As the ongoing Bundibugyo Ebola outbreak spreads across the Democratic Republic of the Congo and Uganda, global attention has understandably focused on the absence of licensed vaccines and therapeutics for this rare species of Ebola virus. Yet one of the outbreak’s most consequential failures has received far less attention: our inability to rapidly and reliably diagnose the pathogen in the first place. The current epidemic, which has caused more than 1,300 confirmed cases and over 375 confirmed deaths in DRC and has spread into neighboring Uganda, initially evaded detection not because clinicians failed to recognize viral hemorrhagic fever, but because many of the diagnostic tools deployed on the frontlines were designed for a different outbreak.

The GeneXpert assays routinely used during Ebola responses were developed primarily to detect Zaire ebolavirus (EBOV), the species responsible for the devastating West African epidemic of 2014-2016 and most subsequent Ebola outbreaks. They did not reliably detect Bundibugyo ebolavirus (BDBV). As a result, patients with Ebola initially tested negative, allowing transmission to continue undetected for critical weeks.

This should be a wake-up call. Now, as the BDBV epidemic continues, Ugandan authorities are investigating cases of Marburg virus disease occurring amid the ongoing Ebola response. According to recent reports, Uganda reported to the World Health Organization that it had detected a Marburg virus disease outbreak in the western part of the country while investigations continue into additional suspected cases.

Whether these reports ultimately represent a larger Marburg outbreak, isolated spillover events, or another pathogen entirely is almost beside the point. The fact that public health officials must even consider the simultaneous circulation of multiple high-consequence pathogens during a single regional emergency highlights a reality that our preparedness systems have failed to fully embrace: Pathogens do not organize themselves according to our testing algorithms. As scientists and physicians who have worked on Ebola outbreaks across Africa, including in West Africa, Uganda, and the Democratic Republic of the Congo, we have watched the current epidemic unfold with a familiar sense of frustration.

Once again, we are learning that one of the greatest barriers to controlling an outbreak is often not the absence of vaccines, therapeutics, or public health expertise. It is the inability to rapidly determine what pathogen we are confronting in the first place. For decades, the global health community’s approach to outbreak diagnostics has been largely reactive and pathogen-specific.

We build tests for the pathogen that caused the last major emergency, deploy them widely, and then discover during the next crisis that a different pathogen is circulating. This cycle has repeated itself across outbreaks of EBOV, Sudan virus, BDBV, and Marburg virus. But the same pattern has characterized our broader approach to infectious diseases.

After SARS emerged in 2003, we developed SARS diagnostics. After the H1N1 influenza pandemic, we expanded influenza testing.