Using AI to Identify Gender-Based Violence in Brazilian Health Records and Enable Earlier Intervention

1

A Recife pilot examined 13,000 documented gender-based violence cases and found that women’s clinic visits began an average of 92 days before violence was recorded.
2

ClarIA analyzes clinical notes and medical histories for linguistic patterns associated with gender-based violence, rather than relying on isolated keywords.
3

ClarIA issues amber alerts for patterns resembling confirmed cases and red alerts for women previously identified or reporting violence.
4

After deployment began in Recife in March, ClarIA operated across 39 clinics and hospitals, and clinicians submitted 8 violence notifications after receiving alerts.
5

Women with unreported assaults faced 8 times the risk of death compared with women whose assaults were reported; more than 60% of identified femicide victims died within 30 days of notification.

1