1. Availability of Outpatient Rehabilitation Services for Children after Traumatic Brain Injury: Differences by language and insurance status
Megan Moore, PhD, MSW Sidney Miller Endowed Assistant Professor in Direct Practice University of Washington WASCLA 2016

2. Nathalia Jimenez, Ali Rowhani-Rahbar, Margaret Willis, Kate Baron, Jessica Giordano, Deborah Crawley, Fred Rivara, Ken Jaffe and Beth Ebel

3. Traumatic Brain Injury in the US
A TBI is caused by
bump, blow, or jolt or a penetrating head injury that disrupts the normal function of the brain
Severity ranges from “mild” to “severe”
Most TBIs are mild
Not all blows or jolts to the head result in a TBI.
CDC 2015

4. 2.5 million people each year seek medical care
Traumatic Brain Injury in the US
2.5 million people each year seek medical care
1/3 of all injury-related deaths caused by TBI
339,462 DoD TBI Worldwide Numbers since
$77 billion indirect and direct medical costs
Leading cause of death and disability in US
UCSF/CDC

5. Causes

6. TBI by Age In 2010 0-4: 58 per 100,000 5-14: 23 per 100,000

7. TBI by Race
African American persons have higher rates of ED visits, hospitalizations and death in most age groups
American Indian/Alaska Native males have highest rate of TBI-related death
CDC 2015

8. TBI Outcomes Cognitive deficits Impaired physical functioning
Mental health symptoms and behavioral problems
Headache, fatigue, sleep disturbance, irritability
Cognitive deficits: recall, information processing speed, attention, executive functioning
NINDS; CDC 2015

9. Availability of Outpatient Rehabilitation Services for Children
Aims:
Explore associations between English proficiency, insurance status and
outpatient rehab availability and travel time to
nearest service
Access to rehab has been associated with better outcomes

10. METHODS
Created a comprehensive statewide database of pediatric providers, services and geographic locations in WA state

11. METHODS Assessed services available to All children
Those with Medicaid insurance
Those needing language interpretation services

12. METHODS
Assessed association between language need and availability of rehab by county
Availability of multilingual services by county characteristics
Defined language need as % of persons older than 5 yrs speaking a language other than English in home (American Community Survey)

13. METHODS
Travel times compared for children with English and Spanish speaking parents
Identified closest provider who met child’s need
We adjusted for child’s age, child’s sex, education of parent, household income and Medicaid status. Household income and Medicaid status were significantly correlated (r=.42, p<0.001). Because our main variable of interest was insurance status, we removed household income from the model and presented those results in Table 4 (R-squared mental health=0.075; PT/OT=0.114; Speech/Language/Cognitive Therapy=0.100). When we included household income in place of insurance status in a sensitivity analysis, results were unchanged (R-squared mental health=0.065; PT/OT=0.127; Speech/Language/Cognitive Therapy=0.119). We also assessed the interaction between Spanish speaking parent and Medicaid status, but the interaction was not statistically significant.

14. RESULTS Provider and Service Availability Mental Health, n (%) 210
All services accepting children
Accepts Children & Medicaid
Accept Children, with Language Services
Accept Children Medicaid & Language Services
Mental Health, n (%)
210
(100) 77
(37) 19
(9) 16
(8)
293 service providers
385 services
In each category there were fewer services for children whose families needed languge interpretations
Mental health services represented more than half of all services available to children, however, only 8% were available to children with Medicaid and language service needs

15. RESULTS Provider and Service Availability
All services accepting children
Accepts Children & Medicaid
Accept Children, with Language Services
Accept Children Medicaid & Language Services
Physical and occupational therapy, n (%) 92
(100) 64
(70) 38
(41)
293 service providers
385 services
In each category there were fewer services for children whose families needed languge interpretations
Mental health services represented more than half of all services available to children, however, only 8% were available to children with Medicaid and language service needs

16. RESULTS Provider and Service Availability
All services accepting children
Accepts Children & Medicaid
Accept Children, with Language Services
Accept Children Medicaid & Language Services
Speech / language / cognitive therapy, n (%) 83
(100) 66
(80) 43
(52) 40
(48)
293 service providers
385 services
In each category there were fewer services for children whose families needed languge interpretations
Mental health services represented more than half of all services available to children, however, only 8% were available to children with Medicaid and language service needs

17. RESULTS Provider and Service Availability
All services accepting children
Accepts Children & Medicaid
Accept Children, with Language Services
Accept Children Medicaid & Language Services
Total pediatric rehabilitatio n Services, n (%)
385
(100)
207
(54)
100
(26) 94
(24)
293 service providers
385 services
In each category there were fewer services for children whose families needed languge interpretations
Mental health services represented more than half of all services available to children, however, only 8% were available to children with Medicaid and language service needs

18. RESULTS Provider and Service Availability
Less than 20% of service providers accepted children with Medicaid and provided language services

19. RESULTS
Availability of multilingual services by county characteristics
Table 2. Univariate associations between measured county-level characteristics and number of multilingual pediatric services
Characteristic PR
95% CI
Percentage of persons using a language other than English in the home (per 10% difference)
0.60
0.43, 0.90
Median household income (per $1,000 difference)
0.95
0.93, 0.97
Percentage of persons living below the poverty level (per 1% difference)
1.00
0.90, 1.11
Percentage of persons with educational attainment of high school or higher (per 1% difference)
1.05
0.99, 1.12
Region of state
East
0.86
0.17, 4.24
Northwest
0.42
0.11, 1.54
Central
Referent
South
0.46
0.12, 1.72
PR: prevalence ratio; CI: confidence interval
for every 10% increase in percentage of persons speaking a language other than English at home, there was a 40% decrease in the number of multilingual services in the county
The only covariate of interest significantly associated with the number of multilingual services was median household income. After adjusting for median household income, the proportion of persons with LEP in each county remained significantly associated with the number of multilingual pediatric services (PR = 0.66; 95% CI: 0.48, 0.90; per 10% difference in percentage of persons with LEP). For every 10% increase in persons speaking a language other than English at home, there was a 34% decrease in availability of multilingual services

20. RESULTS Travel Time to Nearest Rehab Service
For children with Spanish-speaking parents
Travel time was significantly longer across all service types
+16.4 min to mental health
+9.2 min to physical/occupational therapy
+9.4 min to speech/language/cognitive therapy
Average travel time to the closest rehabilitation provider was significantly longer across all service types for children in the cohort with Spanish-speaking parents relative to the children with English speaking parents (Table 4).
Adjusting for age, sex, parental education and Medicaid insurance status, Spanish speaking parents needed to drive an average of 16.4 (CI: 2.0, 30.9) additional minutes to reach the nearest mental health service provider, and had to drive 9.2 (CI: 3.2, 15.2) additional minutes to reach physical, occupational therapy and 9.4 (CI: 3.3, 15.5) additional minutes to speech/language or cognitive therapy.

21. LIMITATIONS Information from providers was self-reported
Sample of children for travel time analysis was small
Likely residual confounders not accounted for in county level analysis
Conducted in WA state
Further research is needed to expand upon results

22. KEY POINTS
<20% of providers accepted Medicaid and provided language interpretation
Mental health services most limited
Multilingual services lowest in counties with greater language diversity
Spanish-speaking families had longer travel times

23. KEY POINTS Access to rehab is associated with better outcomes
Limited availability of services at least partially explains noted disparities in outcomes
Need:
Research in other states
Evidence-based standardization of care transitions
Policies to train and incentivize providers to serve children with Medicaid/interpretation need

24. Partnership between: WASCLA, UW, WSU, HMC
ONGOING COLLABORATION AND FUTURE DIRECTIONS
Addressing Language Needs to Improve the Health Care Access and Utilization by Latinos in Washington (WA) State
Partnership between: WASCLA, UW, WSU, HMC
Funded by UW Latino Center for Health
Contact: