Research paper

238

Undetected congenital permanent hearing loss impairs development in infants by diminishing their ability to acquire language and cognitive skills at an appropriate age.1 Current esti- mates are that 1 to 6 per 1000 infants are born with significant permanent hearing loss.1,2-4 Failure to intervene

at an early age may increase the risk of lifelong learning difficulties, as well as result in significant costs in interven- tion, therapy, special education, med- ical expenses, and lost productivity.5

In 1999, the American Academy of Pediatrics Task Force on Newborn and Infant Hearing endorsed the im-

From the Department of Pediatrics, Women and Infants Hospital, Brown University, Providence, Rhode Island; In- novative Health Solutions Corporation, Brookline, Massachusetts; University of Colorado Health Sciences Center, Denver; Riley Children’s Hospital, Indiana University, Indianapolis; University of New Mexico Health Sciences Center, Albuquerque; and St Elizabeth’s Hospital, Boston, Massachusetts. Supported by Natus Medical Inc.

Submitted for publication Oct 13, 2000; revision received Mar 1, 2001; accepted Mar 16, 2001.

Reprint requests: Betty R. Vohr, MD, Women and Infants Hospital, 101 Dudley St, Provi- dence, RI 02905-2499.

Copyright © 2001 by Mosby, Inc.

0022-3476/2001/$35.00 + 0 9/21/115971

doi:10.1067/mpd.2001.115971

Objective: To investigate the costs and referral rates of 3 universal new- born hearing screening programs: transient evoked otoacoustic emissions (TEOAE), automated auditory brainstem response (AABR), and a combi- nation, two-step protocol in which TEOAE and AABR are used.

Study design: Clinical outcomes (referral rates) from 12,081 newborns at 5 sites were obtained by retrospective analysis. Prospective activity-based cost- ing techniques (n = 1056) in conjunction with cost assumptions were used to analyze the costs based on an assumed annual birth rate of 1500 births.

Results: Referral rates differed significantly among the 3 screening proto- cols (AABR, 3.21%; two-step, 4.67%; TEOAE, 6.49%; P < .01), with AABR achieving the best referral rate at discharge. Although AABR had the lowest referral rate at discharge and the highest pre-discharge costs, the total pre- and post-discharge costs per infant screened (AABR, $32.81; two-step, $33.05; TEOAE, $28.69) and costs per identified child (AABR, $16,405; two-step, $16,527; TEOAE, $14,347) were similar among programs.

Conclusion: Although AABR incurs higher costs during pre-discharge screening, it has lower referral rates than either the TEOAE or two-step pro- gram. As a result, the total costs of newborn hearing screening and diagnosis are similar among the 3 methods studied. (J Pediatr 2001;139:238-44)

plementation of universal newborn hearing screening programs to provide early assessment and intervention.1

Thirty-two states have passed or en- acted legislation recommending or mandating the implementation of uni- versal newborn hearing screening,6

and assessment protocols for screening and diagnosis have been published.7

Successful implementation of uni- versal newborn hearing screening is dependent on the implementation and coordination of numerous program- matic factors. Protocols are necessary to provide efficient, reliable, and valid methods for the evaluation of infants in a cost-effective manner. Three com- monly used protocols are available for universal newborn hearing screening programs: (1) transient evoked oto- acoustic emissions (TEOAE), (2) au- tomated auditory brainstem response (AABR), and (3) a “two-step” pro- gram in which infants are first screened with TEOAE, followed by an AABR screen for those infants who do not pass the TEOAE screen.

The costs of infant hearing programs have been explored in the past, yet many of the published cost-analysis and cost-effectiveness articles refer- ence technologies no longer widely used in universal newborn hearing screening programs.8-10 Although there are several studies that evaluate the cost of TEOAE, AABR, and the two-step program,2,11,12 there is no study that compares all 3 testing pro-

Comparison of costs and referral rates of 3 universal newborn hearing screening protocols Betty R. Vohr, MD, William Oh, MD, Edward J. Stewart, MBA, Judith D. Bentkover, PhD, Sandra Gabbard, PhD, James Lemons, MD, Lu-Ann Papile, MD, and Ronald Pye, MD

AABR Automated auditory brainstem response

TEOAE Transient evoked otoacoustic emissions

239

cedures. Furthermore, limitations of previous reports on the costs of TEOAE and AABR programs do not include complete operating costs8-10

and the total cost per identified child.13

Previous studies have documented the costs of administering individual infant hearing screening programs7,10,11,13-17

or compared the costs of screening with other, standardized tests given to newborns.13-15 The objective of this study is to compare the screening costs of 3 different protocols used in univer- sal newborn hearing screening.

METHODS

Five hospitals participated in the study. Retrospective data including type of hearing screen personnel (dedicated full-time technician, audiology student, or volunteer), hospital discharge policy (24-hour vs 48-hour discharge), and re- ferral rate (failures and incomplete screens) at the time of discharge were collected for 12,081 infants screened at 5 sites. A minimum sample size of 1500 well-baby nursery infants sequentially screened at each site was required for valid estimation of referral rates.

Resource utilization data were col- lected prospectively for at least 100 well-baby nursery infants sequentially seen at each of the 5 sites over a 3- month period (July 1, 1998, to Sep- tember 30, 1998). All sites were moni- tored and audited to ensure data collection was performed properly and consistently across sites.

Resources essential to a screening program, which are not consumed on an infant-by-infant basis, could not be collected by using prospective case re- port forms. These resources, which in- clude items such as time spent by an audiologist managing a program or training staff, were assessed by inter- viewing personnel at each site and es- timating the time. Because of differ- ences in protocols, a series of cost assumptions was established to ensure comparability in interpreting the costs

THE JOURNAL OF PEDIATRICS VOHR ET AL VOLUME 139, NUMBER 2

among the 5 sites. Table I shows a list of the economic model assumptions and relationships that were developed based on an annual rate of 1500 births. Personnel time and equipment re- quirements for the 3 protocols are shown. The model assumed that the time spent by an audiologist on pro- gram management was proportional to the number of technologies used, the number of staff, and the expertise of staff supervised. The times spent on scheduling/follow-up and administra- tive/clerical work were proportional to the referral rate at discharge and were assumed to be 1 hour per week per 3% referral rate (based on 1500 infants). The time spent on training was pro- portional to the number of technolo- gies used and to the number of screen-

ing personnel to be trained and was related to the “complexity” of the tech- nology. The model assumed that two pieces of first-step equipment and one piece of second-step equipment were required for 1500 births, which is con- sistent with the equipment used at the study sites.

In addition, the analysis assumed identical post-discharge re-screen fail rates because it was assumed that all sites used the same post-discharge protocol and had an incidence of hear- ing impairment of 2 per 1000.1 Diag- nostic fail rates were estimated ac- cordingly. It was assumed that each follow-up screen, including interpre- tation and explanation to parents, would require 1 hour and that a diag- nostic examination would require 2.5

Staff/equipment Two-step TEOAE AABR

Audiologist 6 h/wk 4 h/wk 4 h/wk Scheduling/follow-up 1.56 h/wk 2.16 h/wk 1.07 h/wk Administrative/clerical 1.56 h/wk 2.16 h/wk 1.07 h/wk Training time: audiologist 32 h/y 24 h/y 8 h/y Training time: screener 64 h/y 48 h/y 16 h/y Equipment/unit

TEOAE 2 2 2 AABR 1

Staff costs per hour are as follows: audiologist, $27.29; technician, $12.60; paid student, $12.71; and volunteer, $0.00.

Table I. Economic model assumptions per 1500 births

Study sites

A B C D E Two-step Two-step TEOAE AABR AABR

Study N 4684 1551 2777 1540 1529 Annual births 8034 2068 694 1422 1530 Length of stay (h) 48 24-48 48 48 24-48 Screener type FTT S FTT S V Referral rate at discharge (%) 3.65 7.41 6.49 1.70 3.94 Combined referral rate (%) 4.67* 6.49†‡ 3.31

A, Women and Infants Hospital; B, University of New Mexico; C, Memorial Hospital of Rhode Island; D, St Elizabeth’s Hospital; E, Boulder Children’s Hospital; FTT, full-time (dedicated) screener; S, audiology student; V, volunteer.

*P < .001 versus AABR. †P < .002 versus AABR. ‡P < .001 versus two-step.

Table II. Retrospective data set characteristics

240

hours of an audiologist’s time. Costs were assigned to the follow-up screen and diagnostic evaluations by using the 50th percentile charge and the Medicare/Medicaid Fee Schedule charge. These assumptions were based on reviews of practice and interviews with personnel at each site.

Once resource utilization had been documented, each resource was as- signed a cost value in 1998 dollars. Salaries and wages were obtained from the US Bureau of Labor Statis- tics, interviews with personnel at se- lect sites, and published infant hearing screening literature.13 Equipment and

VOHR ET AL THE JOURNAL OF PEDIATRICS AUGUST 2001

screening supplies were valued at the full list price obtained from the manu- facturers. For example, the TEOAE ILO88 was $8500, and the AABR Algo 2e was $17,500. As appropriate, necessary cost estimates were adjust- ed to 1998 dollars by using the Med- ical Services Consumer Price Index. Costs of post-discharge procedures were as follows: TEOAE screen, $50.00; AABR screen at 70 dB/35 dB, $71.06; diagnostic TEOAE, $114.00; diagnostic AABR including tone pips and bone conduction, $145.00; tympa- nometry at 60 MHz, $32.00; and visu- al threshold audiometry, $34.50.

Standard pass/fail protocols were used at all sites. For TEOAE screen- ing, ILO88 machines (Otodynamics, Ltd, Hatfield, England) were used. All AABR screening was performed on ei- ther an Algo 2 or 2c machine (Natus Medical, Inc, San Carlos, Calif). If an infant did not pass the initial TEOAE, AABR, or two-step screen, the screen was repeated.

Outpatient screening and diagnostic procedure assumptions were based on a two-step follow-up protocol pub- lished by Gabbard et al.6 The post-dis- charge protocol called for infants to return for a TEOAE and AABR (35 dB/70 dB) screen. Infants who did not pass the follow-up screen were re- ferred for a diagnostic evaluation con- sisting of diagnostic TEOAE and AABR, tympanometry, and visual au- diometry or speech threshold audiom- etry procedures.

Statistical Analyses Clinical and economic data from the

5 sites were combined into 3 screening models: two-step (TEOAE followed by AABR), TEOAE alone, and AABR alone. Referral rates at dis- charge among programs were com- pared by using the z-statistic for com- parison of proportions. A one-tailed test with an α value of .01 was used to determine whether one referral rate was significantly lower than another. Program costs through discharge and total costs were divided by the sample size of 1500 infants to give an average expected pre-discharge or total cost per infant.

As recommended by current guide- lines for conducting economic evalua- tions of health technologies, sensitivity analyses were conducted to test the ro- bustness of our results.18 Parameters that can vary from program to pro- gram (eg, screening personnel) and parameters that were based on as- sumptions (eg, percent follow-up and diagnostic evaluation costs) were ana- lyzed by using both univariate and multivariate analyses.

Figure. Screen process outcomes for the two-step protocol,TEOAE protocol, and AABR protocol are shown.

241

RESULTS

The number of well-baby nursery in- fants sequentially tested ranged from 1529 to 4684, and the number of annual deliveries at sites ranged from 694 to 8034 (Table II). Two of the 5 sites had a 24- to 48-hour discharge policy, and 3 had a 48-hour discharge policy. Screen- ers included volunteers, students, and full-time technicians. Referral rates ranged from 1.70% for AABR per- formed by student screeners at sites with a 48-hour discharge policy to 7.41% for the two-step protocol per- formed by student screeners at a site with a 24- to 48-hour discharge policy. The AABR referral rate was lower than the TEOAE (P < .002) and two-step (P < .001) rates, and the two-step rate was lower than the TEOAE rate (P < .001).

The Figure shows the pass, fail, and referral rates for each method based on the retrospective data collected for the 3 protocols and expressed per 1500 births. At one two-step screening site, screeners bypassed the TEOAE for 9.9% of all infants and went direct- ly to the AABR.

There were 1056 infants screened in the prospective part of the study: 646 with two-step (2 sites), 100 with TEOAE (1 site), and 310 with AABR (2 sites) to evaluate resource con- sumption including supply utilization, screen time, and number of screens completed. Enrollment was propor- tional to the annual birth rate. The samples represented at least 10% to 15% of the birth rate at each hospital. The first-step initial screen times were similar for the 3 methods, ranging from 9 to 11 minutes. As infants pro- gressed to re-screens or second-step screens, screen time increased, which reflects the fact that either progres- sively less-likely-to-pass infants were being re-screened or the infants were “difficult” to screen. The total mean screens per infant (initial plus any re- screening or second-step screening) were 1.15 for AABR, 1.23 for TEOAE, and 1.37 for two-step.

THE JOURNAL OF PEDIATRICS VOHR ET AL VOLUME 139, NUMBER 2

Table III incorporates the data on screens, screen times, supply consump- tion, and post-discharge costs into a comprehensive cost analysis for each screening method, based on an annual screen rate of 1500 well babies. First- step screen costs (labor and screening supplies only) ranged from $4.28 for two-step to $12.84 for AABR per in- fant. In the two-step and TEOAE pro- tocols, approximately 20% of infants

screened required a re-screen com- pared with 11.5% of infants screened with AABR. First-step re-screen costs were lower than initial screens for all 3 programs because these infants often have only one ear re-screened. Total costs of all in-hospital pre-discharge screens ranged from $24,668 for TEOAE to $35,257 for two-step to $38,137 for AABR. This corresponds to per-infant costs of $16.45 for

Pre-discharge costs Two-step TEOAE AABR

Screening costs, direct No. of infants 1500 1500 1500 First-step screen cost $4.28 $4.71 $12.84 Infants requiring re-screen [No. (%)] 305 (20.3) 285 (19.0) 173 (11.5) First-step re-screen cost $3.94 $3.99 $9.91 No. of first-step re-screens required 1 1.21 1.32 Total first-step screen costs $7,623 $8,786 $21,517 First-step screen costs per infant $5.08 $5.86 $14.34 Second-step screen cost $13.73 N/A N/A Infants with second-step screen [No. (%)] 386 (25.7) N/A N/A Infants with second-step re-screen [No. (%)] 71 (18.3) N/A N/A Second-step re-screen cost $5.37 N/A N/A No. of second-step re-screens required 1.13 N/A N/A Total second-step screen costs $5,720 N/A N/A Second-step screen costs per infant $14.84 N/A N/A Total screening costs $13,343 $8,786 $21,517 Total screening costs per infant $8.90 $5.86 $14.34

Screening costs, indirect Personnel $13,356 $11,331 $8,270 Equipment $8,456 $4,506 $8,081 Supplies $102 $45 $270 Total overhead costs $21,914 $15,882 $16,620

Total pre-discharge screening cost $35,257 $24,668 $38,137 Total pre-discharge costs/infant $23.50 $16.45 $25.42

Referral rate (% fails) 4.67 6.49 3.21 Follow-up No. requiring re-screen 70 97 48

Cost of re-screen $148.29 $148.29 $148.29 Total follow-up screen costs $10,388 $14,436 $7,140 Fail re-screen [No. (%)] 10 (14.3) 10 (10.3) 10 (20.8)

Diagnostic costs $393.73 $393.73 $393.73 Total diagnostic costs $3,937 $3,937 $3,937 No. with confirmed hearing loss 3 (30%) 3 (30%) 3 (30%)

Total post-discharge costs $14,325 $18,373 $11,077 Total pre- and post-discharge costs $49,582 $43,041 $49,215 Cost per birth $33.05 $28.69 $32.81 Cost per identified child $16,528 $14,347 $16,405

Table III. Cost analysis results based on 1500 annual births

242

TEOAE, $23.50 for two-step, and $25.42 for AABR. However, the method with the highest cost (AABR) had the lowest referral rate (P < .01), and the method with the lowest cost (TEOAE) had the highest referral rate. Follow-up re-screen costs were there- fore lowest for AABR and highest for TEOAE because follow-up costs cor- respond directly to referral rate at dis- charge. Because of similar referral rates for diagnostic testing and identical di- agnostic protocols, the diagnostic costs are identical for the 3 protocols. When all pre- and post-discharge costs are combined, the total costs are similar, yielding mean costs per birth of $33.05, $28.69, $32.81 for two-step, TEOAE, and AABR, respectively.

Personnel costs were $22,059 for the two-step, $19,745 for TEOAE, and $14,559 for AABR. Two-step screening required the most personnel, because training is conducted for two types of equipment and infants receive more screens. All programs require audiolo- gist time for managerial training and referral, but AABR does not require audiologist time for interpretation.

Sensitivity analyses to demonstrate the effects of personnel and length of stay on costs are shown in Table III. Screening personnel differ from pro- gram to program and can even vary within programs, depending on budget and schedule issues. Based on our ob- servation that referral rates decrease substantially when dedicated versus non-dedicated screeners are used and the fact that wages for these two groups differ substantially, two multi-

VOHR ET AL THE JOURNAL OF PEDIATRICS AUGUST 2001

variate sensitivity analyses were un- dertaken to assess the impact on mean total cost per birth. In scenario 1, it was assumed that only non-dedicated screeners were used for screening. In this scenario screener wages were re- duced to $0.00 per hour (ie, volun- teers), whereas referral rates for the TEOAE and two-step programs were adjusted upwards from the base case (both AABR sites in our analyses em- ployed non-dedicated volunteers, and therefore the AABR referral rate was not changed in this scenario). In sce- nario 2, it was assumed that only dedi- cated screeners were employed. In this scenario, screening wages were as- sumed to be $12.61 per hour, and re- ferral rates for the two-step and AABR programs were adjusted down from the base case (the TEOAE site in our analysis actually employed dedicated screeners, and therefore the TEOAE referral rate was not changed in this scenario). As shown in Table IV, the rank order of the programs, in terms of mean total cost per birth, changed with the scenarios, indicating that total costs are highly susceptible to changes in personnel and referral rates. AABR had the most stable mean total cost per birth in the 3 scenarios, whereas both the two-step and TEOAE had wider ranges. The AABR referral rate was the lowest in all cases. Additional analyses on program-specific factors (eg, inclusion of infant transport times) and cost assumptions (eg, varying costs of follow-up and diagnostic ex- aminations) did not yield significant variations in total cost.

DISCUSSION

The American Academy of Pedi- atrics Task Force on Newborn and In- fant Hearing revised its recommenda- tions for newborn hearing screening in 1999.1 The guidelines include the fol- lowing concepts: (1) the methodology should identify all infants with signifi- cant bilateral hearing loss, and (2) the referral rate for formal audiologic test- ing after screening should not exceed 4%. As a result, many hospitals throughout the United States are starting or expanding existing hearing screen programs. With current efforts at cost containment in health care, evaluation of costs is an important component of program development.

The screening methods had signifi- cantly different referral rates at dis- charge. AABR had a referral rate at discharge of 3.21%, compared with 4.67% and 6.49%, respectively, for two-step and TEOAE. These find- ings are consistent with recent re- ports.1,2,4,6,11,14,16 The two AABR sites and one two-step site achieved the American Academy of Pediatrics recommendation of a referral rate of <4% at discharge. Lower referral rates result in decreased post-dis- charge follow-up screening, adminis- trative, and scheduling costs, as well as lower parental anxiety associated with unnecessary referral. These data indicate that continued efforts must be made to lower the referral rates at hospital discharge.

Because hospital discharge time and screening personnel differed among

Referral rate at discharge Mean total cost per birth

Two-step TEOAE AABR Two-step TEOAE AABR

Base case 4.75% 6.49% 3.21% $33.05 $28.69 $32.81 Scenario 1* 7.41% 13.71% 3.21% $37.28 $40.52 $29.48 Scenario 2† 3.65% 6.49% 1.57% $31.08 $28.69 $29.64

*Performed by non-dedicated screeners. †Performed by dedicated screeners.

Table IV. Sensitivity analyses results

243

sites, we investigated the effects of these factors on referral rates and costs. Three hospitals had 48-hour stays for newborns, and two sites had 24- to 48-hour stays. Although the age for the initial screen was not affected by these policies, we did identify an ef- fect of the policy on the referral rate. The effect is attributed to the “screen- ing window,” the time between the first screen and discharge. The screen- ing window is smaller for hospitals with 24- to 48-hour discharge policies, and infants are less likely to have a second screen. In fact, the hospitals with 24- to 48-hour discharge policies had higher referral rates. This rela- tionship was identified in both the two-step and AABR sites where refer- ral rates in the hospitals with 24- to 48-hour discharge policies (7.41% and 3.94%) were higher than in the com- parable hospitals with 48-hour dis- charge policies (3.65% and 1.70%). In fact, the scenario analyses confirm that environmental and systems fac- tors other than the screen protocols have an impact on costs.

Screening personnel at the study sites varied from dedicated technicians to students and volunteers. However, in order to remain consistent in the cost- ing of each screening method, all screening personnel were assigned the same cost, and screening timing was valued at the wage rate for paid full- time technicians in the baseline cost analyses. As a result, the AABR screen- ing method is costed conservatively in the analysis, that is, higher than the ac- tual, for comparison purposes because the two AABR sites employed either audiology students or volunteers. A sig- nificant reduction in referral rate was identified for the two-step protocol with dedicated personnel.

The characteristics of the 3 hearing screen protocols are quite similar. The duration of the screen in minutes, which included preparation, was simi- lar for all methods. The initial TEOAE screens averaged 9 and 10 minutes, whereas the initial AABR

THE JOURNAL OF PEDIATRICS VOHR ET AL VOLUME 139, NUMBER 2

screen averaged 11 minutes. In previ- ous studies shorter screen times have been reported.1,2 We suggest that this is due to variations in the definition of screen time used in this study. Screen time in this study began as soon as the infant arrived for the screen and ended when the final pass/fail decision was made. The technology used for screening actually has a smaller effect on the overall screen time than previ- ously reported, and the set-up time and baby state (fussy/moving) deter- mine how long the screen will take.

It is important to note that in evalu- ating follow-up screening and diag- nostic evaluation, several cost assump- tions were made. The first relates to the cost of diagnostic testing; the as- sumption was that the guidelines rec- ommended by the Infant Assessment and Amplification Task Force in Col- orado7 were followed at every hospital in the study and therefore the costs were the same. However, diagnostic protocols differ substantially across the country. In addition, for consisten- cy, the return rates for diagnostic eval- uations were assumed to be 100%. Al- though the rate is not tracked by all study sites, Women and Infants Hos- pital post-discharge follow-up rates between 1993 and 1996 ranged from 74% to 88%.19

Several additional factors should be considered when the results of this study are evaluated. First, the study was conducted at sites with estab- lished, ongoing screening programs, although the data at site C included the first year of screening. The equipment and overhead costs were annualized (amortizing over 5 years) and there- fore do not represent actual startup cash requirements. Second, the results are modeled for a typical program of approximately 1500 newborns per year. It is expected that some economies of scale would exist for larg- er programs because some overhead and equipment costs can be spread across a larger number of infants. Fi- nally, one cost that was not included is

the cost of data management and tracking. Several software programs that are capable of tracking screening data are commercially available and must be considered when regional or state screen systems are developed.

Total pre-discharge screening costs were similar for the two-step and AABR programs, whereas total pre- discharge costs for the TEOAE pro- gram were considerably lower. These costs can be converted to per-infant costs that range from $16.45 to $25.42 and are consistent with estimates from past studies.2,12-14,17 Because re- ferral rates were highest for TEOAE, the highest post-discharge screen costs were for TEOAE, and the low- est post-discharge costs were for AABR, which had the lowest referral rate at discharge. Finally, total costs (pre- and post-discharge) per new- born and cost per child identified with hearing impairment were remarkably similar for the two-step, TEOAE, and AABR programs.

There has been concern that false- positive neonatal screens may have negative effects on parental stress, coping abilities, and the parent-child relationship.20-28 The hospital referral rates reported in this study were 6.49% for TEOAE, 4.67% for the two-step, and 3.21% for AABR. These rates must be contrasted to an antici- pated rate of identification of perma- nent hearing loss of 2 per 1000 or 0.2%. On the basis of these referral rates, TEOAE has the highest number of false-positive screens at the time of discharge. False-positive screens in the hospital may reflect a number of factors including inexperience of the screener, debris in the ear canal, a transient hearing loss, or fluid in the middle ear.

We thank the following individuals for their research assistance during this study: Patricia Moore, Kristen Letourneau, and Cheryl Mc- Dermott (Women and Infants Hospital); Margo Chiappinelli (Memorial Hospital); Beth Lannon, PhD (St Elizabeth’s Hospital); Lorraine Wells (University of New Mexico Health Sciences Center); Vickie Thomson and

244

VOHR ET AL THE JOURNAL OF PEDIATRICS AUGUST 2001

C. J. Grace (Boulder Community Hospital); and Kate Stewart and Kim Powers (Innova- tive Health Solutions).

REFERENCES 1. American Academy of Pediatrics Task

Force on Newborn and Infant Hear- ing. Newborn and infant hearing loss: detection and intervention. Pediatrics 1999;3:527-30.

2. Maxon AB, White KR, Behrens TR, Vohr BR. Referral rates and cost effi- ciency in a universal hearing screening Program using transient evoked oto- acoustic emissions. J Am Acad Audiol 1995;6:271-7.

3. White KR, Vohr BR, Maxon AB, Behrens TR, McPherson MG, Mauk GW. Screening all newborns for hear- ing loss using transient evoked oto- acoustic emissions. Int J Pediatr Otorhinolaryngol 1994;29:203-17.

4. Huynh MT, Pollack RA, Cunningham RAJ. Universal newborn hearing screening: feasibility in a community hospital. J Fam Pract 1996;42:487-90.

5. Downs MP. The case for detection and intervention at birth. Semin Hear- ing 1994;15:76-83.

6. Gabbard SA, Thomson V, Brown AS. Considerations for universal newborn hearing screening, audiologic assess- ment & intervention. Audiology Today 1998:8-10.

7. Turner RG. Factors that determine the cost and performance of early identification protocols. J Am Acad Audiol 1992;3:233-41.

8. Markowitz RK. Cost-effectiveness comparisons of hearing screening in the neonatal intensive care unit. Semin Hearing 1990;11:161-6.

9. Prager DA, Stone DA, Rose DN.

Hearing loss screening in the neonatal intensive care unit: auditory brain stem response versus crib-o-gram; a cost-effectiveness analysis. Ear Hear 1987;8:213-6.

10. Weirather YP, Korth N, White KR, Downs D, Woods-Kershner N. Cost analysis of TEOAE-based universal newborn hearing screening. J Com- mun Disord 1997;30:477-93.

11. Mason JA, Herrmann KR. Universal infant hearing screening by automated auditory brainstem response measure- ment. Pediatrics 1998;101:221-8.

12. Hermann BS, Thornton AR, Joseph JM. Automated infant hearing screening using the ABR: develop- ment and validation. Am J Audiol 1995;4:6-14.

13. Johnson JL, Mauk GW, Takekawa KM, Simon PR, Sia CCJ, Blackwell PM. Implementing a statewide system of services for infants and toddlers with hearing disabilities. Semin Hear- ing 1993;14:105-18.

14. Mehl AL, Thomson V. Newborn hear- ing screening: the great omission. Pe- diatrics 1998;101:E4.

15. White KR, Behrens TR, Strickland B. Practicality, validity and cost-efficien- cy of universal newborn hearing screening using transient evoked oto- acoustic emissions. J Child Commun Disorders 1995;17: 9-14

16. White KR, Mauk GW, Brandt Culpepper N, Weirather Y. Newborn hearing screening in the United States: is it becoming the standard of care? In: Spivak L, editor. Universal newborn hearing screening. New York: Thieme; 1997. p. 225-255.

17. White KR, Maxon AB. Universal screening for infant hearing impair- ment: simple, beneficial, and presently

justified. Int J Pediatr Otorhinolaryn- gol 1995;32:201-11.

18. Siegel JE, Torrance GW, Russell LB, et al. Guidelines for Pharmacoeco- nomic Studies: Recommendations from the Panel on Cost Effectiveness in Health and Medicine. Pharma- coEconomics 1997;11:159-68.

19. Vohr B, Carty, LM, Moore PE, Le- tourneau, K. The Rhode Island Hear- ing Assessment Program: experience with statewide hearing screening (1993-1996). J Pediatr 1998;133:353-7.

20. Bess FH, Paradise JL. Universal screening for infants hearing impair- ment: not simple, not risk-free, not necessarily beneficial, and not present- ly justified. Pediatrics 1994;93:330-4.

21. Sorenson JR, Levy HL, Mangione TW, Sepe SJ. Parental response to re- peat testing of infants with “false-posi- tive” results in a newborn hearing screen. Pediatrics 1984;73:183-7.

22. Fyro K. Neonatal screening: life-stress scores in families given a false-positive result. Acta Paediatr Scand 1988;77: 232-8.

23. Fyro K, Bodegard G. Difficulties in psychological adjustment to a new neonatal screening programme. Acta Paediatr Scand 1988;77:226-31.

24. Mischler EH, Wilfond BS, Fost N, Laxova A, Reiser C, Sauer CM, et al. Cystic fibrosis newborn screening: im- pact on reproductive behavior and im- plications for genetic counseling. Pe- diatrics 1998;102:44-52.

25. Marteau TM, van Duijn M, Ellis I. Effects of genetic screening on per- ceptions of health: a pilot study. J Med Genet 1992;29:24-6.

26. Stewart-Brown S, Farmer A. Screen- ing could seriously damage your health. BMJ 1997;314:533-4.