Gender bias in medical device research: Why disaggregated data matters
By Matt Hanson, Senior Research Executive and Guy Pasquill, Data Analytics Manager
As awareness of gender based differences in health outcomes continues to grow, there is a valuable opportunity to examine Post Market Survey data more deeply. Post Market Surveys are essential tools for monitoring the safety and performance of medical devices in real-world settings, yet the full potential of this data is often underutilised. By analysing the data through a more granular lens, including disaggregation by gender, we can uncover meaningful insights, identify differential responses, and ultimately drive more equitable outcomes for all patient groups. Doing so also supports alignment with regulatory expectations around inclusive and evidence based post-market surveillance.
The absence of gender analysis is a finding in itself
One of the most striking observations from a recent internal review across a range of projects is not what was found, but what wasn’t. Across a diverse set of device types and therapy areas, gender was frequently omitted from the data collection process altogether. In cases where it was collected, it was rarely analysed. This absence of gender-based analysis means that potential disparities in adverse event (AE) occurrence between male and female patients go unnoticed, unreported, and unaddressed.
This issue is especially concerning given the FDA’s 2025 guidance which explicitly states:
"Where available background information or clinical study results suggest there are clinically meaningful sex differences, you should include this information in interim reports and in the results section of your final report. If warranted, you should also submit revised labelling to include this information."
— FDA Guidance for Industry and FDA Staff, March 2025
Therefore, we should be including gender specific findings in post market reports when clinically meaningful differences are suspected. The guidance also encourages sponsors to enrol and retain study populations that reflect the gender specific prevalence of the condition being treated, a standard that many standard Post Market data collection plans do not include.
When gender is considered, differences emerge
In the subset of Post Markey Survey projects where gender data was available and analysed, clear patterns began to surface. Across the reviewed studies, adverse events were more frequently reported in female patients than in male patients. While the magnitude of these differences varied, the consistency of the trend was notable. In fact, only a small fraction of studies showed no gender based variation in AE occurrence.
These findings underscore the importance of disaggregating data by gender. Even when differences appear modest in isolated studies, longitudinal analysis across multiple waves of data collection could reveal more significant trends. This is particularly relevant for devices used in chronic or long-term therapy areas, where cumulative exposure and physiological differences may amplify risk over time.
Unlocking the potential of Post Market data to advance equity
The implications of these findings are clear: gender bias in medical device research is not just a theoretical concern, it’s a measurable gap in the way data is collected, analysed, and applied. But this also presents an opportunity. Post Market studies generate a wealth of real-world data that, when used effectively, can uncover meaningful patterns and differences in device safety and performance across patient populations.
By incorporating gender disaggregated analysis into Post Market activities, we can proactively identify and address potential disparities, enhance regulatory alignment, and support more inclusive, patient-centred healthcare. The good news is that this doesn’t require a complete overhaul, even small adjustments in data collection and reporting practices can lead to significant improvements. For our clients that means using the data we’re already gathering to build a stronger, more equitable foundation for medical technology, one dataset at a time.