E V I D E N C E F R O M T H E N A T I O N A L S U R V E Y O F S S I C H I L D R E N A N D F A M I L I E S

SSI Children with Speechand Language Disorders

Presentation to the IOM Committee to Evaluate the SSI Disability Program for Children with Speech and Language Disorders

Kalman Rupp, Paul S. Davies, and Jeffrey Hemmeter

Social Security Administration

May 18, 2015

Two Data Sources

SSA program data Background data on award cohorts of SSI children from 1985 to

2000

Large sample and can observe program entry patterns over time (flow), but…

…limited information on recipient characteristics and virtually no information on family situation

National Survey of SSI Children and Families (NSCF) Primary focus of this presentation

Rich survey data on child and family characteristics at a single point in time (stock), but…

…sample size limits subgroup analysis and data are nearly 15 years old

SSA Data on Speech and Language Delay

SSA diagnosis codes 3153 (Speech and Language Delays) 7840 (Communication Impairment)

Award cohorts of SSI children (age 0-17 at award) 1985, 1990, 1995, and 2000

At the time of award, SSI children with speech and language delay (SLD) are: More frequently male than other SSI children Younger than other SSI children

Number and percentage of new awards with SLD growing over time (see more recent data presented by Melissa Spencer on 1/26/15)

Gender (% male) Age Group

SLD Gender and Age Distribution by Award Cohort

0

10

20

30

40

50

60

70

80

90

100

1985 1990 1995 2000

% o

f A

war

ds

Award Year

0

10

20

30

40

50

60

70

80

90

100

1985 1990 1995 2000

% o

f A

war

ds

Award Year

0-5

6-12

13-17

Number of new awardsPercentage distribution of new awards

Diagnosis Codes by Award Cohort

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

1985 1990 1995 2000

Year of Award

Other Identified

Physical

ID/MR

Other & Unknown -nonSLD

Other Mental - nonSLD

Other & Unknown -SLD

Other Mental - SLD

Award Year

Diagnosis 1985 1990 1995 2000

Other Mental - SLD 90 142 4,423 9,109

Other & Unknown - SLD 31 775 1,728 2,628

Other Mental - nonSLD 3,081 10,551 53,433 49,394

Other & Unknown - nonSLD 3,292 5,409 23,914 22,667

ID/MR 18,156 31,339 53,584 27,529

Physical 16,428 21,298 28,131 25,212

Other Identified 4,039 6,939 8,345 7,530

N 45,116 76,453 173,559 144,068

National Survey of SSI Children and Families

Conducted by SSA in 2001/2002 through contract with Mathematica Policy Research

Goal was to answer a broad range of research and policy questions about SSI children and their families

Sample frame included SSI children (0-17) and young adults (18-23) in three groups: Receiving payments in December 2000

Receiving payments in December 1996

Former recipients and denied applicants since 1992

NSCF and SLD

We focus on the NSCF sample of children aged 0-17 receiving SSI payments in December 2000 (n=3,651)

About 5% of the sample was identified as having SLD based on SSA diagnosis code 3153 (3.7%, n=135) or SSA diagnosis code 7840 (1.2%, n=44) Applying sample weights, 8.7% of the respondents have SLD

We draw comparisons between SLD and non-SLD children in the NSCF for several key characteristics that cannot be measured in SSA program data, highlighting notable and statistically significant differences between the two groups

Context and Interpretation

NSCF sample is representative of SSI recipients, but quite different from general population SSI means test → low income Disabilities of child meeting SSA screen Parent knowledge and motivation to apply

The average child in our sample has been on SSI for several years

Both SLD and non-SLD recipients are affected by dual challenges of poverty and disabilities One subgroup may have lower poverty rate than the other, but both may have high poverty

rates compared to general population SLD recipients may have higher or lower service needs compared to non-SLD recipients, but

both may have higher service needs compared to children of comparable ages who do not receive SSI

Results are descriptive, do not include regression-adjusted comparisons

Topics

Demographic characteristics of children

Disability and health status of children

Functional limitations

Health insurance and medical utilization

Out of pocket expenses

Services received and unmet needs

School-based services

Characteristics of parents and families

Financial well-being

Care and parental work

Demographics Characteristics of Children

Major difference is that SLD children tend to be relatively younger

SLD children more likely to be boys compared to non-SLD (majority is boys in both groups)

Race and ethnicity distributions are quite similar, but:

• Significantly lower proportion of whites in SLD

• Marginally significant overrepresentation of Hispanic ethnicity in SLD

0% 20% 40% 60% 80% 100%

Male***

Age group

0-5

6-12***

13-17***

Race

White**

Black

Hispanic*

SLD

non-SLD

Award Cohort vs. Cross Section

Characteristics differ depending on perspective: awardee cohort versus prevalence in cross-section of “current” SSI recipients

Awardee cohort perspective presented previously gave age, gender, and diagnosis distribution at first-ever SSI award

NSCF analysis reflects characteristics of December 2000 cross-section of SSI recipients as of the 2001/2002 NSCF interview

Possible sources of differences include differential exit patterns due to mortality and transition to nonrecipient status

Results are consistent – median age for SLD recipients is lower and higher proportion of SLD recipients are boys in both the award data and the cross-section data

Disability and Health Status of Children

Overall distributions for general health status, severity, and ability to do things are remarkably similar based on respondent perceptions (mainly parents)

Differences are not statistically significant and difficult to interpret

0% 20% 40% 60% 80% 100%

General Health Status

Poor

Fair

Good

Very good

Excellent

Missing

Severity of Health Condition

High

Moderate

Mild

No health condition reported

Missing

Disability affects ability to do things

A great deal

Some

Very little

No disability reported

SLD

non-SLD

Functional Limitations

SLD children report significantly less need for help with physical functioning, as might be expected

NSCF is lacking in measurement of social, cognitive, and psychological functioning

0% 20% 40% 60% 80% 100%

Needs help--

Getting in or out of bed****

Getting around inside home***

Using or getting to toilet***

Eating***

Bathing or showering*

Dressing

SLD

non-SLD

Health Insurance and Medical Utilization

Due to Medicaid and SCHIP both SLD and non-SLD children are essentially fully insured

Insurance coverage concerns among those losing SSI in adulthood are expected to be reduced under ACA due to Medicaid expansion and availability of subsidized insurance

A relatively lower proportion of SLD children have 4+ annual doctors’ visits and have been hospitalized during previous year

0% 20% 40% 60% 80% 100%

Any health insurance

Medicaid

SCHIP

Private health insurance

Doctor Visits

0

1

2

3

4**

5 or more**

Missing

Hospital Visits

0***

1***

2

3 or more

Missing

SLD

non-SLD

Out of Pocket (OOP) Expenses

Despite similarities in health insurance coverage, a smaller fraction of SLD children have any out of pocket (OOP) expenses for medical care, services, or therapies

The distributions of the dollar values of OOP expenses (over the 12 months prior to the interview) are similar for SLD and non-SLD children

The mean amount of OOP expenses (among those with positive values) was $423 for SLD children and $862 for non-SLD children

0% 20% 40% 60% 80% 100%

Any OOP expenses**

Medical OOP expenses ($)

None

1-99

100-199

200-499

500-999

1,000 or more

Servcies/Therapies OOP expenses ($)

0

1-99

100-199

200-499

500-999

1,000 or more

SLD

non-SLD

Services Received and Unmet Needs

Significantly higher utilization of physical, occupational, speech services and audiology services among SLD children

Also significantly higher unmet need for physical, occupational, and speech services for SLD children

Unmet need for mental health counseling significantly lower among SLD children

0% 20% 40% 60% 80% 100%

Services received

Physical, occupational, speech***

Respiratory**

Recreational***

Audiology**

Transportation

Respite care***

Mental health counseling

Unmet needs (among not receiving)

Physical, occupational, speech*

Respiratory

Recreational

Audiology

Transportation

Respite care

Mental health counseling**

SLD

non-SLD

School-Based Services

Vast majority ages 3-17 are in school in both groups

Percent of 3-5 years old in school is 94% for SLD children but only 88% for non-SLD children

Statistically significantly higher proportion of SLD children in school in 13-17 age group – relative odds of dropping out lower

For ages 3-17 combined, the majority of both groups has had experience with Special Ed and IEPs

0% 20% 40% 60% 80% 100%

In School

All age 3-17***

Age 3-5

Age 13-17*

Special Ed Ever (age 3-17)

IEP Ever (age 3-17)

SLD

non-SLD

Characteristics of parents and families

Both SLD and non-SLD recipients are clearly disadvantaged: single parent, several children, parents with low education, about half live in household with at least one other person with disabilities

A notable difference is that significantly smaller proportion of SLD children live with both parents – only about 21 percent – and a larger fraction live with mom only

0% 20% 40% 60% 80% 100%

Living Arrangements

Two Parents**

Mom***

Dad*

Other Older Relative

Other

Number of Children in HH

1

2

3

4

5+

Missing

Other HH Member with Disability

Mom Education (if present)

Less than HS

HS or GED

More than HS

Missing

Dad Education (if present)

Less than HS

HS or GED

More than HS

Missing

SLD

non-SLD

Financial Well-Being

Overall SSI children live in families with relatively high poverty rate – significantly higher for families of SLD children

SSI provides roughly half of average family income

Even though parents have relatively weak labor force attachment and low earnings capacity, many are not disabled and earnings is a substantial source of family income for many

Food Stamps and housing assistance are significantly more prevalent among families of SLD children

0% 20% 40% 60% 80% 100%

Poverty Rate**

Total Family Income< 500

500-9991,000-1,4991,500-1,9992,000-2,999

3,000 or more

Mom Earnings (if present)0

1-9991,000-1,999

2,000 or moreMissing

Dad Earnings (if present)0

1-9991,000-1,999

2,000 or moreMissing

Percent of family income fromSSI

Earnings

Percent receivingSocial Security

Other Public AssistanceFood Stamps**

Housing Assistance**Energy Assistance

SLD

non-SLD

Care and Parental Work

Survey asked about caregiving hours specifically related to child’s disabilities

Significantly lower proportion of SLD children received care from family members

66% of SLD children reportedly did not receive any caregiving hours – family or nonfamily

Significantly more caregiving was provided for non-SLD children

Most of the care was provided by family members for both groups

About half of mothers reported no market work for both groups, but 20% worked full time

0% 20% 40% 60% 80% 100%

Family Provides Care***

Total Hours of Care

0***

1-19

20-34**

35 or more***

Missing

Mom Weekly Work Hours (if present)

0

1-20

21-35

36 or more

Missing

SLD

non-SLD

Conclusions – Child Characteristics

Percent of SSI awardees with SLD dramatically increased between 1985 and 2000 Consistent with increase post-2000, but rate of increase declined post-2000 Percent classified as ID/MR substantially decreased Change in diagnostic and/or classification practices ?

Overall SSI children with SLD tend to have characteristics that do not dramatically differ from non-SLD children on SSI on variables that were available for the analysis Most importantly, overall distributions on responses to general health and severity questions are remarkably

similar for SLD and non-SLD children and do not show statistically significant differences SLD children appear significantly less limited on functional limitations measures from the NSCF, but NSCF

functional limitations measures exclusively focus on physical disabilities To get balanced and unbiased picture, more information on severe limitations in social functioning, cognitive and psychological variables would be extremely helpful

Although NSCF contains no comparable data for all (non-SSI) children, both SLD and non-SLD children show high proportion of minorities, high utilization of school-based services, virtually complete health insurance coverage. Both SLD and non-SLD children had modest reported OOP expenses, but small minority in both groups reported OOP expenses that are relatively high compared to SSI income guarantee (Federal Benefit Rate)

Some notable differences. SLD children tend to be younger, have less doctors visits and hospitalizations, received more

physical/occupational/speech services and audiology services

Conclusions – Family Situation

Generally, the family situation of both SLD and non-SLD children on SSI indicates high degree of financial vulnerability Prevalence of poverty is high despite SSI cash benefits providing a large portion of family

income Family characteristics typically associated with poverty are common for both groups. These

include: large proportion of single parent families, high prevalence of low educational attainment of parents, low employment levels, part time work and low wages

About half of SLD children and non-SLD children on SSI live in households with at least one other person with a disability Added source of stress and financial vulnerability Increases caregiving needs

Many report family and nonfamily caregiving hours specifically related to health care needs of both SLD and non-SLD children on SSI However, substantially more report no family or nonfamily caregiving hours for children

with SLD

Classification of Childhood Population bySSI Eligibility and SLD Diagnosis

Usefulness of NSCF and other data sets to address SSI and SLD issues depends on questions of interest

Questions of interest determine the comparison groups that are necessary and relevant

Sample coverage is key

Financially eligible

for SSI under means

test?

Serverely disabled according to SSA criteria?

Yes, SLD Yes, non-SLD Not severe, but

some disablement

Neither severe nor

less severe

disablement

YES

SSI (in NSCF)

severe disablement

+

financially

vulnerable

SSI (in NSCF)

severe disablement

+

financially

vulnerable

less severe

disablement

SLD and/or non-SLD

+

financially

vulnerable

no disablement

+

financially

vulnerable

NO

SLD

severe disablement

+

financially

better off

non-SLD

severe disablement

+

financially

better off

less severe

disablement

SLD and/or non-SLD

+

financially

better off

no disablement

+

financially

better off

NSCF in Context

NSCF is useful to answer questions where within-SSI comparisons are relevant, but broader questions require external standards and comparison with non-SSI populations

Looking to future: Need for more current data

Need for future data collection strategies that integrate SSI and SLD issues in an overall sample framework of all children in the United States Need for sufficient SSI and SLD sample sizes

Accurate identification of SSI receipt is crucial (administrative data matches)

Use variables that are relevant for assessing nature and severity of disabling conditions for both SLD and non-SLD subgroups of children

Assessment of longer-term outcomes over the childhood, adolescent and adult portions of the life-cycle

Contact Information

Kalman RuppSenior Economist, Office of Policy Evaluation and ModelingOffice of Research, Evaluation, and StatisticsKalman.Rupp@ssa.gov

Paul S. DaviesSenior AdvisorOffice of Data Exchange and Policy PublicationsPaul.Davies@ssa.gov

Jeffrey HemmeterDeputy Director, Office of Program DevelopmentOffice of Research, Demonstration, and Employment SupportJeffrey.Hemmeter@ssa.gov

References

Davies, Paul S. and Kalman Rupp (2005/2006), “An Overview of the National Survey of SSI Children and Families and Related Products,” Social Security Bulletin 66(2): 7-20.

Kalman Rupp, Paul S. Davies, Chad Newcomb, Howard Iams, Carrie Becker, Shanti Mulpuru, Stephen Ressler, Kathleen Romig, and Baylor Miller (2005/2006), “A Profile of Children with Disabilities Receiving SSI: Highlights from the National Survey of SSI Children and Families,” Social Security Bulletin 66(2): 21-48.

Rupp, Kalman, Jeffrey Hemmeter, and Paul S. Davies (2015), “Longitudinal Patterns of Disability Program Participation and Mortality Across Childhood SSI Award Cohorts,” Social Security Bulletin 75(1): 35-64.

Davies, Paul S., Kalman Rupp, and David Wittenburg (2009), “A Life-Cycle Perspective on the Transition to Adulthood among Children Receiving Supplemental Security Income Payments,” Journal of Vocational Rehabilitation – Special Issue on SSI Youth in Transition 30(3): 133-151.

Rupp, Kalman and Steve Ressler (2009), “Family Caregiving and Employment among Parents of Children with Disabilities on SSI,” Journal of Vocational Rehabilitation – Special Issue on SSI Youth in Transition 30(3): 153-175.

DeCesaro, Anne and Jeffrey Hemmeter (2009), “Unmet Health Care Needs and Medical Out-of-Pocket Expenses of SSI Children,” Journal of Vocational Rehabilitation – Special Issue on SSI Youth in Transition 30(3): 177-199.

Aron, Laudan Y. and Pamela J. Loprest. Meeting the Needs of Children with Disabilities. Washington, DC: The Urban Institute Press, 2007.