A risk-minimisation measure nobody has measured is only a hope, and PRAC is now asking

The question inside the review

On 12 November, EMA's safety committee opened an Article 31 review of all injectable iron products. It will assess hypophosphataemia and osteomalacia. The third item is the one I want PV teams to read twice: whether current risk-minimisation measures work.

That is not a request for more case reports. It asks whether the tools already in place, such as product-information wording and any additional measures, change what happens to patients. Many PV systems cannot answer that. They can show that a measure was implemented. Showing that it was effective is a different and harder claim.

Implemented is not effective

Most holders keep tidy evidence of delivery: the label was updated, the communication went out, the materials were distributed. Evidence of behaviour is scarcer. Do prescribers test for the risk? Do they act on the result? Do patients at higher risk get identified? A measure that exists on paper but leaves no trace in practice is a hypothesis.

Article 31 is a referral procedure for cases where the Union's interest is at stake. The outcome can range from maintaining authorisations unchanged to varying, suspending or revoking them. I do not know where this review will land, and the records do not say. The scope, though, tells holders what the committee considers the live question. It is no longer only whether the risk exists, but whether the system around it works.

The pattern is wider than iron

The same week, FDA required application holders for every GLP-1 receptor agonist to add a Warnings and Precautions entry on a potential increased risk of NAION. A label change is where the effectiveness question starts. Once a warning is on the label, someone must be able to show whether it reaches the people who need it and changes decisions.

Regulators in both regions are signalling that a warning is an intervention, and interventions get evaluated. A PV team that treats label text as the end of the job will be exposed the next time a committee asks what the text achieved.

The strongest counter-argument

The best objection is that effectiveness measurement belongs to regulators and to class-wide work. A single holder's drug-utilisation study or prescriber survey may be underpowered, hard to compare with competitors' data, and costly. In a review covering every injectable iron product, the committee will set the lens, and holder-run studies launched now may measure the wrong thing. There is also a real risk of generating data that complicates a position without clarifying it.

I accept most of that, but it argues for modest, early work, not for none. Holders do not need a full study to know whether their own measures leave a trace. Internal evidence is cheap to assemble: complaint and case-report patterns, whether reports describe the risk being checked for, and what the existing monitoring recommendations assume about practice. A holder that has looked at this can contribute to a class review. One that has not can only react to it.

What I would do on Monday

Take every product with a measure beyond routine labelling and write down, in one paragraph, how you would know it is working. If the answer is “we distributed the materials”, you have found a gap. For injectable iron holders that gap needs closing now, with the review underway. Everyone else should do the same exercise before a referral or a new label warning forces it.