To compare 2 screening protocols performed concurrently in Minnesota: (1) liquid chromatography tandem mass spectrometry steroid profiling as a second-tier test on positive fluoroimmunoassay (FIA) results; and (2) low-birthweight 3-screen protocol (FIA tests at <48 hours, 2 weeks, 4 weeks) on all infants <1800 g, regardless of result.

Population-based study of all <1800 g infants (n = 8739) born in Minnesota from 2004-2010 comparing newborn screening performance metrics of 2-tier (FIA + liquid chromatography tandem mass spectrometry) protocol (2004-2010) vs 1-tier (FIA) low-birthweight 3-screen protocol (2006-2010). False positive (FP) rates were calculated per infant's final confirmatory result. Protocol results used in different time periods (2004-2005 vs 2006-2010) were compared by 2-sample tests of proportions; results of both protocols for 2006-2010 were compared by McNemar test.

First-tier testing of final dried blood spot result (n = 6625) of the low-birthweight 3-screen protocol during 2006-2010 reduced the FP rate more than 5-fold (P < .0001) compared with 2-tier testing of a single dried blood spot (n = 2114) from 2004-2005. In comparing results (n = 6625) of both protocols from 2006-2010, first-tier testing of final dried blood spot accounted for 23% of FPs; second-tier testing of the first dried blood spot accounted for 77%, yielding significantly more FP results (McNemar test, P < .0001).

Timing of dried blood spot collection rather than assay used played a more important role in reducing FP results of congenital adrenal hyperplasia newborn screening in low birthweight infants.