PEDIATRICS Vol. 106 No. 4 October 2000, pp. 725-735

Otitis Media in Early Childhood in Relation to Preschool Language and School Readiness Skills Among Black Children

Joanne E. Roberts, PhD^{*, , §}, Margaret R. Burchinal, PhD^*, , Sandra C. Jackson, PhD^*, Stephen R. Hooper, PhD^{*, ¶}, Jackson Roush, PhD^{*, §}, Martha Mundy, MS^{*, §}, Eloise C. Neebe, PhD^*, and Susan A. Zeisel, EdD^{*, #}

From ^* Frank Porter Graham Child Development Center;  Department of Pediatrics; ^§ Division of Speech and Hearing Sciences, Department of Allied Health Sciences; Departments of  Psychology and ^¶ Psychiatry; and the ^# School of Nursing, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina.

	ABSTRACT

Top Abstract Methods Results Discussion Conclusion References

Objective.  To examine whether otitis media with effusion (OME) and associated hearing loss (HL) during the first 5 years of life were related to children's language skills during the preschool years and to school readiness skills at entry to kindergarten.

Methods.  In a prospective study, the ears of 85 black children primarily from low-income families and recruited from community-based childcare programs were repeatedly examined from 6 months to 5 years of age for the presence of OME and from 6 months to 4 years of age for HL when well and ill with OME. Assessments were made annually of the children's child-rearing environments at home and in childcare, and children's language skills between 3 and 5 years of age and readiness skills in literacy and math were evaluated at entry into kindergarten.

Results.  Children had either bilateral or unilateral OME ~30.4% and HL 19.6% of the observation time. OME and associated HL were significantly positively correlated with some measures of expressive language at 3 and 4 years of age; however, these direct relationships were no longer significant when the child's gender, socioeconomic status, maternal educational level, and the responsiveness and support of the home and childcare environments were also considered. Further, both OME and HL were moderately correlated with school readiness skills at entry to school, with children having more OME scoring lower in verbal math problems and with children with more HL scoring lower in math and recognizing incomplete words. These associations continued to remain significant even after partialing out the child and family background factors.

Conclusions.  There was not a significant relationship between children's early OME history or HL and language skills during the preschool years. However, children with more frequent OME had lower scores on school readiness measures. These associations were moderate in degree, however, and the home environment was more strongly related to academic outcomes than was OME or HL. These results should be interpreted cautiously when generalizing to other populations.  Key words:  otitis media, language, hearing loss, school readiness skills, children.

The linkage between frequent and persistent otitis media with effusion (OME) during early childhood and delayed language and development continues to be deliberated.^1-4 Several studies have found that a history of OME in early childhood results in lower scores on measures of language skills during the preschool years^5-7 and lower scores on later academic achievement measures in school,^8-10 whereas other studies have not found these associations for language during the preschool years^9-12 or for later academic skills.^13-15 A child learns language by listening to the stream of speech in the environment, attaching meaning to these sounds, and then abstracting the rules of language. The child with OME typically experiences mild to moderate fluctuating hearing loss (HL) and thus receives a partial or inconsistent auditory signal. This may result in misperceiving or not hearing words and may affect the input to the knowledge base or neural substrate on which language learning is built. Persistent or recurrent HL caused by OME during the formative years of language learning has been presumed to be responsible for later language and academic difficulties. Despite the conceptual logic of the hypothesized relationship between OME and language and academic sequelae, the linkage remains controversial because of inconsistencies in study findings and methodological problems of these studies.^1,2,4

In our previous prospective studies^16,19 we did not find a direct association of OME or related HL and language or cognitive development at 1 or 2 years of age. However, children with more frequent OME and associated HL tended to have less responsive home and childcare environments, which were linked to lower language skills at 1 and 2 years of age. The present study follows the same children and examines how children's OME history and hearing sensitivity during the first 5 years of life are related to children's language development from 3 years of age to kindergarten entry and to school readiness skills at kindergarten entry. This study also examines whether children's development during certain periods in infancy and preschool are more vulnerable to OME and associated HL.

	METHODS

Top Abstract Methods Results Discussion Conclusion References

Study Population

The study participants were 85 black children (39 boys and 46 girls) who were in a longitudinal study of children's otitis media and development. All of the study children participated in an earlier study of OME experience,^17,18 and OME and 2-year developmental outcomes,¹⁹ while 58 of these children who participated in an earlier study of OME and 1-year developmental outcomes² were enrolled into this study without previous knowledge of their ear history. Children were recruited from 9 center-based childcare programs between 6 and 12 months of age (mean: 8.2 months) over a 20-month period. All children had no known medical or genetic abnormalities when entering the study. Sixty-seven children were born at full gestational age (>37 weeks); 11 were born at 37 weeks, 6 at 36 weeks, and 1 at 30 weeks of gestational age.

At entry into the study, 72.9% of the families were classified as low-income based on whether family income was <185% of the federal poverty threshold (income <$20 609 for a family of 3), and 67.1% of the primary guardians were single. At the time of the child's birth, the mothers' mean age was 24 years (standard deviation [SD]: 5.5) and ranged from 14 to 38 years. Seven of the children were not living with their birth mother at 1 year of age: 3 lived with a grandmother, 3 with a foster or adoptive mother, and 1 with a father. The highest level of education attained by the caregivers was less than a high school degree (29%), a high school degree (28%), some college or other training after high school (31%), or a college degree (11%). The mean IQ of the primary guardian was 87.1 (SD: 9.8) as measured by the Vocabulary and Block Design Short-Form of the Wechsler Adult Intelligence Scale-Revised.²⁰ One child was later reclassified as Native American based on parent request but was retained in the study sample because the child was similar in socioeconomic status. Study protocols were approved each year by the Academic Affairs Institutional Review Board at the University of North Carolina at Chapel Hill. The child's parent or guardian provided informed consent.

Otitis Media Documentation

Two nurse practitioners who were trained in pneumatic otoscopy by a pediatrician examined children's ears between study entry and 5 years of age. Children's ears were examined weekly for the first 15 months of the study and then, attributable to a protocol change, biweekly until the age of 5 years, and then monthly for the last year of the study. Children who left their original childcare sites remained in the study and had their ears examined monthly. Children were seen for an average of 85.9 ear examinations between study entry and 5 years of age (SD: 23). The interval between ear observations was ~1 week in length for 11.5% of the observations, 2 weeks for 51.6%, monthly for 33%, and 2 months for 3.8%.

Diagnosis of OME was based on pneumatic otoscopy; routine tympanometry (226 Hz probe tone) was used to corroborate the OME diagnosis. OME was diagnosed when fluid was observed in the middle ear and the tympanic membrane was immobile. A flat (type B) tympanogram was characterized by low static admittance (.2 millimhos), or no discernible pressure peak. Overall, 99.7% (n = 14 558 total ear examinations) were successful by either otoscopy or tympanometry. Of these examinations, 92.6% were successful by otoscopy, 94.8% by tympanometry, and 87.7% by both otoscopy and tympanometry. When the findings of otoscopy and tympanometry did not agree, the otoscopic diagnosis was used. Agreement between judgments of mobility on pneumatic otoscopy and tympanometry based on 12 285 ear examinations (excludes examinations when tympanostomy tubes were present) was 91%. Interobserver agreement levels between the primary nurse practitioner and a board-certified otolaryngologist was very high ( = .87) as previously reported.^16-19 Treatment for acute otitis media was with antibiotics. Parents were informed of findings after each ear examination. The children's primary health care providers were notified of all treatment prescribed for study children. Decisions regarding referrals for further evaluation and consideration of tympanostomy tubes were made by children's primary care providers. Additional information about the OME diagnostic and treatment procedures can be found in previous studies.^16-19

The percentages of observation time with unilateral OME, bilateral OME, and unilateral or bilateral OME (Total OME), and time with no OME from entry into the study until 5 years of age were computed for each child. For each OME episode, the date of onset of OME was subtracted from the date of resolution of OME. The date of onset of OME was computed as midway between the last day that the ears were observed as normal and the first day OME was present. OME resolution was computed as midway between the last day that OME was present and the first day ears were normal.

Hearing Assessment

Visual reinforcement audiometry (VRA) administered between 6 months to 21/2 years of age and play audiometry (PLAY) administered between 21/2 and 4 years of age were used to assess hearing sensitivity. For VRA, randomized frequency-modulated pure tones at 500, 2000, and 4000 Hz were presented through a loudspeaker in a calibrated sound field. For PLAY, pure tones at 500, 1000, 2000, and 4000 Hz were presented via earphones. Children's hearing was tested: 1) routinely on entry into the study and every 3 months; 2) during weeks 1, 4, 7, and 13 after diagnosis of OME; and 3) after a change in ear status (eg, bilateral to unilateral OME, bilateral OME to normal). Between study entry and 4 years of age, children's hearing was tested an average of 22.1 times (SD: 4.7). Testing was performed at each child's childcare center in a mobile testing van that housed a single-wall sound suite. Sound level measurements were conducted periodically at each site and mean ambient noise levels were consistently within allowable levels.²¹ The audiologist ranked the validity of each session on a scale of 1 to 5 based on the child's state (eg, cooperativeness, interest in reinforcers, noise level), with 5 representing excellent validity and 1 representing poor validity. Each week the audiologist and assistant were given lists of children to be tested by the nurse practitioner so that they were blind to the child's ear status. Tympanometry was performed after the hearing assessment. Sessions with validity rankings of 1 (nonreliable) were dropped from the data analysis, as were sessions with only 1 frequency for VRA and 2 frequencies in an ear for PLAY (6.7%). Additional details of the hearing testing procedures can be found in previous studies performed.^16,19

A summary index of HL was calculated by averaging thresholds in each session. For VRA, when the mean of thresholds was 25 dB HL in a session, this was considered indicative of significant HL. For PLAY, the thresholds for left and right ear were averaged. When the mean of these thresholds at 500, 1000, 2000, and 4000 Hz was 20 dB HL in a session, this was considered indicative of significant HL. This computation for HL when VRA was administered is different from the method we used previously.^16,19 We used a mean in the present study rather than 2 or more response levels of 25 dB HL as in our previous studies because beginning at age 21/2 years, we had 2 ear data (PLAY). For the VRA data, computation of HL using this new method, compared with the method used in previous studies,^16,19 resulted in similar percentages of HL and was highly correlated (r = .88; P = .0001). The percentage of days with significant HL was then calculated for each child in the same manner as described above for OME.

Language and School Readiness Measures

Children's language was assessed at 3 years old, 4 years old, and at kindergarten entry, and their early school readiness skills were assessed at entry into kindergarten. At 3 years of age, children's language skills were assessed using 2 standardized measures of language development and a language sample. The Sequenced Inventory of Communication Development (SICD)-Revised²² measures overall receptive and expressive communication; a receptive communicative age (RCA) and an expressive communication age (ECA) were computed. The Peabody Picture Vocabulary Test (PPVT)-Revised²³ measures receptive vocabulary, and a standard score was computed. Both measures have adequate levels of reliability. For the language sample, children also played with a fire engine, firehouse, and other toys while they interacted with an examiner. The child's language during 50 consecutive utterances was entered onto a computerized program (CHILDES²⁴) and scored for length in words of each utterance (mean length of response [MLR]) and proportion of different words in each utterance (Word Type). MLR and Word Type are measures of language during conversational speech that provide an index of language growth. At 4 years of age, children were administered the Clinical Evaluation of Language Fundamentals (CELF),²⁵ which measures overall receptive language (RL) and expressive language (EL); standard RL and EL scores were computed. The CELF has adequate levels of reliability and validity. A language sample where children played with a toy playground and people figures while interacting with an examiner was also administered, and the same measures as for the 3 year olds (MLR and Word Type) were computed. At entry into kindergarten, both the CELF and PPVT were administered again. At entry into kindergarten, the children were also administered 3 subtests from the Woodcock-Johnson Psychoeducational Battery (WJPB).²⁶ These included Letter-Word Identification, which assesses reading identification skills in identifying isolated letters and words (eg, "What is the name of this letter?", while pointing to the letter O); Applied Problems, which assesses skills in analyzing and solving verbal math problems (eg, showing a picture of 4 unhatched and 2 hatching eggs and asking, "How many eggs have not hatched?"); and Incomplete Words, which measures a child's ability to identify complete words after hearing a tape recorded word that has 1 or more phonemes missing (eg, "otato" for "potato"). Standard scores for Letter-Word Identification, Applied Problems, and Incomplete Words were computed. One of 4 trained examiners with expertise in speech and language skills and blind to the child's OME and hearing history, administered the assessments within 2 weeks of the child's birthday. Tests were given in the sound room of a mobile test van or in a sound room at a university research center.

Child-Rearing Environment

The Home Observation for Measurement of the Environment-Inventory for Infants²⁷ (HOME) assessed at 9, 18, 30, and 42 months of age the overall quality and responsiveness of the home environment. The HOME, which has high levels of validity and reliability,²⁸ has 6 subscales: emotional and verbal responsivity of the parent, acceptance of the child's behavior, organization of the environment, provision for appropriate play materials, and maternal involvement with the child. A HOME total score was computed for 9, 18, 30, and 42 months by adding together the item scores at each age and an average of the 4 scores (HOME total) was used in the analyses. Over the 5-year period, home visits were conducted by 2 trained nurse practitioners and 2 speech-language pathologists who had achieved interrater agreement of at least 95% before collecting the data.

Quality of the caregiver and of the physical environment in each child's classroom in childcare was assessed using the Infant/Toddler Environment Rating Scale (ITERS)²⁹ during the first and second years that children were in childcare and the Early Childhood Environment Rating Scale (ECERS)³⁰ during the third, fourth, and fifth years that children were in childcare. Both instruments have high levels of reliability and validity.^29,30 The ITERS examines the following: furnishings and display for children, personal care routines, listening and talking, learning activities, interaction, program structure, and adult needs. An average item score was computed for each assessment by dividing the sum of the items scored by the number of items scored. The ECERS examines the developmental appropriateness of classroom practices by assessing routine care needs; furnishings and display; activities and experiences related to motor, cognitive, social, and language development; and adult provisions. The scores for all of the ITERS and ECERS assessments were then averaged and the ITERS/ECERS total was used in the analyses. An examiner who was blind to the children's OME and hearing status rated each classroom during a 3- to 4-hour observation in the spring of each year of the child's first 5 years of life. Over this period, 1 of 5 trained observers who had achieved good interrater reliability (r  .85) with the test developers and trained ITERS observers did the observations.

Data Analysis

In this study, both descriptive and inferential analyses tested the associations among OME or HL, quality of care environments, and infant outcomes. Because both OME and HL were skewed within this sample, a rank transformation was applied to all of the OME and HL variables. There was 1 child who did not have complete hearing data, and this child was included only in the OME analyses. In the descriptive analyses, the linear association between the measures of otitis media, HL, family characteristics, quality of care environments, and child outcomes were examined using zero-order correlations. Pearson correlations were run among rank of Total OME, rank of HL, ratings of the quality of the family and childcare environments, and the child outcomes at each age.

The general analysis strategy for the inferential analyses involved multivariate multiple regression analyses that examined the extent to which amount of OME and HL during early childhood was related to language and academic skills at 3 and 4 years of age and at entry to kindergarten. The multivariate tests analyzed 4 sets of child outcomes: 1) language skills at 3 years oldSICD-ECA, SICD-RCA, language sample-MLR, and language sample proportion Word Type; 2) language skills at 4 years oldCELF-EL, CELF-RL, language sample-MLR, and language sample proportion Word Type; 3) language skills at entry to kindergartenCELF-EL and CELF-RL, and PPVT receptive vocabulary; and 4) academic achievement at entry to kindergartenWJPB Incomplete Words, Letter-Word Identification, and Applied Problems. The predictor variable of interest in 1 set of regressions was the rank of the proportion of time of OME and in the other set of regressions was the rank of the proportion of time with HL. For the 3-year outcomes, the predictors of interest were the ranks of proportions of time with OME and HL from entry to the study in the child's first year through their third birthday. For the 4-year outcomes, the predictors of interest were the ranks of the proportions of time with OME and HL from entry to the study in the child's first year through their fourth birthday. For the entry to kindergarten outcomes, the predictors of interest were the ranks of the proportions of time with OME before the child's fifth birthday and proportion of time with HL before the child's fourth birthday. Because the hearing data were complete only through age 4 years, the hearing analyses stopped at 4 years, while the OME analyses went though 5 years of age. The covariates in the regressions included gender, poverty, maternal education, average HOME total score, and average ITERS/ECERS scores to adjust for the child's gender, whether the family lived in poverty, highest grade of school mother completed, and the responsiveness and support of the home and childcare environments.

When significant multivariate associations between either OME or HL with developmental outcomes emerged, follow-up multiple regression analyses were conducted to determine whether OME or HL during a specific age periodinfancy (6 months to 2 years) versus preschool (2-5 years for OME and 2-4 years for HL) was differentially related to language and cognitive outcomes. The predictors and outcomes were identical to those in the previous analyses except that OME and then HL during the infancy and preschool periods were used instead of OME and HL for the entire early childhood period.

	RESULTS

Top Abstract Methods Results Discussion Conclusion References

Otitis Media Experience

The distributions of bilateral OME, unilateral OME, and Total OME from 6 months to 5 years of age in yearly intervals are shown in Table 1. Total OME decreases considerably over time from 88% in the first year of life, to 55% in the second year, to 20% in the third year, to 15% in the fourth year, to 11% in the fifth year. Except for the fifth year of life, bilateral, compared with unilateral OME, occurred most often. Unilateral OME remained relatively consistent across the age groups. Total OME occurred 62.8% of the time (SD: 26.2) during the infancy period from 6 months to 2 years of age and 15.9% (SD: 22.4) during the preschool period from 2 to 5 years of age and 30.4% (SD: 20.9) across the entire study period of 6 months to 5 years. Children who had more bilateral OME had more Total OME between 6 months and 5 years of age (r = .92; P = .0001). Because Total OME and bilateral OME were so highly correlated, Total OME was used in the analyses of OME and the child rearing and outcome measures as described below. Figure 1 shows the variability in percentage of time with Total OME across the age groups.

View larger version (23K): [in this window] [in a new window]   Fig. 1.   Distribution of study population for percentage of time with Total OME from 6 months to 5 years of age.

Hearing Sensitivity

Children's mean hearing thresholds for VRA were 20.3 dB HL at 500 Hz, 21.0 dB HL at 2000 Hz, and 25.1 dB HL at 4000 Hz. Mean hearing thresholds for PLAY averaged across the right and left ears were 16.5 dB HL at 500 Hz, 12.8 dB HL at 1000 HZ, 10.5 dB HL at 2000 Hz, and 13.3 dB HL at 4000 Hz. Next, we applied the HL criteria of 25 dB HL or greater for VRA and 20 dB HL or greater for PLAY as indicative of HL for these thresholds in the right and left ears. As shown in Table 1, HL occurred in 44.6% of the observation time from 6 to 12 months, 25.9% from 1 to 2 years, decreasing to 14.1% from 2 to 3 years, and 10.8% from 3 to 4 years of age. HL occurred 30.6% of the time (SD: 28.7) during the infancy period from 6 months to 2 years, 12.3% (SD: 21) during the preschool period from 2 to 4 years, and 19.6% (SD: 21.6) across the entire study from 6 months to 4 years of age. Figure 2 shows the variability in percentage of time with HL from 6 months to 4 years of age. Children who experienced more Total OME from 6 months to 4 years of age had more HL (r = .77; P = .0001) over this period.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Distribution of study population for percentage of time with HL from 6 months to 4 years of age.

Otitis Media, Hearing, Quality of Child Rearing, and Developmental Outcomes

As shown in Table 2, the outcome measures were at the population means for expressive language on 1 test (CELF-EL at 4 years of age and kindergarten entry) and somewhat below the population means for all other standardized tests using both age equivalents (SICD-RCA and SICD-ECA at 3 years of age) and standardized scores (WJPB scale scores at kindergarten entry, CELF-RC at 4 years of age and kindergarten entry, and PPVT at kindergarten entry). The outcomes showed moderate to high correlations over time as is expected from repeated assessments of the same or related constructs (see Table 3). The various measures of OME and HL tended to be fairly highly correlated. OME and HL showed only mild associations with some of the child, family, and childcare covariates (see Table 4). Slightly more infant OME and HL were experienced by children living in poverty, children whose mothers had less education, and children living in less stimulating home environments.

Analyses examined the association between developmental outcomes and both OME and HL separately by age and type of outcome. The association between developmental outcomes and both OME and HL were examined descriptively with correlations. The association was examined inferentially with multivariate multiple regressions attributable to the moderate to high levels of correlations among the outcomes. Partial correlations were computed from the regression analyses to provide a more interpretable statistic than the regression coefficients to describe the association.

Language at 3 Years of Age First, we examined the zero-order correlations descriptively. Children with more Total OME scored lower on both the SICD-ECA and SICD-RCA, and children with more HL scored lower on the SICD-ECA and on the proportion of different words types during the language sample (see Table 2 for zero-order correlations). However, the multivariate regressions indicated that neither OME (F[5,68] = .60; P = .70) nor HL (F[5,67] = .77; P = .58) was significantly associated with the 3-year language measures after adjusting for child gender, family poverty status, maternal education, and the quality of the home and childcare environments (see Table 2 for the partial correlations).

Language at 4 Years of Age OME was not correlated with any of the outcomes and HL was correlated with only 1 outcome, proportion of different word types from the language sample. Multivariate regressions indicated that neither OME (F[4,72] = 1.63; P = .17) nor HL (F[4,71] = 1.38; P = .25) was significantly associated with the 4-year language measures after adjusting for the selected covariates (see Table 2 for the zero-order and partial correlations).

Language at Kindergarten Entry Neither OME nor HL was correlated with any of the outcomes. Again, multivariate regressions indicated that neither OME (F[3,56] = .26; P = .85) nor HL (F[3,57] = .33; P = .81) was significantly associated with the 48-month language measures after adjusting for the selected covariates (see Table 2 for the zero-order and partial correlations).

School Readiness at Kindergarten Entry The zero-order correlations indicated that children with more OME tended to score slightly lower on WJPB Applied Problems and that children with more HL scored slightly lower on WJPB Applied Problems and Incomplete Words. The multivariate regressions also indicated that both OME (F[3,63] = 3.00; P = .04) and HL (F[3,64[ = 3.30; P = .03) were significantly associated with the academic achievement scores at entry to kindergarten in analyses that adjusted for gender, poverty, maternal education, and quality of the home and childcare environments. In these analyses, both OME and HL were negatively associated with WJPB Applied Problems and HL was negatively associated with WJPB Incomplete Words (see Table 2 for the zero-order and partial correlations and Table 5, model 1 for the regression coefficients). However, as Table 5 shows, HOME was a much stronger predictor of the school readiness measures than was OME or HL. Figure 3 shows a plot of the association between HL 6 months to 4 years of age and scores on the WJPB Incomplete Words at entry into kindergarten. Figure 4 shows a plot of the association of the scores on the average of the HOME total over time and the WJPB Incomplete Words at kindergarten entry. As can be seen, children who have more HL scored lower on WJPB Incomplete Words (r = .29; P = .0117). However, as the plot shows, the relationship was stronger between the HOME total score and the WJPB Incomplete Words (r = .46; P = .0001).

View larger version (13K): [in this window] [in a new window]   Fig. 3.   Distribution of HL from 6 months to 4 years of age and WJPB Incomplete Words.

View larger version (14K): [in this window] [in a new window]   Fig. 4.   Distribution of HOME total averaged 6 months to 4 years of age and WJPB Incomplete Words.

	DISCUSSION

Top Abstract Methods Results Discussion Conclusion References

In this study, there was a significant positive correlation between both OME and HL with several measures of language development at 3 and 4 years of age; however, these direct relationships were not significant when the child's gender, socioeconomic status, maternal educational level, and the responsiveness and support of the home and childcare environments were considered in the analyses. Yet, both OME and HL were moderately correlated with school readiness measures at entry to kindergarten, with children having more OME over time scoring lower in math skills and children with more HL over time scoring lower in both math and an auditory closure task. These associations continued to remain significant even after partialing out the child and family background factors. The results did not support a critical period of experiencing OME in infancy or the preschool years in which OME or HL matters most. In this study, OME and associated HL were common among black children attending group childcare through 2 years of age, yet the prevalence of OME decreased steadily between 2 years of age and entry into kindergarten.

We did not find a direct association between both OME and associated HL with language outcomes from 3 to 5 years of age and this is consistent with some prospective studies^9,11,12 but inconsistent with other prospective studies^5-7 that did find a direct association. Gravel and Wallace⁹ did not find an association with previous OME history for receptive or expressive communication at 4 years of age for high-risk and low birth weight infants. Similarly, Wright and colleagues¹¹ did not find an association for receptive and expressive language in 2- to 4-year-old low-income children. Roberts et al¹² in a different cohort of low- and middle-income children attending childcare also failed to find an association for receptive and expressive language for 41/2- to 6-year-olds. In contrast, Teele and colleagues⁵ reported lower receptive and expressive language skills for low- and middle-income children with a history of OME versus without a history of OME in a medical practice at 3 years of age. Rach and colleagues⁶ also reported lower expressive language for 2- to 4-year-olds with a history of OME living in The Netherlands. Finally, Maw and colleagues⁷ recently reported that children (mean age: 3 years) with chronic OME who received bilateral ventilating tubes early (and were free of OME), compared with children who had received tubes 9 months later (who may have had continued OME), had higher receptive and expressive language skills 9 months later after their assignment to groups but not 18 months later. The present study findings are difficult to compare with the previous studies that examined direct OME language sequelae, because previous studies did not consider the role of important child and environmental factors that may affect children's language development. The ecological approach used in this study would seem critical to this type of investigation.

The findings of a significant relationship between early OME and verbal math problems and between early HL are both math and auditory closure skills, but no significant findings for early reading skills is similar to the findings of some studies^8-10 but in contrast to others.^13-15 Teele et al⁵ reported lower scores in math and reading for middle-income but not for low-income 7-year-olds in a medical practice with a history of OME, compared with children without an OME history. Gravel and Wallace⁹ also found lower reading scores for 6-year-old children with a history of OME who were high-risk and low birth weight. Further, Lous¹⁰ reported lower reading scores for cohort of low- and middle-income children from Denmark. In contrast, Roberts and colleagues^13,14 in a different cohort of children reported no associations between early OME history and later math and reading skills at entry to kindergarten and in second grade for low-income children who attended childcare. Further, Peters and colleagues¹⁵ also did not find an association with early OME history and later reading or math skills for 7-year-olds, although associations were found for writing and spelling. The findings of the present investigation did consider the role of important child and environmental factors affecting children's early school readiness skills and, thus, are hard to compare with previous studies.

The findings in this study that children with more OME over time scored lower in solving verbal math problems and that children with more HL over time scored lower in math and on an auditory closure task at kindergarten entry is interesting, given the lack of associations with the language measures. The auditory closure task requires the ability to recognize a word, retrieve the word from memory, identify the sounds that are missing, and blend the word together. It involves emergent phonological skills, working memory, and auditory processing. It is possible that children experience a delay in phonological processing secondary to a history of OME and this affects the child's ability to make these associations. Applied problems measure a child's ability to analyze simple math problems presented verbally and to engage in practical problem solving in math. The child must decide what of the verbally presented information to include and exclude to do the calculation. It requires working memory and processing of auditory-based information. Perhaps OME and HL set the stage for the vulnerability of early language-based literacy skills. For both of these WJPB tasks, we cannot tell whether working memory or phonological or auditory processing is affected directly or indirectly by OME, because the tasks on the WJPB do not separate out each of these skills. However, we are looking at each of these processes in other aspects of the current project.

We also found in this study that the home and childcare environments did not mediate the association between OME or HL and the later child outcomes as in our previous studies of the infancy period.^16,19 The lack of support for the mediational hypothesis may be attributable to the fact that once children have established basic language form and functions during the preschool years, the home and childcare environments do not play the same major role in affecting children's development. It may be that during the first 2 years of life caregiver interaction style plays a more important role in affecting parent-child interactions and language development. As children become more proficient at the basic tenants of language and become more independent and self-directed, caregiver interactional style seems less important in the OME developmental linkage, although this is highly speculative.

Based on study results, we found that the amount of OME and HL that children experience during infancy versus the preschool period was not differentially related to children's later achievement skills at entry into kindergarten. We hypothesized that certain periods of development may be more vulnerable to OME effects (eg, when children are learning to count vs learning to add). Yet, our study findings indicated no differential associations, suggesting that through the age of 5 years, the OME associations with early language and school readiness skills are similar across the selected periods.

This study had much strength, compared with previous studies. First, this prospective study followed a group of children who attended childcare from infancy through their preschool years. Second, a standard protocol was used to assess OME status, hearing, and children's development at repeated intervals. Third, children's hearing sensitivity in addition to their OME experience were measured often when children were both well and when ill with OME. Finally, family background variables (eg, child's home and childcare environments) that are known to affect children's development and often were not controlled for in previous studies were measured.

The results of this study should be interpreted cautiously for several reasons. First, characteristics of this study population may restrict generalization to other study populations. The children were recruited from community childcare programs, all children were black, and the children's families were primarily low-income. Also, a high incidence of OME was found in this population of children, attending community-based childcare programs as described by our previous work.^16-18 Second, we measured children's early school readiness at kindergarten entry and the predictive value of these readiness skills remains to be seen. Nonetheless, these types of tasks have been described as critical in the measurement of early literacy development. Third, we did not examine other measures of children's language development (eg, language use), speech (eg, perception or production), attention, auditory processing, or behavior. These other measures as outcomes may lead to different results. Fourth, the significant associations were mild in degree and indicate that multiple factors in addition to the variables studied, likely influence children's development. This ecological approach seems critical to studying longitudinal trends in development. Finally, this study used a correlational design and the results do not imply that OME causes later delays in school readiness.

	CONCLUSION

Top Abstract Methods Results Discussion Conclusion References

The results indicate that there was not a significant relationship between early children's early OME history or HL and language skills during the preschool years. We did find, however, that children with more frequent OME tended to score lower in early math skills and an auditory closure task at entry into school. These associations were moderate in degree and the home environment was more strongly related to school readiness outcomes than was OME or HL. These results should be interpreted cautiously when generalizing to other populations, because the children experienced considerable amounts of OME in early childhood, attended childcare, and were primarily from low-income families. We continue to follow these children into the early elementary school years to explore the interrelationships among children's OME and hearing history, family and childcare environments, language development, and academic skills.

	ACKNOWLEGMENTS

This research was supported in part by the Maternal and Child Health Program (Grants MCJ-370599, MCJ-370649, and R40 MC 00145 Title V, Social Security Act), Health Resources and Services Administration, US Department of Health and Human Services, and National Institute of Health Grant 01R01-CD03817-01A1.

We thank the children and families who participated in this study.

We have greatly appreciated the support of Sarah Henderson in this work.

We also want to express our special thanks to the following audiologists for their advice regarding classification of hearing loss: Drs Judy Gravel, John Grose, Joe Hall, Lisa Hunter, and Robert Nozza.

	FOOTNOTES

Received for publication Jul 23, 1999; accepted Jan 7, 2000.

Reprint requests to (J.E.R.) Frank Porter Graham Child Development Center, University of North Carolina at Chapel Hill, 105 Smith Level Rd, CB 8180, Chapel Hill, NC 27599-8180. E-mail: joanne  roberts{at}unc.edu

	ABBREVIATIONS

OME, otitis media with effusion; HL, hearing loss; SD, standard deviation; Total OME, unilateral or bilateral otitis media with effusion; VRA, visual reinforcement audiometry; PLAY, play audiometry; SICD, Sequenced Inventory of Communication Development; RCA, receptive communication age; ECA, expressive communication age; PPVT, Peabody Picture Vocabulary Test; MLR, mean length of response; CELF, Clinical Evaluation of Language Fundamentals; EL, expressive language; RL, receptive language; WJPB, Woodcock-Johnson Psychoeducational Battery; HOME, Home Observation for Measurement of the Environment-Inventory for Infants; ITERS, Infant/Toddler Environment Rating Scale; ECERS, Early Childhood Environmental Rating Scale.

	REFERENCES

Top Abstract Methods Results Discussion Conclusion References

1. Stool SE, Berg AO, Berman S, et al. Managing Otitis Media With Effusion in Young Children: Clinical Practice Guidelines, Number 12. Rockville, MD: Agency for Health Care Policy and Research, Public Health Service, US Department of Health and Human Services; 1994. Agency for Health Care Policy and Research publication 94-0622
2. Roberts JE, Wallace IF. Language and otitis media. In: Roberts JE, Wallace IF, Henderson F, eds. Otitis Media in Young Children: Medical, Developmental and Educational Considerations. Baltimore, MD: Paul H. Brookes; 1997
3. Ruben RJ, Haggard MP, Bagger-Sjoback D, Complications and sequelae: recent advances in otitis media report. Ann Otol Rhinol Laryngol. 1998; 174:81-94
4. Paradise JL Otitis media and child development: should we worry? Pediatr Infect Dis J 1998; 17:1076-1083 [CrossRef][Medline]
5. Teele DW, Klein JO, Rosner BA, and the Greater Boston Otitis Media Study Group Otitis media with effusion during the first three years of life and development of speech and language. Pediatrics 1984; 74:282-287 [Abstract/Free Full Text]
6. Rach GH, Zielhuis GA, van den Broek P The influence of chronic persistent otitis media with effusion on language development of 2- to 4-year-olds. Int J Pediatr Otorhinol 1988; 15:253-261 [CrossRef][Medline]
7. Maw R, Wilks J, Harvey I, Peters TJ, Golding J Early surgery compared with watchful waiting for glue ear and effect on language development in preschool children: a randomised trial. Lancet 1999; 353:960-963 [CrossRef][Medline]
8. Teele DW, Klein JO, Chase C, Otitis media in infancy and intellectual ability, school achievement, speech, and language at age 7 years. J Infect Dis 1990; 162:685-694 [Medline]
9. Gravel JS, Wallace IF Listening and language at 4 years of age: effects of early otitis media. J Speech Hear Res 1992; 35:588-595
10. Lous J Silent reading and secretory otitis media in school children. Int J Pediatr Otorhinol 1993; 25:25-38 [CrossRef][Medline]
11. Wright PF, Sell SH, McConnell KB, Impact of recurrent otitis media on middle ear function, hearing, and language. J Pediatr 1988; 113:581-587 [CrossRef][Medline]
12. Roberts JE, Burchinal MR, Davis BP, Collier AM, Henderson FW Otitis media in early childhood and later language. J Speech Hear Res 1991; 34:1158-1168
13. Roberts JE, Sanyal MA, Burchinal MR, Collier AM, Ramey CT, Henderson FW Otitis media in early childhood and its relationship to later verbal and academic performance. Pediatrics 1986; 78:423-430 [Abstract/Free Full Text]
14. Roberts JE, Burchinal MR, Collier AM, Otitis media in early childhood and cognitive, academic, and classroom performance in the school-aged child. Pediatrics 1989; 83:773-786
15. Peters SAF, Grievink EM, van Bon WHJ, Schilder AGM The effects of early bilateral otitis media with effusion on educational attainment: a prospective cohort study. J Learn Disabilities 1994; 27:111-121
16. Roberts JE, Burchinal MR, Zeisel SA, Otitis media, the caregiving environment, and language and cognitive outcomes at two years. Pediatrics 1998; 102:346-352 [Abstract/Free Full Text]
17. Zeisel SA, Roberts JE, Gunn EB, A prospective study of the prevalence of otitis media with effusion among African American infants in group child care. J Pediatr 1995; 127:875-880 [CrossRef][Medline]
18. Zeisel SA, Roberts JE, Neebe EC, A longitudinal study of otitis media with effusion among 2- to 5-year old African American children in child care. Pediatrics 1999; 103:15-19 [Abstract/Free Full Text]
19. Roberts JE, Burchinal MR, Medley, et al Otitis media, hearing sensitivity, and maternal responsiveness in relation to language during infancy. J Pediatr 1995; 126:481-489 [CrossRef][Medline]
20. Wechsler D. Wechsler Adult Intelligence Scale-Revised. San Antonio, TX: Psychological Corporation, Harcourt Brace Javonovich; 1981
21. American National Standards Institute. Criteria for Permissible Ambient Noise During Audiometric Testing (ANSI 3.1-1977). New York, NY: American National Standards Institute; 1977
22. Hedrick DL, Prather EM, Tobin AR. Sequenced Inventory of Communication Development-Revised. Seattle, WA: University of Washington Press; 1984
23. Dunn LM, Dunn LM. Peabody Picture Vocabulary Test-Revised. Circle Pines, MN: American Guidance Service; 1981
24. MacWhinney B. The CHILDES Project: Tools for Analyzing Talk. 2nd ed. Hillsdale, NJ: Laurence Erlbaum Associates, Publishers; 1997
25. Semel E, Wiig E, Secord W. Clinical Evaluation of Language Fundamentals. 3rd ed. San Antonio, TX: Psychological Corp; 1995
26. Woodcock RW, Johnson MB. Woodcock-Johnson Psycho-Educational Battery-Revised. Allen, TX: DLM Teaching Resources; 1990
27. Caldwell BM, Bradley RH. Home Observation for Measurement of the Environment-Revised. Little Rock, AR: University of Arkansas; 1984
28. Bradley RH, Caldwell BM, Rock SL, Home environment and cognitive development in the first 3 years of life: a collaborative study involving six sites and three ethnic groups in North America. Dev Psychol 1989; 25:217-235 [CrossRef]
29. Harms T, Cryer D, Clifford RM. Infant/Toddler Environment Rating Scale. New York, NY: Teachers College Press; 1990
30. Harms T, Clifford R. Early Childhood Environment Rating Scale. New York, NY: Teachers College Press; 1980
31. Baron RM, Kenny DA. The moderator-mediator variable distinction in social psychology research: Conceptual, strategic, and statistical considerations. J Pers Soc Psychol. 1986;1173-1182