Guarding a publication plan against non-publication and delay
Publication planning addresses the risk that research results remain unavailable or take too long to reach readers, including negative or commercially unwelcome findings.
Unwelcome and negative findings
All good-quality scientific findings should be published, including negative outcomes and hypotheses that are not confirmed or are disproved. Negative findings can be as informative as positive ones. Absence of an expected phenomenon can be important, and negative findings can encourage new experimental approaches or prevent other researchers repeating work unnecessarily.
Researcher or sponsor interests may discourage publication of results contradicting earlier work, but selectively hiding new knowledge is incompatible with publication ethics and constitutes serious misconduct. Failure to publish human-study findings that could affect human welfare is widely regarded as scientific misconduct.
Researchers and publishers have traditionally been reluctant to publish negative findings, yet their positive contribution depends on availability; both have a responsibility to make them accessible. A 2021 publication-planning account linked their limited availability with the reproducibility crisis and damaging practical and economic consequences.
Industry submission commitments
In 2010, four pharmaceutical manufacturer associations, including the Pharmaceutical Research and Manufacturers of America, stated that all industry-sponsored interventional trials should be considered for publication regardless of whether findings were positive. At minimum, all phase 3 trials and other trial results of significant medical importance should be submitted, including those from discontinued development programmes.
The statement’s submission timings were:
| Product status | Submission timing |
|---|---|
| Already marketed | Within 12–18 months of trial completion. |
| Investigational | Within 12–18 months of regulatory approval or the decision to discontinue development. |
Time from research to publication
A practitioner’s account published in 2008 described three or four years from study planning to journal publication as not unusual, covering planning, obtaining funds, treating and examining patients, and scrutinising and interpreting data.
A 2016 analysis of 76 registered Pfizer trials completed in 2010 for approved prescription products and vaccines found that primary outcomes from 65 trials (85%) appeared in 71 manuscripts. Median time from study completion to publication was 31 months, ranging from 3 to 63 months.
The eleven unpublished trials had different statuses:
- Two had been submitted to at least one journal.
- Two had not yet been submitted.
- Seven had no publication plans because recruitment difficulties had caused early termination.
Publication timing also differed by the journal’s place in the submission sequence:
| Accepted by | Manuscripts | Median months from completion | Range in months |
|---|---|---|---|
| First-choice journal | 31 | 28 | 8–63 |
| Second-choice journal | 19 | 32 | 3–45 |
| Third-choice journal | 13 | 40 | 24–53 |
The publication rate and median time were comparable to earlier analyses combining industry and non-industry research. Earlier research on industry-sponsored studies supporting US regulatory approvals had found that up to 57% remained unpublished five years after approval, with favourable primary outcomes more likely to be published.
Sponsors, authors and journals have opportunities to improve timeliness, but effective changes may require reconsidering the entire publication process.