The COVID-19 has led to deepen changes in children’s daily routines (Prime et al., 2020). During the lockdown phases, preschool children only had their parents around them to promote a positive development and new learning experiences (Spinelli et al., 2020). Furthermore, contact with teachers and peers were not allowed (Spinelli et al., 2020). Even with the advances in vaccination, the re-opening phases have continued to be associated with uncertainty for preschool children who had to deal with new routines and restrictions in early childhood and care services (Sette et al., 2022, 2023).

However, the stressful disruptions related to the COVID-19 pandemic outbreak may have a differential impact on children’s socioemotional adjustment, depending on pre-existing vulnerabilities (Prime et al., 2020). High and stable behavioral inhibition (BI) during early childhood is one of the best predictors of later anxiety (Sandström et al., 2020) and may place individuals at increased risk of experiencing such emotional problems when confronted with stressful events characterized by high levels of uncertainty, like the COVID-19 pandemic crisis (Zeytinoglu et al., 2021). In fact, this temperamentally based wariness is characterized by high levels of caution, fearfulness, and vigilance in front of novel persons, situations, and activities (Fox et al., 2023). The developmental-transactional framework (Rubin & Chronis-Tuscano, 2021) establishes that highly inhibited preschoolers are more likely to experience shyness. This increased fearfulness and heightened self-consciousness during social interactions with unfamiliar peers (Rubin et al., 2009) may generalize to the preschool classroom (Rubin et al., 2018). Within this context, shy-inhibited preschoolers are more likely to display anxious withdrawn behaviors, that is, to isolate themselves from the peer group (Rubin et al., 2009).

Nevertheless, developmental-transactional theory and research acknowledges that parent, peer, and child factors may protect highly inhibited preschoolers against adverse socioemotional outcomes (Rubin & Chronis-Tuscano, 2021; Rubin et al., 2009). Within the family context, warm and sensitive parenting behaviors contribute to enhance children’s independence and emotion-regulation (Smith et al., 2019), reducing children’s avoidance in future social situations (Chronis-Tuscano et al., 2018). Highly inhibited preschoolers growing up in such family contexts are more likely to engage positively with peers (Smith et al., 2019) and to capitalize important developmental opportunities during peer interactions that can enhance their socioemotional skills (Rubin et al., 2018). This may place them in healthier developmental pathways (Rubin & Chronis-Tuscano, 2021; Rubin et al., 2009).

Narrative reviews have highlighted that there has been an increasing interest in evidence-based interventions targeted at BI/AW during the recent years (Chronis-Tuscano et al., 2018). Drawing from developmental theory and research on the transactional influences between parent, child, and peer factors (Rubin & Chronis-Tuscano, 2021; Rubin et al., 2009), these interventions may capitalize early neuroplasticity and prevent the occurrence of chronic emotional problems and comorbidities (Bayer et al., 2018; Chronis-Tuscano et al., 2018). Most interventions have focused on the promotion of children’s socioemotional skills in a peer group of inhibited preschoolers and/or have involved parents in face-to-face or online programs to modify their ways of thinking, behaving, and responding to their children’s needs (Ooi et al., 2022).

Few knowledge syntheses, using systematic review methods, have been conducted on currently available evidence-based interventions targeted at BI during the preschool years. Vallis et al. (2020) conducted a systematic review and meta-analysis about the effectiveness of individual, group, in-person, and online cognitive-behavioral interventions targeted at BI, anxiety symptoms and disorders during the preschool years. The authors identified four selective prevention interventions targeted at BI during the preschool years, published until September 2019: (1) the parent education program Cool Little Kids, delivered in-person (Rapee et al., 2005) and online (Morgan et al., 2016, 2017); (2) the combination of the Cool Little Kids with children’s socioemotional skills training (Social Skills Facilitated Play Program), delivered in-person (Lau et al., 2017); (3) the multicomponent Turtle Program (Chronis-Tuscano et al., 2015), combining Parent–Child Interaction Therapy adapted to anxiety problems with an extension of Social Skills Facilitated Play, delivered in-person; and (4) a parent home intervention, designed by LaFreniere and Capuano (1997). This study concluded that individual and group child-only, parent-only and parent–child cognitive-behavioral interventions were effective in reducing children’s BI and anxiety during the preschool years. However, Vallis et al. (2020) found that online interventions were heterogeneous in terms of therapist support and technological drawbacks and less effective when compared with in-person interventions. More recently, Ooi et al. (2022) conducted a meta-analysis to summarize the effectiveness of psychological interventions targeted at inhibited preschoolers, published until March 2021. In this meta-analysis, the same intervention programs identified by Vallis et al. (2020) were retained. The authors concluded that these intervention programs are effective at reducing BI, anxiety symptoms and diagnosis, when reported by parents and teachers, but not by observers.

Notwithstanding their contribution, prior systematic reviews, and meta-analyses (Ooi et al., 2022; Vallis et al., 2020) have essentially examined the effectiveness of available early interventions in the reduction of children’s negative individual emotional outcomes (i.e., BI, anxiety symptoms and diagnoses, or internalizing problems). The effectiveness of available interventions to improve negative (such as, non-social peer play) and positive domains of social functioning (such as, social engagement with peers, or social competence) needs to be clarified. In fact, children who display a stable pattern from BI to shyness and anxious withdrawal during the preschool years are more likely to be identified as “easy targets” by peers and to experience later peer victimization and exclusion (Rubin et al., 2018). Social disruptions in contacts with teachers and peers during the lockdown (Spinelli et al., 2020) and reopening phases (Sette et al., 2022, 2023) of the COVID-19 crisis highlight the need to examine the effectiveness of available interventions targeted at BI in child social domains of functioning. Furthermore, prior reviews and meta-analyses (Ooi et al., 2022; Vallis et al., 2020) have excluded universal preschool-based interventions with targeted elements for shy-inhibited preschoolers. Universal preschool-based interventions may be useful to counteract parent mental-health help-seeking barriers for internalizing problems (Johnson et al., 2023). In fact, this type of interventions has the advantages to be more accessible, to focus on the whole classroom, to reduce stigmatization and to be implemented in a naturalistic setting, where the vulnerabilities of highly inhibited preschoolers are particularly impairing (Coplan et al., 2018).

To overcome extant limitations, new knowledge syntheses, using systematic review methods, are needed. Recent research found that a stable pattern of BI during early childhood predicted greater worry dysregulation and, in turn, greater anxiety among young adults during the COVID-19 pandemic (Zeytinoglu et al., 2021). Furthermore, few studies also found that children’s shyness was positively associated with self-reported social anxiety, depression, and loneliness during the reopening stage after the first COVID-19 lockdown (Sette et al., 2022, 2023). The heightened vulnerability associated with BI during stressful events, like the COVID-19 pandemic crisis (Zeytinoglu et al., 2021), justifies the need to conduct a rapid review. In this type of knowledge synthesis, systematic review methods are streamlined and accelerated to provide evidence to decision makers and practitioners in due course (Garritty et al., 2021).

This rapid review was guided by the research question “Which psychological interventions have been effective in improving the socioemotional outcomes of inhibited, shy, and anxious withdrawn preschoolers in different settings?”. The specific study objectives were: (1) to synthetize the main characteristics of the studies (i.e., design, sample size and characteristics, screening procedures, intervention outcomes and measures) that examined the effectiveness of currently available interventions targeted at BI, shyness and anxious withdrawal during the preschool years; (2) to synthetize the main intervention characteristics (i.e., type, delivery mode, setting, number and time interval between the sessions, therapeutic approach, and intervention providers); and (3) to summarize the intervention effectiveness in reducing children’s negative individual and social outcomes and in improving children’s positive domains of social functioning.

Methods

Study eligibility criteria

Abstract inclusion followed a set of inclusion and exclusion criteria. Studies were included in the review if they met the following criteria: (1) Randomized Controlled Trial (RCT) design (including pilot RCT, cluster RCT); (2) participants included preschool children (between 3 and 5 years); (3) participants were inhibited, shy or socially-anxious withdrawn preschool children, as assessed using recognized measures; (4) evidence-based targeted psychological interventions, evidence-based selective preventive psychological interventions, or evidenced-based universal preventive psychological interventions with targeted elements for behavioral inhibition, shyness and/or anxious withdrawal; (5) control group without any type of intervention, treatment as usual, waitlist control group, or active control group; (6) studies published in the last 30 years (January 1991 to December 2021); (7) studies published in English; (8) published primary research studies.

No restrictions were established concerning intervention therapeutic approach, intervention therapeutic setting and the geographical location of the studies. Studies that were based on non-randomized controlled designs (quasi-experimental designs), descriptive, observational, and case studies were excluded. All kinds of reviews, protocols, book chapters, editorial letters, guidelines, websites were excluded. The main outcomes were child behavioral inhibition, shyness, social withdrawal, internalizing behaviors, anxiety symptoms, anxiety diagnoses, social skills, social behaviors, social engagement, assessed using validated instruments (teacher and parent questionnaires or diagnostic interviews, laboratory, or naturalistic observations). No secondary outcomes were considered, due to the scope of the review (i.e., rapid review).

Eligibility criteria were limited to primary research studies describing RCTs, published in English during the last 30 years and excluded other types of study designs, due to the scope of review (i.e., rapid review) that was conducted in the present study (Garritty et al., 2021). Inclusion criteria related to the population were established in accordance with the study specific objectives. Inclusion criteria related to the type of interventions (namely, the inclusion of universal preventive interventions with targeted elements for behavioral inhibition, shyness or anxious withdrawal) were defined to overcome existing gaps in previous literature reviews and meta-analyses (Ooi et al., 2022; Vallis et al., 2020).

Search strategy

This rapid review was conducted in accordance with the guidelines from the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA; Page et al., 2021). The guidelines of the International Prospective Register of Systematic Reviews (PROSPERO) were also considered. The protocol was registered and published in PROSPERO (Registration Number: CRD42021290615).

A systematic search was implemented in the PsycINFO, Web of Science, Scopus, and PubMed databases. The search strategy included the following combination of terms (combined with Boolean terms): “behavioral inhibition” OR “behavioural inhibition” OR “shyness” OR “social withdrawal” OR “anxious withdrawal” OR “social reticence” AND “preschool children” OR “preschoolers” OR “kindergarten children” AND “intervention” OR “intervention program” AND “randomized controlled trial” OR “pilot randomized controlled trial” OR “cluster randomized controlled trial” AND “internalizing behaviors” OR “anxiety symptoms” OR “anxiety diagnoses” OR “social skills” OR “social engagement” OR “social behaviors”. In addition, we applied filters for English language and publication date (January 1991 to December 2021).

The primary search strategy was preliminarily validated by the last author, by testing if known relevant records were retrieved and by checking errors of spelling, operator usage and line number combinations (Kleerings et al., 2023).

Study selection

Search results are synthesized in the PRISMA flowchart (Fig. 1). The systematic data search identified a total of 662 records, and 11 additional records were identified via a review of references in the articles that were retrieved. The research team developed and tested screening forms based on the inclusion and exclusion criteria. Prior to the screening process, a calibration exercise was undertaken to pilot and refine the screening form. Two authors independently screened each of the retrieved titles and abstracts against the inclusion and exclusion criteria. Reasons for exclusion were recorded. The remaining 37 articles were read in full to determine their eligibility. Any discrepancy in inclusion decisions was resolved by consensus, with the help of a last author. Thirteen articles met eligibility criteria and were included in the present rapid review

Fig. 1
figure 1

PRISMA Flowchart: This flowchart displays the number of records in the identification, screening and eligibility phases and the number of the records included in the rapid review

Full size image

Data Extraction

The two first authors extracted information about (1) the main studies’ characteristics (i.e., name of the authors, country, intervention name, study design, sample characteristics, screening procedures, child intervention outcomes and measures), and (2) the main interventions’ characteristics (i.e., type of intervention, delivery mode, intervention setting, targeted groups, number and time interval between the sessions, therapeutic approach, intervention provider, intervention integrity). No software was used for data extraction. However, an extraction form was developed by the research team to record the aforementioned criteria. Disagreements between the two first authors were solved through consensus, with the help of the last author.

Results

Study characteristics

The characteristics of the included studies are presented in Table 1. The 13 studies were published between 1997 and 2021.

Table 1 Summary of Studies’ Main Characteristics
Full size table

Interventions and geographical location

Practically half of the studies were conducted in Australia (n = 7) and assessed the efficacy of the Cool Little Kids (CLK) and its adaptations. Two studies were conducted in Canada on a parent home intervention (LaFreniere & Capuano, 1997) and on the Social Skills Facilitated Play (SSFP). Two studies were conducted in China on the SSFP when implemented alone and in combination with the CLK. Two studies on the Turtle Program (TP) were conducted in the USA.

Study designs

With respect to study designs, most studies (n = 9) were pilot RCT or large-scale RCT, with waiting-list control conditions. The other studies used usual care (n = 1), monitoring (n = 1) and active (n = 1) control conditions. In one study, the type of control group condition was not specified (LaFreniere & Capuano, 1997). Five studies on the TP, SSFP and parent home intervention (LaFreniere & Capuano, 1997) included only pre- and post-treatment assessments. The remaining studies included pre-treatment and short-term follow-up assessments (ranging from 2 months to one year).

Study samples

Sample sizes ranged from 16 to 545 children. Most studies (n = 7) included more girls than boys. Four studies included more boys than girls. In two studies, the proportion of girls and boys was equal. Six studies reported on caregivers’ living arrangements or marital status. These studies revealed that most children lived with both parents and most caregivers were married/cohabitating. Nine studies reported caregivers’ education. In these studies, the majority of caregivers (either mothers, fathers, or both) hold a university degree.

Screening procedures

With respect to screening procedures, most studies (n = 9) only used heterogeneous parent-rated questionnaires to identify child high BI, shyness, or anxious withdrawal. Two studies combined parent-rated questionnaires with laboratory assessments. One study combined parent-rated questionnaires with teacher ratings. One study only used teacher ratings. In two studies, children were included if at least one parent was diagnosed with an anxiety disorder (n = 1) or reported clinically significant internalizing symptoms (n = 1). Most studies excluded children with a known diagnosis of developmental disorders (n = 8) or whose parent ratings revealed suspected autism spectrum disorders (n = 2). Four studies excluded families, when parents did not have conversational or reading English skills.

Child intervention outcomes

With respect to child intervention outcomes, nine studies assessed child anxiety symptoms, inhibited-withdrawn and internalizing behaviors, using parent-rated questionnaires. Two studies only relied on parent-rated questionnaires. Six studies combined parent-rated questionnaires with structured diagnostic interviews conducted with caregivers to assess anxiety diagnoses. Five studies used teacher ratings on child anxiety symptoms, non-social and socially competent behaviors. Three studies used classroom observations on child non-social and socially competent behaviors.

Parent-rated and teacher-rated questionnaires, structured diagnosis interviews and classroom observation measures were quite heterogeneous across studies (see Table 1).

Intervention Characteristics

Table 2 displays the main characteristics of the interventions.

Table 2 Main Characteristics of the Interventions
Full size table

CLK

Eight studies focused on the CLK that is based on cognitive-behavioral therapy (CBT). In these studies, the CLK was mostly delivered in group (n = 5) and in face-to-face (n = 6) formats. Facilitators hold academic qualifications in psychology. Three studies used adaptations of the CLK. Two of them combined the CLK with children’s activities, drawn from the SSFP (n = 1) or the Parent–Child Interaction Therapy (PCIT, n = 1). Four studies on the face-to-face CLK reported that the intervention was implemented either in a university clinic (n = 2) or preschool (n = 2) settings. However, studies on the CLK were heterogeneous in terms of group size (from 5 to 16 caregivers), number (from 6 to 11) and time interval (weekly, or fortnightly) of the sessions. Two studies focused on the CLK, delivered online in an individual and self-administered format.

SSFP

Two studies focused on the SSFP, delivered in a group of inhibited preschoolers. In both studies, the SSFP was delivered in a face-to-face format and in preschool settings by trained facilitators with qualifications in education. Nevertheless, these studies differed in terms of the number (from 7 to 14) and time interval (weekly, or biweekly) of the sessions.

TP

Two studies focused on the TP, delivered in parallel groups of parents and children. In these studies, the TP was delivered in a face-to-face format and in a university clinic setting. Both studies reported that the TP consists of eight weekly sessions with 5–6 families, drawn on PCIT and SSFP principles.

Parent home intervention

One study focused on a face-to-face individual home parent intervention (LaFreniere & Capuano, 1997). This intervention was delivered by trained graduate students. The therapeutic approach was based on attachment, attributional and behaviorist therapeutic perspectives. It consisted of 20 sessions, spread over 6 months.

Summary of intervention outcomes in domains of child individual and social functioning

Table 3 summarizes the main intervention outcomes in domains of child positive and negative individual and social functioning, using heterogeneous methods and informants.

Table 3 Intervention Outcomes in Domains of Child Socioemotional Functioning
Full size table

Parent reports

Studies on the CLK and the TP consistently found a significant reduction in parent-rated BI, shyness, anxiety symptoms, anxiety interference, internalizing or emotional problems from pre-treatment to immediate post-treatment or short-term (e.g., 6 months, 1-year) follow-ups. These intervention effects were identified in studies involving waiting-list (Chronis-Tuscano et al., 2015; Kennedy et al., 2009; Lau et al., 2017; Luke et al., 2017; Morgan et al., 2017) and usual care (Bayer et al., 2018) control groups.

The only exceptions were the studies of Luke et al. (2017), Doyle et al. (2021) and Rapee et al. (2005) on the CLK. These studies didn’t find significant reductions in parent-rated internalizing problems, anxiety symptoms and behavioral inhibition from pre-treatment to follow-ups, respectively.

In the pilot study of the CLK online (Morgan et al., 2016), significant reductions in parent-reported children's anxiety, emotional symptoms, and anxiety interference were found in both intervention conditions (clinician support vs. no clinician support) from pre- to post-treatment.

Clinician-rated anxiety diagnoses, using structured interviews with parents

Four studies on the CLK and the TP found a significant reduction in clinician-rated anxiety disorders from pre-treatment to immediate post-treatment or short-term (i.e., 6-months, 1 year) follow-ups. These intervention effects were found in studies involving either waiting-list or monitoring control group conditions (Chronis-Tuscano et al., 2015; Kennedy et al., 2009; Lau et al., 2017; Rapee et al., 2005).

However, two studies on the CLK diverged from these findings. Bayer et al. (2018) only found minor differences from pre-treatment to 1-year follow-up in clinician-rated anxiety disorders in the intervention group when compared with usual care control group. Doyle et al. (2021) found that the reduction in clinician-rated anxiety disorders from pre-treatment to 6-months follow-up was similar in the intervention and waiting-list conditions.

Laboratory assessments

The few studies that used laboratory measures to assess the intervention effects of the CLK in BI found inconsistent findings. Kennedy et al. (2009) found a significant reduction in laboratory-assessed BI from pre- to 6-months follow-up in the intervention group when compared to a waiting-list condition. Conversely, Rapee et al. (2005) didn’t find a significant reduction in laboratory-assessed BI from pre- to 6-months follow-up in the intervention group when compared to a monitoring condition.

Teacher reports

The two studies assessing the TP (Barstead et al., 2018; Chronis-Tuscano et al., 2015) and one study on the CLK (Luke et al., 2017) found comparable findings, using teacher-rated reports. These studies found a reduction in teacher-rated school anxiety, fearful and anxious behaviors from pre to immediate post-treatment and 6-months follow-up assessment. In contrast, one study on the SSFP (Coplan et al., 2010) and the study on the parent home intervention (LaFreniere & Capuano, 1997) didn’t find significant intervention effects in teacher-rated anxiety and anxious withdrawal from pre to immediate post-treatment assessment.

With respect to socially competent behaviors, LaFreniere and Capuano (1997) found a significant improvement in teacher-rated social competence from pre- to post-treatment. Conversely, one study on the SSFP (Coplan et al., 2010) and one study on the CLK (Luke et al., 2017) didn’t find significant improvements in teacher-rated prosocial behaviors and social initiative from pre to immediate post-treatment and 6-months follow-up assessment.

Classroom observations

Two studies on the SSFP and one study on the TP found consistent findings concerning children’s non-social and social behaviors. Trained and blind observers reported significant improvements in peer play interactions, peer initiations (Barstead et al., 2018), observed socially competent and prosocial behaviors (Coplan et al., 2010; Li et al., 2016) from pre- to post-treatment and short-term follow-up (2 months). Coplan et al. (2010) also found significant reductions in observed socially wary behaviors from pre- to post-treatment.

Discussion

Overall, our rapid review identified 13 pilot and large-scale RCTs, published between 1997 and 2021, on selective prevention interventions targeted at BI during the preschool years.

Study characteristics

Our findings show that most reviewed studies were conducted in North America and Australia, with the exception of two studies conducted in China (Lau et al., 2017; Li et al., 2016). These findings are consistent with the conclusions of the narrative review of Chronis-Tuscano et al. (2018), showing that selective prevention interventions targeted at BI during the preschool years need to be evaluated in other cultural contexts. These conclusions converge with the central tenets of the bioecological developmental framework (Bronfenbrenner & Morris, 2006) that establishes that cultural norms may shape the way how parents and peers think, feel, and respond to children’s social behaviors, such as shy-inhibited behaviors.

With respect to samples’ composition, we found that the reviewed studies involved a greater proportion of girls living in intact families, whose caregivers hold university degrees and middle to high incomes. The findings concerning the representation of child sex diverge from the conclusions of the systematic review of Doey et al. (2014) that didn’t reveal differences in the percentages of shy girls and boys but concluded that shy-inhibited behaviors may carry increased negative socioemotional costs for boys when compared with girls. Furthermore, our findings support the conclusions of prior research (Barstead et al., 2018; Chronis-Tuscano et al., 2018) that highlighted the need of partnering with preschools serving more diverse populations in terms of family structure, education, and income. Our findings also show that only a minority of studies provided information on the sociodemographic characteristics of both mothers and fathers. These findings are in line with the meta-analysis of Carnes et al. (2019), showing that most studies only included or reported results on mothers in childhood anxiety treatment. These conclusions are noteworthy, because meta-analyses found that fathers’ overprotective parenting is at least as important as mothers’ and that paternal, but not maternal challenging parenting is associated with less child anxiety (Möller et al., 2016).

From a methodological standpoint, our rapid review reveals that most studies were small-scale pilot RCTs with pre- to post-intervention and/or short-term follow-up assessments and used waiting-list control group conditions. These findings converge with the conclusions of the meta-analysis of Ooi et al. (2022), showing that only half of the studies were rated as having adequate sample sizes and having a trial that is long enough to assess longer-term outcomes (6 months or more).

Our findings also indicate that screening procedures and intervention outcomes assessment relied, to a large extent, on parent reports and used a wide range of measures across studies. These findings support the conclusions of the meta-analysis of Ooi et al. (2022), showing that inhibited, shy, and anxious withdrawn behaviors are conceptualized and operationalized in a heterogenous way across studies. Parent reports are often considered a fast and economical method of obtaining information on child BI in literature (Bishop et al., 2003). However, these findings warrant attention, because prior research found that parents typically display difficulties in identifying children’s internalizing problems (Johnson et al., 2023) and in seeking help for anxiety-related problems, especially in a preventive manner (Mian, 2014). Furthermore, the meta-analysis of Ooi et al. (2022) concluded that parent reports need to be interpreted with caution, since parents were not blind to the condition allocation to which their children were assigned to, due to their active involvement in the intervention in most studies.

Intervention characteristics

The 13 studies that were retained on our rapid review focused on four evidence-based selective prevention interventions targeted at BI: Cool Little Kids (CLK, Rapee et al., 2005), Social Skills Facilitated Play (SSFP, Coplan et al., 2010), the Turtle Program (TP, Chronis-Tuscano et al., 2015) and an intensive parent home intervention (LaFreniere & Capuano, 1997). These four interventions were comparable to those retained in prior systematic reviews and meta-analyses (Ooi et al., 2022; Vallis et al., 2020) and featured two main intervention traditions.

Seven of the 13 studies featured the first intervention tradition that was described in the meta-analysis of Ooi et al. (2022), that is, parent-only education programs. This intervention tradition converges with developmental research, showing that sensitive parenting may enhance the emotion-regulation skills, independence and social engagement of inhibited preschoolers and place them in healthier developmental pathways (Fox et al., 2023; Hane et al., 2008; Lewis-Morrarty et al., 2012). With only one exception (LaFreniere & Capuano, 1997), all studies from this intervention tradition focused on the CLK (Bayer et al., 2018; Luke et al., 2017; Morgan et al., 2016, 2017; Rapee et al., 2005). A noteworthy number of studies on the CLK carefully described facilitators’ training and intervention integrity monitoring. Nevertheless, our findings show that the implementation of the CLK was heterogenous across studies in terms of the intervention type (i.e., group vs. individual), delivery mode (i.e., face-to-face vs. online), group sizes, intervention setting, number, and time interval of the sessions. These results are noteworthy, since the intervention characteristics can be moderators of the intervention effects (Ooi et al., 2022).

Consistent with previous systematic reviews and meta-analyses (Ooi et al., 2022; Vallis et al., 2020), our findings reveal that the CLK was the only selective prevention intervention targeted at inhibited preschoolers that was adapted to and tested in an online format (Morgan et al., 2016, 2017), until december 2021. Intervention monitoring showed that the take-up of clinician support (i.e., support call) in the CLK Online was surprisingly low (Morgan et al., 2017). Limitations in the take-up of clinician support in the CLK Online converge with recent efforts to introduce more intensive online individual and group-based parenting interventions for anxious (Comer et al., 2021) and inhibited preschoolers (Guedes et al., 2023), drawn on PCIT, CBT and videoconferencing. These findings warrant attention, because the meta-analysis of Vallis et al. (2020) found that the heterogeneity of online interventions for anxious and inhibited preschoolers, in terms of clinician support, may influence intervention outcomes.

Two studies (Coplan et al., 2010; Li et al., 2016) featured the second intervention tradition described in the meta-analysis of Ooi et al. (2022), that is, child-only interventions working directly with inhibited preschoolers in a peer group with similar difficulties. This intervention tradition, represented by the SSFP, converges with developmental theory and research, showing that peer interactions enhance the development of children’s age-appropriate social and socio-cognitive skills that place inhibited preschoolers in healthier developmental pathways (Rubin et al., 2018). However, prior studies on the SSFP acknowledged that this intervention remains to be adapted for use as a teacher-delivered classroom-based universal prevention program (Li et al., 2016).

No universal prevention program with targeted elements for inhibited preschoolers was retained for inclusion in our rapid review. These findings diverge from the central tenets of the bioecological developmental framework (Bronfenbrenner & Morris, 2006) that establishes that children’s characteristics (such as, BI) and contextual proximal factors (such as, teacher and peer responses) shape the bidirectional interactions with peers and teachers in the preschool classroom. To the best our knowledge, only the INSIGHTS encompasses a teacher-delivered universal classroom-based intervention component with targeted elements for shyness for kindergarten children living in urban low-income areas (O’Connor et al., 2014). Notwithstanding its contribution, the large-scale RCT on the INSIGHTS have primarily examined the intervention effects on children’s early literacy skills (McCormick et al., 2018; O’Connor et al., 2014) rather than on children’s emotional and social domains of functioning. This explains why the studies on the INSIGHTS were not retained for inclusion in our study.

Four studies that were included in our review combined the two intervention traditions that were identified in the meta-analysis of Ooi et al. (2022), that is, both parent-focused and child-focused approaches. These multimodal intervention approaches encompass diverse adaptations of the CLK to enhance child involvement (Doyle et al., 2021; Lau et al., 2017) and the Turtle Program (Barstead et al., 2018; Chronis-Tuscano et al., 2015). Multimodal intervention approaches converge with the central tenets of the developmental-transactional framework (Rubin & Chronis-Tuscano, 2021), establishing that the transactional child, parent, and peer influences shape the developmental pathways of inhibited preschoolers.

Intervention outcomes

Our findings show that the CLK and the TP were effective in reducing child negative individual outcomes (i.e., anxiety symptoms and diagnoses) at immediate post-intervention assessment, according to parents, teachers, and clinician ratings. Similar conclusions emerged in prior systematic reviews and meta-analyses (Ooi et al.., 2022; Vallis et al., 2020). However, teacher, parent, and clinician-reported improvements in favor to intervention conditions were less consistent in studies on the parent intensive home intervention (LaFreniere & Capuano, 1997), on the CLK (Bayer et al., 2018; Rapee et al., 2005) and its adaptations outside Australia (Doyle et al., 2021; Luke et al., 2017) that examined the intervention outcomes at short-term follow-ups (i.e., 6-months and 1-year). It is possible that selective prevention interventions targeted at BI contribute to modify the more transient expressions of anxiety (Ooi et al., 2022), especially when focusing mainly on parents and on the parent–child relationship. More specifically, children’s wariness, emotional reactivity and avoidance when confronted with unfamiliar persons, situations, and activities (Fox et al., 2023) may remain unchanged (Ooi et al., 2022) and revert to more typical ways of responding across time. The inconsistent changes that were found in the few studies on the CLK that used laboratory assessments of BI appear to support this idea. Comparable conclusions were drawn from the meta-analysis of Ooi et al. (2022) that didn’t find intervention changes in BI, using laboratory assessments. However, inconsistent findings need to be interpreted with caution, due to the heterogeneity of methodological (e.g., measures, assessment times, control groups) and intervention (e.g., number and time interval of the sessions, targeted groups) features across studies.

Our rapid review extends prior systematic reviews and meta-analyses (Ooi et al., 2022; Vallis et al., 2020), by synthetizing the intervention outcomes in children’s positive and negative domains of social functioning. Three small-scale pilot RCTs on the TP (Barstead et al., 2018) and the SSFP (Coplan et al., 2010; Li et al., 2016) found significant short-term improvements in children’s socially wary behaviors, social competence, prosocial behaviors, and positive peer interactions, using classroom observations conducted by trained and blind researchers. In both programs, the intervention sessions took place in a playroom that is as similar as possible to a preschool setting (Coplan et al., 2010; Li et al., 2016). It is possible that this type of intervention programs that promotes age-appropriate play and social skills enhances the generalization of intervention effects to the preschool classroom (Druskin et al., 2023). However, these promising findings need to be interpreted with caution, due to the methodological characteristics (i.e., small sample sizes, short-term assessments) of the studies.

Our findings reveal that pre-to-post intervention changes in teacher ratings on child positive domains of social functioning (i.e., social competence, or prosocial behaviors) were inconsistent across studies (Coplan et al., 2010; LaFreniere & Capuano, 1997; Luke et al., 2017). Intervention changes were only identified in the study on an intensive home parent intervention (LaFreniere & Capuano, 1997), but not in studies on the SSFP, either implemented alone (Coplan et al., 2010) or in combination with the CLK (Luke et al., 2017). These inconsistent findings may be related to the heterogeneity in the conceptualization and operationalization of child positive domains of social functioning across studies. Moreover, the intervention dosage was higher in the study on the intensive home parent intervention (i.e., 20 weeks, LaFreniere & Capuano, 1997) than in the studies on the SSFP (i.e., 6 to 8 sessions, Coplan et al., 2010; Luke et al., 2017). Also, teachers may be more concerned with disruptive behaviors in the classroom rather than with noticing children coping positively when facing affective and emotional challenges and have multiple children to compare with (Fernandes et al., 2020). This may contribute to the non-significant results in teachers’ reports of child positive domains of social functioning.

Strengths and limitations

To advance the current state-of-art knowledge, this rapid review aimed to provide a comprehensive synthesis of the characteristics of RCTs that examined the effectiveness of evidence-based interventions targeted at inhibited preschoolers. Furthermore, the main features of currently available interventions, including universal preventive classroom-based interventions, were synthetized. Instead of focusing only on child individual negative emotional functioning, this rapid review summarizes the intervention outcomes in child positive and negative domains of social functioning.

However, some limitations need to be acknowledged. This rapid review only focused the systematic search in four databases, in studies drawing on pilot or large-scale randomized controlled designs and published in English, between 1991 and 2021. The strategy search was preliminarily validated, but rapid reviews searches may be characterized by a more reduced sensitivity when compared with systematic reviews that aim to find all known relevant records (Kleerings et al., 2023). The review only identified thirteen pilot and large-scale randomized controlled trials assessing the effectiveness of evidence-based interventions targeting BI in children’s positive and negative socioemotional outcomes. In most studies, samples consisted of highly educated mothers, who were married/cohabitating and hold medium-to-high incomes, from North American and Australian contexts. Sample compositions limit the generalizability of the findings on the effectiveness of interventions targeted at inhibited preschoolers to fathers and more diverse populations in terms of family structure, education, income, and culture. Studies were heterogeneous in terms of methods (i.e., instruments, control groups, intervention assessments) and intervention characteristics (i.e., number and time interval of sessions, delivery mode and targeted groups), which makes it difficult to compare and interpret the findings. Due to the scope of the present review (i.e., rapid review), the risk of bias and the quality of the evidence were not evaluated.

Conclusions, recommendations, and future directions

Overall, thirteen pilot and large-scale RCTs on four evidence-based interventions targeted at BI were retained in the present rapid review. Most studies were conducted in samples of married/cohabitating and highly educated mothers and almost half of them focused on parent-only interventions. Only three pilot RCTs on a child-focused approach and on a multimodal intervention found significant post-intervention improvements in observed child social competence and positive peer interactions. Pre-to-post intervention changes in teacher ratings of child social outcomes were less consistent across studies.

This rapid review has implications for future research to overcome existing limitations in the current state-of-art knowledge. Further pilot RCTs need to be conducted in cultural contexts that differ from those where the intervention programs were initially developed. Cross-cultural studies may inform cultural tailoring that may be needed before implementing evidence-based interventions targeted at inhibited preschoolers in new cultural groups (Gonzales et al., 2016). Minor modifications in the way how intervention activities and materials are presented to inhibited-withdrawn preschoolers and their caregivers may be needed to enhance their acceptability (Guedes et al., 2019a, b, 2021). Moreover, large-scale RCTs, with active or usual care control groups and more diverse samples in terms of child sex, family structure, education, and income are needed to examine the generalizability of intervention effects in families who face sociodemographic adversity. The perspectives of both mothers and fathers need to be included or reported, because fathers’ parenting is at least as important as mothers for child anxiety (Möller et al., 2016). The moderating role of child (e.g., sex) and family (e.g., education, income) in the intervention effects need to be examined.

From a methodological standpoint, medium to-long-term follow-up assessments need to be included to examine the maintenance of the intervention effects across time, using a multi-informant (e.g., parents, teachers, clinicians, trained observers), multi-method (e.g., observations of parenting and child social behaviors in naturalistic and laboratory contexts, questionnaires, and diagnostic interviews) and a multi-domain (positive and negative child emotional and social outcomes) approach. Intervention process assessments (e.g., intervention monitoring, integrity, or dosage) need to be conducted to explore the moderating role of intervention characteristics and implementation in the intervention effects. For online interventions, large-scale RCTs need to examine the intervention effects, depending on the clinician support model. For multimodal interventions, the single or addictive effects of parent, child, parent–child, or teacher modules need to be examined to clarify for whom more intensive interventions are needed (Chronis-Tuscano et al., 2015, 2022).

From a clinical standpoint, these findings have implications for further intervention developments after the COVID-19 pandemic crisis, due to the heightened vulnerability associated with BI. This rapid review suggests that novel approaches in community-based intervention strategies and/or add-on motivational modules are needed to enhance parent engagement in underserved communities. Screening procedures at preschools may be particularly useful after the COVID-19 crisis for the early identification of highly inhibited preschoolers. Furthermore, the adaptation of peer-play interventions to a teacher classroom-based format and the development of universal preschool interventions with targeted elements for BI may improve the accessibility of interventions for inhibited children living in underserved communities. Online interventions may also be useful to overcome accessibility barriers. However, clinician alternative support models that can enhance the therapist-client alliance need to be introduced in internet-delivered interventions and their intervention effects need to be explored.