1. What the AMCP Format Is
In the United States, decisions about which medicines a health plan will pay for are made by pharmacy and therapeutics committees at insurers, pharmacy benefit managers, hospital systems and similar organizations. AMCP calls these groups health care decision-makers. To make a coverage decision, the committee needs clinical and economic evidence about the product, and AMCP identifies the manufacturer as a principal source of that evidence.
The Academy of Managed Care Pharmacy (AMCP) first published a standard template for this evidence in 2000. The template is called the AMCP Format for Formulary Submissions, and the document a company prepares against it is called an AMCP dossier or a product dossier. Version 5.0 appeared in the Journal of Managed Care and Specialty Pharmacy in April 20241.
The AMCP Format is guidance, not a legal requirement. A company decides for itself whether to prepare a dossier and what to put in it. A dossier prepared to the AMCP Format is treated as a response to a payer's request for evidence rather than as promotional material, and this determines how and when it can be shared.
A US payer reviewing a new product asks the manufacturer for an AMCP dossier. EvySaif prepares AMCP dossiers and global value dossiers for companies in India, the Gulf and Europe that are entering the US market.
2. The Three Dossier Types
Version 5.0 describes three dossiers. They serve different stages of a product's life and are governed by different rules.
| Dossier type | When it is used | What it may say | How it reaches the payer |
|---|---|---|---|
| Unapproved product dossier | Before the product's first FDA approval | Facts only. No conclusions about safety or effectiveness. | The company may share it proactively, typically 6 to 12 months before expected approval |
| Approved product dossier | Any time after FDA approval | The full clinical and economic value case, including any off-label evidence | Only in response to an unsolicited request from the payer |
| Unapproved use dossier | While the company seeks FDA approval for a new indication of an approved product | Facts only about the new use. No conclusions about safety or effectiveness for that use. | The company may share it proactively while approval is being sought |
The three are connected over time1. A company may start with an unapproved product dossier and update it as trial results arrive. On the day of FDA approval, that dossier is retired and rewritten as the approved product dossier, which is then kept current for the rest of the product's life. If the company later files for a new indication, it may prepare a separate unapproved use dossier for that indication, and once the new use is approved, the approved product dossier absorbs it.
Before approval, the FDA has not decided whether the product is safe and effective, so the dossier reports what the trials found without characterizing the results. After approval, the dossier presents the product's value case.
3. How a Dossier Is Shared
The approved product dossier can contain information that goes beyond the FDA-approved label, including off-label studies. Because of this, the company cannot send it out on its own initiative. A payer has to ask for it first, in writing, without prompting from the company. AMCP calls this an unsolicited request, and FDA's 2011 draft guidance on unsolicited requests governs the reply4. The dossier itself should carry a statement that it is being provided in response to such a request. Appendix C of the AMCP Format gives payers a sample request letter, and payers can ask in that letter to receive updates for a fixed period such as six or twelve months.
The two pre-approval dossiers work differently. FDA's 2018 guidance on manufacturer communications with payers2 and the Pre-Approval Information Exchange Act of 2022, enacted as Section 3630 of the Consolidated Appropriations Act 20233, allow a company to share factual information about a product before approval so that payers can plan budgets. AMCP recommends that this happen 6 to 12 months before the expected approval date. In June 2026 FDA published a revised draft of the payer communications guidance that incorporates the PIE Act and will replace the 2018 version once finalized9; until then the 2018 guidance remains in effect.
AMCP also recommends that the person who presents or discusses the dossier has medical, scientific or clinical credentials, for example a PharmD, MD or PhD, rather than a sales or account role1. Dossiers should be sent electronically, and any economic model should be sent as a working spreadsheet rather than a printout.
4. The Approved Product Dossier, Section by Section
The approved product dossier has six sections. The table below lists each section, what it contains, and the page count AMCP recommends. Where AMCP gives no figure, the cell says so.
| Section | Contents | Recommended length | Maximum |
|---|---|---|---|
| 1.0B Executive summary | Clinical benefits, economic benefits, conclusions | 5 pages | 8 pages |
| 2.1B Product information | Label facts, comparison with competitor products | 5 pages | 10 pages |
| 2.2B Place of the product in therapy | Disease description, current treatments, where the product fits | 5 pages per indication | 10 pages per indication |
| 2.3B Evidence for companion diagnostic tests | Only where a test is linked to the product | 5 pages (product information), 10 pages (place in practice) | 10 and 15 pages |
| 3.0B Clinical evidence | Study summaries and evidence tables | 5 pages per study summary; under 1 page per evidence table row | 10 pages per summary; 2 pages per row |
| 4.0B Economic value and modeling report | Cost-effectiveness and budget impact models with a written report | 12 pages per model | 20 pages per model |
| 5.0B Additional supporting evidence | Guidelines, HTA reports, systematic reviews, compendia, equity, quality measures | 2 pages per study or source | 5 pages per source |
| 6.0B Appendices | References, models, prescribing information, patient information, safety data sheet | No figure given | No figure given |
The page counts are recommendations. AMCP states that a company should not add material to reach them, and that the most frequent complaint from payers is that dossiers are too long1.
Section 1.0B: The Executive Summary
The executive summary is the part of the dossier in which the company states its value case. It opens with the FDA-approved indication and a short account of efficacy and safety from the label and the trials. It then covers four clinical points: efficacy and effectiveness, comparative effectiveness against the alternatives, safety and tolerability, and the unmet need the product meets. The economic part gives the cost per unit and per course of treatment, sets that cost against the expected benefits and savings, and states at least one headline economic result, either a per-member-per-month cost or an incremental cost-effectiveness ratio. Each claim should point to the place in the full dossier that supports it, ideally by hyperlink.
Section 2.0B: Product Information and Place in Therapy
Section 2.1B is mostly label material: names, indications and approval dates, mechanism, dosage forms, contraindications, warnings, adverse events, NDC and HCPCS codes, and the wholesale acquisition cost. AMCP suggests hyperlinking to the prescribing information rather than repeating it. The section also includes a comparison table against the main competitor products, with a sentence explaining how the comparators were chosen. For a biosimilar, the comparison covers the reference product and other biosimilars, together with the evidence of biosimilarity or interchangeability1,8.
Section 2.2B describes the disease and how it is currently treated, then explains where the product fits. It covers epidemiology, clinical presentation, burden of illness, disparities in access and outcomes, and current treatment approaches. It is also where the company describes any support programs that come with the product, any post-marketing obligations such as a Risk Evaluation and Mitigation Strategy, and any evidence that the product works differently in different patient groups. AMCP asks that this section stay brief and link to Section 3.0B for the data rather than repeat it.
Section 3.0B: Clinical Evidence
This section holds the trial evidence, sorted into three tiers so that the dossier stays readable.
| Tier | What goes here | How it is presented |
|---|---|---|
| Pivotal studies and other important trials or real-world studies | Studies that carry the product's evidence base | Full study summary plus a row in the evidence table |
| Smaller or less rigorous studies that still add something | Supporting studies | Evidence table row only |
| Everything else that has been reported | Studies that add little | Listed in the bibliography only |
The section opens with a one-page overview that states the criteria used to sort studies into tiers, so that a reader can see the selection was not made to favor the product. A study summary covers the citation and trial registration number, objective, design, setting, inclusion and exclusion criteria, baseline characteristics, dropouts, treatments, endpoints with primary and secondary clearly separated, effect sizes with confidence intervals, and the limitations the authors themselves reported. Evidence tables carry the same information in compressed form, one row per study, and AMCP suggests landscape layout.
Three practical rules apply. Head-to-head trials against the main comparators should be included wherever they exist. Off-label studies should be included, listed after the on-label ones. And the section should say how well the trial populations represent the patients the payer actually covers, including where diversity data are missing and why.
For biosimilars, AMCP asks the company to state for every study whether it was run with the biosimilar itself or with the reference product, so that the payer can tell direct evidence from extrapolated evidence.
EvySaif writes the clinical evidence section from the clinical study reports and publications, with the study selection criteria documented so that the tiering can be defended; see medical affairs strategy.
Section 4.0B: The Economic Section
AMCP describes three kinds of model.
| Model | Question it answers | Typical time horizon | Main output |
|---|---|---|---|
| Cost-effectiveness model | Is the product good value for the money? | Long enough to capture all important differences in costs and outcomes; lifetime for life expectancy and QALY results | Incremental cost per QALY, per life-year, or per clinical event avoided |
| Budget impact model | Can the health plan afford the product? | 1 to 5 years | Per-member-per-month cost and total budget change |
| Financial model | What does the product do to the pharmacy budget alone? | Short | Net drug spend after discounts and rebates |
AMCP expects the first two. The financial model is optional; AMCP notes that payers have the internal resources to build such models themselves1.
For the cost-effectiveness model, AMCP asks that the analysis be built from the payer's perspective, follow the ISPOR-SMDM good practice reports7, use the simplest structure that captures the disease, discount future costs and health effects, and report clinical events, life expectancy and quality-adjusted life-years separately. Efficacy data should come from randomized trials, with real-world evidence used to adjust for adherence and practice patterns where the method is clearly explained. Expert opinion is not acceptable for the key effectiveness or safety inputs. Both one-way and probabilistic sensitivity analyses are expected, with a tornado diagram to show which inputs drive the result.
For the budget impact model, AMCP points to the ISPOR principles of good practice6. The model should start from a US population that the payer can replace with its own membership size, prevalence and cost figures. Off-label use is left out of the base case and may appear in a scenario.
A written modeling report accompanies each model. It follows a fixed order: introduction, methods, results, limitations, discussion, with a 500-word abstract on the first page that names the drivers of the result. The report should contain, at minimum, a figure of the model structure, a table of every input with its source and range, a table of assumptions, a table of disaggregated results, and a tornado diagram. AMCP recommends the CHEERS 2022 reporting standard as a further check5; EvySaif has a separate guide to CHEERS 2022.
The model itself should be an unlocked Excel workbook in which every calculation is visible, every key input can be changed, subgroups can be run, and one-way sensitivity analysis is automated. If a company will not hand over a working model, the report must say so.
For biosimilars, AMCP states that a full cost-effectiveness model is generally not required, because the biosimilar is expected to perform the same as its reference product. AMCP states that a budget impact model or a cost-minimization analysis may be more relevant1.
EvySaif builds budget impact models and cost-effectiveness models to the AMCP and ISPOR specifications, delivered as unlocked Excel workbooks with a CHEERS-compliant report.
Section 5.0B: Additional Supporting Evidence
This section holds everything that supports the product but is not a clinical trial: clinical practice guidelines (with the focus on US guidelines), health technology assessments and systematic reviews (for example from ICER, NICE or CADTH), compendia listings, published economic studies, evidence on health equity, evidence on quality measures, and anything else relevant. Each item gets a short summary of two pages or so. Evidence already summarized in Section 3.0B is not repeated here.
The equity item is new in version 5.0. AMCP asks companies to discuss barriers to equitable use of the product, such as access to specialists, affordability and stigma, and acknowledges that data on these points may be limited for a new product.
Section 6.0B: Appendices
The appendices hold the reference list, the economic models, the prescribing information, patient information such as medication guides, and the material safety data sheet. Digital therapeutics need an additional privacy and data security appendix.
5. The Two Pre-Approval Dossiers
The unapproved product dossier and the unapproved use dossier follow the same shape as the approved product dossier but are shorter and more restricted. Neither has an executive summary, because an executive summary argues value, and the company cannot argue value before approval. Each opens instead with a table of highlights giving factual milestones: trial names and completion dates, expected FDA submission and approval dates, expected launch date, dosing and administration, disease prevalence, and pricing. For an unapproved product, price is given as a band rather than a figure.
| Section | Unapproved product dossier | Unapproved use dossier |
|---|---|---|
| 1.0 Highlights and overview | 2 pages (maximum 4) | 2 pages (maximum 4) |
| 2.1 Product information | 5 pages (maximum 10) | 5 pages (maximum 10) |
| 2.2 Disease description | No figure given | 5 pages per disease (maximum 10) |
| 3.1 Study summaries | 2 pages each (maximum 5) | 2 pages each (maximum 5) |
| 3.2 Evidence table rows | Under 1 page each (maximum 2) | Under 1 page each (maximum 2) |
| 4.0 Economic information | Price band and directional cost estimates; AMCP notes that models may not be feasible before approval | Current price and any expected change; AMCP notes that models may not be feasible before approval |
| 5.0 Additional supporting evidence | Populated only in limited cases, for example with evidence from use outside the US | 2 pages per source (maximum 5) |
| 6.0 Appendices | References, basic models, safety data sheet | References, models, prescribing information, patient information, safety data sheet |
Where information does not yet exist or cannot be disclosed, the dossier says "N/A" and the company fills the gap in a later update.
6. Keeping a Dossier Current
AMCP treats the dossier as a living document. It should be revised when something significant changes: a new indication is approved or refused, a boxed warning is added, or important new clinical or economic evidence appears. Where nothing has changed, the dossier should still be checked once a year for technical accuracy, for example updated prices. Every version carries its original date and the dates of revisions and reviews, and revised sections should be marked so a payer can find what is new1.
A company does not automatically send an updated approved product dossier to a payer that received an earlier version. The payer's original request can cover updates for a stated period, and a company may send an update on its own initiative only in narrow cases, such as a serious error in the earlier version or a new boxed warning. When a product nears the end of its exclusivity, the company may stop updating the dossier, in which case the last version should be dated and the reason given to anyone who asks for it.
7. What Changed in Version 5.0
Compared with version 4.1 of 2020, version 5.0 makes five additions1.
| Change | What it means for the dossier |
|---|---|
| Pre-approval information exchange guidance | Appendix A sets out what a company should share before approval and when, in line with the 2022 PIE Act |
| Digital therapeutics | A table of highlights, a privacy and security appendix, and guidance on how a software product is described |
| Real-world evidence | Explicit place for observational and registry studies in the clinical and economic sections, with a note on their limitations |
| Health equity | A new item in Section 5.0 on barriers to fair access, and a request for trial diversity data in Section 3.0 |
| Brevity | Page counts are restated with a warning against padding, and hyperlinks are encouraged in place of repetition |
Version 5.0 also keeps the practical guidance from earlier versions: electronic submission, confidentiality agreements where needed, and the use of scientifically qualified staff to present the dossier. AMCP has since opened and closed a public comment period on a further revision of the Format, so a later version should be expected10.
8. A Note for Companies Outside the United States
For a company whose home market is India, the Gulf or Europe, three parts of the AMCP dossier require US-specific input. The economic section requires US unit costs, US practice patterns and a US payer perspective, so a model built for an Indian or European submission has to be rebuilt with US inputs. The comparator set has to be the products US payers cover, which may differ from the comparators in the pivotal trials. The clinical section has to state how well the trial population represents US patients, which AMCP lists as a required element of Section 3.0B.
EvySaif prepares AMCP dossiers as part of its HTA and payer submission service, adapting global value dossiers and existing models to the US format; the structure of a global value dossier is described in a separate guide.
9. Frequently Asked Questions
No. The AMCP Format is guidance. A company may choose not to prepare a dossier. AMCP describes the dossier as the standard document payers request from manufacturers when evaluating a product for coverage.
The payer does. For an approved product, the request must come from the payer in writing without prompting from the company. For products or uses still under FDA review, the company may share information proactively under the PIE Act.
AMCP gives page counts per section rather than a total. Adding the recommended lengths for a single-indication product with one economic model gives about 40 to 60 pages of main text before the appendices.
No. AMCP expects a cost-effectiveness model, a budget impact model, or both, depending on the product. For biosimilars, AMCP states that a budget impact model or cost-minimization analysis may be more relevant than a cost-effectiveness model.
An unlocked Microsoft Excel workbook in which the payer can see every calculation and change every key input.
For a product before approval, 6 to 12 months before the expected FDA decision. For an approved product, as soon as the label and price are available.
Yes. The format is public, and the writing can be done anywhere. The parts that need US-specific input are the cost data, the comparator set and the payer perspective in the economic models.
A global value dossier is an internal master document covering all markets. An AMCP dossier is a market-specific external document for US payers. The global value dossier supplies the source content for the AMCP dossier, which is then restructured into the AMCP sections and given US economic models.
10. Working With EvySaif on US Payer Submissions
EvySaif is a clinician-led medical writing, regulatory affairs, HEOR and drug clinical development consultancy based in Pune, India, serving pharmaceutical, biotechnology and medical device clients across India, the Middle East and Europe. For US market entry, EvySaif prepares AMCP Format 5.0 dossiers, adapts global value dossiers to the AMCP structure, builds budget impact and cost-effectiveness models as unlocked Excel workbooks with CHEERS-compliant reports, and writes the clinical evidence sections from study reports and publications. Contact EvySaif to discuss a US payer submission.
References
- Academy of Managed Care Pharmacy. AMCP Format for Formulary Submissions, Version 5.0: Guidance on submission of pre-approval and post-approval clinical and economic information and evidence. J Manag Care Spec Pharm. 2024;30(4-b Suppl):1-64. https://doi.org/10.18553/jmcp.2024.30.4-b.s1
- US Food and Drug Administration. Drug and device manufacturer communications with payors, formulary committees, and similar entities: questions and answers. Guidance for industry and review staff. June 2018. https://www.fda.gov/regulatory-information/search-fda-guidance-documents/drug-and-device-manufacturer-communications-payors-formulary-committees-and-similar-entities
- Consolidated Appropriations Act, 2023. Pub L No. 117-328, Section 3630 (Pre-Approval Information Exchange). https://www.congress.gov/bill/117th-congress/house-bill/2617/text
- US Food and Drug Administration. Responding to unsolicited requests for off-label information about prescription drugs and medical devices. Draft guidance for industry. December 2011.
- Husereau D, Drummond M, Augustovski F, et al. Consolidated Health Economic Evaluation Reporting Standards 2022 (CHEERS 2022) statement. J Manag Care Spec Pharm. 2022;28(2):146-155. https://doi.org/10.18553/jmcp.2022.28.2.146
- Sullivan SD, Mauskopf JA, Augustovski F, et al. Budget impact analysis: principles of good practice. Report of the ISPOR 2012 Budget Impact Analysis Good Practice II Task Force. Value Health. 2014;17(1):5-14.
- Caro JJ, Briggs AH, Siebert U, Kuntz KM. Modeling good research practices: overview. A report of the ISPOR-SMDM Modeling Good Research Practices Task Force-1. Value Health. 2012;15(6):796-803.
- US Food and Drug Administration. Questions and answers on biosimilar development and the BPCI Act. Guidance for industry. September 2021. https://www.fda.gov/media/119258/download
- US Food and Drug Administration. Drug and device manufacturer communications with payors, formulary committees, and similar entities: questions and answers. Revised draft guidance for industry. Notice of availability, Federal Register, 3 June 2026. https://www.federalregister.gov/documents/2026/06/03/2026-11060/drug-and-device-manufacturer-communications-with-payors-formulary-committees-and-similar
- Academy of Managed Care Pharmacy. AMCP Format for Formulary Submissions. https://www.amcp.org/resource/amcp-format-formulary-submissions-guidance
Last reviewed: September 2026. This article is for general information and does not constitute regulatory or legal advice. The AMCP Format is guidance issued by AMCP; the current version should be consulted before preparing a dossier.