Pinellas Data Collaborative

1
Pinellas Data Collaborative
Cross-system Utilization using Three Years of
Data Paul Stiles, J.D., Ph.D. Diane Haynes,
M.A. 813) 974-9349 voice
[phone] voice
stiles_at_fmhi.usf.edu
haynes_at_fmhi.usf.edu Policy Services
Research Data Center Department of Mental Health
Law Policy Louis de la Parte Florida Mental
Health Institute University of South
Florida 13301 Bruce B. Downs Blvd. Tampa, FL 33612
2
PDC Background

• Established under F.S. 163
• Several organizations in county participating
  including
• BOCC -- DJJ
• DSS -- JWB
• Courts/Sheriff -- DCF
• EMS --FMHI (repository)

3
Data Sets

• Criminal Justice CJIS
• Dept of Social Services DSS
• DCF/ADM IDS
• Medicaid AHCA

• Juvenile Welfare Board JWB
• Child Welfare CW
• Emergency Medical Services EMS
• Baker Act BA

4
On-going Questions/Analyses

• Use and cost of Acute Care (Baker Act) services
  in the county
• Use of Acute Care (Baker Act) by children
• Overlap of identified population (PEMHS) in JWB
  that are in Medicaid
• Network Analysis on Individuals Dealing with
  Substance Abuse
• Continuity of Care Study (GPW Closing)
• How many kids being served by JWB are also
  involved in the Child Welfare system?

5
Cross Systems UtilizationInitial Questions

• What is the measure/degree to which adults and
  children in the 8 systems have caseload overlap
  for over a three year period (1998-2001)?
• What is the measure/degree to which high users in
  the systems have caseload overlap over a three
  year period (1998-2001)?

6
Overview

• The Eight Systems
• The Statistical Method used in this study
• Findings

7
Eight Systems Included in this Study

• Criminal Justice CJIS
• Dept of Social Services DSS
• DCF/ADM IDS
• Medicaid AHCA

• Juvenile Welfare Board JWB
• Child Welfare CW
• Emergency Medical Services EMS
• Baker Act BA

8
Overlap within each system across years - All
Between 1998 and 1999
Between 1999 and 2000
Between 1998 and 2000

• CJIS 35
• DSS 34
• EMS 24
• IDS 58
• MMH 77
• JWB 19
• CW 24
• BA 21

• CJIS 35
• DSS 36
• EMS 24
• IDS 51
• MMH 79
• JWB 15
• CW 15
• BA 15

• CJIS 25
• DSS 21
• EMS 19
• IDS 44
• MMH 64
• JWB 34
• CW 13
• BA 13

9
Overlap within each system across years - HH
Between 1998 and 1999
Between 1999 and 2000
Between 1998 and 2000

• CJIS 28
• DSS 33
• EMS 18
• IDS 55
• MMH 56
• JWB 8
• CW 65
• BA 38

• CJIS 26
• DSS 29
• EMS 19
• IDS 41
• MMH 57
• JWB 3
• CW 58
• BA 18

• CJIS 14
• DSS 13
• EMS 11
• IDS 21
• MMH 40
• JWB 0
• CW 53
• BA 13

10
Statistical Method

• Probabilistic Population Estimation (PPE)
• Caseload Segregation/Integration Ratio (C-SIR)
• This process relies on information in existing
  databases and the agencies do not have to share
  unique person identifiers. It avoids the expense
  of case-by-case matching and sensitive issues of
  client-patient confidentiality.

11
Probabilistic Population Estimation (PPE)

• A statistical method for determining the number
  of people represented in a data set that does not
  contain a unique identifier. The estimation is
  based on a comparison of information on the
  distribution of Date of Birth and Gender in the
  general population with the distribution of Date
  of Birth and Gender observed in the data sets.
• The number of distinct birthday/gender
  combinations that occurred in each data subset
  are counted. The number of people necessary to
  produce the observed number of birthday/gender
  combinations are then calculated.

12
Caseload Segregation/Integration Ratio (C-SIR)

• C-SIR
• C-SIR is a rating between 0 and 100 which
  indicates
• the amount of overlap of clients between
  agencies.
• Zero being no overlap at all and 100 being total
• overlap.

Duplicated Count Unduplicated Count
Duplicated Count Largest Undup. Count
?

100
- 1
- 1
13
One year Overlap/C-Sir (44)
IDS MMH 7,447  
3,996 3,131    
Unique ID Count PPE Count
Population Cross MMH 7,104 7,127

56.06 IDS 11,640 11,443
34.92
14
Findings from last years analysis

• There is very little overlap in users between the
  systems that were looked at.
• The caseload integration/segregation rating in
  this study varied from 5 to 44 on a scale of 0 to
  100. The greatest overlap is between IDS and MMH,
  the mental health systems
• It is the non-high users that are more likely to
  cross multiple systems, not the high users. If an
  individual is a high user in one system, they
  probably are not in the other systems.

15
Three Year Overlap/C-Sir-AdultsTotal Population
16
Three Year Overlap/C-Sir-KidsTotal Population
17
Three Year Overlap/C-Sir-AdultsHeavy Hitters
18
Three Year Overlap/C-Sir-KidsHeavy Hitters
19
Overlap/C-Sir JWB CW
JWB Child
Welfare 48,639
23,720 17,563
20
Overlap/C-Sir EMS IDS/MMH/BA (All Age Groups)
21
We need your help!!

• Help with what systems to cross
• Help with predicted pathways (so that we can look
  at service use patterns over time)
• Help selecting other systems to include (e.g.,
  Education?)
• Help with interpretation

22
Demographics (ALL-Adults)
23
Demographics (ALL-Kids)
24
Reference
  Banks, S. Pandiani, J. (1998). The use of
state and general hospitals for inpatient
psychiatric care. American Journal of Public
Health, 99(3), 448-451.     Banks, S., Pandiani,
Gauvin, L, Readon, M.E., Schacht, L.,
Zovistoski, A. (1998). Practice patterns and
hospitalization rates. Administration and Policy
in Mental Health, 26(1), 33-44.   Banks, S,
Pandiani, J. James, B (1999). Caseload
segregation/integration A measure of shared
responsibility for children adolescents.
Journal of Emotional Behavioral Disorders,
7(2), p 66-17.   Banks, S, Pandiani, J., Bagdon,
W., Schacht, L. (1999). Causes and Consequences
of Caseload Segregation/Integration. 12th
Annual Research Conference (1999) Proceedings,
Research and Training Center for Childrens
Mental Health.   Pandiani, J., Banks, S.,
Gauvin, L. (1997). A global measure of access to
mental health services for a managed care
environment. The Journal of Mental Health
Administration, 24(3), 268-277.