Reflective Supervision and Consultation and its Impact Within Early Childhood-Serving Programs: A Systematic Review

Abstract

Reflective supervision and consultation (RS/C) is regarded as best practice within the infant/early childhood mental health field. Benefits of RS/C on the early childhood workforce and children and families have been demonstrated through case studies, conceptual pieces, and individual research studies. However, findings across studies have not been summarized using gold-standard methodology, thus the state of existing empirical support for RS/C is unclear. This systematic review examined the collective evidence for RS/C across diverse early childhood-serving programs. Electronic databases were searched to identify studies investigating associations between RS/C and professionals’ reflective capacity and well-being, child/family outcomes, and implementation factors. Twenty-eight papers were identified. Studies showed positive associations between RS/C and early childhood-serving professionals’ reflective capacity and well-being, with qualitative studies reporting more consistent results than studies using quantitative methods. Many methodological limitations were identified, including incomplete reporting of study designs and participant characteristics, variability in outcome measures, and lack of randomization and comparison groups. Furthermore, few studies examined child and family outcomes. Therefore, while RS/C shows great promise, it was difficult to ascertain its overall effectiveness from an empirical standpoint. Establishing RS/C as an empirically supported approach will be possible with more rigorous research.

Supporting infant and early childhood mental health (IECMH) through the provision of home- and community-based services is increasingly recognized as a catalyst for healthy development across the lifespan (Zeanah & Zeanah, 2019). This is accomplished in multiple ways. Some early childhood programs, such as home visiting, were explicitly designed to serve children and families experiencing risk factors like poverty, child welfare involvement, and parental substance use; a goal of many home visiting programs is to facilitate safe, responsive caregiver-child relationships in which children may thrive (Stoltzfus & Lynch, 2009). Other programs, such as early intervention, may screen children for developmental delays or disabilities and initiate services to address these specific needs (Sapiets et al., 2021). Early care and education programs may be considered more universal, in that they promote positive social-emotional and academic development in children with a range of backgrounds (Durlak et al., 2011). Although this is a somewhat reductive overview of early childhood systems and programs, these programs all aim to promote healthy development in young children, with approaches as diverse as the numerous settings in which early childhood work takes place. Despite these differences, high-quality early childhood programs generally operate from the unifying belief that relationships are central to young children’s well-being, and that by cultivating supportive relationships between children and their caregivers, children will flourish, and future problems will be prevented (Heffron & Murch, 2010; Heller & Gilkerson, 2009; Tomlin et al., 2016).

From an IECMH perspective, one effective way to foster strong relationships between children and caregivers is by building strong relationships between caregivers and the people who support them (Tomlin et al., 2016). For example, a parent who experiences a supportive relationship with their home visitor may be better able to extend similar support to their child; likewise, a home visitor who enjoys a supportive relationship with a supervisor may be better able to show such support to a parent. This is known as the parallel process, a concept that underlies the highly regarded practice of Reflective Supervision or Consultation (RS/C; Tomlin et al., 2016). We, the authors, fully embrace RS/C as an approach with immense potential to support early childhood-serving professionals and better their practice, as we have experienced and witnessed the benefits of RS/C in our own and others’ professional lives. It is from this stance that we seek to understand in what ways RS/C works for supervisors, supervisees, and ultimately children and families. Thus, we conducted a systematic review spanning early childhood-serving settings to determine whether and how RS/C is associated with its intended outcomes.

Defining and Contextualizing RS/C

In this review, we use the combined acronym RS/C when referring to the distinct reflective approach that can be used by either supervisors or consultants, and use the term “supervisor” and “consultant” interchangeably. When referencing studies that investigated Reflective Supervision or Reflective Consultation specifically, we use the full name of the approach rather than the acronym. The term “Reflective Supervision” generally indicates that a staff member within the direct service provider’s organization, typically a more seasoned individual who also has administrative and/or clinical oversight responsibilities, is providing the supervision (Alliance for the Advancement of Infant Mental Health [AAIMH], 2018). In practice, reflective supervision is often delivered within a blended model incorporating administrative, clinical, and reflective goals. The term “Reflective Consultation” implies that this same reflective approach is being provided by someone outside of the provider’s organization (AAIMH, 2018). Moreover, a reflective consultant may not hold the same level of responsibility as a reflective supervisor.

RS/C is a complex construct that has been described in many nuanced ways in the literature (Heffron & Murch, 2010; Heller & Gilkerson, 2009; Tomlin et al., 2016). Briefly, RS/C is both a way of being—embodying that "how you are” in relationships is as important as “what you do” (Pawl & St. John, 1998)—and a distinct set of skills and strategies that supervisors use (Heffron & Murch, 2010; Heller & Gilkerson, 2009). Specifically, if supervisors can consistently hold their supervisees, remaining present to their emotions, cherishing their strengths, and attending to their needs with genuine curiosity, compassion, and warmth, supervisees are more likely to experience enhanced well-being and subsequently engage in relationships with families in this same manner (Tomlin et al., 2016). The skills and strategies that characterize RS/C are actions that supervisors may take, as well as building blocks in the development of an open, reliable reflective alliance; thus, skills and strategies are inextricably intertwined with a way of being.

For supervision or consultation to be reflective, it must be grounded in three core principles: regularity, collaboration, and reflection (Heffron & Murch, 2010). Regularity implies that RS/C occurs regularly and predictably, which creates a consistent space for the supervisee to think about their work proactively and in-depth. Collaboration embodies a respectful approach to supervision, wherein the supervisee’s perspective is valued, time is spent in collective “wondering,” and challenges are faced together, resulting in the co-creation of plans. Reflection is slowing down and bringing attention to feelings that the work elicits, so that the supervisee learns that their emotions matter and inform their interactions, while also pausing to consider the emotions, past experiences, current circumstances, and complex values, beliefs, identities, and biases that each person holds.

Associations Between RS/C and Reflective Capacity and Well-Being

While reflection is an internal process that involves self-awareness, it is often best facilitated by an experienced person who can offer a safe, supportive space in which the individual can explore (Barron et al., 2022a). Based on theory, anecdotal evidence, and burgeoning research, it is thought that the focus and collaboration around reflection begins to enhance the individual’s reflective capacity (also called reflective functioning or reflective self-efficacy), which in turn allows reflective practice to unfold, wherein the individual puts reflection into action through their decision-making, problem-solving, and responses to challenge (Barron et al., 2022a). Said another way, the reflective supervisor helps a provider process the meaning of feelings that arise during interactions with children, families, and colleagues. Additionally, the supervisor holds space for the provider to explore difficult situations and collaboratively solve problems, which over time results in professional growth.

In addition to growth in reflective capacity, and subsequently professional practice, RS/C is also thought to influence the well-being of early childhood-serving providers. This notion is supported by anecdotal experiences, some research evidence, and theory. Work in the early childhood realm involves navigating complex systems, understanding environmental and cultural influences, and attuning to the interconnected web of relationships, emotions, and histories that each person brings to sensitively implement services. Additionally, early childhood-serving providers are often operating under the strain of personal, organizational, community, and societal stressors, which have only been exacerbated by the COVID-19 pandemic (Morelen et al., 2021). For instance, providers are exposed to a host of adversities experienced by families, including trauma and structural inequities, like poverty, racial discrimination, and oppression (Fitzgibbons et al., 2018). Many providers also personally experience these issues, and are at risk for job stress, secondary traumatic stress or compassion fatigue, mental health difficulties, and burnout (Eaves et al., 2021; Eaves Simpson et al., 2018). Supervision alone cannot relieve all of these concerns without attention to larger, systemic issues (Eaves et al., 2021). However, RS/C may help mitigate effects of stress and promote well-being among early childhood-serving providers. Exploration of how RS/C affects the interconnected—and likely bidirectional—constructs of reflective capacity and well-being is essential, and as a first step, this review has compiled studies focused on these outcomes.

The Importance of Summarizing Empirical Evidence for RS/C

RS/C has been viewed as best practice since the 1990s and is therefore incorporated—and often mandated—as a key feature of many early childhood-serving programs (Eggbeer et al., 2010; Osofsky & Weatherston, 2016; Tomlin & Heller, 2016). RS/C has largely been implemented on the basis of theory and anecdotal evidence, and while this evidence is powerful, rigorous research is needed to truly establish RS/C as an empirically supported practice within the field. This evidence base may instill greater confidence in state and federal stakeholders, open doors for more funding opportunities, promote effective workforce development policies, and reassure families that they are receiving the best care possible.

The literature on RS/C has grown steadily since the 2000s, yet findings across these studies have not been summarized, which prevents the field from making research-informed assertations about RS/C that are drawn from a collective body of evidence. Therefore, we have conducted a systematic review to exhaustively summarize the current evidence pertaining to RS/C across early childhood settings. Our goal was two-fold: first, we sought to clarify the current state of the research on the impact of RS/C within early childhood-serving programs; second, through the process of elucidating the current state of the research, we hoped to provide the field with a roadmap of what has been done and what areas of inquiry remain, with the goal of guiding future research efforts and elevating the overall evidence base for RS/C.

Research on RS/C has lagged behind research on other early childhood-focused practices in large part due to inherent difficulties in measuring a practice that is as much a way of being as it is specific skills and strategies (Eggbeer et al., 2010). Understanding how RS/C affects supervisors, providers, and children and families—and the relationships between each of these entities—is of course contingent on how RS/C is measured. Several comprehensive resources have detailed what is known about measurement of RS/C, and put forth ideas for addressing current measurement issues (Low et al., 2018; Shea et al., 2016; Tomlin & Heller, 2016; Watson et al., 2016b); thus, the current review focused solely on outcomes related to RS/C. Notable efforts are being made to describe and operationalize the nature of interactions between a supervisor and supervisee during reflective supervision, such as the Reflective Interaction Observation Scale (RIOS; Watson et al., 2016b). This work is essential to capturing what it looks like to be engaged in RS/C, and in turn how RS/C affects those who experience it.

Purpose of the Review

The purpose of this systematic review was to aggregate and synthesize findings from studies focused on how RS/C is related to reflective capacity and well-being among early childhood-serving professionals. We were also interested in how training to provide RS/C affected these same outcomes at the supervisor level. However, many studies did not differentiate between receiving training in how to deliver RS/C to providers, and receipt of RS/C itself. Therefore, we did not necessarily draw distinctions between training in RS/C and receiving RS/C in this review, though that should be clarified as this research area advances. To the extent that they were available, we also included studies that explored associations between RS/C and child or family outcomes, or how RS/C affects interactions between the provider and the family. Furthermore, we included studies that sought to understand the factors that affect the successful implementation of RS/C in early childhood-serving programs, as this literature holds important implications for adoption and use of RS/C in relevant settings. Finally, we examined the overall quality of the existing literature to inform and guide the next wave of research efforts. This review addressed the following research questions:

1. What is the current evidence for associations between RS/C and early childhood-serving professionals’ reflective capacity and well-being?

2. What is the current evidence for associations between RS/C and child/family outcomes?

3. What individual, relational, and organizational factors influence the implementation of RS/C within early childhood-serving programs?

Methods

This systematic review is reported in accordance with PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses) guidelines (Page et al., 2021). The research questions and inclusion/exclusion criteria for the review were determined using the PICOS (Population, Intervention, Comparison, Outcomes, Study Design) framework (Eriksen & Frandsen, 2018). Furthermore, review methods were established prior to the conduct of the review.

Eligibility Criteria

To be eligible for this review, studies needed to meet the following inclusion/exclusion criteria within each category of the PICOS framework:

(a) Population/setting: study participants were professionals (supervisors and/or direct service providers) working in early childhood settings, including child welfare, early intervention, home visiting, early care and education, and community mental health agencies; studies that included professionals working primarily in settings with older children or adults (≥ 6 years old) and studies that did not clearly report on the setting or ages of children served were excluded.

(b) Intervention: participants were either receiving RS/C or completing training in how to deliver RS/C; the description of RS/C in study manuscripts was consistent with RS/C as described within the best practice guidelines put forth by AAIMH (2018). Practically, articles had to mention at least two constructs from this document (e.g., parallel process, relationship for learning), reference IECMH, and cite at least one foundational work pertaining to RS/C. Studies that included RS/C as a component of a larger program or model, such as IECMH consultation, were excluded, as effects of RS/C could not be assessed separately.

(c) Comparison: studies could include any comparison condition, including no comparison condition.

(d) Outcomes: primary outcomes for Research Question 1 included reflective capacity, skill, or self-efficacy; reflective functioning or insightfulness; constructs related to professional or personal well-being, such as job/compassion satisfaction, job stress/compassion fatigue, secondary traumatic stress or vicarious trauma, and burnout. Familiarity with the early childhood workforce and relevant literature suggested that these were the most salient outcomes for this population. Primary outcomes for Research Question 2 were any outcomes that pertained to children/families. Primary outcomes for Research Question 3 included barriers and facilitators to implementation of RS/C. Studies that focused solely on measurement (e.g., factor analysis of measure items, validity and reliability of measures) were excluded.

(e) Study design: all quantitative study designs (e.g., experimental, quasi-experimental, observational) with statistical indicators presented in study manuscripts were included; all qualitative study designs with described methods were included. All study designs, not just randomized controlled trials (RCTs), were included in the review given the lack of randomized studies in this topic area.

(f) Origin: any country.

(g) Language: published in English.

(h) Date: there were no limitations on publication date to assess the complete body of literature on RS/C; searches were done in January 2021 and updated in September 2022 (last date searched was September 17, 2022).

Search Strategy

Consistent with PRISMA guidelines, the literature search was conducted in three electronic databases deemed most relevant to RS/C and with wide reach, covering most publications (and dissertations) within the fields of health and education: Web of Science, PubMed (Medline), and APA PsycINFO. Google Scholar was also used to identify relevant studies, and the first author received Google Scholar alerts for any new paper published including “reflective supervision,” “reflective consultation,” or “reflective practice” in the title, abstract, or keywords. Additional searches were conducted using the search functions within Infant Mental Health Journal and Zero to Three Journal, given the focus of these journals on RS/C. The references of all included studies were also examined for additional publications. Search terms were “reflective supervision,” “reflective consultation,” and “reflective practice;” these terms were searched separately to capture all relevant publications.

Screening and Selection

All records identified by searches were exported into Endnote and duplicates were removed. Records were first screened by titles and abstracts by three independent reviewers (the first author and two trained research assistants [RAs]) with excellent interrater reliability (92% agreement). Records that passed this first phase of screening were then retrieved as full-text articles. Full-text articles were screened by two independent reviewers (first author and RA) and checked by the other RA; any discrepancies were discussed by the full group of reviewers to reach consensus. See Figure 1 for details on the screening and selection process, including reasons for article exclusions, per PRISMA guidelines.

Figure 1.

PRISMA flowchart illustrating papers identified and excluded at each stage.

Data Extraction

Two reviewers (first author and RA) independently extracted data from all papers included in the review using a standardized form in Excel, while a third and fourth reviewer checked for accuracy (last author reviewed quantitative articles, and second author reviewed qualitative and implementation-based articles). Extracted data included: author(s); publication date; study aims; sample and setting characteristics; method; intervention characteristics; primary and secondary outcomes.

Study Quality Assessment

The quality assessment of included articles was conducted using the Joanna Briggs Institute (JBI) critical appraisal tools (Moola et al., 2017), which allow assessment of cross-sectional, cohort, and qualitative studies, among other designs, using the same rating system. Two reviewers independently reviewed the quantitative studies, including those that were implementation-focused (first and last authors) and the qualitative studies, including those that were implementation-focused (first and second authors), then discussed discrepancies in ratings until a consensus was reached. Interrater reliability was good, with 93% agreement on overall rating for the quantitative studies and 88% agreement on overall rating for the qualitative studies.

For each item in the relevant JBI checklist, reviewers indicated whether the study appropriately addressed the construct (Yes), did not address the construct (No), or the information was not reported (Unclear), or was not applicable. Reviewers referred to the quality assessment tool’s guidelines for each construct. Assessed constructs varied somewhat based on whether the cross-sectional, cohort, or qualitative checklist was used. The cross-sectional and cohort checklists contained items such as clarity of research question; selection bias; study design (e.g., use of valid and reliable measures, measurement timing, intervention delivered consistently across participants); retention; confounders; statistical analysis. The qualitative checklist contained items such as congruity between philosophical perspective and research methodology; congruity between methodology and data analysis/interpretation; whether influence of the researcher was addressed; adequate representation of participants’ voices. An item assessing ethical issues “were ethical issues considered (e.g., informed consent, approval from appropriate ethics committee, discussion of consent or best interest decisions reported)?” was added to each checklist, except the JBI Qualitative Studies Checklist, which already included this item. An overall quality rating was assigned based on the reviewers’ assessment of the individual constructs: Strong, if 75% or more of the applicable items were rated as “yes,” Mixed if 50-74% of items were rated as “yes,” and Weak if less than 50% of items were rated as “yes.” It should be noted that because cross-sectional and cohort checklists were used, studies could receive a favorable rating even if they did not include a control group or randomize participants to conditions; this may bias study ratings to appear higher than potentially warranted. Similarly, checklists did not include items specific to RS/C, so a study could receive a favorable rating even if they did not adequately describe what RS/C looked like in their study, for example.

Synthesis of Evidence

Narrative synthesis was used to summarize evidence from included studies due to heterogeneity in the studies’ research design, methodology, outcomes, and delivery of RS/C. Further, many studies were qualitative in nature and there were no RCTs. Thus, meta-analysis was not appropriate.

Results

Overview of Articles

Overall, 28 articles representing 24 unique samples were included in the systematic review. Fourteen of these 28 articles presented quantitative results (i.e., Research Questions 1 and 2; see Table 1), 17 presented qualitative results (i.e., Research Questions 1 and 2; see Table 2), and 6 presented results pertaining to implementation of RS/C (i.e., Research Question 3; see Table 3). Of note, several articles presented both quantitative and qualitative results (n = 5) or examined implementation-related factors in addition to outcomes (n = 4), therefore these articles are included in more than one table. The publication dates ranged from 2007 to 2022, with 63% of articles published in 2017 or later. Sample sizes ranged from N = 20 to 139 (M = 55) for quantitative analyses, N = 5 to 97 for qualitative analyses (M = 33), and N = 24 to 139 (M = 71) for implementation-based analyses. Articles varied substantially on the participant characteristics they reported; of the studies that assessed gender and race/ethnicity, the majority of participants self-identified as White, non-Hispanic women. Along with gender and race/ethnicity, education and years of experience in the field are reported in Tables 1-3 for those studies that included them. Some studies assessed supervisory experience (when participants were supervisors), previous experience with RS/C, and IECMH Endorsement status; however, these were inconsistently reported and thus are not included in the tables.

Table 1.

Summary of Included Articles with Quantitative Analyses

STUDY ID Study Design	Setting	Participant Characteristics	RS Experience	Primary Constructs & Measures	Primary Findings	Study Quality
±Begic et al. (2019)a Cross-sectional	Home visiting Idaho, USA	N = 27 home visitors; for n = 20 who completed survey: 7% did not have a bachelor’s degree, 56% had a bachelor’s degree, 11% had a master’s degree	Details regarding the RS experience were not provided	Quality of RS relationship using the Supervisory Working Alliance Inventory-Trainee; secondary traumatic stress and burnout using the Professional Quality of Life Scale; intention to quit using a 2-item scale	Higher quality RS relationships were associated with lower levels of secondary traumatic stress, burnout, and intention to quit	Mixed
±Brown (2016)b Cohort (one group before and after)	Early care and education California, USA	N = 34 teachers; for N = 37 who completed larger study: 97% female; 73% Hispanic/Latinx, 24% White, 3% Asian; 6% with no high school degree, 18% with high school degree or GED, 47% with some college or 2-year degree, 29% with bachelor’s degree	Received RS via small group weekly (45-90 minutes) for 9 months from external facilitators with formal training; for N = 37: n = 18 in first year of RS, n = 19 participated in RS for at least 2 years	Length of time in RS was grouped by one year or less, or two years or more; compassion satisfaction, burnout, and secondary traumatic stress using the Professional Quality of Life Scale	Amount of time in RS did not affect ratings of compassion satisfaction, burnout, or secondary traumatic stress from pre- to post-intervention	Mixed
±Eaves et al. (2022)c Cross-sectional	Home visiting USA (national sample)	N = 139 home visitors; 97% female; race/ethnicity reported for n = 71: 85% White, 11% Hispanic/Latinx, 4% Black; 9% with less than or equal to a high school or other degree, 39% with bachelor’s degree, 50% with master’s degree; 50% with <5 years experience, 13% with 5-10 years experience, 37% with >10 years experience	Participants reported on whether their workplace had a policy providing RS, and they responded yes or no; other details regarding the RS experience were not provided	Workplace policy for providing RS using study-created yes/no item; experience of core components of a RS relationship using 7-item measure based on Tomlin et al. (2014); burnout, secondary traumatic stress, and compassion satisfaction using the Professional Quality of Life Scale	Workplace policy for providing RS was not associated with burnout, secondary traumatic stress, or compassion satisfaction; participants who reported their workplace had a policy more consistently experienced core components of a RS relationship compared to those with no workplace policy; participants who reported more consistently experiencing core components of a RS relationship reported lower burnout and secondary traumatic stress, and higher compassion satisfaction	Strong
*±Frosch et al. (2018)d Cohort (one group before and after)	Early childhood intervention Texas, USA	N = 33 (n = 17 supervisors, n = 16 interventionists); 100% female; 67% with bachelor’s degree, 33% with master’s degree or higher; 24% with ≤5 years experience, 24% with 6-11 years experience, 52% with 12+ years experience	Received RS via small group at 2- to 3-week intervals for 9 months (24 hours total); facilitated by one IECMH-Endorsed reflective supervisor and clinician	Perceived self-efficacy in RS principles using modified Reflective Supervision Self-Efficacy Scale (RSSESS) for Supervisees; work-related stress via 2 study-created items	From pre- to post-intervention, participants reported increases in RS self-efficacy and increases in work-related stress	Mixed
±Hazen et al. (2020) Cohort (one group before and after)	Mixed (child welfare and early care and education) Nebraska, USA	N = 63 (n = 10 child welfare caseworkers or mental health providers; n = 47 child welfare legal professionals; n = 6 early childhood educators); 75% female, 83% White, 5% Black, 3% Hispanic/Latinx; 75% with master’s or professional degree; mean years in field = 3.1 (SD = 1.7)	Five early childhood professionals completed Facilitating Attuned Interactions (FAN) initial training + 6 months of mentorship; these trainees then provided reflective practice small group or one-on-one sessions to study participants twice per month for 9 months (during which trainees also received monthly mentoring)	Reflective practice quality via the Reflective Supervision Rating Scale (RSRS); vicarious trauma using the Vicarious Trauma Scale; professional burnout using the Maslach Burnout Inventory – Human Services Survey	Post-intervention reflective practice quality was not associated with post-intervention reports of burnout; post-intervention reflective practice quality moderated association between pre-intervention vicarious trauma and post-intervention professional burnout, such that participants who reported higher levels of vicarious trauma and higher quality reflective practice reported lower rates of professional burnout compared to those with higher levels of vicarious trauma and lower quality reflective practice	Strong
*±Lepore (2016)b Cohort	Early care and education California, USA	N = 34 teachers; for N = 37 who completed larger study: 97% female; 73% Hispanic/Latinx, 24% White, 3% Asian; 6% with no high school degree, 18% with high school degree or GED, 47% with some college or 2-year degree, 29% with bachelor’s degree	Received RS via small group weekly (45-90 minutes) for 9 months from external facilitators with formal training; for N = 37: n=18 in first year of RS, n=19 participated in RS for at least 2 years	Number of years teachers received RS; frustration with parents using the Index of Teaching Stress – Frustration with Parents Scale; insight coded from qualitative interviews and used in quantitative analyses	Participants who had received RS for over two years decreased in their frustration with parents over the school year, while participants with one year or less increased in their frustration; participants with two or more years of RS showed greater insight when discussing teacher-parent relationships compared to participants with less than one year of RS	Mixed
*Low et al. (2019) Cohort (one group before and after)	Home visiting Rhode Island, USA	N = 26 supervisors; 100% female; 4% with some college, 50% with bachelor’s degree, 46% with master’s degree; 23% with <5 years experience, 31% with 5-10 years experience, 46% with 10+ years experience	This study was part of a larger program evaluation of the state’s MIECHV home visiting network; home visiting supervisors received training in delivering RS via a 2-day foundational training and 12 monthly 2-hour reflective practice groups facilitated by two PhD-level IECMH-Endorsed reflective supervisors and clinicians	Perceptions of one’s own RS competencies and practices using the Supervisor Rating Scale–Adapted (SRS-A), Supervisor’s Professional Development Guide and Self Assessment (SPDG), Supervision Self-Assessment (SSA)	Participants reported increases in their use of reflective practices on the SPDG and SSA (though not the SRS-A) over the course of the year (baseline, mid-year, and end of year)	Strong
*±Meuwissen & Watson (2021) Cohort (one group before and after)	Child welfare Primarily Minnesota, USA, and several other states	N = 40 workers; 97% female; 92% White, 95% Non-Hispanic; 49% with bachelor’s degree, 51% with master’s degree or higher; 49% with 1-5 years experience, 15% with 6-10 years experience, 15% with 11–15 years experience, 21% with 16–30 years experience	Received RC via virtual small groups for 6 monthly hour-long sessions; each group facilitated by one IECMH-Endorsed reflective consultant	Observed use of RS principles using Reflective Interaction Observation Scale (RIOS); perceived self-efficacy in RS principles using Reflective Supervision Self-Efficacy Scale (RSSESS); perceptions of the reflective consultant using Reflective Supervision Rating Scale (RSRS); positive and negative work experiences using the Professional Quality of Life Scale	From pre- to post-intervention, increases in observed Self Openness and Parallel Process (but not Taking Others’ Perspectives) using the RIOS; participants in groups with increases in Self Openness showed increases in self-efficacy, positive perceptions of the consultant, burnout/secondary traumatic stress, and decreases in compassion satisfaction	Mixed
±Morelen et al. (2022) Cross-sectional	Mixed (home visiting, early care and education, early intervention, child welfare, others) Tennessee, USA	N = 123 (24% in home visiting and at least one other setting, 23% in home visiting only, 18% in early care and education only, 8% in early care and education and at least one other setting, 10% in early intervention, 6% in health, 3% in child welfare, 2% in child advocacy, 1% in higher education, 1% in speech and language pathology, and 4% in other profession); 98% female, 2% undisclosed gender; 81% White, 12% Black, 3% Hispanic/Latinx, 2% Multi-ethnic, 1% Asian or Pacific Islander, 2% undisclosed; mean years in field = 13.6	Participants were asked whether or not they received RS; could choose “no”, “yes, for less than 1 year”, or “yes, for more than 1 year”; 32% reported no, 14% reported <1 year, 54% reported >1 year	Internalizing symptoms using the Center for Epidemiologic Studies Depression Scale (CES-D) and General Anxiety Disorder-7 (GAD-7); COVID-19 stress using the Pandemic Stress Index; self-care using the study-created Self-Care Belief and Behavior Questionnaire; burnout using the Burnout Scale of the Professional Quality of Life Scale	Participants who received no RS reported higher levels of internalizing symptoms compared to participants who had received >1 year of RS; no mean differences based on receiving RS on COVID-19 stress, self-care, or professional burnout; RS moderated the pathway between COVID stress and self-care, such that higher levels of COVID stress were associated with lower levels of self-care when participants had no RS or <1 year, but not when participants had >1 year of RS	Strong
*Shea et al. (2016) Cohort (one group before and after)	Community mental health Michigan, USA	N = 29 (n = 13 supervisors; n = 16 supervisees) Of supervisors: 92% female; 46% White, 39% Black; 39% with 1-5 years practice experience, 39% with 6-10 years practice experience; 62% with 1-5 years supervisory experience, 23% with 16-20 years supervisory experience Of supervisees: 100% female; 50% White, 38% Black; 13% with 0 years experience, 31% with <1 year experience, 43% with 1-5 years experience	Supervisors and their supervisees participated in a training series on RS, which included eight 3-hour sessions (24 hours total) for supervisors, with supervisees attending the last four sessions (12 hours); the training series was facilitated by two trainers with expertise in RS and IECMH	Reflective practice skills using the Reflective Supervision Rating Scale (for Supervisees and Supervisors); perceived self-efficacy in RS principles using the Reflective Supervision Self-Efficacy Scale (for Supervisors and Supervisees)	From pre- to post-intervention, supervisees demonstrated increases in 5 of 16 reflective skills per supervisor report; 8 to 10 months following the training series, supervisees maintained increases in 4 of these skills and showed increases in 2 additional skills; there was a decrease in 1 of the skills from pre- to post; supervisees reported increases in self-efficacy on 1 of 17 items; at the 8-to-10-month follow-up, there was a decrease in self-efficacy for one different item From pre- to post-intervention, supervisors demonstrated decreases in 2 of 17 reflective skills per supervisee report, though these were not maintained at 8-to-10-month follow-up, there was a decrease in one different skill; supervisors reported increases in self-efficacy on 16 of 17 items, and maintained these increases on 10 items at 8-to-10 month follow-up	Mixed
*±Shea et al. (2020) Cohort (one group before and after)	Home visiting Michigan, USA	N = 56 home visitors; 100% with master’s degree; mean months in field = 39.5 (SD = 43.1)	This study was part of a larger program evaluation of the state’s home visiting services; participants reported receiving RS (either individual, group, or both) on average 1.6 times per week (SD = 0.8)	Perceived self-efficacy in RS principles using Reflective Supervision Self-Efficacy Scale (RSSESS); therapist characteristics and burnout, job meaning, job satisfaction, and coping with work challenges using the Clinician Profile Form	At the 3-month assessment, reflective practice self-efficacy was positively associated with job satisfaction and negatively associated with job burnout, but not associated with finding meaning and having coping strategies; from 3- to 12-month assessments, participants reported increases in self-efficacy subscales “Use of Reflective Practice Skills” and “Use of Observational Skills,” but not “Use of Supervisory Relationship” or “Use of Self-Awareness”	Strong
*±Shea et al. (2022)e Cohort (one group before and after)	Early care and education USA (national sample)	N = 38 (n = 6 supervisors and program managers; n = 13 grant specialists; n = 19 IECMH consultants); 89% female; 95% White, Non-Hispanic; 5% Hispanic/Latinx; 32% with bachelor’s degree, 68% with master’s degree; 3% with <1 year experience, 18% with 6-10 years experience, 32% with 11-15 years experience, 8% with 16-20 years experience, 39% with >20 years experience	Participated in virtual 2-hour monthly reflective learning groups for 12 months; groups were facilitated by 6 external IECMH-endorsed consultants contracted from the Alliance	Reflective practice self-efficacy using the Reflective Practice Self-Efficacy (RPSE) Scale for IECMH Consultants, the RPSE Scale for Infant-Early Childhood Program Specialists, and the RPSE Scale for Infant-Early Childhood Program Managers and Supervisors; reflective practice skills using the Use of Self and Reflective Practice Skills measure; mindfulness using the Mindful Attention Awareness Scale (MAAS); burnout using the Maslach Burnout Inventory-Human Services Survey	Reflective practice self-efficacy increased from pre-test to 6-month assessment, and from pre-test to 12-month assessment; reflective practice skills increased between 6- and 12-month assessments, and from pre-test to 12-month assessment; there were no changes in mindfulness or burnout (pre-test levels indicated high mindfulness and low burnout)	Strong
*Virmani & Ontai (2010) Cohort (two groups before and after)	Early care and education California, USA	N = 20 (n = 10 teachers at university-based site using RS; n = 10 teachers at university-based site using traditional methods; 100% female; all were college students; all had worked no more than 1 week in their respective classrooms prior to enrolling in the study	The “reflective site” included a 3-hour orientation, 15-minute daily discussion meetings, 1.5 hour weekly discussion meetings led by head teachers and supervisors, and a course seminar on early childhood; the “traditional site” focused on administrative tasks and general child development	Teacher insightfulness via the Insightfulness Assessment, which involved the teacher viewing videotaped interactions (three 5-minute segments) between themselves and a target child and discussing their thoughts and feelings about the interaction; two independent coders unaware of site rated eight components of insightfulness on a 9-point scale (scores were collapsed to represent low, medium, and high category ratings for each component)	At baseline, only 1 of the 7 insightfulness assessment components was associated with site but 2.5 months later all 7 assessment components were associated with site such that insightfulness was higher at the RS site than the traditional supervision site	Strong
*±Watson et al. (2016a)f Cohort (one group before and after)	Home visiting Minnesota, USA	N = 150 (n = 30 supervisors; n = 120 home visitors); N changed across the evaluation period due to attrition and hiring practices (as few as 26 supervisors and 82 home visitors completed measures at time 3. Of supervisors (n = 26): 96% White, 4% other; 58% with bachelor’s degree, 42% with postgraduate degree Of home visitors (n = 66): 100% White, 3% Hispanic/Latinx; 9% with some college or associate’s degree, 70% with bachelor’s degree, 21% with postgraduate degree	This study was part of a broader evaluation of one state’s tiered model to build reflective practice capacity funded by the MIECHV expansion project Two mentors (IECMH-Endorsed reflective supervisors and clinicians) provided initial training to sites, and provided monthly RC to IECMH consultants; IECMH consultants provided monthly individual RC to program supervisors; program supervisors provided weekly individual RS to home visitors; additional supports were also in place (e.g., monthly case conferences)	Reflective functioning using the Kentucky Inventory of Mindfulness Skills (KIMS); burnout (emotional exhaustion, depersonalization, personal accomplishment) using the Maslach Burnout Inventory (MBI)	Over the course of the grant period (assessed at beginning, middle, and end), there were no changes in reflective functioning for supervisors or home visitors; there were no changes in supervisors’ Emotional Exhaustion, Depersonalization, or Personal Accomplishment scores; there were no changes in home visitors’ Depersonalization or Personal Accomplishment scores, but there was an increase in Emotional Exhaustion scores	Strong

Note. IECMH = infant early childhood mental health; for consistency, we use the abbreviation IECMH to refer to work with infants only, young children only, or both, given that some states use the term “infant mental health” while others incorporate “early childhood”. RC = Reflective Consultation. RS = Reflective Supervision. Superscripts indicate articles published from the same sample, or articles that include both quantitative and qualitative (or implementation-focused) results, and thus are presented in multiple tables. The * indicates papers that included reflective capacity as an outcome. The ± indicates papers that included well-being as an outcome.

Table 2.

Summary of Included Articles with Qualitative Analyses

STUDY ID	Setting	Participant Characteristics	RS Experience	Primary Constructs & Measures	Primary Findings	Study Quality
*±§Barron et al. (2022a)g Part I	Mixed (home visiting and early care and education) Michigan, USA	N = 50 (n = 43 home visitors; n = 2 home visiting administrators; n = 4 early care and education teachers, n = 1 consultant); for n = 25 who completed individual interviews: 65% White, 8% Black, 4% Hispanic/Latinx; 2% not reported; for full sample: 2% para-professional, 4% with associate degree, 12% with bachelor’s degree, 80% with graduate degree	Received group only (20%), individual only (8%), or both group and individual (70%) RS; for those receiving group RS, 2% met weekly, 48% met bimonthly, 46% met monthly; for those receiving individual RS, 46% met weekly, 12% met bimonthly, 20% met monthly	Qualitative reports of participants’ professional satisfaction, practice behavior, and perspectives on components of RS they find most important, using semi-structured focus groups (n = 25) and individual interviews (n = 25); grounded theory analysis was used	Participants identified essential components of RS: feelings and safety, feelings of trust, consistency and predictability, nonjudgmental responses, and a commitment to being emotionally present to the experience; two categories emerged as important outcomes from RS: professional wellness (e.g., reduced burnout and vicarious trauma, greater engagement with work, increased motivation for professional development, increased professional efficacy) and personal growth (e.g., increased empowerment, greater emotion regulation, increased reflective capacity); several practice behavior outcomes were identified: bringing up difficult things with families, becoming a better observer, developing relationships with families, sharing perspectives and ideas, re-energizing to keep moving forward	Strong
* ± Begic et al. (2019) a	Home Visiting Idaho, USA	N = 27 home visitors; for n = 20 who completed survey: 7% did not have a bachelor’s degree, 56% had a bachelor’s degree, 11% had a master’s degree	Details regarding the RS experience were not provided	Qualitative reports of participants’ perceptions of risk factors for secondary traumatic stress and burnout, and how RS mitigates negative effects using a semi-structured interview; the framework method of analysis was used	Participants reported that lack of adequate and safe RS affected their ability to process secondary traumatic stress; high quality RS was identified as a protective factor, allowing participants to be more reflective, gain different perspectives, reduce feelings of isolation, and help process their own emotions	Mixed
* ± Brown (2016) b	Early care and education California, USA	N = 20 teachers; for N = 37 who completed larger study: 97% female; 73% Hispanic/Latinx, 24% White, 3% Asian; 6% with no high school degree, 18% with high school degree or GED, 47% with some college or 2-year degree, 29% with bachelor’s degree	Received RS via small group weekly (45-90 minutes) for 9 months from external facilitators with formal training; for N = 37, n = 18 in first year of RS, n = 19 participated in RS for at least 2 years	Qualitative reports of whether time spent in RS was associated with compassion satisfaction, burnout, and secondary traumatic stress using semi-structured interviews; inductive thematic analysis was used	Participants reported that RS improved compassion satisfaction and buffered against some work-related stressors (though not administrative), and increased feelings of self-efficacy	Strong
± Eaves et al. (2021)	Mixed (community mental health, early intervention, others) Colorado & Connecticut, USA	N = 21 (about 69% were clinical social workers or psychologists, 31% were nurses or early intervention specialists); 100% female; predominantly White (<10% Black, Asian, Hispanic/Latinx); 77% with master’s degree or higher	Currently receiving RS in the workplace; additional details regarding RS were not provided	Qualitative reports of participants’ perceptions of whether RS is an effective professional self-care mechanism, using semi-structured interviews; inductive analysis using grounded theory approach was used	86% of participants identified RS as an important component of their self-care regimen, but thought that RS was not enough on its own; 14% of participants discussed supervisory relationships that were not truly reflective, and missing elements (e.g., lack of trust and safety) prevented these experiences from being useful	Strong
*±Frosh et al. (2019)d	Early childhood intervention Texas, USA	N = 31 (n = 16 supervisors; n = 15 interventionists); 100% female; 68% White, 13% Hispanic/Latinx; 16% Black; 71% with bachelor’s degree, 33% with master’s degree or higher; 23% with ≤5 years experience, 23% with 6-11 years experience, 55% with 12+ years experience	Received RS via small group at 2- to 3-week intervals for 9 months (24 hours total); facilitated by one IECMH-Endorsed reflective supervisor and clinician	Qualitative reports of stress and coping using a 3-item project-created questionnaire with open-ended questions about ways participants cope with work-related stress, aspects of their work that are most stressful, and what can be done to reduce work-related stress; inductive analysis techniques to identify themes in responses were used	Participants identified individual, relational, and organizational sources of stress and coping strategies; relational stress related to working with staff was a larger issue for supervisors; from pre- to post-assessment, participants’ responses were both longer and deeper, suggesting greater exploration of thoughts and feelings; participants’ expanded their methods for coping with stress	Strong
* Harrison (2016)	Early care and education Minnesota, USA	N = 15 (n = teachers; n=7 developmental therapists); 94% female; 7% with 4-6 years experience, 7% with 7-10 years experience, 87% with 10+ years experience; 13% with 1-2 years in RC, 27% with 2-4 years in RC, 60% with 5+ years in RC	RC offered via monthly small groups for 90 minutes and optional 30-minute individual/small group sessions; facilitated by outside licensed mental health clinician with extensive IECMH experience	Qualitative reports of how participants experience the RC program using a semi-structured interview; general inductive approach was used	Four overarching themes emerged, resulting in the creation of a change process model defined as release (overwhelming and helpless/hopeless emotions in a safe, calming, validating group), reframe (by asking questions, considering others’ perspectives, and challenging own biases), refocus (on realistic possibilities with a belief in one’s own self-efficacy), and respond (by slowing down, observing, listening, and flexibly meeting the needs of children and parents)	Strong
§ Lepore (2016) b	Early care and education California, USA	N = 20 teachers; for N = 37 who completed larger study: 97% female; 73% Hispanic/Latinx, 24% White, 3% Asian; 6% with no high school degree, 18% with high school degree or GED, 47% with some college or 2-year degree, 29% with bachelor’s degree	Received RS via small group weekly (45-90 minutes) for 9 months from external facilitators with formal training; for N = 37, n=18 in first year of RS, n=19 participated in RS for at least 2 years	Qualitative reports of how RS relates to interactions with parents, and if there are differences between groups in their first year and groups with two or more years of RS, using semi-structured interviews; inductive thematic analysis was used	For participants in both groups, themes emerged related to gains made regarding teacher-parent relationships, including structured time to discuss families, improved ability to communicate with families, greater referrals to collateral services for families, and greater coping with challenges related to interactions with families	Strong
* ± Russ et al. (2020) h	Child welfare Queensland, Australia	N = 24 child protection workers; 75% female; 8% had < 2 years experience, 92% had ≥ 3 years experience	Details regarding the RS experience were not provided	Qualitative reports of participants’ experiences of adversity, resilience, and what influenced and maintained resilience using a semi-structured interview; used an initial deductive process using a priori codes, then used inductive coding drawing themes directly from the data, consistent with constructivist and exploratory approaches	Reflective practice promoted resilience (over and above personal qualities and supportive relationships), contributed to self-awareness, and enabled learning and practice improvement	Strong
* ± Shea et al. (2022) e	Early care and education USA (national sample)	N = 38 consultees (n = 6 supervisors and program managers; n = 13 grant specialists; n = 19 IECMH consultants); 89% female; 95% White; 5% Hispanic/Latinx; 32% with bachelor’s degree, 68% with master’s degree; 3% with <1 year experience, 18% with 6-10 years experience, 32% with 11-15 years experience, 8% with 16-20 years experience, 39% with >20 years experience	Participated in 2-hour virtual monthly reflective learning groups for 12 months; groups were facilitated by external IECMH-endorsed mentors contracted from the Alliance	Consultants’ qualitative reports of consultees’ reflective practice skills, and consultees’ qualitative reports of their approaches to work using open-ended surveys at pre- and post-test; an inductive two-cycle coding process was used	From pre-test to post-test, consultees and consultants reported an increase in consultees’ reflective practice skills and decrease in their use of directive approaches; consultees reported greater use of personal and professional resources; consultees discussed burnout as a central element of their work experience at pre-test, but not at post-test	Mixed
* Summers et al. (2007)	Mixed (early care and education and early intervention) Location unclear	N = 37 (n = 16 home visitors and family specialists; n = 10 supervisors and administrators; n = 8 parents; n = 3 mentors); 43 in-depth interviews were conducted	Individual, regular RC sessions between mentor and home visitor with videotape review; as-needed meetings between mentors and home visitors; group meetings between mentors and supervisors; in one program, center staff met regularly to review videotapes together in an interdisciplinary, case-study model, or reflecting team approach	Qualitative examination of what constitutes successful IECMH mentoring in general, and RC defined with videotaping in particular using an in-depth interview; used a constant comparative coding and analysis method; key informants were interviewed up to three times	Several themes emerged as processes that take place during ongoing RC and video review: increasing recognition (“you don’t know what you don’t know”), increasing observational abilities (“they don’t know what they see”), clarifying intentions (“the difference is between the wanting and the doing”), building identity of oneself as a professional (“when this keeps not working, there’s a reason”)	Mixed
* ± Susman-Stillman et al. (2020)	Mixed (community mental health, home visiting, early intervention, early care and education, child welfare, medical care, others) USA (national sample)	N = 97 supervisors (n = 56 in mental health; n = 15 in home visiting; n = 10 in early intervention or physical/occupational/speech therapy; n = 16 in early care and education, child welfare, medical care, or other); 24% with bachelor’s degree, 67% with master’s degree; 64% with 0-10 years providing RS, 23% with 11-20 years, 13% with >20 years	All participants provided RS to others; additional details regarding RS were not provided	Qualitative reports of perceptions of providing and receiving RS using a 4-item project-created survey with open-ended questions about why participants became a reflective supervisor, what has been most helpful in learning/maintaining RS skills, whether they experienced effects of their participation in RS, and whether they have seen effects of RS for others who have received it; general inductive approach was used	Four themes of how reflective supervisors perceived the impact of RS in relation to well-being emerged: cultivated skills of emotional awareness, empathy and compassion, and self-efficacy; increased reflective capacity and perspective-taking skills; built, experienced, and used supportive relationships; strengthened ability to cope with stress, maintain job satisfaction, and decrease burnout	Mixed
* Tomlin et al. (2016)	Home visiting Indiana, USA	N = 9 (n = 3 speech-language pathologists; n = 2 occupational therapists, n = 2 developmental therapists, n = 1 physical therapist, n = 1 audiologist); 100% female; 100% White; 11% with bachelor’s degree, 89% with master’s degree; 11% with 3-5 years experience, 89% with >8 years experience	Details regarding the RS experience were not provided	Qualitative reports of participants’ reflective skills using a semi-structured interview which required participants to read a vignette and respond to a series of questions; conducted content analysis of the data, with themes developed by one coder and a co-investigator	Participants largely responded to the vignette by describing action-oriented rather than reflective responses, such as directly instructing parents in child development and appropriate parenting, with little to no attention on asking questions to gather more information, wondering about internal experiences of the parent or baby, or linking behaviors or experiences across relationships; participants also largely omitted discussion of their own responses and did not acknowledge a need for consultation/supervision	Mixed
* Veloni (2017)	Mixed (home visiting, doctoral program with focus in IECMH) California, USA	N = 8 (n = 4 home visiting supervisors; n = 4 doctoral students); 100% female	Virtual reflective practice groups; additional details regarding the RS experience were not provided	Qualitative reports of participants’ experiences in virtual RS groups using semi-structured interviews; iterative coding process	Increased understanding of a reflective way of being and increased use of (and confidence in) reflective skills; felt supported, relieved, refreshed, and refueled	Strong
* Watson & Gatti (2012) i	Early care and education Minnesota, USA	N = 14 teachers and related service providers (n = 5 completed interviews); 100% female	Monthly RC groups that met for 2 hours per session for a total of eight meetings across the school year; groups were led by an IECMH specialist with a PhD in clinical psychology	Qualitative reports of specific strategies learned and implemented at RC groups and whether and how the groups helped participants feel supported and effective, as well as whether and how beliefs, thoughts, and practices were influenced by RC using a small group interview format; interviews were analyzed for themes	RC was helpful to reframe perspective, try new things with families, helped to understand relationship between parent and child, helped to understand own emotions, safe place to share, provided additional resources, acknowledged that it is difficult work, learned when to let go of things out of your control	Weak
* ± Watson et al. (2014) i	Early care and education Minnesota, USA	N = 31 teachers and related service providers; n = 15 participated in RS for ≥1 year; n = 16 did not participate in RS; 97% of RS group had some graduate coursework or master’s degree, 94% of non-RS group had some graduate coursework or master’s degree; for all participants, M = 10 years experience	Monthly RC groups that met for 90 minutes, with option to sign up for 30-minute individual or small group sessions; groups were led by same IECMH specialist described in Watson & Gatti (2012)	Qualitative reports of the ways in which RC addresses work stress and the impacts of RC on participants and on the choice of interventions they implement in their work using interviews; interviews were analyzed for themes	Participants reported that RC reduced stress in the moment by providing emotion regulation and release, validation, and affirmation, and over time by clarifying their professional role and increasing perspective-taking; participants noted impacts on their feeling (e.g., less overwhelm, isolation, anxiety, greater empowerment and confidence), thinking (e.g., more reflective, patient, and less judgmental; more perspective taking), and greater trust in their own reactions as data; they reported more use of questions, more thoughtfulness in the use of time and resources, greater presence and flexibility, better listening, and reduction in trying to fix all of a family’s problems	Weak
* ± Watson et al. (2016a) f	Home visiting Minnesota, USA	N = 150 (n = 30 supervisors; n = 120 home visitors); N changed across the evaluation period due to attrition and hiring practices (as few as 28 supervisors and 60 home visitors completed interviews at the end of the grant period. Of supervisors (n = 26): 96% White, 4% other; 58% with bachelor’s degree, 42% with postgraduate degree Of home visitors (n = 66): 100% White, 3% Hispanic/Latinx; 9% with some college or associate’s degree, 70% with bachelor’s degree, 21% with postgraduate degree	This study was part of a broader evaluation of one state’s tiered model to build reflective practice capacity funded by the MIECHV expansion project Two mentors (IECMH-Endorsed reflective supervisors and clinicians) provided initial training to sites, and provided monthly RC to IECMH consultants; IECMH consultants provided monthly individual RC to program supervisors; program supervisors provided weekly individual RS to home visitors; additional supports were also in place (e.g., monthly case conferences)	Qualitative reports of what knowledge supervisors and home visitors felt that they had gained over the evaluation period using interviews at pre- and post-evaluation period; interview data were first analyzed for themes, and then analyzed using bricolage techniques	Five themes emerged: greater ability to pause and reflect (i.e., slowing down and listening); increase in wondering, not fixing; understanding the concept of parallel process; increased focus on the baby; greater ability to “go deeper” (i.e., looking for causes and meaning of behaviors, understanding one’s own responses); regarding burnout, supervisors saw their role as relieving stress for home visitors by providing a safe place for them to express emotion, which led to greater exploration of different perspectives; home visitors also identified feelings of personal accomplishment mitigated burnout	Mixed
* § Williams et al. (2019) j	Community mental health California, USA	N = 34 supervisors (n = 18 marriage and family therapists, n = 9 clinical social workers, n = 7 psychologists); 97% female; mean years in early childhood mental health = 8.3	2-day large group didactic training (11 hours); eight 2-hour small group meetings over 4 months (16 hours); one optional individual session with facilitator to review video of participant’s supervision session; training facilitated by four licensed psychologists (one per small group), who were IECMH-Endorsed and received current RC	Focus groups to ascertain the impact of the training series on participants’ roles as supervisors; focus groups were analyzed for themes	Three core elements of RS were identified: being present for others, the quality of relationships formed with supervisees, and a collaborative approach to supervision; several themes were identified regarding the impact of the training on participants: supervisors became more reflective, less directive, applied a reflective approach across situations, experienced positive personal changes; additional themes were identified regarding the impact of the training on supervisees: supervisors described their supervisees as becoming more expressive and confident, increasingly driving their own answers, and potentially improving the quality of clinical care	Strong

Note. IECMH = infant early childhood mental health; for consistency, we use the abbreviation IECMH to refer to work with infants only, young children only, or both, given that some states use the term “infant mental health” while others incorporate “early childhood”. RC = Reflective Consultation. RS = Reflective Supervision. Superscripts indicate articles published from the same sample, or articles that include both quantitative and qualitative (or implementation-focused) results, and thus are presented in multiple tables. The * indicates papers that included reflective capacity as an outcome. The ± indicates papers that included well-being as an outcome. The § indicates papers that included child and family outcomes.

Table 3.

Summary of Included Articles with Implementation-Focused Analyses

STUDY ID	Setting	Participant Characteristics	RS Experience	Primary Constructs & Measures	Primary Findings	Study Quality
Barron et al. (2022b)g Part II Qualitative	Mixed (home visiting and early care and education) Michigan, USA	N = 50 (n = 43 home visitors; n = 2 home visiting administrators; n = 4 early care and education teachers, n = 1 consultant); for n = 25 who completed individual interviews: 65% White, 8% Black, 4% Hispanic/Latinx; 2% not reported; for full sample: 2% para-professional, 4% with associate degree, 12% with bachelor’s degree, 80% with graduate degree	Received group only (20%), individual only (8%), or both group and individual (70%) RS; for those receiving group RS, 2% met weekly, 48% met bimonthly, 46% met monthly; for those receiving individual RS, 46% met weekly, 12% met bimonthly, 20% met monthly	Qualitative reports of participants’ views of supervisee, supervisor, relational, and contextual characteristics that influence their experience of RS, using semi-structured focus groups (n = 25) and individual interviews (n=25); grounded theory analysis was used	Three domains of influence emerged – individual, relational, and contextual. Individual factors included expectations, past experiences, understanding of RS, perception of administrative/reflective balance, perception of role, intrinsic qualities, supervisor’s level of experience, supervisors’ support of the supervisee’s professional development, supervisors’ ability to ask questions, supervisors’ reflective functioning. Relational factors included sharing vulnerability, availability of both parties, and relationship disruptions. Contextual factors included agency support of RS, format of RS (group vs. individual), issues of diversity, and resource limitations (i.e., cost, job demands, time)	Strong
Eaves et al. (2022)c Cross-sectional	Home Visiting USA (national sample)	N = 139 home visitors; 97% female; 85% White, 11% Hispanic/Latinx; 4% Black; 9% with less than or equal to a high school or other degree, 39% with bachelor’s degree, 50% with master’s degree; 50% with <5 years experience, 13% with 5-10 years experience, 37% with >10 years experience	Participants reported on whether their workplace had a policy providing RS, and they responded yes or no	Workplace policy for providing RS using study-created yes/no item; experience of core components of a RS relationship using 7-item measure based on Tomlin et al. (2014); burnout, secondary traumatic stress, and compassion satisfaction using the Professional Quality of Life Scale	Workplace policy for providing RS was not associated with burnout, secondary traumatic stress, or compassion satisfaction; participants who reported their workplace had a policy more consistently experienced core components of a RS relationship compared to those with no workplace policy; participants who reported more consistently experiencing core components of a RS relationship reported lower burnout and secondary traumatic stress, and higher compassion satisfaction	Strong
Julien-Chinn & Lietz (2019) Cross-sectional	Child welfare Arizona, USA	N = 89 workers; 85% female; 60% White, 9% Black, 1% Asian, 6% Multiracial, 1% Other, 23% Hispanic/Latinx; mean years in child welfare = 5.0	This study was part of a larger project evaluating training in Strengths-Based Supervision	RS quality using 8-item project-created measure; learning culture within one’s agency using the Comprehensive Organizational Health Assessment (COHA); hours spent in group supervision per week	Higher RS quality and more time spent in group supervision were each associated with higher perceived learning culture	Strong
Russ et al. (2020)h Qualitative	Child welfare Queensland, Australia	N = 24 child protection workers; 75% female; 8% had < 2 years experience, 92% had ≥ 3 years experience	Details regarding the RS experience were not provided	Qualitative reports of participants’ experiences of adversity, resilience, and what influenced and maintained resilience using a semi-structured interview; used an initial deductive process using a priori codes, then used inductive coding drawing themes directly from the data, consistent with constructivist and exploratory approaches	Supportive supervisory and peer relationships that acknowledged stressors and encouraged reflection promoted resilience and reflective practice; workplace culture that valued reflective practice created a sense of safety in emotional processing and learning	Strong
Watson et al. (2016a)e Qualitative	Home visiting Minnesota, USA	N = 168 (n = 30 supervisors; n = 120 home visitors; n = 18 grant administrators); N changed across the evaluation period due to attrition and hiring practices (as few as 28 supervisors and 60 home visitors completed interviews at the end of the grant period. Of supervisors (n = 26): 96% White, 4% other; 58% with bachelor’s degree, 42% with postgraduate degree Of home visitors (n = 66): 100% White, 3% Hispanic/Latinx; 9% with some college or associate’s degree, 70% with bachelor’s degree, 21% with postgraduate degree Grant administrators’ demographics not reported	This study was part of a broader evaluation of one state’s tiered model to build reflective practice capacity funded by the MIECHV expansion project Two mentors (IECMH-Endorsed reflective supervisors and clinicians) provided initial training to sites, and provided monthly RC to IECMH consultants; IECMH consultants provided monthly individual RC to program supervisors; program supervisors provided weekly individual RS to home visitors; additional supports were also in place (e.g., monthly case conferences)	Qualitative reports of whether state supports were sufficient for implementing reflective practice; interview data were first analyzed for themes, and then analyzed using bricolage techniques	Supervisors identified the need for additional resources, including an ongoing RS group, greater administrative support, access to relevant topics and literature, and having an IECMH specialist/reflective supervisor embedded in program; grant administrators stated that with the MIECHV grant there were sufficient financial resources for RS, but without that funding it would be difficult to support RS	Mixed
Williams et al. (2019)j Qualitative	Community mental health California, USA	N = 34 supervisors (n = 18 marriage and family therapists, n = 9 clinical social workers, n = 7 psychologists); 97% female; mean years in early childhood mental health = 8.3	2-day large group didactic training (11 hours); eight 2-hour small group meetings over 4 months (16 hours); one optional individual session with facilitator to review video of participant’s supervision session; training facilitated by four licensed psychologists (one per small group), who were IECMH-Endorsed and received current RC	Focus groups to ascertain perceived key ingredients of a reflective culture and descriptions of the reflective culture at participants’ agencies, including supports and barriers to RS; focus groups were analyzed for themes	Agency supports for RS included support from agency leaders and training opportunities; agency barriers included job demands, lack of buy-in from agency leaders, and isolation	Strong

Note. IECMH = infant early childhood mental health; for consistency, we use the abbreviation IECMH to refer to work with infants only, young children only, or both, given that some states use the term “infant mental health” while others incorporate “early childhood”. RC = Reflective Consultation. RS = Reflective Supervision. Superscripts indicate articles published from the same sample, or articles that include both quantitative and qualitative (or implementation-focused) results, and thus are presented in multiple tables.

Several early childhood-serving settings are represented in the 28 articles, with 30% of articles (n = 8) including multiple settings, 26% of articles (n = 7) focused on early care and education, 22% (n = 6) focused on home visiting, 11% focused on child welfare (n = 3), 7% (n = 2) focused on early intervention, and 7% (n = 2) focused on community mental health. All but one study, which was completed in Australia, took place in the United States. With the exception of three studies that recruited nationwide samples, 11 different U.S. states were represented.

Results underlined heterogeneity in both the format of RS/C and whether or how it was described. Although many participants received RS/C in small groups, the frequency varied from weekly (e.g., Brown, 2016; Virmani & Ontai, 2010) to monthly (e.g., Harrison, 2016; Watson et al., 2014), and some were delivered virtually (e.g., Meuwissen & Watson, 2021; Shea et al., 2022; Veloni, 2017). Other participants received individual RS/C (e.g., Watson et al., 2016a; Summers et al., 2007), and still others received a combination of individual and small group RS/C (e.g., Barron et al., 2022a; Hazen et al., 2020; Shea et al., 2020). Some participants had access to additional RS/C supports as needed (e.g., Harrison, 2016; Williams et al., 2019). Participants in four studies attended large group foundational or didactic trainings to learn about RS/C in more depth, and/or how to provide RS/C to their staff (i.e., Low et al., 2018; Shea et al., 2016; Watson et al., 2016a; Williams et al., 2019).

Study Quality Assessment

Of the 14 studies that presented quantitative results, eight (57%) had an overall strong quality, six (43%) had an overall mixed quality, and zero had an overall weak quality (see Table 1). Of the 17 studies that presented qualitative results, nine (53%) had an overall strong quality, six (35%) had an overall mixed quality, and two (12%) had an overall weak quality (see Table 2). Of the 6 studies that presented implementation-focused results, five (83%) had an overall strong quality, one (17%) had an overall mixed quality, and zero had an overall weak quality (see Table 3). Although most studies in this review were rated as strong using criteria for cross-sectional, cohort, or qualitative studies, only one study used a control group (Virmani & Ontai, 2010), and no studies were RCTs, thus effects cannot be solely attributed to RS/C.

Research Question 1

Reflective Capacity in Early Childhood-Serving Professionals

Quantitative Studies.

Nine of the 14 studies using quantitative measures investigated how RS/C was associated with reflective capacity, skill, or self-efficacy (see papers marked with * in Table 1). Studies used a variety of measures for this outcome; four utilized the Reflective Supervision Self-Efficacy Scale, two used the Reflective Supervision Rating Scale, and five used other, distinct measures, with some studies using more than one measure. Results showed positive associations between RS/C and these constructs. From pre- to post-intervention, self-efficacy in reflective practice increased among supervisors and providers in early intervention (Frosch et al., 2018), supervisors, program managers, grant specialists, and consultants in early care and education (Shea et al., 2022), and providers in child welfare (Meuwissen & Watson, 2021). Likewise, home visiting supervisors reported increases in their use of reflective practice and supervision skills on two out of three measures after participating in a foundational training and enhanced skill development groups over the course of a year (Low et al., 2019). In the only study that used a control group design, insightfulness was higher among teachers who received reflective supervision compared to those who received traditional supervision (Virmani & Ontai, 2010). Lepore (2016) also examined teachers’ insightfulness, finding that teachers who had received two or more years of reflective supervision showed greater insight when discussing their relationships with parents compared to teachers with less than one year of reflective supervision. The two studies that examined changes in individual skills (rather than change in a composite score) demonstrated some increases in skills from pre- to post-intervention, but given the number of skills examined and incorporation of both supervisor and supervisee perspectives, findings were mixed (Shea et al., 2016; Shea et al., 2020). Although this body of research, when considered as a whole, suggests positive associations between RS/C and reflective capacity, one study stands as the exception: Watson and colleagues (2016a) did not find changes in reflective functioning among home visiting supervisors or providers following receipt of RS/C.

Qualitative Studies.

Fifteen of the 17 studies using qualitative methods asked participants how RS/C affected their reflective capacity, skill, or self-efficacy (see papers marked with * in Table 2). From participants’ perspectives, RS/C was robustly associated with increases in reflective capacity and improvements in individual skills for both supervisors providers, including slowing down, observing, and listening; wondering instead of fixing; asking more questions; considering multiple perspectives; focusing more on the infant/child; challenging one’s own biases; being less judgmental; being less directive in one’s approach; being flexible; and understanding the concept of parallel process (Barron et al., 2022a; Begic et al., 2019; Harrison 2018; Shea et al., 2022; Summers et al., 2007; Susman-Stillman et al., 2020; Watson & Gatti, 2012; Watson et al., 2014; Watson et al., 2016a; Williams et al., 2019; Veloni, 2017). The ability to explore and understand one’s own emotional responses to the work, regulate emotions, empathize with others, build and rely upon supportive relationships, and reduce isolation also emerged as important themes following exposure to RS/C (Barron et al., 2022a; Begic et al., 2019; Frosch et al., 2019; Harrison, 2018; Susman-Stillman et al., 2020; Watson & Gatti, 2012; Watson et al., 2014; Watson et al., 2016a). Reflective supervision also increased feelings of empowerment and self-efficacy among providers (Barron et al., 2022a; Brown, 2016; Susman-Stillman et al., 2020). In summary, people were able to clearly identify how their reflective capacity, skill, or self-efficacy had changed for the better after receiving RS/C. One study, in contrast, assessed participants’ reflective skills by requiring participants to read a vignette and respond to a series of questions; participants largely responded to the vignette by describing action-oriented rather than reflective responses, though it was unclear the extent to which participants had participated in reflective supervision (Tomlin et al., 2016).

Well-Being in Early Childhood-Serving Professionals

Quantitative Studies.

Eleven of the 14 studies using quantitative measures explored associations between RS/C and well-being of supervisors or providers (see papers marked with ± in Table 1). Several constructs associated with well-being, or lack of well-being, were examined, and while measures varied, five studies used the Professional Quality of Life Scale, three studies used the Maslach Burnout Inventory, and seven studies used other, distinct measures, with some studies using more than one measure. Although results from these studies are heterogenous, there is some evidence to suggest receiving RS/C may be positively associated with well-being and negatively associated with secondary traumatic stress, burnout, and related constructs. For instance, two cross-sectional studies found that experiencing a high-quality reflective supervision relationship was associated with lower levels of secondary traumatic stress and burnout (Begic et al., 2019; Eaves et al., 2022), lower levels of intention to quit (Begic et al., 2019), and higher levels of compassion satisfaction (Eaves et al., 2022) among home visitors. In a cross-sectional study of providers spanning many early childhood settings, those who had received no reflective supervision reported higher levels of internalizing symptoms compared to providers who had received at least one year of reflective supervision (Morelen et al., 2022). Relatedly, in a study of home visitors, Shea and colleagues (2020) found that after receiving reflective supervision, perceived self-efficacy in reflective practice was positively associated with job satisfaction and negatively associated with burnout.

Two studies provided additional context for associations between reflective supervision and well-being through exploration of moderating factors. Hazen et al. (2020) found that post-intervention reflective practice quality moderated the association between pre-intervention vicarious trauma and post-intervention burnout, such that early childhood-serving providers who reported high levels of vicarious trauma and higher quality reflective practice had lower rates of burnout compared to the providers with high levels of vicarious trauma and lower quality reflective practice. Morelen et al. (2022) found that higher levels of self-reported COVID-19 stress were associated with lower levels of self-care only when providers had received no reflective supervision or less than one year of reflective supervision.

Other studies that examined these associations longitudinally, however, revealed conflicting results. For instance, the amount of time that early care and education providers had received reflective supervision did not affect their ratings of secondary traumatic stress, burnout, and compassion satisfaction from pre- to post-assessment (Brown, 2016). Within the same sample, providers who had received reflective supervision for more than two years reported decreases in their frustration with parents over the school year, while providers who had received one year or less of reflective supervision reported increases in frustration (Lepore, 2016). There were no changes in mindfulness or burnout in a sample of early care and education supervisors, program managers, grant specialists, and consultants following year-long participation in reflective learning groups (Shea et al., 2022). Some studies showed change in unanticipated directions: supervisors and providers in early intervention reported increases in work-related stress following participation in reflective supervision small groups for nine months (Frosch et al., 2018), and child welfare providers who participated in small groups for six months and whose small groups demonstrated increases in openness, reported increases in secondary traumatic stress and burnout and decreases in compassion satisfaction (Meuwissen & Watson, 2021). Watson et al. (2016a) found that while there were no changes in home visiting supervisors’ emotional exhaustion, depersonalization, or personal accomplishment scores after receiving reflective consultation, home visitors’ scores on emotional exhaustion increased after receiving reflective supervision; their scores on depersonalization and personal accomplishment did not change. These results suggest that associations between RS/C and well-being are complex, and that professional role and timing may be key in understanding these relations.

Qualitative Studies.

Ten of the 17 studies using qualitative methodology explored associations between RS/C and well-being (see papers marked with ± in Table 2). Respondents were nearly unanimous in their perceptions that RS/C improved their well-being and mitigated negative outcomes. The role of RS/C in reducing burnout emerged as an important theme in interviews with supervisors and providers, such that 4 of the 10 relevant studies identified this outcome specifically (Barron et al., 2022a; Shea et al., 2022; Susman-Stillman et al., 2020; Watson et al., 2016a). Moreover, two studies revealed that receiving reflective supervision was associated with reductions in secondary traumatic stress or vicarious trauma (Barron et al., 2022a; Begic et al., 2019) and one found that reflective supervision improved compassion satisfaction (Brown, 2016). Reflective supervision was also thought to help maintain job satisfaction (Susman-Stillman et al., 2020), increase engagement with work, motivate professional development, and increase professional efficacy (Barron et al., 2022a). Reflective supervision also buffered general work-related stressors (Brown, 2016). Watson and colleagues (2014) showed that reflective consultation reduced provider stress in the moment by providing validation, affirmation, and emotional release and regulation, and over time helped providers clarify professional roles and increase perspective taking. Providers noted feeling less overwhelmed, anxious, and isolated, and more empowered and confident.

Additional benefits of RS/C were identified, including strengthening providers’ ability to cope with stress (Frosch et al., 2019; Susman-Stillman et al., 2020), increasing use of personal and professional resources (Shea et al., 2022), and contributing to provider resilience through increased self-awareness and practice improvement (Russ et al., 2020). One study found that 86% of participants identified reflective supervision as an important self-care component, but thought that reflective supervision was not enough on its own (Eaves et al., 2021). Taken together, these studies provide strong support for the theory that RS/C contributes to well-being, though specific constructs measured varied greatly.

Research Question 2

Child and Family Outcomes

Qualitative Studies.

No studies explored the impact of RS/C on young children and families directly. Three qualitative papers, however, provided initial evidence that RS/C may change how professionals interact with the families they serve, an important preceding step as positive, more effective relationships between providers and families may be a catalyst for change in parent and child behaviors (see papers marked with § in Table 2). In a sample of home visitors and teachers, participants spoke about how reflective supervision facilitated their ability to bring up difficult topics, and ultimately develop stronger relationships, with the families they served (Barron et al., 2022a). Similarly, teachers reported more positive teacher-parent relationships, improved ability to communicate and cope with challenges in interactions with families, and greater referrals to collateral services for families (Lepore, 2016). In a community mental health setting, reflective supervision was identified as a mechanism by which the quality of clinical care may improve (Williams et al., 2019). These studies suggest potential positive effects for families, and ostensibly children, through improved interactions with providers who have experienced reflective supervision. However, these results are limited, and no studies have yet taken the final step of measuring child- or family-level outcomes directly.

Research Question 3

Barriers and Facilitators Affecting Implementation of RS/C

To document the ways in which workplace context affects provision of RS/C, five studies (quantitative and qualitative) examined barriers and/or facilitators at the organization- or program-level that influenced RS/C implementation, as well as participants’ experience of reflective practice (see Table 3). One quantitative study tested the association between receipt of reflective supervision and participant’s perceptions of workplace learning culture. The need for agency support and a workplace culture that valued reflective practice and supervision was the most prominent theme (Barron et al., 2022b; Eaves et al., 2022; Russ et al., 2020; Williams et al., 2019). A study conducted by Eaves and colleagues (2022) provided more nuance to this finding. They found that having a workplace policy supporting the provision of reflective supervision was not directly related to secondary traumatic stress, burnout, or compassion satisfaction in home visitors; however, having a workplace policy was associated with more consistently experiencing core components of a reflective supervisory relationship, which in turn was associated with each of these outcomes (Eaves et al., 2022). Interestingly, participants who reported higher quality reflective supervision and spent more time in group supervision perceived their workplace as having a better learning culture (Julien-Chinn & Lietz, 2019), suggesting a possible bidirectional effect of reflective supervision and workplace culture. Two major barriers to implementation of reflective supervision were other job demands, and thus limited time, and cost (Barron et al., 2022b; Watson et al., 2016a; Williams et al., 2019). Watson and colleagues’ (2016a) study further contextualized the barrier of cost: in interviews, grant administrators stated that with funding from the MIECHV grant for home visiting, there were sufficient financial resources for reflective supervision, but without those funds it would be difficult to support reflective supervision. Four studies each identified other individual, relational, or contextual barriers and facilitators unique to their samples; see Barron et al., 2022b, Russ et al., 2020, Watson et al., 2016a, and Williams et al., 2019 in Tables 2 and 3 for further information.

Discussion

We conducted a systematic review to synthesize the current empirical evidence pertaining to RS/C as an effective practice within early childhood-serving programs. The review aimed to (1) examine associations between RS/C and early childhood-serving professionals’ reflective capacity and well-being (at both the supervisor and provider levels), (2) examine associations between RS/C and child and family outcomes, and (3) investigate how individual, relational, and organizational factors influence the implementation of RS/C in early childhood-serving programs. The review also assessed the quality of the included studies to provide an overview of the current state of the literature. We identified a total of 28 published articles or dissertations generated from 24 unique samples that addressed these research questions.

Despite the adoption of RS/C by many early childhood-serving programs, and the widespread belief that RS/C is a critical component of IECMH best practices, a cohesive evidence base is just beginning to emerge. Other resources have consolidated some of the existing evidence for RS/C or what is known about its measurement (Eggbeer et al., 2010; Osofsky & Weatherston, 2016; Tomlin & Heller, 2016), and argue for the importance of aggregating evidence, but this is the first review using rigorous methodology to comprehensively illustrate what is known about RS/C while also providing an appraisal of study quality.

Overall, the quality of the included articles was moderate to strong, which speaks to a useful, emerging evidence base. However, it is important to note that no studies employed a randomized design, and only one study used a non-randomized comparison group. Therefore, it is not possible to state with certainty that changes in outcomes are due to RS/C. As mentioned before, included articles were appraised using cross-sectional and cohort checklists, which may have biased the studies towards higher ratings, as reviewers were not asked to evaluate studies based on whether they used randomization or comparison groups. Researchers who are interested in designing a RCT that is also ethical, feasible, and acceptable to stakeholders may consider using a “business-as-usual” waitlist control group design. For example, to test the effects of training supervisors in reflective supervision, the waitlist control group would not immediately receive professional development focused on skill building in reflective supervision, but would receive professional development after the experimental group completed the training and outcomes in both groups were assessed. Supervisors in the waitlist control group could still access other professional development opportunities and would continue to supervise their staff using their current approach. Similarly, in a RCT testing the effects of receiving RS/C, providers in the waitlist control group would initially receive supervision as usual while providers in the experimental group received RS/C. Waitlist control providers would then go on to receive RS/C themselves once outcomes were collected. This method also allows within-group comparison of outcomes (e.g., how reflective capacity or well-being change once providers receive RS/C).

When assessing the quality of the included articles, reviewers also did not assign ratings based on factors specific to RS/C. The latter point is important because a significant limitation to the extant RS/C research is a lack of clarity around what RS/C looked like in practice; this shortage of detail, paired with differences in methodology and setting, made it infeasible for this review to ascertain how differences in format and delivery (e.g., group vs. individual vs. both; supervision vs. consultation) differentially affected outcomes. High-quality study designs are clearly needed to establish RS/C as an empirically supported approach, while also demonstrating which characteristics of RS/C are responsible for positive change in outcomes.

Research Question 1

Reflective Capacity in Early Childhood-Serving Professionals

Results suggested that RS/C was associated with improvements in reflective capacity, skill, or self-efficacy, regardless of whether quantitative or qualitative methods were used. Although there were signs of improvement in reflective capacity across studies, methodological variability impacted results. Specifically, a few quantitative studies elected to measure individual skills rather than use a summary score of reflective capacity, and these studies invariably found that RS/C was linked to increases in some skills but not others. Furthermore, whether RS/C was associated with improvements in specific reflective skills was dependent upon whether the supervisor’s or supervisee’s perspective was considered (Shea et al., 2016). It would be useful for future studies to include summary scores of reflective capacity in addition to reporting on individual items, as well as include both supervisor and supervisee perspectives. One of the greatest challenges in conducting research on RS/C is determining how to measure reflective capacity, a complex and multifaceted construct (Tomlin & Heller, 2016). As we seek to measure reflective capacity in a reliable and valid manner, multiple perspectives should continue to be included (e.g., asking supervisors and supervisees to report on their perceptions of their own and the others’ reflective capabilities) and multi-method approaches should be utilized whenever possible (e.g., questionnaires, interviews or focus groups, and observational measures such as the RIOS [Watson et al., 2016b]).

Well-Being in Early Childhood-Serving Professionals

Results showed that in general, RS/C was positively associated with early childhood-serving professionals’ well-being (e.g., job satisfaction, engagement with work, compassion satisfaction, ability to cope with stress) and negatively associated with their distress (e.g., internalizing symptoms, secondary traumatic stress, burnout, intention to quit). However, study design influenced these findings, such that studies using qualitative methodology were much more likely to report associations in the expected directions. This points to the importance of using both quantitative and qualitative methods in studies of RS/C, as participants may perceive and report changes in their well-being that do not appear in quantitative measures (e.g., Watson et al., 2016a). Future research should explore the meaning behind participants’ viewpoints that they are benefiting from RS/C while at the same time quantitative measures are failing to capture these benefits.

Additionally, there may be an important element of time that affects results. It is possible that participating in RS/C for a year is not long enough to see improvements in well-being (e.g., Shea et al., 2022). In fact, participating in RS/C may be associated with increases in distress (e.g., Frosch et al., 2018; Meuwissen & Watson, 2021; Watson et al., 2016a). It is possible that as early childhood-serving professionals learn to become more aware of their own and others’ emotions, they increasingly recognize and resonate with their experiences of distress. We might hypothesize that as professionals continue to learn how to identify distress, cope with it, and use it effectively in their work, they will report reductions in distress; however, it is critical that future studies examine this phenomenon over time to learn how exactly providers manage stress. Furthermore, it will be useful to measure related constructs and how they covary across time. For example, in Frosch and colleagues’ (2018) study, although participants reported greater stress over time, they also reported that reflective supervision had improved their ability to effectively manage that stress. These authors theorized that high levels of stress may always be present among early childhood professionals, thus it is especially important for providers to receive RS/C to mitigate this stress and prevent a cascade of negative outcomes, including burnout and turnover (Frosch et al., 2018). This suggests a need for future research to carefully differentiate the types of stress that providers experience, and examine how RS/C affects each of these types of stress, along with well-being, over time. Importantly, RS/C alone may not be sufficient; Eaves and colleagues’ (2021) study suggests the value of reflective supervision within a larger system of support, including adequate pay and flexible hours.

Research Question 2

Child and Family Outcomes

A central tenet of RS/C is that of the parallel process, or the concept that an interaction at one level of the system will affect an interaction at another level of the system. This review found (1) moderate-to-strong evidence that experiencing RS/C expanded providers’ reflective capacity, (2) limited evidence that how providers think about their work (e.g., increased reflective capacity resulting from RS/C) affected their subsequent interactions with families, and (3) no evidence that changes in provider-family interactions impacted child and family outcomes. Thus, to fully understand whether and how the parallel process transpires within RS/C, more research is needed around the last two links. Specifically, documenting whether RS/C facilitates more effective relationships between providers and families is an important preliminary step in identifying whether RS/C benefits children and families in measurable ways. Indeed, results from three studies identified in this review demonstrated that RS/C holds promise for improving provider-family relationships through better communication and capacity for navigating difficult topics (e.g., Barron et al., 2022a, Lepore, 2016). It is theorized that through such relationships (in concert with effective, evidence-based practices), parents and children may then experience improved functioning, though this idea needs to be tested in future studies as it is absent from the literature. Although it is challenging from a methodological standpoint to obtain data about how RS/C affects provider-family relationships, and in turn, parent-child interactions and a host of relevant outcomes, incorporating multiple perspectives and observational measures will be essential. Importantly, RS/C may affect some aspects of relationships and certain child and family outcomes, but not others; therefore, extensive research is likely needed.

Research Question 3

Barriers and Facilitators Affecting Implementation of RS/C

Finally, this review sought to explore the barriers and facilitators to the implementation of RS/C within early childhood-serving programs. It is well known that promising and effective practices, such as RS/C, are often not implemented or sustained due to organizational, relational, or individual barriers, and this is certainly true in early childhood-serving systems and programs (List et al., 2021). Agency or program support of RS/C was the factor identified most frequently in implementation studies. Eaves et al. (2022) offers additional context for this idea: having a workplace policy to provide RS/C is associated with a greater likelihood of experiencing a positive reflective supervision relationship, and that relationship in turn is associated with well-being. In other words, agency or program policy makes it more likely that RS/C is taking place, though is not on its own sufficient in ensuring quality. Given that time and other work demands were other major barriers to RS/C in early childhood settings, workplace policies that embed RS/C in the day-to-day responsibilities of supervisors and providers may be critical, as is funding to carry out this approach.

Strengths and Limitations of the Current Review and the RS/C Literature

Strengths of this systematic review include the use of rigorous methodology following PRISMA guidelines and the PICOS framework, inclusion of quantitative and qualitative evidence, exploration of associations between RS/C and multiple relevant outcomes, and investigation of factors related to the implementation of RS/C in early childhood-serving programs. Although this review represents the first effort in the field to systematically synthesize the existing research on RS/C, a few limitations of the review itself should be noted. First, this review was limited to studies published in English. Second, and relatedly, it is possible that other studies that explored RS/C within early childhood-serving programs exist but were not identified during the literature search, despite the thorough search method used. Third, although two PhD-level researchers reviewed each article, it is possible that results of a study could have been misinterpreted, or that certain findings were mistakenly left out.

There are several significant limitations to the extant RS/C literature that affect the interpretation of this synthesis. First and foremost, many articles did not provide enough detail on the RS/C that study participants received, making it difficult to know what exactly RS/C looked like and which components of RS/C must be present for it to be effective. For example, there were variations in length of time that participants took part in RS/C, with several studies failing to report length of time at all, as well as whether participants had ever taken part in RS/C prior to the study. Whether RS/C was provided in individual versus group format, the training and credentials of the facilitators, what content participants received, and participant characteristics (e.g., race/ethnicity, education, years of experience in the field as well as in their current role and organization, endorsement status) were also frequently excluded. Exploring how RS/C is associated with outcomes for professionals of color—who may be navigating issues of racial discrimination and structural inequities in addition to the stress that often accompanies caregiving professions, especially for women—is a critical next step for determining the role that RS/C may play in reducing stress and bolstering well-being, especially for populations who have been oppressed (Eaves et al., 2022). While there have been long-standing efforts to understand how RS/C may be a culturally responsive practice, much work remains as it is unclear whether and how RS/C can help supervisors and supervisees reflect on their social identities and issues of power, privilege, and bias (Noroña et al., 2012; Stroud, 2010). It is also unclear whether exploration of these issues in the context of RS/C may then contribute to greater inclusivity and equity for professionals, children, and families within programs.

Another significant limitation to the extant literature on RS/C is that the range of measures employed to assess RS/C and associated outcomes varied substantially, making it difficult to compare results and definitively determine the effectiveness of RS/C. Moreover, several studies used non-validated tools or single items to assess outcomes, limiting validity. As previously mentioned, there were also significant limitations due to study design issues. All but one study failed to include a comparison group, and no studies utilized randomization procedures. Most of the studies relying on quantitative methods had small samples, and thus may have been underpowered to detect effects of RS/C. Although a number of studies employed pre-post cohort designs and examined effects of RS/C after nine months or a year, there was a general lack of long-term follow-up, which is necessary to understand how RS/C affects reflective capacity and well-being over time. Further, given the relatively limited number of studies conducted within each type of early childhood-serving setting (e.g., home visiting, early care and education, etc.), it was not feasible to examine results separately by setting, though there may be important differences. Similarly, it was difficult to distinguish between RS/C trainings that focused on increasing reflective capacity in supervisors and providers, RS/C trainings focused on teaching supervisors how to provide reflective supervision to their staff, and receipt of RS/C itself. Making these distinctions will help inform the field as to how supervisors learn to provide reflective supervision (e.g., through trainings, via their own access to RS/C), and by measuring outcomes at supervisor, provider, and child/family levels we will have a fuller picture of the parallel process in action. Given these serious limitations to the RS/C literature, it is important to balance enthusiasm for RS/C with recognition of the essential questions we do not yet have answers to; by appraising the research on RS/C with a discerning eye, we hope to encourage researchers to continue examining RS/C in rigorous ways that will ultimately benefit the early childhood workforce and the young children and families it serves (see Table 4 for a summary of research recommendations based on gaps identified in this review).

Table 4.

Future Directions for RS/C Research

Identified Gaps	Research Recommendations
Study Design and Reporting Issues
Lack of comparison groups and randomization	Use comparison groups, randomize participants to groups when possible Consider a “business-as-usual” waitlist control design, which may be more ethical, feasible, and acceptable in many settings
Lack of differentiation between training in RS/C and receipt of RS/C	Distinguish between training in RS/C and receipt of RS/C Describe characteristics of training (e.g., duration; qualifications of individual(s) conducting training; components/content of training; learning modalities; supports provided)
Treating reflective supervision and reflective consultation as interchangeable	State whether reflective supervision or consultation (or both) is being provided Describe the nature of the supervision or consultation (e.g., individual, group, or combination; frequency of meetings; length of meetings; duration of meetings over time; content discussed or curriculum used)
Omitting consideration of differences in format and delivery of RS/C	Describe the nature of the RS/C received (see above), which will help future systematic reviews and meta-analyses make comparisons Consider study designs that examine differences in outcomes based on qualities of RS/C (e.g., effects of group vs. individual RS/C, having a consultant in addition to a supervisor)
Differing findings based on whether quantitative or qualitative methods are used	Employ a mixed methods approach Explore the meaning behind participants’ viewpoints that they are benefiting from RS/C while at the same time quantitative measures are failing to capture these benefits in some instances
Lack of longitudinal examination of RS/C and relevant outcomes	Include multiple time points capturing outcomes related to both reflective capacity and well-being (e.g., baseline, mid-assessment, post-assessment, 6-month follow up, 1-year follow up, etc.)
Measurement Issues
Failing to measure reflective capacity, skill, or self-efficacy, as well as the process of RS/C and whether one is a “competent and skillful” reflective supervisor or consultant, in a valid, reliable manner	Use multiple methods (e.g., questionnaires, interviews or focus groups, observational measures such as the RIOS) Use validated tools rather than study-created or single-item measures Continue investigation of how to best measure reflective capacity and related constructs, a longstanding issue in the field (Eggbeer et al., 2010 comprehensively addresses this issue) Consider how reflective capacity and skill as a reflective supervisor/consultant changes over time (e.g., what does the process of becoming more reflective look like?)
Differing findings based on whether supervisors/consultants or supervisees/consultees are surveyed	Include multiple perspectives
Differing findings based on whether individual items or summary scores are used in analyses	Include summary scores of reflective capacity and related constructs rather than individual items
Inconsistency in findings on whether RS/C improves well-being of professionals	Differentiate types of stress and well-being, as well as coping and supports, and use valid and reliable measures to assess these constructs Investigate moderating factors to better understand in what contexts and for whom RS/C is associated with well-being or lack thereof Examine how RS/C affects these constructs, and interactions between constructs, over time
Inclusion Issues
Inadequacy of reporting of race/ethnicity of trainers, supervisors/consultants, and supervisees/consultees	Ask about race and ethnicity in studies
Lack of inclusion of people of color in research on RS/C	Include diverse participants Explore how RS/C is experienced by different groups of people, and how RS/C is associated with outcomes for professionals and families of color
Absence of knowledge on how RS/C affects provider-family relationships and interactions, and subsequently child and family outcomes	Include families, parents, and children in research Use multiple methods to capture changes in provider-family interactions (e.g., provider report, family report, observation) Assess outcomes at parent- and child-levels through multiple methods (e.g., questionnaires, observation, record review)
Implementation Issues
Lack of understanding of factors that affect implementation of RS/C in real-world settings	Continue to explore barriers and facilitators to implementation of RS/C, particularly in settings that have received less research in this area (e.g., early care and education) Consider and test interventions/approaches to address those factors (e.g., improving administration buy-in of RS/C)

Conclusions

In summary, RS/C holds great promise to positively affect people within early childhood-focused programs. Through this review, we found that the current body of research related to RS/C generally shows enhanced reflective capacity and improved well-being among professionals, though with caveats based on study quality. Research focused on child and family outcomes is unfortunately sparse, and examination of how RS/C affects relationships between providers and families, and in turn outcomes for children and their parents, is a clear future direction for the field. Continued investigation of the factors that affect whether RS/C is implemented and sustained within early childhood-serving programs is needed, with attention to several important barriers and facilitators. RS/C has deservedly garnered much interest as a key source of professional development support for early childhood-serving professionals, resulting in its rapidly growing use within numerous programs (Tomlin et al., 2016). Bringing rigorous research methods to bear on understanding the effectiveness of RS/C will help ensure that RS/C as a practice continues to grow in ways that best supports young children, and establishes RS/C as an empirically supported approach. This review was a crucial step toward that goal. As expressed by Tomlin and colleagues (2016, p. 625): “As the work is rich and complicated, so must our efforts to understand the work be similarly complex.”

Statement on Diversity and Anti-Racism.

This systematic review was conducted with an appreciation for diversity and an anti-racist approach. We reported the race, ethnicity, and gender of study participants from the included papers whenever possible. It is important to note that many papers did not include this information, and among those that did, the majority of participants self-identified as White, non-Hispanic women. We addressed the implications of this within the review. For instance, it is unclear whether and how reflective supervision and consultation (RS/C) benefits professionals of color, or improves equity and inclusion within early childhood-serving programs. Some researchers have begun exploring these questions (e.g., Eaves et al., 2021; Eaves Simpson et al., 2018), yet more studies are needed. From the professionals to the young children they serve, people in early childhood-focused settings hold many intersecting identities; thus, exploring the interplay of diversity and the effectiveness of RS/C is a critical task of future research.

Key Findings for Practitioners.

1. Reflective supervision/consultation was generally associated with enhanced reflective capacity and improved well-being in early childhood-serving professionals across settings, but these findings were tempered by methodological limitations in the existing research.

2. Little is known about whether and how reflective supervision/consultation impacts young children and their families, thus more research is needed.

3. Several factors affected whether reflective supervision/consultation was adopted and sustained in early childhood-serving programs, including agency/organizational support, job demands resulting in limited time, and cost.

Statement of Relevance.

Reflective supervision and consultation (RS/C) is a long-cherished practice within the field of infant and early childhood mental health, and has even been mandated in a number of early childhood-serving programs. While the enthusiasm for RS/C is well-deserved, and is also shared by these authors, a comprehensive evidence base for RS/C has not yet been established.

The current systematic review exhaustively summarized the existing empirical evidence for the effectiveness of RS/C as a first step towards understanding the impact of RS/C on the early childhood workforce and young children and families.