Could GLP-1 obesity drugs break pharma’s safety infrastructure?

GLP-1 agonists like Ozempic and Wegovy are a pharma industry dream: mega-blockbuster products that are visibly transforming patient outcomes, particularly now in the context of treating obesity.

But they also present an operational nightmare for companies, as volumes of adverse events overwhelm traditional safety workflows.

Glucagon-like peptide-1 (GLP-1) receptor agonists such as Ozempic and Wegovy now account for more than 7% of all prescriptions in the US, reflecting their rapid adoption beyond diabetes care – particularly in weight management.

The success of GLP-1 agents, which has yet to peak, has created a tidal wave of new safety work for drug developers, marketing authorisation holders and industry regulators, as the products are taken up in ever higher numbers on an international scale.

Eli Lilly’s GLP-1 drugs (including Mounjaro and Zepbound) reportedly generated c.$39.5 billion in revenue in the first nine months of 2025, making them among the best-selling medicines globally.

Traditional pharmacovigilance (PV) systems are struggling to cope with the sheer volume and complexity of adverse event reports generated by these blockbuster products.

In the case of injectables (combination products), there is the further complication of possible device failures.

These can trigger complaints that are peripheral to the medication itself, adding to the PV workload and potentially clouding drug-related safety insights.

These issues pose genuine concerns for the pharma industry, particularly as post-marketing surveillance around GLP-1 products is currently such an active area of research.

Real-world signals currently being closely monitored include those around pancreatitis and ocular safety.

The GLP-1 phenomenon highlights how quickly pharmaceutical infrastructure can become overwhelmed.

This is not a failure of diligence, or of commitment to patient safety, but rather the limitations of legacy systems and approaches to case intake and processing.

Traditional system-supported PV workflows, which involve teams being scaled up to cope with extra demand, were built around relatively predictable patterns of adverse event reporting, primarily via healthcare professionals.

They also assume clear distinctions between drug-related adverse events and device-related product complaints.

GLP-1 products have challenged these parameters. The products have been taken up quickly and widely, and real-world side effects are only now emerging.

These are being reported directly by patients, recorded by frontline care providers and pharmacists, fed back by drug sales teams, and captured as part of patient support programmes (PSPs).

Information is being captured in a multitude of ways, across an array of different channels and sources – some of which may be duplicated.
In among all of this, immediate opportunities to ask key follow-up questions may be missed.

This diversity and inconsistency in reporting is leading to delays in drawing critical insights together, with implications both for regulatory compliance (critical signals must be reported swiftly) and ultimately for patient safety.

On top of these issues are device considerations.

Injectable GLP-1s are combination products, administered by patients themselves without medical supervision.

During reporting potential issues, for instance via a physician, pharmacist or PSP contact, clarity around the source of the problem may be lost.
Was it a faulty injector, or the drug itself?

While the patient may simply want to register a complaint, the pharma company needs more detail to fulfil its regulatory reporting obligations, contain risk and keep patients safe.

Any disconnect could place additional pressure on case intake and management processes, as PV teams strive to correctly capture, classify and route reports promptly, while maintaining data quality.

The limitations of manual PV processes become particularly acute at scale.
Take a single adverse event case that requires assessment and distribution to multiple regulatory bodies.

Each distribution requires careful data entry, quality checking and regulatory submission within strict timelines.

Multiply this by thousands of cases monthly, and long-standing manual approaches soon come undone. The follow-up challenge (capturing a fuller history from the patient), which has long existed, is magnified in the context of GLP-1s.

When patients first volunteer feedback about an adverse event, this is the drug company’s ‘golden moment’ to capture all of the detail that it as well as regulators will need to properly assess and record a safety signal.

If follow-up questions aren’t asked at the time (where possible), the opportunity to capture full and rich detail – which may be important for downstream analysis – risks being lost forever.
Anecdotally, pharmaceutical companies report success rates hovering at around only 10% where attempts to obtain additional information are made at a later point.
In the context of GLP-1s for weight loss, and their mass-scale take-up around the world, reliance on after-the-fact gap-filling is both impractical and risky from a safety perspective.

The emerging challenges with safety intake are not unique to GLP-1 products.

The trend towards combination products is accelerating across multiple therapeutic areas as companies seek to improve outcomes, enhance intellectual property and differentiate offerings.

The shift towards patient self-administration is irreversible as well.
The convenience it offers patients is too valuable, and healthcare systems under pressure increasingly support moving routine medication administration into patients’ homes and daily routines.

Incremental improvements to existing processes are no longer the answer.
This is a challenge requiring a fundamental rethink of how safety data is captured at source.

The more ambitious the therapies being rolled out to patients, the more critical it is that genuine signals can be identified, qualified and circulated swiftly.
‘Smart intake’ approaches are gaining traction as a result.

These involve using optimised, intuitive systems that have been designed to capture safety data directly from reporters in structured formats.

This eliminates subsequent manual data entry, while also ensuring appropriate routing between adverse event and product complaint pathways (e.g. to support safety reporting for combination products).

Thanks to these efficiencies, scalability is built in.

The impact of GLP-1s on PV workloads is indicative of what is to come, as patient therapies continue to advance.

Pharma companies can no longer expect to address soaring case volumes and rising complexity by simply increasing their PV headcount.

To ensure sustainability, they will need to adopt optimised safety intake process automation, so that they can accelerate and improve routine safety data entry, while ensuring high-quality structured capture – at the point that patients or professionals are ready to report.

This will also liberate PV professionals to focus on the critical activities of scientific assessment and signal detection.

Companies that continue to rely on increased team sizes to handle peaks in case volumes risk being perpetually overwhelmed – potentially missing critical safety signals and falling foul of regulatory timelines.

In many ways, this industry’s ability to keep bringing innovative therapies to patients, while maintaining robust safety surveillance, depends on the choices that companies make now.