Child Abuse Prevention Program Grant Proposal

Child abuse is still a significant problem in United States. There are 2.9 million child abuse reports to Child Protective Services in 2005 and 825,000 indicated cases. Child abuse has profound impact on the child’s current and future development. The consequences include health and physical effects, intellectual and cognitive impact, and emotional, psychological and behavioral consequences. To improve parenting practice of first time parents is an important way to address this problem. The rate of child abuse is highest for children between birth and 3 years of age. A new Cognitive Behavioral Parent Training Program for Child Abuse Prevention targeted at first time mothers is formed by absorbing elements from existed parent training programs and adding components of child abuse prevention specifically targeting to first time mothers and infants, This proposal seeks to evaluate the effectiveness of the Cognitive Behavioral Parent Training Program on maternal psychosocial health and child abuse prevention with a three years’ follow up. Randomized control trail will be used to evaluate the effectiveness of this program for reducing child abuse of newborns.

Specific aims
The first aim of the study is to evaluate the effectiveness of the Cognitive Behavioral Parent Training Program for Child Abuse Prevention on maternal psychosocial health. We hypothesize that after attending the training program, maternal psychosocial health such as parenting stress and self-esteem will be improved. The second aim of the study is to evaluate the effectiveness of the Cognitive Behavioral Parent Training Program for Child Abuse Prevention on child abuse prevention. We hypothesize that mothers attends the training program will be less likely to abuse their children compared to those in the control group. The second aim of the study is to evaluate the effectiveness of the Cognitive Behavioral Parent Training Program for Child Abuse Prevention in a relatively long run. We hypothesize that mothers attends the training program will still be less likely to abuse their children compared to those in the control group at follow-up.

Background and significance
Child abuse is still a significant problem in United States. There are 2.9 million child abuse reports to Child Protective Services in 2005 and 825,000 indicated cases (Olds, 2007). A lot of studies have been done to explore effective prevention methods about child maltreatment, including primary prevention, secondary prevention and tertiary prevention. Primary prevention refers to activities designed to impact the whole population and makes child abuse less likely to occur in the first place. Secondary prevention refers to activities designed to reduce risks of abuse for high-risk population at earlier possible stage. Tertiary prevention refers to activities designed for abusers and victims to rehabilitate them and prevent further prevalence of maltreatment in the population. Over the past few decades many primary and secondary child maltreatment prevention strategies have been used. Success of these programs is varied. Typical programs are home visiting programs, parent training programs, abusive head trauma programs, sexual abuse prevention, etc. A lot of studies have been done to evaluate the effectiveness of home visiting programs and the results show that nurse hose visiting program is most effective secondary prevention program identified to date. However, further evaluation with more rigorous methodology and outcome measures is needed for parent training programs (Krugman, 2007).

Generally speaking, training programs for parents of young children have the potential to improve children’s life-course trajectories, to reduce development problems and to reduce the burden of government and society (Olds, 2007). These programs intend to improve parents’ psychosocial health and thus improve children’s development. Parenting programs could be classified to five categories based on the basic theoretical stance and rationale, including behavioral category, cognitive behavioral category, multimodel category, humanistic category and rational emotive therapy based category (Barlow J, 2003). Behavioral category includes programs that teach parents behavioral strategies in order to manage their children. Cognitive behavioral category includes programs that combine behavioral strategies with cognitive strategies. Cognitive strategies help parents to restructure their thinking about themselves and their children. Multimodel category includes programs that have further components in addition to behavioral or cognitive strategies. Humanistic category includes Webster-Stratton programs (Hughes, 2004) and programs with video –tape modeling methods.

We absorb elements from previous parenting training programs and form a Cognitive Behavioral Parent Training Program (CBPTP) for Child Abuse Prevention targeting first time mothers. Previous evaluation studies show that cognitive behavioral parenting programs are more effective than other kinds of parenting programs with respect to child abuse treatment and intervention (Barlow, 2006). Most of parenting programs specifically designed for child abuse problems are used as a tool for treatment of parents who have already exhibited maltreatment behavior and their children who have suffered maltreatment, for example, Abuse-Focused Cognitive Behavioral Therapy (AF-CBT). The most appropriate target population of AF-CBT is physically abusive parents and their school-aged children (Kolko, 1996). Here the CBPTP is given to first time mothers just after the birth of their babies and before maternal abusive behavior exhibited. This program intends to prevent child abuse at this particular risky time, which could be considered as secondary prevention.

A randomized clinical trial will be conducted to evaluate the effectiveness of this program on improving maternal psychosocial health and preventing child abuse with 3 year’s follow up. There are five characters of this program and the study. Firstly, with respect to based theory, it is a cognitive behavioral program. Secondly, with respect to target population, the direct target of the program is first time mothers. Thirdly, with respect to specific aim, the program is designed specifically to prevent child abuse occurrence for newborns. Fourthly, in terms of outcomes evaluation, both maternal psychosocial health and prevalence of child abuse are evaluated. Finally, in terms of following up, this is a longitudinal study with 3 years’ follow-up. The target of the study is first-time mothers because new parenthood is a very important transition point in a person’s life and children 0-3 years old are more likely to be abused compared to those at other age (Children’s Bureau, 2006).

For first-time parents, they may feel unprepared to raise a child and feel stressful when facing such a challenge. Meanwhile, first-time parents are always willing to learn things about parenting. There are many meaningful topics which could be included in the parent education curriculum. Besides parent-child relations, child development, community resources and supports, all aspects of child abuse and neglect issues could be included. Besides, prevalence of mental health problems is generally high in women (Goldberg, 1992) and maternal metal health status is more related to child development. With respect to children, Previous studies also show that rate of child abuse is highest for children between birth and 3 years of age. The abuse rate decreases as age increases. Infants and young children are physically small sized, in early development status, and need constant care. They are particularly vulnerable to abuse (Children’s Bureau, 2006).

Meanwhile, for many parent education programs, to prevent child abuse is often stated as one of the goals these programs intend to achieve, but a lot of them have not included measure of child maltreatment prevalence in evaluation. More research needs to be done to evaluate the effectiveness of parent education programs on child abuse. Program evaluation should include not only parent psychosocial health outcomes but also prevalence of child abuse.

Previous studies suggest that parent training programs can make a significant contribution to the short-term parent psychosocial health. There is a lack of evidence about whether these results could maintain over time. Long-term follow up data about both parent psychosocial health and prevalence of child abuse needs to be collected (Barlow, 2003; Olds, 2007).

Research design and methods
How could parent training programs impact prevalence of child abuse? Parents’ behavior to children is largely influenced by parents’ psychosocial health. Poor mental health of parents is a major risk factor for child maltreatment. Parents who maltreat their children are identified more likely to have low self-esteem, high stress, depression or antisocial behavior compared to parents do not maltreat children (Goldman J., 2003). CBPTP can improve parents’ psychosocial health by restructuring their cognition about parenting, reducing parenting stress, increasing self-esteem, etc. CBPTP is based on an idea that thoughts cause feelings and behaviors. The theoretical bases of CBPTP are social learning theory. According to social learning theory, human behaviors are determined by cognitive factors (knowledge, expectations and attitudes), environmental factors (social norms, access in community and ability to change own environment), and behavioral factors (skills, practice and self-efficacy). CBPTP targets factors in these three domains.

With respect to cognitive factors, in CBPTP, mothers are taught about knowledge about parenting, the impact of parents’ expectation to children. They will be told that it is normal for infants to cry and how to cope with infants. With respect to environmental factors, they will be taught how to get community support. With respect to behavioral factors, they will be taught about parenting skills and asked to do practice homework. As a result of the program, parents get higher self-efficacy and better mental health status. Good psychosocial health would decrease the risk of maltreatment behavior of parents. While parents’ cognition and behavior improving, children benefit. Parent’s attitude and behavior to child would profoundly influence children’s development. Children experiencing abuse would be less confident and perceive themselves as unlovable and a failure. Such kinds of thinking would have negative impact on children’s life in a long run, even after they enter adulthood. In sum, using parenting education as an earlier prevention method would improve children’s wellbeing in the end.

Procedures
Recruitment, screening, random assignment and enrollment
I propose a randomized controlled trial to evaluate the effectiveness of CBPTP for reducing child abuse of newborns. Eligible samples will be first time mothers who have babies 0-2 months old, do not have serious mental illnesses and substance abuse problems, and live in Baltimore. Two ways will be used to find samples. First, flyers about this study will be distributed in community centers of Baltimore. Mothers who are willing to attend this study can contact the community center or our study group directly. Second, information about first time mothers will also be gotten from hospitals in Baltimore. They will be contacted by members of our study group. All mothers who are willing to attend the study and meet the eligible criteria will be asked to do a baseline interview.

Baseline information will be gotten through the interview, including age, race, family income level, whether a victim of child abuse themselves, whether a single parent or not, etc. Pre-tests include parenting stress, parenting self-esteem and anxiety will also be done. The recruitment stage will last 3 months or so. Some demographic characters of the sample can be anticipated. Since samples are all from Baltimore, it is anticipated that the majority of the sample are American Africans and there will be a large proportion of them coming from low income families. Generally, our samples are relatively high risk population for child abuse behavior since first-time parenthood, poverty and un-secure neighborhood are all risk factors for child maltreatment.

After enrollment, a computer program will be used to do random assignment taking into account factors such as age, family income, etc. Participants who are assigned to treatment group will get 16 sessions’ Cognitive Behavioral Parent Training Program. The program will last 16 weeks, one section each week. Participants who are assigned to control group will not attend CBPTP.

CBPTP is a structured, group-based parent training program. The group will be lead by cognitive-behavioral training specialists. It contains 16 sessions, last 4 months. Each session is one hour long and has a specific topic. We will use “the 20/20/20 rule” in each session. In the first 20 minutes, specialists listen to these mothers’ concerns, show understanding and support/encouragement, and let them to give performance feedback. The specialist and mothers review last week’s homework and discuss questions together. In the second 20 minutes, specialists introduce the specific topic of that session. Topics include how to cope with infants, thinking about parent-child relations, understanding parental and family stressors, identifying abuse-specific triggers and managing reactions, relation of parents’ expectations and child development, community resources and supports, effective discipline strategies, etc. In the third 20 minutes, specialists explore these mothers’ understanding about the topic and lead them to discuss the topic. Then home practice exercise (the homework) will be assigned.

Outcome measurement
First time interview will be conducted both for treatment group members and control group members immediately after the completion of the training program. Second time, third time and fourth time interviews will be conducted in the month one year, two years and three years after the completion of the training program both for treatment group members and control group members. Interviewers will be blinded for which group the interviewee belongs to. Measured outcomes include prevalence of child abuse and the same maternal psychosocial health indicators pre-tested at baseline: parenting stress, self-esteem and anxiety. They will got certain amount of compensate for these interviews.

Self-reports of parenting practices will be used to measure prevalence of child abuse. We will use the revised parent-child Conflict Tactics Scale (CTS-PC) at each follow-up to measure self-reported parenting practices (Straus, 1998). These mothers will be instructed to fill out a paper-and pencil version of CTS-PC since they may be reluctant to tell their negative behaviors to interviewers. They will be asked whether they have 27 different behaviors in the past year and other five questions about parenting in the last week. This instrument measures both prevalence (whether an event occurs) and frequency (how often it occurs).

The Parenting Stress Index (PSI) will be used to measure parenting stress. Self-esteem will be measured by the Parenting Sense of Competence Scale (PSOC) (Odom, 1996). Anxiety will be measured by a subscale of the Profile of Mood States (POMS).

Analysis
We will use intention-to-treat approach when conducing analysis. Intention-to-treat means that once a mother is randomly assigned to a group, she will remain in the assigned group for the whole duration of the study. Conditions such as she will not enroll the training group or will drop out earlier when she is assigned to will not be regarded.

For the first step of statistical analysis, we will use t-tests and Chi-square tests to compare the intervention group and the control group with respect to demographic variables and risk factors assessed at baseline. Secondly, we will use generalized estimating equations (GEE) logistic models to examine the effect of CBPTP on the prevalence of child abuse behaviors for two study groups after adjusting for demographic variables and other risk factors for child abuse, including age, income level, race, a victim of child abuse themselves, whether a single parent or not, and variables showing parenting stress level, self-esteem and anxiety. GEE logistic models will be chosen because we have multiple time points of data collection and GEE can take the correlation in measurement for each person into account. GEE logistic models will be built for several different child abuse outcomes from CTS-PC, including serious abuse/neglect composite scale, very serious physical abuse, serious physical abuse, minor physical aggression, psychological aggression, neglect, harsh parenting in the past week and substantiated abuse or neglect report. Treatment condition (i.e., assigned to intervention group vs. control group) will be the primary independent variable. Thirdly, we will use the frequency of child abuse as outcome.

Frequency is a count variable, so log-linear regression will be used to analyze treatment effects. Here treatment condition will still be the primary independent variable and those variable adjusted above will be adjusted too. Fourthly, we could add an interaction term of psychosocial variable (i.e., parenting stress) and treatment variable to these models to see the effect modification of parenting stress/self-esteem/anxiety on child abuse prevalence. Finally, logistic regression model will be used to assess program effect on psychosocial health changes. There would be missing values in CTS-PC reports and the missing data may not be random. The control group may be more likely to skip questions compared to intervention group (DuMont, 2007). We will use multiple imputation method to deal with missing data.