Abstract: Objective To investigate the performance of acute flaccid paralysis(AFP) surveillance system in Shenzhen city.Methods The simplicity,data quality,sensitivity,timeliness,and acceptability of the AFP surveillance system was evaluated based on surveillance data analysis and ad hoc survey.Results A total of 136 AFP cases were reported in Shenzhen city from 2009 through 2012.Among the 14 diseases with clinic manifestation of AFP,polyneuropathy,single neuritis,plexitis,botulism,and transient limb paralysis were considered as much more difficult to be diagnosed.Of all the cases,23.5% were with incomplete information about status of seeking medical service,clinical signs or symptoms,and the history of vaccination.During the 4-year period,the yearly reported incidence of AFP among the children under age of 15 years was 1.8,2.2,4.8,and 4.6 per 100 000,respectively.Among the reported AFP cases,the proportion of the cases with one to four times of outpatient clinic visit before the confirmation of AFP diagnosis was 56.6%,30.2%,3.7%,and 2.9%,respectively.There were 93.8%(15/16) of the staff in centers for disease control and prevention,100%(22/22) of the medical professional in healthcare department of a hospital,and 95.5%(21/22)of the health workers in disease surveillance department considering that the current surveillance system could identify patients with AFP promptly.Conclusion The AFP surveillance system in Shenzhen city runs well,but some aspects of the system need to be improved.