Good morning! Welcome to another edition of Pressure Support, a weekly newsletter simplifying pulmonary and critical care medicine—and the systems that shape it.
Off to the MICU tomorrow! I’m also giving a journal club presentation on the PRAGUE-26 trial, which I wrote about here. I recorded my test-run presentation, which you can watch HERE.
I’m alpha testing this newsletter—so please let me know your feedback.
As a reminder, here’s how the newsletter will be structured:
Main piece of content: this is an article such as Journal Watch, Inefficiency Insights, The Middlemen, Career and Training, and Practice Management. All of this content will relate to the system in which we practice. That’s to say, this isn’t just a “medical take” on an issue. It’s a “systemic” take.
Weekly Digest: these are brief summaries of articles or publications I’ve come across during the week that I found insightful to my own practice.
Pop Quiz: some trivia on pulmonary and critical care, ranging from board-style questions to practical, health system-based questions (like today’s one).
If you have colleagues who’d benefit from this newsletter, share it with them using the link below:
September 16th, 2026
Medicare’s DME Crackdown Needs Better Claims Controls
For pulmonologists, durable medical equipment is part of everyday care. We rely on suppliers to deliver oxygen, PAP devices, nebulizers, ventilators, suction equipment, and the supplies needed to keep those devices working.
The ordering process is already cumbersome:
Document the diagnosis
Enter the order
Attach testing
Send everything to a supplier
The request may still come back because a saturation test was performed under the wrong conditions, a note is missing specific language, or the insurer wants another form.
At the same time, Medicare has a fraud problem within the durable medical equipment, prosthetics, orthotics, and supplies market, commonly called DMEPOS.
According to MedTech Dive, the Trump administration announced a six-month moratorium on new Medicare enrollment for certain DMEPOS suppliers as part of a broader fraud crackdown. A temporary pause gives CMS time to review its controls and keep questionable companies from entering the program.
It may also reduce supplier capacity and slow access for patients if CMS does not pair the crackdown with a more precise way to identify risk.
A 2018 HHS Office of Inspector General audit shows how much of this problem is embedded in the payment process itself:
The OIG reviewed 120,614 Medicare Part B claims for DMEPOS provided while beneficiaries were hospitalized between 2015 and 2017.
Medicare paid suppliers about $34 million for those claims.
The audit concluded that none of the payments should have been made.
During a Medicare Part A-covered inpatient stay, the hospital is generally responsible for providing DMEPOS. Those items are included in the hospital’s inpatient payment. Outside suppliers cannot separately bill Medicare Part B for the same items, except under limited rules for equipment delivered shortly before discharge for use at home.
Medicare’s system still paid the claims. Beneficiaries were also held responsible for $8.7 million in unnecessary deductibles and coinsurance.
The claims edits failed in three different ways:
Missing categories: The system was not designed to catch $30.7 million in claims for prosthetics, orthotics, supplies, and related drugs.
Failed follow-through: The system identified nearly $2 million in DME claims for review, but the claims were still paid or the money was not recovered.
Incorrect rules: Medicare paid another $1.3 million for customized prosthetics because the system applied a skilled nursing facility exception to other inpatient settings.
The OIG estimated that better-designed edits could have saved Medicare $223.1 million from 2008 through 2017. Beneficiaries (our patients) could have avoided another $56.3 million in deductibles and coinsurance.
Improper payment does not automatically mean fraud. Some claims may have resulted from coding mistakes, timing problems, or confusion about Medicare’s billing rules. The audit did not determine the intent behind each claim.
It did identify a clear process failure. Medicare had enough information to know that a beneficiary was hospitalized and that a supplier was separately billing Part B. Its payment system did not consistently connect those facts before releasing the money.
Root Cause Analysis: 5 Whys
The 5 Whys process in root cause analysis involves repeatedly asking "Why?" five times to drill down into the root cause of a problem by exploring the cause-and-effect relationships underlying the issue.
The problem: Medicare pays improper DMEPOS claims and then relies on audits, recoupments, and broad enrollment restrictions to manage the resulting risk.
Why? Medicare’s claims systems do not consistently reject claims that conflict with inpatient status, place of service, supplier eligibility, or payment rules.
Why? Some edits cover only selected product categories. Others flag a claim without ensuring that it is denied or recovered.
Why? DMEPOS billing is divided across suppliers, hospitals, Medicare contractors, and separate Part A and Part B payment workflows.
Why? CMS has added rules and edits over time without creating one process that verifies the supplier, patient setting, coverage rule, and claim before payment.
Why? Medicare manages much of its DMEPOS risk after the claim enters the system. Weak screening and fragmented data allow improper payments through, leaving audits and recoupments to correct them later.
Impact Analysis
Impact analysis is the assessment of the potential consequences and effects that changes in one part of a system may have on other parts of the system or the whole.
Impact Analysis
Patient: Improper billing can leave beneficiaries paying deductibles and coinsurance they never owed. Broad supplier restrictions may also shrink capacity in markets that already struggle to deliver equipment quickly. Delays involving oxygen, noninvasive ventilation, or suction equipment can turn a discharge problem into a clinical problem.
Clinician or Provider: We carry much of the administrative burden even though we do not control supplier enrollment or claims payment. When a request stalls, our teams repeat testing, rewrite orders, send more documentation, and search for another supplier. That work takes time away from clinical care without necessarily improving fraud detection.
System: Medicare absorbs the improper payment, pays contractors to investigate it, tries to recover the money, and adds new controls after the fact. Hospitals and clinics then build manual workflows to navigate those controls. The system pays for the original error and for the administrative response.
The Solution
The solutions surrounding this issue largely have to do with CMS screening suppliers earlier, connecting the data it already has, and stopping invalid claims before payment. This approach protects Medicare (and taxpayer) dollars while allowing physicians and legitimate suppliers to get respiratory equipment to patients on time. More specifically…
Screen suppliers continuously: CMS should verify enrollment, accreditation, licensure, ownership, billing location, and exclusion status before the first claim and at regular intervals afterward. Human labor isn’t necessarily needed for this. AI tools can certainly do this type of routine, redundant checking.
Protect access to time-sensitive respiratory equipment: CMS and Medicare contractors should create an expedited pathway for oxygen, home ventilation, suction, and other equipment required for safe discharge. When a supplier loses enrollment, the patient and clinical team need a rapid transfer to a verified alternative.
Validate the complete claim before payment: Medicare should automatically check inpatient status, discharge destination, place of service, duplicate payment risk, supplier eligibility, and applicable exceptions. A claim that fails one of these checks should not move forward without review. Again, human labor isn’t necessarily warranted. AI tools can do this.
Focus oversight on high-risk billing patterns: Sudden spikes in volume, unusual geographic reach, beneficiary complaints, narrow product concentration, and spending far above peers can trigger targeted review. This directs scrutiny toward the suppliers generating the risk.
Measure access alongside fraud prevention: CMS should track order-to-delivery times, discharge delays, supplier availability, and patient complaints along with prevented payments and recoveries. These measures would show whether program-integrity efforts are creating new barriers to care.
In summary, Medicare needs stronger DMEPOS oversight. The current crackdown may keep some questionable suppliers out of the program, but enrollment restrictions alone will not repair the claims process that allowed improper payments for years.
The VICTORY randomized clinical trial of 238 adults with severe burns found that high-dose IV vitamin C (50 mg/kg every 6 hours for 96 hours) did not improve 28-day mortality or persistent organ dysfunction; the composite outcome was actually worse with vitamin C (40.8% vs 29.7%), with higher 28-day mortality (15.0% vs 7.6%), so the trial was stopped early for futility and possible harm. This is a useful, practice-changing negative trial: it argues against reflexively adding high-dose vitamin C to severe-burn protocols and shows how adequately powered multicenter trials can simplify care by preventing low-value treatment.
A stepped-wedge cluster randomized trial of 1,916 adults with acute hypoxemic respiratory failure found that an integrated telehealth rehabilitation program spanning the ICU, hospital ward, and 2 months after discharge improved 90-day quality of life, reduced mortality by 7.6 percentage points, and shortened mechanical ventilation by about 6 days, although quality of life among survivors was similar. The biggest win is the coordinated, workflow-integrated model: linking liberation, early rehabilitation, and postdischarge follow-up could make recovery support more consistent while reducing fragmentation across the ICU-to-home transition.
In this observational study of adults with COPD and type 2 diabetes, starting an SGLT2 inhibitor was associated with lower incident lung cancer risk than starting a sulfonylurea (HR 0.73, 95% CI 0.60–0.90), along with fewer severe COPD exacerbations and lower all-cause mortality. I think this is a promising signal for a population with high baseline risk, but it’s still an association rather than proof of a lung-cancer prevention effect, so confounding and prospective confirmation matter before changing prescribing.
This randomized trial of 131 people with fibrotic ILD found that 8 weeks of high-intensity interval training (HIIT) during pulmonary rehab improved cycle endurance and 6-minute walk distance, but did not outperform moderate-intensity continuous training; both approaches were safe and well tolerated, with no adverse events. I think the practical takeaway is that HIIT is a reasonable alternative when it fits a patient or program better, but it doesn’t justify replacing standard training or add evidence of superior benefit.
A post hoc analysis of the phase 3 BOREAS and NOTUS trials found that dupilumab consistently reduced moderate-to-severe COPD exacerbations in adults with type 2 inflammation across every season, with the largest reductions in spring and summer. I find this clinically reassuring because it suggests the benefit of targeting type 2 inflammation is durable despite seasonal triggers and supports using dupilumab based on patient phenotype rather than timing treatment around the calendar.
This GBD 2023 analysis estimates that 2.71 billion people—including 767 million children—were exposed to second-hand smoke across 204 countries and territories in 2023, contributing to 1.66 million deaths and 44.8 million DALYs globally; although exposure prevalence fell in many places since 1990, the total number exposed rose with population growth and aging. I think the key PCCM takeaway is that second-hand smoke remains a huge, preventable systems-level driver of cardiopulmonary disease, so tobacco-control efforts and routine exposure screening need to be treated as core prevention rather than an afterthought.
Bronchoscopy in severe pneumonia can materially change clinicians’ diagnostic assessments, showing that bedside gestalt is often uncertain and that invasive sampling may add useful clarity. The practical takeaway for PCCM is to treat diagnostic uncertainty as something to manage explicitly: when the result could change antimicrobials or uncover an alternative process, bronchoscopy may improve the efficiency and effectiveness of care rather than simply add another test.
This article argues that recent data revive concerns about cefepime-associated mortality, but the signal is small, inconsistent between published and unpublished trials, and difficult to separate from differences in patient populations and dosing; the practical takeaway is to avoid abandoning cefepime while staying especially vigilant about renal dosing, neurotoxicity, and prolonged courses. My take is that this is a useful stewardship reminder: in critically ill patients, getting beta-lactam dosing right and watching for neurotoxicity may matter more than treating the cefepime-versus-pip-tazo debate as a simple winner-take-all choice.
The FDA accepted United Therapeutics’ supplemental application for inhaled treprostinil (Tyvaso) in IPF after TETON-1 and TETON-2 showed slower FVC decline and a 29–33% lower risk of clinical worsening at 52 weeks, with cough the most common adverse effect; a decision is expected in April 2027. This is encouraging because it adds a lung-delivered option with a consistent phase 3 signal, but the four-times-daily nebulizer burden and tolerability will matter a lot in real-world IPF care.
This meta-analysis of 30 studies, with prospective validation in an adult primary-care cohort, found that FeNO was the most useful noninvasive biomarker for diagnosing asthma: ≥65 ppb helped rule in disease and <20 ppb helped rule it out, while blood eosinophils alone had weak accuracy and added little to FeNO. My take is that FeNO could make asthma workups more efficient, but clinicians need to account for inhaled corticosteroid exposure, which reduced diagnostic accuracy within two weeks.
Talha Burki describes how Russia’s war is unraveling tuberculosis control in Ukraine, with MSF reporting repeated attacks on health-care workers and infrastructure; Ukraine’s Ministry of Health says more than 2,500 medical facilities were damaged and 327 destroyed, including a major pediatric diagnostic and treatment center. This is a stark reminder that respiratory-care systems depend on resilient, distributed infrastructure and protected frontline teams—without those, even the best TB pathways cannot function.
What’s the biggest barrier to getting DME to patients?
Let me know your feedback!
Keeping it simple,
Jared
