Early Hearing and Communication Development

Chapter VII: Outcomes

Universal Newborn Hearing Screening: The Evidence

Author: Dr. Andrée Durieux-Smith

Permanent childhood hearing impairment (PCHI) has been associated with delays in speech, language development and learning. It has been reported that deaf students graduate from high school with language and academic levels corresponding to those of fourth grade students with normal hearing.^1,2 One of the factors which has been linked with the delays in language and academic development has been the age at diagnosis. Universal newborn hearing screening (UNHS) is seen as a strategy that leads to the early identification of children with PCHI and to the provision of early hearing and communication development (EHCD) programs. EHCD programs are seen as ways to reduce the gap in language and academic skills between hearing and deaf students. The determination of the evidence in support of UNHS programs has, therefore, focused predominantly on two main aspects: (1) Do UNHS programs lead to a greater number of children with PCHI being identified and treated earlier? (2) Does EHCD improve the development of language and communication?

The U.S. Preventive Services Task Force (USPSTF)³ represents one of several initiatives which review evidence to ensure that the development of clinical practice guidelines is based on scientific evidence rather than on expert opinion. Systematic searches of multiple bibliographic research databases help to identify relevant literature in an unbiased and thorough manner. Quality criteria developed by methodologists are used to guide judgments of the strengths and weaknesses of individual studies. Two independent members of the topic team usually review abstracts of all articles. Once a decision has been taken to include an article, information is abstracted on patient population, study design, interventions (where applicable), quality indicators and findings.

The strongest empirical support comes from experimental designs involving randomized controlled trials with large numbers of subjects. This is viewed as the only design that permits clear linkages between the intervention and associated outcomes. The second type of evidence is usually obtained from quasi-experimental designs, often seen in cohort studies. These are prospective studies in which a large group of individuals with a common characteristic are followed over time and a particular outcome investigated. Control groups are usually involved allowing for intergroup comparisons. These types of studies are seen as providing less compelling evidence than randomized controlled studies. A third type of study usually involves the use of non-experimental designs that can often be retrospective in nature. These types of studies are often seen as lacking experimental controls and are often criticized. Once a review of the literature has been carried out, the supporting evidence is reviewed and rated and statements and recommendations are issued.

The USPSTF limits the areas to be reviewed to those conditions that cause a large burden of suffering to society and which can potentially be alleviated by some form of prevention. Good or fair quality evidence for an entire preventive service must include studies of sufficient design and quality to provide linkages that connect the preventive service with the health outcome. Newborn hearing screening is seen as a preventive measure that should be associated with enhanced speech and language development in children with a PCHI. The topic of UNHS has been reviewed by the USPSTF, through the Oregon Health Sciences University evidence-based practice centre and the results of this review appear in an article published by Thompson and colleagues in 2001,⁴ which is summarized in this report.

Universal Newborn Hearing Screening and Early Identification of Hearing Impairment

The ages of diagnosis of children identified in the absence of UNHS and in the presence of UNHS have been addressed in Chapter III in this document. Some of the articles presented in that chapter will be reviewed briefly in this section. The Wessex Universal Neonatal Hearing Screening Trial⁵ reported an increase in the number of cases with a significant hearing impairment who were identified and treated early. Seventy-one more babies with a moderate to severe PCHI per target population of 100,000 were referred before 6 months of age during periods with neonatal screening than during periods without. In addition, UNHS led to an increase in confirmation and management of hearing impairment by 10 months of age. Fifty-seven percent of children with moderate or severe hearing impairment were diagnosed with UNHS in comparison with 14% without UNHS. In the best quality U.S. study,⁶ in the presence of UNHS, the age of diagnosis for mild to moderate and severe hearing impairment was approximately 6 months and below. Both the Wessex trial⁵ and the Dalzell et al. study⁶ have been rated as providing good evidence by the Oregon Health Sciences University evidence-based practice centre. The studies were rated as being well designed - the U.S. study being a cohort study and the U.K. study being a controlled non-randomized trial. Based on the review published by this centre, the USPSTF has stated that there is good evidence that newborn hearing screening leads to earlier identification and treatment of infants with hearing impairments. The next question which needs to be answered is whether early identification and treatment resulting from UNHS improve language and communication development.

Universal Newborn Hearing Screening and Language and Communication Development

Thompson et al.⁴ reviewed the articles which have investigated speech and language development in children with a PCHI identified through UNHS. Their literature search indicates that at the moment there are no prospective, controlled studies that have directly examined whether newborn hearing screening and early intervention give rise to improved speech, language and educational development. Eight recent cohort studies from three intervention programs are summarized in the Thompson et al. paper.⁴ The studies reported used standardized receptive and expressive tests to evaluate the speech and language skills of pre-schoolers.

All of the studies reported statistically significant associations between age of diagnosis and language development at ages 2 to 5 years. Six of the eight studies reported results on children in the Colorado Home Intervention Program. One of these studies⁷ compared the language performance of hearing impaired children born in hospitals with UNHS programs to that of children born in hospitals without. The results showed that the mean scores for expressive, receptive and total language were within normal ranges for the screened group and significantly higher than for the unscreened group. The evidence provided by this study was rated as poor because the authors used a convenience sample, the assessment of outcome was unblinded and the exclusion criteria not specified. In another study by the same group,⁸ which was also rated as poor, children identified before 6 months of age were seen to have language scores at or near their cognitive test scores, whereas children identified after 6 months performed on a significantly lower level than their cognitive test scores. This study was criticized based on the statistical method used in the data analysis, on the fact that no information on dropout rates was provided and that the assessments were not masked.

The articles reviewed by Thompson and colleagues⁴ were all seen as having several limitations. The subject populations were comprised mostly of convenience samples. The inclusion criteria were unclear and the assessments were not blinded. In addition, none of the studies provided information on attrition and follow-up rates. The USPSTF rated the quality of the evidence linking early intervention with language outcomes as inconclusive and the quality of the evidence as fair to poor.

The USPSTF concluded that there is a need for population-based studies that begin with inception cohorts which carefully report outcome on all possible subjects, as well as rates of follow-up and attrition. There is a need for prospective, longitudinal studies which report on the speech, language and education development of children with PCHI over time.

Since EHCD programs are relatively new, there is a need for additional well-controlled research to determine the efficacy of UNHS. Such research is complex and difficult. Many factors have the potential to impact on the communication development of children with a PCHI, such as parental involvement,⁹ degree of hearing impairment, additional handicapping conditions and quality of pediatric care. Many variables can contribute to developmental outcome and not all of them can be included in a research design. Nevertheless, new EHCD initiatives are providing opportunities to prospectively follow children with PCHI who have been identified early, taking into account, as much as possible, the many variables which can have an impact on outcome.

Conclusions

• Some studies have concluded that early identification and strong family involvement improve the development of speech and language in infants and young children with hearing impairment.
• There is a need for more research in this area, in particular whether universal newborn hearing screening (UNHS) programs and early hearing and communication development (EHCD) lead to improved speech, language and education development.

Key References

1. Holt JA. Stanford Achievement Test. 8th ed.: reading comprehension subgroup results. Am Ann Deaf. 1993;138:172-5.
2. Allen TE. Patterns of academic achievement among hearing impaired students: 1974 & 1983. In: Schildroth A, Karchmer AM, editors. Deaf Children in America. Boston, MA: College Hill Press; 1986. p. 161-206.
3. U.S. Preventive Services Task Force. Screening for hearing impairment. In: U.S. Preventive Services Task Force Guide to Clinical Preventive Services. 2nd ed. Baltimore, MD: Williams & Wilkins; 1996. p. 393-405.
4. Thompson D, McPhillips HM, Davis RL, et al. Universal Newborn Hearing Screening: Summary of Evidence. JAMA. 2001;286(16):2000-10.
5. Wessex Universal Neonatal Hearing Screening Trial Group. Controlled trial of universal neonatal screening for early identification of permanent childhood hearing impairment. Lancet. 1998;352:1957-64.
6. Dalzell L, Orlando M, MacDonald M, et al. The New York State universal newborn hearing screening demonstration project: ages of hearing loss identification, hearing aid fitting and enrollment in early intervention. Ear Hearing. 2000;21:118-30.
7. Yoshinaga-Itano C, Coulter D, Thomson V. The Colorado Newborn Hearing Screening Project: effects on speech and language development for children with hearing loss. J Perinatol. 2000;20(Suppl 8):S132-7.
8. Yoshinaga-Itano C, Sedey AL, Coulter DK, Mehl AL. Language of early and later identified children with hearing loss. Pediatrics. 1998;102:1161-71.
9. Moeller MP. Early intervention and language development in children who are deaf and hard of hearing. Pediatrics. 2000;106:E43.

Previous Page | Table of Contents | Next Page