Introduction

Trauma during early childhood was once a very neglected topic for research, clinical practice, and policy [1]. Fortunately, due to the pioneering work by Scheeringa and Zeanah [2, 3], our recognition and knowledge of this important and challenging issue has grown exponentially over the past 20 years. This is critical given the fact that exposure to trauma occurs at the highest rates during early childhood and that young children have a unique set of emotional and physical vulnerabilities that place them at even greater risk of poor outcomes [1, 4••, 5••]. Some of these vulnerabilities include their complete dependence on adults to keep them physically safe and secure, their limited emotional and behavioral self-regulation skills, their limited cognitive and communication skills, the important influence a young child’s immediate relationships has on their mental health and the rapid rate of socio-emotional, cognitive, and neurobiological development that occurs during this stage of life [1, 4••].

We now know that (1) young children also present with many of the same trauma symptoms as older children, adolescents and adults, but due to developmental differences, distress can manifest differently in this young age group; (2) when developmentally modified PTSD assessment tools and diagnostic criteria are used, young children are diagnosed with PTSD at similar rates to older children and adults; (3) there is a high rate of comorbidity that is associated with PTSD in young children; (4) the majority of young children are resilient and/or return to normal levels of functioning following a trauma, but for approximately 10–20%, PTSD can follow a chronic and debilitating symptom trajectory; and (5) the impact of trauma during early childhood must be considered within the context of the parent-child relationship [1, 4••, 5••].

Major advances in this field were made in 2013 when the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) included the first developmental subtype of an existing disorder, PTSD for children 6 years and under [6]. The inclusion of these criteria has important implications for research and clinical practice, therefore the aims of this paper are to (1) review the history of the diagnosis of PTSD in young children and the rationale for the inclusion of the preschool subtype of PTSD in the DSM-5, (2) present and discuss the latest evidence and advances in the validation of the preschool subtype of PTSD, (3) discuss limitations of the current diagnostic algorithm for PTSD in young children, and (4) highlight areas for future research. Our use of the term, young children, throughout this article refers to infants, toddlers, and preschoolers (i.e., < 6 years of age).

History of the Diagnosis of PTSD in Young Children

Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition

The DSM is the most widely used classification system for mental health disorders across the life span [6]. However, one of the criticisms of this classification system is that the diagnostic criteria have mostly been developed based on how psychiatry disorders typically present in adults and clinically and empirically validated and refined using mostly adult populations [7, 8]. Following the recognition that young children can develop PTSD, concerns were raised regarding the validity and suitability of the DSM-IV-TR PTSD diagnostic criteria for this age group [DSM-IV-TR; 8, 9]. When the DSM-IV nosology for PTSD was published in 1994, only minimal modifications were made to accommodate the unique developmental differences in symptom manifestation in children and no children under the age of 15 were included in DSM-IV field trials [10]. The problems with the DSM-IV PTSD criteria for children were further highlighted when research by Scheeringa and colleagues clearly showed that the criteria did not adequately capture the symptom manifestation experienced by infants and preschoolers and underestimated the number of children experiencing posttraumatic distress and impairment [2].

These findings led to the proposal of an alternative PTSD algorithm for young children (PTSD-AA; [2]). The wording of some of the DSM-IV PTSD symptoms were modified to make them more objective, behaviorally anchored and developmentally sensitive to young children. In addition, symptoms that were rarely observed in this age group were deleted and the symptom threshold required for the avoidance/numbing symptom cluster was reduced from 3 to 1. The PTSD-AA was refined based on empirical findings over several years [3, 11] and critiqued and endorsed by a task force of experts on early childhood mental health [12].

DSM-5—Posttraumatic Stress Disorder for Children 6 Years and Younger

In response to concerns raised about the lack of sensitivity of the DSM-IV criteria for young children, the growing empirical validation of the PTSD-AA, and recommendations by leading experts [13], the age-related subtype of PTSD for preschool children, “posttraumatic stress disorder for children 6 years and younger” was proposed and published in the DSM-5 [6]. This is the first developmental subtype of an existing disorder and represents a significant step in the right direction for the DSM taxonomy. Throughout the rest of this review we will use PTSD<6Y when referring to this subtype. Refer to Table 1 for a summary of how the DSM-5 PTSD criteria compares to the DSM-5 PTSD<6Y.

Table 1 Summary and comparison of the similarities and differences between the DSM-5 PTSD, DSM-5 PTSD<6Y, DC:0-5 and ICD-11 diagnostic algorithms
Full size table

PTSD now sits within the ‘Trauma and Stress-Related Disorders’ chapter in the DSM-5 metastructure. Based on studies investigating the latent structure of PTSD, the DSM-IV 3-factor structure was changed to a four-factor structure in the DSM-5 [14]. The symptoms in PTSD<6Y are presented within the same four symptom clusters (Intrusion symptoms, Avoidance symptoms, Negative alternations in cognitions, Alterations in arousal, and reactivity) as the adult criteria. However, as the DSM-5 was greatly influenced by research on the PTSD-AA it specifically sought to eliminate adult symptoms that relied on subjective report and abstractions of personal experience. Important developmental modifications that were made to the DSM-IV PTSD criteria for the PTSD<6Y are discussed in more detail below:

Criterion A: Exposure to a Traumatic Event

The definition of criterion A recognizes that there are some key developmental differences for what makes an event potentially traumatic for a young child. Specifically, attachment relationships are particularly important during early childhood. Therefore, criterion A can be met if the child witnesses or learns that the traumatic event occurred to a parent or caregiving figure.

Criteria B: Intrusion Symptoms

The changes to this cluster are minimal with modifications made in wording to increase face validity and, thereby, improve symptom detection. The old B1 symptom required intrusive memories to be distressing; however, based on research that has found young children do not always manifest distress and may sometimes show no affect or appear “over bright” [2, 3], the DSM-5 now states that this symptom may not appear to be distressing for the young child and that it can be expressed through play re-enactment. It is also noted that it might not be possible to determine the content of frightening dreams (B2) and dissociative reactions (B3) might also manifest in play.

Criteria C: Avoidance of Stimuli OR Negative Alterations in Cognitions

The biggest differences between the DSM-5 PTSD vs. PTSD<6Y diagnostic criteria are in this section. Many of the symptoms in these clusters are highly internalized phenomena that require abstract cognitive capacities (e.g., self-blame) thus making it much more difficult to observe accurately in this age group [14]. The symptoms of “sense of a foreshortened future” and “inability to recall an important aspect of the event” were deleted all together because they are not developmentally appropriate. Therefore, because there are fewer PTSD symptoms to observe in young children, the major change was to require only 1 symptom in either the avoidance symptoms OR negative alterations in cognitions clusters, instead of the threshold of ≥ 1 symptoms for cluster C and ≥ 2 cluster D that is required for older children and adults. The wording of two symptoms from the avoidance cluster and two from the negative alterations in cognitions cluster were modified to enhance face validity and symptom detection. Specifically, the PTSD-6Y avoidance symptoms differ in wording to the adult symptoms given that it is very difficult to identify avoidance of memories and thoughts in this age group. Instead, symptoms in this cluster can be observed as avoidance of (C1) external reminders such as places, activities and physical reminders and (C2) people, conversations, and interpersonal situation reminders. Diminished interest in significant activities (C4) may be observed as constriction in play and feelings of detachment or estrangement is now described as socially withdrawn behavior (C6). The additions in the DSM-5 criteria that were not examined in the studies on the PTSD-AA criteria included changes to the wording of the restricted range of affect symptoms to limit it to only positive emotions (as opposed to positive or negative emotions), and the addition of a new symptom (i.e., “Increased frequency of negative emotional states”).

Criteria D: Alterations in Arousal and Reactivity

Given that many of these symptoms are more behavioral and easily observable, minimal developmental modifications were needed for this cluster. The D1 symptom “irritable behavior and angry outbursts of anger” was modified to include “extreme temper tantrums” to reflect how this symptom typically presents in young children.

Criteria F: Functional Impairment

Functional impairment is met if the symptoms cause clinically significant distress or impairment in relationships with parents, siblings, peers, or other caregivers or with school behavior.

International Classification of Diseases

The World Health Organization’s International Classification of Diseases (ICD) is an international classification standard for reporting diseases, injuries, and health conditions for all clinical and research purposes. The 11th edition was published in June 2018 [15]. The ICD-11 PTSD criteria differ greatly from the DSM-5 criteria as the PTSD diagnosis has been simplified to the following three symptom clusters: (1) re-experiencing the event, (2) avoidance of activities, situations, and people, and (3) persistent perceptions of heightened current threat. The “non-specific” symptoms that are found across other mood and anxiety disorders were removed (e.g., sleep disturbance), leaving a total of only 6 PTSD symptoms which is a significant difference from the 20 symptoms in the DSM-5 (Table 1). Although age-specificity of symptoms is mentioned in the explanatory text of the ICD-11, there are no specific PTSD criteria for preschool-aged children or modifications made to symptom wording to reflect developmental differences across the lifespan. Vasileva and colleagues [16••] have recently reported on the first empirical examination of the proposed ICD-11 algorithm for PTSD compared with the PTSD<6Y in a clinical sample of children aged 1–6 years. They found only a moderate agreement between the two classification systems, with the ICD-11 classifying significantly fewer children with PTSD than the DSM-5 algorithm. The PTSD<6Y was also found to be significantly better at predicting functional impairment than the ICD-11 criteria [16••]. These preliminary findings suggest that, unfortunately, the ICD-11 criteria might not be developmentally sensitive enough to capture PTSD in young children.

Diagnostic Classification of Mental Health and Developmental Disorders of Infancy and Early Childhood

To address the crucial need for developmentally appropriate nosology, the clinically based Diagnostic Classification of Mental Health and Developmental Disorders of Infancy and Early Childhood (DC:0–3, 1994) was the first systematic effort to establish a categorical classification system for disorders of infancy. The overarching aim was to provide a shared language to facilitate understanding of definitions and manifestations of psychopathology and to aid research and clinical practice. Drawing on empirical research and clinical experiences, the DC:0–3 was revised in 2005 (DC:0–3R) and again in 2016, to capture new findings relevant to the diagnosis of psychiatric disorders in children aged 0–5 years (DC:0-5) [17].

Where possible, the disorders have been aligned with the DSM-5 definitions with the long-term goal of having early childhood disorders fully integrated into mainstream nosologies [18]. Therefore, not surprisingly, the PTSD diagnostic algorithm in the DC:0-5 is very similar to the PTSD<6Y (Table 1). The DC:0-5 algorithm has the same four symptom clusters, comparable symptom wording and requires the same thresholds to be met for a PTSD diagnosis to be made. The main differences are that in the DC:0-5, there are more details on how PTSD typically manifests in this age group, more direction around the types of events that may be traumatic for young children, and a more comprehensive and accurate definition of functional impairment. Moreover, there is an additional symptom listed under the re-experiencing cluster (i.e., preoccupation with the traumatic event that can be conveyed by repeated statements or questions) whereas the avoidance cluster is summarized as one symptom rather than two separate symptoms as in the PTSD<6Y. Of note, the DC:0-5 states that a diagnosis should be made with caution in infants less than 12 months as even though they may become symptomatic following a trauma, due to limited representational capacities, they are even less likely to present with all the symptoms of PTSD as defined in the diagnostic criteria.

Validation of PTSD Criteria for Young Children

The PTSD-AA has demonstrated adequate convergent and discriminant validity, and predictive validity (see [19••] for a review). Additionally, over 26 studies across a number of trauma types (i.e., accidental injury, terrorism, natural disaster, war, interpersonal violence, sexual and physical abuse) and different countries have used this criterion and concluded that the PTSD-AA is more developmentally sensitive than the DSM-IV algorithm as it identifies higher rates of PTSD in young children (refer to Table 2).

Table 2 Summary of studies that have examined PTSD prevalence rates in young children between the ages of 0–6 years
Full size table

Following the proposal of the PTSD<6Y by the DSM-5 working group, and prior to its publication in 2013, two separate prospective studies used diagnostic interviews to thoroughly examine the preschool subtype’s diagnostic validity in comparison to the PTSD-AA, DSM-IV and DSM-5 versions [24••, 41••]. Both studies concluded that the PTSD<6Y provided the most developmentally sensitive and valid measure of PTSD in young children (given the similarity, few differences were found between the PTSD-AA and PTSD<6Y). The PTSD<6Y diagnosed PTSD at a higher rate than the DSM-IV and DSM-5 algorithms. Further analysis in both studies revealed that the children who were “misclassified” as not having PTSD were still highly symptomatic and impaired (i.e., still had a mean of 6.4–7.4 symptoms, and two or more domains of impairment), indicating that the lower thresholds set for the PTSD<6Y were not simply diagnosing more mildly symptomatic children. Both studies clearly showed that it was necessary to lower the avoidance threshold from 3 to 1 as the majority of the symptoms in this cluster were endorsed at a very low frequency, especially in comparison to the other symptoms. The deletion of Criterion A2 was also supported as emotional reactions during the time of an event are very difficult to identify in this age group and also because it did not demonstrate good predictive utility. Evidence of convergent validity was demonstrated with significant associations found between a positive PTSD<6Y diagnosis and a high level of comorbidity with other psychiatric disorders, particularly separation anxiety and oppositional defiant disorder [41••, 43]. Finally, the PTSD<6Y demonstrated predictive validity, accounting for a significant amount of variance in emotional and behavioral difficulties and functional impairment 6-months later [24••].

Limitations of the Current Diagnostic Classification Approaches

The inclusion of the PTSD<6Y subtype in the DSM-5 represents enormous progress for the field of infant and preschool mental health. However, there are limitations that need to be considered and improved upon. The DSM-5 has been designed to be a living document so that it can be revised more frequently as relevant scientific findings emerge [14]. Therefore, with the aim of advancing the developmental validity of this diagnosis further over time, we will now discuss some of the limitations and areas that we believe are still needing more clinical and research attention.

Criterion A

Since the PTSD diagnosis was first published in the DSM there has been considerable and ongoing debate over the definition and interpretation of criterion A [44, 45]. A major challenge for the DSM is finding the right balance between a broad versus narrow definition of trauma to ensure that researchers and clinicians are able to accurately distinguish “traumatic” from “non-traumatic” events. In the DSM-5, criterion A is defined as exposure to actual or threatened death, serious injury, or sexual violence. The term “threatened” is included to acknowledge the important role the appraisal process plays during a trauma. The Diagnostic Features section lists potential criterion A events as threatened or actual physical or sexual violence (for children this may include developmentally inappropriate sexual experiences), kidnapping, torture, war, disasters, severe motor vehicle accidents and medical events that involve sudden, catastrophic events (e.g., waking during surgery, anaphylactic shock), and witnessing or learning about events that involved threatened or serious injury, death, or abuse to others (especially if to a primary caregiver).

Some concerns regarding criterion A have been raised because of an ambiguity around how to accurately determine if a child meets the “serious” injury threshold and the requirement that medical events involve “sudden, catastrophic” events. Research has shown that similar to adults, young children develop PTSD following a range of injury types (e.g., burns, dog attacks, falls), cancer, and from painful and frightening medical procedures (Table 2). However, medical-related events that have been associated with PTSD in young children are not always life-threatening and often do not meet objective medical definitions for being “serious”. One study that examined criterion A in a sample of young burned children found PTSD rates varied greatly depending on how a “serious” injury was defined [24••]. Research has also consistently shown that injury severity is not a good predictor of PTSD [46]. One reason for this is because the subsequent medical procedures and hospital experiences are often reported to be more traumatic than the injury itself [29, 47••]. Finally, given that children in this age group often refer to their parents to determine if something is dangerous or threatening and rely on them to help them feel safe and secure, a parent’s reaction during a traumatic event may significantly influence a child’s experience of that event (e.g., a parent visibly distressed and not able to support their child during a painful medical procedure) [21, 48].

However, the fact that DSM-5 includes the term “threatened” in the definition of a traumatic event leaves criterion A more open to subjective appraisal and provides an opportunity to overcome the above-mentioned concerns. This is of great importance for children because it allows researchers and clinicians to consider events where there was no actual likelihood of death, serious injury, or sexual violence but where it was reasonable during the event for the child to experience this event as traumatic. This is especially important for young children who do not always fully understand what is happening to them and may be more likely than adolescents or adults to appraise certain events as threatening and traumatic, even though they are not objectively potentially harmful. The fact that the current definition of criterion A in the DSM-5 offers the possibility to include subjectively threatening events and therefore to appropriately consider developmental aspects in experiencing an event as traumatic is something many clinicians and researchers are still not aware of.

We still have much to learn about how young children appraise threat, what types of events typically lead to the development of PTSD in young children and what the boundaries should be for meeting the threshold of stressor severity (especially for events other than interpersonal trauma and war). However, given that injury and medical events occur at a high frequency during early childhood and that there are many factors that influence how a young child comprehends and experiences threat during these events, we argue that it is important for future revisions to the DSM-5 to consider including more examples and clearer descriptions of the types of events that could meet the “traumatic” threshold for infants, toddlers, and preschoolers. We also think it could be important to highlight how a caregiver’s reaction at the time of the event may buffer or exacerbate their child’s subjective experience of that event.

Factor Structure

Another topic that has always received considerable controversy and debate is around the latent dimensionality of PTSD symptoms. To remain consistent with the adult criteria, the PTSD<6Y has the same four symptom clusters as the adult diagnosis; however, only four symptoms are required to be met from three of those clusters. There have been mixed findings in the literature regarding whether 1-, 2-, 3-, 4-, 5-, 6-, 7-factor or 4-factor dysphoria models provide the best fit for school-aged children and adults [49, 50]. As yet, there are no published studies that have examined the factor structure of PTSD symptoms in young children. Given the developmental differences in symptom manifestation that have been identified in this age group it is quite likely that the DSM-5 model does not provide an adequate fit of symptom clusters in young children. We also assume that in infants (i.e., children below the age of 12 months), the symptom structure is likely to be different compared to toddlers and preschoolers. This is because very little research has specifically investigated if and how trauma symptoms typically manifest during the first year of life. Due to very limited cognitive, motor, and language skills, it is unlikely that infants can actually experience or show many of the PTSD symptoms currently described in the PTSD>6Y. This calls for further differentiation of the PTSD<6Y criteria in future editions/revisions of the DSM-5.

Symptom Presentation

Whilst significant improvements have been made with deleting or modifying symptom wording to focus on behavioral manifestations, there are still several symptoms that are very hard to accurately identify, especially during the first 3 years when children’s language and cognitive abilities are still considerably limited. For example, 18-month-old children typically only have a variety of single words and by 24 months are still only speaking in two to four word sentences [51]. Pretend play is very simple before the age of 2 and starts to become more sophisticated with the emergence of symbolic play (e.g., using a block to represent a car) between 2 and 3 years of age [51]. Parents are typically poor at identifying internalizing symptoms across all age groups, so it is therefore not surprising that the symptoms that require words and/or sophisticated play (e.g., intrusive memories), or close observation to be noticed (e.g., racing heart, sweating, hypervigilance, loss of concentration, diminished interest, and withdrawal) have been endorsed at the lowest frequencies for each of the symptom clusters [24••, 39••, 41••]. Although, interestingly, intrusive recollections were reported more frequently in two studies where the ages ranged from 3 to 6 years [39••, 41••] compared to in a sample where 80% of the children were aged 1–3 years [24••]. Levendosky et al. [52] identified the problem with aggregating data from children across a variety of ages and specifically examined how age influences the types of symptoms and clusters that are endorsed across each year from the age of 1 through 7. Overall, consistent with what would be expected based on developmental milestones, arousal symptoms were the most frequently endorsed symptoms across all ages groups, re-experiencing symptoms peaked at age 4 and avoidance symptoms increased over the first 3 years before stabilizing.

Two of the new symptoms that were included in the DSM-5, increased negative mood states (e.g., guilt) and reduction in positive emotions are yet to be thoroughly examined in this population. Additionally, given the huge variation in what is considered “normal” behavior during early childhood, there is a fine line between pathologizing common and normal symptoms that typically emerge at certain ages (e.g., nightmares, temper tantrums) versus mistakenly assuming that these symptoms are due to, for example, the “terrible twos” rather than potential trauma signs that need to be monitored and assessed. Given the complexities of accurately detecting the presence of PTSD symptoms in this age group, it may be helpful for the DSM-5 in future revisions to give more guidance around what is normal vs atypical development during early childhood.

Functional Impairment

This criterion is not well-defined in the DSM-5 as it provides limited information on how traumatic stress reactions typically impact this age group. The DC:0-5 provides more guidance on how PTSD symptoms, and importantly how caregiver accommodations in response to such symptoms, significantly affect the child’s and family’s functioning (e.g., causes distress to the child, interferes with relationships, limits participation in activities and routines, interferes with normal developmental processes). A clear description of impairment is particularly important for this age group to show how the impact of trauma needs to be considered within the context of the child’s environment, stage of development and caregiving relationships (e.g., functional impairment would look very different in a 5-year old who interacts with a number of peers in different social situations, attends school, and can do things independently in comparison to an 18-month old who is not yet in a child-care setting, has limited opportunities for social interactions outside the family and needs help with most daily activities). Further, clear definitions will help reduce the likelihood of pathologizing transient behavior anomalies and expected individual differences that are particularly prevalent during early childhood [18].

Key Associated Features and Comorbidity

While there has been limited research specifically examining key associated features and comorbidity with preschool PTSD, studies have found that children commonly experience symptoms of increased separation anxiety, oppositional behavior, new onset of fears, more aggressive behavior, and regression in previously acquired skills [11, 43, 52]. It may be that some of these symptoms are reflecting ways in which trauma symptoms actually manifest in this age group, especially in the very young children. Three studies have reported particularly high rates of comorbidity with separation anxiety disorder (SAD) and oppositional defiant disorder (ODD) [11, 28, 43]. Additionally, it was found that all children who received a new-onset non-PTSD diagnosis were also experiencing PTSD symptoms [28, 43]. It is possible these findings are indicating that a PTSD syndrome may be the core post-trauma response [4••]. The high rates of comorbidity may also suggest that our current nosologies may be identifying syndromes that are, in fact, not distinct disorders but varied presentations of underlying syndromes that cannot be characterized adequately by our current descriptive criteria [8]. Transdiagnostic concepts such as the Research Domain Criteria (RDoC) try to address such concerns [53]. Concerns have also been raised about the potential for misdiagnosis in this age group. Specifically, children who present with high emotion and deregulated behavior following a trauma may receive erroneous primary diagnoses such as ADHD or ODD, especially given the difficulties with accurately identifying internalized PTSD symptoms in young children [4••]. Therefore, there is the risk that treatment is targeted towards the easily observable behavioral symptoms without understanding the concurrent underlying PTSD symptomatology. The high frequency of these co-occurring symptoms has important implications for assessment, diagnosis and treatment and so it is essential that researchers and clinicians are aware of these issues when working with young children. Importantly, the DSM-5 has noted that developmental regression may occur in young children (in the Associated Features Supporting Diagnosis section) and that the patterns of comorbidity are different for young children compared to adults, with ODD and SAD predominating. The DC:0-5 also notes that new onset of fears, fearfulness, and angry or aggressive behavior are common associated features seen in this age group. As our field progresses with understanding these complex issues, it will be important for the DSM-5 to be updated accordingly.

Recommendations for Future Research

While the field has made considerable progress there are some limitations with the existing research that are worth noting. The majority of studies have had small sample sizes, differing age ranges (e.g., 1–4 vs. 3–6 years) and utilized cross-sectional or retrospective designs. Due to the different types of trauma selected, there is a great variation in assessment time frames within and across studies. The number and expression of trauma symptoms may vary greatly depending on the child’s age at exposure to the traumatic event and the current age at assessment, how many months post-trauma the assessment occurs and what type of trauma it was (e.g., single event vs multiple, accidental vs. interpersonal). The majority of studies have also only reported on how the prevalence rates of the PTSD-AA compared with the DSM-IV and have not looked at important diagnostic issues related specifically to symptom manifestation, dimensionality of symptom presentations, associated features, development and course, risk and prognostic factors, ethnicity and culture and comorbidity. There is still only a very limited number of studies that examined the validity of the PTSD<6Y criteria which in part might be due to the lack of published and validated assessment measures. Further prospective research using validated diagnostic interviews are therefore needed to:

  1. 1.

    Accurately define what constitutes a traumatic event for infants, toddlers, and preschoolers (as this may differ across the developmental stages).

  2. 2.

    Thoroughly examine the nature, frequency, and trajectories of the DSM-5 PTSD symptom presentations across different ages (<12 months, 1–2 years vs 3–6 years), trauma types, and ethnic groups.

  3. 3.

    Use network analytic methods to explore the dimensionality and structure of PTSD in young children. This approach can be used to help identify which symptoms are most central to the PTSD network, look for direct associations among symptoms, and determine symptoms that are essential for a diagnosis [54, 55••]. It would also be very interesting to use network models to examine comorbidity to address concerns that have been raised about syndromal indistinctiveness [54]. Findings from this type of analysis could provide critical information to inform directions for future diagnostic systems as well as have important implications for assessment and treatment.

  4. 4.

    Thoroughly examine the reliability and convergent, discriminant, criterion, and predictive validity of PTSD<6Y for the different key age groups (e.g., < 1, 1–2, and 3–6 years), range of trauma types and ethnic/racial groups.

  5. 5.

    Accurately define what constitutes functional impairment for infants, toddlers, and preschoolers (as this may differ across the developmental stages).

  6. 6.

    Develop an improved understanding of the relationship between PTSD, associated symptoms and comorbidity.

  7. 7.

    Identify symptoms which best reflect posttraumatic distress vs more global emotional and behavioral distress.

  8. 8.

    Identify the key psychological, physiological, and environmental risk and protective factors.

  9. 9.

    Further examine the association between parent-child relationship factors and psychopathology following trauma and determine the direction of effects.

Conclusions

It is nowadays no longer debatable that young children present with traumatic stress reactions following exposure to a traumatic event. However, while young children present with many of the same trauma symptoms as older children and adults, we know they can manifest differently depending on their age at the time of trauma. It is also much more challenging to accurately assess and diagnose PTSD in children under the age of 6 due to factors related to their stage of development. The DSM diagnostic system has been heavily criticized by infant and child mental health specialists because minimal consideration has been given to incorporating developmental differences in the presentation of psychiatric disorders [8, 18]. Therefore, the inclusion of the empirically supported PTSD<6Y subtype represents a significant step in the right direction for having a developmentally sensitive diagnostic classification system for mental health disorders across the life span. The formal recognition that PTSD presents differently in young children helps raise awareness that this is a serious problem that warrants further attention. Having established criteria enables effective and efficient communication among researchers and clinicians [18]. Further, it helps facilitate more rigorous research into exploring how different developmental stages influence posttraumatic symptom manifestations, and identifying risk factors and mechanisms for the disorder. It will also encourage assessment measures to be developed and validated as well as provide direction for intervention and allow intervention effectiveness to be determined [18].

Considerable progress has already been made in many of these areas. However, although there is growing clinical and empirical support for the preschool PTSD algorithm in the DSM-5 and DC:0-5, as discussed in this review, further work is still needed to modify and validate the criteria further. This is now an opportune time for scientific research to be conducted to continue to refine and improve the PTSD diagnostic criteria. While there are many additional challenges with working with this population, there is the potential to dramatically improve the mental health and well-being of young children and thereby minimize the long-lasting impact trauma can have on children’s emotional, behavioral, social and biological development.