Free scoping read. A written assessment of your manuscript in two business days, at no cost.

Start one
Contents

Academy · Research Publication · Evolution of GPP

The Evolution of Good Publication Practice (GPP) Standards for Medical Writers

Good Publication Practice (GPP) has had four versions in under twenty years, each addressing a gap that practice had exploited. The direction of change has been consistent: greater disclosure, greater author control, broader coverage of the publication programme and less scope for convention to substitute for written commitment. This consistency is the most useful characteristic of the standard for medical writers to understand.

Pubrica AcademyWritten by the medical writing and biostatistics teams Reviewed byPubrica biostatistics team Published9 October 2026 Last checked9 October 2026 Reading time14 minutes Sources31 peer-reviewed

Reading the versions of Good Publication Practice in sequence is more informative than reading the current version alone, because each revision records the practices it was intended to prevent. The written agreement became explicit after a decade of disputes about data control. Plagiarism was introduced as a section once programme-based publishing had made text reuse routine. Non-journal outputs were brought within scope after congress abstracts had for years been managed to a lower standard than the papers they preceded. An understanding of this pattern allows writers to anticipate which current practices are likely to be addressed next.

Learning objectives

  • The state of the literature on sponsored research before GPP existed, and the evidence that prompted the first version
  • The milestones: the 1998 retreat, the 2003 guidelines, GPP2 in 2009, GPP3 in 2015 and the 2022 update
  • How the written agreement developed from a concept into a specified set of responsibilities
  • Six consistent directions of change across the four versions, and the implications of each for current practice
  • Why the consultation figures across versions cannot be interpreted as a growth series, although they are often presented as such
  • Which current undeclared conventions the pattern suggests will be addressed next, stated as inference rather than fact

Common misconceptions

  • “GPP was written by the pharmaceutical industry for the pharmaceutical industry.” The first version arose from a retreat organised jointly by the Council of Biology Editors and industry members, and every subsequent version has involved an external consultation. Authorship has always been mixed.
  • “Each version is a complete rewrite.” Each version addresses specific gaps. Almost every new section in a later version corresponds to a practice that the earlier version had failed to prevent.
  • “The guideline lengthened because guidelines tend to lengthen.” It lengthened in specific areas, namely written agreements, data sharing, plagiarism and non-journal outputs, and each of these areas had a documented problem behind it.
  • “Only the most recent version is relevant.” The operative standard is the version named in the relevant contract and SOPs. Many current agreements still cite GPP3, and some cite no version at all.
  • “Consultation numbers indicate that the standard is gaining authority.” The published figures count different entities, namely companies circulated, individuals invited and experts consulted, and are not comparable across versions.

Publication practice before the introduction of a standard

Two bodies of evidence created the pressure for a guideline, and neither was about industry specifically.

The first was publication bias. Easterbrook and colleagues, in the Lancet, showed that whether clinical research was published was associated with the direction of its findings; Stern and Simes, in BMJ, followed a cohort of research papers and found that publication was not only less likely but also slower where the results were unfavourable 30,31. The implication was general: the published literature was not a representative sample of the research conducted, and the unpublished portion was not missing at random.

The second was authorship. Rennie, Flanagin and Yank argued in JAMA for replacing the author as a unit of credit with a description of contributions, precisely because the author list had ceased to be a reliable account of who had done the work 29. Davidoff and colleagues, writing in Annals of Internal Medicine on sponsorship, authorship and accountability, set out the editors’ position on what a sponsored publication owed its readers 28.

Taken together, these constituted a diagnosis without a procedure. Editors could state what they would not accept, but no one had specified what a sponsor and its collaborators should do, and in what order, before a manuscript existed. GPP was created to fill that gap, and it continues to fill it, which is why it operates alongside rather than within the ICMJE Recommendations.

2003: the first guidelines

The initiative began with a retreat in 1998, organised by the Council of Biology Editors together with industry members; this detail is significant, because it means that the standard was never purely an industry document 26. Wager, Field and Grossman record that the draft guidelines were circulated to 70 companies in 2000, and the result was published in Current Medical Research and Opinion in 2003, volume 19, issue 3, with the authors’ editorial explaining why another set of guidelines was needed at pages 147 to 148 and the guidelines themselves at pages 149 to 154 26,27.

The first version thus occupied six pages. Its brevity is significant, because it was essentially a statement of principles with a small number of procedural requirements: register trials, publish results regardless of direction, follow the ICMJE on authorship, declare the sponsor’s involvement, and do not use ghostwriters 26. It did not, however, specify the mechanisms. It stated that authors should control the content without identifying the document that would establish that control, and that studies should be published without identifying who would decide.

In retrospect, this is the characteristic form of a first-generation standard: correct about ends but silent about means. The subsequent history consists largely of the specification of those means.

2004 to 2009: documentation of the gap

The second half of that decade produced evidence, in an unusually direct form, that a statement of principles was insufficient: internal company documents released in litigation.

Ross and colleagues, in JAMA, examined industry documents from the rofecoxib litigation and described guest authorship and ghostwriting as documented practices rather than suspicions 17. Hill and colleagues, in Annals of Internal Medicine, reviewed internal documents from the ADVANTAGE trial and described a study whose design purpose was marketing rather than inquiry 20. Angell, in JAMA, argued that industry-sponsored clinical research was a broken system; Gøtzsche and colleagues, in PLoS Medicine, asked directly what should be done about ghostwriting in the medical literature; Titus, Wells and Rhoades, in Nature, wrote about repairing research integrity more broadly 16,18,19.

In parallel, the professional bodies defined the writer’s position. Hamilton and Royer set out the AMWA task force position on medical writers’ contributions in the AMWA Journal; Jacobs and Wager published the EMWA guidelines on the role of medical writers in Current Medical Research and Opinion; Norris and colleagues issued the ISMPP position statement in the same journal 21,23,25. Within five years, three professional organisations had stated that the role of the medical writer should be declared rather than concealed.

The empirical evidence was also accumulating. Wislar and colleagues, in BMJ, measured inappropriate authorship across leading general medical journals; von Elm and colleagues, in JAMA, measured duplicate publication in a systematic review corpus 9,24. By 2009, the gaps in the 2003 guidelines were no longer a matter of opinion.

2009: GPP2 and the specification of mechanisms

Graf and colleagues published GPP2 in BMJ, and the change in character is the most important step in this history. GPP2 ceased to describe what should happen and began to specify the documents through which it would happen 12,26.

Foremost among these was the written agreement. GPP2’s written-agreement section lists 3 sponsor responsibilities, 9 author responsibilities and 8 shared responsibilities, a division that converts “authors should retain control” into a set of commitments for which the parties can be held accountable. It also supplied practical tools: GPP2 is supported by 4 tables, including a checklist for articles and presentations and a checklist of basic requirements for written agreements, and cites 29 references 12.

It was also the product of consultation rather than of a committee alone: Graf and colleagues invited 193 people and received 116 sets of comments 12. That response rate, for a document on industry publication practice, itself indicates how contested the field was.

The GPP2 breakdown of responsibilities remains, in Pubrica’s assessment, the most usable account of what a written agreement must settle, which is why a version two generations old is still worth reading. It is used in this way on the GPP topics authors most often miss.

3sponsor responsibilities in GPP2’s written-agreement section
9author responsibilities
8shared responsibilities
4supporting tables, including two checklists

The point at which GPP specified documents rather than intentions. Source: Graf C, Battisti WP, Bridges D, Bruce-Winkler V, Conaty JM, Ellison JM, et al. Good publication practice for communicating company sponsored medical research: the GPP2 guidelines. BMJ.

2015: GPP3 and the expansion of scope

Battisti and colleagues published GPP3 in Annals of Internal Medicine after a notably large consultation: ISMPP invited more than 3,000 publication professionals worldwide to apply for the steering committee or to serve as reviewers 8.

The additions in GPP3 indicate what had become problematic in the intervening six years. It added a section on principles for company-sponsored medical research, a framing change that placed publication practice within research conduct rather than alongside it 8. It added data sharing, which had moved from aspiration to expectation. It added a section on studies that should be published, which addresses the one failure with no external check: a completed study that is never submitted. It also added plagiarism, a section that may appear surprising until it is recognised that programme-based publishing (many related papers, one clinical study report, often one writer) creates pressure for text reuse that an individual academic author does not face in the same form. Das and Panjabi had set out in Perspectives in Clinical Research why medical writers are structurally exposed to precisely that pressure 10.

Each of these four additions closed a gap. None introduced a new principle; each addressed a point at which the earlier text had been sufficiently silent to permit a permissive reading.

2022: extension from the publication to the programme

DeTora and colleagues published the 2022 update in Annals of Internal Medicine, developed with a consultation in which 38 invited experts gave feedback before submission alongside an open reviewer call 4,5.

Three changes are significant. The scope moved from company-sponsored medical research to company-sponsored biomedical research, which brings devices, diagnostics and digital health unambiguously within scope 5. The unit of concern moved from the journal article to the full set of outputs from a programme: congress materials, plain-language summaries, preprints and data-sharing artefacts are handled under the same obligation as the paper. Patient and public contribution is also treated as a contribution to be acknowledged rather than a courtesy to be mentioned.

One limitation of this account should be stated. Counts of the sections and recommendations in GPP 2022 circulate in secondary summaries; Pubrica was able to confirm the update’s existence and publication details, but not its internal structure, from the article itself. A structural figure that cannot be verified is therefore not repeated here, and it should not appear in an SOP. The article itself should be consulted if its architecture needs to be described.

Interpretation of the consultation figures

The participation figures across versions are often presented as evidence that the standard is gaining authority. They should not be interpreted in this way, because they count different entities.

“Circulated to 70 companies” in 2000 counts organisations 26. “193 people invited, 116 sets of comments” for GPP2 counts individuals invited and responses received 12. “More than 3,000 publication professionals invited to apply or review” for GPP3 counts an open invitation to a pool, not participants 8. “38 invited experts” for GPP 2022 counts a targeted expert panel, which ran alongside an open reviewer call 4. Reading 70, 193, 3,000 and 38 as a series leads to the erroneous conclusion that consultation collapsed after 2015.

What the figures jointly support is narrower but still relevant: every version since the first has been developed through consultation beyond the drafting group, and the consultation has been documented on each occasion. That is the appropriate response, internally, when GPP is characterised as one vendor’s opinion. The specific numbers should be cited with their specific definitions, or not at all.

Six consistent directions of change

Across the four versions, the direction of change is sufficiently consistent to serve as a heuristic.

From convention to document. Practices that were tacitly understood became practices that are formally signed. The written agreement is the clearest example.

From the funder to the role. Declaring who paid gave way to describing what the sponsor did, across design, conduct, analysis and the decision to publish.

From the paper to the programme. The unit of concern widened from the manuscript to every output derived from the same data.

From concealment to attribution. The writer moved from invisible, to permissibly mentioned, to named with employer and funder, a shift that the professional bodies promoted and that the reporting-quality work of Gattrell and colleagues in BMJ Open and Mills and colleagues in F1000Research subsequently made easier to defend 6,7.

From publishing well to publishing at all. The obligation shifted from how a study is reported to whether a completed study is reported at all, which is where the publication-bias evidence was eventually reflected in the text 30,31.

From a narrow definition of contribution to a broader one. Statisticians, agencies and most recently patient contributors moved from background to named.

Each direction has been monotonic. None has reversed across four versions in two decades, which provides a reasonable basis for predicting where the next gap will be closed.

The four versions, and what each one closed

VersionWhere publishedConsultation as reportedWhat it closed
GPP (2003)
Wager, Field & Grossman
Current Medical Research and Opinion, 19(3): editorial 147–148, guidelines 149–154Began with a 1998 retreat organised by the Council of Biology Editors with industry members; draft circulated to 70 companies in 2000The absence of any written standard: registration, publication regardless of direction, ICMJE authorship, no ghostwriters
GPP2 (2009)
Graf et al.
BMJ193 people invited; 116 sets of comments receivedThe missing mechanisms — written agreement specified as 3 sponsor, 9 author and 8 shared responsibilities, with 4 supporting tables and 29 references
GPP3 (2015)
Battisti et al.
Annals of Internal MedicineMore than 3,000 publication professionals invited to apply or reviewFour silences: principles for company-sponsored research, data sharing, studies that should be published, and plagiarism
GPP 2022
DeTora et al.
Annals of Internal Medicine38 invited experts gave feedback before submission, alongside an open reviewer callScope limits: “biomedical” rather than “medical” research, non-journal outputs in scope, patient contribution named

Publication details and consultation figures as reported in each source and, for GPP 2022, in the ISMPP announcement. The consultation figures count different things — organisations, invitees, an open applicant pool, a targeted expert panel — and are not comparable with one another. The final column is Pubrica’s characterisation of what each version added, not a quotation.

Illustration of the change in a single clause

The historical account is easier to apply when illustrated in a document. Below is the publication clause of a sponsor–investigator agreement written to first-generation assumptions, followed by the same clause written to current assumptions. The first is not a contrived example; it remains common in legacy templates.

A publication clause written to 2003 assumptions, and to current ones

First-generation

The Institution and Investigator may publish the results of the Study, subject to the Sponsor’s prior review. The Investigator shall submit any proposed publication to the Sponsor for review at least thirty days prior to submission, and shall give due consideration to the Sponsor’s comments. Multi-centre results shall be published before single-centre results. The Sponsor’s contribution shall be acknowledged. Authorship shall be determined in accordance with accepted practice.

Publication is permitted, not required. Nothing here obliges any party to publish an unfavourable result, which is the failure with no external check.

“Due consideration” to sponsor comments leaves editorial control unallocated at precisely the point at which it matters.

Review may extend indefinitely. A thirty-day minimum with no maximum delay and no stated grounds.

“Accepted practice” on authorship names no criteria, produces no record and can be satisfied by whatever occurred.

“The Sponsor’s contribution shall be acknowledged” covers the funding and says nothing about the writer or the statistician.

Silent on data access and on every output that is not a journal article.

Current

The Sponsor shall ensure that the results of the Study are submitted for publication, whether or not those results are favourable to the Sponsor’s products, within twelve months of the primary completion date. The Investigator retains final authority over the content of any publication and over the decision to submit; the Sponsor may review a proposed publication for thirty days and may request a further delay of up to sixty days solely to permit the filing of patent applications, such delay to be requested in writing with reasons. Authorship shall be determined by application of the ICMJE criteria to each individual, recorded in an authorship grid completed before outlining and retained by both parties; authorship shall not be offered, withheld or conditioned on anything other than contribution. All contributors not meeting those criteria, including any medical writer and any statistician, shall be acknowledged by name, employer and the party funding their contribution. The Investigator shall be given access to the full statistical analysis plan and output, and shall be entitled to request further analyses. This clause applies equally to congress abstracts, posters, preprints and plain-language summaries derived from the Study.

Publication becomes an obligation with a deadline, stated as independent of the direction of the result.

Final authority over content and submission is allocated to a named party, and sponsor review is bounded.

Delay has a maximum, a single permitted ground and a written-request requirement, so that it cannot become an indefinite veto.

Authorship is tied to criteria and to a record retained by both parties, which makes it auditable rather than open to dispute.

The writer and statistician are named with employer and funder, which captures the attribution shift of the last two decades in a single sentence.

Data access is specified, and the clause reaches congress abstracts, preprints and lay summaries.

Every difference between these two clauses corresponds to a point that a later version of GPP had to make explicit because the earlier text was read permissively. This is the practical value of the history: the second clause is not stricter out of general caution, but in the five specific places where twenty years of practice demonstrated that a convention would not hold.

Implications for current medical writing practice

Three practical conclusions follow, together with one clearly labelled inference.

Know which version binds you. The operative standard is the version named in the contract and SOPs, not the newest document. A writer working to GPP 2022 on a programme contracted under GPP3 should state this in writing rather than applying the higher standard without comment, because the difference may later matter to another party.

Treat every output as in scope. The direction of change has been unambiguous on this point for over a decade. A congress abstract managed to a lower standard than its paper is the most outdated practice still in common use.

Document what is currently left to tacit understanding. If a working arrangement (who approves the final wording, how long authors are given, what happens if an author dissents) exists only as a shared assumption, the history suggests that it will eventually become a documented requirement, and a version drafted now is likely to be more favourable than one imposed later.

The inference, offered as inference. Applying the pattern, the most exposed gaps at present are those where it is currently conventional to leave a substantive contribution or material influence undeclared. The disclosure of AI-assisted drafting, the status of preprints within a publication plan, and the handling of plain-language and congress material at paper-level rigour are all areas in which practice currently relies on nothing having been written down. Pubrica is not aware of a GPP provision that settles any of them, and none is proposed here. This paragraph should be read as an interpretation of a trajectory, not as a requirement, and used as any trend would be: to decide which undocumented practices to document first. How an organisation can put this into practice is discussed on getting GPP adopted internally.

Key points

Key points

GPP has had four versions in under twenty years, and each addressed a gap that practice had identified. The 2003 guidelines, based on a 1998 retreat and circulated to 70 companies, stated appropriate objectives but left the means unspecified. GPP2 specified them, in the form of a written agreement with 3 sponsor, 9 author and 8 shared responsibilities, together with checklists. GPP3 added principles, data sharing, an obligation concerning studies that should be published, and plagiarism. The 2022 update extended the scope to biomedical research and to every output of a programme. The direction of change has not reversed: from convention to documentation, from funder to role, from individual paper to programme, and from concealment to attribution. Any arrangement that currently relies on tacit understanding is therefore likely to require written documentation.

Application to your study

Application to your study

  1. The version named in contracts and SOPs should be established, together with whether any version is named at all. This is the operative standard, irrespective of the version to which an individual writer works.
  2. The written-agreement breakdown in GPP2 should be consulted, although it is two generations old. It remains the clearest account of the matters that the agreement must resolve.
  3. Legacy publication clauses should be audited for the five first-generation markers: publication permitted rather than required, unlimited review, no maximum delay, authorship determined by “accepted practice”, and acknowledgement limited to the sponsor.
  4. Every clause should be checked for applicability to non-journal outputs. If a clause refers to “publication” but applies only to journal articles, congress abstracts fall outside the standard.
  5. At least one arrangement that currently relies on tacit understanding should be documented in writing. Review timelines and the handling of dissent are the usual candidates. Drafting such provisions internally is preferable to receiving another party’s version.
  6. Consultation figures should not be cited across versions as a series. They count, respectively, organisations, invitees, an applicant pool and an expert panel; each should be cited with its own definition.
  7. The internal section counts of GPP 2022 should not be cited unless they have been verified against the article directly. Secondary summaries circulate figures that Pubrica has been unable to confirm.

Frequently asked questions

The initiative began with a retreat in 1998, organised by the Council of Biology Editors together with industry members, so it was never purely an industry document. Wager, Field and Grossman record that the draft guidelines were circulated to 70 companies in 2000, and the result appeared in Current Medical Research and Opinion in 2003, volume 19, issue 3, with their editorial explaining the need for another set of guidelines at pages 147 to 148 and the guidelines at pages 149 to 154. The first version was short; it was strong on principles but almost silent on the documents and mechanisms that would make those principles operate.

GPP2, published by Graf and colleagues in BMJ, was the version that specified mechanisms: its written-agreement section sets out 3 sponsor responsibilities, 9 author responsibilities and 8 shared responsibilities, and it is supported by 4 tables, including a checklist for articles and presentations and a checklist of basic requirements for written agreements, with 29 references. GPP3, from Battisti and colleagues in Annals of Internal Medicine, widened the territory instead, adding sections on principles for company-sponsored medical research, data sharing, studies that should be published, and plagiarism. GPP2 answered “how”; GPP3 answered “about what else”.

Programme-based publishing creates text reuse pressure that an individual academic author does not face in the same form. A single clinical study report typically generates a primary paper, secondary analyses, congress abstracts and lay summaries, often drafted by the same writer from the same source document, so methods text and background text are routinely reused. Das and Panjabi set out in Perspectives in Clinical Research why medical writers are structurally exposed to this. The section concerns misconduct less than it concerns making a normal production pattern explicit before it becomes a finding in another party’s audit.

The figures are often presented in this way, but they do not support that interpretation, because they count different entities. The 2003 draft was circulated to 70 companies — organisations. GPP2 invited 193 people and received 116 sets of comments — individuals invited and responses received. GPP3 saw more than 3,000 publication professionals invited to apply for the steering committee or serve as reviewers — an open invitation to a pool, not a count of participants. GPP 2022 had 38 invited experts giving feedback before submission, alongside an open reviewer call. What the figures jointly support is that every version has been subject to consultation beyond the drafting group.

No figure is given here. Counts circulate in secondary summaries, and Pubrica could confirm the update’s existence, authorship and publication details, but not its internal structure, from the article itself. Citing an unverified structural figure in an SOP or a training deck is the kind of detail that can undermine a policy document later, so the article should be read and its sections counted directly if its architecture needs to be described. What can be stated about the 2022 update is what it changed: the scope moved to biomedical research, non-journal outputs came within scope, and patient contribution is treated as a contribution.

The version named in the contract and SOPs is the operative standard, not the newest document. For new agreements, GPP 2022, published by DeTora and colleagues in Annals of Internal Medicine, should be named with its citation. Where legacy master agreements and SOPs still cite GPP3, that remains the contracted standard until the documents are amended, and a writer applying a later standard to such a programme should state this in writing rather than doing so without comment. The weakest position is a document that names no version at all, which can be satisfied by any practice.

This is inference, and it is identified as such. The consistent direction across four versions is that anything relied upon as a shared understanding eventually becomes a written requirement, and that any substantive contribution or material influence eventually has to be declared. On that reading, the exposed areas at present are those in which practice rests on nothing having been written down: the disclosure of AI-assisted drafting, the status of preprints within a publication plan, and whether congress and plain-language material is held to paper-level rigour. Pubrica is not aware of a GPP provision settling any of these and does not propose one. The inference is best used to decide which undocumented practice to document first.

GPP2 is, for one specific reason: its written-agreement breakdown into 3 sponsor, 9 author and 8 shared responsibilities is the most usable account available of what the agreement must settle, and later versions assume rather than restate it. The 2003 version merits brief reading as history, because the limited extent of its specification explains why each later version is longer. GPP3 matters because many live contracts still name it. If only one version is read, it should be the version cited in the contract, read alongside GPP2’s written-agreement section.

References

  1. International Committee of Medical Journal Editors. Recommendations for the Conduct, Reporting, Editing, and Publication of Scholarly Work in Medical Journals. ICMJE. 2026. read 9 October 2026. icmje.org
  2. International Committee of Medical Journal Editors. The New ICMJE Recommendations (August 2013). ICMJE. 2013. read 9 October 2026. icmje.org
  3. Committee on Publication Ethics. Core Practices. COPE. read 9 October 2026. equator-network.org
  4. International Society for Medical Publication Professionals. GPP 2022: Good Publication Practice guidelines for company-sponsored biomedical research published in Annals of Internal Medicine (press release). ISMPP. 2022. read 9 October 2026. ismpp.org
  5. DeTora LM, Toroser D, Sykes A, Vanderlinden C, Plunkett FJ, Lane T, Hanekamp E, Dormer L, DiBiasi F, Bridges D, Baltzer L, Citrome L. Good Publication Practice (GPP) Guidelines for Company-Sponsored Biomedical Research: 2022 Update. Annals of Internal Medicine. 2022;175(9):1298-1304. doi:10.7326/M22-1460 read 9 October 2026. doi.org/10.7326/M22-1460
  6. Mills I, Sheard C, Hays M, Douglas K, Winchester CC, Gattrell WT. Professional medical writing support and the reporting quality of randomized controlled trial abstracts among high-impact general medical journals. F1000Research. 2017;6:1489. doi:10.12688/f1000research.12268.2 read 9 October 2026. doi.org/10.12688/f1000research.12268.2
  7. Gattrell WT, Hopewell S, Young K, Farrow P, White R, Wager E, Winchester CC. Professional medical writing support and the quality of randomised controlled trial reporting: a cross-sectional study. BMJ Open. 2016;6(2):e010329. doi:10.1136/bmjopen-2015-010329 read 9 October 2026. doi.org/10.1136/bmjopen-2015-010329
  8. Battisti WP, Wager E, Baltzer L, Bridges D, Cairns A, Carswell CI, Citrome L, Gurr JA, Mooney LA, Moore BJ, Pena T, Sanes-Miller CH, Veitch K, Woolley KL, Yarker YE. Good Publication Practice for Communicating Company-Sponsored Medical Research: GPP3. Annals of Internal Medicine. 2015;163(6):461-464. doi:10.7326/M15-0288 read 9 October 2026. doi.org/10.7326/M15-0288
  9. Wislar JS, Flanagin A, Fontanarosa PB, et al. Honorary and ghost authorship in high impact biomedical journals: a cross sectional survey. BMJ. 2011;343:d6128. doi:10.1136/bmj.d6128 read 9 October 2026. doi.org/10.1136/bmj.d6128
  10. Das N, Panjabi M. Plagiarism: why is it such a big issue for medical writers? Perspectives in Clinical Research. 2011;2(2):67-71. doi:10.4103/2229-3485.80370 read 9 October 2026. doi.org/10.4103/2229-3485.80370
  11. Schulz KF, Altman DG, Moher D. CONSORT 2010 statement: updated guidelines for reporting parallel group randomized trials. Annals of Internal Medicine. 2010;152:726-732. read 9 October 2026. care-statement.org
  12. Graf C, Battisti WP, Bridges D, Bruce-Winkler V, Conaty JM, Ellison JM, Field EA, Gurr JA, Marx M-E, Patel M, Sanes-Miller C, Yarker YE. Good publication practice for communicating company sponsored medical research: the GPP2 guidelines. BMJ. 2009;339:b4330. doi:10.1136/bmj.b4330 read 9 October 2026. doi.org/10.1136/bmj.b4330
  13. Moher D, Liberati A, Tetzlaff J, Altman DG; PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Medicine. 2009;6:e1000097. read 9 October 2026. care-statement.org
  14. Woloshin S, Schwartz LM, Casella SL, Kennedy AT, Larson RJ. Press releases by academic medical centers: not so academic? Annals of Internal Medicine. 2009;150:613-618. read 9 October 2026. archive.emwa.org
  15. Rothman DJ, McDonald WJ, Berkowitz CD, Chimonas SC, DeAngelis CD, Hale RW, et al. Professional medical associations and their relationships with industry: a proposal for controlling conflict of interest. JAMA. 2009;301(13):1367-1372. read 9 October 2026. equator-network.org
  16. Gotzsche PC, Kassirer JP, Woolley KL, Wager E, Jacobs A, Gertel A, et al. What should be done to tackle ghostwriting in the medical literature? PLoS Medicine. 2009;6(2):e1000023. read 9 October 2026. archive.emwa.org
  17. Ross JS, Hill KP, Egilman DS, Krumholz HM. Guest authorship and ghostwriting in publications related to rofecoxib: a case study of industry documents from rofecoxib litigation. JAMA. 2008;299:1800-1812. read 9 October 2026. archive.emwa.org
  18. Angell M. Industry-sponsored clinical research: a broken system. JAMA. 2008;300:1069-1071. read 9 October 2026. archive.emwa.org
  19. Titus SL, Wells JA, Rhoades LJ. Repairing research integrity. Nature. 2008;453:980-982. read 9 October 2026. archive.emwa.org
  20. Hill KP, Ross JS, Egilman DS, Krumholz HM. The ADVANTAGE seeding trial: a review of internal documents. Annals of Internal Medicine. 2008;149:251-258. read 9 October 2026. archive.emwa.org
  21. Norris R, Bowman A, Fagan JM, Gallagher ER, Geraci AB, Gertel A, et al. International Society for Medical Publication Professionals (ISMPP) position statement. Current Medical Research and Opinion. 2007;23:1837-1840. read 9 October 2026. archive.emwa.org
  22. von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Vandenbroucke JP; STROBE Initiative. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. BMJ. 2007;335:806-808. read 9 October 2026. care-statement.org
  23. Jacobs A, Wager E. EMWA guidelines on the role of medical writers in developing peer-reviewed publications. Current Medical Research and Opinion. 2005;21:317-321. read 9 October 2026. archive.emwa.org
  24. von Elm E, Poglia G, Walder B, Tramer MR. Different patterns of duplicate publication: an analysis of articles used in systematic reviews. JAMA. 2004;291(8):974-980. doi:10.1001/jama.291.8.974 read 9 October 2026. doi.org/10.1001/jama.291.8.974
  25. Hamilton CW, Royer MG; AMWA 2002 Task Force on the Contributions of Medical Writers to Scientific Publications. AMWA position statement on the contributions of medical writers to scientific publications. AMWA Journal. 2003;18:13-16. read 9 October 2026. archive.emwa.org
  26. Wager E, Field EA, Grossman L. Good publication practice for pharmaceutical companies. Current Medical Research and Opinion. 2003;19(3):149-154. doi:10.1185/030079903125001767 read 9 October 2026. doi.org/10.1185/030079903125001767
  27. Wager E, Field EA, Grossman L. Good publication practice for pharmaceutical companies: why we need another set of guidelines. Current Medical Research and Opinion. 2003;19(3):147-148. read 9 October 2026. ismpp.org
  28. Davidoff F, DeAngelis CD, Drazen JM, Hoey J, Hojgaard L, Horton R, et al. Sponsorship, authorship, and accountability. Annals of Internal Medicine. 2001;135:463-466. read 9 October 2026. ismpp.org
  29. Rennie D, Flanagin A, Yank V. The contributions of authors. JAMA. 2000;284:89-91. read 9 October 2026. ismpp.org
  30. Stern JM, Simes RJ. Publication bias: evidence of delayed publication in a cohort study of clinical research papers. BMJ. 1997;315:640-645. read 9 October 2026. ismpp.org
  31. Easterbrook PJ, Berlin JA, Gopalan R, Matthews DR. Publication bias in clinical research. Lancet. 1991;337:867-872. read 9 October 2026. ismpp.org

Every source above is a peer-reviewed article, a publisher’s record of one, or a standard issued by the body that owns it. Each carries the date it was last read. Where a standard moves, this article is revised rather than left up with a note.

How to cite this article

Pick your referencing style and copy it. Your university or target journal may vary the punctuation — check its guide if it has one.

Pubrica Academy. (2026, October 9). The Evolution of Good Publication Practice (GPP) Standards for Medical Writers. Pubrica. https://pubrica.com/academy/concepts-definitions/what-are-the-evolutionary-standards-in-gpp-for-medical-writers/

This article is written to be learned from. Cite it if you are referring to what it says; if you are using the evidence behind it, cite the sources listed above instead.

Legacy clause audit: first-generation markers

This checklist should be applied to publication clauses in existing contracts, SOPs and templates. Each marker identifies a point at which a later version of GPP had to become explicit because the earlier wording was read permissively. Identifying them is routine; the important decision is which to correct at the next amendment.

LEGACY PUBLICATION CLAUSE AUDIT
Document: ______________________   Date of document: __________
GPP version named in it: [ ] none  [ ] "GPP" unqualified  [ ] 2003  [ ] GPP2
                         [ ] GPP3  [ ] GPP 2022

MARKER 1 - PUBLICATION PERMITTED RATHER THAN REQUIRED
   Clause says publication "may" occur ................... [ ] found
   No deadline for submitting primary results ........... [ ] found
   No statement that publication is independent of the
     direction of the result ............................ [ ] found
   Replacement wording drafted .......................... [ ]

MARKER 2 - EDITORIAL CONTROL UNALLOCATED
   "Due consideration" / "reasonable account" of sponsor
     comments, with no final authority named ............ [ ] found
   Final authority over content named: ______________________
   Final authority over the decision to submit named: ______________________

MARKER 3 - UNBOUNDED REVIEW OR DELAY
   Review period stated as a minimum only ............... [ ] found
   No maximum delay ..................................... [ ] found
   No stated permitted grounds for delay ................ [ ] found
   Maximum delay and permitted ground drafted: ______________________

MARKER 4 - AUTHORSHIP BY CONVENTION
   "Accepted practice" / "customary" / "as agreed" ....... [ ] found
   No named criteria .................................... [ ] found
   No artefact required (grid, record) .................. [ ] found
   No prohibition on offering or withholding authorship . [ ] found

MARKER 5 - ACKNOWLEDGEMENT LIMITED TO THE SPONSOR
   Only the sponsor's contribution is acknowledged ...... [ ] found
   Medical writer not mentioned ......................... [ ] found
   Statistician not mentioned ........................... [ ] found
   No requirement to name employer and funder ........... [ ] found

MARKER 6 - SCOPE LIMITED TO JOURNAL ARTICLES
   "Publication" undefined, or defined as journal articles [ ] found
   Congress abstracts / posters out of scope ............ [ ] found
   Preprints not addressed .............................. [ ] found
   Plain-language summaries not addressed ............... [ ] found

MARKER 7 - DATA ACCESS UNADDRESSED
   No statement of investigator or author access to the
     analysis plan, output or dataset ................... [ ] found

SUMMARY
   Markers found: ____ / 7
   Priority for next amendment (pick two): ______________________
   Owner: ______________________   Target date: __________

Pubrica AcademyWritten and maintained by Pubrica’s subject matter experts. Articles are revised when the underlying standards change, and each article states the date on which it was last reviewed.

How Pubrica can help

Pubrica’s writers work to the GPP version named in the client’s contract and state in writing which version that is and how Pubrica’s contribution will be declared. Where a programme operates under legacy clauses, the team can annotate publication and authorship wording against current practice, distinguishing gaps of continuing relevance from outdated language that has not yet been replaced.

Find something else

Search the Pubrica Academy

Access is free and requires no account. Enter a topic, or open the full library to filter by category, sub-topic or resource type.

Pubrica can review a publication clause from an existing template and identify the generation of GPP to which it corresponds.

A scoping assessment is free and takes two business days. It provides a written assessment of what the work requires and which specialist would undertake it.

Get a free quote

Regulations

Check your project against this rule

  1. 1Your work
  2. 2Your details
  3. 3Send
What do you need?

Material you send is held under signed confidentiality, with access limited to the assigned team and encrypted transfer. It is not used to train anything and is not passed to third-party tools. Ask for the agreement before you send anything you would rather cover first.