A new analysis published in Health Affairs Forefront on March 20 raises a question the BCI industry has largely avoided: even if the FDA clears an implantable brain-computer interface for clinical use, how do patients actually pay for it?

Authors Allison L. Goldring, Laura K. Findeiss, and Lee A. Fleisher argue that the Centers for Medicare & Medicaid Services has no dedicated coding, coverage, or payment framework for implantable BCIs. The technology is advancing through stroke and ALS applications — translating brain activity into control signals that restore communication and motor function — but the reimbursement infrastructure has not kept pace. CMS’s existing framework was built for conventional devices like pacemakers and cochlear implants, not for brain-computer interfaces that require ongoing software updates, algorithm retraining, and multidisciplinary clinical support.

The current state of BCI reimbursement

As things stand, there is almost no commercial coverage for BCIs in the United States. Anthem Blue Cross’s medical policy (DME.00052) classifies brain-computer interface rehabilitation devices — including the FDA-reviewed IpsiHand Upper Extremity Rehabilitation System — as “investigational and not medically necessary.” Anthem’s stated rationale is that no studies evaluating health outcomes for the IpsiHand have been published in a peer-reviewed journal, despite the device having been reviewed by the FDA. Private insurers and other public programmes frequently follow CMS’s lead on coverage decisions, so Medicare’s silence on implantable BCIs creates a downstream vacuum across the insurance landscape.

The GAO flagged the coverage gap in its technology assessment (GAO-25-106952), identifying eight policy options to address challenges across the BCI sector. Among the report’s findings: some clinical trial participants have had BCIs physically removed after studies ended because there was no funding or institutional support to maintain them.

Patients left without devices

That finding is not abstract. In the BrainGate feasibility study, which ran from 2004 to 2021, six of 14 participants underwent elective explant of their intracranial arrays at the conclusion of their participation. The average implantation duration was 872 days. Separately, the case of “Patient R” in the NeuroVista epilepsy trial became widely reported after the participant’s seizure-prediction implant — which had substantially improved her quality of life after 40 years of refractory epilepsy — was removed when the company terminated follow-up studies. Patient R initially objected to the removal, and ultimately only reluctantly consented after receiving a termination order from the manufacturer and being advised of the safety risks of retaining an unsupported device.

These cases illustrate a structural problem: clinical trial protocols typically include provisions for device removal but not for ongoing maintenance after the trial concludes. If a developer ceases operations or a trial ends without a path to commercial access, participants lose the benefits of their implants regardless of clinical outcomes.

The FDA pathway exists. The CMS pathway does not.

On the regulatory side, the FDA has been proactive. The agency published dedicated guidance for implanted BCI devices in 2021, outlining non-clinical testing requirements and clinical study design for patients with paralysis or amputation. Since then, the FDA has granted investigational device exemptions to Neuralink, Paradromics, and others, and issued 510(k) clearance to Precision Neuroscience’s Layer 7 Cortical Interface for implantation periods of up to 30 days. The regulatory pathway to market, while rigorous, is well-defined.

The reimbursement pathway is not. The GAO report noted that BCI developers found it difficult to engage with CMS about their devices. CMS has since designated a specific point of contact to facilitate early dialogue with developers and has published improved guidance for navigating its coding, coverage, and payment processes. The Health Affairs authors argue this single contact point falls short of what the sector needs given the pace of development.

Who gets affected

Medicare beneficiaries represent a disproportionate share of the patient populations most likely to benefit from implantable BCIs. Stroke is the leading cause of long-term disability in the United States, and the majority of strokes occur in adults over 65. ALS, while rarer, overwhelmingly affects adults in their 50s and 60s. Both are demographics where Medicare is frequently the primary insurer.

Without a coverage determination, an FDA-approved BCI could reach the market with no reimbursement mechanism for the patients who need it most. Hospitals would face uncertainty about whether implantation procedures would be paid for, and clinicians would lack the billing codes to document post-implantation support — the software updates, algorithm adjustments, and rehabilitation sessions that implantable BCIs require on an ongoing basis.

What the authors recommend

The piece calls on CMS to establish a dedicated BCI coverage pathway before devices reach commercialisation. That would mean developing new procedure codes, writing coverage criteria that account for the ongoing maintenance model of neural interfaces, and creating payment structures that reflect the fact that a BCI implant is not a one-time intervention but a managed system.

The contrast with China is pointed. Hubei province has already published formal pricing guidelines for BCI procedures: roughly $900 for an invasive implant, $430 for removal. Whether those prices are sustainable is an open question, but they signal that Chinese regulators are building reimbursement infrastructure in parallel with clinical development — not waiting for the technology to arrive first.