Screening for syphilis in arrestees: usefulness for community-wide syphilis surveillance and control.

Background: Syphilis screening of jail arrestees has been promoted as an effective method for both disease control and surveillance.

Goals: To evaluate the yield of the East Baton Rouge Parish Jail screening program in detecting previously undiagnosed syphilis, to evaluate the program as a means for monitoring community syphilis rates, and to characterize arrestees at greatest risk for syphilis infection.

Study Design: From July 1994 to December 1998, arrestees brought to the East Baton Rouge Parish Jail were screened for syphilis. Annual early syphilis prevalence in screened arrestees was calculated and compared with the annual period prevalence of early syphilis in the general population of East Baton Rouge Parish, as reported by laboratories and health providers. A case-control study of cases detected at the jail from 1995 to 1997 and contemporary controls was conducted.

Results: A total of 50,941 arrestees were booked into the East Baton Rouge Parish Jail, of whom 38,573 (76%) were screened for syphilis. Of the 38,573 arrestees screened, 494 (1.3%) were diagnosed with untreated syphilis. Of these, 299 (61%) were treated for syphilis before release. The estimated prevalence of early syphilis in arrestees decreased by 68% during the study period, from 0.79% in 1994 to 0.25% in 1998. During this time, the East Baton Rouge Parish community rates decreased by 79%, from 150 cases per 100,000 to 31 cases per 100,000. In female arrestees, a booking charge of prostitution was associated with syphilis (odds ratio [OR] 7.0; 95% CI, 1.5, 39.3). In male arrestees, a booking charge of felony theft was associated with syphilis (OR 4.8; 95% CI, 1.8, 13.8). However, only 15 (12%) of the early syphilis cases would have been detected if screening had been based on the booking charges found to be associated with syphilis in this study.

Conclusions: Routine syphilis screening and treatment in jail settings is feasible and identifies many persons with syphilis. Monitoring of syphilis prevalence among arrestees is a useful method for monitoring community prevalence of syphilis. Analysis of booking charges may be useful for determining factors associated with syphilis infection, but not for developing screening criteria.