Difficult Toddler Temperament – Prevalence and Associated Factors at 18-Month Follow-Up of a Birth Cohort

Introduction

Temperament refers to a “behavioral style” or how an individual behaves when facing events and people.¹ Temperament presents early in life, is moderately stable throughout life, and is a result of distinctive biological mechanisms interacting with environmental influences.^2–5 Broadly speaking, temperament can be classified into three categories: “difficult”, “slow-to-warm” and “easy”.^6,7 Children with a difficult temperament are characterized as having negative mood, low regularity in routine, low adaptability and high intensity.⁸ Parental reports are often used to measure child temperament due to the unsuitability of self-report in young children and the costs and poor reliability of child observations.⁹ Studies have shown that parent-reported difficult temperament in infancy and toddlerhood is associated with greater psychopathology in childhood and adolescence, greater depression, lower wellbeing, and unemployment in adulthood.^10–15 Children with difficult temperaments often struggle to regulate their emotions and behaviors, which can lead to difficulties in forming and maintaining positive relationships with peers, teachers, and family members. Over time, these interpersonal challenges can contribute to social withdrawal, loneliness, and even social rejection, which may persist into adulthood.^10–15 Furthermore, difficult temperament in childhood is associated with an increased risk of developing internalizing and externalizing problems such as anxiety, depression, and conduct disorder. These mental health difficulties can have long-term consequences for individuals’ educational attainment, employment prospects, and overall well-being in adulthood. Therefore, understanding the current prevalence of difficult temperament and associated risk factors is important for early identification and targeted interventions.¹⁰

The prevalence of difficult temperament in children is approximately 10%.^7,16,17 However, there is a lack of current prevalence data for difficult temperament among Australian infants and toddlers. Representation of culturally and linguistically diverse (CALD) populations is also missing in this literature.¹⁶ For example, the Australian Temperament Project studies dating from the 1980s do not accurately represent the current multiculturalism that exists in the Australian population today.^18,19 Therefore, prevalence data and associated risk factors for difficult temperament in a culturally diverse and socioeconomically disadvantaged population are needed.²⁰ This information can help shape relevant early interventions to support multicultural parents who perceive their child as having a difficult temperament in order to ensure optimal developmental and behavioral outcomes.

Risk factors that have been associated with difficult temperament include male gender, children with chronic health conditions, low socioeconomic status, certain cultural/ethnic backgrounds, and marital dissatisfaction.^{17–19,21–24} Adverse parent–child relationships, maternal characteristics such as depression, poor “goodness-of-fit” or mismatch between child’s temperament and the response from the environment including the parent’s lack of sensitivity to their child’s temperament are also identified risk factors.^{17–19,22–26} The likelihood of an adverse outcome increases when there are multiple environmental and psychosocial risk factors compounding difficult temperament, particularly when parents are unable to appropriately respond to their child.²⁷ As many of the risk factors regarding temperament were identified 20 to 30 years ago, current evidence is required to understand risk factors that may be Australian affecting children today, including screen time. For example, watching television for more than 2 hours daily has been associated with higher rates of behavioural problems^28,29 particularly in the domains of withdrawn behavior, attention, externalizing behaviours and total problems, and this in turn has been shown to be a precursor for significant adverse behavioural and mental health outcomes.

To address the knowledge gap, this study’s aim was to investigate the prevalence of maternal perceptions of difficult child temperament at the 18-month follow-up of a multicultural birth cohort recruited from South West Sydney (SWS), Australia. A secondary aim was to determine significant biological and psychosocial factors associated with a difficult temperament.

Methods

Participants and Setting

The “Watch Me Grow” (WMG) study is a longitudinal birth cohort study consisting originally of 2025 parent-infant dyads with the main aim of evaluating the uptake of current developmental surveillance in a rapidly growing multicultural population in South West Sydney with significant socioeconomic disadvantage.³⁰

The recruitment, participant characteristics and representativeness of the cohort have been previously reported, and summarized in Figure 1.^31,32 A total of 2025 participants enrolled in the WMG study at birth, and the 18-month follow-up questionnaire data were available for 500 participants.

Figure 1 Initial recruitment and retention of participants.

Data Collection and Measurement of Child and Family Socio-Demographic Risk Factors

Data were collected using self-reported questionnaires which were derived and developed by the team based on existing literature and other questionnaires from other Australian cohort studies, such as the Longitudinal Study of Australian Children (LSAC) (2004 – current). Using the bio-ecological model as a framework, the impact of child, parent and family, and socio-demographic characteristics that are available in the WMG cohort data were analyzed, as outlined in Table 1.³³ A composite measure of household disadvantage was created using three variables – a household income of less than AUD 25,001, mother’s education (less than high school), and/or father being unemployed. Additional information was obtained through electronic medical records, including maternal depression scores.³⁴ Further details on the study variables have been detailed in previous publications.^30,31

Table 1 Child, Family and Socio-Demographic Characteristics Included in the Analysis

Measurement of Temperament

The questions were derived from the temperament questionnaire developed for use in the Longitudinal Study of Australian Children (LSAC) based on the 12-item Toddler Temperament Scale.³⁵ Examples are given below:

1. My child is often fearful or anxious.
2. My child is often angry or defiant.

These statements were rated on a Likert scale of 1 to 5 corresponding to “not at all how I feel” to “exactly how I feel”, respectively. A child was considered to have a “difficult” temperament if 4 or 5 was marked on any one of these questions.

Data Analysis

Descriptive statistics were used to calculate the clinical, psychosocial, and sociodemographic characteristics of the 18-month follow-up group who had questionnaire data (n = 500). Estimates of prevalence of parental perception of difficult temperament were calculated with frequency counts and percentages.

To determine the risk factors associated with difficult temperament at 18 months, univariate logistic regression was initially used to assess the associations between parental perception of difficult temperament and the individual child, family and socio-demographic characteristics and composite measures. Multivariable logistic regression was then used to identify independent predictors for parental reporting of difficult temperament. All factors significant at the p <0.25 level in the univariate logistic regression models were considered in the multivariable model. To ensure a minimum ratio of 10:1 events per variable, we limited the multivariable logistic regression model to a maximum of 3 variables.³⁶ The Bayesian information criterion (BIC) and p-values were used to determine the final multivariable model. Lower BIC values are considered better model fit whilst models with BIC values within 2 units of each other are generally considered similar to each other. All analyses were conducted using Statistical Package for Social Sciences (SPSS – version 29, IBM, Chicago). P-value <0.05 was considered significant.

Results

Participants and Characteristics

The characteristics of the 18-month group with completed questionnaires are presented in Table 2. In the 18-month follow group as a whole (n = 500), the mean age of the participants at the point of data collection was 21 months (S.D. 1.87). Just fewer than half of the participants (47.2%) were male, 9.8% were preterm, and 7.0% were born with low birth weight. Close to 10% experienced poor sleeping (9.2%) and feeding habits (8.4%). Six percent of toddlers had mothers who did not have partners, 14% had not completed high school and approximately half of mothers were born overseas. One-third reported experiencing a stressful life event within the past two years and 16% felt that they received inadequate social support at the 18-month follow-up. Almost 20% of the households were considered to be at a disadvantage at 18-month follow-up, with either a household income less than AUD 25,001 per year, a mother who did not complete high school and/or a father who was unemployed.

Table 2 Characteristics of Participants

Prevalence of Parental Concerns of Difficult Temperament as Indicated by the 18-Month Questionnaire

Of the 500 toddlers, 18 (3.7%) parents scored 4 or 5 on the statement “my child is often fearful or anxious” and 23 (4.7%) scored 4 or 5 with the statement “my child is often angry or defiant” (Table 3). Thirty-six parents (7.3%) scored 4 or 5 on either or both statements, classifying their child as having a “difficult temperament”.

Table 3 Prevalence of Difficult Temperament

Factors Associated with Parental Concerns Indicating Difficult Temperament Developmental Risk on Univariate Logistic Regression

The results of the univariate logistic regression are presented in Table 4. Three risk factors were significant, ie, more than two hours of screen time per day (OR = 2.32, 95% CI: 1.2, 4.6), the child not being read to (OR = 3.37, 95% CI: 1.6, 7.1) and experience of stressful life events within the past two years (OR = 2.05, 95% CI: 1.0, 4.1).

Table 4 Risk Factors for Reported Difficult Temperament in the 18-Month Follow-Up Group

Factors Associated with Parental Concerns Indicating Difficult Temperament Developmental Risk on Multivariable Logistic Regression

Five risk factors with p <0.25 on univariate analysis were considered in the multivariable logistic regression models: screen time >2 hours a day; child not read to; family history of mental health problems; stressful life event; and being a single mother. Based on the BIC and p-values, the final model showed that a toddler who watched more than two hours of screen time per day (AOR: 2.43, 95% CI: 1.2, 4.9); toddlers who were not read to (AOR: 3.92, 95% CI: 1.8, 8.5); and who had a family history of mental health problems (AOR: 2.69, 95% CI: 1.1, 6.5) were associated with higher odds of having difficult temperament (Table 5).

Table 5 Significant Risk Factors Associated with Development of Difficult Temperament

Discussion

This study aimed to investigate the prevalence of maternal perceptions of difficult toddler temperament at 18-month follow-up of a birth cohort recruited from a culturally diverse and socioeconomically disadvantaged community, and to identify risk factors associated with report of a difficult temperament. The prevalence of difficult temperament in this population of toddlers was 7.3%, similar to previous findings in the general population.^7,17 To our knowledge, this is the first study to show associations between toddler temperament and family and home environmental determinants such as increased screen time, not being read to, and family history of mental health problems. The results from this study are important in identifying support needs for toddlers and parents raising children with a difficult temperament to prevent mental health problems in the future.

The current study showed that toddler screen time of more than two hours a day was related to difficult temperament. This is congruent with previous research which has shown a bidirectional relationship between effortful control and screen time in young children.^37,38 There is emerging evidence that excessive screen time can limit social interactions and have adverse effects on health and developmental outcomes, particularly social and language development, and can impact social and emotional growth by interfering with the ability of children to interpret and regulate emotions and lead to aggressive conduct.^39,40 Further, while increased screen time may not have a direct impact on the development of toddler temperament, it could have indirect effects via interacting with the temperamental traits to exaggerate the functional impacts, particularly in the long run. Examples include children with inhibited temperament resorting to more screen time and social media for friendships and social connections rather than engaging in real-time face-to-face interactions. Therefore, future research is needed to determine whether there is a link between difficult temperament and screen time in toddlers or whether the relationship is compounded by sociodemographic factors indicative of psychosocial disadvantage and lack of stimulatory experiences and nurturing care. Behavioral interventions that address temperamental difficulties coupled with parenting support may be helpful for these toddlers and their families.⁴¹ Working with families to understand what factors influence excessive screen time could also help in the provision of relevant, focused interventions.⁴²

A difficult temperament was also found to be associated with children who were not read to. This finding may be explained by parents being overwhelmed with challenges that arise with children with difficult temperaments, which may result in parents having less capacity to spend time reading with their children.⁴³ Previous research has shown that not reading to children is related to multiple risk factors, including child’s low task attention, maternal psychological distress, and low maternal warmth.⁴⁴ Evidence suggests that early reading to infants with a difficult temperament can improve language and literacy skills in kindergarten.⁴⁵ Early reading can also benefit the parent–child relationship, including decreased parental stress.⁴⁶ Therefore, early identification of child and family risk factors, including difficult temperament, may help target families who might benefit from support for early reading, although it is unclear whether a causal relationship exists. Further, parent–child reading interventions which address psychosocial risk factors are recommended.⁴⁴ It may be important for practitioners to ask about the barriers to reading to better inform individualized interventions.⁴⁴ Future research is required to model the relationship between difficult temperament, psychosocial risk factors, and early reading in order to support early identification and intervention for temperamental difficulties in toddlers and their families.

A family history of mental health problems was also related to difficult temperament in this study. While previous studies have shown a relationship between parental mental health problems and child temperament, research is limited regarding family history.^43,47–50 This finding may highlight the relationship between family history of mental health challenges and specifically parental mental health problems, although this was not a focus of this study.⁵¹ Interestingly, the current study found that maternal antenatal depression was not related to difficult temperament. Familial psychopathology may be an important risk factor for difficult temperament that could be identified during the prenatal and postpartum stages. Early interventions that target maternal prenatal and postpartum depression and anxiety may be beneficial to promote positive parent–child interactions and also maternal perceptions of child’s temperament.^52,53 However, if other members of the family have this history, it may be more difficult to identify and intervene. Future research is needed to examine the relationship between toddler temperament, familial psychopathology and parental mental health problems.

In the current study, stressful life events (eg, parental divorce) were related to difficult temperament, a finding that aligns with other research showing a link between stressful life events and emotional and behavioral difficulties in young children.⁵⁴ Positive temperament characteristics have been shown to act as a protective factor for the child to cope with “stressful” parent relationships.⁵⁵ This study may indicate that the converse is also true, whereby children with a difficult temperament may be more likely to act out during a stressful event. It is recommended that clinicians be aware of stressful life events that may be impacting the families they support, and appropriate support and intervention be made available, such as parenting resources.^54,56 Culturally diverse families may be at higher risk for stressful life events and should be given particular attention.⁵⁷

Understanding and responding to a toddler’s needs may be more difficult for parents who have a negative perception of their child’s temperament.⁴³ In this regard, research has shown that parental perceptions of difficult temperament are related to greater parenting stress, lower mindful parenting, lower parenting sense of competence, and lower parenting efficacy.^43,58 Providing parents with education and support for difficult temperament may help to increase parenting efficacy.^59,60 Mindfulness-based interventions for parents are recommended to improve the parent–child bond and perceptions of child temperament.^61–63 Parent–child interaction interventions which focus on emotion regulation may also be important in reducing future psychopathology in children with difficult temperaments.^10,64 Fearful temperament in toddlerhood is related to conflictual co-parenting; therefore, early interventions that address co-parenting conflict may be helpful.⁶⁵ Increasing availability of interventions and improving pathways of care may be particularly important for culturally diverse and socioeconomically disadvantaged families who are less likely and less able to access services.^66,67

Limitations

This study has several limitations. A potential limitation is the validity of the two statements “My child is often fearful or anxious” and “My child is often angry or defiant” as a substitute for formal measurement of temperament type. However, parental temperament rating can be a good proxy despite being subjective due to their own understanding and perceptions of “normal” child behavior.⁹ Given that the statements in the present study were self-reported, there is possibility that the responses may be biased due to social desirability. Additionally, the self-report nature may result in single rater bias. In terms of the longitudinal nature of data and the long-term follow-up, there was a loss of follow-up with only 500 participants at 18 months. Whilst having a sample size of 500 participants, another limitation is the small proportion of children with a difficult temperament (7.3%) which severely limits the power to detect smaller effects. Future research using more robust parental ratings and objective measures is needed to replicate current findings.

Conclusion

Research into difficult temperament in infants and toddlers and ways to optimize their developmental and behavioral trajectories deserve attention. Our findings suggest that excessive screen time, children not engaging in reading and stressful life events can interact with temperamental traits and this is consistent with previous research that has shown that these factors can adversely impact the behavioural trajectory of children. In this regard, our work has shown that excessive screen time of more than 2 hours a day in children as young as 18 months is associated with less time spent by parents in reading to the child and that the excess screen time was more pronounced among those from disadvantaged background characterized by mothers without a partner, not having outdoor equipment at home and fewer than five outings per week.³⁹ This observation taken together with the findings by Cheng et al⁶⁸ that daily and excessive screen time can impact reward sensitivity and the development of inhibitory control would suggest the critical role of these factors in the longitudinal health and mental health outcomes. These authors also suggested that screen time could also impact future habitual seeking behavior due to activation of the dorsal striatum connectivity. This has implications for service providers in including screen time as a critical factor to be included in the developmental assessment of children. Further, in keeping with the Nurturing Care Framework of the World Health Organization (WHO),⁶⁹ it is important for policy makers to plan targeted supports to enhance play and stimulation, particularly for those children from disadvantaged backgrounds without easy access to green space and outdoor leisure and play activities in line with the nurturing care agenda of the WHO.