A meta-analysis of intensive family preservation programs: Placement prevention and improvement of family functioning

Abstract

The aims of the present study were, first, to establish the effect of brief, in-home intensive family preservation programs on prevention of out-of-home placement, family functioning, child behavior problems and social support and, second, to study moderators of these effects. The results of this meta-analytic study, consisting of 20 studies (31,369 participants), show that intensive family preservation programs had a medium and positive effect on family functioning (d = .486), but were generally not effective in preventing out-of-home placement. Intensive family preservation programs were effective in preventing placement for multi-problem families, but not for families experiencing abuse and neglect. Moreover, the effect on out-of-home placement proved to be moderated by client characteristics (sex and age of the child, parent age, number of children in the family, single-parenthood, non-white ethnicity), program characteristics (caseload), study characteristics (study design and study quality), and publication characteristics (publication type, publication year and journal impact factor). The discussion addresses implications for evaluation and practice.

Introduction

Since the 1970s intensive family preservation programs are widely used for families in crisis experiencing imminent risk for out-of-home placement of a child (e.g. Al et al., 2011, Lindsey et al., 2002). The primary aim of these programs is preventing out-of-home placement. In order to do so, the programs focus on ending the crisis, improving family functioning and promoting the use of social support systems (e.g. Kinney, Haapala, & Booth, 1991). Although intensive family preservation programs carry different names, most programs are built on the Homebuilders model that was developed in Washington in 1974 (Kinney, Madsen, Fleming, & Haapala, 1977). Important characteristics of the Homebuilders model are: a quick start of the intervention (within 24 h after referral), small caseloads of social workers and short duration (4–6 weeks). The intervention is intensive and flexible and offers both therapeutic services, for example, training new parenting skills, and concrete services, such as organizing financial support (Berry, 1997, Kinney et al., 1991, Ryan and Schuerman, 2004, Tully, 2008).

Intensive family preservation programs are largely grounded in crisis theory (e.g. Caplan, 1964, Rapoport, 1962). During a crisis, induced by a sudden disturbance of balance (Golan, 1987), family functioning is seriously disturbed and the families' usual coping mechanisms and social support systems are insufficient (e.g. Caplan, 1964). Crisis intervention aims to end the crisis and to provide the family with new forms of coping that diminishes the chance of a new crisis (Rapoport, 1970).

A few therapeutic approaches are common in interventions for families in crisis. The intervention focuses on the whole family in line with the system approach, which assumes that the behavior of individual family members can only be understood from the perspective of family interactions that influence system balance (Watzlawick, Beavin, & Jackson, 1967). In addition, the intervention uses a network approach, taking into account that the family is an open system, which is influenced by, for example, the school and the neighborhood (Bronfenbrenner, 1979). Other approaches adopted by intensive family preservation programs are the (empowering) competence approach and the solution focused approach. The competence approach is aimed at empowerment and fostering skills and strengths of clients (e.g. Graves and Shelton, 2007, Masterpasqua, 1989). The solution-focused approach, in line with the latter, considers the client as the major source of solutions and is aimed at setting goals that are self-concordant and maximize the use of the client's competencies (De Shazer and Berg, 1997, Gingerich et al., 2011).

Family preservation gained popularity after introduction of the attachment theory by Bowlby in 1969. As separating children from their biological parents was thought to cause attachment problems in children (Bowlby, 1969), in-home intervention to improve family functioning became preferred over out-of-home placement, promoting cost-effectiveness as well (Lindsey et al., 2002). Despite the widespread confidence in intensive family preservation programs as the good alternative for out-of-home placement, the positive effects are far from evident. After the introduction of these interventions, many positive results were presented. Evaluation studies reported successful prevention of out-of-home placement, from 71% up to 93% prevention rates (Berry, 1992, Pecora et al., 1987, Reid et al., 1988). However, the positive results were mainly found in studies that did not use control groups, and therefore no conclusions on effectiveness could be drawn (e.g. Lindsey et al., 2002).

In order to establish the effectiveness of intensive family preservation programs, several narrative reviews (Blythe et al., 1994, Fraser et al., 1997, Lindsey et al., 2002, Littell and Schuerman, 1995, Tully, 2008) and two meta-analyses (Dagenais et al., 2004, Miller, 2006) were completed, which all showed mixed results with respect to out-of-home placement. Some promising results concerning improvement of family functioning were presented, however, particularly in uncontrolled studies. Miller (2006) conducted a selective meta-analysis of intensive family preservation programs delivered in Washington State and concluded that only programs that adhere to the characteristics of the Homebuilders model were effective in preventing out-of-home placement and improving child and family functioning.

Not only the mixed results, but also a variety in target group, study design and outcome measures characterize the crisis intervention literature, which makes it difficult to draw definite conclusions about effectiveness. Intensive family preservation programs serve, for example, families experiencing abuse and neglect (Fernandez, 2004), families with substance abuse of parents (Forrester, Copello, Waissbein, & Pokhrel, 2008) and multi-problem families. Prevention of out-of-home placement has been the most often selected outcome measure, but many have argued that out-of-home placement should not be the sole outcome measure in evaluation studies, and that other outcome measures should be included too, such as family functioning (e.g. McCroskey and Meezan, 1997, Rossi, 1992, Thieman and Dail, 1992, Tully, 2008).

Although the Homebuilders model, and family crisis intervention in a broader sense, is used for over forty years now, it is not possible to draw any conclusions about the effectiveness of family crisis interventions on the basis of the available meta-analyses and the reviews. Moreover, outcome measures other than prevention of out-of-home placement have been minimally addressed. Dagenais et al. (2004), in their meta-analysis of family crisis intervention, concluded that program impact on family functioning seems promising. This conclusion, however, was based on a qualitative analysis of differences between effect sizes of single evaluation studies instead of a quantitative analysis of overall mean-effect sizes and a test of moderators that may have an impact on effectiveness of family crisis intervention. It has, therefore, not been established what the overall effect of intensive family preservation programs is, and which factors moderate intervention effects. Examining moderators is important in order to be able to explain the mixed results that have been presented in the literature so far. Identifying factors that may account for the effectiveness of family crisis intervention may help tailoring interventions better to the needs of families that are targeted. Moderator analyses, for example, may help identifying certain subgroups of clients that profit less or more of the intervention than others or certain program characteristics that especially contribute to therapeutic change.

The present meta-analytic study of controlled family preservation studies aims to address the effectiveness of intensive family preservation programs in terms of prevention of out-of-home placement, improved family functioning, social support and reduced child behavior problems by calculating the overall mean-effect sizes of these outcome measures. Additionally, potential moderators of the effects are examined. Client characteristics (child age, parent age, problem type, risk for placement, number of children in the family, and percentages of boys, non-white ethnicity and single parent families), program characteristics (duration, caseload and adherence to Homebuilders), study design characteristics (prospective/retrospective study design, follow-up time, study quality and randomization), and publication characteristics (publication type, publication year and journal impact factor) are addressed.