We prospectively compared a single-tube multiplex polymerase chain reaction (PCR) for detecting alpha-thalassemia with our current approach using 452 blood samples. Initial evaluation of 89 specimens revealed sensitivity and specificity, respectively, for the hemoglobin H inclusion body test (HbH prep) vs PCR for detecting alpha0-thalassemia carriers of 0.79 and 0.96 and for a mean corpuscular volume (MCV) of 82 microm3 (82 fL) or less, 1.0 and 0.45. Detection of all alpha-thalassemia genotypes was significantly lower for HbH prep and MCV (sensitivity and specificity, respectively: HbH prep, 0.48 and 0.96; MCV, 0.87 and 0.47). In a follow-up evaluation of patients with positive HbH prep results or suspected alpha-thalassemia prescreened by low MCV, the sensitivity and specificity, respectively, of HbH prep vs PCR increased to 0.97 and 0.93 for alpha0-thalassemia and 0.83 and 0.92 for any alpha-thalassemia. PCR detected alpha-thalassemia in 37.2% of 298 suspected alpha-thalassemia cases with suggestive indices but negative HbH prep results and no detectable hemoglobinopathy. This multiplex approach was more sensitive than the HbH prep for detecting all alpha-thalassemia genotypes, particularly alpha+-thalassemia; was particularly valuable for identifying carriers of alpha0-thalassemia at risk for offspring with hemoglobin Bart hydropsfetalis, regardless of other diagnosed hemoglobinopathies; and is an ideal adjunct to standard clinical screening protocols for detecting alpha-globin deletions.