Multiple puncture skin test and mantoux test in Southeast Asian refugees.

Skin test reactions to PPD applied by Mantoux techniques were compared with reactions to tuberculin tine test (PPD-tine); tuberculin, old tine test (OT tine); Aplitest; and Mono-Vacc, tuberculin, old (Mono-Vacc) in newly arrived refugees from Cambodia and Laos. The reaction to Mantoux test was accepted as the "true reading" and compared to the reaction size to one of the multiple puncture tests (MPT). A 2 X 2 table was constructed and sensitivity, specificity, false negative, and false positive rates computed over a broad range of cut-off points for MPT. The MPTs were very sensitive (100 percent to 78 percent) but lacked specificity (78 percent to 18 percent) when a cut-off of 1 mm was used. Predictably, as the cut-off is moved to larger reactions, sensitivity decreases and specificity increases. The relationship between the two is emphasized in a receiver-operator characteristics analysis. The MPTs are not intended for diagnostic use. Reaction to the MPT should be interpreted with the same careful consideration that clinicians use to interpret reactions to the Mantoux test.