Paternal Depression Symptoms During Pregnancy and After Childbirth Among Participants in the Growing Up in New Zealand Study

Key Points

Question  What demographic, social, relationship, and health factors are associated with depression symptoms among men whose partners are pregnant or recently gave birth?

Findings  In a cohort study of 3523 men living in New Zealand, 2% scored higher than 12 on the Edinburgh Postnatal Depression Scale during the third trimester of their partners’ pregnancy and 4% scored higher than 9 on the 9-item patient health questionnaire 9 months after childbirth. Fathers who were stressed or in poor health had elevated depression symptoms during their partner’s pregnancy and 9 months after the birth of their child; postnatal paternal depression symptoms were also associated with adverse social and relationship factors.

Meaning  The effect of parental depression on families and children can be lowered if symptoms in both men and women are identified early, treated, or prevented.

Importance  Antenatal and postnatal depression are known to be common and associated with poor outcomes for women and their children. There is little evidence on depression symptoms among men during the perinatal period.

Objective  To identify characteristics associated with depression symptoms among men whose partners were pregnant and subsequently gave birth.

Design, Setting, and Participants  A longitudinal cohort study provided data from a demographically diverse sample of 3523 New Zealand men who completed interviews during their partner’s pregnancy and 9 months after the birth of their child. Participants were drawn from a cohort whose partners were pregnant women with a due date between April 25, 2009, and March 25, 2010, who were enrolled in the Growing Up in New Zealand study. Data analysis was conducted from September 1, 2015, to January 8, 2016.

Main Outcomes and Measures  Depression symptoms were measured using the Edinburgh Postnatal Depression Scale and the 9-item Patient Health Questionnaire; elevated depression symptoms were defined as scores higher than 12 and 9, respectively.

Results  The mean (SD) age of the participants at the antenatal interview was 33.20 (6.25) years (range, 16-63 years). Elevated antenatal paternal depression symptoms affected 82 fathers (2.3%) and were associated with perceived stress (odds ratio [OR], 1.38; 95% CI, 1.30-1.47) and fair to poor health during their partner’s pregnancy (OR, 2.06; 95% CI, 1.18-3.61). Elevated postnatal paternal depression symptoms affected 153 (4.3%) of fathers and were associated with perceived stress in pregnancy (OR, 1.12; 95% CI, 1.08-1.17), no longer being in a relationship with the mother 9 months after childbirth (OR, 6.36; 95% CI, 2.28-17.78), having fair to poor health at 9 months (OR, 3.29; 95% CI, 2.10-5.16), being unemployed at 9 months (OR, 1.86; 95% CI, 1.11-3.10), and a history of depression (OR, 2.84; 95% CI, 1.69-4.78).

Conclusions and Relevance  Expectant fathers were at risk of depression symptoms if they felt stressed or were in poor health. Rates of elevated depression symptoms were higher during the postpartum period and were associated with adverse social and relationship factors. Identifying fathers most at risk of depressive symptoms and when best to target interventions (antenatal or postnatal) may be beneficial to men and their families.

Nearly 9% of men report a lifetime presence of depression and 2.6% report a depressive episode within the past year.¹ Pregnancy and childbirth may be high-risk periods for male depression.^2,3 Yet, interest in perinatal depression has focused on women despite policies to improve the sex balance in research.⁴ Studying mothers and fathers should be a guiding principle in developmental research. Therefore, we follow-up our group’s maternal perinatal depression studies^5,6 with this article on paternal antenatal and postnatal depression symptoms.

Maternal perinatal depression is linked to physiological changes (eg, hormone fluctuations).⁷ Expectant and new fathers also experience biological and ecological stressors, including changes to brain circuits, structure, and hormones,^8-11 that can increase their risk of depression symptoms.^12,13 Thus, some of the biological mechanisms underlying perinatal depression may increase the risk in both parents.

Paternal depression is associated negative child outcomes, including emotional and behavioral problems.¹⁴ Although paternal depression does not expose fetuses and infants to the same intrauterine/physiological risks as maternal depression, paternal genetic and psychosocial factors may act directly on the child and indirectly through their effect on maternal well-being.^5,15

Reported depression rates among men during their partner’s pregnancy vary from 2% to 19%,^16-18 compared with 12% to 19% for women.¹⁹ As with mothers, there is debate about whether paternal antenatal depression symptoms (ADS) reflect ongoing prenatal depression rather than onset during pregnancy. Risks for maternal ADS include unplanned pregnancy, domestic violence, lack of social or relationship support, and past mental illness.^6,19,20 There is less evidence on paternal ADS, but risks include anxiety, daily “hassles,”¹² and having a depressed partner.²¹ Different factors are likely to contribute to the risk of depression in each of the perinatal periods.^22-24

An estimated 3% to 12% of men experience postnatal depression symptoms (PDS) compared with 13% to 18% of women.^13,25,26 The extent to which PDS are a continuation of ADS or prenatal depression remains unclear owing to a lack of longitudinal studies.¹⁹ Risks for maternal PDS include domestic violence, low socioeconomic status, physical illness, neuroticism, increased parity, multiple births, and ADS.^6,19,20 Many of these factors stem from prepregnancy or pregnancy circumstances.^27,28 Paternal PDS risks include low income, unemployment, renting, maternal depression, and marital conflict.²⁶ Lack of social support and cohesion is also strongly associated with maternal^29-31 and paternal^32,33 perinatal depression. Increasing social support is therefore a key area for intervention.³⁴

Evidence suggests that depression in either parent increases the likelihood of the partner developing perinatal depression symptoms.^{12,16,21,32,35-37} However, the influence of paternal depression on women seems to be greater than vice versa.^12,38 Unlike women, there is not a strong link between past mental health problems and paternal perinatal depression.³⁸ More prospective longitudinal studies that include partnered men and women and span both perinatal periods are needed to confirm these findings.^15,16

The present study investigated ADS and PDS among expectant and new fathers. The association between maternal and paternal depression symptoms was explored.

Participants were drawn from a cohort whose partners were pregnant women, living in 3 contiguous District Health Board regions in New Zealand who had a due date between April 25, 2009, and March 25, 2010, who were enrolled in the Growing Up in New Zealand study.³⁹ Of the 6167 women who were interviewed during pregnancy for the study, 5834 women (94.6%) had a partner and 4402 partners (75.5%) agreed to take part.⁴⁰ Prospective parents in the study were evenly distributed among socioeconomic groups, and most participants (97.5%) were married or cohabiting with a study mother. The New Zealand Ministry of Health Northern Y Regional Ethics Committee approved the study and written informed consent was obtained. Participants who completed the antenatal interview received a gift bag.

A face-to-face, computer-assisted interview was completed at 3827 participants’ homes during the third trimester of their partner’s pregnancy (ie, an interviewer used a computer to administer a personalized questionnaire). Of the 4402 partners who agreed to take part in the study, 1 gave no ADS data and 575 participants completed the interview retrospectively after childbirth; hence, these individuals were excluded, leaving a sample of 3826.

A second interview (n = 3549) was conducted approximately 9 months after childbirth. Thus, 277 participants (7.8%) participated antenatally but not postnatally. Those who remained in the study were older, more likely to be of European ethnicity, married, employed, born in New Zealand, have a bachelor’s degree or higher, and live in a low deprivation area compared with those not completing the postnatal interview (all P < .05). A final sample of 3523 men was included in these analyses (11 women, 10 nonbiological fathers, and 5 fathers not in a relationship with a study mother at the time of the antenatal interview were excluded).

European or other (including Middle Eastern, Latin American, and African) participants composed 69.7% (n = 2451) of the sample, followed by Māori, 9.5% (n = 333) Pacific Island, 9.0% (n = 317), and Asian, 11.8% (n = 414). The mean (SD) age of the men at the antenatal evaluation was 33.20 (6.25) years (range, 16-63 years).

Paternal ADS (PADS) were measured using the Edinburgh Postnatal Depression Scale (EPDS), a tool also validated for use in men. The EPDS has 10 items scored 0 to 3 (maximum score, 30), with a higher score indicating higher symptoms of depression.^41,42 In line with many studies,^14,16 participants with an EPDS score higher than 12 were said to have PADS corresponding to an indication of probable depressive disorder.⁴¹ Maternal ADS/PDS were also assessed, using the EPDS cutoff score of 12/13.

Paternal PDS (PPDS) were measured using the 9-item Patient Health Questionnaire (PHQ-9), a brief standardized depression tool⁴³ with excellent psychometric properties. Each PHQ-9 item is scored 0 to 3 (maximum score, 27); higher scores indicate more and/or greater-severity symptoms of depression.^44-46 Participants with a PHQ-9 score higher than 9 were said to have PPDS, corresponding to an indication of mild to severe depression.⁴⁷ The EPDS and PHQ-9 have good concordance (87% for these cutoff scores).⁴⁸

Self-reported paternal antenatal health variables included perceived general health during the partner’s pregnancy (5-point Likert scale, from excellent to poor, and from the 36-Item Short Form Health Survey⁴⁹), smoking, alcohol consumption, disability or current health problems, and whether they had ever been diagnosed as having depression or anxiety. Participants also completed the 10-item Perceived Stress Scale, with a higher score indicating higher levels of stress.⁵⁰ Warmth/Hostility Scale⁵¹ and Conflict scale⁵² scores were combined to create a relationship environment score (potential range, 15-105). Family Stress scale⁵³ and informal Family Support Scale⁵⁴ items were combined to create a family environment score (potential range, 12-60). An external environment score was calculated using the Neighborhood Integration scale⁵⁵ and external Family Support Scale⁵⁴ items (potential range, 16-86). Higher scores on these composites indicated a better environment.

At the postnatal interview some of these measures were repeated, including employment, socioeconomic status, health, and smoking, and the Warmth/Hostility, Conflict, Family Stress, Family Support, and Neighborhood Integration scales. Participants were asked whether they were still in a relationship with the mother of their Growing Up in New Zealand study child.

χ² Tests were used to determine univariate associations between dependent variables and categorical independent variables. Wilcoxon signed rank test for 2 related samples and Mann-Whitney tests examined associations between dependent variables and nonparametric continuous independent variables, and 2-tailed, independent-samples (unpaired) t tests were used for parametric continuous independent variables. Factors associated with PADS/PPDS at P < .01 were included in multivariable analyses using binary logistic regression. Statistical significance in regression analyses was given at P < .05.

The following analyses were carried out: (1) rates of PADS (EPDS score >12) and PPDS (PHQ-9 score > 9), (2) univariate associations between maternal depression symptoms and PADS/PPDS, (3) univariate associations between independent variables and PADS/PPDS (to determine which independent variables to include in multivariable analyses), (4) multivariable analyses of pregnancy and prepregnancy independent variables and PADS (to explore risks for PADS), (5) multivariable analyses of prepregnancy and pregnancy independent variables and PPDS (for comparison with our group’s report on maternal depression), and (6) multivariable analyses of prepregnancy, pregnancy, and postnatal independent variables and PPDS (to determine whether factors associated with PPDS change according to whether they occur before or after childbirth). Data analysis was performed from September 1, 2015, to January 8, 2016. Statistical analysis was conducted using SPSS, version 23.0 (IBM Corp).

Rates of PADS and PPDS were 2.3% (n = 82) and 4.3% (n = 153), respectively (Figure 1). PADS was associated with PPDS (χ² = 62.67; P < .001). During the perinatal period, 217 men (6.2%) experienced either PADS or PPDS, and 3306 (93.8%) did not have elevated depression symptoms. Eighteen (22.0%) men who experienced PADS also had PPDS, and 18 (11.8%) of those with PPDS had also experienced PADS.

Antenatally, 291 (8.3%) participants reported a clinical diagnosis of depression at some time in their life. Men with PADS were more likely to report a history of depression than were those without PADS (21 of 82 [25.6%] vs 270 of 3441 [7.8%]; χ² = 33.35; P < .001). Similarly, men with PPDS were more likely to report a history of depression than were those without PPDS (43 of 153 [28.1%] vs 248 of 3379 [7.4%]; χ² = 83.12; P < .001).

Men with PADS were more likely to have a partner with maternal ADS than were men without PADS (21 of 82 [25.6%] vs 284 of 3425 [8.3%]; χ² = 30.25; P < .001). Similarly, men with PPDS were more likely to have a partner with ADS (26 of 153 [17.0%] vs 279 of 3354 [8.3%]; χ² = 13.87; P < .001) or PDS (27 of 153 [17.6%] vs 207 of 3362 [6.2%]; χ² = 31.09; P < .001) than were those without PPDS.

Overall, male participants had lower mean (SD) total antenatal EPDS scores (3.31 [3.51]) than their pregnant partners (5.67 [4.59]; z = –25.47; P < .001). For participants who met the criteria for ADS, EPDS total scores were similar for men and their female partners (both 15.5 [2.5]). However, men without PADS had lower mean antenatal EPDS scores (2.92 [2.94]) compared with their female partners without ADS (4.70 [3.48] z = –22.13; P < .001).

Table 1 and Table 2 report the univariate associations between the independent variables and PADS and PPDS. Paternal ADS was significantly associated with the following paternal variables: perceived stress, relationship environment, family environment, ethnicity, self-reported diagnosis of a history of clinically diagnosed anxiety, health status, alcohol consumption during their partner’s pregnancy, relationship with the baby’s mother, and being born outside of New Zealand. There was a significant association for age, educational level, disability, current smoking, workforce status, and unplanned pregnancy. Paternal PDS was significantly associated with the following paternal variables: age; perceived stress score during the partner’s pregnancy; relationship, family, and external environments; ethnicity; educational level; self-reported diagnosis of a history of anxiety; disability; workforce status; socioeconomic status; smoking; health status; relationship status; and unplanned pregnancy.

The risk of PADS increased for men with higher perceived stress (odds ratio [OR], 1.38; 95% CI, 1.30-1.47) and fair to poor health (OR, 2.06; 95% CI, 1.18-3.61). Participants who reported drinking less than usual during their partner’s pregnancy were more likely to experience PADS (OR, 2.08; 95% CI, 1.04-4.18) (Tables 1 and 2).

The risk of PPDS increased for men with higher antenatal perceived stress (OR, 1.14; 95% CI, 1.10-1.18), who smoked during their partner’s pregnancy (OR, 1.72; 95% CI, 1.14-2.58), had fair to poor health antenatally (OR, 1.69; 95% CI, 1.13-2.54), or had a history of depression (OR, 2.80; 95% CI, 1.71-4.59). Taking these factors into account, men with PADS were not more likely to experience PPDS (Tables 1 and 2).

Antenatal variables that could have changed between the interviews (employment; relationship and health status; maternal ADS and PDS; deprivation; relationship, external, and family environments; and smoking) were replaced with postnatal measures. Tables 1 and 2 report that, when considering prepregnancy, antenatal, and postnatal factors, the risk of PPDS increased for men with higher antenatal perceived stress (OR, 1.12; 95% CI, 1.08-1.17); more difficult postnatal relationship environment (OR, 0.97; 95% CI, 0.95-0.99) or family environment (OR, 0.93; 95% CI, 0.91-0.96); those no longer in a relationship with their child’s mother (OR, 6.36; 95% CI, 2.28-17.78); who were unemployed at the postnatal interview (OR, 1.86; 95% CI, 1.11-3.10); smoked at both time points (OR, 1.66; 95% CI, 1.04-2.64); had fair to poor postnatal health (OR, 3.29; 95% CI, 2.10-5.16); or had a history of depression (OR, 2.84; 95% CI, 1.69-4.78). Figure 2 demonstrates the different risks for PADS and PPDS.

Our rates of PADS (2.3%) and PPDS (4.3%) were in line with other studies using the same measures.^14,56 Reported rates of maternal ADS and PDS in the Growing Up in New Zealand Study were higher at 11.5% and 8.0%, respectively.^5,6 Furthermore, 8.3% of the male participants reported a history of depression, compared with 17.9% of women.⁵ These findings are consistent with the current evidence base.^26,57-59

Antenatal EPDS scores for participants not reaching the depression cutoff score were lower for men than for women. It is possible that symptoms in men are not continuously distributed and only men at risk of developing PADS have mood difficulties during their partner’s pregnancy, while women more generally have mood difficulties. This suggests that only at-risk men would benefit from intervention, but all pregnant women (regardless of meeting the criteria for a clinical diagnosis) may benefit from support.

Our PADS rate of 2.3% is similar to estimated rates of depression in the general male population.¹ As such, pregnancy-specific depression symptoms may be of less concern for men than for women. However, our sample of men differed from the general population in that all men had a partner at the start of the study. Being married or cohabiting lowers the risk of depression; therefore, the general rate of depression among partnered men may be lower compared with the rate in the general male population.^60,61

We found higher rates of PPDS (4.3%) compared with PADS (2.3%). The rate of incident postnatal depression (PPDS but no PADS) was 3.8%. This difference may be attributable to being a new father, but it may be that the PHQ-9 had greater sensitivity in our sample compared with the EPDS. Using different scales limited our ability to fully explore longitudinal risks for incident PPDS or the depression symptoms in both pregnancy and after childbirth. As such, it is not clear whether the risks for PPDS that we identified apply to incident PPDS or preexisting symptoms.

In previous studies of mothers, Pacific or Asian ethnicity, stress, anxiety, a history of depression, and difficult relationship and family environments increased the risk of depression symptoms.⁶ In the present study, paternal perceived stress in pregnancy and health status were consistently associated with paternal depression symptoms. Other risks for PPDS included a history of depression, postnatal unemployment and relationship status, and postnatal relationship and family environments.

Figure 2 shows the risks for maternal and paternal perinatal depression identified by this study and previous reports from the Growing Up in New Zealand study. These risks reflect existing evidence^3,26,62 supporting the idea that men and women have different risks in each perinatal period. This finding has implications for the timing and targeting of screening and intervention. Expectant fathers should seek support if they experience unemployment and/or relationship or family difficulties following the birth of their child. Men who have a history of mental health problems or who are stressed or unwell during their partner’s pregnancy should be assessed for PPDS. Men should also be encouraged to give up smoking during their partner’s pregnancy and continue not to smoke after the birth of their child.

Although associated at the univariate level, maternal ADS and PDS were not a risk for PADS and PPDS in multivariable analyses; this finding differs from other studies.^{3,12,16,26,56} Perhaps the quality of the maternal-paternal relationship (included in our study) is a more important risk for paternal depression than maternal depression per se.²¹ We included men who were no longer in a relationship with their child’s mother at 9 months. Only including couples who are still partnered when PDS are measured may affect other study results on the association between maternal and paternal depression. Our divergent finding may reflect the ethnic makeup of other studies; many were predominantly European or gave little ethnicity information.^{3,12,16,26,56} We found that Pacific ethnicity was a strong indicator of maternal ADS and PDS.^5,6 It is possible that Pacific Island culture protects men against the high levels of depression symptoms among their Pacific partners. Fathers in the Growing Up in New Zealand study were more likely to identify as Māori or Pacific Islander compared with study mothers. This difference may have been due to cultural differences in willingness to take part, more Māori and Pacific Island men having partners of different ethnicities, or men with multiple ethnicities being more likely to prioritize Māori or Pacific Island as their main ethnicity and it may have affected our findings.

Reflecting the worldwide situation, New Zealand women are assessed for PDS in the first 6 weeks after childbirth. However, there is no maternal screening during pregnancy and no perinatal screening for men. Although the rates of PADS and PPDS are lower than in mothers, some fathers struggle in the perinatal period. In our study, the postnatal period was a time of higher risk compared with the antenatal period. Paternal perinatal depression was not more prevalent in any one ethnic group and was experienced by those with a range of educational level backgrounds, suggesting that screening needs to be targeted at all fathers.

Discussing the risks of depression with expectant mothers and fathers would provide information about where to seek help and social support should one of them develop symptoms. This communication could be the only way to reach fathers who are no longer with the mothers postnatally. Our studies suggest that family- and couple-based interventions that focus on improving relationship quality and alleviating stress in both men and women may benefit those at risk of developing ADS or PDS.

Follow-up diagnostic assessment of participants with PADS or PPDS was not conducted and we could not determine whether participants were receiving treatment for depression. The data on prepregnancy and early pregnancy variables may have been affected by recall bias. Our results may not be generalizable to the first and second trimesters of pregnancy or to the period immediately following childbirth; these factors may have affected our findings on links between maternal and paternal depression.²¹

It was difficult to make direct comparisons of PADS and PPDS because we used different measures in each perinatal period. More factors were associated with PPDS than with PADS, but it is unclear whether this is because there are different risks for men in each period or because different measures were used. The 7.8% of participants who were not assessed postnatally were significantly different from those who completed both interviews, suggesting some selection bias.

The main strength of the study was the large sample, high participation rates, low attrition, and good representation of minority ethnic groups and those with low socioeconomic status.^39,63,64 Previous longitudinal studies have been smaller and did not assess PADS.^26,65 The Avon Longitudinal Study of Parents and Children (N = 7018) is more comparable to ours, although participants were assessed at different time points.³³

Only relatively recently has the influence of fathers on children been recognized as vital for adaptive psychosocial and cognitive development. Given that paternal depression can have direct or indirect effects on children, it is important to recognize and treat symptoms among fathers early, and the first step in doing that is arguably increasing awareness among fathers about increased risks.

Accepted for Publication: December 11, 2017.

Corresponding Author: Lisa Underwood, PhD, Centre for Longitudinal Research–He Ara ki Mua, School of Population Health, University of Auckland, PO Box 18288, Auckland 1743, New Zealand (l.underwood@auckland.ac.nz).

Published Online: February 15, 2017. doi:10.1001/jamapsychiatry.2016.4234

Author Contributions: Dr Underwood had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Underwood, Waldie, Morton.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Underwood, Peterson, Verbiest, McDaid, Morton.

Critical revision of the manuscript for important intellectual content: Underwood, Waldie, D’Souza, Verbiest, Morton.

Statistical analysis: Underwood, D’Souza, Verbiest.

Obtained funding: Waldie, Morton.

Administrative, technical, or material support: D’Souza.

Study supervision: Waldie, Peterson, Morton.

Conflict of Interest Disclosures: None reported.

Funding/Support: The Growing Up in New Zealand study has been funded by the New Zealand Ministries of Social Development, Health, Education, Justice, and Pacific Island Affairs; the former Ministry of Science Innovation and the former Department of Labour (both now part of the Ministry of Business, Innovation, and Employment); the former Ministry of Women’s Affairs (now the Ministry for Women); the Department of Corrections; the Families Commission (now the Social Policy Evaluation and Research Unit); Te Puni Kokiri;New Zealand Police; Sport New Zealand; the Housing New Zealand Corporation; and the former Mental Health Commission, University of Auckland, and Auckland UniServices Limited. Other support for the study has been provided by the New Zealand Health Research Council, Statistics New Zealand, the Office of the Children’s Commissioner, and the Office of Ethnic Affairs.

Role of the Funder/Sponsor: These funders had a role in the design and conduct of the study via a formal policy forum. The funders had no role in the collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; or the decision to submit the manuscript for publication.

Disclaimer: The views reported in this article are those of the authors and do not necessarily represent the views of the Growing Up in New Zealand study investigators.

Additional Contributions: The antenatal and 9-month data collection waves of the Growing Up in New Zealand study were designed and conducted by the Growing Up in New Zealand study team led by the University of Auckland. We acknowledge the contributions of the original study investigators.