Infants with Congenital Disorders Identified Through Newborn Screening — United States, 2015–2017

Newborn screening (NBS) identifies infants at risk for congenital disorders for which early intervention has been shown to improve outcomes (1). State public health programs are encouraged to screen for disorders on the national Recommended Uniform Screening Panel (RUSP), which increased from 29 disorders in 2005 to 35 in 2018.* The RUSP includes hearing loss (HL) and critical congenital heart defects, which can be detected through point-of-care screening, and 33 disorders detected through laboratory screening of dried blood spot (DBS) specimens. Numbers of cases for 33 disorders on the RUSP (32 DBS disorders and HL) reported by 50 U.S. state programs were tabulated. The three subtypes of sickle cell disease (SCD) listed as separate disorders on the RUSP (S,S disease; S,beta-thalassemia; and S,C disease) were combined for the current analysis, and the frequencies of the resulting disorders were calculated relative to annual births. During 2015-2017, the overall prevalence was 34.0 per 10,000 live births. Applying that frequency to 3,791,712 live births in 2018,† approximately 12,900 infants are expected to be identified each year with one of the disorders included in the study. The most prevalent disorder is HL (16.5 per 10,000), and the most prevalent DBS disorders are primary congenital hypothyroidism (CH) (6.0 per 10,000), SCD (4.9 per 10,000), and cystic fibrosis (CF) (1.8 per 10,000). Notable changes in prevalence for each of these disorders have occurred since the previous estimates based on 2006 births (2). The number of infants identified at a national level highlights the effect that NBS programs are having on infant health through early detection, intervention, and potential improved health, regardless of geographic, racial/ethnic, or socioeconomic differences.

disorders included on the RUSP and to compare these updated prevalence estimates with those previously reported.
The current study is based on data reported for 33 of the 35 disorders included on the RUSP among infants born during 2015-2017, the most recent years of available national data. States reported aggregate numbers of confirmed cases of 32 RUSP DBS disorders to the Association of Public Health Laboratories' Newborn Screening Technical assistance and Evaluation Program (NewSTEPs) (3), a Health Resources and Services Administration-funded data repository. States were requested to apply uniform case definitions established by clinical and public health experts and adopted by NewSTEPs (4). All 50 state programs reported data to NewSTEPs; however, several states were unable to report data for some of the DBS disorders included in this study. In addition, four disorders (severe combined immunodeficiency, glycogen storage disease type II [Pompe disease], mucopolysaccharidosis type 1, and X-linked adrenoleukodystrophy) were added to the RUSP since 2006, for which screening was implemented in some states during the 3-year data collection time frame. Aggregate numbers of confirmed cases of HL among 2015-2017 births were reported from 48 states to CDC's Early Hearing Detection and Intervention (EHDI) Hearing Screening and Follow-up Survey (HSFS) (5). § Colorado did not report HL data for the 2015-2017 period, and Minnesota reported data for 2017 only. The District of Columbia did not report data for any of the DBS NBS disorders, so it was excluded from the analysis.
Because SCD is generally considered a condition comprising multiple subtypes, ¶ the three SCD subtypes on the RUSP were combined into a single disorder for the current assessment. Two RUSP disorders were not included in this assessment: critical congenital heart defects, because few states require reports (6) (2). These data were supplemented with HL data reported to HSFS for 2006.
Prevalence estimates for each of the disorders and all disorders combined are presented for all 50 states (Table), and the prevalence of each disorder is presented by state in a heat map (Supplementary Table, https://stacks.cdc.gov/view/cdc/93107). During 2015-2017, the birth prevalences for any of the disorders, any of the DBS disorders, and HL were estimated at 34.0, 17.5, and 16.5 per 10,000, respectively. Prevalences of individual DBS disorders varied from 0.01 to 6.0 per 10,000. The most prevalent DBS disorders were CH (6.0 per 10,000), SCD (4.9 per 10,000), and CF (1.8 per 10,000); together, these accounted for 73% of all cases of DBS disorders.
The estimated 2006 prevalence of any DBS disorder on the RUSP, other than type 1 tyrosinemia, was 15.6 per 10,000 (6,439 per 4,138,349) births (2). The estimated number of HL cases in 2006 based on HSFS data was 4,097 (9.9 per 10,000) (Table). Thus, the total 2006 prevalence estimate for any of the assessed disorders on the RUSP was 25.5 per 10,000 infants. The RUSP disorders prevalence estimate of 34.0 per 10,000 reported here for infants born during 2015-2017 is a 33% increase since 2006, with more than three-quarters (78%) of that increase driven by HL.
Notable changes in prevalence between 2006 and 2015-2017 occurred for several disorders. Among the more prevalent DBS disorders, the observed rate during 2015-2017 was lower than the modeled rate in 2006 for CF (-1.19 per 10,000) and higher for SCD (0.65 per 10,000) and CH (0.79 per 10,000). The observed rate for HL during 2015-2017 was substantially higher (16.5 per 10,000) than the rate based on 2006 HSFS data (9.9 per 10,000). Variable prevalences by state were observed, with HL, CH, and SCD being the most prevalent in most states (Supplementary Table,  Although the overall prevalence of DBS disorders increased from 2006 (15.6 per 10,000) to 2015-2017 (17.5 per 10,000), changes in individual disorder prevalence estimates varied. Random variation and small numbers might have affected the estimates for each period. Notable changes in prevalence for each of the three most prevalent DBS disorders were observed. First, the lower prevalence of CF during 2015-2017 compared with 2006 might reflect a reduction in live births with CF under the influence of widespread neonatal and prenatal screening and reproductive counseling (7). Second, the increase in CH prevalence might be a continuation of long-term trends related to a higher proportion of U.S. births to Hispanic parents, among other factors (8). Finally, the higher prevalence of SCD might reflect more births to parents originating from countries where SCD is relatively common § § (9). Variations in prevalence of individual disorders across states might reflect differences in the geographic distribution of disease-causing genetic variants and differences in screening methods, case definitions, follow-up, and reporting practices.
The findings in this report are subject to at least four limitations. First, the prevalence rates calculated for the newest disorders that have been added to the RUSP are based on data from only a few states across a short period, making these rates less robust than the rates for the other disorders. However, changes in estimated prevalences for these rare disorders are not likely § § https://www.pewsocialtrends.org/2015/04/09/a-rising-share-of-the-u-s-blackpopulation-is-foreign-born/. to have a large impact on the overall rate of infants identified with an NBS disorder. Second, the study did not include two disorders on the RUSP (spinal muscular atrophy and critical congenital heart defects) because of lack of data; future studies could incorporate these disorders once reliable national data are available. Third, differences among the reporting practices of both NBS and EHDI programs potentially limit the interpretation of these data. Although NewSTEPs recommends uniform case definitions (5), not all NBS programs applied these definitions to the cases submitted. Finally, state prevalence estimates for individual disorders might be affected by 1) newborns born in one state and screened in another or 2) newborns from surrounding states born and screened in a state other than their resident state; however, national-level estimates would not be affected. A strength of the study is that the prior prevalence estimates of NBS disorders are modeled estimates based on four states, whereas the current study relies on reported numbers of disorders from 50 states.
The number of infants identified by NBS at a national level highlights the scope of the effect that NBS programs are having as they identify infants at risk for significant morbidity and mortality and refer them for recommended intervention. NBS continues to be a major public health achievement, offering population-based early detection, intervention, and potential improved outcomes to all infants, regardless of geographic, racial/ethnic, or socioeconomic differences (10).