Multi-Informant Universal Mental Health Screening for Preschool-Aged Children by Parents and Educators: A PRISMA Systematic Review

Early intervention refers to a broad array of academic, medical or social support or treatment programs designed to enhance a young child’s development (Ramey & Ramey, 1998). Programs can be delivered to at-risk children, with the aim of being preventative, or targeted treatment programs implemented with children with known or diagnosed problems. Identifying children who may be at risk of MH difficulties, or recognising specific behaviour or emotional difficulties as requiring further investigation is the first step in early intervention. Early identification of children with MH concerns often leads to intervention, thus disrupting the often stable trajectory of MH difficulties from childhood to adulthood.

Parents and other primary caregivers play a significant role in help seeking for their children, that is, the recognition of emerging MH problems and facilitating access to services (Schnyder et al., 2020). The parental help-seeking process has been identified as (1) Parents’ recognition of MH difficulties, (2) Parents’ recognition of a need for professional help, (3) Parents’ actively seeking help, and (4) Family accessing and receiving necessary help/support (Reardon et al., 2017). However, parents do not always recognise behavioural or emotional difficulties as requiring further investigation, nor do they always appreciate the need for early intervention despite the established evidence base (Kowalenko, 2012). This is particularly pertinent in the preschool age range in which differentiating between psychopathology and normal early childhood development is a major challenge (Sim et al., 2019). A nationally representative survey of Australian parents of children under 18 years found that only 35% of parents were confident that they would recognise the signs of MH difficulties in their children and fewer than half (44%) knew where to access professional help if their child was experiencing difficulties, with confidence rates significantly lower in parents of younger than older children (Rhodes, 2017). A longitudinal study has shown that parents who recognise their child’s problem behaviour are more likely to access professional help for their child at age 3 (Oh et al., 2015). Thus, this evidence demonstrates the importance of recognising MH concerns as the first step towards parental help-seeking, including further assessment and accessing early intervention services.

Children with untreated MH difficulties can experience far-reaching negative effects throughout adolescence and into adulthood. In children as young as 3, behavioural and emotional difficulties have influenced wellbeing in later childhood and early adulthood across domains such as employment, education, criminal activity, physical and MH (Caspi et al., 1996; Jones et al., 2015). However, left untreated, the implications of low rates of identification mean that children in need of MH services frequently do not receive appropriate intervention and face poor outcomes.

Numerous papers have outlined the poor rates of help-seeking particularly amongst parents of young children (Ellingson et al., 2004), low rates of professional service utilisation amongst preschoolers with clinical diagnoses (Lavigne et al., 2009) or unmet MH needs in children more generally (Hiscock et al., 2020; Kataoka et al., 2002). In fact, the highest rates of unmet MH needs in the United States are in children under 6 years, who are also children of colour and from low-income families (Kataoka et al., 2002).

Given overburdened healthcare systems and the long-term ramifications of untreated MH difficulties, UMHS offers a cost-effective mechanism for identifying young children in need of support (Humphrey & Wigelsworth, 2016). UMHS can be an effective tool for identifying MH concerns by improving recognition and increasing access to early intervention services, thereby facilitating help-seeking by parents or other informants (Humphrey & Wigelsworth, 2016). UMHS provides early identification of individuals with MH symptoms, who might require further follow up and it may also include referral pathways to appropriate services (Lavigne et al., 2016). It can be conducted as a single assessment with questionnaires which are scored according to clinical cut-offs, or involve multistage assessments in which children are identified in a first stage and receive further follow-up or assessment in subsequent stages. As with other standardised assessments, it can be completed individually by children, parents, educators, or other health professionals. UMHS does not make diagnoses of specific disorders and is not a diagnostic tool.

UMHS in the preschool years is important because it provides a critical opportunity to identify children at risk of MH difficulties and provide early intervention prior to difficulties becoming more entrenched and severe (Moore et al., 2022a). This is especially important since early intervention, implemented in early childhood, has demonstrated high rates of cost effectiveness (Heckman, 2008). Despite the potential benefits of UMHS for preschoolers, there is a paucity of research examining the efficacy, acceptability, and accuracy of UMHS in this developmental period (Anderson et al., 2019). A recent review by Becker-Haimes et al. (2020) of free, brief, and accessible child MH measures identified only two UMHS measures appropriate for use with preschool-aged children which were rated as having “excellent” psychometric accuracy, the Pediatric Symptom Checklist (PSC; Jellinek et al., 1988) and the Strengths and Difficulties Questionnaire (SDQ; Goodman & Goodman, 2009). Of these measures, only the SDQ was also identified as suitable for multi-informant use by parents and teachers. The current review builds on the work of Becker-Haimes et al. (2020) by specifically evaluating the effectiveness and acceptability of multi-informant screening measures in the preschool age range.

Multi-informant report is a hallmark of developmental research and practice as it captures information about the child from multiple sources. Therefore, it is likely that UMHS for preschoolers will be particularly effective when multiple reporters are included (Scott et al., 2011). Such approaches incorporate the unique perspectives of the child’s behaviour and wellbeing across settings such as the home and school or childcare by different caregivers (Anderson et al., 2019). Given the challenges of identifying normative versus concerning emotions and behaviour in the preschool-age bracket, secondary informants such as education or childcare providers are ideal as they may offer important and distinct insight into child MH in addition to parent perceptions (Feeney-Kettler et al., 2011; Smith, 2007). Educators are often familiar with children’s social interactions, behaviour and wellbeing as they spend significant amounts of time observing children in naturalistic environments and in play with similar-aged peers. In Australia and the United States, the majority of 4–5 year old children are enrolled in preschool or early childhood education and care, attending between 15–21 h per week on average (Australian Bureau of Statistics, 2021; National Center for Education Statistics, 2016, 2018), again emphasising that educators are well-placed to conduct UMHS with an accessible cohort of children (DeLucia et al., 2022). However, there remains a need for children with MH difficulties to be systematically identified and referred by educators as part of an integrated model for child MH assessment (Casale & Reyes, 2023).

In contrast to primary care settings or formal MH services, preschools, pre-kindergarten, childcare, long daycare or early learning centres (henceforth referred to as ‘preschools’) are an ideal setting for UMHS since they are less encumbered by issues of access and stigma (Desta et al., 2017). Like schools for older children, preschools offer a unique opportunity to address the MH needs of children (der Embse & De Los Reyes, 2024). Easy access to preschools means that many parents turn to educators or preschool staff in the first instance to discuss MH concerns they have about their child(ren). In fact, research shows that parents frequently seek help from an informal source such as educators, friends or family, prior to seeking help from professional medical specialists, education or MH services (Pavuluri et al., 1996). Yet preschool educators have been subject to minimal research regarding their perspectives on child MH, and their role in identifying and supporting young children with socio-emotional or behavioural concerns (Croft et al., 2015). Instead, research to date on educator involvement in UMHS has focussed on primary and secondary school teachers. Research with older children has shown that over-reliance on single sources of information and decision-making based on non-scientific factors (i.e., factors which are not evidence-based such as small standardisation samples which are not representative of the target population) contributes to inequitable school MH decision-making (von der Embse & De Los Reyes, 2024). These findings are likely to be relevant for younger children as well, therefore emphasising the need for validated, multi-informant assessment.

Identifying emotional and behavioural concerns in children can be undertaken with varying degrees of accuracy by parents and other primary caregivers each with their own level of biases, knowledge and understanding of the child. Whilst preschool educators have a referent classroom group by which to compare the child, they may have little to no experience in the area of child MH (Stormont & Stebbins, 2005). Screening practices that rely exclusively on either parent only or educator only reports omit important information about child behaviour in different settings and are at risk of misidentifying children with MH concerns (Eklund & Dowdy, 2014; Stefan & Miclea, 2017). Thus, in order to identify children who may require support and additional MH services, parents and educators need clinically useful and effective screening measures that allow for multi-informant reports to help them accurately identify children at risk.

A range of screening tools currently exist for identifying MH difficulties in young children; however, little is known about the multi-informant measures that exist, and the clinical utility and effectiveness of these measures. The clinical utility of these measures needs to be assessed to make value judgements of UMHS in this population group. However, “clinical utility” is a poorly defined construct in the UMHS literature and has been used to encompass an array of broad concepts such as feasibility, practicability, acceptability, perceived utility and treatment effectiveness (Murphy et al., 1996; Proctor et al., 2011; Schubiner et al., 1994). This review defines “clinical utility” as whether the intended screening outcomes have perceived clinical value, that is whether they are helpful and accurate in identifying children at risk of MH difficulties (Gall et al., 2000; Humphrey & Wigelsworth, 2016; Schubiner et al., 1994). Psychometric properties of screening measures can thus be used to assess the accuracy or validity of measures (Humphrey & Wigelsworth, 2016).

When screening tools are inaccurate and thus lack clinical utility, they can incorrectly categorise children as at-risk when in fact they are not, an error that is known as false positives. Over-identification and risks of high false-positives in UMHS can result in wasted resources where children are unnecessarily triaged to receive intervention and/or stigmatisation of children who do not require help (Sawyer et al., 2014). Conversely, under-identification in the form of failing to identify children who are in fact at-risk, the error of false negatives, carries the risk of denying children access to treatment that may assist them. Over- and under-identification can be measured by the psychometric properties of predictive validity, including specificity and sensitivity. Reviewing measurement properties can be a time-consuming task requiring an understanding of psychometrics. This review’s extraction of psychometric properties will, therefore, provide an accessible summary of the current literature.

In evaluating the clinical utility of multi-informant UMHS, the incremental validity of parent and educator report is of particular interest in considering the unique perspectives of each informant and what they add to an understanding of child MH. Adding educators’ perspectives to parent-report has the potential to increase the validity of early identification, whilst also raising the possibility of discrepant ratings between informants (Croft et al., 2015; De Los Reyes et al., 2015). The presence of multi-informant discrepancy has been well-reported and in fact offers important, domain-relevant information specific to each informant’s context e.g., home or school (De Los Reyes et al., 2023). However, the incremental, predictive validity of multi-informant approaches relative to the use of single informant report warrants further attention, especially in relation to young, preschool-age children.

Evaluating the effectiveness of an intervention can be assessed by its ability to achieve its intended outcomes (Andrews, 1999). More specifically, in the context of UMHS, effectiveness can be defined by the improvement of early identification of child MH concerns (Brinley et al., 2024). UMHS is thus considered effective if it improves the identification or diagnosis, referral, or treatment of child MH difficulties, which can be measured by rates of accurate identification of MH difficulties or risk, uptake of referral rates to MH services, or service uptake.

Whilst there are a number of previous reviews of screening by educators for Autism Spectrum Disorder in children, including preschool aged children (e.g., DeLucia et al., 2022), a recent systematic review examining the effectiveness and cost-effectiveness of school-based identification models including UMHS has found there is little research on the identification of MH difficulties in preschool-aged children (Anderson et al., 2019). As such there is a need to examine the literature regarding UMHS effectiveness in terms of service referral and uptake following identification.

The acceptability of UMHS can impact uptake and implementation of screening in clinical and community settings (Harrison et al., 2013). It has been argued that UMHS in young children is inappropriate and ineffective, may increase parental anxiety, stigmatise children with labels, and lead to the over-medication of children (Frances, 2012; Jureidini & Raven, 2012). These arguments made against UMHS have in some instances halted the implementation of UMHS in preschool-age children previously (Alexander & Mazza, 2015). As such it is essential to investigate the perceived acceptability of UMHS amongst users if it is to be adopted and implemented widely as an early intervention strategy.

The current review is primarily interested in users’ perspectives of acceptability, that is, those of parents and educators. One element of acceptability comprises whether screening is considered appropriate, and/or perceived as helpful or useful. This may involve perceptions about whether the screening is wanted, needed or socially significant (Humphrey & Wigelsworth, 2016) or whether it is agreeable, palatable, or satisfactory (Proctor et al., 2011). Satisfaction can also encompass ‘usability’ of the tool (e.g., level of satisfaction with the length of screening, whether the language was easy to understand); whether informants felt comfortable or distressed by the screening; and whether informants would recommend screening to others, or if they would complete screening again in future (Brinley et al., 2024). Acceptability data, expressed as parental or educator attitudes towards screening, can be collected quantitatively or qualitatively. Importantly, acceptability encompasses characteristics that are associated with the likelihood of adoption of a screening measure (Glover & Albers, 2007; Kamphaus et al., 2007). This review will investigate the acceptability of UMHS broadly, and specific screening measures, as assessed by parents and educators.

No systematic review to date has evaluated multi-informant UMHS measures by parents and educators of preschool-aged children. As such, there is a need to identify what measures exist for children in this age range, examine the clinical utility and effectiveness of these measures, and also to examine the acceptability of UMHS more broadly.

The aim of this systematic review is to examine multi-informant UMHS for preschool-aged children (3–5 years) by both parents and educators. The review will answer the following questions:

The protocol for this systematic review was registered with the International prospective register of systematic reviews, PROSPERO, on 22 December, 2022 (Registration number CRD42022383426).

The aim of this review was to systematically examine research conducted on multi-informant UMHS for preschool-aged children by both parents and educators. We identified which measures have been utilised in the preschool population and examined their clinical utility in terms of predictive and incremental validity and effectiveness. The review also examined the acceptability of these measures amongst parents and educators. The recency of included studies suggests that this is an emerging area of research, conducted mainly in the United States and in wealthy, industrialised nations. Identified screening measures were highly varied and for some aspects examined, such as incremental validity, effectiveness and acceptability, there was a paucity of research making definitive conclusions difficult. However, there were 11 included studies and 10 measures identified for multi-informant UMHS in the preschool period, with promising findings for some measures suggesting that UMHS can accurately identify child MH concerns using parent and educator reports.

In order to assess the clinical utility of screening, this review examined the predictive and incremental validity of MH screening measures. Several screening measures demonstrated strong predictive validity for identifying children at-risk of MH difficulties. However, as is common for predictive validity of screening measures, several studies reported either high specificity or high sensitivity due to the tendency of sensitivity and specificity to be inversely related, and the challenge of developing a measure with maximal precision. Interestingly, the research showed that more educator-report measures had strong predictive validity (Girio-Herrera et al., 2015; Kettler et al., 2017; Stefan & Miclea, 2017) compared to parent-report (Kettler et al., 2017; Takayanagi et al., 2016). Using the benchmark of at least 70% sensitivity, 70% specificity and 50% PPV for developmental screening measures (Aylward, 1997), the ECS, IRS and PBSS met these standards for educator ratings (Girio-Herrera et al., 2015; Kettler et al., 2017; Stefan & Miclea, 2017). On the other hand, only the PBSS met this benchmark for parent-report (Kettler et al., 2017). However, results for this measure were mixed as earlier studies found that the combined, multi-informant predictive validity for the PBSS did not meet standards (Feeney-Kettler et al., 2011).

Surprisingly, the preschool version of the widely used SDQ demonstrated poor predictive validity with low sensitivity across both parent and educator reports and borderline acceptable to good specificity for parents and educators (Ezpeleta et al., 2013). Only the hyperactivity subscale had acceptable sensitivity and specificity at the recommended cut points and the study reported poor educator-report screening accuracy in terms of AUC analysis. Other comprehensive reviews of the SDQ as a screening measure amongst preschool populations have found a number of problems with its predictive validity, internal consistency of its subscales, test–retest reliability and criterion (concurrent) validity and have cautioned its use within non-clinical populations (Kersten et al., 2016; Lavigne et al., 2016). The preschool version of the SDQ may not have strong clinical utility or other psychometric properties as it has not been specifically developed for preschool samples, but further research is required to investigate this.

Only one included study reported on the incremental validity of adding educators’ ratings to parent ratings, finding a significant improvement in the identification of socio-emotional outcomes over time (Moore et al., 2022a). Contrary to previous research with older children, this study found that parent ratings did not significantly improve the prediction of outcomes. Previous findings with older children reported that parent ratings of preschool children added significant variance to child outcomes when considered in addition to later kindergarten teacher ratings (Owens et al., 2015) and a large, longitudinal study of children aged 4–11 years highlighted the power of combining informant reports to predict child outcomes across a number of domains (Verhulst et al., 1994). When evaluating the predictive and incremental validity of these measures, it is important to consider the criterion variables used in each analysis, and the context in which they reference child behaviour, such as the home, education or care setting. When criterion measures are mapped across multiple settings, it is easier to achieve the ideal model of child assessment in which child wellbeing is captured from multiple perspectives (De Los Reyes et al., 2023; Makol et al., 2020). There is promising work emerging which examines the use of screening measures with preschool children through parent, educator and child psychologists’ ratings, highlighting the value of gaining multi-faceted assessments of child wellbeing, but there remains an opportunity to research this further (Gustafsson & Sund Levander, 2024).

These results contribute to the evidence that whilst educators often have minimal MH training, their knowledge of a referent classroom group of children may be advantageous in their assessment of child wellbeing and development, given strong clinical utility for educator-report. Future research should focus on developing and evaluating UMHS measures, with high screening accuracy and clinical utility for use in preschool populations by educators.

Examining the effectiveness of UMHS adds to the weight of evidence for multi-informant screening because in the context of UMHS, improved identification is clearly linked to increased diagnosis, referral and/or treatment and thus, improved outcomes for children. Yet our review found no studies reporting on the effectiveness of multi-informant raters. This follows a consistent pattern of omission in the literature in which effectiveness data, in terms of screening outcomes or the consequences of screening, are rarely reported adequately or at all (Houri & Miller, 2020). There is a distinct lack of data reporting screening effectiveness and limited research measuring the effectiveness of UMHS in the child MH space (Brinley et al., 2024). In the case of screening amongst preschool children, none of the included studies investigated the effectiveness of multi-informant screening. This paucity of research remains despite multiple calls that we need to better understand the outcomes and consequences of screening. By effectively understanding outcomes, we can better understand the impact of UMHS (Houri & Miller, 2020). Thus, there is a clear need for future research to examine the effectiveness of screening in preschool populations with additional longitudinal studies measuring MH outcomes.

Given multi-informant report has long been the hallmark of developmental research and despite previous calls to action from highly cited research (e.g.,Achenbach et al., 1987; De Los Reyes et al., 2015), there remains a substantial gap in the study of multi-informant UMHS measures. There is a clear need for more research to demonstrate the incremental value of combining parent and educator reports in UMHS. Given demands on educators’ and parents’ time, lack of MH training for preschool educators and limited MH resources available to educators, the value of combined contributions needs to be made explicit. By demonstrating the predictive power of multi-informant reports, researchers can build the case for UMHS and improve child socio-emotional outcomes.

Critics have raised concerns that UMHS is not warranted in young children as it stigmatises them (Frances, 2012; Jureidini & Raven, 2012), and is neither wanted by parents nor educators (Humphrey & Wigelsworth, 2016). However, there is a mounting case refuting these conjectures. All three studies included in this review that investigated acceptability of UMHS reported strong support from parents and educators, for either specific screening measures or UMHS more generally in the preschool-age population. Importantly, each of these respondents evaluated the acceptability of UMHS after having completed screening for their children. A recent review of parent UMHS for children found that parents generally regarded UMHS as appropriate, across a range of contexts including schools, paediatric hospital settings, and well-child visits, although not all studies required parents to actually complete UMHS for their child (Brinley et al., 2024). The findings of the current review, therefore, add to existing research indicating that UMHS is, in fact, considered highly acceptable. Overall, the positive support from parents and educators for screening provides further evidence for the contention that preschools are ideal settings for screening, as they are accessible, less encumbered by MH stigma and a natural fit given parents frequently turn to educators for MH advice for their children (Desta et al., 2017; Pavuluri et al., 1996).

Whilst the studies in this review largely focussed on the acceptability of specific measures, there is a need to review screening more generally and in the current context. It may be possible that parents and educators have a growing awareness and understanding of early intervention for supporting child wellbeing, particularly in the aftermath of the global pandemic that had and continues to have significant ramifications on the wellbeing of children. More qualitative and quantitative research should be conducted in this area to identify perceived barriers, facilitators and perceived acceptability of multi-informant screening of preschool-aged children.

This review was the first to identify multi-informant UMHS measures utilised by parents and educators in preschool child populations and the first to evaluate their clinical utility, effectiveness and acceptability. However, there are a number of limitations that need to be kept in mind when interpreting the results. Studies that were excluded from this review included those that were not published in English, did not report predictive validity, effectiveness or acceptability data or were not peer reviewed. The exclusion criteria for child age may have limited the results and inadvertently excluded studies that included preschool-aged populations. It was often difficult to discern if 3–5 year old children were included in studies with broad-age range samples. The ages of “preschool” children were not always specified and differed between states or countries, as the age of school entry differed across jurisdictions. Secondary source or missing data were not sought from any authors to provide clarification. Moreover, some aspects of the reviewed literature were difficult to synthesise because of the disparate measures utilised. Many studies reported on author-developed measures, with varying identification methods, scoring and subscale outcomes, making synthesis difficult. Finally, risk of bias presents a possible limitation in the area of recruitment bias across the included studies. In evaluating the methodological quality of included studies using the MMAT, some studies may have been biased by not recruiting a representative sample of the target population (Hong et al., 2018). Given UMHS is intended for use amongst general, non-clinical populations in the community, it is of utmost importance that screening measures are tested in samples which are as representative as possible. The authors suggest that recruitment methodology should match implementation of UMHS in practice in order to minimise confounding variables in testing the efficacy of universal screening of whole populations. For example, UMHS measures should be implemented universally with as few participant eligibility criteria as reasonable and without interference from non-systematic factors such as teacher-nomination.

Parents and educators of preschool-aged children are integral to facilitating help-seeking for young children in need of MH support. They have a central role in early identification of MH problems and are a primary part of the pathway to enabling a child to access healthcare practitioners (De Los Reyes et al., 2015). However, based on the findings of this review, there are few measures that can be recommended to professionals in preschool and early childhood health settings for use with preschool-aged children. Based on the clinical utility of the included measures, the PBSS was the only measure to reach benchmark standards for predictive validity for both educator and parent report, however, the measure has not been published or made publicly available (Kettler et al., 2017; Owens et al., 2015). Having demonstrated incremental validity, the patented BASC-3, BESS and the freely available PSC-17 warrant further investigation to determine whether their multi-informant predictive validity are found to be acceptable. It is important to note that the widely used SDQ did not meet acceptable standards for predictive validity and as such, requires further research prior to being used for UMHS purposes in preschool children.

Based on the limited research examining incremental validity of preschool UMHS measures, it would appear that educator ratings are important and add significant incremental value to identifying children where problem behaviour occurs across multiple settings. Given how resource-intensive the implementation of multi-informant UMHS is, there is a strong need for further evidence that examines the incremental validity and effectiveness of UMHS. Future studies that investigate incremental validity need to carefully select gold-standard comparison measures which span various child contexts so that one informant is not inadvertently favoured over others (Haynes & Lench, 2003).

An important implication for future research is the examination of measures in terms of their ease of implementation and use in clinical practice. Some research in the area of measurement-based care has differentiated between the constructs of clinical utility and validity, defining “clinical utility” in terms of ease of implementation and usefulness for clinicians, and “validity” in terms of the evidence for specific measures (McLeod et al., 2022). If adopted in UMHS, this distinction may be of value as the ease of implementation and usefulness for clinicians and non-clinicians such as educators may be an important factor as to whether UMHS is more widely implemented.

Moreover, there is a need for additional studies examining UMHS, especially those which are representative and broader in scope. Large, population-level studies can provide important normative data that are inclusive of different cultural groups, children in urban and regional contexts and validated across ages. Thus, the results of this review suggest there continues to be a need to examine existing measures and develop new multi-informant UMHS measures for preschool-aged children that are psychometrically valid and reliable, accurate for identifying children at risk of MH difficulties. Such research should also enable further examination of acceptability of UMHS in preschool settings.

In examining the available UMHS measures utilised in preschool-aged children by parents and educators and by evaluating the psychometric evidence presented in studies, this review found promising results in terms of the clinical utility of measures to identify child MH concerns. However, with few measures reporting on all constructs of interest, there remains a need for further research on clinical utility, effectiveness and acceptability of measures. Accurate identification of young children at risk of MH problems is critical to providing early intervention and disrupting the trajectory of MH problems, and research to identify valid screening measures that can be easily implemented by teachers and educators is an important first step.