Effect of the FDA Safety and Innovation Act on racial and gender diversity in neurosurgical device trials.

ABSTRACT

OBJECTIVE: The US FDA uses evidence from clinical trials in its determination of safety and utility. However, these trials have often suffered from limited external validity and generalizability due to unrepresentative study populations with respect to clinical patient demographics. Section 907 of the FDA Safety and Innovation Act (FDASIA) of 2012 attempted to address this issue by mandating the reporting of certain study demographics in new device applications. However, no study has been performed on its effectiveness in the participant diversity of neurosurgical device trials. METHODS: The FDA premarket approval (PMA) online database was queried for all original neurosurgical device submissions from January 1, 2006, to December 31, 2019. Endpoints of the study included racial and gender demographics of reported effectiveness trials, which were summated for each submission. Chi-square tests were performed on both endpoints for before and after years of FDASIA passage and implementation. RESULTS: A total of 33 device approvals were analyzed, with 14 occurring before SIA implementation and 19 after. Most trials (96.97%) reported gender to the FDA, while 66.67% reported race and 63.64% reported ethnicity. Gender breakdown did not change significantly post-SIA (53.30% female, p = 0.884). Racial breakdown was significantly different from the 2010 US Census for all races (p < 0.001) both pre- and post-SIA. Only Native American race was significantly different in terms of representation post-SIA, increasing from 0% to 0.63% (p = 0.0187). There was no significant change in ethnicity. CONCLUSIONS: The FDASIA, as currently written, does not appear to have had a significant impact on the racial or gender diversity of neurosurgical device clinical trial populations. This may be due to the noncompulsory nature of its guidance, or a lack of more stringent regulation on the composition of clinical trials themselves.