Impact of Degree of Hearing Loss and Very Early Enrollment in EI on Language Betty Vohr, MD Julie Jodoin-Krauzyk MED, MA Richard Tucker BA, Mary Jane Johnson.

Impact of Degree of Hearing Loss and “Very

Early” Enrollment in EI on LanguageBetty Vohr, MD

Julie Jodoin-Krauzyk MED, MA Richard Tucker BA,

Mary Jane Johnson MED, Deborah Topol, BA,

Marianne Ahlgren PhD, Women & Infants Hospital, RIHAP,

Family Guidance, Rhode Island School for the Deaf

Brown Medical School

Funding

• Cooperative Agreement from the Centers for Disease Control and Prevention

• #URs/CCU12033-01

• AUCD from the Centers for Disease Control and Prevention

Background

• 33 babies are born each day in the US with HL

• 12,000 to 16,000 babies are born each year with HL

• The rate at birth is 2-3 per 1000

• NICU rate is 10 per 1000

• Well baby rate is 1 per 1000

Approximately 95% of newborns have their hearing screened in the United States

We are continuing to learn about the outcomes of these babies and factors predicting outcome

Moderate, Severe to Profound HL

• There is strong evidence that congenital bilateral moderate, severe and profound hearing loss compromises communication skills, language and behavior, at school age.

Mild Hearing Loss

• There is increasing evidence that unilateral or mild/minimal hearing loss < 40 dB also impacts academic and language outcomes at school age. This impact is important since in some states infants with minimal or unilateral HL are not eligible for early intervention.

Hearing Loss and Early Intervention

• There is evidence that infants with congenital HL without other disabilities who receive early intervention services before 6m will have language development similar to hearing peers.

Yoshinaga-Itano 1998; Moeller 2000.

Age of Identification In Rhode Island over time has decreased significantly

• Prior to hospital screening 30 to 60m

• 1986 24-30m

• 2007 3m

Rhode Island Data

Objectives• To determine the language outcomes of infants with: • Moderate to profound bilateral HL (Mod/P HL)

• Mild (< 40dB) or unilateral hearing loss (Min HL)

• Normal hearing controls

• To identify factors contributing to language scores

Hypotheses

• Infants with Min HL will have language scores similar to hearing infants at 12-16 m.

• Infants with Mod/P HL will have less optimal language outcomes at 12-16 m.

• Infants enrolled in EI ≤ 3 m will have better language scores at 12-16 m compared to infants enrolled > 3 m.

Study Design

•Resources/Support

•Early Intervention

• Degree of HL

•NICU

Language

at

12-16 months

Mod/Prof HL

Mild HL

Typical Hearing

Study GroupsStudy Groups MediatorsMediators OutcomeOutcome

Subject Recruitment

• Eligible families were identified in the RI Hearing Assessment database (RITRACK).

• Families were recruited through the Family Guidance Program (statewide specialty EI program for children with HL).

• Families not in the program were recruited by mail.

• Infants born between 10-02 and 1-05 were eligible

Methods

• Infants with HL were identified and enrolled.

• Matching for Controls was based on a hierarchical matching procedure for gender, DOB ±90days, birthing hospital, NICU status, maternal education, race, and health insurance.

• 33 infants with all degrees of HL were enrolled.

• Informed consent was obtained for all subjects.

Study Subjects

30 children with Hearing Loss were seen at 12+4 m.

Children Mothers

Moderate/Profound HL 18* 15

Mild HL 12 12

Hearing controls 96 85

* 6/18 have profound HL

Home visits were conducted

Assessments at 12+4 months

1. Family Resource Scale (Dunst & Leet)

2. Family Support Scale (Dunst et al)

3. Parenting Stress Index (Abidin)

4. Impact on the Family (Stein & Reissman)

5. Family Demographics

6. MacArthur-Bates Communicative Inventory: Words & Gestures (CDI)

MacArthur Communicative Inventory (CDI)

• Words and Gestures were completed by the mothers• Part I: Early Words includes 33 questions about child

comprehension of initial phrases and a 396 vocabulary checklist.

• Part II: Actions and Gestures has a 63 gesture checklist.• Percentile scores are available for:

phrases understood, words understood, words produced, and early , late, and total gestures.

Administration

• English speaking families: standard CDI

• Native Spanish speakers: the normed Spanish version

• ASL families: The ASL version of the CDI

• If signed vocabulary was understood or used by the child, it was counted accordingly on the English version.

• Scores are reported separately for 2 non-native English speakers: 1 Control – Mandarin language 1 Mod HL – American Sign Language

Data Analysis

1. Three-Way Analysis

– Mod/P HL vs. Mild HL vs. Controls

– ANOVA to analyze differences among group means

– Ҳ 2 to analyze differences among proportions

2. Regression Models were run to show relationships of predictors with outcomes for the total cohort and for the group with HL controlling for confounders.

Maternal Characteristics*Mod/P HL

(n=15)

Mild HL

(n=12)

Control

(n=85)

P

Age 32 ± 7 32 ± 6 32 ± 6 0.999

Married 80% 83% 87% 0.748

≤high school 27% 16% 11% 0.390

Primary English 87% 92% 98% 0.124

Medicaid / No Ins. 20% 16% 17% 0.909

Socio-economic status

41 ± 16 41 ± 13 45 ± 11 0.309

*Mothers with multiples are counted once and assigned to most severe HL group

Child CharacteristicsMod/P HL

(n=18)

Mild HL

(n=12)

Control

(n=96)

P

Female 39% 42% 36% 0.929

White 94% 92% 95% 0.689

NICU 83% 33% 55% 0.019

<37 weeks 83% 42% 46% 0.011

EI 100% 83% 28% 0.001

Types of Hearing Loss14

1

3

6

4

2

0

2

4

6

8

10

12

14

16

SNHL COND AN

Mod-Prof

Mild

*

**

*1 mild bilateral **1 mild bilateral

Mean Chronological* Age at Visit

Group Mod/P HL Mild HL Control P

Age (m) 16.4±2 15.1±2 15.6±2 0.237

Range (11-17) (11-17) (11-17)

* Scores were derived from corrected ages for premature infants

Percent Early Intervention ≤ 3 months

0

10

20

30

40

50

60

70

80

90

100

Mod/ P HL Mild HL

Perc

ent

p=0.176

6/18(33%)

7/12 (58%)

Maternal Report at 12+4 months

Mod/P HL Mild HL Control P

Total Support 36.4±9 37.3±11 35.2±11 0.75

Total Resources 121±11 126±7 125±13 0.54

Total Impact 26.8±5 25.2±4 24.9±4 0.19

Total Stress 64.6±11 67.8±15 67.9±17 0.70

MacArthur CDI % Scores at 12+4 monthsScore

N

Mod/P HL

17

Mild HL

9

Control

77

P

Phrases Understood % 20±25+* 52±22 41±30 0.002

Words Understood % 17±24* 49±25 35±28 0.012

Words Produced % 20±21* 49±34 36±26 0.015

Early Gestures % 29±30* 59±30 44±31 0.060

Late Gestures % 12±8+* 42±30 33±28 0.006

Total Gestures % 12±10+* 45±30 34±28 0.003+ versus Control * versus Mild HL

Words Understood: Percent < 10th

0

10

20

30

40

50

60

70

Mod/P

Mild

Con

65% P<0.0086

25% 27%

16.4 m 15.1m 15.6m

Words Produced: Percent < 10th

0

10

20

30

40

50

60

Mod/PMild

Con

53% P<0.0650

25% 25%

16.4 m 15.1m 15.6m

Mean Number of Words Produced at Visit

0

5

10

15

20

25

Mod/PMild

Con

5

P<0.0114#

20 17

16.4 m 15.1m 15.6m

# Analyses of log transformed data

Imitation Skills Present (yes/no)

0

10

20

30

40

50

60

70

80

Mod/P

Mild

Con

41%

NS

67% 65 %

16.4 m 15.1m 15.6m

Labeling Skills Present (yes/no)

0

5

10

15

20

25

30

35

40

45

50

Mod/P

Mild

Con

6%42% 45 %

P=.009

16.4 m 15.1m 15.6m

Early and Later Gestures Produced: Percent< 10th

0

10

20

30

40

50

60

70

Early Gestures Later Gestures

Mod/PMild

Control

P=0.1178 P=0.0953

53 25 28

%

59 33 32

Two Additional Children with CDI Data

Language ASL Mandarin

Study Group Mod/P HL Control

Words

Understood 43 41

Produced 26 2

Total Gestures 24 21

# Words Produced at Visit for Infants with HL Enrolled in EI ≤ 3 m and Hearing Controls

02468

10121416182022

HL in EI ≤3m Controls

HL in EI ≤3m

Controls

NS

20 17

Mean # Words Produced Reported at Visit by Age of Entry to Early Intervention for Children with HL

0

5

10

15

20

25

12 months

≤3months

>3months

P=0.02

20 Words

4 words

< 10th Percentile for Words and Phrases Understood by Age of Entry to EI

0

10

20

30

40

50

60

Words understood <10th% Phrases understood<10th%

≤3 months>3 months

38 56 38 49

NS NS

Gestures Percentile for Children with HL by Age of Entry to EI

0

10

20

30

40

50

60

70

Early Later Total

≤3 months>3 months

.0001

.02 .01

Total Cohort: Regression to Predict CDI

Words Understood %Predictor b p• NICU -14.7 .01• Mod/P HL -16.0 .03• Mild HL 9.3 .32• SES - 0.3 .22• Support - 0.2 .56• Resources - 0.3 .19

Model R2=.19 .002

Total Gestures %Predictor b p• NICU -13.4 .01• Mod/P HL -20.7 .004• Mild HL 6.4 .48• SES -0.3 .22• Support 0.1 .61• Resources -0.3 .09

Model R2=.21 .0009

Group with HL: Regression to Predict CDI

Words Produced %

Predictor b p• NICU 11.8 .32• Mod/P HL -25.1 .03• EI ≤ 3 m 37.4 .0008

Model R2=.56 .0003

Early Gestures %

Predictor b p• NICU -3.6 .77• Mod/P HL -14.4 .21• EI ≤ 3 m 40.7 .0006

Model R2=.59 .0002

Conclusions

• Infants with Moderate to Profound HL who have hearing parents have significantly more language delays at 12+4m than infants with Minimal HL and controls.

• Infants with HL enrolled in Early Intervention ≤ 3m have better language skills at 12m than infants enrolled > 3m. These skills are similar to hearing controls.

Strengths of StudyStrengths of this study are :

• Infants with all degrees of HL are included

• “Very early” enrollment in EI is evaluated

• The study includes control children with typical hearing.

Weaknesses of the study are:

• The language information at 12+4 m was obtained by parent report and not direct assessment.

• The majority of children with Moderate-Profound HL were NICU infants.

Conclusions

• All infants diagnosed with permanent HL benefit from specialized intervention programs which meet their developmental and communicative needs and optimize outcomes.

CDI: Imitation and Labeling Examples

• Imitation: child parrots or imitates things s/he has heard including phrases and words s/he is just learning

• Labeling: child goes around proudly naming or labeling things in her/his environment

CDI: Early Gestures Examples

• Initial Communicative Gestures like pointing, waving, putting arms up to be picked up, shaking or nodding head, reaching arm to show or give toy, shrugging shoulders as if "all gone" or "where'd it go", blowing kisses, etc

• Initial Games/Routines like peek-a-boo, chase, singing, dancing, etc.

CDI: Late Gestures Examples• Actions with Objects such as brushing teeth, putting

phone to ear, throwing a ball, pouring liquid, making a plane fly, using a spoon, etc;

• Pretending to be a Parent such as putting a stuffed animal or doll to bed, pretending to feed it, rock it, talk to it, etc;

• Imitating Adult Actions such as housework, reading, driving, shoveling, putting on glasses, etc.