Maternal well-being and its association to risk of developmental problems in children at school entry

In the early years, unique opportunities exist to influence children's development trajectories and their families [1–4]. Birth to five years is a critical time in development when language, cognitive, emotional, social, behavioural and physical skills become the basis for the scaffolding of new skills throughout further experiences and education into adult life [4–6]. Childhood development has a demonstrable influence on long-term outcomes, affecting academic performance, health, coping skills and successes throughout the life course [2, 5, 6]. Evidence indicates that children at highest risk of developmental problems benefit from early identification and intervention [1–4]. Early interventions not only enhance the well-being of children and the resources of families but also benefit society by preventing and minimizing developmental problems and their costly consequences [1, 3, 7–11].

With an estimated 780 million young children around the world affected by an intellectual disability, the impact of developmental problems in society is widespread [4]. Understanding who is at risk in our community, and who would benefit from early intervention, allows for optimal allocation of limited resources. Investigating factors affecting child development at the time of school entry may reveal opportunities to tailor early intervention programs for the greatest effectiveness, social benefit and economic gain.

It is well-known that sociodemographic characteristics of the family, such as low income and less maternal education, put children at higher risk for poor outcomes [10, 12–14]. Recent evidence also suggests that, neighbourhood factors, such as neighbourhood income and cohesion, may also impact child development [15]. Males are at increased risk for developmental problems [14], and exposure to substances in utero impacts birth outcomes and child development [16, 17].

Maternal well-being and parenting have also been associated with child development. Harsh or inconsistent parenting approaches as well as maternal stress and anxiety have been associated with child behavioural outcomes [9, 18–21]. A maternal history of abuse may also be associated with infant temperament and child behaviour [22, 23]. Post-partum depression has been associated with poor cognitive and emotional development in infants, particularly among males, and these effects may persist into early childhood [14, 19, 24–28]. Predictable, appropriate responses to infants from adults are critical to optimal child development, and poor maternal mental health, in particular, can interfere with the creation of early infant-maternal social interactions that are stimulating and supportive [18–20]. Furthermore, parental models of attachment, family dysfunction, and family structure (e.g. lone parenting) have also been associated with parent-child interactions and behaviour problems in children [9, 12, 13, 29]. Lack of maternal social support may also adversely impact child cognitive and behavioural development, most notably when mothers are dealing with depression or have a history of abuse [22, 30, 31]. The relationships among these maternal and family factors are complex as some factors may mediate the impact of others [22, 29]. Furthermore, risk for delay and poor outcomes are cumulative such that children exposed to a greater number of risks have poorer outcomes than children exposed to less risk [13].

The study described in this paper follows up a cohort of mothers and children that have been studied on two previous occasions--during the perinatal period and when children were three years of age. Based in Calgary, an urban centre of over 1 million people in Alberta, Canada, the original Community Perinatal Care Study was a randomized controlled trial (RCT) of three types of prenatal care [32]. The results showed that additional prenatal support from nurses and home visitors could increase the use of community based resources and the access of pregnancy related information. The additional support, however, did not translate into changes in alcohol and tobacco use, post-partum depression or birth outcomes [32].

When the children in the Community Perinatal Care Study were followed up in a study at three years of age, the participants constituted a demographically low risk sample with regards to maternal education and income [14]. Of note, 73% of all Canadian families with kids under six years of age have a household income greater than or equal to $40,000 [33]. The follow-up study at three years found that 11% of children screened at high risk of developmental problems [14]. Type of prenatal care received in the RCT did not have an impact on the child's risk for developmental problems at three years of age [14]. However, a male child who had a history of ear infections and was living in a low income environment with a mother with poor mental health (defined as a history of postpartum depression, abuse, and/or lack of contentment during pregnancy), had a predicted probability of 53% of having developmental problems at three years of age [14]. With all other variables held constant, this risk was reduced to 19% if the mother reported good mental health during the prenatal and early postpartum period, suggesting a meaningful potential benefit of interventions designed to address maternal mental health and well being [14].

The current study was conducted to answer questions about the health and development of a community based sample of children as they became of school age. The objectives of this study were to:

For the purposes of this paper, developmental problems refer to a child's difficulties affecting development in any one, any combination, or all (i.e., global dysfunction) of the following domains: cognitive, social, language, motor, academics.

This study indicates a direct relationship between the well being of mothers and the development of their children at school entry [14]. Maternal history of abuse reported during the prenatal period and past parenting morale were independent predictors of high risk for childhood developmental problems at school entry. Mothers of children at high risk for developmental problems were also more likely to have lower scores in measures of current mental health and parenting morale, as well as current and past social support.

This study and the previous follow-up study at three years suggests that the direct relationship between the well-being of mothers and the development of their children begins in the early years and persists through to school entry [14]. These findings suggests that children may not 'rebound' from early threats to development that occur before age 3 simply through maturity, independence or through increased interactions outside the home. This work highlights the critical importance of maternal well being and parenting morale in child development, and emphasizes that these can be unrelated to economic security and maternal education.

In particular, children of mothers who reported a history of abuse while pregnant were more than twice as likely to be at risk for developmental problems at three years and at five years [14]. These findings suggest that identifying women with a history of abuse may be one way to identify women whose children are at higher risk of developmental problems. Based on the findings from this study, identification of these women could take place as early as the prenatal period. However, further studies are required to better understand how we can identify and intervene with women in the prenatal period to remediate barriers to optimal parenting and child development outcomes [44–46]. Nevertheless, the finding that about a third of this population reported a history of abuse speaks to the potentially pervasive experience of abuse across all socio-economic domains and highlights opportunities for improved access to services and supports to address past abuse.

It is important to note that maternal history of abuse, poor parenting morale, and poor maternal social support were all associated with highest risk of developmental problems (Path A) at school entry in the bivariate analysis, but maternal history of abuse and poor parenting morale were more predictive in the logistic regression model of a child screening in Path A when accounting for other factors. This may be due to the relationship among these variables. Past research has demonstrated that women in abusive situations often have poor social support, resulting from the social systems that create an environment conducive to abuse or the social isolation imposed on women by those who abuse them [46, 47]. Women who report a history of abuse, regardless of current socio-demographic circumstances, may be less willing or able to engage in supportive social relationships due to their view of self or others [46].

Children at high risk for developmental problems were more likely to have mothers who had low parenting morale when their children were three years old. It may be that mothers with low parenting morale are particularly sensitive to difficulties that may be emerging in their children, or that a child's emerging developmental difficulties and a mother's parenting morale influence each other in a reciprocal fashion. Considering this, support to parents that address sense of competence and morale may positively impact parental well being and child development.

In this study, 15% of children screened at high risk of developmental problems, similar to the proportion found when this cohort of children was three years of age. Although most children had had routine medical checkups, only about half of those who screened at high risk of developmental problems based on parents' concerns had been previously referred for assessments or interventions. The common practice among physicians for developmental surveillance is subjective clinical observation which identifies only 30 - 50% of children with developmental delay [48, 49]. Physicians' developmental screening often focuses on sensory deficits, with less attention to developmental and behavioural/emotional problems. Reliable screening approaches would greatly improve the rate of identification [50]. Finding solutions for providing accessible, valid and reliable identification of developmental and behavioural/emotional problems with appropriate interventions for children and families at risk would benefit all of society.

Although the women most likely to participate in the study had high levels of education and household incomes, the sample aligns with the income level of three-quarters of all Canadian families with kids under six years of age and the educational attainment of three-quarters of Canadian women giving birth [33, 51]. The results then are generalizable to the majority of Canadian families with children under six. Furthermore, 15% of children screened at highest risk of developmental problems, as would be anticipated in a population based setting.

Despite our attempts to follow-up women in the study, mothers who were younger, had lower education and income, had poor physical health, were single or divorced, and smoked were more likely to be unreachable or not respond to follow-up attempts. Thus, these findings cannot be generalized to this more vulnerable population. In other research, these factors have been associated with increased risk for poor infant outcomes and child developmental problems, and consequently, the results potentially underestimate the proportion of the total population of preschool children at risk of problems [10, 12–14, 19, 52–54].

It is important to consider that the factors associated with women who were not retained in the cohort (young maternal age, low education and income, poor physical health, being single, and smoking) are similar to the characteristics of women who are difficult to retain in studies and in longitudinal research [55]. These factors may signal complex health, lifestyle, and social issues that these women face which make it difficult to retain them [56]. Also, the original study was a community based study that was not initially designed to be a longitudinal follow-up study. Thus, traditional strategies to retain women (e.g. incentives, changes of address cards, and routine follow-up) were not implemented immediately after the first study. However, retention strategies were implemented between the follow-up study at 3 years and at 5 years (e.g. routine contact, asking women to inform us of an upcoming change in contact information). In all three studies, the participation rates were over 60%, and the women not retained in the cohort appear to be similar among all three studies.

Because this sample included few women who had a household income less than $40,000 and none who had less than a high school education, the current study did not find differences in developmental risk by education or income level. Nevertheless, this study raises the consideration that although children may not be at risk for developmental problems due to maternal education or family income, other factors may place children at risk for developmental problems. It is noteworthy that the women in this study would be commonly defined as a "low risk" population considering their demographic information. However, one in four of these mothers had mental health issues and almost one in three had a history of abuse, which in turn may have placed their children at risk for developmental problems.

Notably, the PEDS is not diagnostic, but is rather a screening tool based on parent report of concerns. It assigns a level of risk of developmental problems based on those concerns. Thus, children who are identified as being at high risk for developmental problems may not actually be encountering developmental problems. Similarly, children whose parents do not express concerns about development may be delayed or at risk of behavioural/emotional problems. This possibility for misclassification by the PEDS must be considered when interpreting the regression model in this paper, yet the magnitude of the odds ratios and significance and consistency of the findings related to the impact of maternal factors suggest a notable relationship to child development [14, 18–20, 25, 28, 52–54].

This study and the previous follow-up study indicate that the well-being of mothers is associated with developmental risk in children through the preschool years and at school entry. There is a consistent influence of maternal well-being on infant and child development, and an opportunity to better understand how women's well-being during pregnancy and early childhood can be optimized. Identifying women with a history of abuse and determining what strategies are needed to support their well-being and parenting has the potential to improve the life course trajectory for children and families. Thus, ongoing research and evaluation of interventions that enhance parenting and early assessment of child development is warranted. Failure to intervene when risk can be identified may limit efforts to optimize child development by school entry.