Misclassification of the stages of syphilis: implications for surveillance.

SHORT
SUMMARY: Syphilis cases were reviewed to see if reported stages met the Centers for Disease Control and Prevention case definition. Classification was excellent for primary and secondary and good for late latent, but half of early latent and unknown duration were misclassified. New surveillance definitions are suggested, comments requested.
BACKGROUND: Uncertainty when staging latent syphilis should lead clinicians to call it late latent (requires more treatment) and disease investigators to call it early latent (priority for partner investigation). Accurate surveillance requires consistent case definitions.
OBJECTIVE: Assess validity of reported syphilis stages.
METHODS: Record reviews in 6 jurisdictions to determine if reported cases met the Centers for Disease Control and Prevention case definitions.
RESULTS: Nine hundred seventy-three records from 6 jurisdictions in 2002 showed excellent agreement for reported primary (94.0%) and secondary (95.4%), good agreement for late latent (80.2%), and poor agreement for early latent (48.4%) and unknown duration (49.7%). Unknown duration (age < or =35 and nontreponemal test titer > or =32) was often misinterpreted to mean "not known." Early latent (within the past year, documented: seroconversion, fourfold titer increase, symptoms, or contact with an independently documented early syphilis case) was often misinterpreted to include patients with risky behavior, young age, or high nontreponemal test titers.
CONCLUSIONS: The unknown duration stage should be dropped. Surveillance of latent syphilis would be more consistent if cases were reported as having high or low titers on nontreponemal test. Alternative approaches are solicited from readers.