Search tips
Search criteria 


Logo of pediatricsLink to Publisher's site
Pediatrics. 2011 October; 128(4): 715–722.
PMCID: PMC3182843

False-Positive Newborn Screening Result and Future Health Care Use in a State Medicaid Cohort



To compare health care visit rates between infants with false-positive and those with normal newborn screening (NBS) results.


We analyzed administrative claims of Medicaid-enrolled infants born in Michigan in 2006 and calculated the average number of outpatient, emergency department, and hospital visits for infants aged 3 to 12 months according to NBS results. We calculated an adjusted incidence rate ratio for each visit category, adjusting for covariates and accounting for interaction effects.


Of the 49 959 infants in the analysis, 818 had a false-positive NBS result. We noted a significant interaction between gestational age and NBS results. We found that preterm, but not term, infants with false-positive results had more acute outpatient visits than their counterparts with normal NBS results. We found no difference in adjusted rates of other visit types (emergency department, inpatient, outpatient well) between infants with false-positive and normal NBS results, regardless of gestational age.


Increased rates of acute outpatient visits among preterm infants with false-positive NBS screening results may be attributable to underlying chronic illness or parental anxiety. The absence of increased health care utilization among term infants may be unique to this Medicaid population or a subgroup phenomenon that was not detectable in this analysis.

Keywords: newborn screening, false-positive results, health care service utilization, infants


There has been long-standing concern that parents of children with false-positive results for newborn-screening (NBS) tests overuse health care services because they believe their child to be persistently vulnerable to illness. Previous studies that have explored this issue have yielded conflicting results.


These analyses revealed that Medicaid-insured term infants with false-positive NBS results did not have more health care visits than infants with normal NBS results. However, preterm infants with false-positive NBS results had more acute outpatient visits than those with normal NBS results.

Newborn screening (NBS) is a mandatory state-run public health program by which newborns are screened for inherited disorders. In 2005, the American College of Medical Genetics recommended that states expand their NBS panels to screen for 29 disorders. Nearly all states have complied with this recommendation.1 However, screening for additional rare disorders increases the number of false positive results even when a highly specific test is used.2

There is concern that false-positive results increase parents' anxiety about their child's future health, leading to greater use of health care services.3 A recent study did not find evidence of increased health care visits (outpatient, emergency, inpatient) in children with false-positive NBS results compared with those with normal NBS results.4 However, this study may have been subject to recall and selection bias because it involved voluntary participation and assessment of health care use by use of parental report. With that in mind, our objective was to conduct a population-based study by using state Medicaid claims to explore whether infants with false-positive NBS results had greater future health care use than infants with normal NBS results.


This study was approved by the institutional review boards of the University of Michigan and the Michigan Department of Community Health.

Source Population

The source population for this study was all infants born in Michigan in 2006 who had ≥11 months of Medicaid eligibility and no other health insurance .

Matching Medicaid Claims With NBS Data

The institutional review boards of the University of Michigan and the Michigan Department of Community Health required that the process of matching infants' Medicaid claims with their NBS result be conducted by the Michigan Department of Community Health and that all subsequent data analysis be conducted on a deidentified data set. To improve the validity of the match, a previously validated 3-step linkage process5 was used: (1) NBS records were matched to live birth records; (2) Medicaid claims were matched to live birth records; and (3) NBS records were linked to Medicaid claims. Concordant/discordant data were investigated by hand after the final linkage was completed. Disorders included in the Michigan NBS panel are listed in Appendix 1.


We excluded infants from our analysis who met any of the following criteria: died before 12 months of age; eligible for Medicaid but did not have any claims or encounters; unknown gestational age; true-positive NBS results (Fig 1).

Study population.

Independent Variables

Gestational Age, Birth Weight

Gestational age and birth weight information were obtained from each infant's NBS card. Internal validity checks between the NBS card and the vital records, for which we used a cutoff of ≥37 weeks for term infants, demonstrated excellent concordance (98% concordance). For those infants for whom gestational age was not available from the NBS card, we determined gestational age by using International Classification of Diseases, Ninth Revision, codes for gestational age from birth claims. We identified and excluded incompatible combinations of gestational age and birth weight by using a previously published method.6

Child Gender

The infant's gender was obtained from the Medicaid beneficiary table.

Birth Date

Because this analysis was conducted on a deidentified data set, we had to estimate each infant's birth date. We did so by subtracting the age in months at the time of service from the date of service to calculate a birth date for each claim. The earliest calculated birth date among an infant's set of claims served as the estimated birth date for that infant.

Utilization Outcomes

We measured health care use by counting the number of visits (represented by submitted claims) for outpatient (well and acute visits), emergency department (ED), and inpatient services. If a child had multiple visits within the same visit category (outpatient well, outpatient acute, ED, inpatient) on a given day, then only 1 visit was counted in the analysis. If an infant had multiple visits in different visit categories on a given day, each visit was counted individually with 1 exception: if an infant had an ED visit with a date of service that was identical to the admission date for an inpatient claim, then we assumed that the infant had been admitted to the hospital from the ED and categorized this visit as an inpatient visit or hospitalization.

We counted only visits from 3 to 12 months of age because infants with false-positive NBS results may use health care services in the first 2 months of life for evaluation of their positive screening result. Because preterm infants often have lengthy hospital stays after birth, we checked to see how many preterm infants were still in the hospital at 2 months of age. Of those preterm infants for whom we had birth admission claims (n = 4840), 95% were discharged from the hospital before 2 months of age.

We classified a claim as an outpatient visit for acute care if the claim contained any of the following Current Procedural Terminology (CPT) evaluation and management (E&M) codes: 99201–99205 (new patients); 99211–99215 (established patients); and 99241–99245 (office consultation). We used the following CPT E&M codes to identify outpatient visits for well-child care: 99381–99386 (new patients) and 99391–99397 (established patients).

We analyzed ED visits, observation care visits, and urgent care visits as 1 visit category because all of these visits occur in similar care settings and similar services are provided. We classified a claim as an ED visit if a claim contained either of the following: a revenue code (045x) or a CPT E&M code (99281–99285). We classified a claim as an observation care or urgent care visit if the claim had any of the following codes: 99217–99220, 99234–99236, or G0381–G0383 (CPT codes) or 700 (cast room), 760–762 (specialty room/observation room), or 769 (other treat/observation room) (revenue codes).

We identified inpatient visits as claims that had a billing revenue code of 100 through 179 or 190 through 219 and 2 of the 3 following elements: admission date; discharge date; length of stay. When an infant was transferred to another hospital during an admission, we counted the infant as having a single hospitalization.

Chronic Illness

Because chronic illness can confound the relationship between NBS result type and health care utilization, we accounted for chronic illness among the study population by using the Clinical Risk Group (CRG) software system developed by 3M (Wallingford, CT). With this software we analyzed administrative claims for visits, procedures, and medication use and assigned children to 9 different health status groups (healthy, significant acute illness, and 7 chronic illness groups). In previously published studies this software was used to assess chronic illness in a cohort of pediatric patients who were covered by Medicaid.7,8 For this analysis, we created a dichotomous chronic illness variable: no chronic illness (a CRG status of healthy or significant acute) versus chronic illness (any of 7 CRG chronic illness health status levels). Each individual assigned to a health status level also had a primary chronic disease (PCD) that represented the most significant chronic condition for which the infant was under active treatment. We have provided a list of the top 5 PCDs for infants with false-positive and normal NBS results (Appendix 2).

Statistical Analysis

Our objective for this study was to compare outpatient, ED, and inpatient visits between infants with false-positive NBS results and infants with normal NBS results. We performed χ2 tests on demographic characteristics stratified according to NBS result type (Table 1). We also calculated the number of infants with false-positive results for various disorders or metabolites, categorized these results either as an endocrine or a metabolic abnormality, and stratified them according to gestational age (Table 2).

Demographics of the Study Population
Disorders/Analytes for False-Positive Results

We calculated the average and total numbers of visits in each health care visit category according to gestational age (preterm versus term). Using negative binomial regression to account for overdispersion,9 we calculated an unadjusted and adjusted incidence rate ratio (IRR) for each visit category. The adjusted model included covariates for gender, birth weight, gestational age, and chronic illness. Prompted by our clinical experience, we explored an interaction effect between gestational age (preterm versus term) and NBS screening result (false-positive versus normal). We then calculated the predicted rates of visits on the basis of the presence of the interaction. By slicing interactions, we tested the difference in visit rates between infants with false-positive and infants with normal NBS results for each gestational age (eg, preterm and term) in those visit categories in which the interaction was significant.


Study Population

After we applied exclusion criteria to the matched study population, there were 49 141 infants with normal NBS results and 818 with false-positive NBS results. Of the term infants with false-positive results, 282 had an endocrine abnormality and 178 had a metabolic abnormality (Table 2). Infants with false-positive results were more likely to be male, have a birth weight of <2500 g, and have a chronic illness (Table 1).

Health Care Visits

We calculated the visit rates for infants for each type of heath care visit category (Table 3). In unadjusted analyses, there was a statistically significant increase in acute outpatient visits (IRR: 1.36 [95% confidence interval (CI): 1.26–1.46]), emergency visits (IRR: 1.16 [95% CI: 1.05–1.28]), and hospitalizations (IRR: 2.71 [95% CI: 2.14–3.44]) for infants with false-positive NBS results but no difference for outpatient well visits (IRR: 1.02 [95% CI: 0.97–1.07]). After adjustment for gender, birth weight, gestational age, and chronic illness, infants with false-positive NBS results remained more likely to have additional acute outpatient visits (IRR: 1.07 [95% CI: 1.00–1.15]) but not outpatient well visits (IRR: 1.0 [95% CI: 0.96–1.05]), emergency visits (IRR: 0.95 [95% CI: 0.86–1.05]), or hospitalizations (IRR: 1.19 [95% CI: 0.96–1.47]).

Health Care Visit Rates (Unadjusted) of Children According to Newborn Screening Result and Gestational Age

However, we noted differences in visit rates across different health care visit categories according to gestational age (Table 3). We then found a significant interaction between gestational age and NBS result type (eg, false-positive or normal) in the adjusted model for outpatient acute visits (P = .004) but not for any of the other types of health care visit categories. The adjusted predicted mean rates for acute outpatient visits were highest for preterm infants who received false-positive results even after we adjusted for chronic illness (Fig 2). After adjustment of data for confounders and interaction effects, preterm infants with false-positive NBS results had more acute outpatient visits than preterm infants with normal NBS results (P = .001). However, term infants with false-positive NBS results did not have more acute outpatient visits than term infants with normal NBS results (P = .64).

Figure 2. Adjusted predicted mean visit rates for children according to NBS result and gestational age.


Since the early days of NBS, there has been concern that parents of infants with false-positive NBS results overuse health care services because they believe their child to be persistently vulnerable to illness.10 Previous studies that have used voluntary enrollment of study participants instead of a population-based methodology have yielded conflicting results.3,4

In this study, which represents the largest and only population-based analysis to date, we found that only preterm infants with false-positive NBS results had more acute outpatient visits than their preterm counterparts with normal NBS results. We did not find such a difference among term infants. Furthermore, regardless of prematurity status, we found no difference in outpatient well visits, ED visits, or hospitalizations between infants with false-positive and those with normal NBS results.

There are a number of possible explanations for our findings. First, it is possible that parents of premature infants with false-positive NBS results make more acute outpatient visits because they are overanxious about both the child's history of prematurity and the false-positive NBS result. However, others might counter that parents of preterm infants deal with so many concerns about their child's tenuous health that parents are unlikely to pay much attention to a false-positive NBS result. Second, it is conceivable that preterm infants with false-positive NBS results have more acute outpatient visits because of their underlying chronic illness. Premature infants are known to have a higher rate of false-positive results because of illness-related stress11 and the use of screening cutoffs based on studies of term infants.12,13 So it is possible that, despite our best efforts, we were unable to fully control for the confounding influence of chronic illness on health care use in preterm infants with false-positive results. Nonetheless, additional studies should be performed to examine the psychosocial effects of a false-positive NBS result on parents of preterm infants. Finally, although case reports have identified cases in which infants with false-positive results have been inappropriately diagnosed as disease free,14 this phenomenon is likely to be uncommon and thus is not likely to account for our study findings.

We did not find any differences between infants with false-positive and those with normal NBS results for other health care visit types (eg, well visits, ED visits, hospitalizations), regardless of prematurity status. There are a number of plausible explanations for this null finding. First, it is possible that the phenomenon of parental anxiety that leads to increased health care use15 is present only in certain subgroups, such as parents who had difficulty conceiving, those with pregnancy complications, or even first-time parents. These subgroups may be “primed” to become concerned about the lasting significance of their child's false-positive NBS result. If so, it is unlikely that these subgroups will be identified in an aggregate analysis of a study population.

Alternatively, it may be that there are certain types of false-positive results, such as those for disorders that can be more immediately life-threatening, such as MCADD (medium chain acyl CoA dehydrogenase deficiency). Hearing that their newborn may have a potentially life-threatening illness leave a more powerful and lasting effect on parents' perceptions of their child's health than if they had been told that that the disorder was not immediately life threatening.16 We had neither sufficient numbers of these disorders nor the necessary parental medical and social information to explore either of these hypotheses in the current analysis.

It is conceivable that our findings are unique to a continuously enrolled Medicaid population. Children with private insurance have greater outpatient health care use than children with public insurance.17 Although some of this discrepancy in use may be because of lack of access to care among the publicly insured, it could also be fueled by the phenomenon of the “worried well.” Future studies will need to examine the relationship between false-positive NBS results and health care use in a privately insured population.

There are limitations of this study that should be noted. As evidenced by the fact that every child in our analysis did not have a claim record for their birth admission (n = 8729 [17%]), it is likely that our collection of claims did not represent all of the health care visits for the study population. Nonetheless, we have no reason to suspect there would be a differential bias for missing claims according to NBS result. As noted above, we may not have completely controlled for chronic illness, and so our results may be explained, in part, by residual confounding. Finally, because we examined only health care use in the first 12 months of life, children may not have had enough opportunities to seek acute care for illness. As a result, we might have failed to detect a difference between infants with false-positive and normal NBS results that would become evident in the next few years of life when the ratio of acute to well health care visits rises.

In addition, it is important to recognize that increased health care utilization is only one potential manifestation of parental anxiety about false-positive NBS results. Although we did not find a link between false-positive NBS results and increased health care visits, we cannot rule out that parental anxiety about false-positive results may have been present and could conceivably have been manifested through other actions, such as decisions to forego future childbearing or restriction of the child's activity.18 Additional studies are needed to explore the presence and scope of maladaptive parental behaviors that might result from persistent parental anxiety regarding false-positive NBS results.


We believe that our study findings highlight the potential complexity of the relationship between false-positive NBS results and future health care use. We were unable to find a difference in health care visits between term infants with false-positive results and those with normal NBS results in a Medicaid population. However, because a parent's worry about his/her child's health predicts health care use for that child,15 it may be that the relationship between false-positive NBS results and health care use exists only in certain subgroups of the population who are prone to being more worried about their child's health. Nonetheless, in this study we have demonstrated that it is important that future studies of the effect of false-positive results on future health care use account for the potential interaction effect between NBS result and gestational age.


This project was funded by the Robert Wood Johnson Health and Society Scholars Small Grant Award, the Michigan Institute for Clinical and Health Research, and an intramural grant through the department of pediatrics at the University of Michigan. This work was conducted with a research license from 3M corporation and was not funded by the 3M corporation. Dr Tarini was supported by the Clinical Sciences Scholars Program at the University of Michigan and a K23 Mentored Patient-Oriented Research Career Development Award from the Eunice Kennedy Shriver National Institute of Child Health and Human Development (K23HD057994).

We acknowledge Lisa Cohn and Dr John Neff for their helpful comments with data analysis and Dr Piero Rinaldo for his thoughtful comments during the preparation of the manuscript.


Disorders Included in the Michigan NBS Panel, 2006

Phenylketonuria (PKU)
Isovaleric acidemia (IVA)
Benign hyperphenylalaninemia (H-PHE)
2-Methylbutyryl-CoA dehydrogenase deficiency (2MBG)
Biopterin cofactor biosynthesis (BIOPT [BS])
3-Methylcrotonyl-CoA carboxylase deficiency (3MCC)
Defects of biopterin cofactor regeneration (BIOPT[Reg])
3-OH 3-CH3 glutaric aciduria (HMG)
Maple syrup disease (MSUD)
3-Methylglutaconic aciduria (3MGA)
Homocystinuria (HCY)
Beta-ketothiolase deficiency (BKT)
Hypermethioninemia (MET)
Glutaric acidemia type I (GA I)
Citrullinemia (CIT)
Propionic acidemia (PA)
Citrullinemia type II (CIT II)
Methylmalonic acidemia (mutase deficiency) (MUT)
Argininosuccinic acidemia (ASA)
Methylmalonic academia (MA), also known as cobalamin A or B deficiency (CblA, CblB)
Tyrosinemia type I (TYR I)
Methylmalonic academia (MA), also known as cobalamin C or D deficiency (CblC, CblD)
Argininemia (ARG)
Multiple carboxylase deficiency (MCD)
Carnitine:acylcarnitine translocase deficiency (CACT)
2-Methyl 3 hydroxy butyric aciduria (2M3HBA)
Carnitine palmitoyltransferase II deficiency (CPT II)
Malonic acidemia (MAL)
Carnitine uptake defect (CUD)
Isobutyryl-CoA dehydrogenase deficiency (IBG)
Carnitine palmitoyltransferase IA deficiency (liver) (CPT 1A)
Congenital adrenal hyperplasia (CAH)
Short-chain acyl-CoA dehydrogenase deficiency (SCAD)
Congenital hypothyroidism (CH)
Glutaric acidemia type II (GA II)
Galactosemia (GALT)
Medium-chain acyl-CoA dehydrogenase deficiency (MCAD)
Biotinidase deficiency (BIOT)
Long-chain L-3-OH acyl-CoA dehydrogenase deficiency (LCHAD)
Sickle cell anemia (HbSS)
Trifunctional protein deficiency (LCHAD/TFP)
HbS/C disease (HbS/C)
Very long-chain acyl-CoA dehydrogenase deficiency (VLCAD)
HbS/β-thalassemia (HbS/β-Th)
Medium-chain ketoacyl-CoA thiolase deficiency (MCKAT)

CoA indicates coenzyme A.

Data source: Korzeniewski SJ, Young WI, Andruszewski K, Hawkins HC. Michigan Newborn Screening Program, Annual Report, 2006. Lansing, MI: Michigan Department of Community Health. Available at:


Top 5 PCD Categories According to NBS Result Type

PCD Categoryn
Infants with false-positive NBS results
    Major respiratory anomalies65
    Other chronic pulmonary diagnoses29
    Blindness, visual loss, and chronic eye diagnoses, major/moderate23
    Chronic endocrine, nutritional, fluid, electrolyte, and immune diagnoses12
Infants with normal NBS results
    Major respiratory anomalies352
    Ventricular and atrial-septal defects340
    Other chronic pulmonary diagnoses318
    Hydrocephalus, encephalopathy, and other brain anomalies181

A portion of this work was presented at the Pediatric Academic Societies Meeting; May 1-4, 2010; Vancouver, British Columbia, Canada.

The views expressed herein do not necessarily represent those of the University of Michigan, the Eunice Kennedy Shriver National Institute of Child Health and Human Development, the Michigan Department of Community Health, or 3M.

FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.

Funded by the National Institutes of Health (NIH).

newborn screening
emergency department
Current Procedural Terminology
evaluation and management
incidence rate ratio
confidence interval


1. National Newborn Screening and Genetics Resource Center National Newborn Screening Status Report, 2010. Austin, TX: University of Texas Health Science Center; 2010
2. Tarini BA, Christakis DA, Welch HG. State newborn screening in the tandem mass spectrometry era: more tests, more false-positive results. Pediatrics. 2006;118(2):448–456 [PubMed]
3. Waisbren SE, Albers S, Amato S, et al. Effect of expanded newborn screening for biochemical genetic disorders on child outcomes and parental stress. JAMA. 2003;290(19):2564–2572 [PubMed]
4. Lipstein EA, Perrin JM, Waisbren SE, Prosser LA. Impact of false-positive newborn metabolic screening results on early health care utilization. Genet Med. 2009;11(10):716–721 [PMC free article] [PubMed]
5. Korzeniewski SJ, Grigorescu V, Copeland G, et al. Methodological innovations in data gathering: newborn screening linkage with live births records, Michigan, 1/2007–3/2008. Matern Child Health J. 2010;14(3):360–364 [PubMed]
6. Dombkowski KJ, Leung SW, Gurney JG. Prematurity as a predictor of childhood asthma among low-income children. Ann Epidemiol. 2008;18(4):290–297 [PubMed]
7. Neff JM, Sharp VL, Muldoon J, Graham J, Myers K. Profile of medical charges for children by health status group and severity level in a Washington State Health Plan. Health Serv Res. 2004;39(1):73–89 [PMC free article] [PubMed]
8. Neff JM, Sharp VL, Muldoon J, Graham J, Popalisky J, Gay JC. Identifying and classifying children with chronic conditions using administrative data with the clinical risk group classification system. Ambul Pediatr. 2002;2(1):71–79 [PubMed]
9. Cameron LD, Sherman KA, Marteau TM, Brown PM. Impact of genetic risk information and type of disease on perceived risk, anticipated affect, and expected consequences of genetic tests. Health Psychol. 2009;28(3):307–316 [PubMed]
10. Rothenberg MB, Sills EM. Iatrogenesis: the PKU anxiety syndrome. J Am Acad Child Psychiatry. 1968;7(4):689–692 [PubMed]
11. Torresani T GA, Scherz R, Burckhardt JJ, Harras A, Zachmann A. Improving the efficacy of newborn screening for congenital adrenal hyperplasia by adjusting the cut-off level of 17-hydroxyprogesterone to gestational age. Screening. 1994;3:77–84
12. Zytkovicz TH, Fitzgerald EF, Marsden D, et al. Tandem mass spectrometric analysis for amino, organic, and fatty acid disorders in newborn dried blood spots: a two-year summary from the New England Newborn Screening Program. Clin Chem. 2001;47(11):1945–1955 [PubMed]
13. Slaughter JL, Meinzen-Derr J, Rose SR, et al. The effects of gestational age and birth weight on false-positive newborn-screening rates. Pediatrics. 2010;126(5):910–916 [PubMed]
14. Browning MF, Larson C, Strauss A, Marsden DL. Normal acylcarnitine levels during confirmation of abnormal newborn screening in long-chain fatty acid oxidation defects. J Inherit Metab Dis. 2005;28(4):545–550 [PubMed]
15. Janicke DM, Finney JW, Riley AW. Children's health care use: a prospective investigation of factors related to care-seeking. Med Care. 2001;39(9):990–1001 [PubMed]
16. Grob R. Is my sick child healthy? Is my healthy child sick? Changing parental expexzriences of cystic fibrosis in the age of expanded newborn screening. Soc Sci Med. 2008;67(7):1056–1064 [PubMed]
17. McCormick MC, Kass B, Elixhauser A, Thompson J, Simpson L. Annual report on access to and utilization of health care for children and youth in the United States: 1999. Pediatrics. 2000;105(1 pt 3):219–230 [PubMed]
18. Bergman AB, Stamm SJ. The morbidity of cardiac nondisease in schoolchildren. N Engl J Med. 1967;276(18):1008–1013 [PubMed]

Articles from Pediatrics are provided here courtesy of American Academy of Pediatrics