A Network Meta-Analysis of Triple Therapy Inhalers Used in Moderate to Severe Chronic Obstructive Pulmonary Disease
Author(s)
Shahariar Mohammed Fahim, PhD, MS1, Abigail Wright, PhD, MSc1, Belen Herce-Hagiwara, BA1, Jeffrey Tice, MD2, David M Rind, MD, MSc1, Foluso Agboola, MBBS, MPH1.
1Institute for Clinical and Economic Review, Boston, MA, USA, 2University of California San Francisco, San Francisco, CA, USA.
1Institute for Clinical and Economic Review, Boston, MA, USA, 2University of California San Francisco, San Francisco, CA, USA.
Presentation Documents
OBJECTIVES: Previous network meta-analyses (NMAs) comparing triple therapy inhalers for the treatment of chronic obstructive pulmonary disease (COPD) reported inconsistent findings due to differences in inclusion criteria and statistical methods. We evaluated the comparative effectiveness of triple therapy inhalers for COPD using the most recent evidence and methodological guidance.
METHODS: We systematically identified randomized trials (RCTs) that ensured participants in all arms of the trial were prescribed the same number of inhalers with the same dosing schedule in order to assess the efficacy of different triple therapy regimens independent of treatment adherence. We performed Bayesian NMA using a random-effects model to estimate the treatment effect for annualized moderate to severe COPD exacerbation rate. The rate of moderate to severe COPD exacerbations was modeled using the log-transformed rate ratio and standard error; the output was relative risk with 95% credible intervals generated using a normal likelihood with an identity link.
RESULTS: We identified 16 publications describing 18 RCTs with nine unique triple therapies that met our criteria. The NMA results showed no statistically significant differences between any single-inhaler or multi-inhaler triple therapies on moderate to severe exacerbation rates. The single-inhaler fluticasone furoate/umeclidinium/vilanterol and multi-inhaler budesonide/formoterol plus tiotropium had lower point estimates than other available triple therapy combinations, but these differences were not statistically significant.
CONCLUSIONS: All nine triple therapy regimens included in this review provide similar clinical benefits in reducing the rates of moderate to severe exacerbations. In comparison to previous NMAs, the results of this NMA highlight that conclusions depend on the selection of studies (e.g., open-label vs. blinded), controlling the number of inhalers across treatment arms, and choice of statistical model (e.g., fixed vs random effects). Differences in methodologic choices likely explain prior inconsistent results. Explicit rationale for NMA decisions should be provided in future NMAs.
METHODS: We systematically identified randomized trials (RCTs) that ensured participants in all arms of the trial were prescribed the same number of inhalers with the same dosing schedule in order to assess the efficacy of different triple therapy regimens independent of treatment adherence. We performed Bayesian NMA using a random-effects model to estimate the treatment effect for annualized moderate to severe COPD exacerbation rate. The rate of moderate to severe COPD exacerbations was modeled using the log-transformed rate ratio and standard error; the output was relative risk with 95% credible intervals generated using a normal likelihood with an identity link.
RESULTS: We identified 16 publications describing 18 RCTs with nine unique triple therapies that met our criteria. The NMA results showed no statistically significant differences between any single-inhaler or multi-inhaler triple therapies on moderate to severe exacerbation rates. The single-inhaler fluticasone furoate/umeclidinium/vilanterol and multi-inhaler budesonide/formoterol plus tiotropium had lower point estimates than other available triple therapy combinations, but these differences were not statistically significant.
CONCLUSIONS: All nine triple therapy regimens included in this review provide similar clinical benefits in reducing the rates of moderate to severe exacerbations. In comparison to previous NMAs, the results of this NMA highlight that conclusions depend on the selection of studies (e.g., open-label vs. blinded), controlling the number of inhalers across treatment arms, and choice of statistical model (e.g., fixed vs random effects). Differences in methodologic choices likely explain prior inconsistent results. Explicit rationale for NMA decisions should be provided in future NMAs.
Conference/Value in Health Info
2025-05, ISPOR 2025, Montréal, Quebec, CA
Value in Health, Volume 28, Issue S1
Code
SA63
Topic
Study Approaches
Topic Subcategory
Meta-Analysis & Indirect Comparisons
Disease
SDC: Respiratory-Related Disorders (Allergy, Asthma, Smoking, Other Respiratory)