Child abuse is an important psychological, educational, and legal issue worldwide. There is ample evidence suggesting that past experiences of neglect and child physical, psychological, and sexual abuse increase the risk of a range of psychiatric disorders, substance use disorders, suicide attempts, sexually transmitted infections, and risky sexual behaviour in later life [1Norman RE, Byambaa MDeR, Butchart A, Scott J, Vos T. The long-term health consequence of child physical abuse, emotional abuse, and neglect: A systematic review and meta-analysis. PLoS Med 2012; 9: e1001349.
[http://dx.doi.org/10.1371/journal.pmed.1001349] [PMID: 23209385] ]. The annual incidence of infant abuse reported to the National Child Abuse and Neglect Data System in the United States was 23.2 cases per 1,000 [2Brodowski M, Nolan CM, Gaudiosi JA, et al. Nonfatal maltreatment of infants: United States, October, 2005 - September 2006. Mortal Morbidity Wkly Rep 2008; 57: 536-339.]. In 39% of cases the infants were less than 1 month old, and about 80% of instances occurred within 1 week after childbirth. More than half of the reported cases were of neglect. Hence, neonates and infants are at particular risk of child abuse. In Japan, however, despite the clinical importance of infant and neonatal abuse, virtually no empirical studies have been conducted on abuse of infants aged less than 1 month.

Exploration of the factors affecting the risk of abusive parenting is a prerequisite for implementing effective preventive strategies. Previous studies revealed that psychological factors were related to abusive parenting style. Such factors include parenting stress [3Peltonen K, Ellonen N, Pösö T, Lucas S. Mothers self-reported violence toward their children: a multifaceted risk analysis. Child Abuse Negl 2014; 38(12): 1923-33.
[http://dx.doi.org/10.1016/j.chiabu.2014.10.016] [PMID: 25459986] , 4Ishii Y, Maruyarna A, Anme T, McCall ME. Maternal stress, parenting factors, experiences in day care, and developmental outcomes in 5-year-old children in day care in Japan. Jpn J Hum Sci Health-Social Serv 2011; 18: 148-59.], undesired pregnancy [5Baydar N. Consequences for children of their birth planning status. Fam Plann Perspect 1995; 27(6): 228-234, 245.
[http://dx.doi.org/10.2307/2136174] [PMID: 8666086] , 6Barber JS, Axinn WG, Thornton A. Unwanted childbearing, health, and mother-child relationships. J Health Soc Behav 1999; 40(3): 231-57.
[http://dx.doi.org/10.2307/2676350] [PMID: 10513146] ], bonding impairment [7Brockington I. Maternal rejection of the young child: present status of the clinical syndrome. Psychopathology 2011; 44(5): 329-36.
[http://dx.doi.org/10.1159/000325058] [PMID: 21734436] ], and parental psychopathology [8De Bellis MD, Broussard ER, Herring DJ, Wexler S, Moritz G, Benitez JG. Psychiatric co-morbidity in caregivers and children involved in maltreatment: a pilot research study with policy implications. Child Abuse Negl 2001; 25(7): 923-44.
[http://dx.doi.org/10.1016/S0145-2134(01)00247-2] [PMID: 11523869] -10Walsh C, MacMillan H, Jamieson E. The relationship between parental psychiatric disorder and child physical and sexual abuse: findings from the Ontario Health Supplement. Child Abuse Negl 2002; 26(1): 11-22.
[http://dx.doi.org/10.1016/S0145-2134(01)00308-8] [PMID: 11860159] ] and personality disorders [11Smith CL, Spinrad TL, Eisenberg N, Gaertner BM, Popp TK, Maxon E. Maternal personality: Longitudinal Associations to parenting behavior and maternal emotional expressions toward toddlers. Parent Sci Pract 2007; 7(3): 305-29.
[http://dx.doi.org/10.1080/15295190701498710] [PMID: 18174914] , 12Laulik S, Chou S, Browne KD, Allam J. The link between personality disorder and parenting behaviors: A systematic review. Aggress Violent Behav 2013; 18: 644-55.
[http://dx.doi.org/10.1016/j.avb.2013.07.017] ]. There are at least three important correlates of neonatal/infant abuse, the first being maternal depression. There is ample evidence that mothers with depression are more likely to show maladaptive parenting styles [9Shay NL, Knutson JF. Maternal depression and trait anger as risk factors for escalated physical discipline. Child Maltreat 2008; 13(1): 39-49.
[http://dx.doi.org/10.1177/1077559507310611] [PMID: 18174347] , 13Brennan PA, Le Brocque R, Hammen C. Maternal depression, parent-child relationships, and resilient outcomes in adolescence. J Am Acad Child Adolesc Psychiatry 2003; 42(12): 1469-77.
[http://dx.doi.org/10.1097/00004583-200312000-00014] [PMID: 14627882] -27Zajicek-Farber ML. Postnatal depression and infant health practices among high-risk women. J Child Fam Stud 2009; 18: 236-45.
[http://dx.doi.org/10.1007/s10826-008-9224-z] ]. Maternal child abuse may be correlated with maternal depression. For example, De Bellis et al. compared caregivers of maltreated and non-maltreated children or adolescents [8De Bellis MD, Broussard ER, Herring DJ, Wexler S, Moritz G, Benitez JG. Psychiatric co-morbidity in caregivers and children involved in maltreatment: a pilot research study with policy implications. Child Abuse Negl 2001; 25(7): 923-44.
[http://dx.doi.org/10.1016/S0145-2134(01)00247-2] [PMID: 11523869] ]. The lifetime prevalence of maternal major depression was significantly higher when offspring were maltreated (72%) than when they were not (22%). In a large community study, Walsh, MacMillan, & Jamieson reported that physical abuse had occurred in 41% of respondents with at least one parent with a history of depression, compared to only 23% of respondents with parents who were not depressed [10Walsh C, MacMillan H, Jamieson E. The relationship between parental psychiatric disorder and child physical and sexual abuse: findings from the Ontario Health Supplement. Child Abuse Negl 2002; 26(1): 11-22.
[http://dx.doi.org/10.1016/S0145-2134(01)00308-8] [PMID: 11860159] ]. The proportion of childhood sexual abuse showed a similar trend: 18% of respondents reporting a history of parental depression were sexually abused, as opposed to 6% of respondents with no history of parental depression. In a longitudinal study of newborns followed up until age three years, Windham et al. found a correlation between maternal depression and severe child physical abuse and assault on the child’s self-esteem [28Windham AM, Rosenberg L, Fuddy L, McFarlane E, Sia C, Duggan AK. Risk of mother-reported child abuse in the first 3 years of life. Child Abuse Negl 2004; 28(6): 645-67.
[http://dx.doi.org/10.1016/j.chiabu.2004.01.003] [PMID: 15193853] ]. These previous studies all indicated substantial association between the risk of child abuse and maternal depression, but all used correlational statistical analyses. Hence, it is unknown whether maternal depression leads to child abuse or vice versa, or if the two are confounded by one or more other variables. Longitudinal follow-up studies incorporating causal path models are still required.

The second important correlate of abusive parenting is bonding disorder. Mothers of newborns may experience aversive feelings in their early relationships with their infants. These include lack of affection or hatred towards the babies, impulses to harm them, or rejection or neglect [7Brockington I. Maternal rejection of the young child: present status of the clinical syndrome. Psychopathology 2011; 44(5): 329-36.
[http://dx.doi.org/10.1159/000325058] [PMID: 21734436] , 29Brockington I. Diagnosis and management of post-partum disorders: a review. World Psychiatry 2004; 3(2): 89-95.
[PMID: 16633463] -31Kumar RC. Anybodys child: severe disorders of mother-to-infant bonding. Br J Psychiatry 1997; 171: 175-81.
[http://dx.doi.org/10.1192/bjp.171.2.175] [PMID: 9337956] ]. About a quarter of mothers referred to psychiatrists for child abuse had disorders of the mother–infant relationship [32Brockington IF. Motherhood and mental Health. Oxford: Oxford University Press 1996; pp. 327-66.]. When bonding disorder is severe, mothers may try to escape or to seek permanent transfer of infant care within or outside of the family. Bonding difficulty, and in particular its subscale Anger and Rejection, is associated with trait anger [33Kitamura T, Takegata M, Haruna M, et al. The Mother-Infant Bonding Scale: Factor structure and psychosocial correlates of parental bonding disorders in Japan. J Child Fam Stud 2015; 24: 393-401.
[http://dx.doi.org/10.1007/s10826-013-9849-4] ]. In a study of a Japanese community population of mothers of 3-month-olds, Kitamura, Ohashi, Kita, Haruna, & Kubo revealed that an abusive parenting pattern was associated, albeit weakly, with maternal depression and bonding difficulty [34Kitamura T, Ohashi Y, Kita S, Haruna M, Kubo R. Depressive mood, bonding failure, and abusive parenting among mothers with three-month-old babies in a Japanese community. Open J Psychiatr 2013; 3: 1-7.
[http://dx.doi.org/10.4236/ojpsych.2013.33A001] ]. In a non-recursive structural equation model, they revealed that an abusive parenting pattern was predicted by the other two variables. However, this again was a cross-sectional study.

The third potential factor affecting abusive parenting after childbirth is unwanted pregnancy, though there is minimal empirical evidence regarding this relationship. Nevertheless, there is indirect evidence suggesting that unwanted pregnancy has an adverse effect on mothers’ parenting styles. Postnatal depression is more likely in women with unintended pregnancies at both 3 and 12 months after childbirth [35Mercier RJ, Garrett J, Thorp J, Siega-Riz AM. Pregnancy intention and postpartum depression: secondary data analysis from a prospective cohort. BJOG 2013; 120(9): 1116-22.
[http://dx.doi.org/10.1111/1471-0528.12255] [PMID: 23651010] ]. Kokubu, Okano, & Sugiyama reported that negative attitudes towards pregnancy predicted bonding difficulties within the 5 days after delivery in a Japanese non-clinical maternal population [36Kokubu M, Okano T, Sugiyama T. Postnatal depression, maternal bonding failure, and negative attitudes towards pregnancy: a longitudinal study of pregnant women in Japan. Arch Women Ment Health 2012; 15(3): 211-6.
[http://dx.doi.org/10.1007/s00737-012-0279-x] [PMID: 22526407] ]. Hence we were interested in whether negative attitudes towards pregnancy in expectant and new mothers are associated with both bonding disorder and depression after childbirth, and whether negative attitudes towards pregnancy lead to foetal/neonatal abuse. This latter effect, if it exists, may be mediated via bonding disorder and depression after childbirth.

In this study, we longitudinally investigated a Japanese nonclinical maternal population with a focus on how emotional abuse towards infants was influenced by depression, bonding disorder, and the women’s response to pregnancy. We hypothesized that negative maternal attitudes towards pregnancy would increase the risk of emotional abuse of neonates and that this increased risk would be mediated by maternal depression and bonding disorder during and after pregnancy. Furthermore, we hypothesized that maternal depression and bonding disorder at one time point in the perinatal period would predict each other at a later time point.

A main finding of this study was that emotionally abusive parenting 1 month after childbirth was predicted by bonding disorder but not depressive symptoms at 5 days after childbirth. In accordance with previous investigations, a bivariate correlation matrix showed that emotional abuse and depression were significantly correlated with each other. However, our SEM revealed that this link was spurious.

These findings have public policy implications. In Japan, “Sukoyaka Oyako 21 [Healthy Parents and Children 21]”, an early 21^st century national plan for the health of mothers and children, has as its goals the healthy development of children and a reduction of parental child-raising anxiety. This policy dictates that local governments should dispatch home visitors to each mother of a newborn baby (Whole Infant Family Visit) to provide physical as well as psychological support to the dyad. It has been recommended that the EPDS be used to screen for postnatal depression because the condition is believed to be a risk factor for infant abuse [51Ministry of Health Lobour and welfare. Sukoyaka Oyako 21 dai 2 kai tyukan hyouka torimatome ni tuite [The 2nd interim evaluation report of Sukoyaka Oyako 21 Plan. Tokyo, Japan. Available from: http://www.mhlw.go.jp/shingi/2010/02/dl/s0205-13a.pdf 2009., 52Ministry of Health Lobour and welfare. Nyu-youji zenko houmon guidline [The guideline of the Whole Visits for infants-toddler]. Tokyo, Japan 2015. Available from: http://www.mhlw.go.jp/bunya/kodomo/kosodate12/03.html ]. However, this study showed that depression did not have a direct effect on neonatal emotional abusive parenting. Just focusing on postnatal depression may be insufficient approach in preventing neonatal or child abuse. Hence we propose that in clinical and community settings, more focus should be placed on bonding disorder especially as a risk factor for emotional abusive parenting. Unfortunately, the assessment of maternal bonding and bonding disorder in the community postnatal care system has been scarcely done. In addition to the use of the EPDS, the use of instruments such as the PBQ by health visitors or midwives may be a possibility. Further, we previously showed that mothers with postnatal depression who did not seek medical support scored significantly higher in bonding failure than those who did [53Kitamura T, Yamashita H, Yoshida K. Seeking medical support for depression after the childbirth: A Study of Japanese community mothers of 3-month-old babies. Open Wom Health J 2009; 3: 1-14.
[http://dx.doi.org/10.2174/1874291200903010001] ]. The more the government announces the use of depression screening as a means to identify mothers at risk of infant abuse, the less depressed mothers of infants will seek professional help with fear that they may be viewed as a “abusive mother. Therefore, policy makers should be very careful to distinguish between child abuse prevention and promotion of mothers’ general mental health. Identification of mother with depression should be clearly separated from identifying mother at risk of child abuse, however, it is of great importance. Care programs for women identified as with bonding disorders are another very important clinical issue [54McGregor J, Casey J. Enhancing parent-infant bonding using kangaroo care: A structured review. Evidence Based Midwifery 2012; 10: 50-6.].

This longitudinal study revealed longitudinal associations between a negative response towards the present pregnancy, depression, and bonding disorder during the perinatal period, and emotional abuse at 1 month after childbirth. We found that both depression and bonding were moderately to strongly predicted by scores at earlier time points. These results mean that depression and bonding disorder occur over the entire course of the perinatal period, not just during the postnatal period, and are consistent with the findings of previous longitudinal studies [36Kokubu M, Okano T, Sugiyama T. Postnatal depression, maternal bonding failure, and negative attitudes towards pregnancy: a longitudinal study of pregnant women in Japan. Arch Women Ment Health 2012; 15(3): 211-6.
[http://dx.doi.org/10.1007/s00737-012-0279-x] [PMID: 22526407] , 55Yoshida K, Yamashita H, Conroy S, Marks M, Kumar C. A Japanese version of Mother-to-Infant Bonding Scale: factor structure, longitudinal changes and links with maternal mood during the early postnatal period in Japanese mothers. Arch Women Ment Health 2012; 15(5): 343-52.
[http://dx.doi.org/10.1007/s00737-012-0291-1] [PMID: 22733162] ]. Our results also indicate that both depression and bonding disorder during pregnancy are linked to the woman’s negative attitude towards her pregnancy when she learnt that she was pregnant, confirming the result of a previous study [36Kokubu M, Okano T, Sugiyama T. Postnatal depression, maternal bonding failure, and negative attitudes towards pregnancy: a longitudinal study of pregnant women in Japan. Arch Women Ment Health 2012; 15(3): 211-6.
[http://dx.doi.org/10.1007/s00737-012-0279-x] [PMID: 22526407] ]. This suggests that perinatal health professionals should pay more attention to women’s affection and attitudes towards both the pregnancy and the foetus, beginning at the onset of pregnancy. In Japan, almost all women receive periodic medical check-ups during pregnancy, enabling adequate observation by health professionals. When pregnancy is initially confirmed, obstetricians, midwives, and nurses have ample opportunity to evaluate and support women’s desire for pregnancy and child rearing, including their motivations for raising a child, and to identify any potential ambivalence [56Brückner H, Martin A, Bearman PS. Ambivalence and pregnancy: adolescents attitudes, contraceptive use and pregnancy. Perspect Sex Reprod Health 2004; 36(6): 248-57.
[http://dx.doi.org/10.1363/3624804] [PMID: 15687083] -59Söderberg M, Lundgren I, Christensson K, Hildingsson I. Attitudes toward fertility and childbearing scale: an assessment of a new instrument for women who are not yet mothers in Sweden. BMC Pregnancy Childbirth 2013; 13: 197.
[http://dx.doi.org/10.1186/1471-2393-13-197] [PMID: 24165014] ]. This type of clinical service is not a routine practice in Japan. The pregnant woman’s affection towards her foetus is another very important topic that perinatal health professionals should inquire about when providing psychological support to this population. If the above issues are not addressed by health care providers in common clinical settings, many pregnant women may miss the chance to express negative attitudes towards the current pregnancy and foetus.

There was a significant covariance between depressive moods and bonding disorder at both day 5 and 1 month after childbirth. In addition, there were significant covariances between depressive mood, bonding disorder, and abusive parenting at 1 month after childbirth. We speculate that these three variables often coexist. Our SEM analysis, however, revealed that depressive mood and bonding disorder were causally independent. This analysis suggests the presence of confounders that explain the coexistence of depression, bonding disorder, and abusive parenting. Such unidentified confounding factors were not evaluated in this study, however, and should be identified in future research. They may include the women’s experience of being reared by their own parents when they were children, attitudes towards fertility, and personality, to list just a few.

One of the strengths of this study is its longitudinal design, which enabled us to disentangle causal links between variables. When studies are cross-sectional, the results are correlational and thus determining causality is almost impossible.

Nevertheless, our investigation was not free from drawbacks. Our model could explain only about 5% of the variance in abusive parenting. We failed to identify the third “culprit” factor (other than women’s negative attitudes towards the pregnancy) that explains the correlation between depression, bonding failure, and abusive parenting. There may be important factors that were not included in this study. Future investigations should include a greater variety of information, such as the women’s personality traits, psychopathology, substance abuse, histories of abuse, characteristics of relationships with significant others, socioeconomic problems, and the support of intimate partners.

In this study, we focused only on neonatal emotional abuse. The very restricted children’s age range may require that caution be exercised when interpreting the results. The results of this study cannot be said to relate to any period outside of the neonatal period or to any kind of abuse outside of emotional abuse as assessed with the CTS. The relationship between maternal bonding difficulties and abusive parenting may differ in children of different age ranges.

Another major limitation of this study is the questionable validity of self-report of maternal abusive behavior in this population. Future studies should incorporate information gathered from the other resources such as spouses and family members and child protection agencies. Another weakness of the present study is lack of validation of scales used in the study such as the MAAS.

Taking these drawbacks into consideration, the present study provides preliminary evidence that 5% of variance of emotionally abusive parenting in the postnatal period was predicted by bonding disorder during the early days after childbirth. Perinatal depression failed to predict neonatal abuse at 1 month after childbirth. Women’s negative attitudes towards the present pregnancy are likely to increase the risk of emotionally abusive parenting via bonding failure.

Bonding failure may be a risk factor for neonatal emotional abuse. Negative feelings about the current pregnancy are another risk factor for parenting after birth, via bonding failure. We should pay more attention to women’s prenatal attitudes towards their pregnancy and foetus, a practice that is not routinely performed in Japan.