Youth Mental Health Narratives: Novel Variables

Discover novel trends from narratives about youth suicides using machine learning techniques. #health

$25,000 in prizes
Completed nov 2024
372 joined

NVDRS Data Sharing Agreement


In order to access the competition dataset, you must first agree to this National Violent Death Reporting System (NVDRS) Data Sharing Agreement (DSA). You may not use the competition dataset if you do not accept this DSA. By signing up for the competition, accessing the competition dataset, or both, you hereby agree to the terms of the DSA.

The National Violent Death Reporting System (NVDRS) Data Sharing Agreement (DSA) was created to govern the protection and use of sensitive or potentially identifiable NVDRS data, as required by the NVDRS Data Re-release Plan. Prior to release of NVDRS Restricted Access Data (RAD) by the Centers for Disease Control and Prevention (CDC), a DSA must be established for external users.

1. Terms Governing Use, Protection, and Reporting of Data

Prior to the release of any data, requestors, their collaborators, and an authorized institutional business official must agree to comply with all of the terms and conditions described below. The requestor’s home institution might also require approval from the body of the requestor’s home institution that is charged with the ethical review and approval of research projects. It is the requestor’s responsibility to check with his or her institution and obtain any necessary approvals (e.g., IRB approval) for your records. CDC requires institutional IRB approval only when the proposal includes linking of NVDRS data with other data that contains personally identifiable information.

a. Use Limited to Research Project

The principal investigator(s) and all collaborators agree that the data will be used solely for the purpose approved by CDC. Furthermore, the principal investigator(s) and all collaborators agree to refrain from any attempt to link NVDRS RAD to any other dataset without prior permission from the CDC (any intention to link NVDRS RAD with other data should be specified in the initial request or follow-up request for the RAD and approved by the CDC). Researchers seeking to conduct additional analyses not specified in the approved proposal or to receive additional data years or variables that do not substantively change the scope of the approved project may contact the program at nvdrs-rad@cdc.gov to complete a project amendment request.

b. Non-transferability of Agreement

Substantive changes made to the project (including, but not limited to, the appointment of a new principal investigator to complete the project, the inclusion of additional collaborators who will have access to the data, or a change in the principal investigator’s institutional affiliation) require the execution of a new DSA, or an amendment to the existing DSA. It is the sole responsibility of the principal investigator to alert the CDC of such changes (within 21 days of the change). If the principal investigator of NVDRS RAD changes positions or leaves an agency or institution, CDC should be notified within a week, and the principal investigator will be required to destroy the data as directed by the CDC.

c. No Disclosure of Data

The principal investigator(s) and all collaborators agree to employ all reasonable efforts to maintain the confidentiality of the data, with such efforts to be no less than the degree of care to preserve and safeguard its own data. The principal investigator(s) and all collaborators further agree not to disclose, reveal, or give the data, with or without charge, to any entity or any individual not listed in section III.k. without prior approval from the CDC.

In the event that the principal investigator is required by judicial or administrative process to disclose the data, the principal investigator must: (1) immediately notify the CDC’s National Center for Injury Prevention and Control (NCIPC) and allow CDC a reasonable time to oppose the process; and (2) work in collaboration with the CDC to maintain confidentiality of the data. (The NVDRS Data Re-release Plan calls for expedited review and fast-track processing in the event of selected public health emergencies).

d. Non-identification of Subjects

The principal investigator(s) and all collaborators of NVDRS RAD agree to the following confidentiality restrictions:

  • NVDRS data will be used solely for statistical analyses related to the approved project. No attempt will be made to identify specific individuals, households, or institutions. Data lists at the individual level will not be published or distributed.
  • In the event of inadvertent discovery of the identity of any person during the course of the proposed project, the principal investigator(s) will (1) send an email to the RAD Help Desk at nvdrs-rad@cdc.gov which will then be routed by CDC to notify the NCIPC Associate Director for Science; (2) safeguard or destroy the identifying information as directed by the CDC; and (3) make no use of knowledge of the discovery. The identifying information must not be disclosed to any other individual or party.
  • State VDRS data provided to CDC are protected under an Assurance of Confidentiality pursuant to Section 308(d) of the Public Health Service Act. An Assurance of Confidentiality is a formal confidentiality protection for data maintained by CDC authorized under Section 308(d) of the Public Health Service Act.
  • If the data use agreement is violated, the principal investigator will be restricted from using NVDRS data in the future. (https://www.cdc.gov/rdc/Data/b4/section308.pdf)
  • The data files provided to the researcher must be stored on, and accessed from, the secure computer system of the researcher’s affiliated organization or institution. This means that the researcher should store files on a server that is behind a firewall, has data encryption, permits file access only to the approved researchers, and has encrypted network communications. The researcher would then access the files from a password-protected institutional desktop or laptop. If a secure computer system is not available, the researcher should store files on an encrypted, password-protected, stand-alone computer or laptop protected by anti-virus and anti-malware software.
  • The inadvertent disclosure of potentially identifying information is to be avoided by using the following guidelines for the release of statistics derived from the requested dataset. For any data release format:
    • Annual counts and rates must be suppressed for cities or counties of fewer than 100,000 people.
    • Cells showing or derived from fewer than 10 deaths must be suppressed, but “zero” cells may be shown. Cell “suppression” will take one of two approaches: 1) combining row or column categories so as to eliminate the small cells, or 2) suppressing the small cell, another cell in the same row, another cell in the same column, and a fourth cell at the intersection of the row and column containing the second and third suppressed cells. Suppression of the second and third additional cells is necessary to prevent derivation of the small cell by subtraction from the row or column totals. Suppression of the fourth cell is necessary to prevent derivation of the second or third cells by subtraction. Beyond these specific guidelines, it must not otherwise be possible to derive identifying information by subtraction or other calculation from a table, or combination of tables, in any release format.
    • Rates are not to be computed for cells containing fewer than 20 deaths (or cases).
    • The disclosed data should never permit identification when used in combination with other known data.

e. Maintenance of Data Security and Oversight

The principal investigator(s) and all collaborators must ensure that data security measures to secure the data, preserve confidentiality, and prevent unauthorized access are enforced and maintained at all times during possession of NVDRS RAD. The principal investigator shall ensure that no unauthorized person has access to the contents of NVDRS RAD files or to any files derived from RAD. Upon request, the principal investigator agrees to permit the inspection by the CDC of the physical storage, management, and handling of RAD files (at reasonable hours) and any other information relating to the DSA.

f. Notification of Pending Publications

The principal investigator agrees to notify the CDC in advance as to when and where a publication of a report (or other public disclosure) from the project will appear. In addition, the principal investigator agrees to provide the CDC, in advance of its appearance, a copy of any manuscript or other public disclosure document. CDC will respect the embargoed information and requests that this information is provided for awareness and record keeping purposes.

g. Non-endorsement Liability

The principal investigator(s) and all collaborators agree not to claim or imply Governmental endorsement of the research project, the entity, or personnel conducting the research project. Any published material derived from NVDRS data must acknowledge the CDC as the provider of the data and participating NVDRS states as the sources of the data. Published materials must also include a disclaimer that credits any analyses, interpretations, or conclusions reached by the author (i.e., the principal investigator(s) and any collaborators who received the data) to that author and not to the original sources of the data (i.e., NVDRS participating states) or to the CDC. The disclaimer should take the following form: “The National Violent Death Reporting System (NVDRS) is administered by the Centers for Disease Control and Prevention (CDC) by participating NVDRS states. The findings and conclusions of this study are those of the authors alone and do not necessarily represent the official position of the CDC or of participating NVDRS states.”

h. Termination and Disqualification

The CDC, in its sole discretion, may terminate the DSA if it determines that the principal investigator(s) and/or any collaborators are in violation of any condition of the DSA and such violation is not remedied within 30 days after the date of written notice of the violation. Furthermore, failure to comply with the DSA may result in the disqualification of the principal investigator(s) and collaborators from having access to the NVDRS data. Violations should be addressed by sending an e-mail to nvdrs-rad@cdc.gov.

i. Duplication of Research

The principal investigator(s) and all collaborators of NVDRS RAD acknowledge that other researchers have access to NVDRS data in the form of public-use datasets and RAD and that duplication of research is a distinct possibility.

j. Destruction of All Sensitive Files at Project Completion

The principal investigator agrees to destroy all NVDRS RAD files, and all derived files three years from the receipt of the data unless otherwise specified. Before the 3 year period expires, the principal investigator can apply to extend this destruction date. Researchers can renew after the allowable access period for up to 2 one-year extensions.

The Participant warrants and represents that he/she/they has the requisite power and authority to enter into this DSA and to perform according to its terms, and that the Participant agreeing to this DSA by signing up for the competition has authority to do so.