Data Integrity Policy Statement

Department of Pathology

Revised June 1st, 2019

Data Integrity Policy Statement

A major obligation in the responsible conduct of research is ensuring scientific data integrity as mandated by Stanford University and federal funding agencies (e.g., NIH). A summary of Stanford’s policies for the responsible conduct of research can be found here: (doresearch.stanford.edu/research-scholarship/responsible-conduct-research)


Ultimately, the principle investigator (PI) is responsible for the integrity of scientific data in their lab and across managed research projects.  Although problems with data integrity may not be a result of the PI’s direct actions, they can still adversely affect the PI in terms of the ability to obtain/maintain funding, publish research papers, protect intellectual property, and more seriously involve disciplinary actions against members of their research organization.  There are several responsibilities of the PI to ensure scientific data integrity including: 1) overall data reliability; 2) reporting misconduct; 3) retention of scientific data; and 4) access to research data. Official Stanford University policies regarding these topics and links to resources for compliance and best practices are described below.

Clinical Research Data (HIPAA)

Clinical research involving patient health information may be subject to regulation under the Health Insurance Portability and Accountability Act (HIPAA). Such information is held to higher standards regarding where it can be stored and how it can be accessed and shared. These must be considered if research datasets contain such qualified information.

Responsibilities of the PI

1) Scientific Data Reliability

The responsible conduct of research includes maintenance of high quality standards while acknowledging possible mistakes. Beyond human errors or negligence, there may be deliberate research misconduct, which is defined as "fabrication, falsification, plagiarism in proposing, performing, or reviewing research, or in reporting research results."

While PIs are not necessarily responsible for meticulously scrutinizing every piece of data generated in their research operations, they do have a duty for timely correction of honest mistakes and immediate reporting of research misconduct (see 2 below). It is recommended to utilize the resources below to maximize transparency/communication within the lab to minimize errors and misconduct.

 

1) Reporting Errors or Misconduct

Ideally, if scientific data are well organized (see Resources below) and a culture of information sharing is engrained in research operations, errors in data reporting or misconduct will be minimized.  For incidental errors that have been included in research publications, grants, or data repositories the PI should work with the respective editor, program officer, or database administrator to issue a correction in a timely fashion.

For actual research misconduct in data representation the PI is obligated to immediately report the incident through appropriate channels at Stanford (below).

 

3) Retention of Research Data

Maintaining well organized repositories of research data is generally a good practice that can benefit the long-term continuity of a research program. It can also be a required component of funding mechanisms and publications, and can be a critical component of intellectual property prosecution.

Stanford requires that all data relevant to funded projects be retained for a minimum of 3 years following the project completion (defined as end of funding). In addition, any project involving trainees (postdocs or students) must retain all data until completion of the trainee’s program at Stanford. Lastly, data retention relevant to intellectual property disclosures, in particular written records of early ideas and experiments (see Lab Notebook guidelines), must be retained to support patent prosecution (up to 20 years from initial disclosure). 

 

4) Access to Research Data

Data aggregation and management are emerging as key components of many research activities.  Sharing data reinforces open scientific inquiry, encourages diversity of analysis and opinion, and promotes new research.  Moreover, sharing of scientific data and detailed protocols is becoming an intrinsic component of grant productivity and publication in top-tier scientific journals.

While Stanford has no official policy on data access, it does provide numerous tools for sharing and collaborating on simple to complex data types within and outside the university (see Resources).  Beyond just good practice, PIs should consider the data sharing and reporting standards of independent scientific journals.  The recommended standards for sharing of published data should meet or exceed those laid out by the Nature Publishing Group (below). 

Resources

Stanford resources for Storing, Sharing, Organizing data within the University:

 

Stanford Resources for Outside Data Sharing:

 

Non-Stanford Outside Data Sharing Resources: