|
Selecting Screening Equipment Reference Table
|
| Issue |
Automated ABR |
DPOAE |
TEOAE |
| 1. Cost of Equipment
Cost of equipment ranges from about $4,000 to $25,000 per unit. Because prices change frequently, specific figures are not listed here. Hand-held units tend to be the most inexpensive, while units which include both AABR and OAE technology tend to be the most expensive.
|
(See list of
screening equipment) |
(See list of
screening equipment) |
(See list of
screening equipment) |
| 2. Cost of Supplies
(Included here is the cost of all necessary supplies and reoccurring expenses (e.g., calibration) for doing screening. It does not include supplies for communicating the results of screening with parents or pediatricians, printing educational materials, or other ancillary materials. Costs are estimated per baby based on reported usage by typical programs with 1000-4000 births per year.)
|
$5.00-10.00 per baby includes the costs of disposable earphones and electrodes |
$.50-1.50 per baby includes the costs of disposable tips for the probe assembly, calibration, and probe replacement |
$1.00 per baby includes the costs of disposable tips and for the probe assembly and replacing the probe assembly every 750 babies |
| 3. Initial Training of Screening Technicians
(Although it is possible to start any program by reading the literature which comes with the equipment and teaching yourself, most programs find that hands-on, competency-based training by someone who is already experienced with that particular equipment and has used it successfully is the best way to begin a program. Estimated times are based on the experience of operational programs and includes only the initial training of screening technicians. Regular supervision with additional upgrading of skills should be included in addition to this initial training.
|
2 hours |
4 hours** |
4 hours** |
| * Unless otherwise specified, most of the commentary for AABR only applies to the Algo 2 since it is the most frequently used automated ABR screener.
** Recently released hand-held units for OAE require only about one hour of training.
|
| 4. Time to do Screening per Baby
(This is often misunderstood because the term "screening time" is used by people to refer to different aspects of the screening process. As used here, it is the total amount of time devoted to screening babies and includes getting the baby ready for screening, talking to the parents if necessary, setting up the equipment, conducting the screening, recording information about the baby so results can be retrieved later, etc. "Screening time" is best computed by taking the total number of hours worked by screening technicians and dividing that time by the number of babies screened during that period. Numbers for each device are based reports of well-established programs.)
|
15-40 minutes per baby |
10-30 minutes per baby |
10-30 minutes per baby |
| 5. What is Being Measured?
(None of the devices is a direct measure of hearing. Instead, each one measures slightly different physiological mechanisms which are related to hearing. Issues related to this are discussed below.)
|
| 5a. What Degree of Hearing Loss is Likely to be Detected? |
As used in most programs, the Algo 2 uses a 35 dB nHL click and, consequently, would probably miss children with very mild sensory hearing losses (25 or 30 dB). An alternative mode for the Algo 2 measures at 40 dB and 70 dB. |
Although there is not unanimous agreement among researchers, most believe that with the proper parameters, hearing loss as low as 25 dB nHL can be detected with DPOAEs. |
There is substantial agreement that TEOAEs will be detected if hearing threshold is 25 dB nHL or better. |
| 5b. Is Frequency Specific Information Available?
(In addition to indicating whether or not a child has a hearing impairment, some people are interested in knowing at what frequencies that hearing impairment is likely to occur. Others argue that the purpose of a screening test is not to provide detailed information about the nature of the loss, but to identify those children who need further diagnostic tests, during which information about frequency and severity of hearing loss can be determined.)
|
The Algo 2 is a dedicated screening device. Screening is a selection procedure for diagnostics where hearing loss is confirmed and its characteristics defined. No frequency specific information is obtained by click evoked auditory potentials screening, but is available through completion of diagnostic ABR follow-up where it is used to make treatment decisions. |
DPOAEs have the best potential for providing frequency specific information, and some argue that DPOAEs can be used as a diagnostic tool. However, this has not been sufficiently demonstrated. There is general agreement that DPOAEs provide more information about the higher frequencies (6-10 kHz) than do TEOAEs, but most would agree that the improved information in these higher frequency areas is not very critical for newborn hearing screening. |
TEOAEs provide information about the frequencies at which emissions are detected between 1 and 5 kHz. However, the absence of an emission at a particular frequency does not always correspond to a hearing loss at that frequency. |
| 5c. What is Being Measured? |
The AABR provides information about the auditory pathway up to the brainstem (including the middle ear, the inner ear, and the VIII nerve). |
DPOAEs provide information only up to and including the cochlea. Hence, infants with central auditory processing problems would not be discovered. Although definitive prevalence data are not available, most experts agree that this represents less than 1% of all children with hearing loss, or less than 3 children per 100,000 in the general population. |
TEOAEs provide information only up to and including the cochlea. Hence, infants with central auditory processing problems would not be discovered. Although definitive prevalence data are not available, most experts agree that this represents less than 1% of all children with hearing loss, or less than 3 children per 100,000 in the general population. |
| 6. Scoring Criteria and Ease of Interpretation
(Because DPOAEs and TEOAEs produce a wave form for each infant, users must decide what constitutes a pass or a refer. Because the widespread use of these techniques is fairly recent, there is not universal agreement on what criteria should be used. In practice, however, this lack of agreement affects a very small number of infants, since in most cases emissions are clearly present or clearly absent, and it's only the relatively small number of infants around the cut point where disagreement occurs.
|
The Algo 2 matches the ABR to a template derived from the waveforms of normally hearing neonates to 35 dB nHL click stimuli. The algorithm employs binomial sampling and a statistical test to determine that data collected sufficiently discriminates between the presence of a response + noise vs. pure noise at > 99% level of confidence. There is no operator interpretation needed. Studies which have compared the results of the Algo 2 with expert scoring of conventional ABR have found agreement ranging from 83% to 98%. |
DPOAEs are the most recent of the techniques, and, not surprisingly, there is a lot of disagreement about what constitutes a pass or a refer. Most people have tended to use fairly conservative pass criteria until more data are available. The numerical criteria are easy to interpret, and most programs use technicians to make this determination in a few seconds per baby. |
Although they have been used extensively since the early 1990's, there are still many different pass/refer criteria being used in TEOAE-based newborn hearing screening programs. The most frequently used criteria recommended by NCHAM is a very conservative criteria. Using this numerical criteria, interpretation is straight forward and is done in most programs by technicians in a few seconds per baby. |
| 7. Flexibility of Administration |
Because it was intended to be a completely automated system, the Algo 2 is designed to have very little flexibility. It is possible to screen at either 35 dB or at 40 dB and 70 dB, and it's possible to screen both ears simultaneously or each ear separately. |
There is lots of flexibility in how the test is administered. Unfortunately, there is not unanimity about what parameters are best for screening (e.g., the different primaries to be used for f1 and f2, the intensity of the stimulus, or how many data points per octave are required for an adequate test). |
Although there is a great deal of flexibility with regard to collecting TEOAE information, parameters used in screening programs are usually those recommended by NCHAM (e.g., QuickScreen, low frequency filter, peak stimulus between 78 and 83 dB SPL). |
| 8. Flexibility of Use |
The Algo 2 is a dedicated screener designed for use only with infants. Consequently, it can only be used for screening newborns. |
In addition to being used for infant screening, DPOAE equipment is used with children and adults for monitoring the effects of surgery and drug administration and various diagnostic applications |
In addition to being used for infant screening, TEOAE equipment is used to screen hearing with children and adults for monitoring the effects of drug administration and various diagnostic applications |
| 9. Referral Rates
(Screening is designed to identify a small group of at-risk infants who will require further diagnostic testing. As in all screening programs, it is expected that some children who have normal hearing will be referred for further diagnostic testing, but the lower this number is, the better.
|
Reported referral rates at the time the infant leaves the hospital for programs using the Algo 2 equipment range from 1% to 10%, with an average of about 4%. |
Reported referral rates at the time the infant leaves the hospital for DPOAE programs range from 4% to 15%, with an average of about 8%. Since most DPOAE programs do a two-stage screening process where those who do not pass before discharge from the hospital are rescreened before referring them for diagnostic testing, the percentage referred for diagnostic testing is about 1%. |
Reported referral rates for infants at the time they are discharged from the hospital range from 3% to 12%, with an average of about 7%. Since most TEOAE programs are a two-stage screening program, with infants who are referred at the time of discharge from the hospital being screened a second time before being referred for diagnostic assessment, the percentage of infants referred for diagnostic assessment ranges from 1/2% to 1%. |
| 10. Screening in Noisy Situations
(Noise which interferes with screening can come from the external environment or from the baby. Because newborn nurseries can be quite noisy, many people have questions about the effects of noise on newborn hearing screening procedures [this is especially true for intensive care nurseries.])
|
The Algo 2 manual recommends choosing a baby in a favorable state "sleeping, having been fed recently" for most efficient screening. An artifact reject system automatically interrupts data collection when ambient noise > 50 dB SPL at 2000 Hz and automatically resumes when conditions meet criteria again. Thus, the Algo 2 screens in noisy settings, but noise may slow data collection. |
The key to screening in noisy situations is achieving good probe fit. Not all DP equipment provides feedback regarding adequacy of probe fit. Most DP units have artifact reject systems which exclude noisy data from averaging. Thus, the equipment can be used in noisy settings, but data collection is slower. Because DPOAEs measure one frequency at a time, they are more susceptible than TEOAEs to a response at that frequency being obscured by noise. Babies do not need to be asleep, but a noisy baby will slow data collection substantially. |
The key to screening in noisy situations is achieving good probe fit. The ILO88 provides excellent real-time information to monitor probe fit and has an artifact reject system which excludes noisy data from averaging. Thus, the equipment can be used in noisy settings, but data collection is slower. Babies do not need to be asleep, but a noisy baby will slow data collection substantially. |
| 11. How Many Children with Hearing Loss will Pass the Screen?
(These children are often referred to as false negatives and reported as a measure of the test's sensitivity. It's important to minimize the number of infants in this category. While no screening test is perfect, ideally, as few children as possible should be in this group. This does not refer to children who have late onset losses, but instead is only concerned with those children who have impaired hearing at the time of the test and still pass the screen.
|
Infants with very mild losses (25 to 30 dB will likely pass the screening, as will infants with high frequency losses, reverse slope losses, or precipitous losses. |
Children with neural or central auditory pathology or children having reverse slope losses may pass. |
Children with neural or central auditory pathology or children having reverse slope losses may pass. |
| 12. Cost Per Infant Screened
(Although there have been numerous reports in the literature and anecdotal reports about the cost per baby screened in newborn hearing screening programs, most of these analyses are based on gross estimates of time devoted to different tasks or have been incomplete [e.g., have ignored fringe benefit costs for personnel, indirect costs, supervisory costs, or costs associated with supplies and equipment]. How the program is organized can also have a big impact on the cost per baby. Because of such factors, people trying to interpret reported costs should be very cautious and remember that cost per baby is primarily a function of how long it takes to do the tasks, coupled with the hourly rate of people doing the work and the cost of supplies, equipment, and facilities.
|
Reported costs range from $15 to $75 per baby. |
Costs per baby are not available for DPOAE programs, but they should be similar to those reported for TEOAE. |
Reported costs range from $8 to $30 per baby. |