By Mark Helfand, M.D., M.S.a; Diane Thompson, M.S.b; Robert Davis, M.D., M.P.H.c; Heather McPhillips, M.D., M.P.H.d; Tracy L. Lieu, M.D., M.P.H.e; Charles J. Homer, M.D., M.P.H.f.
Address correspondence to: Mark Helfand, M.D., M.S.; Division of Medical Informatics and Outcomes Research, Oregon Health & Science University; 3181 S.W. Sam Jackson Park Road; Portland, OR 97201. E-mail: helfand@ohsu.edu
Select for copyright, and reprint information.
The summaries of the evidence briefly present evidence of effectiveness for preventive health services used in primary care clinical settings, including screening tests, counseling, and chemoprevention. They summarize the more detailed Systematic Evidence Reviews, which are used by the U.S. Preventive Services Task Force (USPSTF) to make recommendations.
Introduction
Methods
Results
Comment
Acknowledgments
References
Notes
Each year, approximately 5,000 infants are born in the United States with moderate-to-profound, bilateral permanent hearing loss (PHL). Estimates of the incidence of moderate, severe, and profound congenital PHL among newborns range from 1 in 900 to 1 in 2,500.1-7 Congenital PHL is associated with delayed language, learning, and speech development.8-12 This delay is measurable as early as age 3 years13 and has consequences throughout life. On average, deaf students graduate from high school with language and academic achievement levels below those of fourth-grade students with normal hearing.14,15
Diagnosis and treatment are often delayed until ages 1 or 2 in children with congenital PHL, particularly among children at low risk for PHL.16-20 Current theory views auditory stimuli during the first 6 months of life as critical to development of speech and language skills.21-23 Advocates of universal newborn hearing screening (UNHS) believe that earlier application of available therapies, such as speech and language therapy, amplification, and family support, could reduce or eliminate the gap in language skills between deaf and hearing children.24,25
Selective screening of high-risk newborns is an alternative to UNHS. The incidence of PHL varies with race, birthweight, and other risk factors. Among infants in a neonatal intensive care unit (NICU), the risk of moderate-to-severe PHL is 10 to 20 times higher than in the general population.26 In addition to NICU admission, the Joint Committee on Infant Hearing high-risk guidelines specify four other risk factors (Table 1).27 From 10 percent to 30 percent of newborns meet these criteria, which can identify 50 percent to 75 percent of all cases of moderate-to-profound bilateral hearing loss.2
In 1995, the U.S. Preventive Services Task Force (USPSTF) found insufficient evidence to recommend UNHS.28 It argued that, among low-risk infants, the prevalence of hearing impairment was very low, and substantial numbers of infants would be misclassified. It found that evidence for the efficacy of early intervention in patients diagnosed by screening was incomplete, but endorsed selective screening of high-risk newborns based on the higher prevalence of disease in this group.
Since 1995, many health care professionals and Federal health care agencies have advocated UNHS, which is now mandated by law in 32 States.16,27,29 Is widespread support for UNHS now justified? To update the USPSTF recommendations, we critically reviewed recent evidence to identify strengths, weaknesses, and gaps in the evidence supporting UNHS.
Table 1. Risk factors for sensorineural hearing loss in newborns*
*Joint Committee on Infant Hearing (JCIH) criteria for identifying infants at high risk for hearing loss.27 |
We focused our literature search on key questions underlying the clinical logic behind screening for hearing impairment in newborns (Figure 1, 95 KB; Text Version). The logic assumes that screening tests are accurate; that screening reduces delays in diagnosis and treatment; that earlier treatment results in better language function within the preschool period; and that this improvement in early language function will improve educational, occupational, and social function later in life. Moreover, for UNHS to be preferred over selective screening, the potential benefits of early detection and treatment must be realized in the subgroup of newborns who have no risk factors and would not otherwise be screened.
We searched MEDLINE®, CINAHL, and PSYCINFO for relevant papers published in English from 1994 to September 2000, using the keywords hearing disorders and infant or newborn combined with terms for screening and relevant treatments, such as early intervention, amplification, and American Sign Language (select Appendix 1 for complete search strategy). The search was updated quarterly through August 2001. We also examined reference lists of review articles,7,30-37 and queried experts. To identify articles published before 1994, we relied on systematic reviews published in 199628 and 1997.18
The first author and at least 1 other author reviewed titles and abstracts of 864 articles and selected 340 articles as possibly relevant to 1 of the key questions. Of these, we selected to include in evidence tables:
Ten studies of the yield of universal screening programs,24,38-46 1 study of the accuracy of OAEs and ABR in high-risk infants,47 and 8 studies of language outcomes13,48-54 met these inclusion criteria. Two authors abstracted data on population, test performance, outcomes, and methodological quality from each included study. We classified each study as "good," "fair," or "poor" using prespecified criteria developed by the USPSTF for grading the internal validity of studies and the overall evidence for each link in the analytic framework (Appendix 2).55 When necessary, we sought additional information needed to apply the criteria from authors.
We constructed a mathematical model of the likely benefits and harms of UNHS versus selective screening of 10,000 newborns, estimating prevalence, sensitivity and specificity, compliance, treatment effect size, and other model parameters from the included studies. We used this evidence to prepare a technical report, summarized by the present article. The project team included investigators from the Oregon Health & Science University Evidence-based Practice Center, the University of Washington Departments of Pediatrics and Epidemiology, and two representatives from the USPSTF. The entire 13-member Task Force discussed the review, examined and rated the quality of 4 key studies of early intervention, and provided overall guidance.
Does UNHS improve the yield of screening, compared with selective screening of high-risk newborns?
Ten publications24,38-46 provided information about the yield of UNHS and the performance of OAE and ABR in actual screening programs (Table 2). The 10 studies include 1 controlled trial, 4 State-based programs, and 5 hospital-based programs. Overall, screening detected 1 case of moderate-to-profound PHL for every 465 to 925 infants screened; from 779 to 2,794 low-risk, and 86 to 208 high-risk, newborns were screened to find 1 case. In these studies, the proportion of infants diagnosed to have significant hearing loss who had no risk factors varied from one-fourth to over two-thirds. Screening the low-risk or well-nursery population resulted in identification of 5 of 27 (18.5 percent) hearing-impaired infants in the Wessex trial,38 7 of 22 infants (32 percent) in the Whipps Cross study,44 8 of 15 (53 percent) hearing-impaired infants in the Hawaii study,39 and 2 of 6 (33 percent) in North Carolina.46 All of the U.S. studies that reported results for low-risk and high-risk groups separately defined "high-risk" as those who had NICU admission. The New York program examined differences between the NICU and well-baby nursery in detail. Overall, 1 in 884 newborns screened had bilateral hearing loss. In the NICU, where 90 percent of babies had other risk factors, 1 in 125 had hearing loss. In the well-baby nursery, where 30 percent had risk factors, 1 in 1,042 had hearing loss.42,56
How often do false-negative and false-positive screening test results occur?
Either the OAE or the ABR is used as the initial test in screening. Criteria vary for defining a "pass" or "fail" on the initial screening test, and results are sensitive to equipment, the tester's training, and ongoing quality control. Most programs use a two-stage approach, in which an infant who fails the initial test is retested and is referred for audiologic evaluation only if he or she fails the second test. False-negative and false-positive rates can be calculated based on the results of the initial test or on the overall results of the two stages of screening.
Three studies, a controlled trial of UNHS in Wessex,38 a hospital based program in England,44 and a report of statewide screening in Rhode Island,24 provided some information about sensitivity and about the false-negative rate of the screening test (1-sensitivity). These studies reported the number of cases missed by screening and eventually diagnosed by other means, but they did not make a comprehensive effort to follow babies who had normal screening test results. The false-negative rates were 15 percent,38 6 percent,24 and 11 percent.44
A nonrandomized, controlled trial of screening was conducted at four hospitals in the Wessex district of the United Kingdom.38 Over 3 years, neonatal screening alternated with usual care every 4 to 6 months in four maternity hospitals. During the periods of neonatal screening, 21,279 of 25,609 eligible children (83 percent) had a transient evoked otoacoustic emission (TEOAE) test, followed by ABR testing for those with a positive TEOAE test. Newborns with positive ABR results were referred for audiological testing. All children in both the screened and unscreened groups received the existing screening program—the health visitor distraction test (HVDT)—at about 8 months of age. Children who did not pass the HVDT were also referred for audiological testing. The two-stage screening protocol identified 23 of 27 (85 percent) infants who proved to have hearing loss upon followup.57 In the Rhode Island program, 5 of 79 (6 percent) infants with bilateral hearing loss passed birth screening but were diagnosed between 5 and 22 months of age by other means.24
False-positive rates vary among centers and depend on the strategy and timing of testing. In the Wessex trial, the false-positive rate of the overall, 2-stage screening procedure was 1.5 percent (specificity = 98.5 percent). Therefore, for every 1,000 normally hearing newborns who completed screening, 15 were referred for a full audiologic evaluation because of false-positive screening test results. The false-positive rate fell from 1.9 percent on the first postnatal day to 1.1 percent on days 2 through 4.
If an infant has a positive result on the screening test, how likely is it that the infant has hearing loss? Because the prevalence of congenital hearing loss is low, there are many more false positives than true positives; as a result, the positive predictive value (PPV) (number of infants with hearing loss and a positive test divided by the total number testing positive) is also low.
The programs in Table 2 used a two-stage screening protocol, in which an infant who fails the initial test (an OAE or ABR) is retested, either in the hospital or as an outpatient within 12 weeks of discharge, and is referred for audiologic evaluation if he or she fails the second test. The PPV can be calculated for either the first stage or the second stage of screening. If both stages are performed while the infant is in the hospital, the PPV of the second-stage test determines who will be recalled for followup testing as an outpatient. In one good-quality study, the overall PPV for the second-stage screening test was 6.7 percent.38 In the well-baby nursery, the PPV was 2.2 percent, meaning that 1 of every 45 infants referred for outpatient audiologic evaluation eventually proved to have moderate-to-profound bilateral SNHL. For high-risk babies the PPV was 20 percent (18/90). None of the other studies in Table 2 provided sufficient data to determine the PPV for moderate-profound bilateral PHL.
The gold standard determination of permanent hearing impairment for validating results of screening tests is a combination of otolaryngological and audiological consultation, diagnostic ABR testing, and other electrophysiological testing.58 These assessments have traditionally been performed after 6 months of age, but in some programs are done as early as 2 months of age. The reliability of the gold standard—behavioral and/or audiologic evaluation—increases with the age at which it is performed.
In the Wessex trial, the first audiometric examination was done when the babies were between 8 and 12 weeks of age. Of 158 infants who screened positive, 27 were diagnosed to have permanent sensorineural hearing loss; in 2 of these cases (7.4 percent), however, the "gold standard" diagnosis was wrong, and the babies proved to have normal hearing when re-examined at 4 months or 10 months of age.57 This suggests that, even when formal diagnostic evaluation is performed following screening evaluations, 2 of 27 infants were misdiagnosed. Based on the 95 percent confidence interval (CI), the number of misdiagnosed cases could range from 0.24 to 6.5 infants. In another study,44 5 of 17 (29 percent) infants initially diagnosed to have moderate PHL were later found to have only mild hearing loss. None of the other studies in Table 2 followed patients long enough to determine when the audiometric diagnosis of PHL was incorrect.
In UNHS programs, how many children are identified and treated before 6 months?
One indicator of the benefit of UNHS is the number of additional cases of significant hearing impairment that are diagnosed early. The Wessex trial did not directly compare the rate of early diagnosis and treatment for UNHS to that of selective screening of high-risk newborns. It did compare UNHS to no newborn screening, followed in both groups by HVDT at 8 months of age. In the Wessex trial,38 for infants with moderate-to-profound hearing impairment, UNHS increased rates of referral to an audiologist by age 6 months (an increase of 51 per 100,000; CI, 7.4-94.0 per 100,000; P=.03), but did not increase rates of confirmation of diagnosis (P=.22) or initiation of management within 10 months (P=.08). Among those with moderate or severe hearing loss, however, screening led to highly significant increases in confirmation and management by 10 months of age. With UNHS, 13 out of 23 (57 percent) children with moderate or severe impairment were diagnosed by 10 months, whereas during the period of time without UNHS, only 2 out of 13 (14 percent) with hearing impairment were identified by then.38 UNHS did not reduce the rate of diagnosis after 18 months, either overall (5/27 for UNHS vs 6/26 for the control group) or in the moderate-to-severe subgroup.
How much of the overall benefit in the Wessex trial can be attributed to screening low-risk infants? Compared with selective screening of high-risk newborns, universal screening diagnosed an additional 13 cases of moderate-to-profound bilateral hearing loss per 100,000 screened, or 1 additional case for every 7,692 screened, before 10 months of age.
Because the 9 other studies in Table 2 were uncontrolled, the effect on the timing of diagnosis, compared with selective screening of high-risk newborns, cannot be estimated. Some of them reported decreases in the age at diagnosis over time. During the 4 years of UNHS in Rhode Island, the mean age of hearing loss detection decreased from 13.3 months prior to implementing UNHS to 5.7 months by year 4.24 In Hawaii, the average age of hearing-loss identification and fitting with hearing aids decreased as the percent of the population screened by UNHS increased. When 19 percent of the population was screened, the mean age of identification was 12 months and the mean age of amplification was 16 months. In the last year of the program, when 95 percent of the population was screened, the average ages of identification and amplification were 3 months and 7 months, respectively.59 In the Whipps Cross study,44 performed in the United Kingdom, the mean age at amplification was 4.2 and 13.8 months for children with profound and moderate hearing loss, respectively.
Four of the eight observational studies reported the mean age at the time of treatment. For hearing aid fitting, the mean age for all patients was 5.7 months,24 5.8 months,39 and 7.5 months60 in the three U.S. studies. In the Whipps Cross study,44 performed in the United Kingdom, the mean age at amplification was 4.2 months for children who had profound hearing loss and 13.8 months for children who had moderate hearing loss. None of these estimates included children who, although screened, did not return for followup testing or treatment (that is, they were not calculated on the appropriate intention-to-treat basis). None of the studies reported information about the technical success of fitting, including how often the hearing aids were used.
The ages of diagnosis in the screening studies were all considerably earlier than those reported in a national survey.20 The validity of this comparison is limited, because estimates from the screening studies did not include children who, although screened, did not return for followup testing or treatment (that is, they were not calculated on the appropriate "intention-to-treat" basis). These cases are included in surveys of the time lags in usual care, which are assessed retrospectively and therefore include children diagnosed at a later date, making it likely that the age at diagnosis will appear better in studies of screening than in surveys of usual care.
No prospective, controlled study directly examined whether newborn hearing screening results in improved speech, language, or educational development. None of the State-based programs described in Table 2 reported the outcomes of treatment for infants identified to have hearing impairment.
One retrospective cohort study compared language performance in hearing-impaired children detected by UNHS (n=25) to unscreened children (n=25).51 All study subjects were participants in the Colorado Home Intervention Program (CHIP), a program that provides hearing aids and home visits for children with hearing loss.61,62 Children born after 1996 in a hospital that employed UNHS and who did not have significant cognitive delays were compared with children born since 1992 in hospitals without a UNHS program. Subjects were matched on degree of hearing loss (mild, moderate, moderately-severe, profound), cognitive quotient (CQ), and age at time of speech language evaluations. The 2 groups were similar in gender, ethnicity, presence of multiple disabilities, mode of communication, education of primary caregiver, and chronological age.
Mean scores for expressive, receptive, and total language were within normal range for the screened group and 18 to 21 points higher (P <.001) than the unscreened group (expressive language 82.9 [SE 3.7] vs 62.1 [SE 4.3]; receptive language 81.5 [SE 3.7] vs 66.8 [SE 4.0]; total language 82.2 [SE 3.3] vs 64.4 [SE 3.9]). A 20-point gap is more than 1 standard deviation lower than normal for age, which would indicate that a child with average intellect would have the language abilities of a child who had an IQ of 80. Children identified prior to 6 months (whether in the screened or unscreened group) had a smaller gap between language development and cognitive ability than children identified after 6 months. Language development was within normal range for 56 percent of the screened group compared to 24 percent of the unscreened group.
While this study used relevant, validated measures of language outcomes and controlled for several important potential confounders, the creation of the study groups and description of the patients limited the conclusions that could be drawn. Eligibility for the screened group was determined by the availability of an assessment of language outcomes at 2 to 4 years of age. Because the groups were drawn from different hospitals and time periods, factors other than exposure to UNHS might have influenced outcomes. Selection of subjects and assessment of outcome were unblinded, and neither the number of excluded subjects, nor the reasons for exclusion, are reported.
Older studies comparing early-identified to late-identified children with impaired hearing consisted of clinical series or case-control studies of highly selected patients, with heterogeneous causes of hearing loss, inconsistent definitions of early diagnosis, incompletely defined treatment regimens, and inadequate control for potential confounders.18,28 None of these older studies examined the outcome of delayed diagnosis in children who have no risk factors for hearing impairment at birth.
Table 3 summarizes methodologic aspects and results of 8 retrospective cohort studies from 3 intervention programs.13,48-54 All of these studies used standardized receptive and expressive tests to evaluate speech and language skills in preschool children, and all reported statistically significant associations between the age at the time of diagnosis and language development at 2 to 5 years of age. Adjusted mean scores for expressive and receptive language were 15 to 20 points higher in groups of children identified and treated early compared to the later identified groups.
Five studies reported speech and language results for children enrolled in the Colorado Home Intervention Program.13,48-50,52 The most widely cited of these studies compared 72 hearing-impaired children identified prior to 6 months of age to 78 hearing-impaired children identified after 6 months.49 After adjustment for cognitive function, children whose hearing losses were identified by 6 months of age demonstrated significantly better receptive, expressive, and total language scores than children identified after 6 months of age. For children with normal cognitive abilities, this language advantage was found across all test ages, communication modes, degree of hearing loss, and socioeconomic strata. The children identified before 6 months of age had language scores at or near their cognitive test scores, whereas children identified after 6 months of age performed, on average, 20 points lower on language scores than cognitive scores. Children with low cognitive abilities (CQ <80) experienced a smaller improvement in total language, but no statistically significant improvement in receptive and expressive language abilities.
The groups of early and late diagnosed children differed: late-identified children were more likely to be cognitively impaired, to have severe or worse hearing loss, to use sign language, and their mothers were less likely to have finished high school. The statistical method used in the analysis did not simultaneously adjust for more than 2 factors and may not have removed the influence of these differences. Additionally, the study did not provide data on dropout rates in the 2 groups, and outcome assessments were not masked.
Another CHIP study evaluated factors related to expressive language development in a group of 113 deaf and hard-of-hearing children.52 It reported that expressive vocabulary was higher with increased age, increased CQs, identification of hearing loss by the age of 6 months, and having a hearing loss as the only medical condition.
Additional evidence for the effect of early identification and treatment was provided by a retrospective study of 112 children enrolled for at least 6 months in a diagnostic early intervention program in Nebraska.53 After adjustment for family involvement, degree of hearing loss, and nonverbal IQ, children enrolled prior to 11 months had stronger vocabulary and reasoning skills than children enrolled at later ages. At age 5, family involvement accounted for 57 percent of variance in vocabulary, and age of enrollment accounted for 11.5 percent. In one study, a retrospective series of 80 children in a home intervention program in Washington State,54 early enrollment was associated with better language skills at 3 years of age. The relevance to newborn screening is low because only nine subjects were enrolled before 12 months of age.
The studies in Table 3 had several important limitations. The study populations were composed of convenience samples. That is, the studies compared children who were identified early and late by means other than UNHS, rather than children whose age at identification and enrollment was determined primarily by whether or not they were screened. None of the studies had clear criteria for inclusion, none had blinded assessments, and all selected children for inclusion based on the availability of a language assessment between ages 2 to 5. This could introduce bias: early-identified children who remained in the program may have had better results than early-identified children who were not available for followup. Because of these limitations, selection bias cannot be confidently ruled out as an explanation for the findings.
Moeller53 found family involvement an important contributor to language development. Since other studies did not adjust for this factor, they may have overestimated the association of early enrollment with language development. None of the studies provides information on attrition or followup rates. The USPSTF rated the strength of evidence linking early treatment with improved language function "inconclusive" and the quality of evidence as "fair/poor."
Screening
Potential adverse effects of false-positive screening tests include misdiagnosis, parental misunderstanding and anxiety, and unfavorable labeling. As noted earlier, the "gold standard" determination of PHL is imperfect; in expert hands, as many as 7 percent of infants diagnosed to have PHL may eventually prove to have normal hearing. The frequency of misdiagnosis in everyday practice settings has not been studied.
Another potential adverse effect of screening is parental anxiety. In the Wessex trial, parents whose babies were screened had similar anxiety and attitudes to parents in the unscreened group. It should be noted that, before screening was done, parents in the screened group received information about the benefits of early identification and gave informed consent for the procedure. In the Whipps Cross hospital study,63 among parents whose infants failed the initial screen and received a second test, 2 of 57 (3.5 percent) reported they were very worried. In a survey at a regional hospital in Logan, Utah (n=169), parents indicated acceptance of newborn screening for their infants, 98.2 percent of parents said they would give permission for screening, 95.3 percent would prefer screening even if the baby failed, and 84.9 percent felt that anxiety caused by failing a screening test would be outweighed by the benefits of early detection.64 In another survey of non-NICU infants who failed hospital screening, 46 of 49 (94 percent) parents of infants who had false-positive screening test results approved of UNHS. However, 4 of 49 (8 percent) mothers said they treated their child differently (e.g., spoke louder or clapped their hands), and 7 of 49 (14 percent) reported "lasting anxiety" after the second screening exam even though hearing was normal.46 No study attempted to assess the effect of parental anxiety or changes in parental behavior on infants' development or on the parent-infant relationship.
Treatment
The harms of early intervention have not been adequately studied. As noted by the previous USPSTF, differing ethical and philosophical attitudes about deaf awareness and culture have led to controversy about the content of early interventions.28 The argument for early intervention is based on the prevailing theory of language development, which holds that early auditory input is an important precursor of language development. An opposing viewpoint expressed in the literature is that, during infancy, nonverbal communication, joint attention, shared experiences, and mutual understanding are more important precursors of language development than are hearing speech and forming sounds. Proponents of this view theorize that early intervention could harm infants because it leads parents to focus on "means of communication the child has the least prerequisites for" and on the baby's disability instead of its competencies.65 Because there are no randomized trials of different management strategies, it is impossible to assess the merits of these concerns.
Table 4 summarizes the benefits and harms of UNHS and selective screening in a hypothetical cohort of 10,000 newborns. With UNHS, an additional 7,800 screening tests would be done, resulting in the diagnosis of 6 additional cases of moderate-to-profound hearing loss diagnosed before 10 months of age. Of these, three additional cases would be treated before 10 months of age. Thus, the number needed to screen (NNS) to detect 1 additional case before 10 months would be 1,441, and the NNS to treat 1 additional case before 10 months would be 2,401. With UNHS, 254 newborns would be referred for audiological evaluation because of false-positive second-stage screening test results, versus 48 for selective screening. Of these, one would be falsely diagnosed to have PHL.
Of the 6 additional early-diagnosed, low-risk newborns, how many would actually benefit from early treatment? The data needed to estimate this—the probabilities of a poor language outcome with and without early treatment—are not known. To use a hypothetical example, if 50 percent of low-risk newborns would have poor language ability if diagnosed after 10 months, and early intervention reduced this by 50 percent, then the NNS to prevent 1 additional case of delayed language acquisition would be 6,771.
Table 5 summarizes the evidence for each of the major assumptions underlying the case for UNHS. Several gaps in information about UNHS effectiveness remain. It is clear that modern screening tests for hearing impairment can improve identification of newborns with PHL, but as many as 10 percent of newborns with normal or temporarily impaired hearing will require a second screening test. From 1 percent to 3 percent of newborns will be referred for audiological assessment; over 90 percent of those referred are false positives. The consequences of these false alarms have not been adequately evaluated, nor has the reliability of audiological and behavioral assessment—the reference standard used to make the definitive diagnosis of hearing impairment—been adequately assessed in the setting of UNHS.
A clearer picture of the consequences of delayed diagnosis in low-risk newborns would strengthen the case for universal screening. Epidemiologic studies indicate that language development is often delayed in children with congenital hearing impairment and that the diagnosis of hearing impairment is often delayed. However, no study has examined language development in infants who were diagnosed at 1 or 2 years of age and who had no other disabilities and no risk factors for hearing impairment at birth.
Because the frequency and severity of poor language outcomes in this group is uncertain, only adequately controlled trials can establish the efficacy of early intervention. Several retrospective cohort studies show that, by 2 to 4 years of age, children who have had hearing aids and other therapy in the first 6 months of life had better language skills than those who have had hearing aids and other therapy for shorter periods of time.
None of these studies compared an inception cohort of newborns offered UNHS to infants managed by usual care (including selective screening). While they are better than older studies, these studies had unclear criteria for selecting subjects, making it impossible to exclude selection bias as an explanation for the findings. The hypothesis that early intervention is a predictor of language acquisition is plausible, but the studies do not establish that screening low-risk newborns is the important factor.
As use of UNHS rapidly expands, it is important to conduct longitudinal studies of UNHS to address these gaps in its scientific basis. Further randomized trials of UNHS seem unlikely to be conducted in the United States. Although it would be possible to compare States with and without UNHS, such studies would be prone to uncontrollable confounding due to differences among States. However, better evidence about the effectiveness of UNHS is needed and could be obtained via time-series, population-based studies that begin with inception cohorts and carefully report outcomes in all possible patients, as well as rates of loss to followup. Speech, language, and scholastic achievement of deaf and hard-of-hearing children should be followed over time. States that have UNHS should conduct such population-based studies to evaluate whether the long-term language outcomes of deaf children improve as the age of identification decreases.