Europe Created a Joint Assessment to Reduce National Methodology. Every Agency Answered by Building More of It.

On 23 July, France’s health authority renamed a department. The real-world data coordination cell it had created in 2021 became the Mission Méthodes, Europe, Données, folded into a strategic expertise pole inside its evaluation directorate. As announcements go, it is the sort of thing that gets filed under administration and forgotten by lunchtime.

Read the justification, though, and it says something the harmonisation project has never admitted. The HAS explains that the European work has produced a significant increase in workload, requiring the organisation to adapt and reinforce its expertise. And it states the ambition plainly: to strengthen its methodological influence within European assessment networks, and to contribute to building the standards that will structure access to innovation in France and in Europe.

That is not a national agency absorbing a European system. That is a national agency positioning to write it.

And France is not the outlier. It is the latest.

The Joint Clinical Assessment was sold on a simple proposition: one clinical assessment instead of twenty-seven, less duplicated effort, more efficiency. Eighteen months into implementation, look at what the member states actually did.

Germany amended its benefit assessment ordinance in March 2025, building a referencing solution and a formal delta-dossier mechanism to bolt the European report onto AMNOG. The Netherlands ran a continuous gap analysis of its entire national process, constructed a project-governance structure around implementation, developed new procedures and templates, and hired dedicated JCA assessors — while identifying capacity, resources and financing as open problems. Belgium completed its implementation guidance alongside the third edition of its economic evaluation guidelines. Italy stood up a scientific and economic committee and a dedicated working group. Ireland’s national centre expects around twenty-five joint assessments and ten scientific consultations a year, and its chief pharmacist has said publicly that finding the experts to conduct them is a key challenge.

And the European level is funding the shortfall it created: a capacity-building programme aimed at training some two hundred assessors, because there are not enough. The head of the Commission’s own HTA unit has described the state of the system as a start-up — this, after two decades of preparatory cooperation.

Not one agency in Europe responded to the arrival of joint assessment by doing less methodological work.

The volume explains part of it. The first year produced ten joint assessments and seven scientific consultations. The 2026 work programme scales that towards fifty, with as many as one hundred and one technologies potentially eligible. Orphan medicines enter scope in 2028, everything else by 2030. A tenfold increase in assessments does not reduce the demand for assessors; it multiplies it.

But volume is the boring half of the explanation. The interesting half is what the agencies are actually building capacity for.

Here is the mechanism nobody says out loud.

Before the regulation, a national agency’s methodology governed one country. IQWiG’s scepticism about indirect comparisons bound Germany. The HAS’s doctrine on real-world data bound France. Each was sovereign at home and irrelevant abroad, and the sensible level of investment in methodological doctrine was calibrated accordingly.

The Joint Clinical Assessment changed the denominator. A method adopted at European level now conditions the evidence requirements facing every developer selling into twenty-seven markets. The prize for authorship went up by more than an order of magnitude — and it went up for every agency simultaneously.

So they invested. Rationally, and in exactly the way anyone would predict.

We already know who got there first. German bodies took a leading role in developing the JCA methodology; the guidance on indirect treatment comparisons substantially reflects the approach IQWiG and the G-BA have used for years — an approach more demanding than most other national systems. The submission template itself, developed under the leadership of the French, German and Irish agencies, ended up structurally resembling the German national dossier. As we have written before, the European standard did not emerge from a neutral methodological consensus. It arrived pre-loaded with one country’s doctrine, because that country’s institutions were in the room, with staff, when the drafting happened.

The European method was not settled by argument. It was settled by attendance.

Which makes the MiMED announcement legible. France is not building a research function. It is building a delegation. A unit named for Methods, Europe and Data, created explicitly to strengthen methodological influence in European networks, is a diplomatic instrument wearing a scientific title — and its most obvious target is the one area where the European guidance is thinnest and French capability is strongest: the use of real-world data, where France holds one of the richest national claims systems in the world and the European level has so far been at its most restrictive.

That is a contest, not a compliance exercise. And it is now formally staffed.

The efficiency case rested on a premise that has quietly failed: that the European assessment would replace national work. It does not.

Member states retain autonomy over the decision. The scoping phase now starts earlier than the national process ever did. The joint report must be given due consideration, but the delta dossier exists precisely because the report does not answer everything a national body wants to know — and because it may not even arrive in time for the national submission. Pricing, reimbursement, and economic evaluation were never in scope to begin with. The clinical assessment moved up a level; nothing moved out.

There is decent evidence that even the shared template is not efficient on its own terms. Analysis of the German dossier requirements that shaped it found that developers now report a median of 2.6 times more adverse-event analyses than under the previous template, while IQWiG disregards around 33% of those analyses and 73% of the subgroup categories submitted. A great deal of the work being demanded is not being used — and that template was the model for Europe’s.

To be fair to everyone involved, the capacity building is real and necessary work, and a well-made shared method genuinely is better than twenty-seven inconsistent ones. Agencies participating heavily is not cynicism; you want expertise in the room. The regulation is eighteen months old and being judged against a twenty-year ambition.

But the efficiency claim should now be retired, because the evidence is in. Harmonisation did not reduce the value of national methodological capacity. It multiplied it — and every agency that could afford to respond, did.

Which leads to the part that ought to trouble the people who designed this.

If the European method is determined by which agencies have the assessors to shape it, then influence over the standard tracks institutional capacity — and institutional capacity tracks national wealth. Germany, France, the Netherlands and Ireland can build units, hire assessors, and send people to draft templates. The member states that most needed a shared assessment, because they never had the capacity to run a rigorous national one, are precisely the states least able to influence what that shared assessment now demands of them.

Harmonisation was meant to level the field. What it has produced so far is a single field, on which the largest agencies play and the smallest ones receive.

Three things follow.

First, your methodological interlocutors multiplied rather than consolidated. The instinct after January 2025 was to assume national agencies mattered less. The opposite happened: they now matter at home and in the drafting of the standard that binds everywhere. Track which bodies serve as assessors and co-assessors on which files, because that is where doctrine is actually made — not in published guidance, which follows.

Second, watch the real-world evidence question specifically. It is the largest open methodological gap at European level, it is where France has just staffed a unit, and it is where the current guidance is most restrictive. If the doctrine loosens over the next three years, it will be because a national agency with capacity argued it loose. If it does not, that is also a decision, and it will have been made the same way.

Third, budget for both layers, permanently. The delta dossier is not a transitional artefact of an immature system; it is a structural consequence of assessment moving up while decision-making stayed put. Anyone modelling European submission cost on the assumption that joint assessment will eventually collapse the national work is modelling a system that was never designed.

Europe set out to stop twenty-seven agencies doing the same methodological work twenty-seven times. It ended up creating a single method worth twenty-seven times more to control — and then watched the best-resourced agencies compete to control it. France has now said so in an organogram.

The joint assessment did not make national methodology obsolete. It made it the most valuable thing a national agency owns.

Share

Beyond Approval publishes weekly strategic intelligence on market access, HTA, pricing, and regulatory shifts that reshape how medicines reach patients.