Objective. Universal Newborn Hearing Screening began in 2 Honolulu hospitals in 1992, and by 1999, all 14 civilian birthing facilities in Hawaii were providing screening. Examination of 1998 Hawaii data indicated that approximately 13% of infants who did not pass initial hearing screening in the hospital did not return for the indicated follow-up. The purpose of this study was to determine the epidemiologic profile of infants who were born in 1999 and did not return for follow-up.
Methods. A population-based, cohort study of the hearing screening completion rates among the 13 civilian birthing facilities in Hawaii that provided data to the Department of Health was conducted. Analysis included a bivariate analysis of the demographic characteristics of infants who completed the screening/follow-up process compared with those who did not and logistic regression modeling to ascertain the demographic profile of infants at high risk for being lost to follow-up.
Results. Of 12 456 infants, hearing screening data could be linked to the birth certificate file, and a final disposition regarding completion of the screening/follow-up process was determined for 10 328 (83%). Less than 2% (n = 176) of the linked infants failed to complete the screening/follow-up procedures. Low birth weight and white infants and infants born to women who had not completed high school were approximately twice as likely not to complete the screening as were their normal birth weight or nonwhite counterparts.
Conclusions. Failure to complete the hearing screening follow-up may be related to cultural differences that have been previously reported in other maternal and child health studies of the diverse populations in Hawaii. The results of this study will allow the Hawaii Newborn Hearing Screening Program to target its efforts and limited resources toward infants who are at higher risk of not completing the screening and who may need special attention to encourage their mothers to complete the screening process, and to move quickly with rescreening infants whose initial tests are positive so that infants are not lost to follow-up.
In 1990, Hawaii legislation mandated that the Hawaii Department of Health develop methods to establish, implement, and evaluate a statewide program for early identification of hearing loss and intervention for children younger than 3 years.1 Newborn hearing screening began voluntarily in 2 Honolulu hospitals in 1992, and by 1999, all 14 civilian birthing facilities in Hawaii were providing screening. The goal of the program is to provide universal newborn hearing screening/follow-up by 1 month of age; initiate audiological evaluation to confirm infant hearing loss including type, configuration, and degree to infants who screen positive for hearing loss by 3 months of age; and begin intervention for infants identified with a hearing loss by 6 months of age. The specific procedures for newborn hearing screening in Hawaii using otoacoustic emissions screening have previously been described.1
“Infants are generally screened within the newborn nursery, or, if the nursery’s noise level is unacceptable, in an adjacent room. Best results are obtained after the first 24 hours following the birth. If an initial response is not obtained, several efforts are made to secure a response. If a response still cannot be secured, the infant is scheduled to return for a rescreen as soon as possible. If responses cannot be obtained during the second screening, the infant is referred for a diagnostic auditory brainstem evaluation … Parents and pediatricians are informed whenever responses are not obtained from the newborn. For those newborns for whom screening cannot be completed before discharge, parents are notified that the child was not screened and parents are offered the opportunity to return for outpatient screening.”
Examination of 1998 Newborn Screening Program data indicated that approximately 13% of infants who did not pass initial hearing screening in the hospital did not return for follow-up to complete the screening process. Although the Newborn Screening Program worked extensively with the birthing hospitals in 1999 to decrease the number of infants lost to follow-up, there was still some concern about what more could be done for the infants who seemed to be “falling through the cracks.” Because early intervention has been deemed either to minimize or to prevent the negative effect of hearing loss,2,3 it is important to identify all children who might benefit from the program. Furthermore, if families who are at risk of not completing the hearing screening/follow-up process can be identified, then more intensive efforts, including health education, can be directed to them. Thus, the purpose of this study was to determine the epidemiologic profile of infants who did not complete hearing screening/follow-up in 1999.
We conducted a population-based, cohort analysis of the hearing screening/follow-up completion rates among 13 of 14 civilian birthing facilities in Hawaii. We compared the infants who were not screened or were screened but did not pass the hearing screening and did not complete the follow-up procedures with infants who did not pass the initial newborn hearing screening and completed the follow-up process. Because the necessary demographic data were not consistently reported in the tracking system, we linked the 1999 hearing screening and tracking data from the Department of Health’s HI*TRACK system to the birth certificate file.4
A bivariate analysis of the demographic characteristics of infants who completed the screening/follow-up process compared with those who were lost to follow-up was conducted using χ2 test statistics. To assess the independent effects of the sociodemographic variables on the outcome, we constructed a logistic regression model adjusting for education (<12 years, 12+ years), ethnicity of the mother (white, Asian, Filipino, Hawaiian, all others), infant birth weight (<2500 g, ≥2500 g), and location of birth (obstetric level 2 or 3 medical center on Oahu; community hospital on Oahu; community hospital on the islands of Kauai, Maui, Molokai, and Hawaii, referred to as “neighbor islands”). We estimated model coefficients by using unconditional maximum likelihood methods, and we estimated relative risks by calculating adjusted odds ratios (aORs).
Of the 17 091 births in Hawaii in 1999, 12 456 infants were born in the 13 civilian hospitals that provide hearing screening data to the Hawaii Department of Health. (Although the 1 civilian hospital not included in this study did conduct hearing screening in 1999, the results were not transmitted to the Hawaii Department of Health.) Hearing screening tracking data could be linked to the birth certificate file for 10 328 (83%) of these infants. The age and marital status of the mother, and gender and screening/follow-up rates of the infants whose birth certificates could not be linked to HI*TRACK files were similar to those of infants whose files could be linked. However, some differences were noted between the infants with linked files and those with unlinked files when birth weight, hospital of delivery, maternal education, and maternal race were examined (Table 1).
This study included 1013 infants who missed the hearing screening in the hospital, had an inconclusive initial result, or were referred for follow-up. All of the potential outcomes of the initial screening are provided in Fig 1. Sociodemographic characteristics of the mothers and infants are presented in Table 2. No significant differences in age or marital status of mother or gender of infant were noted between the infants who failed to complete the hearing screening/follow-up process and those who completed it. However, infants born weighing <2500 g were less likely to complete the hearing screening than were infants weighing 2500 g or more (P < .05). Infants who were born in community hospitals throughout the state were less likely to complete the screening than were infants who were born in the medical centers on the island of Oahu designated as obstetric level 2 or 3 (P < .05), although these latter hospitals delivered the majority of the births in the study and in the state. Significantly lower completion rates were also noted for lower educational levels as well as for white mothers.
We entered maternal education, infant birth weight, maternal ethnicity (white versus other), and hospital of delivery in the logistic regression models (Table 3). Women who had not completed high school were less likely to have their infant complete the hearing screening/follow-up than were more educated women (aOR: 1.8; 95% confidence interval [CI]: 1.2–2.8). Low birth weight infants were twice as likely not to complete the screening/follow-up compared with their normal birth weight counterparts (aOR: 2.0; 95% CI: 1.2–3.4). White infants were also more likely not to complete the screening/follow-up compared with nonwhite infants (aOR: 1.8; 95% CI: 1.2–2.7). After adjustment, infants who were born in rural, community hospitals were not different with respect to completing the screening/follow-up process when compared with infants who were born in major medical centers on Oahu.
The review of newborn hearing screening surveillance data in Hawaii revealed several interesting results. First, the age of the infant’s mother was not a significant predictor of failing to bring the infant in to complete the hearing screening/follow-up process. Often in maternal and child health programs, the youngest mothers, notably teenagers, are at the highest risk for adverse reproductive and parenting outcomes.5 However, in our study, we found that mother’s lack of a high school education was a more reliable predictor of infants who might fail to complete the hearing screening follow-up. Second, white infants were more likely to have an incomplete hearing screening profile, similar to findings in an earlier study in Hawaii that reported that white pregnant women were less likely to receive prenatal care.6 These authors speculated that a small percentage of white women may choose “alternative lifestyles,” preferring nontraditional health care rather than traditional services.6 If this speculation is true, then it could explain the increased odds of failing to complete hearing screening follow-up among white infants in our study.
One of the strengths of our study is that it is population-based, including births from 13 of the 14 civilian birthing facilities in the state. Nevertheless, the findings are limited by the fact that the 2 hospitals not included in the study, 1 civilian (11.1% births) and 1 military (15.4% births), accounted for more than one fourth of the live births in Hawaii in 1999. Another limitation was the lack of information in the surveillance data about the medical status of the infants screened. It is likely that infants who were low birth weight were more medically fragile than their normal birth weight counterparts and were unable to complete the screening/follow-up process. Furthermore, more of the infants whose HI*TRACK files could not be linked with birth certificates were low birth weight compared with those in the study (9.8% vs 7.8%). However, we have no reason to believe that the infants who were excluded from the study had a higher risk of experiencing hearing loss than infants who could be linked, because 93% of the excluded infants passed their initial hearing screening tests (compared with 90% of the included infants).
Although we were unable to link 2128 (17%) of the birth certificates of the 12 456 infants who were born in the 13 hospitals in our study, we have no reason to suspect that any nondifferential biases were introduced. Less than 2% of infants (176 of 10 328) whose records were linked failed to complete the screening/follow-up procedures. The percentage was the same among the 2065 infants whose records could not be linked.
Hawaii’s newborn screening law was amended by the 2001 State Legislature to mandate that “all newborn infants shall be screened for hearing impairment for early identification of children with hearing loss and for the promotion of language and communication.” The new law further allowed the Hawaii Department of Health to clarify roles and responsibilities for agencies involved in service delivery to these newborns. Because of this law, all the hospitals in the state must now report surveillance data to the Hawaii Department of Health. Thus, it is anticipated that the findings from this study can be updated to include births in the additional 2 hospitals in Hawaii.
Hearing loss is a serious problem that affects 4.3 of every 1000 infants born in Hawaii in 1999. Early identification, diagnostic evaluation, and intervention services are important for a child’s acquisition of the fundamental language and communication skills needed for school success.7,8 Activities to refine further Hawaii’s system of early hearing detection and intervention will continue. A major change will be updating the data management information system at birthing facilities statewide to improve data collection.
The results of this study will allow the Hawaii Newborn Hearing Screening Program to target its efforts and limited resources toward ensuring that pediatricians are aware that infants who are at higher risk of not completing the follow-up testing may need special attention to encourage their mothers to complete the process. Moreover, efforts must be made to move quickly with rescreening infants whose initial tests are positive so that a final disposition can be determined and infants who need intervention are not lost to follow-up.
- ↵Finitzo T, Diefendorf AO. The state of the information: evidence gathering in infant hearing programs. Am J Audiol.1997;6 :91– 94
- ↵HI*TRACK. Tracking and Data Management for Newborn Hearing Screening. Logan, UT: National Center for Hearing Assessment and Management, Utah State University; 1999
- ↵Ventura SJ, Curtin SC, Mathews TJ. Teenage Births in the United States: National and State Trends, 1990–1996. National Vital Statistics System. Hyattsville, MD: National Center for Health Statistics; 1998
- ↵Mason JA, Herrmann KR. Universal infant hearing screening by automated auditory brainstem response measurement. Pediatrics.1998;101 :221– 228
- Copyright © 2003 by the American Academy of Pediatrics