Eliminating access to anonymous HIV antibody testing in North Carolina: effects on HIV testing and partner notification.

Anonymous HIV testing may attract persons who might otherwise not be tested but may hinder partner notification. We evaluated the effects on North Carolina's HIV testing and partner notification programs of policy changes that eliminated and later restored anonymous testing in 82 counties. We used an interrupted time-series design to compare counties eliminating with counties retaining anonymous testing. We analyzed HIV testing and partner notification data from before, during, and after elimination of anonymous testing. After elimination of anonymous testing in 82 counties, the mean monthly level of testing (+/- SE) increased by 45%, or 548 (+/- 123) tests per month, while in 18 counties that retained anonymous testing, there was a 63% increase, or 802 (+/- 162) tests per month (p > .05). Among men of all races, testing increased by 16%, or 155 (+/- 35) tests per month, in counties that eliminated anonymous testing; and by 51%, or 305 (+/- 42) tests per month (p < .05), in counties that retained anonymous testing. After elimination of anonymous testing, both county types experienced similar increases in the rate of partners notified. However, partner notification was more successful if the index patient was tested confidentially; 2.7 times as many partners per index patient were notified and counseled. There was no effect on testing or on partner notification rates following restoration of anonymous testing. Substantial community opposition to eliminating anonymous testing was encountered. The policy change appeared to result in a slight decrease in testing among men and a slight increase in partners notified. Programs considering the elimination of anonymous testing should weigh these potential gains and losses, as well as the impact on relationships between the public health and advocacy communities