Can we measure risk in home visitation? An examination of the predictive validity of the Healthy Families Parenting Inventory (HFPI)
Can We Measure Risk in Home Visitation?
Authors:
Cara Kelly, Craig LeCroy
Arizona State University, United States
Article Information
Keywords: Child welfare, home visitation, child abuse and neglect, assessment, measurement
Abstract
Objective: The study examines the predictive validity of the Healthy Families Parenting Inventory (HFPI) in predicting future risk of child abuse and neglect.
Method: Participants included 2,088 parents in a statewide home visitation program, mostly low-income mothers with various socio-demographic risk indicators. HFPI was administered at enrollment and followed by monitoring through administrative data for about one year.
Results: Pre-intervention HFPI scores showed incremental predictive validity for future maltreatment reports.
Conclusions: HFPI can be effectively used during home visitation to identify needs and services aimed at reducing child maltreatment likelihood.
Introduction
Maltreatment (including neglect, physical, sexual, and emotional abuse) persists as a significant social issue despite various intervention efforts.
Evidence-based early childhood home visitation programs primarily aim to prevent maltreatment by enhancing child well-being and supporting vulnerable families in the U.S. (Azzi-Lessing, 2017).
These programs focus on at-risk families with children under five, building supportive caregiver relationships to improve family welfare.
Studies indicate considerable variation in risk levels among these families (Duggan et al., 2007; Peacock et al., 2013), yet little focus has been on validating assessment protocols for accurate risk identification.
1.1. High-risk Children and Home Visitation
Research highlights that many families enrolled have significant maternal and child risk factors for maltreatment (Jacobs et al., 2016; Easterbrooks et al., 2019).
Common factors:
Mothers often live in poverty, are teen parents, undereducated, or suffering from mental health or substance issues (Ammerman et al., 2010).
Infants often experience congenital disabilities, low birth weight, or are born preterm (LeCroy & Milligan Associates, 2016).
Chronic stress from community violence, poor housing, and lack of resources heightens risks (Fortin et al., 2012).
Untreated maltreatment has long-term adverse effects on development (Barlow & Scott, 2010).
Home visitation, when implemented thoroughly, can alter negative parenting practices and improve child and family well-being (Caldera et al., 2007; DuMont et al., 2008).
1.2. Home Visitation Program Impact on Risk Factors
Home visitation models aim to promote child development, reduce maltreatment risk, and enhance parental competence (Avellar & Supplee, 2013).
Engaging families through preventive services helps to mitigate child maltreatment occurrences (Daro & Donnelly, 2002).
Flexible programs can create tailored service plans suitable for individual family needs, contrary to public welfare systems (Duggan et al., 2000).
1.3. Measuring Risk in Home Visitation
Assessment tools have emerged as vital in determining family risks and guiding service interventions (Shlonsky & Gambrill, 2014).
Tools exist for protecting children and managing service demands, but the reliability and validity of these instruments in early childhood contexts are questionable (Kirk et al., 2005).
Understanding the multi-dimensional nature of maltreatment risk is necessary for effective interventions (Ogles et al., 2002).
1.4. Background of the Healthy Families Parenting Inventory (HFPI)
HFPI is a 63-item assessment created to evaluate risks and protective factors within home visitation settings (LeCroy & Milligan Associates, 2004).
Administered within 30 days of program start and every six months, scores range from 63 to 315 for overall and 5 to 50 for each of the nine subscales: 1. Social Support
Problem Solving/Coping
Depression
Personal Care
Mobilizing Resources
Role Satisfaction
Parent/Child Interaction
Home Environment
Parenting Efficacy
Items are scored on a Likert scale indicating frequency, with designated “red flag” and “strength indicator” items guiding intervention plans.
Reliability for HFPI subscale and total scores is established (Krysik & LeCroy, 2012).
1.5. Purpose of the Current Study
Prior studies show risks for maltreatment, yet family identification in home visitation needs clarification.
This study aims to validate HFPI's effectiveness for predicting future maltreatment in home visitation, potentially influencing policy and practice to better serve high-risk families.
Methods
2.1. Study Context
The study involves 2,088 parents with young children in a voluntary statewide child abuse prevention program.
Families volunteered, with criteria of having at least one child under three months and meeting socio-environmental risk factors via a two-stage screening process.
2.2. Data Collection Procedures
Data was extracted from a maintained database for longitudinal evaluation, combining socio-demographic, HFPI scores, and public child welfare data.
The Institutional Review Board approved study procedures.
2.3. Sample Characteristics
Most participants were female, low socio-economic, with 67.8% single, and less than half graduated high school.
Average household size: ~4; average children prior: 1; 8% faced maltreatment reports within a year.
2.4. Measures
HFPI scores evaluated for predictive validity using total composite and subscale scores.
Maltreatment defined by official reports, allowing the study to assess risks regardless of substantiation status (Hussey et al., 2005).
2.4.1. Analytic Approach
Bivariate and multivariate analyses determined HFPI's predictive validity.
T-tests compared averages between maltreatment and non-maltreatment groups, while correlations and logistic models assessed risk relationships.
Results
3.1. Bivariate Analyses: T-Tests
Differences in HFPI scores between maltreated and non-maltreated families were significant across various subscales (social support, problem-solving, etc.), with effect sizes evaluated via Cohen’s d (0.17 to 0.26).
3.2. Bivariate Analyses: Correlations
Point-biserial correlations showed significant relationships between HFPI scores and maltreatment outcome, affirming HFPI's relevance.
3.3. Multivariate Analyses: Logistic Regression
HFPI's total composite score was negatively predictive of maltreatment risk (AOR = 0.99) with specific subscales showing mixed results regarding risk increase or decrease (e.g., personal care AOR = 0.92).
Discussion
4.1. Predictive Validity of HFPI
The study affirmed HFPI's predictive validity in assessing families for maltreatment risk based on various interacting factors.
4.2. Implications for Policy
Public child welfare can benefit from early identification via home visitation to divert families from engaging with the welfare system.
4.3. Implications for Practice
HFPI can guide practitioners in effectively serving families based on their risk scores, aiding in the development of individualized treatment plans.
Limitations
Limitations related to using official maltreatment reports raise questions about generalizability and actual maltreatment rates.
Future research should evaluate HFPI's predictive validity across diverse samples and follow-up periods.
Conclusions
Validating HFPI presents an opportunity for early childhood home visitation programs to monitor and improve outcomes related to family risk for maltreatment.
References
Over 70 references cited regarding early childhood intervention, assessment, risk factors, and program effectiveness related to home visitation and child welfare.