In the rush to address COVID-19 institutions face increased risk of data manipulation and research fraud

With the rush to understand the novel coronavirus, and grapple with its widespread effects on public health, the pressure to deliver the “right” results continues to grow and new research studies seem to be published daily delivering new conclusions about the disease. While research-based studies and public health data are often accepted at face value as objective evidence – and rarely scrutinized closely by the public at large – those relying on this new wave of science should approach with caution.  In the scramble to both cure the COVID-19 disease and restore the US economy to pre-COVID levels of activity, researchers from all disciplines are being asked to operate on incomplete information and forgo the usual rigors of the scientific process. In the more extreme cases, scientists are even being asked to manipulate data to produce results that fit political benchmarks.

Take, for example, the high-profile dismissal of a Florida Department of Health employee who claims that she was fired after she refused to manipulate data contained in the state’s central data dashboard, a repository for publicly tracking COVID-19 cases and related information in Florida.  The top public health researcher who managed the Geographic Information System team at Florida’s Department of Health claims that she was removed from the project after she refused a request to manually change data to “drum up support for the plan to reopen.” Her termination raises serious questions of the integrity and reliability of the data that is serving as the basis for local officials’ decision whether to reopen the economy.

While the US Department of Health and Human Services has long cautioned against such data manipulation as reprehensible research fraud, this newly-reported trend of manipulating critical public health statistics may spark even greater scrutiny by federal oversight agencies and lead to harsher consequences, including criminal prosecution, for individuals and entities that are suspected to have deliberately manipulated COVID-19-related research records.

The pressures to produce data that confirms the public’s bottom line is not unique to public health research. Researchers tasked with studying the novel coronavirus and its effects are facing increasing pressure to abandon traditional safeguards to expedite publication of new scientific “discoveries” that answer pressing questions about COVID-19. More and more frequently, research teams are opting to forego traditional journals and publish new data in pre-print online websites where researchers can post studies that have not yet been reviewed or published. The danger, of course, is that these new studies do not benefit from the scrutiny implemented under the traditional peer review process, which is credited with producing stronger and better supported scientific conclusions. While traditional scientific journals are doing their part to meet the demand for COVID-19 research – often expediting their internal review process and relaxing customary processes in favor of fast results – the pre-print phenomenon continues to gain traction in the scientific community.

Antibody tests are another example of the potential consequences of rushed and unvalidated science. Under the guise of “eliminating red tape” in order to get tests for COVID-19 antibodies on the market, the FDA permitted drug companies and others to “self-validate” the effectiveness of their own tests in detecting such antibodies, without any FDA review. The result: Dozens of such tests – highly sought by the public in the hope of establishing immunity to the virus – have recently entered the market and have become more available to the public.

But the Centers for Disease Control and Prevention has now reported that many of these tests have results that are wrong 50% of the time. This makes such tests worse than useless, as millions of people potentially could have received false positives on which they improperly relied in believing they had antibodies, were immunized from the virus, and thus could return to a normal lifestyle, without precautions and without fear of getting contaminated or contaminating others. The FDA’s abdication of its vital responsibility to review and approve such tests before they are used, in order to protect the public, has undoubtedly hastened the spread of the virus.

While science is doing its best to meet the public’s demand for answers, those institutions who oversee the research cannot lose sight of the importance of data integrity. The need for rapid results and increased transparency could not be greater as we battle the COVID-19 crisis and scientists continue to be a critical piece of fighting the pandemic. But scientists engaged in the research efforts and journals and institutions championing those results must remain diligent about the quality of research published and continue to employ well-established experimental rigors to the extent they are able. More importantly, institutions funding research should remain vigilant of the external pressure to produce data that supports predetermined hypotheses and not abandon traditional research integrity measures to ensure the accuracy and validity of all data that emanates from their labs. The failure to ferret out manipulation and fraud in a rush to obtain results may have long-lasting and serious consequences, opening up those responsible for publishing unverified results to criminal and civil enforcement after the dust settles.