A global value dossier (GVD) is the master evidence and argumentation document for a health technology: the single source that captures everything known about the disease, the product, its clinical and economic evidence, and the value story those facts support. Affiliate and country teams then adapt it into the local artifacts that actually win access: HTA submissions, formulary dossiers, payer presentations, pricing negotiations.
Despite how central the GVD is to market access, most explanations of it stay abstract. This article does the opposite: it explains what a GVD is for, walks through its structure section by section with what each section must contain and the mistakes that weaken it, shows how the 2026 European assessment landscape changes what a good GVD looks like, and closes with a worked example: an annotated skeleton for a hypothetical product that you can pattern your own dossier against.
What a GVD is, and what it is not
The GVD is an internal strategic document. Payers never see it. Its readers are your own market access, medical affairs, and affiliate teams, and its job is to arm them: every claim they will make locally, matched to the evidence that substantiates it, with the counterarguments anticipated.
That purpose separates it from three documents it is often confused with. It differs from a regulatory dossier, which demonstrates safety, efficacy, and quality to a licensing authority against statutory requirements. It differs from a local HTA submission, which answers one agency's template with one country's comparators, costs, and population. And it differs from a promotional slide deck, because a GVD that reads like marketing fails at its actual job: affiliate teams need the unvarnished evidence base, including its weaknesses, to prepare for the scrutiny their submissions will face.
A useful mental model is a hub and spokes. The GVD is the hub, written once and maintained centrally. Each local submission is a spoke: it draws facts, arguments, and evidence tables from the hub and adds the local layer of comparators, epidemiology, costs, and template formatting. Done well, this saves each affiliate hundreds of hours and, more importantly, keeps the global value story consistent while every country tells it in its own required format.
When to build it
The GVD's value peaks at launch, and its content decisions reach back into development. The strongest practice is to draft a first GVD around phase 2 to phase 3 transition, when evidence gaps identified in dossier drafting can still influence the evidence generation plan: a missing comparison, a patient-reported outcome payers will ask about, a subpopulation analysis. A GVD first drafted after pivotal readout is limited to describing whatever evidence the program happened to generate, while one drafted earlier gives the team a chance to close gaps while trials can still be designed or amended. It is then updated at each major milestone: pivotal data, regulatory submission, approval, launch, and on a maintenance cycle afterward as new data and new competitors arrive.
The structure, section by section
Formats vary across companies and consultancies, and the underlying anatomy is consistent. A complete GVD contains the following, typically in this order.
1. Executive summary and value proposition
One to three pages stating the value proposition and the core value messages, each cross-referenced to the section that substantiates it. Write this last, and hold it to one rule: no message appears here that does not have an evidence trail inside the dossier. The value proposition itself should name the population, the comparator context, the clinical benefit, the humanistic benefit where genuine, and the economic consequence, in language precise enough that a reader could challenge it. A vague value proposition at the top propagates vagueness into every submission built from it.
2. Disease overview and burden of illness
The clinical description of the disease, its epidemiology (incidence, prevalence, and how they vary across the launch geographies), its natural history, and its burden: clinical burden in morbidity and mortality, humanistic burden in quality of life and caregiver impact, and economic burden in direct and indirect costs. This section earns its length only when every element connects forward to the value story. Epidemiology feeds budget impact models. Burden data substantiates unmet need. Natural history frames what disease progression costs, which is what prevention or delay is worth. Burden sections written as literature tourism, comprehensive and unconnected, are the most common form of GVD bloat.
3. Unmet need and current treatment landscape
What treatments exist, how treatment pathways actually run, what guidelines recommend, and where current options fail: efficacy ceilings, tolerability problems, adherence barriers, populations left behind. This section defines the gap the product claims to fill, so it must be written with an honesty that survives payer scrutiny: overstating unmet need in a crowded class is one of the fastest ways for a dossier, and the submissions built on it, to lose credibility.
4. Product profile
Mechanism of action, posology and administration, the regulatory status and label (and differences between labels across regions, which local teams need flagged), and the product's place in the treatment pathway. Where administration itself carries value, such as an oral option in an injectable class or a once-weekly regimen replacing daily dosing, this is where that story starts, to be substantiated later with adherence and preference evidence.
5. Clinical evidence
The heart of the dossier: the trial program presented study by study, with design, population, endpoints, results, and critical appraisal, followed by the comparative evidence picture. Three practices distinguish a strong clinical section.
Present the totality, including unfavorable findings, failed endpoints, and safety signals, each with context and, where legitimate, mitigation. An affiliate team that first meets a negative finding in an assessor's critique, rather than in its own dossier, has been left unprepared by the document whose purpose was preparation.
Build the comparative picture deliberately. Payers assess relative effectiveness, and where head-to-head trials against the locally relevant comparator do not exist, the dossier must carry the indirect evidence: network meta-analyses or other indirect treatment comparisons, presented with their methodology, assumptions, and limitations stated plainly. Because comparators differ by market, a good GVD often maintains a comparator-by-market matrix showing which comparison each affiliate will need and what evidence supports it.
Pre-empt the methodological critique. Every HTA agency will appraise the evidence critically; the GVD should do it first, stating the limitations and the best available responses, so local teams negotiate from prepared positions.
6. Economic evidence and models
The economic case: the cost-effectiveness or cost-utility analysis with its structure, inputs, assumptions, and results; the budget impact model; and the health state utility values and cost inputs that populate both. The global models are built for adaptation, so this section should document what is core and what is locally adaptable: model structure and clinical inputs usually travel, while costs, comparators, discount rates, and sometimes utilities are swapped locally. Include the sensitivity analyses prominently, because the parameters the results are most sensitive to are where local pricing and access negotiations will concentrate.
7. Humanistic evidence: PROs, preference, and real-world data
Patient-reported outcomes from the trial program, health-related quality of life evidence, patient and clinician preference studies, and real-world evidence where it exists. This section has grown in weight: assessment bodies increasingly expect patient-relevant outcomes alongside clinical endpoints, and real-world data increasingly carries part of the comparative argument, particularly post-launch. Present each source with the same critical appraisal standard as the trial data.
8. Value messages and the evidence map
The distilled output: the value messages, usually organized by audience or by theme (clinical, humanistic, economic), each stated in one sentence and mapped to its substantiating evidence with references. This is the section local teams use daily, and its quality test is mechanical: a reviewer should be able to take any message, follow its references, and find that the evidence says what the message claims. A message that outruns its evidence will be caught locally, in front of a payer, at the worst available moment.
9. Objection handling
The anticipated payer objections, stated as sharply as a skeptical assessor would state them, each with the best evidence-based response. Weak dossiers skip this section or soften the objections. Strong ones write the objection column in the voice of IQWiG, NICE, or HAS, since assessors of that caliber will eventually raise the same points in earnest.
10. Local adaptation toolkit
Guidance for the spoke-building: which sections map to which HTA templates, what must be localized (epidemiology, comparators, costs, pathways), the evidence hierarchy rules for markets with strict requirements, and contact points for the global model and data requests. Some organizations split this into a separate document; wherever it lives, its absence is why affiliates rebuild from scratch and value stories drift.
The 2026 layer: what EU joint clinical assessment changes
Since 12 January 2025, new cancer medicines and advanced therapy medicinal products entering the EU centralized procedure undergo a joint clinical assessment (JCA) under the HTA Regulation (EU) 2021/2282: one EU-level assessment of relative clinical effectiveness that all member states receive and must give due consideration in their national processes. The scope widens on a fixed schedule: orphan medicines join in January 2028 and all new medicinal products in 2030, and the framework now extends toward devices, with an October 2025 implementing regulation setting procedural rules for JCAs of selected high-risk medical devices and IVDs. The system is no longer theoretical: the first JCA report was endorsed in April 2026, for the pediatric oncology product tovorafenib.
For the GVD, the JCA changes three things.
First, timing. The JCA dossier is prepared in parallel with the EMA marketing authorization application, with information submitted to the HTA secretariat alongside the EMA filing. A GVD that historically matured between approval and launch now needs its clinical evidence, comparative analyses, and objection handling mature at submission, roughly a year earlier than older planning assumed.
Second, the PICO problem. The JCA scope is defined through PICO questions (population, intervention, comparator, outcomes) consolidated from member state input, and a single product can draw many PICOs spanning different comparators and subpopulations. The GVD's comparator-by-market matrix and its indirect comparison evidence stop being background assets and become the direct feedstock for the JCA submission. A dossier that already maintains this matrix arrives at JCA scoping with most of the work done, while a dossier built around a single comparator story for the whole of Europe leaves the team assembling multi-comparator evidence under procedural deadlines.
Third, division of labor. The JCA covers relative clinical effectiveness at EU level, while economic evaluation, pricing, and reimbursement remain strictly national. Two rules govern the interface: member states must give due consideration to the JCA report in their national processes, and they may not re-request information already submitted at EU level, which makes the JCA dossier's content decisions binding on the whole European campaign. The GVD therefore does not shrink: its clinical core feeds the JCA once, and its economic and adaptation layers feed the same national processes as before, now anchored to a shared clinical assessment that every national submission must be consistent with. Consistency management across the hub and spokes, always good practice, is now a compliance property.
A worked example: annotated skeleton for a hypothetical product
To make the structure concrete, here is how the sections take shape for a hypothetical product: an oral, once-daily disease-modifying therapy, "Product X," for a chronic progressive condition currently managed with an injectable biologic and generic symptomatic therapy.
Executive summary. Value proposition: in adults with the condition, Product X delays progression comparably to the injectable standard of care, with oral once-daily administration, a lower discontinuation rate, and a cost profile that offsets drug cost through reduced administration and monitoring burden. Five value messages follow, each with a section reference.
Disease overview. Prevalence and incidence with ranges across launch markets; progression described in stages with the annual cost of each stage, because the economic model's health states mirror them; quality of life by stage from published utility studies; caregiver burden quantified for the later stages.
Unmet need. Guideline pathway diagram; evidence that a documented share of patients discontinue or underdose the injectable comparator, with adherence data; the resulting progression outcomes in sub-optimally treated patients. The unmet need claimed is specific: an efficacious option patients can stay on, and no broader.
Product profile. Mechanism in one page; label summary with the EU/US label differences flagged for local teams; position claimed in the pathway: alternative first-line disease-modifying option, and explicitly not claimed as superior efficacy, because the evidence will not carry that claim.
Clinical evidence. The two pivotal trials in full, including the secondary endpoint that missed significance, with its confidence interval and the honest discussion of what can and cannot be said. The head-to-head trial used the injectable comparator relevant to most, but not all, markets; for the markets where a different biologic dominates, the network meta-analysis is presented with its assumptions, heterogeneity assessment, and the sensitivity of results to one older trial, stated before any assessor states it.
Economic evidence. A cost-utility model whose health states follow the disease stages from section 2; base case results per market archetype; the budget impact model showing the offset structure (drug cost up, administration, monitoring, and progression costs down); one-way sensitivity analysis showing results are driven by the progression delay estimate and the comparator discontinuation rate, which tells every local team where their negotiation will live.
Humanistic evidence. Trial PRO results on fatigue and treatment burden; a discrete choice experiment showing patient preference for oral administration and the strength of that preference; early real-world persistence data from the first launch market, appraised with its limitations.
Value messages. Twelve messages in three groups (clinical, patient, economic), each one sentence, each mapped to evidence. Example, economic group: "In the base case analysis, Product X was cost-effective against [comparator] at conventional thresholds, driven by reduced administration and progression costs (section 6.3, tables 6.4 to 6.7)."
Objection handling. Eleven objections including: the missed secondary endpoint; the NMA's reliance on the older trial; the absence of long-term progression data beyond the trial horizon and the PMPF-style registry commitment that answers it; the price premium against the incumbent's expected biosimilar erosion. Each with a response, and where the accurate response is a commitment rather than a rebuttal, it says so.
Local adaptation toolkit. Mapping tables to the major HTA templates; the JCA-facing extract of the clinical section for the EU; the list of locally required inputs with sources suggested per market; version control and the update calendar.
The point of the example is the connective tissue: disease stages that become model health states, adherence evidence that becomes both an unmet need argument and a model input, weaknesses that appear once in the clinical section and are answered once in objection handling. These connections are what turn ten chapters into a single argument, and their absence is usually visible within the first ten pages of a weak dossier.
Frequently asked questions
How long should a GVD be? Common practice lands between 100 and 300 pages plus appendices, and length is the wrong target. The dossier should run only as long as the substantiated value story requires. The executive summary and value message sections must stay short enough to be used daily; the evidence sections behind them carry the depth.
Who writes it? The core team is HEOR, market access, and medical affairs, with clinical development and biostatistics contributing, and affiliates consulted early on comparators and local evidence needs. Many companies partner with a specialist writing group for the drafting and evidence synthesis while retaining strategy internally. The failure mode to avoid is a dossier written by one function alone: purely medical GVDs miss the economic architecture, and purely access-led ones lose evidential discipline.
How does a GVD differ from an AMCP dossier? The AMCP Format dossier is a US-specific artifact: a standardized evidence dossier provided to US payers and pharmacy benefit committees on request, following the AMCP Format's prescribed structure. In hub-and-spoke terms it is a spoke, and a well-built GVD should generate it quickly.
Do devices and diagnostics need GVDs? Increasingly, yes. Reimbursement for high-value devices and diagnostics runs through the same evidence logic, and with high-risk devices and IVDs entering the EU joint assessment framework, structured value documentation is moving from good practice toward necessity in that sector too.
How often should it be updated? On event and on calendar: at every major data readout, approval, or competitive entry, and on a scheduled review at least annually, with version control that lets affiliates know what changed. An outdated GVD quietly propagates outdated claims into live submissions, which is worse than having none, because it carries authority.
What if our comparative evidence is limited? This is common, especially for products approved on single-arm trials or against placebo in classes with established comparators. The dossier's job is then to assemble the best defensible comparative picture: indirect comparisons, real-world evidence, and transparent argumentation about what the evidence can support. We are preparing a dedicated article on that specific problem, preparing value dossiers and HTA submissions with limited comparative effectiveness data, as a follow-up to this one.
How EvySaif builds value dossiers
EvySaif Research and Medical Affairs Solutions develops global value dossiers and their spokes: evidence synthesis and systematic literature reviews, network meta-analyses and indirect treatment comparisons, cost-utility and budget impact models built for local adaptation, value message development with evidence mapping, and the local submissions themselves across Europe, MENA, and India. Our teams combine HEOR modeling, medical writing, and regulatory experience, so the dossier's clinical, economic, and strategic layers are built to agree with each other.
If you are scoping a first GVD, refreshing one for a new indication, or preparing for a JCA-era European launch, write to info@evysaif.com or use the contact page.
This article reflects the position as of 12 August 2026, including the EU HTA Regulation (EU) 2021/2282 JCA rollout status. Practices and requirements evolve; verify current expectations for your product and markets before acting.