We conducted a pilot study of potential sources of incorrect laboratory reports of human immunodeficiency virus type 1 testing using blind proficiency testing. Sets of three serum samples, including one serum sample with negative reactions in antibody tests, one serum sample with positive reactions, and one that gave false-positive results with certain testing kits, were sent as routine patient specimens to testing laboratories. Half the laboratories reported the serum sample positive for human immunodeficiency virus antibodies as "indeterminate"; one laboratory rendered a final positive report without supplemental testing. On the report forms, the actual laboratory results were often obscured and intermingled with information, sometimes incorrect, such as identifying the agent as "HTLV-III" (human T-cell lymphotropic virus type III) and advising that a test with positive results is evidence of exposure to the virus. Many of these reports have the potential to confuse, rather than to enlighten, the requesting physician.