflammatory bowel disease; IPTW, inverse probability treatment weighting; MRSA, methicillin-resistant Staphylococcus aureus; NCFBE, non-cystic fibrosis bronchiectasis; NTM, nontuberculous mycobacteria; PCD, primary ciliary dyskinesia; PVD, peripheral vascular disease; RD, rheumatologic disease.
Mortality risk
Patients with NCFBE had a 77% higher risk of death than non-NCFBE comparators through end of study (HR 1.77 [95% CI 1.74-1.80]; p < .0001; Table 2 and Supplement Figure S2). Risk of death increased with age and was highest among patients aged ≥65 years (HR 11.03 [95% CI 10.36-11.74]; p < .0001). Black patients had a 53% higher risk of death than White patients (HR 1.53 [95% CI 1.50-1.55]; p < .0001). Patients with NCFBE and comorbid COPD had a 42% higher risk of death than patients without NCFBE (HR 1.42 [95% CI 1.40-1.44]; p < .0001).
Table 2.
Mortality risk among patients with NCFBE vs comparators through end of study.
| Mortality | NCFBE cohort vs non-NCFBE comparator cohort | |||
|---|---|---|---|---|
| Hazard ratio | 95% CI | p Value | ||
| Overall | 1.77 | 1.74, 1.80 | <0.0001 | |
| Age group (ref: 18-34 years) | 35–54 years | 5.53 | 5.19, 5.88 | <0.0001 |
| 55–64 years | 5.79 | 5.46, 6.15 | <0.0001 | |
| ≥65 years | 11.03 | 10.36, 11.74 | <0.0001 | |
| Sex (ref: Male) | Female | 0.99 | 0.98, 1.00 | 0.1751 |
| Race (ref: White) | Black | 1.53 | 1.50, 1.55 | <0.0001 |
| Asian | 0.26 | 0.24, 0.29 | <0.0001 | |
| Other | 0.59 | 0.48, 0.73 | <0.0001 | |
| Unknown | 1.32 | 1.30, 1.34 | <0.0001 | |
| Payer channel (ref: Commercial) | Medicare | 1.44 | 1.41, 1.47 | <0.0001 |
| Medicaid | 1.28 | 1.25, 1.31 | <0.0001 | |
| Unknown/other | 0.45 | 0.42, 0.48 | <0.0001 | |
| Region (ref: Northeast) | North central | 0.79 | 0.78, 0.81 | <0.0001 |
| South | 0.65 | 0.64, 0.66 | <0.0001 | |
| West | 0.81 | 0.79, 0.82 | <0.0001 | |
| Other | 0.53 | 0.34, 0.82 | 0.0047 | |
| Comorbidities | COPD | 1.42 | 1.40, 1.44 | <0.0001 |
COPD, chronic obstructive pulmonary disease; NCFBE, non-cystic fibrosis bronchiectasis; ref, reference.
Survival post follow-up period
Among patients who survived longer than the 12-month follow-up period, those with NCFBE had a lower probability of survival than non-NCFBE comparators at 1, 3, and 5 years post follow-up (Table 3). The probability of survival at 1 year post follow-up was 92.1% among patients with NCFBE (vs 97.7% in comparators, p < .0001) and decreased to 75.2% at 5 years post follow-up (vs 93.9% in comparators, p < .0001). The probability of survival in patients with NCFBE decreased by 10.1% (vs 2.4% in comparators) from 1 to 3 years post follow-up and 9.2% from 3 to 5 years post follow-up (vs 1.5% in comparators).
Table 3.
Probability of survival during the follow-up period and at 1, 3, and 5 years post follow-up among patients with NCFBE vs comparators.
| NCFBE cohort | Non-NCFBE comparator cohort | p Value | |||
|---|---|---|---|---|---|
| N | % (95% CI) | N | % (95% CI) | ||
| Follow-up period (index to 12 months) | 117,178 | 94.5 (94.3–94.6) | 306,678 | 98.5 (98.4–98.5) | <0.0001 |
| 1 year follow-up survival (12-24 months postindex) | 112,548 | 92.1 (92.0–92.3) | 305,633 | 97.7 (97.6–97.7) | <0.0001 |
| 3-year follow-up survival (12-36 months postindex) | 89,049 | 82.8 (82.5–83.0) | 236,898 | 95.3 (95.2–95.4) | <0.0001 |
| % Change: Follow-up years 1 to 3 | 10.1 | 2.4 | - | ||
| 5-year follow-up survival (12-60 months postindex) | 61,149 | 75.2 (74.9–75.5) | 134,156 | 93.9 (93.8–94.0) | <0.0001 |
| % Change: Follow-up years 3 to 5 | 9.2 | 1.5 | - | ||
NCFBE, non-cystic fibrosis bronchiectasis.
Exacerbation-specific outcomes among patients with NCFBE
In the 12-month follow-up period, 63.6% of patients with NCFBE had ≥1 exacerbation, including 42.8% who experienced ≥2 exacerbations and 30.2% ≥ 3 exacerbations (Figure 2). After IPTW weighting, 61.0% of patients with NCFBE had ≥1 exacerbation, including 38.3% who had ≥1 exacerbation requiring hospitalization or IV antibiotic use.
Figure 2.
Exacerbations during 12 and 24 months of follow-up among patients with NCFBE, based on (a) exacerbations requiring oral antibiotic use or exacerbations requiring hospitalization or IV antibiotic use and (b) frequency of exacerbations. Note. groups in panel A are not mutually exclusive. IV, intravenous; NCFBE, non-cystic fibrosis bronchiectasis.
After controlling for baseline variables, patients with any exacerbations were more likely to have comorbidities or symptoms, including COPD, alpha-1 antitrypsin (AAT) deficiency, cough, graft-vs-host disease (GVHD), and pulmonary tuberculosis than patients without exacerbations (all p < .0001; Supplement Figures S3–S5). Certain comorbidities were associated with >3.0 times higher odds of exacerbations requiring hospitalization or IV antibiotic use, particularly AAT deficiency, GVHD, and pulmonary tuberculosis (all p < .0001). Factors associated with reduced odds of having any exacerbations included Black race, evidence of chronic kidney disease, hypertension, obesity, or pregnancy. Congestive heart failure and peptic ulcer disease were associated with reduced odds of having exacerbations requiring oral antibiotic use.
After controlling for baseline variables, odds of increased exacerbation frequency were 2.0 to 3.5 times higher in patients with cough, primary ciliary dyskinesia (PCD), GVHD, history of immune deficiency (AAT deficiency or common variable immunodeficiency), COPD, or pulmonary tuberculosis compared with patients without the respective symptom or comorbidity (all p < .0001; Supplement Figures S6–S8). Odds of increased exacerbation frequency were 1.8 times higher in patients with P aeruginosa (p < .0001) and were 2.9 times higher among patients aged 55 to 64 years than patients aged 18 to 34 years. Patients on Medicaid had >2 times the risk of increased exacerbation frequency than commercial payers.
HCRU among patients with NCFBE
In the 12-month follow-up period and after adjustment for baseline variables, patients with NCFBE incurred higher all-cause and respiratory-related HCRU than non-NCFBE comparators in terms of outpatient office, outpatient hospital, ED., and inpatient visits (all p < .0001), as well as for respiratory-related pulmonologist visits (p < .0001) (Table 4; unadjusted HCRU is in Supplement Table S3). However, inpatient length of stay was shorter among patients with NCFBE, both for all-cause (p < .0001) and respiratory-related visits (p = .0085).
Table 4.
HCRU in patients with NCFBE vs comparators during 12-month follow-up.
| HCRU a | NCFBE cohort (N = 117,718) | Non-NCFBE comparator cohort (N = 306,678) | p Value (cases vs comparators) | |||
|---|---|---|---|---|---|---|
| LS means | 95% CI | LS means | 95% CI | |||
| All-cause healthcare utilization | Outpatient office visits | 9.63 | 9.17–10.12 | 5.17 | 4.93–5.43 | <.0001 |
| Outpatient hospital visits | 1.96 | 1.74–2.22 | 0.67 | 0.6–0.76 | <.0001 | |
| ED visits | 1.71 | 1.57–1.86 | 0.93 | 0.85–1.01 | <.0001 | |
| Inpatient visits | 4.32 | 3.71–5.03 | 1.37 | 1.18–1.6 | <.0001 | |
| Length of inpatient stay, days | 8.36 | 6.01–11.62 | 9.96 | 7.52–13.21 | <.0001 | |
| Respiratory-related healthcare utilization | Outpatient office visits | 2.72 | 2.47–3.00 | 0.81 | 0.73–0.89 | <.0001 |
| Outpatient hospital visits | 2.72 | 2.34–3.15 | 0.45 | 0.39–0.52 | <.0001 | |
| ED visits | 1.09 | 0.90–1.31 | 0.47 | 0.39–0.56 | <.0001 | |
| Pulmonologist visits | 0.02 | 0.01–0.04 | 0.004 | 0.00–0.01 | <.0001 | |
| Inpatient visits | 1.97 | 1.56–2.49 | 0.37 | 0.29–0.46 | <.0001 | |
| Length of inpatient stay, days | 5.16 | 2.67–9.98 | 6.95 | 3.74–12.92 | 0.0085 | |
aPatients without HCRU for a service were included. Zero-inflated negative binomial models were used for length of stay; Poisson distribution was used for visits.
ED, emergency department; HCRU, healthcare resource utilization; LS means, least square means; NCFBE, non-cystic fibrosis bronchiectasis.
HCRU among patients with exacerbations
Exacerbation-related HCRU for patients with exacerbations was generally higher than respiratory-related HCRU for patients without exacerbations, in terms of outpatient office, ED, pulmonologist, and inpatient visits, as well as exacerbation-related length of stay (all p < .0001) (Supplement Table S4; unadjusted HCRU is in Supplement Table S5). However, there were fewer exacerbation-related outpatient visits among patients with exacerbations than respiratory-related outpatient visits among patients without exacerbations (p < .0001).
HCRU increased with exacerbation frequency (Supplement Table S6). Patients with 1 exacerbation incurred higher all-cause HCRU than patients without exacerbations. Patients with ≥2 or ≥3 exacerbations also had generally higher HCRU compared with patients without exacerbations, except for length of inpatient stay, which was similar.
Discussion
In this real-world claims-based study, which aimed to assess patient survival, disease burden, and HCRU associated with NCFBE, risk of death was 77% higher in patients with NCFBE than in non-NCFBE comparators, and the probability of survival over time was significantly lower (75.2% vs 93.9% at 5 years). The difference in survival was observed despite a higher proportion of patients having severe comorbidities, such as metastatic carcinoma, in the comparator group. Prior studies on survival in patients with NCFBE were conducted in single centers with small patient groups and reported varying survival estimates. In an outpatient study of sequential patients with NCFBE in the UK (N = 91), 4-year survival was 91%, decreasing to 68.3% by 12 years. 16 In newly diagnosed patients with NCFBE in Belgium (N = 245), overall mortality after 5.2 years was estimated at 20.4%. 23 In a retrospective cohort study of patients with NCFBE in Turkey (N = 56), mortality was 35.7% at 5.4 years. 24 A prospective study of consecutive outpatients with NCFBE in Turkey (N = 98) reported 1-year survival rates of 97%, which was reduced to 58% by 4 years. 15
In this study, key factors increasing the risk of death were advanced age, Black race, and having comorbid COPD. The survival rates were based on an employed, insured population, with approximately 28% of patients ≤64 years of age (mean, 69 years). In comparison, survival rates reported in an older population (mean age, 77 years) of Medicare patients with NCFBE were lower (63% at 5 years) than this study. 9 Other studies have also noted the association of advanced age with increased mortality.15,16,23–25 Studies that assessed patient age among survivors and nonsurvivors with NCFBE found that surviving patients were younger (mean age 59.0–59.7 years) than nonsurvivors (mean age 71.0–72.1 years).15,26 The presence of comorbidities has been associated with increased mortality in other reports. 24 A prospective observational cohort study of patients with NCFBE in Europe reported that an increase of one comorbidity equated to a 17% increase in mortality. 27 The presence of COPD has been noted as an important factor for increased risk of death in other reports.23,26,28 This study found that Black patients with NCFBE had increased risk of death compared with White patients despite having fewer exacerbations reported. This discrepancy may reflect a higher risk of death from other causes or underreporting of exacerbations in Black patients because of socioeconomic factors, decreased access to care, and social drivers of health inequalities. 29
All-cause and respiratory-related HCRU were generally higher in patients with NCFBE than comparators. Other real-world studies have found that patients with NCFBE used more healthcare resources than matched controls.18,30 The higher HCRU in patients with NCFBE may be attributed to increased medical appointments, hospital days, and use of antibiotics. 30 By contrast, the current analysis noted that length of inpatient stay, both all-cause and respiratory-related visits, was shorter in patients with NCFBE than in comparators. Other reports have noted a recent trend toward shorter length of stay for patients hospitalized with COPD or NCFBE and suggested the decline may be related to improved quality of care and focus on stabilization and early discharge. 31
Among patients with NCFBE, HCRU was considerably higher in those with exacerbations than without. Notably, HCRU increased as number of exacerbations increased. These findings support other reports that identified exacerbations as a main driver of HCRU associated with NCFBE,12,17 with hospitalizations related to exacerbations constituting a major component of the resource burden of NCFBE. 19 Notably, length of inpatient stay was similar between patients with 2 or 3 exacerbations, suggesting a consistent standard of care after ≥2 exacerbations, although future claims-based studies of patients with multiple exacerbations would be needed to confirm this.
Limitations
The use of IPTW to balance patient characteristics among cohorts is increasing rapidly in literature, 32 although it is associated with limitations. For example, results may be affected by residual systematic differences in baseline characteristics between case and comparator patients, even if predictors of artificial censoring and the outcome are accounted for by the correction. 32 There may also be issues with extreme weights 33 ; however, this can be addressed by weight stabilization (applied in this study) and/or truncation. 34 IPTW is also sensitive to misspecifications of the propensity-score model biasing the effect estimate. 34 HRs can be estimated using IPTW with less bias compared with propensity-score stratification or adjustment using the propensity score. 34 Despite these limitations, IPTW allows all eligible patients to be retained, thus increasing the effective sample size, whereas unmatched individuals are often discarded during propensity-score analysis.33,34 Consistent with other propensity score-based analysis methods, IPTW can be used to summarize all patient characteristics to a single covariate (the propensity score). These methods are warranted in analyses with either a large number of confounders or a small number of events. 34
Additionally, this study was subject to inherent limitations of a claims database, including data entry and coding errors that may result in misclassification of patients. The presence of a diagnosis code on a claim cannot be considered confirmatory of disease, and the presence of a claim for a filled prescription does not indicate the medication was consumed or taken as prescribed. Medications filled over the counter or provided as samples by the physician are not observed in claims data, and patients paying out of pocket may not be represented. Specific to this study, patients with PCD, which can be a cause of NCFBE, 1 were captured in both the NCFBE and non-NCFBE comparator groups; in the comparator group, this likely represents individuals who had a diagnosis claim for PCD without a separate diagnosis claim for NCFBE. The advantage of using a comprehensive claims database is that it incorporates all medical and pharmacy claims and allows for longitudinal analysis of a large sample, thus providing unique information across the continuum of care. Despite being of interest, the impact of P aeruginosa infection could not be assessed when comparing patients with NCFBE with non-NCFBE comparators because of the small number of patients recorded as having the infection. The impact of COVID-19 was not controlled for in this study, which may have disproportionally impacted the NCFBE group that used more recent data.
Conclusions
NCFBE was associated with a significant mortality burden, particularly among older patients, Black patients, and those with comorbid COPD, and the risk of death increased over time. HCRU for NCFBE was high and increased with exacerbation frequency. These results highlight the need for more effective treatments to manage NCFBE and reduce the frequency of exacerbations. Current management of NCFBE focuses on preventing infections, optimizing mucus clearance, and addressing underlying comorbidities. Opportunities to potentially improve patient care include readdressing adherence to individuals’ current management of NCFBE, a better understanding of socioeconomic factors to address lifestyle concerns in risk management and disparities in geographic access to specialty care, and an increased awareness of neutrophil-mediated inflammation and the patient burden associated with NCFBE.
Supplemental Material
Supplemental Material for Real-world disease burden, mortality, and healthcare resource utilization associated with bronchiectasis by Sofia Shoaib, Joseph Feliciano, Elliott C Dasenbrook, Jason Maynard, Lakshmi Batchu, Maitreyee Mohanty, Melanie Lauterio and Alexandra J Feld in Chronic Respiratory Disease.
Acknowledgments
Medical writing support was provided by Jon Carthy, PhD, of Envision Pharma Group and funded by Insmed Incorporated.
Declaration of conflicting interests: S.S., J.M., L.B., and A.J.F., are employees of STATinMED, a paid consultant for Insmed Incorporated. J.F., M.M., and M.L. are employees and shareholders of Insmed Incorporated. E.C.D. reports receiving consulting fees from Insmed Incorporated in the last 24 months.
Funding: The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by Insmed Incorporated.
Supplemental Material: Supplemental material for this article is available online.
Ethical Statement
This study was performed in accordance with relevant guidelines and regulations. Since this study did not involve the collection, use, or transmittal of individually identifiable data, institutional review board approval was not required. Only analyzed deidentified data were used in the study, which are a priori exempt from the Federal Policy for the Protection of Human Subjects (1991) and does not meet the identification criteria necessary to be privileged under the Health Insurance Portability and Accountability Act (HIPAA) of 1996. Nonetheless, both the dataset and the security of the offices where the dataset was kept met HIPAA requirements, and by default, this study was in accord with the same.
ORCID iD
Sofia Shoaib https://orcid.org/0009-0006-6185-3965
References
- 1.O'Donnell AE. Bronchiectasis – a clinical review. N Engl J Med 2022; 387(6): 533–545. [DOI] [PubMed] [Google Scholar]
- 2.Chalmers JD, Elborn S, Greene CM. Basic, translational and clinical aspects of bronchiectasis in adults. Eur Respir Rev 2023; 32(168): 230015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.O'Donnell AE. Medical management of bronchiectasis. J Thorac Dis 2018; 10(suppl 28): S3428–S3435. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Chalmers JD. New insights into the epidemiology of bronchiectasis. Chest 2018; 154(6): 1272–1273. [DOI] [PubMed] [Google Scholar]
- 5.Weycker D, Hansen GL,, Seifer FD. Prevalence and incidence of noncystic fibrosis bronchiectasis among US adults in 2013. Chron Respir Dis 2017; 14(4): 377–384. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Martinez-García MA, Oscullo G, Posadas T, et al. Pseudomonas aeruginosa and lung function decline in patients with bronchiectasis. Clin Microbiol Infect 2021; 27(3): 428–434. [DOI] [PubMed] [Google Scholar]
- 7.Chalmers JD, Aliberti S, Filonenko A, et al. Characterization of the “frequent exacerbator phenotype” in bronchiectasis. Am J Respir Crit Care Med 2018; 197(11): 1410–1420. [DOI] [PubMed] [Google Scholar]
- 8.Hill AT, Haworth CS, Aliberti S, et al. Pulmonary exacerbation in adults with bronchiectasis: a consensus definition for clinical research. Eur Respir J 2017; 49(6): 1700051. [DOI] [PubMed] [Google Scholar]
- 9.Feliciano J, Lewing B, Mohanty M, et al. Survival outcomes in US Medicare patients with bronchiectasis by rate of baseline exacerbations. Abstract presented at: 6th World Bronchiectasis & NTM Conference, New York, NY, July 18-20, 2023. [Google Scholar]
- 10.Martínez-García M, Máiz L, Olveira C, et al. Spanish guidelines on treatment of bronchiectasis in adults. Arch Bronconeumol 2018; 54(2): 88–98. [DOI] [PubMed] [Google Scholar]
- 11.Hill AT, Barker AF, Bolser DC, et al. Treating cough due to non-CF and CF bronchiectasis with nonpharmacological airway clearance: CHEST expert panel report. Chest 2018; 153(4): 986–993. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Polverino E, Goeminne PC, McDonnell MJ, et al. European Respiratory Society guidelines for the management of adult bronchiectasis. Eur Respir J 2017; 50(3): 1700629. [DOI] [PubMed] [Google Scholar]
- 13.Flume PA, Basavaraj A, Garcia B, et al. Towards development of evidence to inform recommendations for the evaluation and management of bronchiectasis. Respir Med 2023; 211: 107217. [DOI] [PubMed] [Google Scholar]
- 14.Chalmers JD, Goeminne P, Aliberti S, et al. The bronchiectasis severity index. An international derivation and validation study. Am J Respir Crit Care Med 2014; 189(5): 576–585. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Onen ZP, Gulbay BE, Sen E, et al. Analysis of the factors related to mortality in patients with bronchiectasis. Respir Med 2007; 101(7): 1390–1397. [DOI] [PubMed] [Google Scholar]
- 16.Loebinger MR, Wells AU, Hansell DM, et al. Mortality in bronchiectasis: a long-term study assessing the factors influencing survival. Eur Respir J 2009; 34(4): 843–849. [DOI] [PubMed] [Google Scholar]
- 17.Germino R, Liao L. Resource utilization associated with non-cystic fibrosis bronchiectasis patients who are being treated for exacerbations. Value Health 2016; 19(3): A116. [Google Scholar]
- 18.Guan W-J, Han X-R, de la Rosa-Carrillo D, et al. The significant global economic burden of bronchiectasis: a pending matter. Eur Respir J 2019; 53(2): 1802392. [DOI] [PubMed] [Google Scholar]
- 19.Goeminne PC, Hernandez F, Diel R, et al. The economic burden of bronchiectasis – known and unknown: a systematic review. BMC Pulm Med 2019; 19(1): 54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Navari RM, Nelson WW, Shoaib S, et al. Real-world treatment outcomes, healthcare resource use, and costs associated with antiemetics among cancer patients on cisplatin-based chemotherapy. Adv Ther 2023; 40(7): 3217–3226. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.The US Department of Health and Human Services . Summary of the HIPAA privacy rule. US Department of Health and Human Services, 2003. Accessed April 24, 2024:https://www.hhs.gov/sites/default/files/privacysummary.pdf [Google Scholar]
- 22.Flume PA, Feliciano J, Lucci M, et al. Pulmonary exacerbations in insured patients with bronchiectasis over 2 years. ERJ Open Research 2023; 9(4): 00021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Goeminne PC, Nawrot TS, Ruttens D, et al. Mortality in non-cystic fibrosis bronchiectasis: a prospective cohort analysis. Respir Med 2014; 108(2): 287–296. [DOI] [PubMed] [Google Scholar]
- 24.Çiftçi F, Şen E, Saryal SB, et al. The factors affecting survival in patients with bronchiectasis. Turk J Med Sci 2016; 46(6): 1838–1845. [DOI] [PubMed] [Google Scholar]
- 25.Seitz AE, Olivier KN, Steiner CA, et al. Trends and burden of bronchiectasis-associated hospitalizations in the United States, 1993-2006. Chest 2010; 138(4): 944–949. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Mateus SP, Ribeiro-Alves M, Salles REB, et al. Mortality and comorbidities in patients with bronchiectasis over a 3-year follow-up. Medicine (Baltim) 2022; 101(52): e32537. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.McDonnell MJ, Aliberti S, Goeminne PC, et al. Comorbidities and the risk of mortality in patients with bronchiectasis: an international multicentre cohort study. Lancet Respir Med 2016; 4(12): 969–979. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Goeminne PC, Scheers H, Decraene A, et al. Risk factors for morbidity and death in non-cystic fibrosis bronchiectasis: a retrospective cross-sectional analysis of CT diagnosed bronchiectatic patients. Respir Res 2012; 13(1): 21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Richardus JH, Kunst AE. Black-white differences in infectious disease mortality in the United States. Am J Public Health 2001; 91(8): 1251–1253. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Diel R, Chalmers JD, Rabe KF, et al. Economic burden of bronchiectasis in Germany. Eur Respir J 2019; 53(2): 1802033. [DOI] [PubMed] [Google Scholar]
- 31.Ford ES. Hospital discharges, readmissions, and ED visits for COPD or bronchiectasis among US adults: findings from the nationwide inpatient sample 2001-2012 and Nationwide Emergency Department Sample 2006-2011. Chest 2015; 147(4): 989–998. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Austin PC, Stuart EA. Moving towards best practice when using inverse probability of treatment weighting (IPTW) using the propensity score to estimate causal treatment effects in observational studies. Stat Med 2015; 34(28): 3661–3679. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Allan V, Ramagopalan SV, Mardekian J, et al. Propensity score matching and inverse probability of treatment weighting to address confounding by indication in comparative effectiveness research of oral anticoagulants. J Comp Eff Res 2020; 9(9): 603–614. [DOI] [PubMed] [Google Scholar]
- 34.Chesnaye NC, Stel VS, Tripepi G, et al. An introduction to inverse probability of treatment weighting in observational research. Clin Kidney J 2021; 15(1): 14–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental Material for Real-world disease burden, mortality, and healthcare resource utilization associated with bronchiectasis by Sofia Shoaib, Joseph Feliciano, Elliott C Dasenbrook, Jason Maynard, Lakshmi Batchu, Maitreyee Mohanty, Melanie Lauterio and Alexandra J Feld in Chronic Respiratory Disease.