Carolyn S. Perchuk RN, MN

1
Families First School Readiness

• Carolyn S. Perchuk RN, MN
• Winnipeg Regional Health Authority
• University of Manitoba

www.gov.mb.ca/.../familiesfirst/evaluation.html
2
Background

• Family First Program
• Home visiting program in Public Heath
• 1999
• Health Child Manitoba
• Research based
• The purpose of the program is to decrease
  child maltreatment

ReferenceGreat Kids Inc.(2004)
3
Evaluation of Families First Program ...

• Parenting
• Increased positive parenting (ES 0.81)
• Decreased hostile parenting (ES - 0.53)

Reference Healthy Child Manitoba (2010)
4
NOT Evaluated Families First effect on School
Readiness.
5

• Research Question
• Is there a relationship between families
  participating in the Families First home visiting
  program and an increase in their childs school
  readiness on entering kindergarten as assessed by
  the EDI?

6

• Education is a social determinant of health

7
Evidence academic ability in K predictive of
long term
school readiness grade 3 success
complete grade 9 graduate grade 12
8
Identify family Risk factors
Intervene to build skill and improve
environment
Improve parent child attachment
Improve Outcomes
9
The Manitoba Families First Program

• Nurses and para professionals
• Working together
• Prenatal and postpartum
• Universal screen
• In-depth parent survey or fsc
• Family centered
• Curriculum
• Voluntary
• 3 years

10
Methodology
11

• Quantitative research design
• Secondary data analysis of data from Healthy
  Child Manitoba Office (HCMO) data base

12
Measurement Tools
13
Screening Tools' Sensitivity and Specificity
Specificity
Sensitivity
Children not in Care
Children in Care
83 scored  not at risk  On Families
First Screen
77 scored  at risk  on Screen
http//www.umanitoba.ca/centres/mchp/report.htm
14
Family Stress Checklist

• Parent's childhood experience
• Lifestyle behaviours and mental health
• Parenting experience
• Coping skills and support systems
• Current stresses
• Anger management skills
• Expectations of infant's development,
• milestones, and behaviours
• Plans for discipline
• Perception of new infant
• Bonding and attachment

ReferenceGreat Kids Inc.(2004)
www.ecdip.org/earlylanguage/
15
2003/2004
Program Group
Control Group
Positive screen 3 Positive FSC 25 Was
enrolled in the FF program
Positive screen 3 Positive FSC 25 Receive NO
program
No program Due to not Enough resources
No program Due to Refused Services
16
The Early Development Instrument
Five Domains

1. Physical health and wellbeing
2. Social Competence
3. Emotional Maturity
4. Language and Cognitive Development
5. Communication Skills and General Knowledge

Score in each domain 0-10
17
Data accessed through data sharing agreement with
the MB government SPSS software used for data
analysis (alpha .05)
18
Imputation For Missing Data

• Used Sequential regression multiple imputation
  (SRMI)where other variables are used as
  predictors for missing values
• Multiple imputations (10 cycles) as accounts for
  statistical uncertainty in the imputations-
  cycles improve outcome variables

19

• T test to Assess for Homogeneity of 2 Groups
• - Childs age at time of EDI
• FSC score
• Age of mother at the birth of the child
• Last two statistical difference but not clinical

20
Chi square to assess for Homogeneity of the 2
Groups

• screened prenatally
• low education level of mother
• on social assistance/financial difficulty
• mothers history of depression
• history of abuse as a child for mother or father
  of baby
• no prenatal care before 6 months
• family screened prenatally
• lone parent family
• teen parent
• alcohol or drug use of mom during pregnancy
• current substance use by mother
• social isolation
• violence between parents.

21
Results of Chi Square

• Important to consider not significantly
  different
• Teen mother
• Lone parent
• Low education mom
• Social isolation
• On social assistance/financial difficulties
• Depression of mom
• Violence between parents

22
Results of Chi Square

• Significant differences
• -no prenatal care (1611)
• -alcohol use by mother in pregnancy-higher
  control (4836)
• -drug use by mom during pregnancy-higher control
  (2517)
• -mother has history of child abuse- more in
  control (4129)
• -father has history of child abuse- more in
  program (1227)

23
Multiple Linear Regression Effect Independent
Variables
Y a (b1)(x1) (b2)(x2) (b3)(x3) (b4)(x4)
Independent Variables childs gender age of
child at EDI screened prenatally maternal age
alcohol use by mother in pregnancy drug use by
mother during pregnancy teen parent low
education mother lone parent on social
assistance/financial difficulty no prenatal care
before 6 mos mother hx depression current
substance use by mother social isolation
violence between parents mother has hx of being
abused as child father has hx of being abused as
a child fsc score in families first program.
FF Program
No FF Program

One model for each EDI domain
24
Results of Data AnalysisMultiple Regression
AnalysisGender and childs month of birth at EDI
significant for all

• DOMAIN
• Physical Health and Wellbeing
• -In Families First p.057
• -low education mother p.005
• -alcohol use by mother p.043

25
Implications

• Physical Health and Wellbeing
• In Families First (FF) marginal significance
• Answer research question that there is only a
  threshold relationship between FF and one domain
  of the EDI school readiness tool

26
DomainSocial Competence

• Low education of mother p.002
• Social Isolation/lack supports p.030
• Mothers history of child abuse p.009

27
Domain Emotional Maturity

• Low education of mother p .003
• Social isolation/ Lack of supports p .013
• Mothers history of child abuse p .041

28
DomainLanguage and Cognitive Development

• Low education of mother p .002
• On social assistance/financial difficulties p
  .004
• Violence between parents p .052

29
Domain Communication and General Knowledge

• Low education of mom p .003
• Social isolation/lack of supports p .002
• Violence between parents p .007

30
Implications

• Low education of mom
• Significant in all domains
• Significant when control for variables recognized
  in other research as affecting academic success
  ie. Teen parent at time of first birth, on income
  assistance (Brownell, 2010)
• Child welfare system involvement was not used in
  this study but significant in Brownells (2010)

31
Implications

• Home Visiting
• Research identifies difficulty to achieving
  consistency may be affecting outcomes ie. varying
  dosage- early in program
• If enrol in program but dont ever engage or
  engage sporadically could affect outcomes
• Decreased number enter prenatally and research
  has shown greater success with prenatal

32
Further Research

• Compare school readiness between children of
  teen moms who complete high school verses those
  who dont
• Look at dosage of program and effect
• Does FF influence the variables that were
  identified as significant
• Include child protection as a variable to note if
  it is significant or does it change the
  significance of others
• Use propensity scoring to increase group
  similarity
• Use School ready verses not ready as outcome
  variable
• Qualitative component to explain results- explore
  with stakeholders and users, including parents

33
Recommendations

• Policy and Program
• Explore strengthening FFs influence on areas
  that promote school readiness
• Consider increased resources to help parents
  complete high school- ? More childcare or baby
  labs in high schools
• Query if could strengthen FFs influence on
  mothers returning to or staying in high school
• Transportation for mother/child to school program
• Tutoring programs for mothers who are struggling
  to upgrade their skills

34
Acknowledgement

• The presenter would like to acknowledge the
  Healthy Child Manitoba Office including Teresa
  Meyer, Senior Research Analyst, and the Manitoba
  Governments contribution, in providing access to
  the Research Project Data Set. The results and
  conclusions are those of the authors and no
  official endorsement by Manitoba Government is
  intended or should be inferred.
• The presenter would like to acknowledge the
  supportive direction and guidance of committee
  members Dr. Benita Cohen and Dr. Christine Ateah,
  University of Manitoba Faculty of Nursing. Dr.
  Mariette Chartier, Manitoba Centre for Health
  Policy, University of Manitoba.

35
www.clipartof.com/details/clipart/18549.html