Products Liability Insurance for Medical Devices

Products liability is the largest single coverage exposure on most medical device manufacturers’ balance sheets. A single adverse event involving a Class II or Class III device can produce a defense and indemnity profile that no other line on the program comes close to matching. The exposure is structural, the wording is technical, and the underwriting conversation is closer to a regulatory dialogue than to a typical commercial insurance placement.

What makes medical device products liability different from generic product liability is the layered framework around it. Three classical defect theories (manufacturing, design, failure to warn) overlay on a federal regulatory regime (FDA classification, QMSR-aligned quality systems, MDR reporting), which then overlay on a state-by-state tort landscape with widely varying statutes of repose and preemption postures. The placement has to account for all three layers at once.

This walks through what products liability actually covers, how device class and indication drive the underwriting, the three defect theories and what they mean for policy wording, the distribution and vendor dimensions, where products interacts with recall and clinical trial coverage, and the founder mistakes that surface most consistently at claim.

Why Medical Device Products Liability Is Structurally Different

Three structural factors set the line apart from any other commercial products liability placement.

The exposure is long-tail. Medical devices are not consumed and discarded. Implantables remain in patients for years or decades. Multi-year-use devices sit in active service well past the policy period in which they were sold. State statutes of repose vary from 10 years to indefinite. The claim that arrives in 2026 may attach to a unit sold in 2014. Occurrence-based products liability is the standard wording precisely because the alternative (claims-made) would leave manufacturers with a coverage gap they cannot solve with retroactive dates.

The duty of care is layered. Generic product manufacturers owe a reasonable-care duty to the end user. Medical device manufacturers owe a duty that flows through the design control framework, the quality management system, the post-market surveillance obligations, and (for prescription devices) the learned intermediary doctrine that routes the warning duty through the physician. A products liability defense for a medical device is fundamentally a regulatory defense as much as it is a tort defense.

Federal preemption is class-dependent. Riegel v. Medtronic established that Class III PMA devices receive federal preemption of most state-law tort claims under 21 U.S.C. § 360k(a). Class II 510(k)-cleared devices generally do not. The same product defect, alleged against a PMA device versus a 510(k) device, produces materially different defense postures and coverage outcomes. The insurance conversation has to begin with classification.

The Coverage Trigger and What Products Actually Responds To

Products liability is third-party coverage. It responds when a third party (a patient, a hospital, a provider, a downstream user) alleges bodily injury or property damage arising from a defect in the manufacturer’s product. The policy provides defense and, where the manufacturer is liable, indemnity for the alleged harm.

The standard form for medical devices is occurrence-based. The relevant policy is the one in effect at the time of the bodily injury, not the time the claim is reported or the policy purchased. For an implantable that fails 8 years post-implant, the responsive policy is the one in effect during the period of injury. This is structurally important: the manufacturer needs continuous coverage with no gaps over the entire field life of the product, because any gap creates an uninsured period for events that occurred during that window. The full trigger picture across products and the claims-made lines is covered in occurrence versus claims made for a medical device company.

What products liability does not cover:

• First-party costs of pulling product back from the field. That is recall coverage, a separate line.
• Pure economic loss without bodily injury or property damage. Loss-of-use claims and revenue claims that do not attach to physical harm sit outside the standard products grant.
• Claims arising from professional services rendered by the manufacturer (clinical training, application support beyond product use, software-as-service offerings). Those routes through Tech E&O or medical professional liability depending on structure.
• Cyber-rooted bodily injury where the policy contains a cyber exclusion. Section 524B cyber device obligations have made this an active wording area for connected devices.

How Indication and Device Class Drive Underwriting

Underwriters do not price medical device products liability as a single category. They price by indication, class, and patient population.

Class I devices (low risk, mostly exempt from premarket review) sit at the lower end of the underwriting spectrum but are not de minimis. A Class I device with broad consumer distribution can produce volume-driven exposure even with low per-unit severity.

Class II devices (510(k) cleared, moderate risk) cover the largest portion of the device market. Underwriting attention varies by indication: a non-invasive monitor, a surgical instrument, an implanted Class II device, and a software-bearing Class II device all sit at different points on the severity curve. The 510(k) clearance does not establish a safety record; it establishes substantial equivalence to a predicate. Underwriters evaluate the underlying safety profile separately.

Class III devices (PMA, highest risk) carry the highest per-claim severity but receive federal preemption protection for most state-law claims involving the device’s safety and effectiveness. Underwriting attention focuses on areas preemption does not reach: manufacturing defect claims, off-label use, post-approval study compliance, and any deviation from the PMA-approved labeling.

Indication is the dominant variable within class. A Class II device for diagnostic imaging carries a different profile than a Class II implantable orthopedic device, which carries a different profile than a Class II infusion pump. The underwriting conversation begins with the FDA classification but is driven by what the device does, where it sits in the patient care pathway, and what the consequences of failure look like.

Patient population matters at the same level. Devices used on pediatric, elderly, or critically ill populations attract higher attention than equivalent devices on healthy adult populations.

The Three Defect Theories and Why They Matter for Wording

US tort law recognizes three classical defect theories. The policy wording should respond cleanly to all three.

Manufacturing defect. The unit deviated from the manufacturer’s intended specification. The design is sound, the labeling is adequate, but this particular unit (or this lot, or this batch) was made wrong. Manufacturing defect claims are typically the most clearly covered under products liability, and the QMSR-aligned quality management system documentation is the manufacturer’s primary defense. Process validation, lot release records, and CAPA history determine whether the claim succeeds at trial.

Design defect. The unit performed as designed, but the design itself was unreasonably dangerous. Design defect cases are harder to defend because they attack the underlying engineering judgment, and they typically involve multiple plaintiffs (every unit shares the design). For Class III PMA devices, Riegel preemption substantially limits state-law design defect claims. For 510(k) Class II devices, design defect remains the dominant claim theory.

Failure to warn (marketing defect). The labeling, instructions for use, or post-sale communications did not adequately warn the user of a known or knowable risk. For prescription medical devices, the learned intermediary doctrine routes the duty to warn through the prescribing physician, which structurally reshapes the defense. Failure-to-warn claims attach to the labeling history, the post-market surveillance program, and the timeliness of any safety communications. Where post-market evidence accumulates and the labeling is not updated, failure-to-warn exposure compounds.

Policy wording variation across the three theories is meaningful. Some forms grant broad coverage for all three; others sub-limit or exclude design defect claims for specified product categories; still others exclude punitive damages associated with willful misconduct. Reading wording against the three theories is part of the placement review.

Distribution, Geography, and Vendor Risk

The geographic and channel structure of distribution shapes the underwriting and the policy wording.

Domestic distribution. US distribution is the baseline. State-by-state variation in statutes of limitations, statutes of repose, comparative fault rules, and punitive damages availability creates a heterogeneous exposure even within US-only operations. The placement should reflect the states where the device is sold and used, not just the manufacturer’s headquarters.

International distribution. Outside the US, the EU MDR regime introduces parallel post-market surveillance and vigilance obligations under Articles 87 to 91. Reporting timelines run to 2 days for serious public health threats, 10 days for death or unanticipated serious deterioration, and 15 days for serious incidents. The MIR form 7.3.1 became mandatory in May 2026 with EUDAMED-aligned data structures. International coverage extensions, jurisdiction language, and choice-of-law provisions in the policy materially affect whether a non-US claim is defended cleanly. If the device sells into Europe, the program has to reach further, because the EU MDR adds jurisdiction exposure a US policy may not answer.

Distributors and downstream resellers as additional insureds. Distribution agreements routinely require additional insured status for the distributor on the manufacturer’s products liability policy. The wording on additional insured grants varies: some forms grant broad AI status for the distributor’s vicarious liability; others limit AI grants to specified events. Distribution agreements should be drafted against the actual AI grant the manufacturer’s policy provides, not against generic AI language. Signing a distributor adds specific insurance requirements, covered in what insurance you need before signing a distribution agreement. Shipping the device is a narrower question, since commercial auto does not cover a delivered device that later causes harm.

Contract manufacturers and component suppliers. When the manufacturing process is outsourced, products liability still attaches to the manufacturer named on the 510(k) or PMA. The risk transfer from the manufacturer to the CMO via contractual indemnification is a contract law mechanism, not an insurance mechanism. The CMO’s own products liability policy, naming the manufacturer as an additional insured, is the operational backstop. Single-source CMO arrangements without confirmed AI status on the CMO’s policy are a structural gap.

How Products Liability Coordinates with Recall and Clinical Trial Coverage

Products liability does not stand alone in a medical device program.

Recall. Recall is the first-party coverage for the cost of pulling product from the field after a quality or safety event. A single field action can produce both a recall response (first-party retrieval, notification, business interruption) and a products liability response (third-party bodily injury or property damage claims arising from units used before retrieval). The two policies coordinate but do not substitute for each other. A program with strong products limits but a thin recall component leaves the manufacturer exposed on the first-party costs, and the typical recall coverage gaps surface at claim when sub-limits and triggers were not engineered against the actual exposure.

Clinical trial liability. Pre-commercial coverage for products under investigation routes through a separate sponsor-side clinical trial liability policy with IRB-aligned wording. The transition from clinical to commercial requires products liability coverage to attach at first commercial shipment with appropriate retroactive dates relative to the clinical trial coverage period. Coverage gaps at the clinical-to-commercial seam are a frequent finding in diligence reviews. Before commercial product liability attaches, the device may be in a study, where a clinical trial needs its own coverage in the program.

Cyber. For connected and software-bearing devices, the Section 524B cybersecurity obligations and the increasing frequency of cyber-rooted device incidents make the products-cyber interface an active wording question. Some products policies contain broad cyber exclusions; others contemplate bodily injury arising from cyber events. The dovetail with the cyber policy needs to be reviewed against the specific device architecture.

What Founders Most Often Get Wrong

Assuming the limit scales with the company. Products liability limits are driven by the exposure profile of the device, not by the size of the company. A 25-employee company with a Class III implantable is not appropriately covered at the same limit as a 25-employee company selling a Class I monitor. Stage-of-funding heuristics fail at the products liability line.

Treating 510(k) clearance as a safety record. Clearance establishes regulatory permission to market, not actuarial safety. Underwriting reviews the underlying safety profile separately, and so do plaintiffs at trial.

Buying only on price during pre-launch. A products program assembled at the lowest premium without attention to the wording on additional insureds, vendors, international coverage, and the three defect theories produces a placement that performs poorly when the first claim arrives. Wording matters more than headline limit at this line.

Letting the program lapse between clinical and commercial. The seam between clinical trial liability and commercial products liability is where coverage gaps most often form. The retroactive date on the products policy must align with the latest clinical trial activity period for the cleared indication.

Underestimating the long-tail. A product retired from the market in 2024 can still produce claims through 2034 or later depending on jurisdiction and discovery rule. Tail coverage decisions cannot be deferred until the company is wound down. And for a device whose algorithm keeps changing after clearance, confirm the policy covers the version in the field, not only the one you cleared.

A Note on Placement

MedTech Coverage works with medical device manufacturers on products liability programs structured around FDA classification, indication-specific risk profile, distribution footprint, and the wording dimensions that determine how the policy performs at claim. Coverage is placed through Tower Street Insurance’s appointments with the specialty life sciences markets that underwrite medical device products liability.

If a products liability placement is being assembled before a 510(k) or PMA commercialization, restructured for international distribution, or evaluated against an acquisition due diligence process, a structured coverage review produces a working document calibrated to the device’s actual classification, indication, distribution geography, and the post-market obligations the manufacturer carries.