Current Status and Evidence Base of Training for Foster and Treatment Foster Parents

Child welfare services in the United States serve over 500,000 youth in out-of-home care at any point in time (AFCARS, 2005). The majority of these youth are served via some form of foster care. Youth who enter foster care are likely to have experienced some form of child abuse or neglect and/or other traumatic events (e.g., domestic violence, impaired caregiver, traumatic loss) (Hazen, Connelly, Kelleher, Landsverk, & Barth, 2004; Henry, Cassett, Auletta, & Egan, 1991). It is well known that youth in foster care display a wide range and depth of behavioral, developmental, social, and educational problems (e.g., Farmer et al. 2001; Clausen, Landsverk, Ganger, Chadwick, & Litrownik, 1998; Sawyer & Dubowitz, 1994; Smucker, Kauffman, & Ball, 1996).

Given the severity of foster children's trauma histories and resulting difficulties, it seems apparent that individuals who are becoming foster parents should receive solid training in the range of domains that will be necessary to successfully ‘parent’ these children during the time they reside in foster care. Such a position has a long history of wide endorsement and support (Christiansen & Fine, 1979; Galaway, Nutter, & Hudson, 1995; Ruff, Blank, & Barnett, 1990; Runyan & Fullerton, 1981). A recent study using data from Caring for Children in Child Welfare, a supplemental study to the National Survey of Child and Adolescent Well-being (NSCAW) designed to evaluate the impact of child welfare, Medicaid, state Child Health Insurance Plans (CHIP) programs and policies on service use indicates that there is tremendous variation in the training that foster parents receive (Hurlburt, Leslie, Barth, & Landsverk, in press).

In a recent companion article (Barth, et al., 2005), we reviewed the evidence base for parent-training interventions that are currently being used or have potential to be used with families who come into contact with child welfare agencies because of allegations of abuse/neglect. This review identified a range of parent training approaches with varying levels of ‘evidence’ (Chambless & Ollendick, 2001; Hoagwood, 2003; Kratochwill & Stoiber, 2002). This review also pointed out the strong disjuncture between training for ‘biological’ parents and ‘substitute’ parents. Few of the identified approaches in the parent-training review had any evidence base with foster or substitute parents. Only in the last three years have any of the identified approaches (i.e., Parent-Child Interaction Therapy, The Incredible Years) been the subject of empirical investigation with a focus on foster parents (Linares, Montalto, Li, & Oza, 2006; McNeil, Herschell, Gurwitch, & Clemens-Mowrer, 2005). Aside from these recent investigations, a search of training programs for foster parents identified a distinct set of training programs developed specifically for foster parents. Therefore, the current review and evaluation examines training specifically for foster parents.

Dimensions for Examining Foster Parent Training

The initial review of parent training focused specifically on interventions designed to help parents involved with child welfare develop increased competence in parenting maltreated children. This required efforts to understand how parent training fit within the child welfare services context, which includes consideration of information about case management and legal processes, as well as the usual training on dyadic care for children. As such, it quickly became evident that a review of training programs for foster parents would also need to be broad. An evaluation of training for foster parents and treatment foster parents must include the relevant dimensions that potentially influence the development and implementation of such training. For the current review, we included legal requirements, recommendations and practice parameters developed by relevant organizations, as well as empirical support for the potential training protocols.

A Note about Foster Care and Treatment Foster Care

Foster care, as traditionally viewed, is an element of child welfare services that involves placement of a child in a substitute home environment when the child’s parents are unable or unwilling to provide appropriate care. Foster care is intended to be a time-limited placement on the way to determining one of the following three permanency plan options: reunification with the biological parent, conversion of the foster home to a legally-permanent guardianship or adoption, or placement of the child into another legally permanent family. Foster parents, in this model, are viewed as parent-substitutes. They provide for the child’s physical, emotional, developmental, and social needs during the period the youth resides with them. While it is recognized that youth in foster care display a variety of behavioral, emotional, developmental, and social difficulties that foster parents will confront, foster parents have not traditionally been viewed as directly responsible for addressing or ameliorating these problems. They are only reimbursed for the board and care they provide and there is often no clear expectation that they will participate in any form of mental health interventions with children in their care (Cain & Barth, 1990).

Treatment Foster Care (TFC), in contrast, has developed explicitly as a treatment-oriented approach for youth with problems. In this model foster parents (sometimes referred to as treatment parents, therapeutic parents, professional parents, etc.) are seen as front-line therapeutic agents who are responsible for working with other professionals in the youth’s life to develop and implement a comprehensive treatment plan (Chamberlain, 1994, 2002; Meadowcroft, Thomlison, & Chamberlain, 1994). Therefore, in addition to meeting the physical, emotional, developmental, and social needs of a youth, treatment foster parents are expected to employ strategies designed to decrease problematic behaviors and develop prosocial behaviors. Thus, they typically receive additional compensation, support, training, and ongoing support in order to accomplish these goals (Chamberlain & Mihalic, 1998; Pecora & Maluccio, 2000).

The use of the term ‘foster care’ for both of these types of out-of-home placement is both appropriate and confusing. On one hand, both types of placement are intended to be relatively short-term placements that lead to permanency. That is, in both cases, foster parents (or treatment foster parents) bring a youth into their home and their family and treat the youth as a part of that unit. Additionally, both are expected to provide opportunities for youth to experience a positive family environment where he/she can participate in a well-functioning family unit and the broader community. In the classic foster parent role, neither is intended to be the long-term permanent solution for youth. Yet, the innovations of “concurrent planning” and “foster-adopt” programs are now offering potential adoptive parents the option of first becoming foster parents to a child (Brooks & Webster, 1999). Treatment foster care programs work more like the foster care of old and are less likely to conclude with adoption (Berrick, Needell, Barth, & Jonson-Reid, 1998). Rather, TFC is intended to fill an immediate need to improve a child’s functioning and, therefore, increase the child’s likelihood of safely and successfully returning home, living with relatives, or being adopted by another family.

The common usage of ‘foster care’ for both services has made distinctions and research confused. Even the Foster Family-based Treatment Association’s (the largest national organization of treatment foster care organizations) annotated review of the TFC literature includes research from both types of care. This co-mingling of literatures is both symptomatic of and contributes to the ongoing confusion in the field. Treatment Foster Care, in its ideal type (e.g., Chamberlain, 2002; Foster Family-based Treatment Association, 1995; Hawkins & Breiling, 1989) is a clearly identified and defined approach that involves a heavy focus on treatment. ‘Real world’ implementation of TFC, however, is a much more heterogeneous service. Recent work in both North Carolina and Maryland (Bruns & Kiser, 2005; Farmer, Burns, Phillips, Angold, & Costello, 2003) has shown tremendous variation in programs that are designated (and reimbursed) as TFC. They varied on nearly all examined domains, including training for treatment parents, supervision, guardianship and referral sources for youth, and goals/tenure of placement.

This ‘fuzzy edge’ between regular foster care and treatment foster care makes it difficult to clearly delineate where one ends and the other starts. Also, contemporary work (discussed later in this paper) is building upon the evidence-base of treatment foster care to bolster and strengthen regular foster care. In an attempt to include the broader range of research and evidence that may be relevant to foster parent training, we have included both ‘regular/traditional’ foster care and TFC. So that we do not contribute to the ongoing obfuscation of these approaches, we will refer to ‘traditional/regular’ foster care simply as foster care; we will refer to treatment foster care as ‘treatment foster care’ or with the acronym TFC.

We recognize, too, that kinship foster care is another important component of the array of out-of-home care options offered through child welfare services. We have not specifically distinguished training for kinship care providers for a number of reasons. First, kin receive the least foster parent training (Hurlburt et al. in press) and may receive children into their home without any license or prior preparation (Geen & Berrick, 2002). When they are trained, kin may be included in standard foster parent training and may, less commonly, be included as treatment foster parents (Kerman, Wildfire, & Barth, 2002). Therefore, for the purposes of this review we will not distinguish between training programs for kinship and nonkinship foster parents.

Legal Requirements

William Grimm (2003), from the National Center for Youth Law, recently published a thorough review of the legal requirements and status of foster parent training throughout the nation. Support for the general idea of foster parent training is contained in federal policy via the Foster Care Independence Act of 1999 (H.R. 3443). This legislation states that “before a child in foster care…is placed with prospective foster parents, the prospective foster parents will be prepared adequately with the appropriate knowledge and skills to provide for the needs of the child, and that such preparation will be continued, as necessary, after the placement of the child” (Grimm, 2003, p. 3). While this language clearly specifies the intent to train foster parents, it also leaves wide latitude for interpretation by states for implementation. In line with this federal requirement, provision of foster parent training is currently contained in administrative codes in nearly all the states. Only one state allows foster parent training to be voluntary, and one other does not require that training be completed as a condition for licensure (though it does require training ‘before the end of the first year of licensure’). The remaining 48 states (and the District of Columbia) all require some form of foster parent training as part of the licensing process for foster parents. All of these states also require some in-service training to maintain licensure.

Beyond this general and nearly universal support for training, however, there is clearly wide variation in the provision of training. The majority of states specify the number of pre-service training hours required, with such specification ranging from 4 to 30 hours (Grimm, 2003). A recent summarization of state requirements by the National Foster Parent Association (www.nfpainc.org) shows that annual in-service training requirements range similarly across the states, from 6 – 20 hours. There are also nearly universal holes in documentation of compliance with training requirements. Most reviews of the training process and compliance in states is completed via self-evaluation by the agencies providing both the training and the service (Grimm, 2003). In addition, there are numerous conditions under which the training requirements may be waived or modified (including kinship care, provisional licensure, and other ‘special situations’).

Specific Legal Requirements Related to Treatment Foster Care (TFC)

As with ‘regular’ foster care, training for treatment foster care is supported by federal and state statutes, but varies widely. There has not been a comprehensive review of TFC training requirements throughout the nation. A recent summary by the National Foster Parent Association suggests that many states do not differentiate training for therapeutic foster parents from training for regular foster care parents. At least 6 states require additional pre-service training hours for TFC parents, and 8 states require more annual in-service training hours for therapeutic foster parents. In general, additional pre-service requirements add approximately 8–10 hours of pre-service training, and annual in-service requirements are 6–30 hours more than for regular foster parents. Often, this additional training is intended to provide additional information and skills related to effectively parenting the types of youth who are placed in TFC (those with severe behavioral problems, histories of delinquency, and other ‘special needs’). Most states that mandate a particular curriculum (usually MAPP^® or PRIDE^®), described below, also require this for TFC parents. Additional training is typically provided by the local public or private agency that runs the treatment foster care programs.

Professional Standards

At present, professional standards for foster parent training have been operationalized via training curricula developed by professional organizations in the area of child welfare. Two curricula -- Model Approach to Partnerships in Parenting Group Preparation and Selection of Foster and/or Adoptive Families (MAPP/GPS) and Foster Parent Resources for Information, Development, and Education (PRIDE) -- are widely used and viewed as ‘gold standards’ for the field. MAPP, the older of the two, was developed in the mid 1980s (revised in early 1990s) by the Child Welfare Institute (Mayers-Pastzor, 1987). PRIDE was developed in 1993 (revised in 2003) by the Child Welfare League of America, in part in recognition of views that standardized training, such as that provided by MAPP, was offering an important service and that additional materials, of its kind and quality, were needed (Menzer & Zobel, 2002).

Both are designed to provide pre-service training for foster parents as well as to provide guidance on screening and selecting potential foster parents by agencies. Therefore, a substantial portion of the information covered in both curricula concentrates on helping families decide if they want to be foster parents, rather than on developing skills for best serving youth in foster care. The two programs are similar in length (27 hours for PRIDE, 30 hours for MAPP) and share similar foci. The MAPP curriculum is built around 12 ‘key skills’ while the PRIDE curriculum focuses on 5 ‘competencies.’ Both include a wide focus on the knowledge and skills necessary to work within the child welfare systems and emphasize core values of foster care (e.g., building strengths, building connections, developing relationships, supporting children’s needs). Both have been criticized for their relatively substantial attention to procedures and policies and relatively brief attention to issues involved in effectively meeting the needs of troubled youth (particularly their scant focus on managing difficult behaviors).

The MAPP and PRIDE curricula have been widely adopted across the nation. Currently, 26 states formally require foster care agencies to use either MAPP or PRIDE as their pre-service training curricula.

Professional Standards for Treatment Foster Care

In 1995 (updated in 2004) the Foster Family Based Treatment Association (FFTA) published a set of Program Standards for Treatment Foster Care (FFTA, 1995, 2004). These standards were developed via a series of meetings among directors and developers of leading programs in Treatment Foster Care. The Standards specify a wide range of dimensions, with Standards categorized into three broad areas: Standards related to Program; Standards related to Treatment Parents; and Standards related to Children, Youth, and Families. The FFTA Standards are, in many ways, consistent with core values and principles underlying PRIDE and MAPP. These include a strength-based focus, an affirmation of children’s natural families and their needs for permanent homes, and of cultural diversity. Given Treatment Foster Care’s focus on ‘treatment,’ not just parenting, the Standards also include specific guidelines on staffing, supervision, respite care, and treatment planning.

In terms of training, however, the FFTA Standards are rather brief. They recommend 30 hours of pre-service training and 24 hours of annual in-service training (FFTA, 1995). Agencies are instructed to provide …”at least 30 hours of primarily skill-based training consistent with the program’s treatment methodology and the service needs of the children.” (FFTA, 1995, p. 16). There is also reference to a Professional Development Plan for each treatment parent, but with few details or guidelines for development or completion.

In addition to guidelines in the FFTA Standards, current thinking about TFC is heavily influenced by work by Chamberlain and colleagues at the Oregon Social Learning Center (Chamberlain, 1994; 2002; Chamberlain & Mihalic, 1998). Their version of TFC, known as Multidimensional Treatment Foster Care (MTFC), provides the primary evidence base for TFC and, therefore, provides a potential empirically supported approach to training. Training for Chamberlain’s model of TFC includes initial training and ongoing supervision. It begins with a 3-day orientation for new treatment foster parents. After this initial intensive introduction, new treatment foster parents receive extensive and frequent supervision, support, feedback, and interaction with a supervisor. Much of the actual ‘training’ in this model, therefore, comes from this ‘in vivo’ learning, rather than from formal classroom training (Chamberlain, 1994; Chamberlain and Mihalic, 1998). Training, both formal and ‘in vivo,’ includes a central focus on principles of social learning and Parent Management Training (Chamberlain, 2002; Patterson, 1976 & 1982; Patterson, Chamberlain, & Reid, 1982). Using intensive and real-time problem solving, supervisors concentrate on assisting foster parents with effectively handling frequently encountered youth behavior problems.

General Dimensions of Foster Parent and Treatment Foster Parent Training

All of this focus on required training hours often ignores the central question of “training about what?” Given the complex roles that foster parents and treatment foster parents are asked to play (Rhodes, Orme, & McSurdy, 2003; Wells & D’Angelo, 1994; Wells, Farmer, Richards, & Burns, 2004), it is not surprising that training often addresses a wide array of topics, and that different training approaches differentially emphasize these domains.

As will be shown in the review of the evidence base below, there are a wide range of training programs that have been developed for foster parents. Also, as shown in recent work, there is considerable variation in how training is actually provided (Farmer, Burns, Dubs, & Thompson, 2002; Hurlburt et al. in press). However, there is also a great deal of similarity among the variously named and developed training approaches. For the sake of simplicity and relevance, we review the core components of training for MAPP and PRIDE and, in addition, we review the training curriculum for Chamberlain’s Multidimensional Treatment Foster Care, the only version of TFC that currently has an evidence-base of effectiveness (Chamberlain, 2003; Chamberlain, Leve, & DeGarmo, 2007).

This brief outline of training curricula shows both overlap and substantial differences in the most frequently used curricula. All curricula recognize that foster or treatment foster parents will encounter new situations and challenges in their role. All include material on children’s challenging behaviors, ways to address these, and information about the broader child-serving system in which foster care or treatment foster care is embedded. Beyond these commonalities, though, the approaches are quite different in tone, content, and emphasis. MAPP is heavily infused with material to prepare foster/adoptive parents for the challenges that they and their family will encounter by becoming foster or adoptive parents. It strives to create well-informed foster/adoptive parents who know what they are getting into and who know how to work within the child-serving system.

The PRIDE competencies, in contrast, focus almost exclusively on knowledge/skills that foster parents will require to meet the needs of the foster children who are placed with them. PRIDE goals (not displayed above) focus on more general values and assert the importance of meeting the range of children’s needs, strengthening families, improving foster and adoptive care, and sharing resources among child-serving agencies. PRIDE training provides broad coverage of many areas, all believed, by the developers and users, to be central to the welfare of children in care.

MTFC’s approach to training is strikingly different. If also focuses on meeting foster families’ and youths’ needs, but its emphasis is very specific, skill oriented, and serves to orient treatment foster parents to the detailed and structured way in which youth will be treated in TFC. The emphasis is on specific approaches, setting an expectation for ongoing supervision and support, and working with other professionals on behalf of the youth in their care.

To oversimplify, but characterize these three approaches: MAPP makes sure that foster parents know what they’re getting into; PRIDE makes sure they grasp underlying values and associated competencies; and MTFC makes sure that treatment foster parents understand how to apply the behavioral principles and structures that will form the foundation of their work with youth.

Evidence Base for Foster Parent and Treatment Foster Parent Training

Training for foster parents is currently mandated by federal law and supported by state statutes in nearly all states. A recent review, however, suggested substantial problems with the current status of training (Grimm, 2003). As noted above, approximately half of the states have mandated use of one of the two leading training curricula for foster parents – MAPP or PRIDE. Before it is possible to more fully understand and evaluate the adequacy of current training, a review of the empirical base of the potential curricula and approaches is needed.

To do this, we used a multi-pronged approach to identify empirical articles to assess the status of the ‘evidence base.’ We began with an electronic search (using PsycInfo, Social Work Abstracts, and Social Science Citation Index) to identify possible candidates. Using ‘foster parents’ and ‘training’ as the key words, we identified 79 potential articles. One of these was eliminated because it was not available in English. Abstracts were read for the other 78 to determine eligibility. Given previous reviews (Grimm, 2003; Puddy & Jackson, 2003) that decried the dearth of research on foster parent training, we employed relatively unrestrictive criteria for inclusion (i.e., there were no restrictions on sample size, year of publication, research design). We did, however, require that the research had to appear in a peer-reviewed format and that it must include outcome measures that assessed either foster parent behavior/success or child-level behavior/success. From the initial list of 78, 30 articles were retained and are included in Table 2. Citations were removed from the list for two primary reasons: they provided only a program description or conceptual discussion (with no assessment of any outcomes), or they were doctoral dissertations (without subsequent publication in peer-reviewed venues).

Table 2.

Annotated Listing of Foster Parent Training Research

Type of Training	Authors	Year	Research Design	Sample Size	Sample Char’s	Outcome Domains	Results
PRIDE	Christenson & McMurtry	2007	Pre-Post	228	Perspective foster and adoptive parents in Idaho (69 kinship, 159 non-relative)	Perceived competence, in the five PRIDE competency areas; knowledge	Increased knowledge and perceived competence particularly for kinship caregivers
MAPP	Lee & Holland	1991	Quasi-Exp.	29 (17 intervention; 12 comparison)	Foster parents in Georgia	Foster parent knowledge/ behavior	No statistically significant differences between groups. Both groups appear to be in ‘average’ range
MAPP/GPS	Puddy & Jackson	2003	Quasi-Exp.	82 (62 intervention; 20 no-intervention comparison)	Foster parents: (85% Caucasian, 13% African American 63% female)	Foster parent knowledge/ behavior	MAPP-trained group showed improvement in 4 of 12 program- identified goals, 3 of 22 basic parenting skills. Conclusion: insufficient for training foster parents
PRIDE	Herczog, van Pagee, Pasztor	2001	Case Studies of Countries	n/a	9 countries	Factors affecting transportability of PRIDE to other countries	PRIDE can be adapted for other countries. Transportability affected by program content, transfer methodology, commitment of transferring organization
Psycho- education for foster parents of sexually abused youth	Barth, Yeaton, & Winterfelt	1994	Quasi-Exp.	27 (15 intervention group, 12 control group)	Foster parents of sexually abused youth	Child psychosocial functioning & sexualized behavior, foster parent satisfaction	Almost no improvements in child functioning or changes in sexualized behavior in either group; high foster parent satisfaction
Behavioral training	Boyd & Remy	1978	Quasi-Exp.	Approximately 105 foster families of 267 youth (intervention and comparison condition)	Foster parents in San Francisco Bay area	Placements, licensure	Training was associated with fewer aborted placements and increased probability that foster parents would remain licensed
Multimodal in- service	Burry	1999	Quasi-Exp.	88 (28 intervention, 60 non-equivalent comparison)	Foster parents of infants with prenatal substance effects	Competency to care for infants with prenatal substance effects	In-service training increased skill and knowledge attainment. Did not achieve goals for increased efficacy, social support, and intent to foster
Stipends and support (behavior management training & supervisor support)	Chamberlain, Moreland, & Reid	1993	RCT	72 (3 conditions: (1) Stipend + support (n = 31); (2) Stipend only (n = 14); & (3) usual care (n = 27))	Foster parents of 4–7 year olds in Oregon	Parenting behaviors, discipline, retention	Child behavioral outcomes and successful days in care better for ‘stipend +’ than for other conditions. Stipend only had some benefits (FP retention; child outcomes)
Foster parents teaching foster parents	Cobb, Leitenberg, & Burchard	1982	Quasi-Exp.	48 (30 intervention; 18 no-intervention comparison)	Foster parents (22 couples, 4 foster mothers)	Communication and conflict resolution	Intervention group showed significantly greater skills than those in no-intervention comparison group; FPs trainers were equally effective
Attachment and Biobehavioral Catch-up (ABC)	Dozier, Peloso, Lindhiem, Gordon, Manni, Sepulveda, Ackerman, Bernier & Levine	2006	RCT	60 (30 ABC intervention; 30 educational intervention comparison)	Foster parents of young foster children (0 to 36 months) in two mid- Atlantic states (93% female; 63% African American, 32% Caucasian)	Cortisol levels, child behavior problems	Cortisol levels for ABC group lowered to be similar to those of typically developing children. ABC intervention effect for behavior problems of toddlers
Nova Foster Parent Preservice Training Curriculum	Fees, Stockdale, Crase, Riggins- Caspers, Yates, Lekies, & Gillis- Arnold	1998	Pre-Post	48	Foster mothers in Iowa (selected during pre- service training) 81% Caucasian	Satisfaction with foster parenting	FPs who viewed training as more useful were more satisfied with foster parenting one year later
Early Intervention Foster Care Program (EIFC)	Fisher, Burraston, & Pears	2005	RCT	90 foster children (47 randomized to EIFC group, 43 randomized to regular foster care comparison condition)	90 3–6 year old children in foster care (majority Caucasian) and their foster parents	Placement outcomes (type of placement, time in placement)	EIFC group had fewer failed permanent placements (EIFC: 90% success rate vs. 64%), EIFC mitigated negative effect of multiple prior placements
Foster parent skills training program	Guerney	1977	Quasi-Exp.	132 (75 in intervention, 57 in no-intervention comparison condition)	Foster parents in Pennsylvania	Foster parent acceptance and sensitivity	Training can be offered effectively Trained foster parents showed more acceptance, reports of desirable parenting responses, less undesirable responses
Foster parent skills training program	Guerney & Wolfgang	1981	Quasi-Exp.	400 (321 experimental, 79 control)	Foster parents in Pennsylvania	Foster parent acceptance and sensitivity	Trained foster parents showed more acceptance, reports of desirable parenting responses, less undesirable responses
Behavioral vs. reflective group training	Hampson & Tavormina	1980	Quasi- Experimental	42 (17 in behavioral groups, 17 in reflective groups, and 9 in a wait-list control)	Foster mothers in a rural area (38% Caucasian, 62% African American)	Foster parent attitudes and behavioral skills	Reflectively trained gained primarily in attitudes; behaviorally trained gained in behavioral skills and reduced behavior problems in child
Individual vs. group training (behavioral and reflective)	Hampson, Schulte, & Ricks	1983	Quasi-Exp. (some random assignment)	29 foster parents making up 18 foster families (9 group training; 9 in-home individualized training)	Foster parents in Dallas	Foster parent attitudes, ratings of child, observations of foster parent/child dyad	Improvements in nearly all measured areas for both groups. Not significantly different between groups for foster parent attitudes or use of behavioral techniques. Home- trained FPs felt more positive about gains of training
Systematic skills training	Levant & Slattery	1982	Quasi-Exp.	14 (8 intervention, 6 no-intervention comparison)	Lower SES foster mothers (13 African American, 1 Caucasian) in Massachusetts	Variety of child measures, placement stability	Children of mothers in the intervention group showed significant improvement on only one outcome (emotionality-tension)
Foster Parents as Mental Health Para- professionals	Levant, Slattery, and Slobodian	1981	Quasi-Exp.	15 (9 intervention, 6 no-contact comparison)	1st time foster mothers (14 African American, 1 Caucasian) in Massachusetts	Variety of child measures, placement stability	Children of mothers in the intervention group showed significant improvement on only one outcome (emotionality-tension)
The Incredible Years (parent management training)	Linares, Montalto, Li, & Oza	2006	RCT	128 parents; 64 children (80 parents received training, 48 assigned to ‘usual care’ condition)	Foster parent and biological parent pairs of 64 pre- and school-aged children (57% Latino, 33% African American; 50% foreign born) in New York City	Parenting, co- parenting, child externalizing problems	Intervention was successful for increasing positive parenting and collaborative co- parenting. Intervention group also had a trend for less child externalizing problems
Cognitive Behavioral Parent Training	MacDonald & Turner	2005	RCT	117 foster parents (67 intervention group, 50 wait list control group)	Foster parents in southwest England (majority Caucasian, single parent families)	Parent knowledge of behavioral principles, parenting skills, child behavior, foster parent satisfaction with the intervention, placement disruption	Training had limited impact on FP behavior, no impact on child behavior or placement disruptions. Small increase in behavioral knowledge for intervention group, increased self-report of some of the behavioral skills; control group more likely to one of the behavioral skills and had fewer disrupted placements
Parent-Child Interaction Therapy (parent management training)	McNeil, Herschell, Gurwitch, & Clemens- Mowrer	2005	Pre-post	30	Foster parents (29 mothers) of preschool and school aged youth (2–8 year olds)	Child behaviors, parenting skills, foster parent training and outcome satisfaction	Improvements in child behavior problems, high reported use of skills (80%) at follow up, high satisfaction with workshop
Multimedia training on dealing with anger outbursts	Pacifici, Delaney, White, Cummings, & Nelson	2005	RCT	74 (half assigned to multimedia training; half to a no intervention control group)	Foster parents (80% Caucasian)	Foster parent knowledge and perception of anger outbursts, satisfaction with the program	Increased knowledge and perceived confidence in having parenting skills to handle anger outbursts at 3- week follow up, high satisfaction with DVD training
Web-based training on lying and sexualized behavior	Pacifici, Delaney, White, Nelson, & Cummings	2006	RCT	97 (half assigned to complete lying course first, half completed sexualized behavior course first)	Foster parents in California (69% female, 76% Caucasian)	Foster parent knowledge and perception (confidence, comfort, objectivity) of lying and sexualized behavior, satisfaction	Increased knowledge (both lying, sexualized behavior) and increased perception of behavior (lying only) at 1-week follow up; high workshop satisfaction
Behavior modification	Penn	1979	Pre-Post	7	Foster parents (4 couples, 3 foster mothers)	Child behaviors	131 focal behaviors changed in positive direction, 27 in negative direction
Managing challenging behavior	Pithouse, Hill-Tout, & Lowe	2002	Quasi-Exp.	106 (53 intervention; 53 no-intervention comparison	Foster parents in South Wales	Foster parent behavior; child behavior	Training had limited impact on children’s behavior and on FP’s behavior
Positive Parent Training	Rinn, Markle, & Wise	1983	Quasi-Exp.	13 (intervention vs. drop-outs)	Foster parents	Goal attainment on target behaviors	Trained group showed higher goal attainment and greater improvement than drop- outs at 1-month and 1- year follow-ups
‘Specialized’ Training	Sanchirico, Jablonka	2000	Survey of foster parents	650	Foster parents in NY state	Involvement of foster parents in parent-child contact	Specialized training and support can increase involvement of foster parents in parent-child contact
Nova University foster parent training	Simon & Simon	1982	Quasi-exp.	65 (foster parents licensed before and after mandatory training)	Foster parents in Florida	Placement failures	Trained homes half as likely as untrained to have disrupted or failed placements (11% vs. 23%)
Parent-Child Interaction Therapy (parent management training)	Timmer, Urquiza, & Zebell	2006	Pre-Post	75	75 non-relative foster parents; 98 biological parents in California	Child behavior, caregiver psychological distress and parenting stress, caregiver potential for abuse	Overall, FPs equally likely to finish PCIT; distressed FPs more likely to finish; FPs somewhat more likely dropout if severe child behavior problems. For both FPs and bio parents completion of PCIT = improved child behavior, reduced caregiver stress and distress.
Family Life Education: Sexual Abuse	Treacy & Fisher	1993	Pre-Post	20	Foster parents of youth with sexual abuse histories	Foster parent knowledge and attitudes	FPs showed significant increases in comfort, knowledge, and sense of competence

The entries in Table 2 are arranged using two factors. First, they are grouped by type of training being investigated. Second, there are arranged alphabetically (by author) within training type. It is quickly evident that despite MAPP and PRIDE’s widespread adoption as training curricula, there is virtually no research to support their use. There have been two published evaluations of MAPP – the first (Lee & Holland, 1991) showed no statistically significant effect of training and the second (Puddy & Jackson, 2003) showed small gains in only a few of the program-identified goal areas. Both of these studies relied on relatively small samples (N = 29 and N = 82, respectively), were quasi-experimental designs, and neither assessed potential influences on child-related outcomes.

Only two peer-reviewed studies of PRIDE were identified, and inclusion of one (Herczog, van Pagee, & Pasztor, 2001) required a slight bending of eligibility criteria in terms of outcomes because, rather than assessing effectiveness of the PRIDE curriculum, this article assessed transportability of the curriculum to other countries. The second article (Christenson & McMurtry, 2007) demonstrated higher levels of foster-parent reported competency in many of the five PRIDE competency areas after receiving the PRIDE training. This evaluation, like those of MAPP, also has significant limitations. Although conducted with a larger sample than any of the MAPP trials (N = 228), this study did not include a comparison group or examine foster parent behavior/success or child-level behavior success.

The rest of Table 2 examines research on various other types of foster parent training (i.e., not PRIDE or MAPP). These are listed alphabetically by author, because there was such diversity in the types of training that it was difficult to develop a ‘training type’ typology that meaningfully categorized the included pieces.

Before reviewing the results, it is important to note that most of these studies focused on training for non-relative foster parents. Few articles reported including kinship caregivers, and none of the studies focused exclusively on training for kinship caregivers. In the US, youth residing in kinship care make up approximately one-fourth of the population of youth in foster care (AFCARS, 2005). Yet, recent research indicates that in addition to youth in kinship care receiving lower rates of mental health services, they also receive significantly less training (both pre and in-service) (Hurlbert, et al., in press).

In addition to the limited research available on kinship caregivers, it also is interesting to note that a third of the articles included in this review (11 of 30) were written more than 20 years ago, and in fact, were clustered between 1979 and 1983. Also notable, in the last five years, there has been a relative surge of research on foster parent training. During this period, another third of the articles (12 of 30) were published, including two of the four papers focusing on MAPP and PRIDE.

Many of the early studies included very small samples, were quasi-experimental (with or without a comparison/control group), and assessed a very limited range of outcomes. However, despite these short-comings, most reported at least preliminarily promising findings – e.g., improved foster parent knowledge/attitudes, placement stability, child behaviors. However, additional research was not conducted on any of the programs to more fully explore effectiveness. This may be because at that time, foster parent training was considered a prominent mechanism to addressing placement instability, which has long been identified as a central problem of child welfare services (Maas & Engler, 1959). During the 80s and 90s, the solutions to foster care instability and drift predominantly were addressed through legal reforms to more rapidly move children out of temporary foster care and a greater emphasis on placement prevention.

However, family preservation’s influence is waning because of little evidence of effectiveness (Littell & Schuerman, 2002; Schuerman, Rzepnicki, & Littell, 1994), and placement instability remains an issue of concern to policy makers and service providers. Indeed, placement moves are one of eight federally mandated outcome indicators for which states and localities must account. The increase in recent studies of foster parent training and other research currently underway (see future directions section) suggests renewed interest in training as a mechanism for improving outcomes for youth. These recent studies typically have more sophisticated designs, larger sample sizes, and assess a wider range of outcomes (e.g., foster parent behavior/success, child behavior). In addition, research on many of the programs evaluated in these studies is ongoing (e.g., Multidimensional Treatment Foster Care for Preschoolers, Attachment and Biobehavioral Catch-up) and should, over time, provide information about more varied outcomes and longer term outcomes (Dozier, personal communication).

Looking across the trainings evaluated in the 29 studies included in this review, it is clear that trainings vary widely, both in approach and content. They included some interventions that targeted behavior management skills (e.g., Linares, Montalto, Li & Oza, 2006; Penn, 1978), overall training programs (e.g., Guerney, 1977; Guerney & Wolfgang, 1981), foster parenting of sexually abused children (Barth, Yeaton, & Winterfelt, 1994), and a variety of specific questions focused on training methods, foster parent cognitions, or demographic subgroups of foster parents (e.g., Cobb, Leitenberg, & Burchard, 1982; Hampson & Tavormina, 1980; Levant & Slattery, 1982).

In terms of outcomes, most of these studies examined foster parents’ self-reported knowledge and/or attitudes immediately following training, with little or no assessment at later points in time. For the most part, these studies show that training increases these subjective factors in the short-term (e.g., Burry, 1999; Fees et al. 1998; Pacifici, Delaney, White, Cummings, & Nelson, 2005; Pacifici, Delaney, White, Nelson, & Cummings, 2006; Treacy & Fisher, 1993). Although increases in knowledge and/or attitudes holds promise for potentially impacting foster parent behavior/skill, follow up studies are needed that directly examine these outcomes, in addition to impact on child behavior/functioning.

A limited number of studies examined impact on children’s behavior, and results were mixed. Generally, these studies tended to be on the ‘more recent’ end of the continuum, with studies in the last five years often having the most sophisticated designs and most positive results. Pithouse, Hill-Tout, and Lowe’s (2002) study of training aimed at managing challenging behavior showed limited impact on parents’ capacity or children’s conduct. Chamberlain, Moreland, and Reid’s (1993) study of increased stipends and support, in contrast, showed positive effects on children’s behavior. This study has received considerable attention in recent reviews (e.g., Grimm, 2003), but its findings should be viewed carefully. The intervention included a group that received increased stipend and support as well as a group that received increased stipend only (both were compared to usual care foster parents). While the increased stipend showed some positive effects, these appeared to be most positive for families and children in the ‘stipend plus support’ group (additional support in this study included ongoing consultation and supervision beyond basic foster parent training).

Three of the studies conducted in the last five years involved utilizing interventions originally developed for other populations and show positive results on a wide range of outcomes. Chamberlain’s MTFC has been adapted by Fisher to address the needs of preschool-aged children in foster care. This intervention, Multidimensional Treatment Foster Care for Preschoolers (MTFC-P; formerly called Early Intervention Foster Care) includes a broad-based approach to training [including 20 hours of pre-service training, intensive ongoing support and consultation to treatment foster parents, and training for long-term placement resources (birth or adoptive parents)] (Fisher, Ellis, & Chamberlain, 1999; Chamberlain & Fisher, 2003). Results from a recent randomized trial show positive effects on parenting, children’s behavioral adjustment, cortisol levels, and reduced placement failures, compared to placement in regular foster care (Fisher, Burraston, & Pears, 2005).

In addition to this adaptation of MTFC, two studies in the last five years examined programs with longstanding evidence for treating disruptive behavior disorders in the general population: The Incredible Years (IY) (Webster-Stratton & Hammond, 1997) and Parent-Child Interaction Therapy (PCIT) (Eyberg & Robinson, 1982). These programs also were highlighted in our 2005 review as holding promise/relevance child welfare (Barth et al., 2005). The randomized trial of IY (N = 64 youth and their foster and biological parent(s)) showed an increase in foster parent-reported positive discipline skills and clear expectations, collaborative co-parenting between foster and biological parents, and a trend for improved child behavior problems, when compared to a child welfare services as usual condition (Linares et al., 2006). Two studies examined PCIT. In the first, (McNeil, Herschell, Gurwitch, & Clemens-Mowrer, 2005; pre- post design, N = 30), foster parents reported high use of the parenting skills as well as significantly improved child behavior problems at follow up. In the second (Timmer, Urquiza, & Zebell, 2006; pre post design, N = 75), foster parents were equally as likely as biological parents (N = 98) to complete PCIT, and foster parents (and biological parents) reported improved child behavior on two different measures. In addition, caregivers (both foster and biological) who completed PCIT reported improved caregiver psychological distress and parenting stress. Interestingly, although IY was administered in its usual format (12 weeks, 2-hours each session), in one PCIT evaluation (McNeil et al., 2005), delivery was modified so that the approximately 14-week treatment (1 hour each session each week, with parent-child practice in between sessions) was administered in a two-day workshop in an attempt to better accommodate foster parent schedules.

Also in the last five years, Dozier and colleagues developed and began evaluating a manualized but flexible approach that focuses specifically on infants and young children (ages 0 to 30 months) in foster care. Their approach is built on research in attachment and emotional regulation in young children and how training may be enhanced to better meet these needs (Attachment and Biobehavioral Catch-up; ABC) (Dozier, Dozier, & Manni, 2002; Dozier, Higley, Albus, & Nutter, 2002). The first randomized trial indicates that infants and children residing with foster parents who received ABC had cortisol levels in the normal range and also that toddlers had fewer behavior problems at follow up compared to an educational intervention comparison group (Dozier et al., 2006).

Pacifici and colleagues (Pacifici et al., 2005 & 2006) conducted two studies examining a web-based training program for foster parents. These studies do not have a sophisticated design or outcomes on foster parent behavior/success or child behavior; however, to our knowledge they appear to be the first to examine the use of web-based training for foster parents. As with other trainings included in this review (all person-to-person), foster parents reported satisfaction with the training and also reported increased competence in dealing with the focus problems (i.e., lying, anger outbursts, sexualized behavior). Given the costs associated with in-person trainings, the viability of using technology to accomplish training goals should be examined further. Although, as with many of the other evaluations included in this review, studies need to move beyond satisfaction and self-report of competence.

Training for Treatment Foster Care

There is currently little data on training for treatment foster parents. As noted above, the majority of ‘evidence’ for treatment foster care comes from Chamberlain’s group. Data from this body of work shows that treatment foster parents adhere to a distinct paradigm of treatment and implement treatment that is substantially different than that in other treatment placements (i.e., group homes; Chamberlain, Ray, & Moore, 1996). As noted above, Chamberlain’s MTFC model includes both preliminary training and ongoing supervision/support. The focus of training (pre-child placement) and ongoing supervision/support (after the child is placed) involves ensuring that parents can effectively deliver high quality behavioral interventions with the youth in their care.

Recent work in a representative sample of TFC agencies in North Carolina suggests that training in TFC agencies is quite heterogeneous (Farmer, et al., 2002). Agencies provided, on average, 20 hours of pre-service training and varied widely on their annual in-service training. Data suggest that more pre-service training was associated with better outcomes for youth [as measured by the Parent Daily Report (PDR) and Behavioral and Emotional Rating Scale (BERS)]. As in Chamberlain’s model, more training was associated with more ongoing supervision and better relationships between treatment foster parents and supervisors. However, unlike Chamberlain’s model, in which training and support/supervision focuses specifically on increasing the behavior management skills of parents, training in NC agencies was highly eclectic. Additionally, many agencies reported difficulty finding time to include much specialized training beyond the state’s required MAPP training.

Conclusions

At present, there is remarkably little empirical evidence for the type of training that foster parents receive. MAPP and PRIDE, the two most widely used training curricula, have virtually no empirical support. A variety of other training curricula, approaches, and concepts have been evaluated. Many of these appear to have some pretest-posttest impact, but few have employed a rigorous research design, sufficient sample size, or wide range of outcome measures. Much of the research was conducted more than two decades ago, but little follow-up or continuation to improve the evidence base is apparent. However, most recently, progress has been made in testing evidence-based treatments, designed for other populations, with foster parents. In addition, these recent studies have all evaluated the impact of training on either foster parent behavior/skill or youth behavior or both. The quality of these recent studies evaluating training for foster parents (all but two are randomized trials) and the interventions themselves (4 build on or are EBPs for other populations) appear to suggest that scientific rigor being applied to foster parent training has increased. With this increase in scientific rigor, our ability to evaluate and determine effective ways to prepare foster parents for effectively caring for youth has also improved. This increased scientific attention and rigor is both exciting and necessary, as research continues to document the significant emotional, behavioral, developmental, and health needs of youth in foster care (Landsverk, Burns, Stambaugh, & Reutz, 2006). In addition, as child welfare legislation around the country appropriate continues to find fault with the current state of services and care for youth in foster care (Katie A. v. Bonta, 2002; Braam v. State of Washington, 2004), this additional empirical attention to foster parent training may provide some guidance for states as they work to improve services and revise standards.

Notably, research on training of therapeutic foster parents, outside of that conducted by Chamberlain and colleagues, is quite limited. Here, research has focused on program-level and child-level processes and outcomes, without rigorous testing of the training methods employed. In the area of TFC, potentially the biggest problem is that Chamberlain’s MTFC has significant evidence of effectiveness, yet there are only 35 MTFC programs in the US (62 programs in total). As there are over 1,500 TFC programs nationwide, MTFC programs make up only about 2% of all programs. Other than converting existing TFC programs to MTFC, which may not be possible for some agencies due to a number of financial and organizational barriers, we are left with little direction from the research literature on how to improve training and practice in these programs. At this point, the limited research outside of that conducted by Chamberlain provides us with little information about whether similar outcomes could be obtained using a slightly different treatment package in TFC. For example, can TFC parents deliver high quality behavioral interventions if they do not employ MTFC’s signature points and levels system? Although Farmer’s research on TFC in NC, discussed in the following section, may provide data to answer some of these questions, more research is needed.

Current Activities and Future Directions

This lack of an empirical foundation, coupled with concerns about the quality of care for youth in out-of-home placements, and strongly held beliefs about the importance of foster parent training (from advocates, policy makers, and providers), have come together to encourage a range of current activities to advance knowledge about effective training for foster and treatment foster parents.

Some of the most positive findings for foster parent training as a vehicle for improving outcomes for youth come from ongoing work by Chamberlain and colleagues. Chamberlain, Price, and others have drawn from the evidence-base of treatment foster care to infuse aspects of the MTFC model into training for regular foster parents. Foster parents participating in the Keeping Foster Parents Trained and Supported (KEEP) intervention received 16 weeks of training, weekly homework, and weekly telephone calls to monitor progress and problems. Information gained from phone calls is used to tailor interventions to the types of problems that foster/kin parents were facing. Data at the end of the intervention phase showed positive effects (compared to conventionally-trained parents) on parenting skills, children’s behavior problems, placement stability, and family reunification. Results indicate that improvements in behavior problems were accounted for by changes in parenting, and particularly for youth with high levels of behavior problems (Chamberlain et al., in press) Finding indicate that youth whose foster parents’ received the KEEP intervention were more likely to have a positive exit from foster care (e.g., be reunified with biological parents) and, in the case of youth with multiple prior placements, receiving the KEEP intervention mitigated the negative impact of multiple placements on stability (Price et al., in press). Furthermore, the KEEP intervention appears to have similar results even when implementation and supervision is out of the developers’ hands (Chamberlain, Price, Reid, & Landsverk, in press). In addition to their recent work on KEEP, Chamberlain and colleagues are engaged in work to disseminate MTFC to sites throughout the country and internationally. Data from such efforts may help illuminate key factors associated with successful dissemination and implementation.

Farmer and colleagues (Farmer, Murray, Dorsey, & Burns, 2005) have recently finished a randomized trial to improve training of TFC treatment parents, supervisors, and clinicians in a statewide sample of existing TFC programs. Previous work (Farmer et al., 2002) showed tremendous variation in foster parent training among these agencies, and substantial deviations from Chamberlain’s behaviorally-focused training. The randomized trial examines the impact of increased and focused training to prepare and support treatment foster parents. This intervention, Together Facing the Challenge, provides a structured format for supplementing usual pre-service and in-service training for treatment foster parents, by bringing together key elements of parent-training from Chamberlain’s model with increased training and ongoing consultation for treatment foster parents’ supervisors about better ways to provide consultation and support their treatment foster parents (Farmer et al., 2005).

Discussion

There is widespread rhetorical and philosophical support for providing training to foster parents, but little empirical support for the utility of the most common programs provided. Current policy in nearly every state requires that foster parents receive pre-service training and most states require some ongoing in-service training. In approximately half of the states, a specific curriculum is specified by statute, and nearly all specified programs are MAPP or PRIDE. The evidence base for both of these curricula is very sparse.

Although we have no reason to doubt that basic foster parent training messages (e.g., maintaining a safe house with active fire detection devices, avoiding corporal punishment, and cooperating with child welfare workers) are being effectively communicated, it seems that foster parent training needs to be two-pronged. Many of the messages conveyed in MAPP and PRIDE are necessary for orienting foster parents to their job and to basic requirements, as well as for licensing homes, and thus make up the first prong. These programs, however, likely are insufficient for meeting the often loftier goal of foster parent training—that foster parents have the necessary skills to effectively care for and maintain the children residing in their home. Recent literature paints a consistent picture that youth in foster care enter such settings with high levels of behavior problems and a wide range of difficulties related to their histories of maltreatment, multiple placements and caregivers, and other disabilities (e.g., developmental delays) (Landsverk et al., 2006). Effectively ‘parenting’ such children is clearly challenging. When foster parents cannot care for these children effectively, they end up moving from foster home to home (and eventually to more restrictive and costly settings), with their behavior worsening, making it less likely that the next foster parent will be effective in being able to maintain them in their home, particularly without the needed training to do so.

The second prong of foster parent training should focus specifically on providing skills-based training to assist foster parents in managing these often difficult and wide ranging behaviors with which youth present. The broader evidence-base for a wide variety of curricula and approaches is very limited. On the positive side, it appears that, in general, some training is better than no training. However, aside from the most recent studies, the range of examined outcomes is very narrow (often foster parent knowledge, perceived competency) and inconsistent.

The best evidence that this second goal can be achieved effectively comes from studies conducted very recently and ongoing studies. Many of the examined approaches involved a focus on parenting skills or behavioral approaches, and these appear to have positive effects on foster parent behavior/skill, child behavior, and for the MTFC-based approaches, placement outcomes. The utilization of some EBPs originally designed for slightly different populations (e.g., PCIT, IY, MTFC) appears to hold the greatest promise for positively impacting foster parent behavior/skills and youth behavior. Additionally promising is a newly developed intervention that specifically targets a specific age/developmental group and their unique needs (ABC). One factor that all of these interventions have in common is that they are provided to foster parents after a child is placed in their home, and not before. A second common factor among most of these recent interventions/trainings is the opportunity for foster parents to practice skills, most often with the youth in their care, and receive coaching and feedback on skill implementation. The fact that studies in our review that include this factor had positive outcomes mirrors findings from a recent meta-analysis on parent training as prevention in child welfare (National Center for Injury Prevention and Control, 2004). In this meta-analysis, practice with the child was the intervention component associated with the most positive outcomes both for the child and parent.

However, despite the promise of some of these interventions, more work is necessary before the evidence-base can be used to definitively guide practice in this area. As we have found in North Carolina and in talking to others around the nation, it is difficult to find time to incorporate in-depth training, such as those reviewed above, when a state already requires a specific curriculum (e.g., MAPP or PRIDE). There are clearly elements of MAPP and PRIDE that appear, a priori, to be beneficial for both screening and training prospective foster parents. However, there also is relatively little attention in these curricula to specifics of parent management training and to formalized follow-up, consultation, and support that seem to be crucial to evidence-based TFC, and indeed, to other parent-training approaches (e.g., Chaffin, et al., 2004; Chamberlain & Mihalic, 1998). Current widely utilized approaches also provide primarily didactic training, with relatively little attention to contemporary views of appropriate and effective adult learning approaches (e.g., problem-based learning, interactive presentations). As data on effective training approaches continues to accumulate, it will be necessary to determine minimal levels of such in-vivo and ongoing training/consultation to support the best possible foster parenting and outcomes for youth. In addition, it may be important to determine whether, given the limited data supporting its effectiveness, hours allotted for MAPP/PRIDE training could be lessened, allowing more time to be spent on skills-based training once foster parents have a youth placed in their home. Currently, it seems that there are three areas in foster parent training that require more research attention to advance the field: content (e.g., predominantly behavioral training or other), timing (e.g., provided pre-placement, after placement, or both), and delivery (e.g., in-person vs. other forms, interactive, practice-based vs. lecture/classroom).

Conducting solid scientific investigations in this area is difficult, and advancing the evidence base will not come easily or quickly. Yet, with creative approaches and substantial effort/resources, barriers could be surmounted. Legal mandates about training throughout the nation set parameters on what and who can serve as comparison groups. Yet, comparisons can still be made between basic and enhanced training programs. Tremendous instability of placements as well as turn-over of foster parents makes long-term evaluations difficult to conduct or interpret. Yet, more sophisticated efforts to explain placement instability are emerging (James, Landsverk, & Slymen, 2004; Rubin, O’Reilly, Luan, & Localio, 2007; Wulczyn, Kogan, & Harden, 2003) and could be enhanced further, if data systems began to integrate foster parent information with placement data. In addition new approaches to modeling complex longitudinal data with attrition minimize these problems (Bryk & Raudenbush, 2002; Verbeke, Molenberghs, 2000; Singer, Willett, 2003). Legal requirements around permanency planning add complexity to sample identification and definition, because of differences between expected lengths of placements and allowable documentation of such plans. Shortages of foster parents and treatment foster parents and perceived burden of mandated training leave little time or willingness to increase training or shift to more demands for follow-up or supervision. However, this was a concern was raised in the 1980s for why MAPP and PRIDE could not be required, and time has shown that foster parents who are serious about fulfilling the obligations of their role are willing to take additional time for training.

Change is difficult for existing organizations, and dedicated leadership and support are essential for successfully changing training and evaluating its impact. However, the timing appears to be ripe for such leadership and change. Foster care and treatment foster care are well-established entities, often not content with their current approaches and outcomes, and are often facing a myriad of procedural, legal, administrative, and contextual factors that make change difficult. Some states (e.g., Illinois, New York) are beginning to hold foster care agency providers accountable for their performances and to compensate them according to their performance (Cross, Leavey, Mosley, White, & Andreas, 2004; Fischer, Green, Kihara, Thrush, & Warren, 2001). Also, as states are looking for effective and cost-efficient alternatives to more expensive residential treatment, there is new focus and critique on TFC as a potential solution (e.g., Maryland, North Carolina; Hahn et al. 2004). These emerging issues and new incentives to improve agency performance may provide catalysts to creatively work to change and improve the research landscape. Conducting research on these types of ‘real world’ services requires large commitments of time, resources, support, and patience. It also requires advocacy, at the highest levels, about the importance of having training for foster and treatment foster parents supported by evidence, rather than continued via inertia.

Table 1.

Overview of Foster Parent and Treatment Foster Parent Training Curricula

	MAPP	PRIDE	MTFC
Hours of pre-service training	30	27	20
Required in-service training	n/a	n/a	Ongoing (frequent contact with supervisor and weekly meetings with other treatment parents)
Training Objectives/ Competencies	7 Objectives	5 Competencies	5 Core Areas
	1. Ensure the family assessment is objective and described in specific, behavioral terms	1. Protect and nurture children	1. Overview of TFC a. professional approach b. teaching opportunities c. relationship building d. confidentiality e. treatment team
	2. Increase amount of responsibility prospective foster parent and adoptive parents take in the decision-making process….	2. Meet children’s developmental needs and address developmental delays	2. Using a 4-step approach a. knowing when a problem is a problem b. defining problem behavior c. examine antecedents d. change the consequences that maintain the problem
	3. Set the foundation for partnership between parents and social service agency staff	3. Support relationships between children and their families	3. Procedures for using a 3-level point system a. learning levels b. home-school link
	4. Provide opportunity for prospective foster/adoptive parents to make an informed decision about their ability to foster or adopt….	4. Connect children to safe, nurturing relationships intended to last a lifetime	4. Working with the child’s natural family a. family therapist’s role b. common sources of stress on the team c. treatment foster care policies and procedures
	5. Prepare foster/adoptive parents for the initial and long-term impacts of a new child in their family system	5. Work as a member of a professional team	5. Putting it all together a. daily support and ongoing training b. foster parent support
	6. Give new foster/adoptive parents some guidelines and practice to deal with issues that most often cause placement disruptions
	7. Prepare adoptive parents for the life-long issues they will confront as their child matures….