Table 1

Pathogen exclusion criteria

To remove noise and address potential confounding driven by changes in outbreak detection capacity over time, we applied strict exclusion criteria to the viral zoonotic pathogens in our epidemiological database (online supplemental table 1). This box shows the exclusion criteria applied to viruses in the epidemiological database, the rationale for each exclusion criterion, and examples of viruses that were excluded based on each criterion. Note that pathogens may be excluded by multiple criteria.
Exclusion criterionRationaleExclusion examples
Has caused 100 or more annual cases for five consecutive yearsReporting effort for endemic pathogens varies substantially by country and throughout time, with endemic diseases in low-income countries often being significantly underreported.25 Advances in public health capacity, surveillance technology, and surveillance effort over time could show an increase in outbreaks and deaths caused by these pathogens without a true increase in their occurrence.MERS Coronavirus, Lassa virus, Monkeypox virus, Hantavirus
Fewer than 50 reported deathsPathogens that have caused minimal human mortality are generally not prioritised for diagnostic testing development and surveillance. These pathogens are more likely to go undetected; however, advances in healthcare and diagnostic technology are likely to increase the probability of detecting a spillover event over time. This criterion also excludes most non-vector-borne pathogens that have no documented human-to-human transmission, as the focus of this study is on pathogens that may cause significant epidemics.Hendra virus, Lujo virus, Whitewater Arroyo virus
Vector-borneThe mechanism of spillover for vector-borne pathogens is different from non-vector-borne pathogens and may be influenced by different factors. Additionally, vector control/eradication programmes are likely to influence the frequency and severity of vector-borne pathogen spillover events.Venezuelan equine encephalitis, Crimean-Congo haemorrhagic fever virus, Zika virus
InfluenzaThere are specifically targeted influenza surveillance programmes that have increased significantly over the time period being analysed,26 which could confound any temporal increase seen in spillover events or number of deaths.2009 H1N1 pandemic