Statistics from Altmetric.com
QUESTION: Are anxiety and depression at 14 years related to poverty, marital break-up and maternal anxiety and depression during early childhood?
Longitudinal cohort study.
Pregnant women were enrolled at an average of 18 weeks gestation, reinterviewed 3–5 days after birth and again when their child was 6 months, 5 years and 14 years old. Children were assessed at 14 years. Complete data were available for 4434 families.
Assessment of risk factors
Poverty and martial break-up were assessed using maternal self report. The depression subscale from the Delusions Symptoms States Inventory was used to assess maternal anxiety and depression. Mother’s marital relationship distress was categorised using Dyadic Satisfaction items from the Spanier Dyadic Adjustment Scale.
Main outcome measures
Anxiety and depressive symptoms at 14 years assessed using Child Behaviour Checklist and Youth Self report subscales.
After controlling for poverty and marital relationship factors, maternal anxiety and depression during early childhood predicted high anxiety and depression symptoms at age 14. The effect was small, but increased with repeated exposure to maternal depression. Poverty, marital break-up and distressed marital relationships during the child’s first five years also increased the risk of adolescent anxiety slightly. Stable single parent status was not a risk factor. There were no gender differences in risk factors, apart from poverty that had more of an effect on girls.
Maternal anxiety and depression, marital conflict, marital break-up and poverty during a child’s first 5 years have a small but significant association with depression in adolescence.
Numerous studies have found a correlation between parental psychopathology and negative outcomes in children,1 although the strength of this link and the mechanisms underlying it are not well understood. Spence et al provide further evidence of a link between maternal depression and adolescent internalising symptoms. The large sample and reasonable retention rate are strengths. The researchers examined several variables related to both maternal and adolescent depression and anxiety, provide a thoughtful discussion, offer several alternative explanations and suggest questions for future research.
There are some limitations, however. Why have the independent and dependent variables been dichotomised rather than using the entire range of scores? There is a lack of information about the extent of maternal psychopathology and other mediators or proximal predictors during the years between assessments at 5 and 14 years. Without controlling for the level of internalising symptoms at age 5, it is not possible to rule out simple continuity of symptoms. Nor can we assess the incremental contribution of other predictors (poverty, marital status, maternal symptoms) beyond the level of symptoms at age 5 when predicting internalising symptoms at age 14.
The study raises important research and clinical issues. As in past research, there was a lack of concordance between parent and child report. Moreover, links between predictors and adolescent outcomes varied as a function of the reporter. This reinforces the importance of obtaining information from multiple informants. We should not assume, however, that data from different reporters should be combined or that one reporter is more accurate than another. The finding that more chronic exposure to maternal depression predicts worse outcomes for children is also consistent with past studies. Maternal depression is heterogeneous in terms of severity, chronicity and recurrence. Evidence from this study and others2 suggests that repeated or prolonged exposure to maternal depression may be particularly detrimental to children’s functioning. Clinicians therefore need to intervene early with depressed mothers to reduce the duration and relapse of symptoms. Moreover, pediatricians, obstetricians and primary care physicians need to assess maternal depression and make appropriate referrals for treatment.
Stable single parent status was not associated with increased risk of adolescent internalising symptoms, whereas marital conflict or break-ups during the child’s first 5 years did predict adolescent symptoms. Thus, marital status may be less important than marital distress and instability in predicting children’s emotional well-being. Marriage per se may not be as critical for children’s outcomes as a low conflict home with healthy parenting by either single or married parents.
For correspondence: S Spence, School of Psychology, University of Queensland, Brisbane, Australia.
Sources of funding: Not specified
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.