Awaiting definitive diagnosis before scheduling healthcare visits complicates HIV screening and referral. Clients screened by rapid tests as initially reactive often fail to return to receive definitive test results, are not linked to care and enter care late or not at all.To evaluate statewide, (1) the accuracy of a single-visit, two test HIV rapid testing algorithm (RTA) and (2) its effect on referral to care for positive clients.A two-test RTA was implemented at 24 sites in New Jersey beginning in December 2008. All clients with a reactive rapid HIV test were offered a second rapid HIV test, and RTA results were compared with Western blot (WB). Referral to care occurred based upon two sequential positive rapid tests.The RTA program has screened 51,413 individuals obtaining 426 reactive rapid test results; 394 (92.5%) were reactive by a second rapid test, 32 (7.5%) had a negative second rapid test. Twenty-eight individuals refused WB testing. Of 369 RTA-positive individuals who have WB results, 368 (99.5%) were confirmed positive. Of RTA-positive clients, 290 (73.6%), including 25 (6.6%) who refused Western blot, were immediately referred for care including one individual with a false-positive RTA.The RTA reduced false positive results by 6.2% and agreed with WB results 99.5% of the time. Improved referral to care compared to traditional rapid HIV screening occurs when immediate referral is based on RTA verification of a preliminary positive rapid test. WB confirmation is not essential for effective screening and contributes to difficulties linking individuals to care.