Competitive Bidding Program not linked to supplemental oxygen use, COPD outcome changes

This study by Duan and associates is statistically sound. Unfortunately, that faint praise is really all that can be offered. The authors make a variety of inexplicable assumptions and irrelevant analyses on their way to an inaccurate conclusion that jeopardizes efforts to actually improve the sorry state of oxygen therapy in America.

The biggest problem is the choice of primary outcomes (new oxygen prescriptions and discontinuations of oxygen therapy). The authors assert that these measures can be used to represent oxygen therapy “access.” They then claim that since their analysis did not demonstrate any change in the probability of initiation or discontinuation of oxygen therapy, Medicare’s CBP had no impact on oxygen therapy utilization. However, this overlooks critical context. Because of Medicare’s payment rules, a durable medical equipment (DME) supplier receives the same payment for “portable oxygen” whether they provide a portable concentrator, compressed gas cylinders or a liquid oxygen system. The DME company is allowed to provide whatever modality it deems most cost-effective, rather than what is most appropriate for the patient. Suppliers are then able to replace equipment with devices that are less costly to service and support (regardless of suitability) as reimbursement cuts demand, rather than discontinuing service altogether. This creates the illusion of preserved “access,” as the authors define it, indifferent to clinical utility.

Ironically, the answer to making this an impactful, beneficial paper is buried within it. Throughout the discussion section, there are sentences like, “Therefore, it remains unclear from our study whether the CBP was differentially associated with either low-value or high-value oxygen prescribing,” “limitations of claims data precluded evaluation of certain quality and patient-centered outcomes” and “we did not evaluate how the CBP changed the DME supplier market, which may have affected the quality of care.” Had the authors leaned into these uncertainties, this paper could have been an opportunity to generate additional discussions about why we don’t know what we don’t know about the oxygen therapy population. It could also have highlighted the aspects of reform efforts such as the SOAR Act that are, in fact, aimed at making supplemental oxygen a more value- and evidence-based therapy, preserving clinical quality while still reducing costs. Instead, the myopic drive to apply a binary answer to a nebulous question has done a remarkable disservice to patients, prescribers and the health care system at large, discrediting major resources invested in reform efforts.

Given the significant structural problems with the paper, it is frankly surprising that it was selected for publication by such a prestigious journal. However, that is less an indictment of the journal and more a sad commentary on the lack of understanding about oxygen therapy in the medical community. It is difficult to blame peer reviewers for missing context when that context has rarely (if ever) been presented. That makes the true takeaway from this paper not its conclusions, but the dire need to broaden understanding of oxygen therapy and the real-world impact of barriers to its access.

As we noted in our study, our primary outcomes are the best available measure of access through claims data, although we acknowledge that this does not fully characterize access to supplemental oxygen. Despite these limitations, we believe the outcomes nevertheless provide valuable insights into supplemental oxygen use patterns.

We do not view our study as applying a binary answer to the policy discussion, but rather providing important data using rigorous methods to inform policy, and as a result, bringing forward more nuance to ongoing policy considerations. Various approaches to improving supplemental oxygen care through policy could be considered, and our study sheds light on one legislative approach, which is the CBP that was designed to contain costs. However, broader re-examination of supplemental oxygen care overall may offer other approaches, such as improving supplemental oxygen appropriateness based on clinical indications. Such an approach could, in fact, get us closer to the shared objective of improving patient care, while being mindful of health system resource utilization.