Coaching With Parents in Early Intervention

An Interdisciplinary Research Synthesis

Abstract

The purpose of this article was to synthesize intervention studies using coaching with parents in early intervention with a focus on (a) definitions and descriptions of coaching with parents; (b) characteristics of families and coaches; (c) parameters such as settings, contexts, dosage, and professional development related to coaching; and (d) child and family outcomes. Through a search of articles from 2011 to 2013, the authors identified 8 studies that met search criteria. Results indicate that there is no common definition/description for the term “coaching with parents in early intervention.” Furthermore, the use of the term “coaching with parents” spans a continuum that on the one end can best be described as relationship-directed process and on the other end an intervenor-directed process. This continuum continues to create confusion for practitioners in appropriate use of coaching with parents. In addition, although positive outcomes are noted for infants and toddler who experience disabilities and their families, it is difficult to discern the direct impact of coaching with parents. The studies, as a whole, tend to give broad descriptions for the use of coaching but little information on how to individualize given specific child and family outcomes. Finally, the intended outcomes of coaching with parents are not fully articulated. Given these results, this synthesis has found the need for increased research efforts in the area of coaching with parents with focus on clarification of the term “coaching” and the parameters that affect its effectiveness.

“COACHING WITH PARENTS” is a term defined in a variety of ways in the field of early intervention (EI). The generally accepted definitions encompass a wide variety of adult learning strategies intended to “promote parents' abilities to support child learning and development within contexts of everyday activity settings” (^{Rush & Shelden, 2011}, p. 175). Coaching with parents became an accepted and oftentimes expected practice in EI for infants and toddlers with disabilities and their families between the years of 2000 and 2010, which was the first decade with widespread use of the term in the literature (^{Dunst, Trivette, Raab, & Masiello, 2008; Friedman, Woods, & Salisbury, 2012; Kaiser & Hancock, 2003; Rush & Shelden, 2011; Rush, Shelden, & Hanft, 2003; Wallace & Rogers, 2010; Woods, Kashinath, & Goldstein, 2004}). During this same time period, promotion of coaching with parents appeared in many professional policy statements and guidance documents (^{American Speech-Language-Hearing Association, 2008; National Early Childhood Technical Assistance Center [NECTAC], 2008}).

Coaching with parents was first justified in Individuals with Disabilities Education Act (IDEA) and its focus on partnering with families in the delivery of EI services (^{IDEA, 2004}). Second, the practice emerges from the adult learning literature in which coaching is aligned with the adult learning principles of acquisition, mastery, application of meaning to one's own experience, and emphasis on novel or relevant problems (^{Knowles, Holton, & Swanson, 2011}. Coaching with adults (including both parents and professionals) is one of four adult learning methods that result in positive outcomes for adult learners (^{Trivette, Dunst, Hamby, & O'Herin, 2009}).

A third justification for the use of coaching with parents began within a movement in the field in the late 1990s in which the phrases “parent training” and “parent education” generated much discussion and disagreement in terms of use as an approach for working with parents. One group of researchers argued for a reintroduction of the term and a renewed focus on parent education, which they believed had been greatly deemphasized in the preceding 15 years (^{Kaiser et al., 1999; Mahoney et al., 1999}). Others recommended that “parent training/education” was a term that was not in need of resurrection and the use of practices associated with the term in the field during the 15 prior years were too directive and resulted in ineffective and inflexible relationships with families (^{Winton, Sloop, & Rodriguez, 1999}). For these researchers, the term “conveys a formal, one-way flow of information from one who knows (the professional expert) to one who does not know (the parent)” (^{Winton, Sloop, & Rodriguez, 1999}, p. 159). In addition, parent training did not reflect the type of partnership practices called for by family-centered practice including the concept of shared decision making between professionals and families. Parent leaders argued that traditional parent training often infused parent–child relationships with an overemphasis on pedagogical approaches and transformed natural home environments into unnatural didactic environments (^{Turnbull, Blue-Banning, Turbiville, & Park, 1999}).

Following the debate about parent training, ^{Rush et al. (2003)} published an article calling for a coaching process that embodied collaboration with families in natural settings. This process would require reconceptualization of the role of early interventionists from expert telling parents what to do in a top-down fashion to a coach who would be a collaborative partner working alongside parents (^{Rush et al., 2003}). Furthermore, this approach envisioned enrollment in an EI programs as parents “initiating an opportunity for a coaching conversation related to informational, emotional and material supports” (p. 39). This view of coaching was to replace the “hierarchical, power-over relationships” (p. 37) with which parent training and parent education had been associated (^{Rush et al., 2003}).

Prior to the first decade (2000–2010) of an emphasis on coaching, ^{McBride and Peterson (1997)} documented home interventionists coaching with parents only 0.4% of the time. In contrast, interventionist spent more than 50% of the time working directly with the child and 24% of the time providing or asking for information. The remaining time was spent with transition of topic (6%), general conversation (1.8%), listening (7%), observing interaction (7%), facilitating child's play (3%), modeling for parent (0.5%), or in no interaction (0.4%). Research on coaching during this first decade (2000–2010) focused on promotion of coaching with parents as a general concept. Researchers worked to situate the use of coaching within the concepts of family-centered practice. This set the stage for widespread acceptance of the use of coaching as a preferred practice to support parents in EI. The adoption of coaching did not increase much over the first decade despite research describing its process of implementation and anticipated outcomes (^{Campbell & Sawyer, 2007; Rush et al., 2003; Salisbury, Woods, & Copeland, 2010}). Researchers at the end of the decade still conjectured that the “adoption and use of coaching in EI ... has been limited and challenging for EI providers (^{Friedman et al., 2012}, p. 63).

During the initial years of a second decade of research (2011–2014) on coaching with parents in EI, researchers have directed activities toward understanding barriers that inhibit the use of coaching by trying to understand more about how coaching is implemented in the field (^{Campbell & Coletti, 2013; McWilliam, 2012; Salisbury & Cushing, 2013; Wilcox & Woods, 2011; Woods, Wilcox, Friedman, & Murch, 2011}). Researchers have conducted observations of interventionists engaged in coaching with parents to find what strategies are being used under the practice umbrella of coaching and to analyze strategies that work and do not work within home visit settings (^{Campbell & Coletti, 2013; Friedman et al., 2012; Salisbury & Cushing, 2013; Sawyer & Campbell, 2012}). A challenge that has dominated this research is the lack of a clear definition of coaching. The need exists to have an operational definition and to reach consensus on that definition (^{Friedman et al., 2012}).

Currently, the definition most frequently cited by EI professionals, technical assistance providers, and researchers is derived from the definition that was put forth by ^{Rush et al. (2003)}. Since that time, this team of authors has further refined both the definition and the characteristics. Coaching is defined as:

An adult learning strategy in which the coach promotes the learner's (coachee's) ability to reflect on his or her actions as a means to determine the effectiveness of an action or practice and develop a plan for refinement and use of the action in immediate and future situations. (^{Rush & Shelden, 2011}, p. 8)

The characteristics associated with this definition are (a) joint planning, (b) observation, (c) action, (d) reflection, and (e) feedback (^{Rush & Shelden, 2011}). These authors identify parents as possible recipients of coaching and that “coaching of parents can promote their confidence and competence in supporting child learning and development” (^{Rush & Shelden, 2011}, p. 4).

^{Friedman et al. (2012)} offered operational definitions for a set of caregiver coaching strategies that they had identified through their research. These strategies share some commonalities with the coaching characteristics offered by ^{Rush and Shelden (2011)}. These strategies include (a) conversation and information sharing; (b) observation; (c) demonstrating; (d) direct teaching; (e) caregiver practice with feedback; (f) joint interaction (g) guided practice with feedback; (h) problem solving; and (i) child focused. The coaching strategies proposed by this team arise from a conceptual framework that incorporates both principles of adult learning and family-centered practices (^{Friedman et al., 2012}).

In addition to the second-generation need to define and operationalize coaching, another question in need of more investigation in this second decade is how the use of coaching with parents impacts child and family outcomes. ^{Wallace and Rogers (2010)} highlighted the need for comparative studies of the effects of different coaching approaches with families and the need for examination of how to individualize for each child and each family given their unique characteristics and circumstances.

This research synthesis reviews second-generation interventions designed for infants and toddlers and their parents in the effort to analyze the current definition and use of coaching with parents. We addressed the following research questions as they pertain to coaching with parents in EI literature published in 2011–2013:

• What are the definitions and descriptions of coaching with parents?
• What are the characteristics of families and coaches who have participated in the studies?
• What are the parameters such as settings, contexts, dosage, and professional development related to coaching with parents?
• What are the child and family outcomes that have resulted from the use of coaching with parents?

METHODS

Search terms and sources

We conducted a computerized search of databases including Academic Search Complete, Educational Resources Information Center (ERIC), ProQuest Nursing, and Allied Health Source, Psychological Abstracts (PsycINFO), PubMed, and Sage Journal Search for peer-reviewed journal articles on coaching with parents in EI from 2000 to 2013. For these searches, we used the following key words and descriptors in various combinations: coaching, parents, caregivers, infants, toddlers, “early intervention,” and “Part C.” One challenge of synthesizing literature related to coaching with parents is that this method of interacting with parents goes by a variety of names. For example, researchers may use the terms, “parent education” or “parent training,” rather than “coaching”; however, our intent in this review was to synthesis studies that specifically use the term “coaching.”

This computerized search process resulted in 864 studies from which we identified the studies that met the following levels of inclusionary criteria. First, we included all studies that met the following inclusionary criteria: (a) published after 2000 in a peer-reviewed journal; (b) used the term “coaching”; (c) conducted with parents/caregivers of infants and toddlers with disabilities, developmental delay, or at high risk for developmental delay; (d) a majority of child participants in study were between the ages of birth to 3 years; (e) study focused on sessions delivered at least partially in the context of home visits; and (f) reflected at least one of a variety of EI disciplines. The decision to include those studies conducted in the context of a home visit using at least one EI discipline was included to capture studies conducted in alignment with requirements for programs under Part C of IDEA. Using these inclusion criteria, we excluded 783 studies from the previously identified 864 studies. Therefore, we identified 81 studies that met the first-level inclusion criteria.

Second, we selected only those studies of the 81 that included (a) empirical research reporting use of coaching with parents in EI and (b) child and/or family outcomes. This resulted in inclusion of 35 studies.

Our third-level inclusion criterion was to include only those studies published between 2011 and 2013. The purpose of this narrow focus is to review early results of the coaching with parent interventions emerging from the second-generation coaching literature. This will assist in identifying any emerging evidence of a unified definition of coaching with parents. This focus will also provide information as to how the use of coaching with parents impacts child and family outcomes as researchers and practitioners begin to use what was learned in the second-generation research. Our final database resulted in inclusion of eight studies of coaching in EI published in seven journals. Of the eight articles, seven articles represented randomized control trial and one article was a case study design (^{Salisbury & Copeland, 2013}).

SYNTHESIS RESULTS

In this section, we provide a descriptive summary of the eight articles as the basis for highlighting the use of coaching with parents in EI. The summary includes the following sections: (a) definitions/descriptions; (b) characteristics of families and coaches; (c) settings, contexts, dosage, and professional development related to coaching with parents; and (d) parent and child outcomes.

Definitions/descriptions

Table 1 summarizes the definitions and descriptions of coaching with parents.

Table 1: Definitions, Description, Setting, Context, and Dosage of Coaching

Table 1: (continued) Definitions, Description, Setting, Context, and Dosage of Coaching

Table 1: (continued) Definitions, Description, Setting, Context, and Dosage of Coaching

Table 1: (continued) Definitions, Description, Setting, Context, and Dosage of Coaching

Each of the studies offers a different definition/description of coaching; however, the studies share enough similarities to allow for categorization (see Table 1). The studies divide between those that have an explicit definition of coaching (^{Blauw-Hospers, Dirks, Hulshof, Bos, & Hadders-Algra, 2011; Salisbury & Copeland, 2013; Vismara, Young, & Rogers, 2012}) and those that have an implicit definition of coaching (^{Kaiser & Roberts, 2013; Landry et al., 2012; Roberts & Kaiser, 2012; Romski et al., 2011; Welterlin, Turner-Brown, Harris, Mesibov, & Delmolino, 2012}).

The studies with an explicit definition involved a relationship-directed process, and the studies with an implicit definition can be characterized as having an intervenor-directed protocol. The three studies that had an explicit definition related to a relationship-directed process (^{Blauw-Hospers et al., 2011; Salisbury & Copeland, 2013; Vismara et al., 2012}) described coaching as a process in which the intervenor and the family jointly planned the intervention and collaboratively identified and implemented a variety of interventions during the caregiver and child-determined routines and activities. The primary strategies described in these three studies were joint interaction, reciprocal feedback, and reflection. In these studies, coaching was characterized as engagement in conversations with parents to learn how to use existing routines and materials to jointly plan interventions, in contrasted to coaching being tied to one specific intervention. Furthermore, two of the studies included a description of the relationship between the intervenor and the parent. ^{Blauw-Hospers et al.} defined the coaching relationship as “an ongoing, equal partnership in which the family defines the priorities for intervention, the coach supports the family members in developing their own way of caring for the infant and in improving personal coping skills” (2011, p. 1327). ^{Salisbury and Copeland (2013)} defined the coaching strategies of conversation and information sharing as a bidirectional conversational strategy with primary focus on establishing and maintaining the relationship between the caregiver and the EI provider.

Five studies implicitly defined, through their description, an intervenor-directed protocol (^{Kaiser & Roberts, 2013; Landry et al., 2012; Roberts & Kaiser, 2012; Romski et al., 2011; Welterlin et al., 2012}). In these studies, parents were taught strategies dictated by a specific intervention curriculum and carried out primarily in practitioner-chosen home routines with practitioner-chosen home materials (^{Kaiser & Roberts, 2013; Landry et al., 2012; Roberts & Kaiser, 2012; Romski et al., 2011; Welterlin et al., 2012}). In three of these studies, as the intervention progressed, more parent-chosen routines and materials were integrated into the intervention (^{Kaiser & Roberts, 2013; Landry et al., 2012; Roberts & Kaiser, 2012}). Coaching in these studies was used to describe an interaction that occurred after initial parent training was completed. Coaching was used as an interaction to support parents to achieve implementation fidelity in the use of a specific intervention curriculum with their children. Coaching most commonly included directive feedback on how parent could work to improve implementation of the strategy. Thus, this type of coaching can be characterized as an intervenor-directed coaching protocol that primarily used the strategies of direct teaching and modeling by the intervenor.

Characteristics of parents, children, and coaches

Parents

Table 2 summarizes the characteristics of parents. Eight studies reported parent gender, with 92% of the participants being mothers. The majority of the households were two-parent households. The reported ages of the parents ranged from 30 to 45 years, with a mean of 33 years. Seven studies reported race, with 53% of participants being European American, 22% Hispanic, 21% African American, and 3% Asian. Reported parental education indicated that approximately half of the families were high school graduates and half had some college to graduate-level college education. Regarding income in reported studies, families were typically at low- to middle-income range.

Table 2: Characteristics of Parents and Children

Table 2: (continued) Characteristics of Parents and Children

Children

Table 2 summarizes the characteristics of children. All studies included children who were at least between 3 and 36 months old. Of the studies reporting gender, 60% were male. Of the eight studies, two studies restricted their samples to children who experienced autism spectrum disorders and included 15 children (^{Vismara et al., 2012; Welterlin et al., 2012}). The remaining six studies reported children with or at substantial risk for a range of developmental disabilities or delays. The studies reporting etiology included autism, Down syndrome, cerebral palsy, expressive and receptive language impairment, significant intellectual disability, and significant motor disability.

Coaches

Coaches included teachers, therapists, and researchers who served in the role of coaches to parents. Two studies (^{Salisbury & Copeland, 2013; Vismara et al., 2012}) reported on characteristics of coaches. All coaches held a minimum of a masters' degree. Only one study reported race, gender, and years of service, with five of seven coaches being European America and six of seven coaches being female. These coaches ranged in years of service from 4 to 17 years (^{Salisbury & Copeland, 2013}).

Settings, contexts, dosage, and professional development

Table 1 summarizes the settings, context, dosage, and professional development related to coaching with parents in EI.

Setting

Each of the eight studies included a home component. This was an inclusion criterion for review. Five of the eight studies included sessions conducted only in the home (^{Blauw-Hospers et al., 2011; Landry et al., 2012; Salisbury & Copeland, 2013; Vismara et al., 2012; Welterlin et al., 2012}). In three studies, sessions both in the clinic and in the home were included (^{Kaiser & Roberts, 2013; Roberts & Kaiser, 2012; Romski et al., 2011}). The session site in each of these three studies transitioned from the clinic to the home as the intervention progressed.

Context

Only one study was conducted within the context of a Part C program (^{Salisbury & Copeland, 2013}). The participants were families currently enrolled in the program. The other seven studies were conducted in the context of a university study, with parent volunteers recruited from other ongoing research studies or from the community (e.g., flyers or physician referral).

Dosage

The coaches based the number of sessions on the treatment condition outlined in the study. They typically held weekly sessions lasting from 20 to 90 min and ranging from 11 to 36 weeks in duration. One study included biweekly sessions of 1 hr for 3 months (^{Roberts & Kaiser, 2012}). In two studies, the sessions were dependent on individual child and family characteristics and ranged from one to two times per week (^{Blauw-Hospers et al., 2011}; Salisbury & Copeland, 2012).

Professional development

Four studies described significant effort devoted to training of coaches. One study described extensive training, as part of the existing program, in the use of caregiver coaching as an approach component (Salisbury & Copeland, 2012). This training included participation in weekly reflective supervision sessions and a bimonthly community of practice. These coaches were trained to fidelity to use the 10 coaching strategies outlined by ^{Friedman et al. (2012)}. Other studies described training coaches to fidelity in the use of coaching strategies using face-to-face training and the use of a protocol manual to guide ongoing use of coaching (^{Kaiser & Roberts, 2013; Roberts & Kaiser, 2012; Romski et al., 2011; Vismara et al., 2012}). Examples of activities to ensure fidelity in parent training strategies include (a) use of video recording and (b) use of self- and observer-implemented parent training implementation assessments. These studies offered no additional information on what the training entailed.

Parent and child outcomes

Table 3 summarizes parent and child outcomes.

Table 3: Parent and Child Outcomes

Table 3: (continued) Parent and Child Outcomes

Parent outcomes

The majority of parent outcomes reported were positive. Most outcomes focused on parents' ability to attain fidelity in the use of strategies related to the intervention protocol (^{Kaiser & Roberts, 2013; Landry et al., 2012; Roberts & Kaiser, 2012; Vismara et al., 2012; Welterlin et al., 2012}). In addition, these studies noted some personal outcomes for parents including (a) increased ratings of responsiveness, (b) increased positive parental perceptions of success about child communication, (c) decreased perception of severity of child's language difficulties, (d) stronger working alliances with primary interventionist, and (e) decreased parental stress. Two of the relationship-directed process studies reported outcomes for parents including (a) increased perception of personal capability, (b) parent-reported increased perception of motor and social skills in their children, (c) increased feelings of self-efficacy, (d) increased engagement by parents, and (e) increased incorporation of educational actions (i.e., spending brief amounts of time playing in child-preferred activities; incorporating variation and trial and error in daily activities) into daily routine (^{Blauw-Hospers et al., 2011; Salisbury & Copeland, 2013}).

Two studies reported outcomes that were less positive. One study reported a decreased parent perception of child language ability when parents did not receive the option of augmented communication support as compared with parents who did (^{Romski et al., 2011}). A second study reported decreased maintenance in the use of intervention strategies by parents for those routines that were not among those directly taught by the coach during intervention. Researchers noted a need to focus on increased generalization across more routines during intervention (^{Kaiser & Roberts, 2013}).

Child outcomes

Most of the studies reported developmental gains for children, with some of the improvements reaching statistical significance. Two studies noted gains across all developmental domains (i.e., gross motor, fine motor, cognitive, communication, self-help, and social-emotional; ^{Blauw-Hospers et al., 2011; Salisbury & Copeland, 2013}). Other studies noted gains in (a) language (^{Kaiser & Roberts, 2013; Landry et al., 2012; Roberts & Kaiser, 2012; Romski et al., 2011; Vismara et al., 2012}), (b) social-communicative strategies (^{Vismara et al., 2012}), (c) motor behaviors (^{Blauw-Hospers et al., 2011}), (d) book reading skills (^{Landry et al., 2012}), (e) improvement in child independent work skills (^{Welterlin et al., 2012}), and (f) improvement in adaptive skills (^{Welterlin et al., 2012}). One study that focused on parent fidelity to intervention noted reduced child outcomes between 6 and 12 months (^{Kaiser & Roberts, 2013}). This coincided with a reported reduction in parents' use of strategies.

DISCUSSION

The primary purpose of this article was to review the current definitions, descriptions, and use of coaching with parents alongside the parent and child outcomes resulting from the use of coaching with parents. We found eight articles that met the inclusion criteria for our synthesis.

Early intervention studies, as a whole, offer a continuum of definitions and descriptions for the use of the term “coaching with parents.” This continuum appears to have, at one end, a relationship-directed process and, at the other end, an intervener-directed protocol. The studies in this synthesis fall along this continuum sometimes sharing common elements and at other times using the same term, “coaching with parents,” to mean two very different activities. Five studies in this synthesis would fall primarily toward the intervener protocol end of the continuum. The researchers in these studies have developed an evidence-based intervention that they intend to coach parents to replicate with fidelity. One example is the ^{Kaiser and Roberts (2013)} study in which parents are trained to use the Enhanced Milieu Teaching strategies. The researchers in these five studies determine which intervention to use, as well as how and when to use it. These studies have many similarities with traditional parent training studies except it appears that the term “coaching” is now used to describe direct feedback given to parents as they try out prescribed strategies.

The other three studies outline a primarily relationship-directed process. These studies view coaching as a process that encompasses the ideals of family-centered practice. An example of this is the caregiver coaching found in the ^{Salisbury and Copeland (2013)} study. The what, how, and when to intervene result from a shared decision-making process. Thus, coaching is a different process than the traditional parent training.

In the studies that described the process for training coaches, four of the studies (^{Kaiser & Roberts, 2013; Roberts & Kaiser, 2012; Salisbury & Copeland, 2013; Vismara et al., 2012}) included fidelity training for the use of coaching. Only the ^{Salisbury and Copeland (2013)} study is set in the context of a Part C EI program. In terms of dosage, sessions with parents typically occurred at least weekly ranging from 20 to 90 min in length.

About characteristics, first, with respect to families' characteristics, coaching is used across families with diverse backgrounds (e.g., race, social economic status, and education). With respect to children's characteristics, coaching is used with parents of children with a wide range of diagnoses and functioning levels (e.g., autism, cerebral palsy, Down syndrome, and significant intellectual disability). In reference to the characteristics of the coaches, a variety of disciplines were represented. All coaches held a minimum of a master's degree, and all received some level of professional development to use coaching with parents.

For parent and child outcomes, the researchers reported that the effectiveness of coaching with parents is one contributing factor in a multicomponent intervention for the majority of the parent and child outcomes measured. The most frequently occurring parent outcome was mastery of the strategies that comprised the intervention curriculum. Other outcomes were (a) increased positive perception of their child's ability, (b) increased ratings of parent responsiveness toward their child, (c) decreased parental stress, (d) increased perception of personal capability and feelings of self-efficacy, and (e) a stronger alliance between the parent and the coach. Finally, for the majority of the studies in which coaching is used with parents, children made developmental progress. Children with varying diagnoses and levels of individual functioning made developmental gains in a variety of domains.

Implications for researchers

First, future studies should clearly define and then examine the effect of coaching as separate component. As previously mentioned, regardless of how each study defines coaching, it is difficult to untangle the impact of coaching with parents in EI from the other intervention components in these multicomponent studies. For example, in the studies by Kaiser and Roberts, coaching is only one of several activities used in parent training. In addition to in vivo coaching during intervention implementation with children, parents attended classes, reviewed modules, and observed therapists using the strategies with their children. In the study by ^{Salisbury et al. (2013)}, coaching is one component of a triadic service delivery approach, which also includes family-guided, routines-based intervention and embedded intervention in naturally occurring activity settings. “When EI providers are uncertain about what constitutes coaching and what it looks like as a collaboration practice, it becomes considerably more difficult for them to understand what about their practice needs to change to effectively coach caregivers” (^{Friedman et al., 2012}, p. 63). These authors join in the call for consensus on a universal definition and description of coaching along with consideration of settings, contexts, dosage, and professional development when using coaching with parents in EI. There is much variation in how coaching is described in the literature. For example, coaching, as described in the work of ^{Rush and Shelden (2011)}, has been described at times as an interaction style for working with parents and at other times as a service delivery model (^{Chai, Zhang, & Bisberg, 2006; McWilliam, 2012}). A consistent operational definition across researchers and practitioners is needed to build a strong research base.

Second, parallel to the need for a clear definition of coaching is the need to detangle the newer emphasis on coaching from traditional emphasis on parent training. If coaching and parent training are two different approaches, then these differences must be made clear.

Third, researchers should explicate desirable short-term, intermediate-, and long-term child and family outcomes that are expected as a result of coaching and then provide psychometrically sound measurement of these outcomes. In reference to child outcomes, expectations usually involve increase in developmental skills to promote functioning in daily routines and activities. Paired with measurement of developmental skills, outcomes should be measured on successful participation of the child in home and community settings through a focus on the three child outcomes mandated by the Office of Special Education Programs. In addition, because previous literature links coaching to parent–child relationship development, research needs to provide data on exactly how use of coaching impacts the parent–child relationship from the child and parent's perspective. Improvements in social-emotional functioning, including the children's attachment to their parents, should be one outcome studied.

Also related to outcomes, more work needs to be done in investigating the impact of coaching on family outcomes. The studies in this research analysis most often document parent outcomes as they pertain to achievement of child outcomes (e.g., parents being able to replicate strategies with fidelity). ^{Bailey, Raspa, and Fox (2012)} argue that a focus on attainment of child outcomes is necessary but not sufficient, given the focus on support of the family that EI can offer. Measurement of coaching with parents should include the impact of this practice on the original five recommended family outcomes developed from the work at the Early Childhood Outcomes Center (^{Bailey et al., 2006}). These five outcomes include how families (a) understand their child's strength, abilities, and special needs; (b) know their rights and advocate effectively for their children; (c) help their child develop and learn; (d) have support systems; and (e) access desired services, programs, and activities in their community. This measurement must be done in such a way that it is clear what the impact of coaching with parents is on these outcomes. Within the studies of this synthesis, it is difficult to ascertain which parent outcomes were a result of coaching and which were a result of other intervention components.

It is worth noting that another often-missing measure is the impact of coaching on parents' quality of life. Wallace and Rogers pointed out:

Parents of infants and toddlers with ASD are not community intervention providers; they are parents of an infant or toddler just diagnosed with a serious chronic developmental disorder. They are experiencing a tragic and life-altering event, one with long-term effects on everyone in the family. They need information, support, and services for their child. How do we support them in this part of their lives and pass on intervention skills? (2010, pp. 1316–1317)

In relation to parental quality of life, the relationship-directed process studies reported greater gains. One of the studies reported that the intervention “did not disrupt or interfere with parents' style of relating to or caring for their children” (^{Vismara et al., 2012}). Parents in the Salisbury and Copeland study reported, “They had developed a stronger sense of self-efficacy, direction, and support” (2013, p. 73). Parent shared statements such as “It gave me a sense of security, gave us guidance” and “I learned not to panic when unpredictable things happen.” In extending this work, one potential measure to document family quality of life, resulting from the use of coaching with parents, is the Beach Center Family Quality of Life Scale (^{Poston et al., 2003}). This psychometrically strong scale contains five subscales: (a) Family Interaction; (b) Parenting; (c) Emotional Well-being; (d) Physical/Material Well-being; and (e) Disability-Related Support (^{Hoffman, Marquis, Poston, Summers, & Turnbull, 2006}).

Fourth, the majority of studies did not offer coaching adaptations in light of specific individual child and parent characteristics. Seven of the studies offered the same protocol and the same dosage to each family, with little discussion of individualization in light of parent or child characteristics. The study by ^{Salisbury and Copeland (2013)} offered more individualization as it was conducted with families that had an Individualized Family Service Plan in place. Within the studies, 53% of families were of European American background and 47% were families from culturally or linguistically diverse (CLD) backgrounds; however, none of the studies describe culture or language as a variable to consider when coaching CLD families. The need for culturally responsive coaching is an issue that has received little discussion in the literature to date. This issue needs to be investigated in ensure that coaching with parents is a strategy that is responsive to the long-held EI commitment and legal requirement to design services that meet the unique individual needs of every child and family.

Fifth, only one study focused on the use of coaching in a Part C setting. Given the nationwide movement by technical assistance/professional development providers, state leadership, and local program leadership to promote coaching practices within Part C programs, coaching in Part C contexts should be a priority for research (^{NECTAC, 2008, 2011}). Research in this setting should focus on how coaching with parents is implemented while ensuring compliance with the IDEA, in particular to focus on how coaching with parents is conducted to ensure protection of parental rights and procedural safeguards. In addition, coaching in the context of weekly home visits over the entire course of a child's enrollment brings unique challenges and opportunities for children, parents, and coaches that differ substantially in terms of topic and focus from that found in a short-term research-led intervention. These differences need to be articulated and then measured.

Finally, related to the issue of the need for more research in the context of Part C settings is the need for more research on the type and intensity of professional development required for coaches in these settings to maintain fidelity to coaching practices. Friedman et al. postulated that providers have had “limited training in how to coach caregivers ... to be effective, EI providers need to strengthen and broaden their specific knowledge and skills about how to collaborate with and coach caregivers during intervention sessions to build caregiver capacity” (2012, p. 63). However, emerging research indicates that with “professional development and ongoing support for using collaboration and coaching practices,” (^{Friedman et al., 2012}, p. 63) providers can be effective at coaching with parents. The majority of research studies in this synthesis showed evidence of extensive training for coaches. The Salisbury and Copeland (2012) study, in particular, gave specific detail on what activities occurred within one program to achieve high-quality coaching. Coaches in this program received professional development in caregiver coaching along with and situated within other core elements of EI (e.g., family-guided routines-based intervention). In addition, they receive ongoing support in the form of participation in weekly reflective supervision sessions and in a bimonthly community of practice meeting (^{Salisbury & Copeland, 2013}).

Implications for practitioners

Although coaching with parents is often used to increase children's developmental outcomes, it also shows promise for improving parent outcomes. Although only a few studies used coaching with parents to enhance parent outcomes, positive outcomes resulted in these areas. This evidence suggests that coaching with parents can facilitate outcomes for parents such as increased positive perceptions of their ability to respond to their children's needs, increased self-efficacy, and stronger sense of partnership with the coach with whom they are working. Therefore, coaching with parents is a promising practice.

The synthesis did not provide one universal definition of coaching for practitioners to follow. However, the synthesis found that both coaching as a relationship-directed process and coaching as an intervenor-directed process had a positive benefit on child outcomes. Furthermore, these findings support that whichever process is used, professional development in the use of coaching with parents and implementation with fidelity are essential in obtaining positive parent and child outcomes.

The synthesis cannot yet provide a confirming answer as to how intensive coaching with parents must be implemented to show positive impact on child and family outcomes. The synthesis found, however, that the most commonly reported frequency was weekly between 20 and 90 min. This finding supports a fairly high level of frequency with families.

Coaching with parents is feasible to implement in the home setting with parents and children who reflect a variety of characteristics. Parents, including those from low socioeconomic status backgrounds, showed increases in parent outcomes. In addition, children with a variety of developmental levels and across different diagnoses, including children with significant disability, showed developmental gains in multiple areas.

The synthesis results do not provide direction for how coaching may be individualized according to individual parent and child circumstances and characteristics. Until such a time that this research is conducted, coaches must rely on what should already be in place in EI programs to guide their coaching conversations and decisions in terms of existing focus on family-centered practices that honor the uniqueness of each family and child.

Limitations

One limitation is the restriction of search criteria to only those studies that use the word “coaching” within the title or text of the article. We acknowledge that sometimes other terms are used to describe strategies or components similar to coaching. As stated in the purpose, this synthesis is designed to review those studies that use the term “coaching.” This is the term currently used by and with EI practitioners and the term that needs to be more fully defined and described.

A second limitation is the inclusion of only articles that were published from 2011 to the present. Narrowing the criteria this way resulted in only eight articles being identified for review. Expansion of these criteria would have resulted in the identification and inclusion of additional articles. Our basis for restricting the date was to focus on those studies that were conducted after the first generation of information on coaching to document how researchers are translating what was learned about coaching in the first decade into intervention and new research questions.

CONCLUSION

The overall use of coaching with parents in EI supports previous research that indicates that parents can successfully implement interventions with their children; as a result, children's skills improve across the developmental domains measured. To effectively discuss and describe coaching in EI, a key set of tasks is to identify a universal definition for coaching and to explore the continuum of coaching approaches/strategies and make a decision, as a field, that coaching is a relationship-directed process or an intervener-directed process or whether it can be defined as both. This will most likely involve detangling the newer use of coaching from traditional parent training. A universal definition of coaching will enhance the quality of research, which can accurately measure the impact of use of coaching with parents on child and family outcomes. Most importantly, consensus must be sought on the intended outcomes of coaching for children and parents in EI. What is known is that coaching, to this point, seems to do no harm. Infants and toddlers make developmental progress comparable with progress made with therapist-directed interventions, sometimes with increased outcomes and sometimes with comparable gains. What is not known is what are the comprehensive outcomes for parents. The synthesis documents that parents learn to implement interventions. The review also revealed that, in at least one instance, the parent–child responsiveness increased. In another instance, parents reported increased self-efficacy. The synthesis did not offer enough information about the benefit to parents through achievement of family outcomes. If EI is to be of value to families, to match the call set forth by Bailey et al. to “go beyond just family satisfaction with services and increase in child outcomes, to helping families” (2012, p. 216), then measurement of all aspects of intervention including coaching with parents must be from the family outcome lens.