Geomapping Approaches to Planning Service Systems & Improving Accessibility

1
Geomapping Approaches to Planning Service Systems
Improving Accessibility

• Illinois Department of Children Family
  Services/Northwestern University
• Dana Weiner, Ph.D.

2
GIS Approaches to Planning/Evaluation

• Statewide Provider Database
• Geomapping Tools
• Service Accessibility Study

3
Statewide Provider Database

• Online Search Tool
• Geocoded service delivery locations
• Highly detailed descriptions of programs and
  services
• Broad user base invested in helping to maintain
  current data

4
SPD Contents

• Currently contains information on
• 1,320 agencies
• 2,700 programs across the state
• Over 15,000 services
• Includes programs with and without DCFS contracts
  
• Includes data on which of state departments
  contract for services

5
SPD Contents

• Program Type
• Mental Health
• Substance Abuse
• Domestic Violence
• Parenting
• Non-clinical
• Early Childhood
• General Medical

• Target Population
• Deaf/hard-of-hearing
• Developmentally disabled
• Young Children
• Teen parents
• Sexual offenders
• Trauma survivors
• GLBTQ
• Foster care

6
Service Types

• Anger Management
• Advocacy
• Art therapy
• Case management
• Crisis intervention
• Drug testing
• Educational testing
• Family counseling
• Food/meals
• Gang awareness

• GED preparation
• Group counseling
• Health education
• HIV/AIDS counseling
• Individual counseling
• Life/Independent Living Skills
• Medication compliance
• Medication management

7
Service Types

• Neuropsychological testing
• Parenting assessment
• Parenting skills training
• Peer support group
• Services for parent/guardian
• Psychiatric evaluation
• Psychological testing

• Psychosocial assessment
• Recreational activities
• Tutoring/mentoring
• Violence prevention
• Vocational assessment/training

8
Program Information

• Intake
• Eligibility
• Geographic
• Financial
• Clinical
• demographic

• Frequency of participation
• Discharge
• Features addressing barriers to access
• Staffing patters credentials
• Use of evidence-based practices

9
(No Transcript)
10
10
11
11
12
12
13
13
14
(No Transcript)
15
(No Transcript)
16
(No Transcript)
17
17
18
(No Transcript)
19
(No Transcript)
20
(No Transcript)
21
(No Transcript)
22
(No Transcript)
23
(No Transcript)
24
(No Transcript)
25
(No Transcript)
26
(No Transcript)
27
(No Transcript)
28
(No Transcript)
29
(No Transcript)
30
Accessibility Scores Why Bother?

• A score can allow us to study the relationship
  between access and other things, such as outcomes
  like placement stability
• Scores can be used to understand gaps more
  precisely than visual maps
• Scores can be used in efforts to maximize the
  impact of contracts

31
Previous Research Does distance matter?

• Costs and disease burden increase as distance
  from providers increases (Billi et al)
• Conflicting findings on whether the utilization
  of community-based services can reduce the
  utilization of more costly inpatient mental
  health services (Curtis Fortney et alMobley)
• Proximity increases utilization (Allard)

32
Barriers to accessing services

• Attitudes cultural norms
• Perceptions of service availability
• Transportation/driving ability/license
• Language/acculturation
• Hours of availability

33
Shortcomings of prior GIS work on accessibility

• Studied in areas other than child welfare
  (veterans, teen parents, managed care
  participants)
• Access aggregated to broad areas, such as census
  tracts or zip codes
• Imprecise calculations without details on
  provider and client locations

34
Our Method

• Incorporates variation in travel impedance by
  land use type
• Uses individually-derived access scores (rather
  than those computed generally for an area and
  applied to all individuals within that area).
• Tests relationship between access and other
  predictors of placement stability for kids
  receiving wraparound services
• Uses SOC as a unique setting in which to study
  the impact of access
• Broad range of service possibilities
• Flexibility in model of service delivery

35
Model for Calculating Accessibility

• Gravity models
• 2-Step Floating Catchment Area (Wang Luo)
• Kernal Density (Guagliardo)

36
Measuring Service Accessibility

• Mental health (797) and non-clinical (366)
  provider locations from SPD (71 locations offered
  both)
• Modified gravity model
• Divided the state into 3 land use types
• Based on land use type, measured each childs
  distance to providers within a radius of
  reasonable distance
• Divided distance to each provider by a rate of
  decay P value
• Summed discounted values for each child to arrive
  at a score signifying their access

37
Reasonable Distance
Rural
Lt. Urban
10 m
Urban
5 m
2.5 m
38
Sample

• 1448 participants in SOC during 2007-2009
• 48 (n699) female
• Ages ranged from 2 to 20 with an average age of
  10.2 (SD4.6) years
• 57.4 African American, 32.3 White, 8.6
  Hispanic, .1 Asian youth and .1 Native American
  youth

39
(No Transcript)
40
(No Transcript)
41
(No Transcript)
42
(No Transcript)
43
Independent variables

• Age
• Clinical characteristics CANS scale scores
• Risk behaviors
• Emotional/behavioral needs
• Strengths
• Trauma experiences
• Trauma stress symptoms
• Acculturation
• Life domain functioning
• Service Accessibility

44
Overall Predictors of Placement Disruption
45
Predictors by Land Use
46
What does this say about access?

• Access was the only variable that exhibited a
  pattern by land use type
• Access means by land use type rural.20, light
  urban.44, and urban.75
• Access had the largest effect in rural areas,
  followed by light urban, and was not significant
  in the urban land use analysis (statistically
  significant, F (df2) 89.81, plt.001)

47
Implications

• Delivering wraparound in rural areas has
  additional challenges
• Equitable distribution of resources cannot
  necessarily be accomplished by
• distributing providers evenly over a service area
• broadening a provider's coverage
• Strategic approaches call for rural providers to
  increase
• home-based services
• transportation services

48
Limitations

• Comprehensiveness of provider data
• Single outcome measure
• Caregiver variables not included
• Data not analyzed between SOC providers

49
Future directions

• Incorporate features addressing barriers
• Relate specific needs to specific services
• Predict other outcomes
• Supplement CYCIS moves with other data for a more
  qualitative understanding of placement disruption
• Develop a threshold access score

50

• By using provider and Child access scores we can
  rationalize how and where to allocate funding
• Optimize current contracts by placing them with
  providers that children can easily reach.
• - i.e. The light green dots.
• Identifies areas where DCFS needs to recruit new
  providers, or encourage providers to relocate, in
  order to improve service access for children.
• - i.e. The dark blue dot.
• Eliminates waste by indentifying contracted
  services that may be at locations which are
  inaccessible to children.
• - i.e. The dark green squares.