Peek behind the paper: costs, exacerbations and pneumonia after initiating combination tiotropium and olodaterol versus triple therapy for COPD
In this feature, lead study author Swetha Palli (Boehringer Ingelheim Pharmaceuticals, CT, USA) discusses the comparative costs, exacerbations and pneumonia associated with combination tiotropium and olodaterol versus triple therapy for individuals with chronic obstructive pulmonary disease (COPD).
Please could you introduce yourself?
I am a Product Director in the Health Economics and Outcomes Research Value Demonstration team at Boehringer Ingelheim Pharmaceuticals (CT, USA). As lead author for this study, I am thrilled to be discussing the importance of comparing combination tiotropium and olodaterol (TIO + OLO) versus triple therapy from a real-world perspective. This study was a collaboration between Boehringer Ingelheim Pharmaceuticals and Optum (MN, USA).
What prompted you to conduct this study?
Since its advent in 2014–2015, the use of dual bronchodilator therapy – comprising a long-acting muscarinic antagonist plus a long-acting β2-adrenoceptor agonist (LAMA + LABA) – continues to lag behind the use of other mono and dual maintenance therapies for the management of chronic obstructive pulmonary disease (COPD), despite its superior efficacy.
Conversely, overuse of triple therapy – comprising LAMA + LABA + inhaled corticosteroids (ICS) –persists across the COPD severity spectrum, while the Global Initiative for Chronic Obstructive Lung Disease (GOLD; WI, USA) recommends this regimen only for the most severe of patients, who experience further disease exacerbations after escalating to dual bronchodilator therapy.
...the use of dual bronchodilator therapy continues to lag behind the use of other mono and dual maintenance therapies for the management of chronic obstructive pulmonary disease, despite its superior efficacy."
In this investigation, we wanted to expand the existing evidence by conducting a rigorously matched, real-world study comparing dual- and triple-therapy regimens, in order to assess their differential impact. The study focuses on the tiotropium plus olodaterol (TIO + OLO) dual-therapy regimen for the primary analysis and any free or fixed dose combination of LAMA + LABA in the secondary analysis.
What are some of the challenges associated with employing real-world data sources, such as Medicare data, and what limitations may these impose on the results of your study?
While administrative databases have been frequently and successfully leveraged to conduct real-world evidence studies, there are some inherent limitations with this approach. The biggest limitation continues to be the absence of specific clinical measures for determining the severity of any given condition. For instance, accurate COPD disease assessment would require results from spirometry tests, the COPD Assessment Test and the Modified British Medical Research Council Questionnaire, as well as detailed knowledge of patients’ tobacco histories, COPD phenotypes and/or blood eosinophil counts.
While our study requirements for participant inclusion – comprising at least two diagnosis codes for COPD on separate dates of service, plus at least one pharmacy claim for a COPD maintenance medication – strengthen the odds of any eligible patient to have clinical COPD, possession of a medical claim with a code is itself not usually a positive proof of disease. A diagnosis on a medical claim could have been miscoded or included as a rule-out criterion. Similarly, the presence of pharmacy claim demonstrates that a prescription was dispensed; this does not guarantee that a patient actually took a medication as prescribed.
What was the rationale behind the screening criteria you employed to define your study cohort and what limitations may these impose?
The study intended to identify a pool of COPD patients first, before narrowing down to those initiating triple therapy or TIO + OLO therapy. To this end, we required multiple distinct medical claims involving COPD diagnosis, plus COPD maintenance therapy initiation, to increase the odds of an eligible patient having COPD.
When these kinds of comparative analyses are performed using real-world data, a large number of clinically relevant patient characteristics could potentially influence outcomes."
We also excluded patients with more than two distinct claims for asthma, cystic fibrosis and/or lung cancer, to increase the sensitivity of our study results and ensure that the results were attributable to COPD alone.
Our selection of four treatment groups – LAMA monotherapy, LABA + ICS, LAMA + LABA and triple therapy – based on the first observed regimen during the identification period, before focusing on the two cohorts of interest, means that we did not include patients on LAMA monotherapy or LABA + ICS who might have escalated to either of the treatments under consideration. Furthermore, the selection of study participants covered by only Medicare Advantage Part D, limits the generalizability of the study findings to patients covered by other insurance types.
Nevertheless, these criteria aimed to ensure that we were assessing the true comparative effectiveness of TIO + OLO versus triple therapy initiation in a COPD population covered under a very relevant insurance type.
What was the importance of employing inverse probability treatment weighting to balance the patient cohorts prior to assessing patient outcomes?
When these kinds of comparative analyses are performed using real-world data, a large number of clinically relevant patient characteristics could potentially influence outcomes. As such, confounding by indication is a very important issue that needs to be addressed. Inverse probability treatment weighting is an established method for minimizing potential confounding factors in observational studies before assessing a treatment effect.
What may the implications of your study be; how do you hope your study results may impact on clinical decision making and/or prescribing practices?
This study found that, compared with triple therapy, TIO + OLO was associated with significantly lower COPD-related healthcare use and substantially lower COPD-related health plan costs per patient per annum. This was explained by TIO + OLO’s association with lower inpatient use, severe exacerbation rates, and incidences of pneumonia or bronchiolitis.
Taken together, we hope these findings provide important real-world insights to decision makers and prescribers on the positive impact of prescribing according to GOLD recommendations, while avoiding unnecessary steroid exposure for COPD patients."
Furthermore, less than 50% of study participants who received triple therapy did so in concordance with GOLD recommendations; more than half of individuals initiated triple therapy as first-line treatments or had no history of severe exacerbations in the preceding 12 months.
Taken together, we hope these findings provide important real-world insights to decision makers and prescribers on the positive impact of prescribing according to GOLD recommendations, while avoiding unnecessary steroid exposure for COPD patients. We hope these results encourage health plans, systems and clinical care practitioners to evaluate their own levels of GOLD-discordant triple prescribing, and identify strategies and tactics to reduce this excess use, while potentially reducing their total cost of COPD care.
Do you have further RWE-based studies planned in the area?
An on-treatment analysis is currently being conducted as a follow-up to validate the current results. Given the high proportion of GOLD-discordant triple therapy usage we found, we would also like to stratify these analyses by exacerbation history and/or maintenance-naïve status, to evaluate the differential margin between TIO + OLO and triple therapy across these subgroups.
The opinions expressed in this feature are those of the interviewee/author and do not necessarily reflect the views of The Evidence Base® or Future Science Group.