Paternal Depression-article

Articles

Paternal depression in the postnatal period and child development: a prospective population study Paul Ramchandani, Alan Stein, Jonathan Evans, Thomas G O’Connor, and the ALSPAC study team*

Summary Background Depression is common and frequently affects mothers and fathers of young children. Postnatal depression in mothers affects the quality of maternal care, and can lead to disturbances in their children’s social, behavioural, cognitive, and physical development. However, the effect of depression in fathers during the early years of a child’s life has received little attention. Methods As part of a large, population-based study of childhood, we assessed the presence of depressive symptoms in mothers (n=13 351) and fathers (n=12 884) 8 weeks after the birth of their child with the Edinburgh postnatal depression scale (EPDS). Fathers were reassessed at 21 months. We identified any subsequent development of behavioural and emotional problems in their children (n=10 024) at age 3·5 years with maternal reports on the Rutter revised preschool scales. Findings Information was available for 8431 fathers, 11 833 mothers, and 10 024 children. Depression in fathers during the postnatal period was associated with adverse emotional and behavioural outcomes in children aged 3·5 years (adjusted odds ratio 2·09, 95% CI 1·42–3·08), and an increased risk of conduct problems in boys (2·66, 1·67–4·25). These effects remained even after controlling for maternal postnatal depression and later paternal depression. Interpretation Our findings indicate that paternal depression has a specific and persisting detrimental effect on their children’s early behavioural and emotional development.

Introduction The findings of research done in people and in animals indicate that early maternal influences are crucial in the development of offspring.1–4 Maternal postnatal depression affects the quality of maternal care and predicts disturbances in children’s later social, behavioural, cognitive, and physical development.5–8 However, the effect of depression in fathers during the early months of their children’s lives has received little attention. There is research9,10 to show that adolescent children of depressed fathers have increased rates of psychopathology, but little is known about the possible effects of paternal depression early in children’s lives.10,11 This gap in knowledge is of concern for various reasons. First, depression is a major worldwide health problem. It is the fourth biggest cause of disability and, by 2020, the global burden of depression will probably be second only to cardiovascular disease.12 Depression often affects parents and is associated with an adverse effect on their children’s development.5,7,13 A proportion of fathers do report depression after the birth of a child,14 suggesting it could be an important health issue. Second, there is good evidence5 that maternal postnatal depression affects children’s development across a range of domains. Symptoms of depression, including low mood, irritability, and feelings of hopelessness, in either mothers or fathers, are likely to interfere with the ability to provide responsive parenting. The quality and sensitivity of parenting during a child’s first year of life could have persisting effects on development.2 Third, www.thelancet.com Vol 365 June 25, 2005

Lancet 2005; 365: 2201–05 See Comment page 2158 *Study team listed at end of paper University of Oxford Department of Psychiatry, Warneford Hospital, Headington, Oxford OX3 7JX, UK (P Ramchandani MRCPsych, Prof A Stein FRCPsych); Academic Unit of Psychiatry (J Evans MD) and Department of Community Based Medicine (ALSPAC study team), University of Bristol, Bristol, UK; and Department of Psychiatry, University of Rochester, Rochester, NY, USA (T G O’Connor PhD) Correspondence to: Dr Paul Ramchandani paul.ramchandani@ psych.ox.ac.uk

research and some governmental policy—for example, the National Service Framework for Children, Young People, and Maternity Services in England15—are beginning to recognise the important role that many fathers play in their children’s early development16 and are encouraging greater involvement of the father in their upbringing. Our aim, therefore, was to assess paternal depression after childbirth. We postulated that paternal depression in the postnatal period would be associated with an increased risk of behavioural and emotional problems in children aged 3·5 years. In view of research findings17–19 that boys might be more vulnerable to the effects of postnatal depression in their mothers, we further hypothesised that the effect of paternal depression at this age would be greater in boys than in girls.

Methods Participants The Avon Longitudinal Study of Parents and Children (ALSPAC)20 is a longitudinal cohort study set up to collect comprehensive data on a large population sample of children and their parents from early pregnancy through childhood. Pregnant mothers were recruited from the Bristol area of the UK. All pregnant women due to deliver their baby between April 1, 1991, and Dec 31, 1992, were eligible, and an estimated 85–90% were recruited. We sent questionnaires to mothers and fathers at regular points during and after pregnancy. The 2201

Articles

questionnaires assessed a wide range of factors, from environmental toxins and psychosocial risk to physical and psychological development. In this part of the study, we focus on parental depression during the postnatal period and subsequent child emotional and behavioural adjustment. All participants provided informed consent, and the study was approved by the ALSPAC Ethics and Law committee and local research ethics committees.

Procedures We used the Edinburgh postnatal depression scale (EPDS) to assess mothers and fathers at 8 weeks after the birth of their baby. Fathers were assessed again at 21 months. The period 6–12 weeks is often used in the assessment of postnatal depression.17,18,21 The EPDS is a well validated, widely used, self-report questionnaire that consists of ten items.21 Although the EPDS was developed to screen for depression in women postnatally, it is also useful in the assessment of mothers outside the postnatal period and has been validated in men.22,23 The scale is not a diagnostic method, though scores of more than 12 do identify a diagnosis of major depressive disorder with a high specificity (95·7%) and sensitivity (81·1%);24 we used this cut-off, and describe the group that scored more than 12 as the depressed group. We measured disturbance of children’s emotional and behavioural development at age 3·5 years (42 months) with maternal reports on the Rutter revised preschool scales.25 Every item on this measure describes a characteristic or behaviour—eg, “is worried, worries about many things” and “fights with other children”— with three possible responses—Yes certainly, Yes sometimes, and No. Individual items combine to form three problem scales (emotional problems, conduct problems, and hyperactivity), and a prosocial behaviours scale. All problem behaviours combine to give a total problems scale. We have used a cut-off of the top 10% of scores on every scale to indicate high-scorers. The % high scores Depressed

Not depressed

Crude OR (95% CI)

Adjusted OR (n=5939)*

Maternal depression (n=9507) Emotional 21·9% Conduct 20·7% Hyperactivity 12·5% Prosocial 10·9% Total 20·7%

12·1% 10·7% 6·8% 8·9% 7·8%

2·03 (1·71–2·41) 2·18 (1·82–2·60) 1·97 (1·59–2·45) 1·25 (1·00–1·57) 3·08 (2·57–3·69)

2·03 (1·61–2·56) 2·05 (1·60–2·63) 1·68 (1·24–2·28) 1·09 (0·79–1·50) 2·76 (2·14–3·55)

Paternal depression (n=7030) Emotional 18·5% Conduct 16·5% Hyperactivity 11·2% Prosocial 10·8% Total 16·9%

12·9% 10·6% 7·1% 9·1% 8·5%

1·52 (1·10–2·11) 1·66 (1·18–2·35) 1·65 (1·10–2·47) 1·22 (0·81–1·84) 2·19 (1·55–3·08)

1·27 (0·87–1·87) 1·84 (1·27–2·68) 1·68 (1·08–2·62) 1·24 (0·78–1·98) 2·09 (1·42–3·08)

*Adjusted for effects of social class, degree of maternal education, and depression in other parent.

Table 1: Parental postnatal depressive symptoms and emotional and behavioural problems in children aged 42 months

2202

10% cut-off has been used in previous research.26 Furthermore, a more recent development of the Rutter questionnaires,27 the strengths and difficulties questionnaire (SDQ), uses a 10% cut-off, and has been tested in numerous studies, including the National Survey of Mental Health of Children and Adolescents.28 SDQ scores above the 90th centile indicate problem behaviours and are associated with a substantially increased risk of psychiatric disorder.29 Although we did analyses with both the highest 10% cut-off and the continuous scores as the outcome measure, only the analyses that used cut-off scores are presented. The continuous scores revealed similar associations, and the findings are available from PR.

Statistical analysis We calculated prevalence of depression in mothers and fathers and then did univariate analyses to examine the associations between postnatal depression in each parent and later emotional and behavioural problems in their children. We did logistic regression analyses to assess the effect of paternal depression on child outcome, controlling for social class, maternal education, and depression in the other parent. Degree of maternal and paternal education were highly correlated, and repetition of the analyses with paternal and maternal educational levels combined in the regression model did not substantially affect the odds ratios (ORs). We have, therefore, used degree of maternal education throughout for consistency. We did further analyses to control for the effects of later paternal depression (measured at 21 months post-birth) to see whether any effects of paternal depression noted were specific to the postnatal period. To assess the specificity of effects by sex of the child, we tested statistical interactions for fathers’ depression by sex with likelihood ratio tests, and report separate ORs for boys and girls. We undertook analyses for the whole sample and then separately for the sample with the small number of second twin births excluded (1%, n=114 of 10 024), to remove any effects of familial clustering. The results were unchanged, and are reported for the whole sample.

Role of the funding source The sponsors of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication.

Results The overall ALSPAC cohort consisted of 13 822 women with live children 28 days after birth. 13 351 mothers agreed to participate in the study and were sent a questionnaire 8 weeks after the birth of their child. 12 884 of respondents had partners. Data were available on mothers’ postnatal depressive symptoms for 11 833 www.thelancet.com Vol 365 June 25, 2005

Articles

(89%) of these women, and on fathers’ postnatal depressive symptoms in 8431 of 12 884 (65%) male partners. More than 99% of these men were identified as the father of the child. Questionnaires were completed about children’s behaviour at 42 months for 10 024 children (75% of those who were in the study at age 8 weeks). High amounts of depressive symptoms in mothers (73% vs 81%), but not in fathers (82% vs 83%), were associated with a lower chance of remaining in the study at 42 months of follow-up. 1203 (10%) mothers and 303 (4%) fathers scored more than 12 on the EPDS. Scores on the EPDS for mothers and fathers were significantly correlated (Pearson correlation 0·27, p0·001). Paternal depression was strongly associated with an increased risk of high total problems scores on the Rutter preschool scales (OR 2·19, 95% CI 1·55–3·08) and with high scores on all three problem subscales (emotional, conduct, and hyperactivity), but not on prosocial behaviour scores (table 1). When we reassessed these associations controlling for social class, degree of education, and maternal depression, they remained significant for conduct problems, hyperactivity, and total problems (2·09, 1·42–3·08), but not for emotional symptoms (table 1). Maternal depression was also associated with an increased risk of high total scores on the Rutter preschool scales (3·08, 2·57–3·69), and with high scores on all three problem subscales (table 1). These associations remained significant when analyses were controlled for social class, degree of education, and paternal depression. When we controlled for depression in fathers at 21 months post-birth, the association between fathers’ postnatal depression and subsequent conduct problems (1·73, 1·06–2·85) and hyperactivity (1·96, 1·12–3·43) in the children remained (table 2). The association with the total-problems score was no longer significant. The association between paternal depression and later behaviour (conduct) problems was stronger in boys than in girls (likelihood ratio test 5·26, p=0·022). The separate ORs for boys and girls are shown in table 3. Maternal depression was associated with high problem scores across all domains of child psychological OR (95% CI)*

Emotional Conduct Hyperactivity Prosocial Total

Paternal depression

Maternal depression

0·87 (0·51–1·51) 1·73 (1·06–2·85) 1·96 (1·12–3·43) 1·47 (0·81–2·67) 1·59 (0·92–2·75)

1·88 (1·42–2·48) 2·17 (1·63–2·90) 1·62 (1·12–2·33) 1·23 (0·78–1·62) 3·10 (2·32–4·14)

*Adjusted for effects of social class, degree of maternal education, depression in the other parent, and later paternal depression. 4430 children assessed.

Table 2: Paternal depression and subsequent emotional and behavioural problems in children (controlling for later paternal depression at 21 months)

www.thelancet.com Vol 365 June 25, 2005

OR (95% CI)* Boys

Girls

Likelihood ratio test for interaction by sex

p

Maternal depression (n=5939) Emotional 1·84 (1·31–2·57) Conduct 2·18 (1·58–3·01) Hyperactivity 1·66 (1·11–2·49) Prosocial 1·16 (0·79–1·71) Total 2·64 (1·87–3·71)

2·24 (1·62–3·11) 1·92 (1·29–2·87) 1·77 (1·12–2·80) 0·91 (0·50–1·64) 3·00 (2·06–4·38)

0·822 0·632 0·093 0·413 0·085

0·365 0·427 0·761 0·512 0·770

Paternal depression (n=5939) Emotional 1·31 (0·75–2·27) Conduct 2·66 (1·67–4·25) Hyperactivity 2·06 (1·16–3·66) Prosocial 1·40 (0·79–2·47) Total 2·91 (1·77–4·80)

1·23 (0·71–2·11) 1·10 (0·56–2·17) 1·35 (0·66–2·75) 1·06 (0·45–2·47) 1·38 (0·73–2·61)

0·001 5·262 0·710 0·393 3·059

0·993 0·022 0·399 0·530 0·080

*Adjusted for effects of social class, degree of maternal education, and depression in other parent.

Table 3: Differential effects of parental depression on boys and girls

functioning, irrespective of sex. However, depression in fathers was associated only with raised levels of problems for boys. The association between depression in fathers and later emotional and behavioural problems in their children seems to be largely a result of an increased risk to their sons. This association is more pronounced for behavioural symptoms (conduct and hyperactivity) than for emotional symptoms (worry and sadness).

Discussion Our findings indicate that children of fathers who have depression during the postnatal period are at increased risk of behavioural problems at age 3·5 years, even after maternal depression and other factors had been controlled for. The increased risk associated with depression in fathers during the infant’s early months also remains after controlling for later paternal depression, suggesting that paternal depression in the early months of a child’s life might be a particular risk factor for adverse development. Our findings further indicate that the association between paternal depression and child behaviour problems is stronger in boys than in girls. This notion warrants further investigation. We also noted an association between depression in mothers and higher rates of behavioural and emotional problems in their children. This association differed from that seen with depression in fathers in two respects; first, an effect was seen across all the domains of child emotional and behavioural functioning measured in this study and, second, there was no difference in the magnitude of effects seen in boys or girls. Our study has various strengths. We report data for a large population sample, which has been followed-up for 3·5 years. We applied widely used, well validated questionnaire measures, and controlled for important confounding factors. Furthermore, paternal symptoms of depression and outcomes in children were rated by different people, substantially reducing the risk of rater 2203

Articles

bias as a potential confounding factor. The rate of depression in mothers noted by us is comparable to that noted in other studies (13%).30 Few studies of fathers have been done. Most are small and, consequently, there is a wide variation in the recorded rates of depression (1–26%).11 The main limitation of our study was that formal psychiatric diagnoses could not be done. However, the EPDS has high sensitivity and specificity for depressive disorder. There also remains a possibility that some of the associations described were the result of some unmeasured confounding factor. Finally, the response rate for fathers was lower than that for mothers, and so some degree of response bias is possible. If depressed fathers were less likely to take part in the study, underestimation of the associated risk of subsequent child behaviour problems might have arisen. Lower response rates from fathers than mothers are common in studies of children and parents. There is a fairly large body of research suggesting a link between postnatal depression in mothers and adverse development of their offspring. Our results confirm earlier findings6,17 of an association between depression in mothers and behavioural and emotional problems in their children. Additionally, the results with respect to depression in fathers add important new findings to the published work. Research has only recently begun to address the possible effect of fathers on children’s early development.16 In particular, there has been little research on depression in fathers when their children are very young.10,11 There are several possible explanations for our findings. First, paternal depression might have a direct effect on the way fathers interact with their children, as has been reported for postnatal depression and other psychiatric conditions in mothers.5,31 This notion would not be surprising given the pervasive effects of depression on psychological and social functioning. An effect on early interactions might itself lead to an increased risk for adverse development in the children. Second, this association might be due to indirect effects. For example, depression in fathers could be linked to marital conflict, which has been linked with behavioural problems in children. Third, genetic factors could play a part, though findings of studies5 of maternal postnatal depression have shown a clear degree of environmentally mediated effects. Finally, the possibility that some part of the association noted between depression in fathers and increased risk of behavioural problems in their children is due to childto-parent effects needs to be considered. Theoretically, early differences in the child before the onset of parental depression—ie, in the first few weeks of life— could have increased the risk of depression in the parent. However, there are no validated measures of child behaviour so early in life and behaviour at this age is not stable over time. One of the strengths of this study, which examines parental depression so early in 2204

the child’s life, is that parental depression almost certainly occurs before the development of behavioural problems and certainly before they become established, hence indicating that any causal relation is largely in the direction of parent-to-child. The specific relationship between a boy’s behavioural development and depression in their father is striking. Perhaps boys are specifically sensitive to the effects of parenting by fathers, potentially because of differential involvement by fathers with their sons. Alternatively, boys might be more vulnerable than girls to a range of adverse parental influences, including depression in their fathers. Findings of published work on the specificity of sex effects are inconclusive. Although some studies17,18 of maternal postnatal depression have reported a preponderance of problems in boys, others, including ours, have not. The influence of fathers in very early childhood might have been underestimated in the past. Although our findings need to be confirmed and expanded, we believe that after the birth of a child, depression in fathers, as well as depression in mothers, should be actively considered. Depression could compromise the ability of fathers both to care responsively for their children and to undertake other roles in the family. The potential for an effect of paternal depression on child development indicated by this research emphasises the fact that addressing paternal depression in its own right is important. From a research perspective more detailed studies of paternal depression in the postnatal period, including direct observations of father-to-child and family interaction, would be informative. Interventions for paternal depression could then be developed to ameliorate its effect on children. Fathers play an important part in the early life and development of their children. In the case of paternal depression, this effect seems especially notable in boys. Contributors P Ramchandani and A Stein had the idea for these analyses, to which J Evans and T G O’Connor contributed. P Ramchandani did the analyses, and all four authors contributed to writing of the report. The ALSPAC study team designed the study and collected and entered the data. Conflict of interest statement PR, AS, and TGO’C declare that they have no conflict of interest. JE has received payments to give lectures from pharmaceutical companies who market antidepressants. Acknowledgments We thank all the mothers and fathers who took part, the midwives for their cooperation and help in recruitment, and Helen Doll for statistical advice. The ALSPAC study team comprises interviewers, computer technicians, laboratory technicians, clerical workers, research scientists, volunteers, and managers who continue to make the study possible. This study was funded by the Medical Research Council, the Wellcome Trust, UK government departments, medical charities, and others. The ALSPAC study is part of the WHO initiated European Longitudinal Study of Pregnancy and Childhood. PR holds a Special Training Fellowship in Health Services and Health of the Public Research, funded by the Medical Research Council, UK. AS is supported by the Wellcome Trust.

www.thelancet.com Vol 365 June 25, 2005

Articles

References 1 Suomi SJ. Early determinants of behaviour: evidence from primate studies. Br Med Bull 1997; 53: 170–84. 2 O’Connor TG, Marvin RS, Rutter M, et al. Child-parent attachment following early institutional deprivation. Dev Psychopathol 2003; 15: 19–38. 3 Rutter M, O’Connor TG, and the English and Romanian adoptees (ERA) study team. Are there biological programming effects for psychological development? Findings from a study of Romanian adoptees. Dev Psychopathol 2004; 40: 81–94. 4 Newport DJ, Stowe ZN, Nemeroff CB. Parental depression: animal models of an adverse life event. Am J Psychiatry 2002; 159: 1265–83. 5 Murray L, Cooper P. Intergenerational transmission of affective and cognitive processes associated with depression: infancy and the preschool years. In: Goodyer I, ed. Unipolar depression: a lifespan perspective. Oxford: Oxford University Press, 2003. 6 Goodman SH, Brogan D, Lynch ME, et al. Social and emotional competence in children of depressed mothers. Child Dev 1993; 64: 516–31. 7 Downey G, Coyne JC. Children of depressed parents: an integrative review. Psychol Bull 1990; 108: 50–76. 8 Rahman A, Iqbal Z, Bunn J, et al. Impact of maternal depression on infant nutritional status and illness: a cohort study. Arch Gen Psychiatry 2004; 61: 946–52. 9 Beardslee WR, Kellner MB, Lavori PW, et al. The impact of parental affective disorder on depression in offspring: a longitudinal follow-up in a nonreferred sample. J Am Acad Child Adolesc Psychiatry 1993; 32: 723–30. 10 Kane P, Garber J. The relations among depression in fathers, children’s psychopathology, and father-child conflict: a metaanalysis. Clin Psychol Rev 2004; 24: 339–60. 11 Goodman JH. Paternal postpartum depression, its relationship to maternal postpartum depression, and implications for family health. J Adv Nurs 2004; 45: 26–35. 12 Murray CJ, Lopez AD. Regional patterns of disability-free life expectancy and disability-adjusted life expectancy: global burden of disease study. Lancet 1997; 349: 1347–52. 13 Ramchandani P, Stein A. The impact of parental psychiatric disorder in children. BMJ 2003; 327: 242–43. 14 Deater-Deckard K, Pickering K, Dunn JF, et al. Family structure and depressive symptoms in men preceding and following the birth of a child. Am J Psychiatry 1998; 155: 818–23. 15 Department of Health. National service framework for children, young people, and maternity services: core standards. London: Stationery Office, 2004. 16 Lamb ME, Lewis C. The development and significance of fatherchild relationships in two-parent families. In: Lamb ME, ed. The role of the father in child development, 4th edn. New Jersey: Wiley and Sons, 2004.

www.thelancet.com Vol 365 June 25, 2005

17 18

19 20

21

22

23

24

25

26

27

28

29

30 31

Sinclair D, Murray L. Effects of postnatal depression on children’s adjustment to school. Br J Psychiatry 1998; 172: 58–62. Sharp D, Hay DF, Pawlby S, et al. The impact of postnatal depression on boys intellectual development. J Child Psychol Psychiatry 1995; 36: 1315–36. Zahn-Waxler C. Warriors and worriers: gender and psychopathology. Dev Psychopathol 1993; 5: 79–89. Golding J, Pembrey M, Jones R, et al. ALSPAC: the Avon Longitudinal Study of Parents and Children, 1—study methodology. Paed Perinatal Epidemiol 2001; 15: 74–87. Cox J, Holden JM, Sagovsky R. Detection of postnatal depression: development of the 10-item Edinburgh postnatal depression scale. Br J Psychiatry 1987; 150: 782–86. Matthey S, Barnett B, Kavanagh DJ, et al. Validation of the Edinburgh postnatal depression scale for men, and comparison of item endorsement with their partners. J Affect Disord 2001; 64: 175–84. Cox JL, Chapman G, Murray D, et al. Validation of the Edinburgh postnatal depression scale (EPDS) in non-postnatal women. J Affect Disord 1996; 39: 185–89. Murray L, Carothers AD. The validation of the Edinburgh postnatal depression scale in a community sample. Br J Psychiatry 1990; 157: 288–90. Elander J, Rutter M. Use and development of the Rutter parents’ and teachers’ scales. Int J Methods Psychiatric Research 1996; 6: 63–78. Plomin R, Price TS, Eley TC, et al. Associations between behaviour problems and verbal and nonverbal cognitive abilities and disabilities in early childhood. J Child Psychol Psychiatry 2002; 43: 619–33. Goodman R, Scott S. Comparing the strength and difficulties questionnaire and the child behavior checklist: is small beautiful? J Abnormal Child Psychol 1999; 27: 17–24. Meltzer H, Gatward R, Goodman R, Ford T. Mental health of children and adolescents in Great Britain. London: Stationery Office, 2000. Goodman R. Psychometric properties of the strengths and difficulties questionnaire. J Am Acad Child Adolescent Psychiatry 2001; 40: 1337–45. O’Hara MW, Swain AM. Rates and risk of postpartum depression: a meta-analysis. Int Rev Psychiatry 1996; 8: 37–54. Stein A, Woolley H, Murray L, et al. The influence of psychiatric disorder on the controlling behaviour of mothers with one yearold infants: a study of maternal eating disorder, postnatal depression and a healthy comparison group. Br J Psychiatry 2001; 179: 157–62.

2205