BIOETHICS

The Bioethics Imperative XXIX
Snowballs! Cases Made Worse by Subsequent Actions

“Mokita”: The truth we all know and agree not to talk about. Papua New Guinea

In a conversation in Washington, DC, in January 2007, James Kroll, head administrator for the National Science Foundation (NSF) Office of Inspector General (OIG), used the term “snowball” for cases that became worse once the OIG began to investigate an allegation. He was kind enough to relay to me some examples of this type of case.

Case 1. An allegation of fraud was self-disclosed by a university supported by NSF funds. This university found that one of its grants managers had embezzled $500,000, a large portion of which came from NSF grants. The administrator had embezzled the funds by cutting honorarium and stipend checks in his wife’s name and cashing them at a local credit union. The NSF OIG concurred with the university’s conclusion, and the employee was prosecuted locally and sentenced to four years in jail.

As part of NSF’s investigation, the OIG asked the university how it had arrived at the dollar amount of NSF funds the administrator had embezzled, because the university’s documentation did not support that figure. As the OIG looked further into the university’s management of NSF grants, they found gross inconsistencies in the documentation supporting claimed costs. This problem was compounded by a lack of documentation for alleged cost sharing associated with these grants. Each time NSF asked the university for further documentation, the final expense numbers changed. After an extensive investigation, the university admitted that they did not have adequate documentation to substantiate most of the costs on these grants.

After a lengthy period of negotiation, the university agreed to settle the matter. In total, $3.3 million was either refunded to the U.S. Treasury or “put to better use” (i.e., “deobligated” and reapportioned to new awards) by NSF.

Case 2. NSF OIG received an allegation of plagiarism in an NSF proposal. Analysis revealed approximately a paragraph of allegedly plagiarized text. Normally, this would not be enough to warrant a full-fledged investigation and would result in a warning. However, in comparing the source document with the proposal, NSF investigators found a pair of figures in both documents that looked similar. Further analysis revealed that the principal investigator (PI) had copied the data figures from the source document, altered them slightly, and included them in his own proposal. However, he went one step further in his proposal and described these data in a way that suggested that the data were preliminary results generated in his laboratory. In addition, the description of the altered data was completely different from the description in the source document.

Finally, both on his CV and in his proposal, the PI referenced two manuscripts that were “in submission” to a professional journal. Wanting to determine if the PI had submitted the altered data as part of the manuscripts, NSF instead found that the manuscripts had never been submitted to the journal.

On the basis of the evidence, NSF debarred the subject for two years for falsification of data and plagiarism.

Next time: “Snowballs!” continued—two more cases.

Dina Mandoli
mandoli@u.washington.edu