The effectiveness of parent education and home visiting child maltreatment prevention programs

 

You are in an archived section of the AIFS website 

 

Content type
Policy and practice paper
Published

April 2006

Researchers

Prue Holzer, Jenny R. Higgins, Leah Bromfield, Daryl Higgins

Contents


Abstract

This paper investigates the effectiveness of child maltreatment prevention programs. The paper begins with an overview of the different types of prevention programs (for example, primary/universal, secondary and tertiary interventions) and the way in which programs can be evaluated. Different modes of evaluation are described (that is, process, impact and outcome evaluations) and the benefits and limitations of adopting certain methodological approaches are outlined. Particular attention is given to two widely used prevention programs:parent education and home visiting programs.Parent education and home visiting program evaluations are critiqued in order to determine whether these programs are effective in preventing child maltreatment and the degree of evidence that exists to support these claims. The paper concludes by discussing the implications of the present findings and their applicability to practice.

Introduction

In an attempt to reduce the rates of child maltreatment, government bodies, non-government organisations, and community alliances have implemented a variety of child maltreatment prevention programs (MacLeod & Nelson, 2000). The aim of this paper was to investigate evaluations of two commonly used child maltreatment prevention programs: parent education and home visiting programs.

Prior to discussing the effectiveness of parent education and home visiting programs, we present an overview of the different types of child maltreatment prevention programs currently in operation. Following this, key terms and research concepts (for example, elements of a rigorous evaluation) are defined in 'plain language'. In this way, we hope to provide readers who are unfamiliar with program evaluation and effectiveness research, the necessary information to determine whether or not an evaluation has been conducted appropriately. The latter half of the paper specifically addresses the effectiveness of parent education and home visiting programs and the evidence that exists to support conclusions concerning their utility.

For the period 2004-2005 there were 252,831 notifications to statutory child protection services in Australia (Australian Institute of Health and Welfare, 2006). This is more than double the number of notifications for the period 1999-2000: 107,134 (Australian Institute of Health and Welfare, 2001)1. Thus, attention has been drawn to the need for prevention. Prevention programs aim to: prevent the occurrence of child maltreatment; prevent recurrences of child maltreatment in contexts where it has already occurred; and to reduce the likelihood of inter-generational transmission (that is, victims of childhood maltreatment becoming perpetrators of child maltreatment) (Doll, Koenig, & Purcell, 2004; Prentice, 2002).

Previous research has shown that parent education and home visiting programs are the most widely used child maltreatment prevention programs in Australia (Tomison, 1998a; Tomison & Poole, 2000). In addition, parent education and home visiting programs are strategies used primarily with families in which children have been identified as 'at risk' of maltreatment. Therefore, practitioners are likely to frequently refer families in which children may be vulnerable to maltreatment to programs such as these. Given the widespread use of parent education and home visiting programs, there is a need to consider the effectiveness of these interventions in preventing child abuse and neglect.

The purpose of this paper was to review evidence on the effectiveness of two different types of child maltreatment prevention programs: parent education2 and home visiting programs3. Specifically this paper addresses the following questions:

  • Are parent education and home visiting programs effective in preventing child maltreatment?
  • What are the components of a successful parent education and home visiting program?
  • How credible is the evidence base?

To this end, literature (both Australian and international) was sourced through existing National Child Protection Clearinghouse resources and via a systematic search of relevant electronic databases4. Articles published from 1990 onwards, which documented the results of prevention program evaluations, were sought.

Prevention

Child maltreatment prevention programs aim to prevent the occurrence or re-occurrence of child abuse and neglect. Prevention programs attempt to do this by: increasing community awareness and knowledge of the nature, prevalence, and possible indicators of child maltreatment; implementing early intervention strategies in situations where risk factors for child maltreatment are apparent; and attempting to reduce the long-term impact of maltreatment in contexts where maltreatment has already occurred.

The targeting of prevention programs at different groups with varying degrees of risk for child maltreatment is referred to as a 'composite approach' to prevention. A composite approach to prevention originated in the public health model of disease prevention. In the public health model of disease prevention, preventative interventions are described as either: primary, secondary, or tertiary interventions (Tomison & Poole, 2000). Child maltreatment prevention programs are also commonly categorised as primary, secondary, or tertiary interventions/prevention programs.

Primary prevention

Primary (or universal) interventions are strategies that target whole communities in order to build public resources and attend to the factors that contribute to child maltreatment. The Australian Childhood Foundation's 'Every Child is Important' campaign is an example of a primary intervention (Australian Childhood Foundation, 2004). The program uses television, radio and print material to educate the community in relation to the importance of a child's early years and the need for a child to have a safe and secure home environment.

Secondary prevention

Secondary interventions target families who are 'at risk' for child maltreatment5.

Where families are at risk for child maltreatment (due to the presence of one or more risk factors for child maltreatment), secondary approaches prioritise early intervention. Secondary interventions generally involve early screening to detect children who are most at risk, followed by a combination of interventions (for example, home visiting, parent education, and skills training) to address the risk factors for child maltreatment (Thyen, Thiessen, & Heinsohn-Krug, 1995).

The Community Bubs program, developed by Family Life, is an example of a secondary intervention. The Community Bubs program provides a home-based support service for highly vulnerable families in Melbourne, Victoria. The program targets families in which parents are limited in their capacity to provide for the wellbeing of their infant. Community Bubs provides intensive professional assistance for up to one year. (For more information on the Community Bubs program visit, <http://www.familylife.com.au/bubs.html>).

Tertiary prevention

Tertiary interventions target families in which child maltreatment has already occurred. Tertiary interventions seek to reduce the long-term implications of maltreatment and to prevent maltreatment recurring. Given that tertiary interventions operate once child maltreatment has occurred or is believed to have occurred, they have been assessed as reactive and 'after-the-fact' approaches (Thyen, Thiessen, & Heinsohn-Krug, 1995). Tertiary interventions (for example, statutory child protection services which operate in each state and territory throughout Australia) initially dominated the child protection domain. However, primary and secondary interventions have gained increasing attention as government bodies, non-government organisations, and community alliances have recognised the importance of proactive strategies, which intervene before maltreatment occurs (Thyen, Thiessen, & Heinsohn-Krug, 1995).

The emphasis on primary and secondary interventions followed greater awareness of the cost of maltreatment (both to individuals and the community) and to the publication of research, which suggested that tertiary level child protection services are not as successful as is often assumed (Geeraert, Noortgate, Grietens, & Onghena, 2004). To illustrate, Geeraert et al. (2004) reported that the difficulty of treating abusive parents at the tertiary level is more pronounced than at the primary or secondary prevention levels, as abusive parenting may have become a fixed pattern of parent-child interaction.

Prevention programs and their target groups

Prevention programs differ in their focus. Some strategies involve working directly with the parent (such as parent education programs), while other programs target children (such as school-based personal safety programs). In contrast, home visiting programs target the family unit, but may differ as to whether the program goals relate to outcomes specifically for the child, the parent, or the family. Table 1 illustrates the way child maltreatment prevention programs can be conceptualised in a public health framework (that is, primary, secondary, and tertiary interventions) according to their level of operation or focus (that is, children, parents or community).

Table 1 Types of child maltreatment prevention programs
 Intervention level
FocusPrimarySecondaryTertiary
ChildPersonal safety programsAssertiveness training for 'at risk' childrenTherapeutic programs (eg, group or individual therapy for abused children)
Parents/FamilyUniversal nurse home visiting programsParent education programsChild protection service referrals (eg, anger management programs
CommunityGeneral media awareness campaignsTargeted media campaigns in 'at risk' communtiesIntensive community interventions (eg, alcohol zero tolerance zones)
Note: The 'public health' model of conceptualising child maltreatment prevention programs, although useful,has some limitations. Not all programs can be neatly classified into distinct categories. For example, therapeutic interventions for maltreated children can be considered tertiary interventions (as maltreatment has already occurred) and also secondary interventions (as addressing the consequences of maltreatment may reduce the likelihood of inter-generational effects) (Tomison & Poole, 2000).

Evaluation

In this section of the paper we define general concepts associated with program evaluations, specifically, the elements of a rigorous evaluation and how to determine whether program evaluation findings are reliable.

The aim of child maltreatment prevention programs is to prevent children from experiencing abuse or neglect; but do they work? In order to answer this question, evaluation is an essential component of any prevention program. Evaluations attempt to demonstrate whether or not a program is of benefit to participants (for example, by reducing the incidence of child maltreatment or improving family functioning). In addition, evaluations can illustrate why some programs are effective, while others are not (for example, by identifying the essential characteristics of an effective or efficacious prevention program)6.

Given there are a limited pool of funds available for child welfare programs, it is important that those prevention programs that are funded actually work to ameliorate the risks of child maltreatment. Consequently, there is a growing demand for programs to be evaluated and their effectiveness demonstrated (Tomison, 2000). Program evaluations can also provide policy makers and service providers with information to facilitate sound strategic and service planning.

Types of evaluation

The increasing demand for 'evidence-informed' services has resulted in a larger number of prevention programs undergoing evaluation (Tomison & Poole, 2000). However, the quality of these evaluations may vary. Therefore it is important for research users to be able to assess whether an evaluation has been conducted well and whether the evidence from the evaluation is credible. In this section, the authors describe the elements of evaluation and identify what makes a good evaluation.

There are several ways program evaluations can be conducted. The three main categories of evaluation are: process, impact and outcome evaluations (Tomison & Richardson, 2004)7.

Process evaluations

Process evaluations consider the way in which a program is implemented or practiced. That is, process evaluations investigate whether a program is doing what it intended to do in a consistent fashion. Tomison (2000) explained that process evaluations essentially investigate a program's 'integrity'. The aim of a process evaluation is to determine whether certain program changes (such as increased procedural consistency) would enhance program delivery. Process evaluations are used to answer questions such as, 'are all service providers administering the program in the same way?'and 'how much of the intervention was provided and by whom?' Process evaluations provide useful information for service delivery planning, however, cannot answer the question of whether or not a program is effective (Tomison & Richardson, 2004).

Impact evaluations

Impact evaluations are the most common form of program evaluation. Impact evaluations measure the direct effect of a program according to its operational aims and objectives (Tomison & Richardson, 2004). For example, an impact evaluation of a parent education program would assess whether participating parents have improved their parenting skills in ways that the program aimed to promote (for example, a decrease in the use of physical punishment in favour of more constructive parenting practices). That is, impact evaluations attempt to answer questions such as 'do participants exhibit an increase in their knowledge and/or parenting skills?'

Outcome evaluations

Outcome evaluations attempt to measure the direct consequences of the program under investigation on the underlying goal (Tomison and Richardson 2004). The difference between an impact and an outcome evaluation is that an impact evaluation looks at the direct aim (for example, parenting skills) whereas an outcome evaluation considers the underlying goal (for example, child abuse prevention). An outcome evaluation attempts to answer questions such as, 'does the program reduce the incidence of child abuse and neglect?' (Tomison & Richardson, 2004).

Outcome evaluations enhance impact evaluations by enabling investigation as to whether the assumptions underlying the direct aims of the program are accurate. For example, do increases in parental knowledge and skills (the direct aim of a parenting program) result in a reduction in the incidence or prevalence of child maltreatment (the overarching goal or purpose of the program). To directly assess these elements, an outcome evaluation in needed (Rossi, 1997).

Designing a rigorous evaluation

To have confidence in an evaluation, policy makers and service providers must be able to determine whether the evaluation has been conducted appropriately (that is, whether an evaluation has been conducted in a way that enables individuals to find out whether the program actually works). The way in which an evaluation is conducted is referred to as the 'research design'.

Essential elements of a rigorous evaluation

Impact and outcome measures. To determine whether a program has achieved its objectives, evaluations need to measure a program's success according to what it hoped to achieve. For example, if a program is designed to improve family functioning, an evaluation will need to assess aspects of family functioning, such as parent/child interactions or levels of family stress/hostility. Alternatively, if a program hoped to reduce the incidence of child maltreatment, an evaluation will need to assess the program's influence of the occurrence of child abuse and neglect. Thus, the measure of a program's success should reflect the goals a program hoped to achieve. Measures used to assess a program's success are referred to as 'outcome measures'.

Pre- and post-test research designs. To ascertain whether a program has resulted in 'change', outcome measures need to be assessed both before and after participation in the program. This method of conducting an evaluation is referred to as a 'pre- and post-test design'.

Comparison group. To determine whether participation in a particular prevention program is effective, an evaluation needs to compare outcomes for participants with either non-participants or participants in a comparison group receiving a different kind of intervention. For example, people grieving after trauma may find that their trauma decreases overtime regardless of whether or not they participate in grief counselling. Thus, for a program to document its effectiveness beyond the passing of time or other influences (for example, maturity, insight, changing circumstances) an evaluation will need to simultaneously compare the outcome measures of participants and nonparticipants. This helps in drawing a causal link between participation in an intervention and the observed/reported impact or outcome. The non-participant group is referred to as a 'comparison' or 'control' group.

Follow-up. A prevention program may prove effective during and immediately after participation, however, do these outcomes extend beyond the short-term? To determine whether a program has a lasting effect (such as improved family functioning or a reduction in the incidence of child maltreatment across a child's lifetime) an evaluation will need to conduct follow-up assessments on the same outcome measures. This type of study is referred to as a 'longitudinal research design'.

Avoiding design flaws

When testing program effectiveness, it is important that an evaluation measures only the influence of the program. That is, the findings of an evaluation should not be swayed by other events or circumstances in participants' lives, for example:

  • Are the participants also attending another program or service?
  • Do participants differ in their social supports or family arrangements?
  • Do participants have a greater motivation to change than non-program participants?
  • Are there differences between program participants and non-program participants in terms of the severity of problems (for example, are those with the most severe problems more likely to receive priority access)?
  • Are people assessed as needing welfare services likely to be different from the general community in terms of personal characteristics such as income, education, or health?

In practice, there is no 'perfect' way to conduct an evaluation of the effectiveness of child maltreatment programs. However, researchers should be aware of the limitations of the evaluation method they adopt and take steps, where possible, to reduce the influence of these constraints. For example, in order to avoid the influence that pronounced differences in personal characteristics or participant circumstances might have on the results of an evaluation ('confounding effects'), evaluators could attain a sample of participants who are representative of the community in which they live. In addition, evaluators could attempt to match participants who partake in the program with people who do not (but who are being used as a comparison group within the evaluation). In conducting an evaluation in this way, an attempt is being made to 'contain' or 'control' the influence these 'confounding effects' may have on the results (for example, swaying the results in such a way that a program appears effective when it is not).

Examples of different methods of evaluation and their ability to control for design flaws.

  1. Compare program participants' test scores before and after program completion, in this way holding some 'personal characteristics' constant. However, research designs that adopt this approach do not contain a comparison group. Thus, it is not possible to demonstrate that changes in outcome measures are primarily due to program participation.
  2. Compare program participants with a community sample (also known as a 'normal' sample). This approach would, in theory, demonstrate change that resulted from program participation. However, it is possible that a community or normal sample would differ from program participants on a number of factors (for example, income, education, health) and would not possess the same risk for child maltreatment as the program group.
  3. Compare changes on the outcome measures of program participants with people on the waiting list, and/or people participating in a modified or less intense version of the same program. This approach would allow for a comparison between different groups, however, there may be significant differences between these groups, which influences the results. For example, people on the waiting list might have been assessed as having lower needs than those in the program.
  4. Match participants and non-participants (both of which are deemed to be 'at risk' of child maltreatment) according to personal characteristics and other circumstances such as income, education, family arrangements, or age. This approach would enable a comparison between two similar groups (in contrast to the comparison between the participant and community sample). However, there may be differences (for example, in motivation, insight and self-awareness) which influences program results.
  5. Randomly assign people who are 'at risk' of child maltreatment and who want to participate in a prevention program to either the participating group or the comparison/control group. The 'random assignment' of participants to either the program or control group, refers to a process whereby participants are allocated in such a way that there is no systematic bias or pattern, thus any bias or pattern found in the data would be the result of chance (this sort of controlled research design is an example of an efficacy trial). The strongest evidence of a program's effectiveness is found in research that randomly assigns participants to a treatment or control group and attempts to take into account statistical differences in personal characteristics such as income, age, and education.

Limitations when evaluating child maltreatment prevention programs

Rigorous research designs are difficult to achieve when evaluating child maltreatment prevention programs, as there are many ethical and practical considerations in research of this nature. Researchers need to consider, for example, whether it is ethical for the comparison or control group, or the program being evaluated, not to receive additional services. In addition, researchers need to give thought to ethical issues concerning confidentiality, mandatory reporting, the potential for harm, and gaining informed consent. Furthermore, there may be a multitude of factors affecting families deemed to be at risk of child maltreatment, which may not be readily controlled within an evaluation study (Tomison, 2000). For example, a service provider attempting to evaluate a prevention program may not be able to 'control' or 'contain' the influence of participant circumstances beyond the program (such as changes in employment or home environment).

Example of a rigorous evaluation: The Triple P-Positive Parenting Program

A large-scale evaluation of the Triple P-Positive Parenting Program compared the efficacy of three levels of the program in reducing disruptive child behaviour and promoting competent parenting. The aim of the evaluation was to examine the effects of targeting marital conflict and parental depression in family-based early intervention programs.

Sample Selection. Participants were recruited through a community based media campaign that included newspaper stories, posters and flyers displayed in child care centres, kindergartens and the like. In total, 305 families with a preschool aged child who exhibited behavioural problems participated in the study. The 305 participating families were selected from a larger pool of respondents as they were deemed to be at greater risk for conduct problems and family difficulties. The participating families were thought to be at greater risk than other respondents on the basis of elevated rates of disruptive child behaviour, parental conflict, maternal depression, single parenthood, and financial pressures. That is, the program was a secondary prevention program because it targeted families assessed as being at heightened risk for maltreatment.

Program design. Participants were randomly assigned to one of four treatment groups: the self-directed Triple-P group; the standard Triple-P group; the enhanced Triple-P group; or the control group.

The treatment groups were incremental:

  • the self-directed treatment group received the materials from a ten-session parenting workshop that addressed core child behaviour management strategies;
  • the standard Triple-P treatment group received the same materials as the self-directed group, however, the weekly interventions were administered by a trained practitioner. Thus, the standard Triple-P group also received active skills training;
  • the enhanced Triple-P group received partner support and coping skills training in addition to the interventions contained within the self-directed and standard treatment groups; and
  • participants allocated to the control condition did not receive any treatment

Parents completed semi-structured interviews, self-report measures, and home observations prior to the commencement of the study. These procedures were repeated post-intervention and at follow-up one year after completion of the program.

Impact measures. Program impact was calculated according to: clinically observable changes in parent-child interactions (assessed through a 30-minute home video recording of parent-child interactions); parent self-report measures; and a daily parent report of child behaviour.

Of note, the evaluation did not include an outcome measure (for example, the number of reports of child maltreatment prior to the program and after), thus it is not clear whether the impact of the program (that is, the benefits exhibited concerning improved knowledge and parenting skills) would translate to a reduction in the prevalence of child maltreatment.

Source: (Sanders, Markie-Dadds, Tully, & Bor,2000).

 

The effectiveness of parent education and home visiting programs

In the next section, we review the effectiveness of parent education and home visiting programs. In so doing, the following questions are addressed:

  • Do parent education and home visiting programs work?
  • How credible is the evidence base?
  • What are the successful components of these programs?

PARENT EDUCATION

Parent education programs are based on the premise that interventions that promote caring, consistent, and positive parenting are central to creating safe and supportive environments for children (Sanders & Cann, 2002). Research has documented that the risk of child maltreatment is heightened when parents lack necessary child rearing skills, social supports, and knowledge of child development (Tomison, 1998b). Thus, parent education programs are designed to: increase parental knowledge of child development; assist parents in developing parenting skills; and normalise the challenges and difficulties inherent in parenting (Sanders, Markie-Dadds, Tully, & Bor, 2000).

Parent education programs, in a similar way to the public health model of disease prevention, operate on a number of levels. For example, community awareness strategies operate at the primary level (such as the Australian Childhood Foundation's Every child is important' campaign), whereas group training sessions and one-on-one programs with parents (such as, the Triple P-Positive Parenting Program) operate at the secondary or tertiary level (Australian Childhood Foundation, 2004; Sanders, Markie-Dadds, Tully, & Bor, 2000). Despite these differing approaches, all parent education programs are thought to assist families primarily by increasing parental knowledge and reducing parental stress. Parent education programs achieve these results by training parents in behavioural management techniques, problem solving, and personal coping skills (Tomison, 1998b).

Policy makers and service providers began to embrace parent education programs in Australia in the late 1980s. At first, Australian service providers adopted existing parenting programs from other countries. Since that time, an increasing number of Australian programs have been developed and a growing number of overseas programs have been adapted to suit the Australian context (Tomison, 1998b).

At present, parent education programs are an essential component of early intervention with 'at risk' families (Tomison, 1998b). However, until recently, few rigorous outcome evaluations had taken place. Tomison noted that in the absence of such research, the overall success of parent education programs in preventing child maltreatment was unclear. However, more recent audits of evaluative research indicate that progress has been made in these areas (Tomison & Poole, 2000).

Do parent education programs work?

Parent education programs emphasise the development of parenting skills and competence, and aim to achieve the application of these attributes to various care-giving contexts (Tomison, 1998b). As parent education programs are often a component of other types of child maltreatment prevention strategies (such as home visiting programs or family support programs), the parent education programs reviewed in the present paper were limited to those that operate independently of other program types and primarily in a centre-based or community-based setting. Twenty articles concerning parent education program evaluations met the inclusion criteria for the present study, three of which were meta-analyses. Eighteen articles reported successful results following participation in a parent education program.

Programs reviewed

The parent education program evaluations reviewed in the studies sourced for the present paper included:

  • the Nurturing Program (offered by SAFEchild in the USA);
  • the ESPACE Program (Canada);
  • the Multilevel Selected Prevention Program (USA);
  • the Parent Education Program for Teen Mothers (USA);
  • the Child Parent Enrichment Project (USA);
  • the Systematic Training for Effective Parenting Program (Canada);
  • Project 12 Ways (USA);
  • the Head Start Education Program (USA);
  • the Triple P-Positive Parenting Program (Australia);
  • the Signposts for Building Better Behaviour Program (Australia);
  • the Parents Under Pressure Program (Australia);
  • the DePelchin Children's Centre Parent Education Program (USA);
  • the Parenting between Cultures Program (Australia); and
  • the Parent Effectiveness Training Program (Australia)

(For a detailed review of each program, see Holzer, Bromfield, Richardson and Higgins 2006).

The results of the evaluations included:

  • fewer incidents of child maltreatment (however, only a small number of studies directly measured this outcome);
  • a reduction in the prevalence of negative/unhelpful parenting attributions (for example, a parent attributing a child's behaviour to malicious intent);
  • a greater ability to use positive/productive discipline strategies rather than punitive strategies;
  • increased parental competence and self-efficacy; and
  • greater parental knowledge/awareness of child development, risk factors for maltreatment, and child outcomes following abuse and neglect (Holzer, Bromfield, Richardson, & Higgins, 2006).

Was the evidence base credible?

The majority of evaluations reviewed in this study documented favourable results following participation in parent education programs. However, these results were based on differing research designs. Some evaluations possessed rigorous research designs, whereas other evaluations were not conducted in such a rigorous fashion (for example, non-randomised treatment and control groups, the use of retrospective measures which relied on participant recall rather than a 'true' pre- and post-program design).

In addition, most studies did not include outcome measures that directly assessed whether there had been improvements in child wellbeing or a reduction in the incidence of child maltreatment. For example, most studies assessed whether program participation led to attitudinal change such as greater disapproval of punitive disciplinary strategies, rather than directly assessing a program's influence on child maltreatment through, for example, statutory child protection service data or child health data. Thus, although the majority of evaluations of parent education programs had favourable results, the direct influence of parent education programs in reducing the incidence of child maltreatment remains somewhat speculative, as this outcome was generally not measured directly.

What makes a good parent education program?

Although there are criticisms of each, the key features of successful programs included:

  • targeted recruitment;
  • a structured program;
  • a combination of interventions/strategies; and
  • a strengths-based approach.

Targeted recruitment

Successful programs specifically targeted families identified as 'at risk' for child maltreatment. In some studies, families were identified as 'at risk' for child maltreatment on the basis of self-report responses on scales measuring parent anger, parent attitude, family stress, and child abuse potential (for example, Bugental et al. 2002). In other studies, a family's degree of risk was determined on the basis of income, level of parental education, parental age, parental unemployment and the absence of social support (for example, Peterson, Tremblay, Ewigman and Saldana 2003). Most programs identified and recruited families through agency and/or hospital referral or through family self-referral.

As Kelly (2000) explained, targeted recruitment is important to ensure that programs designed to treat 'at risk' families actually identify and service these families. In the absence of targeted recruitment practices, prevention programs may source participants who do not require the use of such interventions (for example, programs which recruit parents largely on the basis of selfreferral may not be servicing the most 'at risk' parents). In short, for programs to demonstrate effectiveness in treating 'at risk' parents, they need to effectively target 'at risk' parents. For the most part, the parenting education programs reviewed in the present study, specifically recruited parents 'at risk' for child maltreatment.

The appropriate targeting of secondary prevention programs at populations identified as being at greater risk for maltreatment does not negate the need for universal or primary prevention programs. This issue is discussed in greater detail in the implications section of this paper.

Structured programs and length of participation

All of the parent education programs reviewed in this study possessed a clearly defined structure or sequence of intervention. However, there was some variety across programs in the nature and length of participant involvement. For example, some parent education programs encompassed educational workshops over the course of an afternoon, such as Herbert, Lavoie and Parent's (2002) evaluation of the ESPACE parent education workshop. In contrast, others programs consisted of tailored interventions involving ongoing parent participation over several weeks or months, such as Britner and Reppucci's (1997) evaluation of the Parent Education Program for Teen Mothers.

For the most part, the more intense and prolonged programs were more effective than short-term programs in reducing the prevalence of child maltreatment. Thus, Kelly's (2000) observation that the positive impacts of prevention programs are enhanced when treatment regimes are longer rather than shorter is applicable to the present findings. Operationally, 'longer' programs equate to programs with treatment periods in excess of 4-6 weeks, for example, the Multilevel Selected Prevention Program (Peterson, Tremblay, Ewigman, & Saldana, 2003).

A combination of interventions/strategies

The most successful programs were comprised of a combination of parent education strategies. Programs that included parent skills training, cognitive retraining, child development information, and concrete services, were generally more effective than programs that had a more narrow focus. Programs that focused exclusively on improving parental content knowledge were not as effective as programs that offered a combination of interventions. Several authors of program evaluations have noted the benefits of combining parent education strategies (for example, education concerning prevalence rates, possible indicators of maltreatment, and child outcomes) with other complementary initiatives such as medical assistance, employment programs, behavioural/skills training, and therapeutic interventions. Although the evidence base was mixed concerning the added benefit of combining basic parent education strategies with complementary initiatives, the majority of papers reviewed in this study supported their inclusion (Britner & Reppucci, 1997; Kelly, 2000; Pecora, Fraser, & Haapala, 1992).

Strengths-based approach

For the most part programs that incorporated a strengths-based approach (meaning an approach that identifies parents' existing skills and strengths and builds on these) achieved more positive results than programs that operated from a deficit perspective. MacLeod and Nelson's (2000) study was particularly illustrative of this point. Their meta-analysis revealed that programs incorporating client involvement and a strengths-based approach were significantly more effective than programs without these features (MacLeod & Nelson, 2000). Programs that adopt a strengths-based approach emphasise parental skill and proficiency, rather than focusing on shortfalls. In so doing, programs with this approach aim to further enhance parenting strengths and their application to all parenting/care-giving contexts (for example, the Triple P-Positive Parenting Program).

The adoption of a strengths-based approach in child protection reflects a broader trend in social welfare away from an exclusive focus on 'risks' and 'deficits' (DePanfilis & Wilson, 1996). DePanfilis and Wilson (1996) noted that a strengths-based perspective has become the preferred mode of practice in a number of diverse contexts. However, maltreating families who come into contact with child protection services tend to differ from families often referenced in the strengths-based practice literature. That is, maltreating families and families at greatest risk of child maltreatment tend to be faced with long term, chronic and multiple challenges. It may be more difficult for practitioners and families to identify strengths in these contexts. Similarly, parental risk factors for child maltreatment (such as mental health problems and drug/alcohol use) may exacerbate the difficulty of establishing a trusting and open dialogue with parents. Thus, although a strengths-based approach was favoured in the literature reviewed, the challenges of practicing in this framework with 'at risk' families must be acknowledged.

Summary

In brief, the parent education programs reviewed in the present study improved parenting competence, effectively addressed risk factors for child maltreatment, and in some instances where direct measurements were made (for example, through child protection service data), resulted in fewer incidents of child maltreatment. Thus, on the basis of the literature sourced, practitioners and policy makers have reason to be optimistic about the effectiveness of parenting education programs.

A successful parent education program: The Triple P-Positive Parenting Program

Aim. The Triple P-Positive Parenting Program was developed at the University of Queensland in Brisbane, Australia. The program aims to 'prevent severe behavioural, emotional, and developmental problems in children by enhancing the knowledge, skills and confidence of parents' (Sanders, Markie-Dadds, & Turner, 2003). The program fosters the development of parenting competence and parenting self-efficacy through a comprehensive, multi-level, and strengths-based education approach.

Structure/design. The Triple P-Positive Parenting Program is comprised of five levels of intervention, which operate on a tiered continuum of increasing intensity. The program targets families with children aged from birth to 12 years of age. The first level of the program operates as a primary or universal intervention. Level 1 of the program involves print and electronic media-based parent information campaigns. The campaigns seek to increase community awareness of available parenting resources and to increase parents' receptivity to participating in the Triple-P Positive Parenting Program by 'normalising' help seeking behaviours.

Subsequent levels of the program are more selective and illustrate a secondary child maltreatment prevention program. Level 2 of the program offers targeted interventions for specific concerns such as a child's developmental or behavioural difficulties. Strategies such as face-to-face, telephone, and group services are available to parents. The services offered at this level of the program are fairly contained, with discrete advice given for specific and relatively mild concerns. Typically, parents attend one or two primary health care sessions, which provide early anticipatory developmental guidance.

The third level of the Triple P-Parenting Program includes similar services, however, this level is geared toward families with a child who possesses mild to moderate behavioural difficulties. Level 3 incorporates active parent skills training (for example, rehearsal and self-evaluation) and clinical consultations, in addition to the services outlined in level 1 and 2. At this level, parents are provided with four intervention sessions. Thus, the structure of the program is incremental with each level of the program designed to assist parents in increasingly challenging circumstances.

The fourth level of the program offers intensive training in positive parenting skills. This level targets parents with children who possess more severe behavioural difficulties. The fourth level of the program offers 8-10 sessions of individual, group or self-directed parent training. Interventions at this level focus on parent-child interaction and the application of parenting skills to a broad range of situations.

The fifth level of the program is geared towards families with children who exhibit persistent behavioural problems and concurrent family dysfunction (for example, parental depression or relationship difficulty). Strategies at this level are intensive and individually tailored to a family's presentation. The training parents receive includes: parenting skills training; mood and stress management training; and, partner support training.

An enhanced Triple P program (referred to as Pathways Triple P) is available to parents at risk of maltreating their children. This level contains the features of all prior levels and includes modules that target factors associated with child maltreatment. Training modules at this stage of the program include attribution retraining (that is, exploring/challenging the meanings and explanations a parent attributes to their child's behaviours) and interventions specific to anger management.

The developers of the Triple P-Positive Parenting Program designed the tiered approach in order to enhance the program's utility and ensure the program had wide reach in the community. The multilevel approach reflects the fact that children and families present with varying levels of dysfunction and behavioural disturbance, thus interventions need to be flexible in order to address differing family needs.

(Sanders, Markie-Dadds, & Turner,2003)

Note: The Triple-P Positive Parenting Program is tiered primarily based on a child's needs (that is, the degree of behavioural difficulty a child exhibits determines the level of intervention a family receives). Thus the implicit assumption is that the greater a child's needs, the greater the risk of child maltreatment, and therefore the greater the need for parenting skills training. That said, the program's explicit goal is to enhance competent parenting, rather than to reduce the prevalence of child maltreatment.

What other issues need to be considered when implementing parent education programs?

As a result of this synthesis of parent education programs, policy makers and service providers have reason to be optimistic about the benefits of these programs. However, it is important to note that the explicit goal of many parenting programs is to enhance competent parenting, rather than to reduce the prevalence of child maltreatment. It is highly likely that, in enhancing competent parenting, parent education programs will reduce the prevalence of child maltreatment. However, given that most programs do not specifically focus on the reduction of child maltreatment, evaluations have not generally employed direct measures of program influence on the prevalence of child maltreatment. Thus, caution should be exercised when attempting to extrapolate from the present findings to direct outcomes concerning the prevention of child maltreatment, as this outcome cannot be unequivocally assumed.

HOME VISITING

Home visiting programs, both in Australia and internationally, are diverse and provide a broad range of interventions designed to improve family functioning and/or alleviate the potential for child maltreatment.

The majority of home visiting programs are early intervention services aimed at supporting prenatal women or mothers with young children (see Black, Kemp and Sampson 2004 for an evaluation of an Australian program, the Karitane Volunteer Home Visiting Program, which uses an early intervention approach and targets a general, rather than an 'at-risk', population). An early intervention approach reflects a greater awareness of the importance of children's development during the first years of life, the importance of the role of the parents in shaping children's early years, and the subsequent impact of these years on the health and development of the child as s/he ages. With increasing recognition of the importance of early intervention, government policies are moving toward increasing home visiting services as a preventative intervention strategy (Vimpani, Frederico, Barclay, & Davis, 1996).

Home visiting can be defined as 'a mechanism for the delivery of a variety of interventions directed at different outcomes' (Bull, McCormick, Swann, & Mulvihill, 2004 p.1). The interventions may be provided to all mothers and their newborn infants, to parents and children with specific problems, to disadvantaged families, or to families where children are considered to be 'at risk' of child maltreatment.

Home visiting can be incorporated into a primary prevention intervention strategy, while also having the capacity to cater for the needs of 'at risk' families at the secondary or tertiary level (Vimpani, Frederico, Barclay, & Davis, 1996). Family preservation programs may use an intensive home visiting component over a shorter time frame and work with families to avoid children being in placed into care. The programs reviewed in this paper work with families over the longer term, such as from birth until the child is two years old, and home visitors see the family regularly throughout this period. Programs that target 'at risk' families focus on areas such as parenting skills, parental attitudes and knowledge, parent-child interactions, and strengthening social connectedness. While preventing child abuse is not the focus of many of these services, they may contribute indirectly to a reduction in the incidence of child maltreatment by ameliorating the risk factors associated with maltreatment. In addition, home visiting may be able to identify families at a heightened risk for child maltreatment and refer them for additional services (Scott, 2005).

A national audit of child abuse prevention programs in Australia, conducted by Vimpani et al. (1996), found that a home visiting component was incorporated into one quarter of all child abuse prevention programs. The majority of the home visiting services were performed by nonprofessional volunteers who relied on back-up support by professionals. However, in Australia, there have been very few evaluations of the effects of home visiting as a preventative strategy for child abuse and neglect, and very little systematic research into the effects of home visiting on reducing the risk of child maltreatment (Tomison & Poole, 2000; Vimpani, Frederico, Barclay, & Davis, 1996).

Do home visiting programs work?

Home visiting programs have the advantage of bringing services into the home rather than requiring families to seek out services within the community. What makes home visiting services different from other preventative intervention strategies is that such services allow home visitors an opportunity to observe the environment in which families live, identify and tailor services to meet the needs of families, and build relationships in ways that may not be possible with other types of intervention (Gomby, Culross, & Behrman, 1999).

Twelve publications describing home visiting program evaluations were reviewed in the present study, four of which were meta-analyses.

Programs reviewed

The home visiting program evaluations reviewed in the studies sourced for the present paper included:

  • The Nurse Home Visiting Program (USA);
  • The Healthy Start Program (USA);
  • Healthy Families America (USA);
  • The Head Start program (USA);
  • Parents as Teachers (USA);
  • The Home Instruction Program for Preschool Youngsters (USA);
  • The Comprehensive Child Development Program (USA); and
  • Community Child Health Nurse home visiting program for newborns (Australia)

(For a detailed description of each program, see Higgins, Bromfield, Richardson and Higgins 2006).

Most studies reported some degree of effectiveness. The results of the evaluations included:

  • fewer incidents of child maltreatment (when this outcome was directly measured);
  • enhanced parental knowledge and parenting skills;
  • improvements in children's cognitive and social development; and
  • increased linking of parents to health care and other services (Higgins, Bromfield, Richardson, & Higgins, 2006).

The home visiting programs reviewed in this study aimed to improve parenting competence and enhance child development. Programs provided parents with education regarding child development and parenting techniques, as well as practical assistance such as linking families to services and social supports. One program also aimed to improve circumstances for mothers as well as children, by encouraging mothers to defer future pregnancies and return to further education. Interventions were generally directed at low-income families, young mothers and families considered to be 'at risk' for child maltreatment. Six programs explicitly aimed to prevent child maltreatment. Two programs addressed this issue indirectly through improving parenting skills and welfare, which in turn was expected to enhance the welfare of the children.

Of the eight programs reviewed in this study, none was successful in achieving positive results in relation to all program aims. One program (the Nurses Home Visiting Program) was successful in reducing the prevalence of child maltreatment and improving mothers and children's measurement outcomes on health, wellbeing and behavioural variables. A further three programs were successful in improving parenting skills and had some success in reducing incidents of child maltreatment. Two programs, (which were not designed to reduce child maltreatment) had some success in improving parenting knowledge and skills, and two programs were not effective at all (Higgins, Bromfield, Richardson, & Higgins, 2006).

Was the evidence base credible?

In general, the favourable results of a plethora of home visiting evaluations in Australia, Europe and the US have led to widespread acceptance of claims that home visiting prevention strategies are effective in reducing the potential for child abuse. However the positive results derived from these evaluations were based on a range of research designs, some of which do not give reliable results. Most program evaluations did not utilise randomly assigned control or comparison groups, which provide the strongest and most reliable evidence of a program's effectiveness. Many home visiting programs used less rigorous evaluation methods, such as self-reporting, or comparisons between the participant group and those that withdrew from the program. Thus the positive findings drawn from these program evaluations may not be borne out if the programs were subject to more rigorous evaluation methods (Chaffin, 2004; Gomby, Culross, & Behrman, 1999).

In this study, the eight program evaluations that were reviewed used the following research designs:

  • randomised control groups (Community Child Health Nurse home visiting program for newborns, Nurse Home Visiting Program, Healthy Start Program, Parents as Teachers, Healthy Families America);
  • non-randomised control groups (The Comprehensive Child Development Program, Home Instruction Program for Preschool Youngsters, Healthy Families America); and
  • pre and post-test comparisons only (Head Start, Healthy Families America).

Of these designs, the evaluation comprising only a pre and post-test comparison of participant data presents the most significant methodological problems. In the absence of a control group, it is not possible to determine whether the positive findings were a direct result of the program, or whether they would have occurred over time (for example, due to the development of greater insight and maturity or changing circumstances of the participants).

Another significant methodological issue is whether the evaluations used appropriate outcome measures that enabled researchers to test the effectiveness of the program. For example, the evaluations of the Community Child Health Nurse home visiting program for newborns and the Head Start program used inadequate outcome measures. The evaluation of the Community Child Health Nurse home visiting program for newborns - a program which aims to reduce risk factors for, and incidents of, child maltreatment - did not assess the program's impact on the incidence of neglect, an area that evaluations of other programs have indicated show some positive impact. Similarly the evaluation of the Head Start program - a program designed to reduce parents' potential for child abuse - did not contain any child health or wellbeing measures, or measures of child abuse potential such as Milner's Child Abuse Potential Inventory (Milner, 1994). Thus, researchers are unable to determine whether these programs achieved more favourable outcomes for children by reducing the occurrence of child maltreatment.

Many of the studies also had mixed findings in relation to the outcome variables measured. For example, the evaluation of the Head Start program used both quantitative and qualitative evaluation criteria, and only the qualitative data (home visitor reports) indicated an improvement in parenting skills. While this may be the case, it is also possible that home visitors' expectations of improvements may have influenced their beliefs that improvements in parenting skills had taken place.

It may also be difficult to replicate the success of a home visiting program in one community to another geographical area or sub-population, such as applying aspects of successful programs in the US to an Australian context. Where positive findings were found in one area (such as improvements in infant health) or sub-populations, they were not consistently replicated in similar areas or populations in subsequent or similar studies. Further, findings from evaluations of home visiting models indicate inconsistent results, even when their target population and goals were similar. Thus, the findings from one program model cannot be generalised to another (Gomby, Culross, & Behrman, 1999).

What makes a good home visiting program?

The content and style of delivery of home visiting programs are designed to meet specific goals relevant to each program's specific aims and target population. Home visiting programs also differ in how often families are visited, how long they receive the service and the qualifications of staff who implement the service. Home visiting can be a stand-alone intervention, or a component of a broader intervention program that incorporates a range of strategies directed at meeting the program's goals (Bull, McCormick, Swann, & Mulvihill, 2004; Vimpani, Frederico, Barclay, & Davis, 1996).

The key features of successful programs included:

  • programs that targeted an 'at risk' population;
  • programs where services were delivered by more highly trained and qualified home visitors;
  • programs where home visitors were experienced in dealing with the complex needs of many 'at risk' clients;
  • programs of long enough duration to impact upon parenting or risk factors that contribute to child maltreatment;
  • programs that matched program designs to the needs of the client group; and
  • programs that focused on improving both maternal and child outcomes.

Targeting an 'at risk' population

Programs that targeted families that were 'at risk' of child maltreatment and/or were most socioeconomically disadvantaged, were more likely to have positive results than those that targeted a universal population. Programs that targeted an 'at risk' population showed some positive results in improving parenting skills and reducing incidents of child maltreatment. However the programs that showed positive outcomes for 'at risk' populations were not administered to a universal population, so it is unknown whether a universal population would also have positive outcomes if they received the same services.

This finding suggests that in a climate of limited financial resources, targeting home-visiting services toward 'at risk' families would enable funds to be channelled into programs for populations who will benefit the most, rather than allocating funds to a broader section of the public who may not benefit from the service, and who may be more able to independently access services within the community.

Who should deliver home visiting programs?

One of the main issues debated by researchers is whether the positive results of some home visiting programs and not others are due to the use of either professionally trained nurses or 'paraprofessionals' (visitors with no professional qualifications but specific training related to their home visiting role) (Gomby, Culross, & Behrman, 1999). Of the above studies, only the Nurse Home Visiting Program in the USA used qualified nurses to conduct home visits. Programs employing paraprofessionals, such as high school graduates or mothers with specific training in their home visiting responsibilities, tended to be less effective. Olds, Robinson, O'Brien, Luckey, Pettitt, Henderson, Ng, Korfmacher, Hiatt and Talmi (2002) argued that the failure of the other programs to meet their intended aims is their use of paraprofessionals, rather than trained nurses, to deliver the home visiting services. When Olds and colleagues trialled the Nurse Home Visiting Program in a subsequent Denver study using both nurses and paraprofessionals they found that families who were visited by nurses had more positive results than those visited by paraprofessionals (Korfmacher, O'Brien, Hiatt, & Olds, 1999; Olds et al., 2002). Other researchers argue that the relationship between the home visitor and the parent is more central than the education level of the visitor, and that well supported, resourced and trained paraprofessionals can be equally as effective as nurses (Gomby, Culross, & Behrman, 1999). However this assertion has not been empirically tested.

Home visitor experience in addressing program goals with clients

That the Healthy Start Program in the USA did not achieve more positive results may be attributed to an incompatibility between the clients on the one hand, and the services the paraprofessional home visitors were expected to deliver on the other. The home visitors were required to 'build trust' with their clients and provide advice on parenting and other issues. At the same time visitors were expected to identify and address the complex issues 'at risk' families often face such as substance abuse, domestic violence and mental illness - issues that home visitors felt the least competent to address. On the basis of the evidence available it appears that highly trained and experienced home visitors are required to best serve families with multiple and complex needs (Gomby, Culross, & Behrman, 1999).

Appropriate duration of program

As effective programs often work with clients who have multiple and complex needs, it is important that the program be of long enough duration to impact on the factors that contribute to child maltreatment. The significance of duration is evidenced by the success of programs that have included home visiting in both the antenatal and postnatal periods. This finding highlights the importance of strategies to ensure client participation for the duration of the program, such as matching parental needs to program design and home visitor skills.

Matching the program design to the client group

Another important finding from these evaluations was the importance of matching the program's design and implementation with the program's goals and target population in order for the program to achieve positive results. The Healthy Start Program, for example, was designed to reduce child maltreatment by targeting an 'at risk' population (Gomby, Culross, & Behrman, 1999). The program used paraprofessionals to deliver an 'empowerment' intervention approach, which relied on clients to identify their own risk factors and needs accurately in order to receive appropriate assistance. The program had only minimal success in achieving its desired goals. This finding raises questions as to whether the program's expectations of families challenged by multiple and complex issues were too high, and whether the families may have been better served by a formal assessment of their needs to identify and target appropriate interventions, a finding also reached by other researchers working with families with complex needs (DePanfilis & Wilson, 1996).

Improving maternal and child outcomes

Home visiting programs were also more likely to be successful where the focus was on improving both maternal and child wellbeing. Home visiting programs that attempted to improve the mothers' life chances as well as reduce the risk of child maltreatment showed improvements for both mothers and children.

Summary

The programs that were most successful were those where home visitors were more highly trained and qualified, such as nurses, rather than paraprofessionals. Results indicate that home visitors need to have the necessary skills, experience and training to address the complex issues that many socioeconomically and 'at risk' families face, such as mental health issues, substance abuse and domestic violence.

Home visiting programs were also more likely to be successful when targeted towards a specific client group and when the focus was on improving both maternal and child wellbeing. Home visiting programs that addressed improving the mothers' life chances as well as reducing the risk of child maltreatment, showed improvements for both mothers and their children.

The most convincing evidence of the potential for home visiting to reduce child maltreatment comes from results from the Nurse Home Visiting Program (below).

A successful home visiting program: The Nurse Home Visiting Program (USA)

Program aims and program design. The Nurse Home Visitation Program (NHVP) was designed to improve maternal health-related behaviours during and after pregnancy (Olds, Henderson, Chamberlin, & Tatelbaum, 1986a). Program designers believed that improvements to maternal health and behaviour would lead to improvements in children's health and wellbeing and also reduce incidents of child maltreatment.

The program targeted families from socioeconomically disadvantaged communities. The program was initially delivered by professionally trained nurses who worked with a cohort of 400 expectant, first-time mothers residing in the county of Elmira, in New York State (The Elmira study). At the time of the study, Elmira was the poorest county in the US with the highest rates of child maltreatment reports and substantiations in the US (Olds et al., 1999). The study was later replicated in Memphis, Tennessee (The Memphis study) and then in Denver, Colorado.

Program structure/design. Nurses visited families from weekly to monthly depending on the program's protocols and the availability of the mothers, from the second trimester of pregnancy until the oldest child was 24 months old. The program, guided by the principles of attachment theory, assisted mothers to develop responsive, engaged and sensitive care giving in the early years of their child's life.

Nurses helped mothers to identify how they were parented themselves, and how they would like to parent their children. Nurses aimed to develop trusting, empathic relationships with mothers and other family members (Olds et al., 1999). The Elmira mothers also received education in areas such as: the impact of maternal behaviour (such as dieting, smoking, alcohol and drug use) on the baby during pregnancy, infant health and wellbeing (such as teaching mothers to understand and interpret their baby's behaviour), childhood development and effective parenting strategies.

In addition, the program also aimed to modify known risk factors associated with child abuse and neglect by linking parents with social supports and appropriate services. Risk factors included low education and unemployment, multiple births, marital conflict, and isolation. Nurses also assisted mothers to develop realistic personal goals and achievable objectives, and provided mothers with information on contraception, childcare options and career choices. Thus, the program focused on improving both maternal and child wellbeing.

What other issues need to be considered when implementing home visiting programs?

Results also highlight the importance of careful consideration of the program's design and implementation, carefully matched to the goals of the program and the program's target population. Given that many programs do not seem to have the capacity to meet the needs of clients with complex issues, programs may have to reskill workers and focus on known risk factors when targeting 'at risk' families (Chaffin, 2004). Further research is necessary in order to identify which changes actually reduce the risk of maltreatment so that programs can focus their interventions accordingly.

Limitations of parent education and home visiting evaluations

The results of both parent education and home visiting program evaluations were largely supportive of their effectiveness. However, a number of studies were subject to similar methodological limitations. Many studies had small sample sizes and experienced high attrition rates, particularly those with longitudinal designs which measured parent/child outcomes over an extended period. Thus, it was difficult for researchers to document whether short-term program benefits were maintained. This finding is consistent with previous observations that rigorous experimental conditions are often difficult to obtain given the realities of providing preventative services to families at risk of child maltreatment (Tomison, 2000).

There also appeared to be an over-reliance on self-report measures as evidence of program effectiveness (for example, parents reporting attitudinal change such as a decrease in the acceptability of using physical punishment or increased knowledge of child development). It is unknown whether changes in parental attitude or increases in parental knowledge actually translate to fewer incidents of child maltreatment. Thus future program evaluations could consider incorporating more direct measures of child wellbeing and/or child maltreatment (for example, statutory child protection service data and child medical records) in order to establish with greater confidence, the utility of prevention programs.

Finally, in many of the articles reviewed while compiling this paper, it was unclear who had been involved in conducting the evaluation/s. That is, it was unclear whether an independent body had been enlisted to evaluate the effectiveness of a particular program or whether someone affiliated with the program had devised and conducted the evaluation. To avoid any accusation of bias, it is desirable for evaluations to be conducted independently.

Limitations of the present study

It should be noted that the literature search conducted for the purpose of this paper, extended to published articles only. MacLeod and Nelson (2000) noted that literature searches, which only retrieve published articles, are more likely to generate favourable results. Thus there is a possibility that the findings of this paper are subject to a publication bias. Therefore the present findings concerning the effectiveness of parent education and home visiting programs may appear more favourable than they are in reality.

Although a number of articles were retrieved according to the search criteria outlined earlier, the majority did not qualify as rigorous impact or outcome evaluations. Thus, the evaluations summarised in this report represent the minority of articles sourced through the present literature search. This finding supports previous observations that, despite the increasingly extensive use of primary and secondary preventative interventions, many have not yet been subject to rigorous evaluation (Chaffin, Bonner, & Hill, 2001; Tomison, 2000). In addition, the majority of program evaluations were sourced from international literature. There were limited published evaluations of Australian programs, thus further research is required to determine the applicability of international prevention programs to the Australian context.

Typically, in the public health model of disease prevention, impact and outcome evaluations incorporate a measure of 'effect size'. The term effect size refers to a statistical measure of the extent to which an intervention has impacted upon participants. The statistical measure of effect size (for example, Cohen's d) ranges between zero and one, where an effect size of 0.2 is small; 0.5 is medium; and 0.8 is large. By calculating effect sizes we can compare how effective a program is in relation to other programs. However, given that there is a paucity of quantitative research evaluating child maltreatment prevention programs (in part, due to the hidden nature of child maltreatment and the difficulty of conducting research in this area) there is insufficient data available for which effect sizes could be calculated (for further details concerning the computation and limitations of effect sizes, see http://web.uccs.edu/lbecker/Psy590/es.htm

Implications

The present findings have implications for the direction of future parent education and home visiting programs. Specifically, on the basis of the literature reviewed, policy makers and practitioners have reason to be optimistic about the benefits of parent education and home visiting programs. This report demonstrated that both modes of intervention may reduce the incidence of child abuse and neglect.

The majority of evaluations documented favourable program results, however, some program aims (for example, the Triple P-Positive parenting Program, which aims to decrease children's problem behaviour by enhancing competent parenting) did not relate specifically to child maltreatment prevention. Accordingly, the measurement variables used in these evaluations concerned the promotion of competent parenting. Thus, further research (for example, an outcome evaluation) is required to explore the precise relationship between the promotion of competent parenting and the prevention of child maltreatment8.

Outcome evaluations, which directly demonstrate a program's effectiveness in reducing the prevalence of child maltreatment, will enable policy makers and funding bodies to make informed choices about program funding. Outcome evaluations will also enable comparisons between the effectiveness of particular programs in preventing, reducing and/or repairing the after effects of child maltreatment with the purported costs of child abuse and neglect (for example, the cost of providing tertiary services and the future lost productivity of severely abused children) (James, 1994). Comparisons of this nature could provide further funding impetus.

Further research is also needed to investigate the cost/benefit of prevention programs (particularly the more resource intense interventions, such as long-term home visiting programs) relative to other program types. Arguably, if evaluations demonstrate that less resource intense programs achieve results equivalent to more resource intense programs, then further attention should be given to determining which options are most cost effective.

For child protection practice, this paper illustrated that there is variation in the degree of effectiveness of parent education and home visiting programs. However, factors found to enhance program success were identified, including: targeted recruitment of 'at risk' families; program goals matched to the needs of families requiring the service; the use of a combination of strategies/interventions; and intense/lengthy program involvement. These findings are particularly relevant to service providers and/or policy makers who intend to examine the effectiveness of a prevention program that they provide to 'at risk' families. As a starting point to a more extensive and rigorous evaluation, service providers could begin by investigating whether their program contains these attributes.

Although the majority of programs reviewed in this paper were secondary programs targeted at 'at risk' families, it was not the intention of this paper to advocate for the use of secondary interventions at the expense of universal or tertiary approaches. Primary, secondary and tertiary interventions are vital for a comprehensive and integrated approach to the prevention of child maltreatment. For example, the universal maternal and child health services that exist in some form in every state and territory provide basic paediatric health assessments for pre-school aged children (that is, a primary intervention). These services may also be used as a means of identifying those families who have need for secondary level intervention and enable families to access these services without experiencing any stigma. Universal maternal and child health services may also act as a source of referrals to tertiary interventions for families where the maternal and child health nurse suspects that maltreatment has occurred.

In a keynote address, Professor Dorothy Scott (2005) argued that 'universal maternal and child health services are excellent platforms from which to deliver primary, secondary, and tertiary child protection prevention strategies'. The potential benefits of a comprehensive and integrated child abuse prevention platform include:

  • cost-benefits, as families receive appropriate interventions to meet their needs (that is, they are not over or under-serviced);
  • enhanced secondary service utilisation through non-stigmatising entry to secondary prevention services (if all families are accessing some form of intervention, it will not be apparent to others whether they are accessing different levels [primary or secondary] of intervention); and
  • ultimately a reduction in the demand for tertiary interventions.

The relatively high rate of participant attrition was recognised as a limitation of many of the evaluations reviewed in the present paper. However, if the program evaluation attrition rate is representative of the typical rate of program attrition, then further investigation is required into how service providers engage and maintain parent/family involvement in prevention programs. Arguably, the parents/families who are least able to maintain their involvement in a prevention program evaluation (for example, because of repeated changes in home address making it difficult for evaluators to make contact with the family) are the parents/families most in need of the interventions. Thus, this finding suggests that not only do service providers need to investigate how effective their program is in preventing child maltreatment, but also how effective their program is in engaging and maintaining the involvement of those families at greatest risk.

This paper has provided strong evidence of the need for program evaluation. For example, given the increasing demand for 'evidence-informed' practice and the limited pool of funds available for child welfare programs, rigorous evaluations enable service providers to systematically document the benefits of their program and justify the continuation of the services they provide. In addition, investigating whether prevention programs actually work (through outcome evaluations) should ensure that the prevention programs that are implemented positively influence the lives of families they service.

Conclusion

The purpose of this paper was to investigate evaluations of child maltreatment prevention programs. Particular attention was given to parent education and home visiting programs, as they are two examples of early preventative interventions widely practiced within the child protection field.

Although no one program is effective in eliminating all incidents of child maltreatment, attributes of effective parent education programs and home visiting programs were highlighted. The most successful parent education programs contained targeted recruitment; a structured and lengthy program; a combination of interventions/strategies; and a strengths-based approach. The most successful home-visiting programs were delivered by highly trained professionals (for example, nurses); contained targeted recruitment strategies; programs goals that matched client needs; and were designed to improve both maternal and child wellbeing.

In conclusion, it was shown that both parent education and home visiting programs can improve parents' knowledge, skills and supports and may be effective in preventing child abuse and neglect. However, parent education and home visiting programs should be seen as part of a comprehensive approach to child maltreatment prevention that includes primary, secondary and tertiary interventions. Arguably, the most effective service provision (in terms of program utilisation, participant outcomes and cost effectiveness) targets the 'right' intervention to the 'right' audience. In fact, the promulgation of primary and secondary approaches in the prevention of child maltreatment, may help to alleviate the strain placed on an already over burdened tertiary child protection system (Scott, 2005).

Note: This paper is based on research that was commissioned by the Australian Government Department of Families, Community Services and Indigenous Affairs on the efficacy of child maltreatment prevention programs. Other information and resources developed from this research are available on the National Child Protection Clearinghouse website, visit <www.aifs.gov.au>

References

  • Australian Childhood Foundation. (2004). Every child is important: A booklet for parents. Melbourne, Vic: Australian Childhood Foundation.
  • Australian Institute of Health and Welfare. (2001). Child protection Australia 1999-00 (No. 27). Canberra, ACT: Australian Institute of Health and Welfare.
  • Australian Institute of Health and Welfare. (2006). Child protection Australia 2004-2005 (No. 38). Canberra, ACT: Australian Institute of Health and Welfare.
  • Black, M., Kemp, L., & Sampson, V. (2004). Karitane Volunteer Home Visiting Program Evaluation, from http://chetre.med.unsw.edu.au/files/Karitane%20Final%20Report%20for%20printing.pdf
  • Britner, P. A., & Reppucci, N. D. (1997). Prevention of child maltreatment: Evaluation of a parent education program for teen mothers. Journal of Child and Family Studies, 6(2), 165-175.
  • Bromfield, L. M., & Higgins, D. J. (2004). The limitations of using statutory child protection data for research into child maltreatment. Australian Social Work, 57(1), 19-30.
  • Bromfield, L. M., & Higgins, D. J. (2005). National comparison of child protection systems. Child Abuse Prevention Issues, no.22, 1-21, from www.aifs.gov.au/nch/issues/issues22.html
  • Bugental, D. B., Ellerson, P. C., Lin, E. K., Rainey, B., Kokotovic, A., & O'Hara, N. (2002). A cognitive approach to child abuse prevention. Journal of Family Psychology, 16(3), 243-258.
  • Bull, J., McCormick, G., Swann, C., & Mulvihill, C. (2004). Ante- and post-natal home visiting programs: A review of reviews. Evidence briefing (1st edition - February), from <http://www.hda.nhs.uk/documents/home_visiting.pdf>
  • Chaffin, M. (2004). Is it time to rethink Healthy Start/Healthy Families? Child Abuse & Neglect, 28(6), 589-595.
  • Chaffin, M., Bonner, B. L., & Hill, R. F. (2001). Family preservation and family support programs: Child maltreatment outcomes across client risk levels and program types. Child Abuse & Neglect, 25 (10), 1269-1289.
  • DePanfilis, D., & Wilson, C. (1996). Child protective services: Applying the strengths perspective with maltreating families. The American Professional Society on the Abuse of Children, 9(3), 15-20.
  • Doll, L. S., Koenig, L. J., & Purcell, D. W. (2004). Child sexual abuse and adult sexual risk: Where are we now? In L. J. Koenig, L. S. Doll, A. O'Leary & W. Pequegnat (Eds.), From child sexual abuse to adult sexual risk: Trauma, revictimization, and intervention (pp. 3-10). Washington, DC: American Psychological Association.
  • Geeraert, L., Noortgate, W. V. D., Grietens, H., & Onghena, P. (2004). The effects of early prevention programs for families with young children at risk for physical child abuse and neglect: A meta-analysis. Child Maltreatment, 9(3), 277-291.
  • Gomby, D. S., Culross, P. L., & Behrman, R. E. (1999). Home visiting: Recent program evaluations-analysis and recommendations. The Future of Children, 9(1), 4-26.
  • Hebert, M., Lavoie, F., & Parent, N. (2002). An assessment of outcomes following parents' participation in a child abuse prevention program. Violence and Victims, 17(3), 355-372.
  • Higgins, J., Bromfield, L. M., Richardson, N., & Higgins, D. J. (2006). Child abuse prevention: What works? The effectiveness of home visiting programs for preventing child maltreatment. Research Brief no.2, Online www.aifs.gov.au/nch/pubs/researchbrief/rb2.html
  • Holzer, P. J., Bromfield, L. M., Richardson, N., & Higgins, D. J. (2006). Child abuse prevention: What works? The effectiveness of parent education programs for preventing child maltreatment. Research Brief no.1, Online www.aifs.gov.au/nch/pubs/researchbrief/rb1.html
  • James, M. (1994). Child abuse and neglect: Incidence and prevention. Issues in Child Abuse Prevention no.1, Online <http://www.aifs.gov.au/nch/pubs/issues/issues1/issues1.html>
  • Kelly, R. F. (2000). Family preservation and reunification programs in child protection cases: Effectiveness, best practices, and implications for legal representation, judicial practice, and public policy. Family Law Quarterly, 34(3), 359-391.
  • Korfmacher, J., O'Brien, R., Hiatt, S., & Olds, D. (1999). Differences in program implementation between nurses and paraprofessionals in prenatal and infancy home visitation: A randomised trial. American Journal of Public Health, 89(12), 1847-1851.
  • MacLeod, J., & Nelson, G. (2000). Programs for the promotion of family wellness and the prevention of child maltreatment: A meta-analytic review. Child Abuse & Neglect, 24(9), 1127-1149.
  • Matthews, J., & Hudson, A. (1998). Evaluation of parent education programs. Victorian Parenting Centre News, 1(2), 4-5.
  • Milner, J. S. (1994). Assessing physical child abuse risk: The child abuse potential inventory. Clinical Psychology Review, 14(6), 547-583.
  • Olds, D., Henderson, C. R., Chamberlin, R., & Tatelbaum, R. C. (1986a). Preventing child abuse and neglect: A randomised trial of nurse intervention. Paediatrics, 78, 1238-1244.
  • Olds, D., Henderson, C. R., Kitzman, H. J., Eckenrode, J. J., Cole, R. E., & Tatelbaum, R. C. (1999). Prenatal and infancy home visitation by nurses: Recent findings. The Future of Children, 9(1), 44-65.
  • Olds, D., Robinson, J., O'Brien, R., Luckey, D., Pettitt, L. M., Henderson, C. R., et al. (2002). Home visiting by nurses and by paraprofessionals: A randomised controlled trial. Paediatrics, 110(3), 486-496.
  • Pecora, P. J., Fraser, M. W., & Haapala, D. A. (1992). Intensive home-based family preservation services: An update from the FIT project. Child Welfare 71(2), 177-188.
  • Peterson, L., Tremblay, G., Ewigman, B., & Saldana, L. (2003). Multilevel selected primary prevention of child maltreatment. Journal of Consulting and Clinical Psychology, 71(3), 601-612.
  • Prentice, K. (2002). To study innovative models of childhood sexual abuse: Primary, secondary and tertiary prevention programmes. Bundaberg, QLD: Bundaberg Area Sexual Assault Service.
  • Richardson, N., & Tomison, A. M. (2004). Evaluating child abuse prevention programs. Child Abuse Prevention Resource Sheet no.5, Onlinewww.aifs.gov.au/nch/sheets/rs5.html
  • Rossi, P. H. (1997). Program outcomes: Conceptual and measurement issues. In E. J. Mullen & J. L. Magnabosco (Eds.), Outcomes measurement in the human services: Cross cutting issues and methods (pp. 20-34). Washington, DC: National Association of Social Workers Press.
  • Sanders, M., & Cann, W. (2002). Promoting positive parenting as an abuse prevention strategy. In K. D. Browne, H. Hanks, P. Stratton & C. Hamilton (Eds.), Early prediction and prevention of child abuse: A handbook (pp. 145-163). West Sussex, England: John Wiley & Sons.
  • Sanders, M., Markie-Dadds, C., Tully, L. A., & Bor, W. (2000). The Triple P-Positive Parenting Program: A comparison of enhanced, standard, and self-directed behavioural family intervention for parents of children with early onset conduct problems. Journal of Consulting and Clinical Psychology, 68(4), 624-640.
  • Sanders, M., Markie-Dadds, C., & Turner, K. M. T. (2003). Theoretical, scientific and clinical foundations of the Triple P-Positive Parenting Program: A population approach to the promotion of parenting competence (No. 1). St Lucia, QLD: The University of Queensland.
  • Scott, D. (2005). Towards a national child protection research agenda and its translation into policy and practice. Seminar presentation at the Australian Institute of Family Studies.
  • Streiner, D. L. (2002). The 2 'Es' of research: Efficacy and effectiveness trials. Canadian Journal of Psychiatry, 47(6), 552-556.
  • Thyen, U., Thiessen, R., & Heinsohn-Krug, M. (1995). Secondary prevention: Serving families at risk. Child Abuse & Neglect, 19(11), 1337-1347.
  • Tomison, A. M. (1998a). Child protection towards 2000 revisited: Trends in child protection and child abuse prevention. Changing families, challenging futures: 6th Australian Institute of Family Studies Conference, Online <www.aifs.gov.au/institute/afrc6papers/tomison.html>
  • Tomison, A. M. (1998b, Spring 1998). Valuing parent education: A cornerstone of child abuse prevention. Child Abuse Prevention Issues no.10, Online www.aifs.gov.au/nch/issues10.html
  • Tomison, A. M. (2000). Evaluating child abuse prevention programs. Child Abuse Prevention Issues no.12, Online www.aifs.gov.au/nch/issues12.html
  • Tomison, A. M., & Poole, L. (2000). Preventing child abuse and neglect: Findings from an Australian audit of prevention programs, Online ,www.aifs.gov.au/nch/pubs/auditreport.html.
  • Vimpani, G., Frederico, M., Barclay, L., & Davis, C. (1996). An audit of home visitor programs and the development of an evaluation framework. Canberra, ACT: AGPS.

Notes

  1. Although fluctuations in child protection service data could reflect policy changes over this period, the difference is substantial, and may reflect an increase in the underlying incidence of reported child maltreatment (for a description of the child protection systems in each jurisdiction, see Bromfield and Higgins 2005; and for a discussion of the limitations of child protection service data, see Bromfield and Higgins 2004).
  2. Parent education programs are systematic interventions designed to impart information, awareness, and skills in relation to different aspects of parenting at various stages of child development.
  3. Home visiting programs are designed to provide information and assistance to women during pregnancy and after birth. Home visiting programs tend to concentrate on issues concerning health, nutrition, child wellbeing, and mother/child interaction. Home visiting services are provided in a mother's own home.
  4. Although many child maltreatment prevention programs are currently in operation in Australia, few appeared to have been subject to rigorous evaluation (for an overview of prevention programs in Australia, see Tomison and Poole 2000). Thus, the results of this paper are largely based on published evaluations of international child maltreatment prevention programs.
  5. The term 'at risk' is used in this paper to mean families who exhibit risk factors for child maltreatment. Risk factors for maltreatment include, poverty, parental mental health problems, marital discord, family violence, and parental drug and alcohol use. However, the presence of these 'risk factors' does not necessarily mean that a child will experience maltreatment. With few exceptions, the majority of studies reviewed in this report used the term 'at risk' to mean the capacity or likelihood of abuse and/or neglect occurring.
  6. Efficacy trials investigate whether an intervention works under ideal circumstances, while effectiveness studies investigate whether an intervention works when applied in the 'real world' (Streiner, 2002). Streiner conceptualised efficacy and effectiveness studies on a continuum, with most research designs falling somewhere in between. Consistent with Streiner's observation, most studies reviewed in the present paper were not 'pure' effectiveness or efficacy trials, but fell somewhere in between. For consistency the term 'effectiveness' will be used throughout this paper.
  7. Researchers have used different terminology to define different types of evaluations. For example, some researchers have used the terms 'process, impact and outcome' evaluations (Tomison & Richardson, 2004), while other researchers have used the terms 'input, process and product' evaluations (Matthews & Hudson, 1998) to mean essentially the same thing. Of note, the terms 'impact' and 'outcome' are often used interchangeably in the program evaluation literature. However, a clear distinction between 'impact' and 'outcome' evaluations has been drawn throughout this paper.
  8. An extensive population outcome evaluation of the Triple P Program is currently underway in the US. The five-year research project, which commenced in 2003, aims to evaluate the effectiveness of the Triple P Program in reducing the prevalence of child maltreatment at the population level. As the 'Building Connections' trial has yet to be completed, outcome data is not yet available. However, further information on the aims and methodology of the 'Building Connections' trial can be accessed at http://www.pfsc.uq.edu.au/01_about_us/International_research.html.
Acknowledgements

Prue J. Holzer is a Research Officer with the National Child Protection Clearinghouse at the Australian Institute of Family Studies.

Jenny Higgins is a Research Officer with the National Child Protection Clearinghouse at the Australian Institute of Family Studies.

Leah Bromfield is a Research Fellow and Assistant Manager of the National Child Protection Clearinghouse at the Australian Institute of Family Studies.

Nick Richardson is a Senior Research Officer with the Australian Institute of Family Studies (formerly with the National Child Protection Clearinghouse).

Daryl Higgins is a Senior Research Fellow and General Manager of Research at the Australian Institute of Family Studies.

The authors thank Dr Fiona Arney and Dr Joe Tucci for their valuable suggestions and feedback on this paper.

ISBN

0 642 39536 5

Share