Real-world evidence is a vital tool for informing treatment strategies in chronic obstructive pulmonary disease

The current model for managing chronic obstructive pulmonary disease (COPD) generally considers inhalers from the same therapeutic class to be clinically equivalent.1 When a long-acting muscarinic antagonist is indicated, for example, treatment guidelines generally do not distinguish between which long-acting muscarinic antagonist is selected. Yet, assumptions of intraclass clinical equivalence among inhalers are rarely tested in large head-to-head randomised controlled trials.2 The absence of robust comparative data leaves open the possibility of differential outcomes within classes due to unexpected effects from minor variations in the active pharmaceutical compounds or the complex devices that deliver these compounds to the lungs.

In such circumstances, when clinical trial data are unavailable, high quality evidence generated from healthcare data collected in routine practice (such as insurance claims or electronic medical records) can be an indispensable resource for evaluating the risks and benefits of different treatment approaches.345 In a study published in The BMJ, we used insurance claims to compare the two single-inhaler triple therapies available on the US market: budesonide-glycopyrrolate-formoterol (Breztri Aerosphere), a twice daily metered-dose inhaler, and fluticasone-umeclidinium-vilanterol (Trelegy Ellipta), a once daily dry powder inhaler.6 We found that the former was associated with a slightly higher risk of moderate or severe COPD exacerbations compared with the latter, and a similar risk of pneumonia hospitalisations.

In the absence of head-to-head trials comparing these drugs, our research provides data to inform prescribing decisions, formulary design, clinical treatment guidelines, and educational initiatives by health systems. Although the observed advantages of fluticasone-umeclidinium-vilanterol were modest—with budesonide-glycopyrrolate-formoterol associated with a 9% higher risk of moderate or severe COPD exacerbations and a 29% higher risk of severe exacerbations—such benefits reflect important potential population-level gains for reducing COPD morbidity, improving quality of life, and lowering healthcare costs.

One distinct advantage of rigorous healthcare database studies is that they can be used to compare the performance of therapies as patients actually use (or misuse) them in routine clinical care. Such studies, for example, may pick up benefits related to patient adherence with different dosing regimens or inhaler types. Randomised controlled trials, by contrast, which generally seek to ensure consistent and technically sound inhaler use among study participants, may miss these signals. Indeed, two hypotheses for the small benefit that we observed among patients receiving fluticasone-umeclidinium-vilanterol are its convenient dosing (once daily rather than twice daily) and user-friendly delivery (which requires only deep inspiration and not timing inhaler actuation with inspiration). Of course, the observed differences in our study could also be due to minor variations in the pharmaceutical compounds themselves, and further research is needed to verify and evaluate these potential differences. But well designed observational studies can detect different outcomes that may result from user errors, which randomised controlled trials often do not seek to assess. Our study provides physicians, payers, guideline-writing committees, and health systems with data reflecting the lived reality of COPD management.

Evidence from large healthcare database studies can also provide reassurance to stakeholders who may have reasons to favour a particular therapy beyond its clinical effect. For example, dry powder inhalers are associated with just one twentieth of the carbon footprint of their metered-dose counterparts.789 Health systems seeking to reduce their greenhouse gas emissions may feel more confident, in light of our findings, when taking measures to promote increased prescribing of fluticasone-umeclidinium-vilanterol relative to budesonide-glycopyrrolate-formoterol. These findings help verify that our duties to protect the environment are consistent with our duties to provide the best possible patient care; the inhaler with lower greenhouse gas emissions was no worse (and indeed performed slightly better) than the inhaler with higher emissions.

Fortunately, the environmental advantage of fluticasone-umeclidinium-vilanterol may be short lived. AstraZeneca has announced the completion of clinical studies for a version of budesonide-glycopyrrolate-formoterol containing a novel propellant, which is associated with far lower greenhouse gas emissions than current versions and a lifecycle carbon footprint approximating that of dry powder inhalers.10 Such efforts to develop low-carbon inhalers are laudable, in part, because some patients prefer or require metered-dose inhalers. Yet, the greener formulation of budesonide-glycopyrrolate-formoterol will likely be patent-protected well beyond the older version and fluticasone-umeclidinium-vilanterol.11 Physicians and health systems may thus face a choice between an expensive brand-name version of one product (a metered-dose inhaler with a low carbon propellant) and less expensive generic versions of another (a dry powder inhaler). While this novel formulation is not expected to have a different therapeutic effect than the original version, future claims-based studies could compare the low carbon formulation of budesonide-glycopyrrolate-formoterol against fluticasone-umeclidinium-vilanterol. Our study suggests that patients and health systems with financial reasons for favouring the dry powder inhaler over the metered-dose alternative will likely be on firm clinical ground when doing so.

Rigorous post-marketing studies offer a valuable tool for analysing clinical differences between therapies for COPD, which in turn may shape how clinicians, payers, and health systems choose between products with varying trade-offs, whether environmental, financial, or otherwise. Because manufacturers have little incentive to conduct head-to-head trials probing intraclass differences, longitudinal healthcare database studies represent a vital next step for reconsidering the current treatment paradigm in COPD.34