Table 1 Recommendations for updating data sharing policies to address emerging policy and research developments from the past decade

Participant-level data sharing with researchers

 • Pharmaceutical companies commit to sharing with qualified researchers’ patient-level data from clinical trials for medicines and indications approved in the US and EU.

 • Each company will establish a scientific review board who are not employees of the company.

 • Data requests will be evaluated against a description of the data being requested, hypothesis being tested, research rationale, analysis plan, publication and posting plan, qualifications and experience of the team, and a description of conflicts of interest, including potential competitive use of the data and the source of any research funding.

 • Companies will implement a system to provide applicable data and protocols to help facilitate the research.

1. Participant-level data from any clinical trial result submitted to support drug approvals should be eligible for sharing (irrespective of continuing follow-up). Companies should endeavour to facilitate the sharing of clinical trials not directly supporting medicine approvals within a clearly defined timeframe of primary result completion/publication.

2. Companies should aim to only assess if trials are in scope for participant-level data sharing. All decisions on the legitimacy of a data request should be evaluated by an independent scientific review panel.

3. Companies should outline the date on which their trial consent procedures were last updated and provide an example form to avoid issues with future data sharing.

4. Companies should maintain public lists of sponsored trials that are eligible/ineligible for participant-level data sharing.

5. Where possible, companies should provide full CSRs, data dictionaries, data derivation documents, protocols, SAPs, and anonymisation guides with requests to help facilitate valid secondary research.

Sharing of clinical study reports

 • To help patients and healthcare professionals understand the results of clinical trials and the evidence used to approve a new medicine (US and EU), pharmaceutical companies will make publicly available, at a minimum, the synopses of CSRs for clinical trials.

 • Companies will evaluate requests for full CSRs.

1. While initiatives to share result synopses are admirable, given the extent of extra clinical information and detail contained in CSRs, full CSRs from all clinical trials submitted to support medicine approvals should be publicly available for download.

2. Subsequent versions of CSRs should be made available when prepared.

3. Both the FDA and EMA have acknowledged resource difficulties in disseminating CSRs; thus, it is likely companies need to engage in processes that facilitate public downloads.

Sharing of protocols and statistical analysis plans

 • Pharmaceutical companies commit to sharing with qualified researchers’ protocols from clinical trials for medicines and indications approved in the US and EU.

1. Companies need to make SAPs and protocols of all published clinical trials publicly available, and consideration should be given to sharing within 6 months of enrolling the first participant.

2. Updated versions of SAPs and protocols should be available when prepared.

Sharing results with trial participants

 • To help inform and educate patients about the clinical trials in which they participate, pharmaceutical companies will work with regulators to adopt mechanisms for providing a factual summary of clinical trial results to research participants.

1. All trial participants should be provided a lay summary reporting trial results within 12 months of primary outcome completion. These lay summaries should also be made publicly available at that time.

2. Subsequent summaries should be prepared for follow-up outcomes.

3. Study protocols should include plans for lay summaries.

Publishing clinical trial results

 • All clinical trials should be considered for publication irrespective of whether the results were positive or negative. At a minimum, results from all phase 3 trials and any trial results of significant importance should be published.

1. All clinical trials must have result summaries published to the trials registry site within 12 months of the primary outcome completion, with efforts to make a scientific journal publication available within the same timeframe.

2. Result summaries and scientific journal publications should occur for follow-up outcomes.

3. Publishing of clinical trial results should occur regardless of study outcomes or phase.

4. Study protocols should include plans for publications.

Public data sharing policies

 • Companies following the 2013 PhRMA/EFPIA Principles for Responsible Clinical Trial Data Sharing will certify on a publicly available website that they have established policies and procedures to implement these data sharing commitments.

1. Towards harmonising terminologies and processes, companies should have public data sharing policies providing precise and detailed information on policies and procedures (including web links for access) to sharing participant-level data, full CSRs, protocol/SAPs, lay summaries, CSR synopses, reporting of results on clinical trial registries, and scientific journal publications.

2. Policies should be written with subheadings and numbered criteria, providing clear information on what data will be shared, when, and under what conditions for each data item.

3. To facilitate cross-referencing between documents, clinical trial registration and internal trial numbers/names should be included in all publications, product information leaflets, participant-level data, CSRs, protocols/SAPs, and lay summaries.