Effective data management practices are essential for a well-run research program. They help ensure accurate, reliable, and retrievable data are available to: advance the formation/testing of scientific hypotheses; facilitate collaboration with co-investigators in large research projects; and support future patent pursuits.
Data includes:
laboratory notebooks/worksheets;
memoranda;
original notes or exact copies of notes associated with original observations;
videotapes;
clinical protocols;
spectra;
computer files;
images;
any other records necessary for reconstruction and evaluation of study results.
Sound data management practices also promote compliance with internal/external requirements for data access and retention and facilitate investigators’ abilities to promptly/effectively respond to external requests for data during reviews of grants/manuscripts and scientific misconduct investigations. Strategies for promoting effective data management are briefly addressed below and include: documenting the study design and methodology; maintaining data notebooks; complying with data ownership and retention requirements; and following data sharing and access requirements.
Documenting the study design and methodology – Research records should document the experimental design and methodology, statistical tests used, data that do/do not support hypotheses, and rationales for excluding experimental data/outliers. These practices facilitate future replication of experiments and minimize unintentionally biasing results from preconceived expectations.
Maintaining data notebooks – Data notebooks should preserve the original data, document the research process, preserve the chronology of experiments/data collection, and minimize risks of data loss. Standard laboratory practices to preserve data integrity include:
using bound notebooks with numbered pages or heavy grade paper for loose-leaf notebooks to minimize tearing/damage of pages;
careful inclusion and preservation of original photographs, computer outputs, negatives, etc.;
making entries/corrections in permanent ink;
signing and dating entries;
utilizing an index to facilitate tracking multiple notebooks;
routine reviews of notebooks by PIs and supervisors;
signatures of witnesses for data that may be required to support patent pursuits.
Complying with data ownership and retention requirements – Data generated by faculty, staff, trainees, or visiting scholars in internally or externally funded UTHSCH research are considered UTHSCH assets. Data are all information associated with the research regardless of the form or media (HOOP 23.06), and include laboratory notebooks/worksheets, memoranda, original notes or exact copies of notes associated with original observations, videotapes, clinical protocols, spectra, computer files, images or any other records necessary for reconstruction and evaluation of study results.
Data must be retained for a minimum of five years after publication in final form or after submission of final reports to research sponsors, and PIs, as agents of UTHSCH, have primary responsibility for preserving the data and complying with UTHSCH’s retention policy. In some instances, funding agencies may require longer retention periods that must be followed. If a PI transfers to a new institution during this time period, the original data may be transferred, but UTHSCH and collaborators must have access to the data during the required retention period.
Transfer of tangible research resources requires a materials transfer agreement approved by the Executive Vice President for Research Affairs. Trainees/visiting scholars participating in UTHSC-H research projects may not take original data/data notebooks when they leave the institution because original data must remain with the PI. They may however make arrangements with PIs to make copies of data to take with them. Trainees/visiting scholars must also ensure data are available to PIs during their tenure at UTHSCH.
Following data sharing and access requirements – Scientists in general have an obligation to share and broadly disseminate research findings for a variety of reasons that include: promoting the advancement of science; allowing other investigators to confirm, extend, and/or question research results and hypotheses; avoiding duplication of research requiring costly data collection; and facilitating the translation of research into products/procedures that improve health. UTHSC-H faculty, staff, trainees, and visiting scholars must also satisfy institutional, Federal, and State requirements to provide co-investigators, the institution, research sponsors, the research community, and the public varying levels of access to their research data and/or resources.
At the institutional level, data must be available to collaborators and university administrators, and they also may be requested by research sponsors. At the federal level, NIH “expects and supports the timely release and sharing of final research data from NIH-supported studies for use by other researchers” (http://grants1.nih.gov/grants/policy/data_sharing/index.htm). NIH also requires investigators to provide “prompt access to unique research resources that arise from biomedical research” supported by NIH, and those resources include synthetic compounds, organisms, cell lines, viruses, cell products, cloned DNA, DNA sequences, mapping information, crystallographic coordinates, and spectroscopic data (http://grants.nih.gov/grants/guide/notice-files/not96-184.html).
Research data from federally funded research “cited publicly and officially by a Federal agency in support of an action that has the force and effect of law” may also be requested under the Federal Freedom of Information Act (http://www.hhs.gov/foia/45cfr5.html). Finally, access may be required under the State of Texas Public Information Act (http://www.capitol.state.tx.us/statutes/go/go055200toc.html). In some instances, there may be legitimate reasons to restrict or delay access to research data or resources such as: protecting legitimate proprietary interests as allowed under the Bayh-Dole Act; protecting rights and privacy of human subjects; avoiding premature release of preliminary data; and maintaining confidentiality of data prior to publication.
