The following case study is based on an actual, ongoing request for data. It highlights some very negative aspects of uncontrolled data sharing.

Under a state-specific Public Records Statute, data collected by public employees, including faculty at the state university, become a public record and therefore the property of the state. In addition, this state law specifies that such data must be made available upon request.

Recently, an epidemiologist at the state university completed a large health and environment survey in a population residing in small communities adjacent to large livestock-breeding facilities. They noted a correlation between proximity to these livestock-breeding facilities and adverse health outcomes.

The lobby group for this particular segment of the agricultural industry wanted to see the data from this survey. The attorneys for the lobby group sent a letter to the investigators requesting the data. In the letter, the attorneys requested any and all descriptions of the scope of the study, the work conducted, the results, opinions, and other conclusions drawn from the study. They also requested any and all documentation of the results, including notes and other records from site visits and interviews; samples and tests; any calculations or other forms of analysis; any documents defining the scope of the study; information on sources of support for the study; and the identities of all persons who worked on or contributed to the study, including interviewees. The attorneys wanted to evaluate the veracity of the findings linking proximity to livestock-breeding facilities and health outcomes. Moreover, they wanted to assess whether reporting findings were defamatory. They requested that the data be delivered within 5 days. If the data were not delivered within that time frame, they were filing an action in the state court to compel the investigators to provide the data and to seek other remedies.

The investigators were suitably alarmed by the letter and brought it to the attention of the attorneys for their university. Upon advice of counsel, the investigators provided a redacted data set, one stripped of individual identifiers and sampling maps to protect the identity of the subjects and communities that participated in this study.

The attorneys for the agricultural lobby found these data to be not fully responsive to their request. They wanted to know the areas where the research was conducted and the identity of the farms whose effects were being measured. Specifically, they were requesting the paper copies of questionnaires, paper copies of coding sheets that were sent to the data entry service, study area maps, larger area maps of the study areas, and the electronic data base and paper copies of the master interview log sheet. While they had retreated from the earlier position of asking for subjects’ names and addresses, the investigators feared that location data sought in the second request plus the redacted data set would allow the attorneys to pinpoint participants. The attorneys, in turn, claimed that without this data, they could not investigate the validity of the results. Again, the attorneys mentioned the possibility of taking action with regard to defamation of character of the livestock breeding industry.

Upon advise of counsel, the investigators provided paper copies of the questionnaire, paper copies of coding sheets for interview data, and the electronic database and paper copies of the master interview log sheet. However, the investigators removed all personally identifying information before making copies. They feared that the powerful livestock breeding group would harass the participants in this study, as they had harassed others critical of their industry. None of the maps were released because they would put the identity of individual research subjects at risk.

This is where this case currently stands.