An evaluation of the reliability of HIV partner notification disposition coding by disease intervention specialists in the United States.

BACKGROUND: The reliability of CDC HIV partner notification (PN) disposition codes has not been evaluated.
METHODS: Disease Intervention Specialists (DIS) working for health departments in high HIV/STD-morbidity metropolitan areas completed a questionnaire that presented vignettes describing PN interviews. Questionnaires asked DIS to indicate whether they would record a disposition and what codes they would assign to each partner.
RESULTS: A total of 136 DIS from 28 of 29 eligible states participated. Partner 1: The index case says he will inform his partner of his HIV diagnosis and, at follow-up, reports that the partner has tested negative. Seventeen percent of DIS indicated they would record a partner disposition. DIS used 7 different codes to define the PN outcomes. Partner 2: The index case says she will inform her partner, who attends the clinic, indicates no history of testing, and tests HIV-negative. 93% of DIS reported they would record a disposition, 90% of whom used code 6, "Not Previously Tested, New Negative." Partner 3: The index case with partner 2 (above) agrees to have DIS notify her second partner. When contacted, the partner tells DIS that he had previously tested negative and will arrange to be tested himself. He subsequently reports testing HIV-negative, but DIS do not confirm this. Seventy-three percent of DIS recorded a disposition for the partner, of whom 84% used code J, "Located, Refused Counseling and Testing."
CONCLUSIONS: CDC HIV PN disposition codes are reliable for simple scenarios with verified outcomes, but less reliable when DIS elicit partner-reported outcomes.