Table 3.

Monte Carlo simulation model estimates of number of nonsyndromic critical congenital heart defect (CCHD) cases in 2012 in the United States estimated to be prenatally diagnosed, timely detected, and true positives and false negatives resulting from CCHD screening through pulse oximetry, assuming universal implementation of CCHD screening in all states, in the primary analysis and under scenarios of low and high prenatal diagnosis prevalence

			Results of Monte Carlo simulation using 10,000 iterations Mean (95% uncertainty interval) of the estimated number of CCHD cases in 2012a,b that would be…
Analysis									CCHD screening through pulse oximetry
	Born alive		Prenatally diagnosed			Timely detected			True positives			False negatives
	N	Per 10,000 births	N	(%)	Per 10,000 births	N	(%)	Per 10,000 births	N	(%)	Per 10,000 births	N	(%)	Per 10,000 births
Primary	5,965	15.09	1,800 (1,580–2,020)	(30%)	4.55	2,410 (2,150–2,680)	(40%)	6.10	875 (705–1,060)	(15%)	2.21	880 (700–1,080)	(15%)	2.23
If prenatal diagnosis were universally…
"Low" (~19%)c	5,970	15.10	1,095 (895–1,310)	(18%)	2.77	2,750 (2,415–3,100)	(46%)	6.96	1,105 (885–1,350)	(19%)	2.80	1,020 (805–1,260)	(17%)	2.58
"High" (~ 42%)c	5,965	15.09	2,455 (2,155–2,785)	(41%)	6.21	1,985 (1,720–2,270)	(33%)	5.02	740 (575–925)	(12%)	1.87	785 (610–975)	(13%)	1.99

Notes: Differences in the number born alive are due to rounding

^aCalculated using the number of live births in 2012 (3,952,937)²⁶

^bModel inputs were for each specific CCHD type and these estimates for all CCHD combined were derived by summing across the CCHD categories for each of the 10,000 simulations, then calculating the mean and 95% uncertainty interval across the 10,000 simulations for a given parameter

^cSee Methods and eTable