Dr' Jeff St' Pierre

1
Presented by
Residential Treatment within a System of Care

• Dr. Jeff St. Pierre
• with support from a cast of hundreds

2007
2
Child and Parent Resource Institute Catchment and
Mission
Our mission is to enhance the quality of life of
children and youth with complex mental health or
developmental challenges and to assist their
families so these children can reach their full
potential.
3
How to get real world research started it takes
patience!

• History (a special thanks to the OCHS)
• CPRI a regional multi-disciplinary DH/MH
  tertiary CMH centre in the SWR
• 1990s Management (thank you Claire
  Wallace-Piccin) and Psychology at CPRI intensive
  inpatient services adopts CBCL then SCIS as a
  standard parent report of symptoms
• - MCSS now MCYS we adopt BCFPI/CAFAS provincial
  protocol
• 2002 All inpt clients were tracked with a
  standard database longitudinally for 3 years,
  with pre-post-follow-up data collection
• multiple tracking measures parent report of
  symptoms, clinician rating, teacher and parent
  scales of behaviour and social skills, parenting
  stress, service satisfaction, medications,
  academics
• 2006 This research supported by CPRI and UWO
  funded by MCYS and the Provincial Centre of
  Excellence at CHEO Christine Cullion, Dr. Alan
  Leschied, Dr. Shannon Stewart, Professor Andrew
  Johnston and Steve Cook

4
CPRI residential Sample

• began as program evaluation problem with
  missing data!
• linked with provincial initiative around BCFPI,
  CAFAS which greatly improved human resources and
  data collection
• obtained seed grant used it to hire someone to
  consistently track data which greatly improved
  data collection
• CPRI committed to keeping Ms. Cullion on after
  seed money ended, which allowed continued
  fidelity in data collection
• using the sample data, joined with UWO to get
  major funding from Centre of Excellence, which
  allowed for long term follow-up data collection
• Limitations
• heterogeneity of sample, life events, services
• effectiveness not efficacy work no control
  group no measure of messy, real world treatments
• multivariate statistical nightmare! retained
  HLM expertise

5
Adoption of BCFPI and CAFAS at CPRI

• Our intake social workers were cautious about the
  need to adopt a brief clinical screening tool,
  on clients that had already been screened at
  multiple levels as in need of specific services
  already identified in the clinical referral.
• The BCFPI seemed designed to direct clients to
  cost effective evidence based front line tools,
  such as child behaviour management courses or
  readings that would prevent ongoing difficulties.
  In contrast, we are asked to understand co-morbid
  disorder profiles and complex multiple agency
  (multiple Ministry!) case management (i.e. more
  coordination than triage).
• Our clinical staff were sceptical during CAFAS
  training that this index would have sufficient
  high end variability to serve the tertiary care
  population
• 1. Would a brief parent report of current
  symptoms add any incremental validity to in-depth
  multi-disciplinary assessments already on file
  offering co-morbid diagnostic profiles? - answer
  - ?
• 2. Equally important, and of interest in this
  presentation, would a brief parent report of
  symptoms be a good addition to the CAFAS outcome
  measurement and client satisfaction outcome
  tools?

6
Residential Treatment within a System of Care

• clients age 7-17
• 3 to 1 males to females
• referred for tertiary out-of-home care due to
  severe behaviour plus x and y and z
• comorbidity and severe family distress is the
  norm
• TX - residential cottages split by age and
  gender
• behavioural milieu staffed by child and youth
  counsellors coupled with psychiatric treatment,
  multidisciplinary assessment, individual, family
  and group therapy
• heavy psychotropic mix typically the first step
  is to reduce dosage or of medications in use
  (stimulants, anti-psychotics, anti-depressants,
  mood stabilizers, anti-anxiety)
• onsite treatment classrooms very small ratio,
  school readiness skills training focus with full
  attendance (often the first classroom success
  the child has experienced)

7
Residential Treatment within a System of Care

• prior to admission family typically has received
  psychotropic medication therapies, special
  education services, case management services,
  parent training or support
• child is typically at risk of school or home
  breakdown or has already experienced this
• post-discharge further service is expected from
  mental health, child welfare, special education,
  and possibly corrections
• at discharge Does your child require further
  mental health services 85 Yes

8
(No Transcript)
9
Pre-admission BCFPI Parent T scores
severe family distress

• Structured phone interview with parents
  pre-admission
• T scores have a mean of 50, standard deviation 10
• Scores over 70 are above the 98th percentile
• externalizing symptoms 83
• internalizing symptoms 70
• family adaptation 101
• Symptoms and quality of life impact
• This data emphasizes that we must consider child
  AND family outcomes as distinct entities in
  treatment planning and outcome tracking
  (functional outcomes)

extreme acting out behaviour
10
Brief Child and Family Phone Interview
(BCFPI) Comparison to Ontario Data
CPRI Ontario

• T gt 70
• is clinical range
• (above 98th percentile)

(N4918) Resid
Outpatient Regulation of Attention,
Impulsivity Activity 74.43
65.15 Cooperativeness 76.58 68.33 Conduct 93
.66 67.79 Externalizing 83.07 69.87 Separation
from Parents 63.86 59.39 Managing
Anxiety 59.82 58.63 Managing Mood 75.16 65.19
6 Mood 3 Self Harm Indicators 79.79 68.26 In
ternalizing 69.93 63.72 Total 6 mental health
domains 80.18 69.13 Social participation 84.49 6
9.58 Quality of Relationships 75.66 62.79 Schoo
l Participation Achiev. 79.67 63.00 Global
Functioning 86.07 68.49 Family
Activities 113.67 75.77 Family
Comfort 82.50 72.15 Global Family
Situation 101.05 77.03
11
families of teens and young children off the scale
Ontarios Province-wide aggregated Brief Child
and Family Parent Interview (BCFPI) data for
various agencies
CPRI teens Conduct
12
Family Activities Restricted

• What predicts families with young children
  utilizing out-of-home treatment? There is some
  initial evidence here that they experience
  greater personal restrictions, isolation, and
  embarrassment.
• BCFPI Fact items
• how frequently has Johns behaviour prevented
  you from taking him out shopping or visiting?
• how frequently from having friends, relatives
  or neighbours to your home?
• how frequently his brothers/sisters from having
  friends, relatives or neighbours to your home?

how frequently made you decide not to leave him
with a sitter?
13
BCFPI comparisons
Inpatients (incomplete data)

• Community partners appropriately refer those with
  great need to out-of-home treatment, according to
  parent report
• Our pre-adolescent referral base appears to
  experience very high parental distress
• There is considerable variability that we can
  analyze
• Provincial Data from Peter Pettingill at BCFPI

14
BCFPI T Scores Intake 7 months
post-discharge
15
BCFPI T scores across gt3 years in the System of
Care Intake 7 months post-disch gt 2
year follow p-d
16
Outcomes CAFAS estimate of treatment impact

• CAFAS mean level of functioning according to
  clinical case coordinator ratings
• Totals for the 8 scale sum of 149 residential
  clients at CPRI at referral (T1) then at
  discharge
• Compared to provincial average at intake, there
  is evidence that those most at risk are referred
  for residential services at CPRI
• Risk decreased and home and school functioning
  improved at discharge

at discharge, community treatment needed
17
Chronicity and Pervasiveness

• Another way to summarize this sample is looking
  at CAFAS scores - by the designation
  Pervasiveness of Impairment (Hodges, 2004).
  Pervasiveness of impairment which Hodges
  describes as cross-setting consistency of
  externalizing behaviours, is a predictor of
  treatment outcome in Kay Hodges state-wide
  research efforts. She reports that children and
  youth scoring as moderate (20) or severe (30)
  impairment across all three subscales of school,
  home, and behaviour toward others, represent
  roughly 30 of her clinical referral sample in
  Michigan.
• Here, using Hodges pervasiveness subtype, 76
  of our mental health inpatient referrals met this
  criterion.

18
N70
19
Outcomes Conners Teacher Rating Scale 6 months
post

• T scores
• These may be different teachers, different class
  settings reporting before and after treatment
• These scores need to be matched to parent report
  to examine both settings

clinical cut-off
mean
20
Outcomes Teacher SSRQ Standard Scorespre,
during, 6 months post

• Multiple teacher, parent and staff ratings of
  social skills are collected, and all note
  significant deficits on average
• These children and youth have significant social
  skills handicaps in assertiveness,
  responsibility, cooperation
• Scores approach the handicapped range (70) at
  referral, CPRI classroom teachers report gains
  during treatment, some maintenance of these skill
  gains reported by teachers after 6 months in
  community school

mean
plt.001
Mean 100, s.d. 15
21
A System of Care

• Compared to the limited literature available,
  these appear to be good treatment effect sizes
  within a system of care follow-up in a
  high-service-user tertiary sample of children and
  youth with extreme co-morbid symptomatology,
  significant family stressors and life skill
  deficits
• This impact is a result of a mix of outpatient
  and inpatient mental health, education, and child
  welfare services
• Client satisfaction ratings are positive
• Multi-rater context by collecting data beyond
  the provincial dataset we can speak to
  informant specific outcomes (e.g., school
  versus home), and clarify service delivery
  variability across cases and systems/regions
• The means reported are messy no data cleaning
  our goal is to use HLM to track predictors of
  success and failure
• Questions to Dr. Jeff St. Pierre 519-858-CPRI
  ext. 2021

22
Answering questions from our front line staff

• we can use this data collection to answer
  internal clinical questions
• For example

23
CYWs It seems the clients have more problems
than the old days

• Psychologists at CPRI collected mental health
  residential data (primarily C19, Genesis, Boys
  unit) back in 1989-91, 1995-96 using the Child
  Behaviour Checklist (CBCL)
• We can compare this similar parent report measure
  to the present study which uses the BCFPI
• Because they are not identical measures, it is
  difficult to answer clearly, but it looks like
  yes our single-point-of-access intake does send
  us clients reporting more behaviour problems and
  greater parenting stress

24
A COMPARISON OF RESIDENTIAL REFERRAL DATA OVER
THE YEARS AT CPRI Mean scores on parent report
measures
25

• For these treatment effect size charts, small Ns
  in the earlier two evaluations may indicate a
  biased sample. The 2006 question set was similar
  but not identical to the earlier evaluations. The
  residential treatment programs at CPRI changed
  over these years in several ways.

26
(No Transcript)