Shortly after the severe acute respiratory syndrome (SARS) epidemic subsided, nine experts were invited by this journal to predict whether the epidemic would recur.¹ A review of their opinions two years after the epidemic may prove insightful.

SARS has not recurred in epidemic proportions since it ended in July 2003. Since September 2003, several laboratory associated cases occurred: in Singapore, Taiwan, and Beijing, the latter involving two secondary and five tertiary cases.² In addition, three sporadic cases occurred in Guangzhou from December 2003 to January 2004.³ One patient, a restaurant waitress, was linked to wildlife (civets) while the source of infection of the other cases remained unclear. No more SARS cases were reported since 30 April 2004.⁴

Of the nine experts, three predicted the recurrence of the epidemic in autumn/winter. Three (including the author) correctly predicted that the epidemic was unlikely to recur. Two correctly predicted the return of SARS, but did not predict whether as an epidemic or smaller outbreaks, and one abstained from prediction. Most experts agreed on the importance of vigilance in public health and hospital measures. None mentioned the possibility of laboratory outbreaks. The analogy of the epidemiology of SARS to the periodicity of influenza epidemics made by some experts was misplaced—SARS peaked in April–May 2003, the end of the respiratory virus season. The assumption that SARS might be more efficiently transmitted in winter has been queried. More evidence that civets are animal reservoirs is found. Antibodies to SARS-CoV among civets in Guangdong farms were found.⁵ Clinically inapparent infections were detected among animal market traders (highest in civet traders) in Guangzhou but a healthy human SARS carrier (either patients who recovered, or asymptomatic seropositive people) does not exist. These findings put the focus on wild animals and the laboratory as the most probable sources of future outbreaks. That sporadic cases continued to surface (both in the laboratory and in the community) is an ominous signal. Further research of the role of wild animal species that have close contact with humans is needed. While civets might be responsible for one case in Guangzhou,⁴ the unknown source of infection of the other community acquired cases shows the shortcomings of contact tracing and disease investigation procedures, and also suggests the possibility of yet unidentified sources in the environment.