Prenatal depression predicts postpartum maternal attachment in low-income Latina mothers with infants
Highlights
► Predictors of postpartum attachment in Latina immigrant mothers with infants were examined. ► Predictors included: depressive symptoms, pregnancy intention, feelings about the pregnancy, and the quality of the partner relationship. ► Lower attachment was predicted by depressive symptoms late in pregnancy; pregnancy intention was marginally predictive of attachment, with lower scores being associated with unwanted pregnancies. ► Study fills a gap in our understanding of the role of depressive symptoms during pregnancy in shaping mothers’ early attachment to their infants.
Introduction
Forming a secure attachment to a caregiver during infancy is a key step in social and emotional development and an important predictor of subsequent relationship patterns (Berlin et al., 2005, Bowlby, 1969). Early infancy is a critical time in maternal attachment formation, and disturbances in maternal–infant bonding within the first few months of an infant's life have been shown to affect later infant attachment (Cassidy et al., 2005, Klier, 2006). Insecure infant attachment is related to a variety of physical, emotional, social and learning problems throughout the child's lifetime (Carlson et al., 2003, Zeanah and Smyke, 2005). Identifying early antecedents of risk for insecure attachment could help target preventive efforts (Greenberg, 2005). The purpose of this paper is to examine the predictors of early infant attachment in a low-income Latina sample of predominantly immigrant women.
Ainsworth, Blehar, Waters, and Wall (1978) research on attachment using the Strange Situation paradigm, an observational method of assessing infant attachment behaviors beginning around 12 months, laid the foundation for an extensive empirical literature on infant attachment styles (Karen, 1994, O’Connor and Byrne, 2007). However, much less is known about the precursors of attachment in the prenatal period and early infancy. The “building blocks” of attachment may originate with a woman's responses to learning that she is pregnant, as expressed through maternal–fetal attachment and emotional involvement with the fetus during pregnancy (Alhusen, 2008, Figueiredo and Costa, 2009). Maternal–fetal attachment has been defined as “the extent to which women engage in behaviors that represent an affiliation and interaction with their unborn child” (Cranley, 1981, p. 282) and includes cognitive, affective, and altruistic attributes (Shieh, Kravitz, & Wang, 2001). Maternal emotional involvement with the unborn child generally grows throughout pregnancy and helps prepare women for the transition to motherhood (Figueiredo & Costa, 2009).
Attachment is a bidirectional and transactional process, resulting from both maternal and infant feelings and behaviors. Feldman, Weller, Leckman, Kuint, and Eidelman (1999) highlighted that maternal mental states, thoughts, feelings, worries, hopes, and representations of the infant unique to the bonding period are just as critical for understanding maternal attachment as specific attachment behaviors. While most studies have focused on infant-to-mother attachment behaviors rather than mother-to-infant attachment cognitions and behaviors, more studies are beginning to examine maternal attachment both prenatally (Alhusen, 2008, Figueiredo and Costa, 2009) and postnatally (Condon, Corkindale, & Boyce, 2008).
In addition, parental feelings of attachment may be an important predictor of later infant attachment behavior. A study of infants between 11 and 15 months showed that the postpartum attachment scores of both mothers and fathers were significantly related to the Attachment Q-Sort, an observational measure of infant attachment behavior (Feldstein, Hane, Morrison, & Huang, 2004). However, additional work is necessary to understand maternal experiences of attachment and to identify predictors of maternal attachment behaviors.
Disturbances in maternal–infant bonding within the first few months may be due to a number of factors, some of which occur even before a baby is born, including depression during pregnancy, whether the pregnancy is intended, and the quality of the partner/marital relationship. The impact of these factors in early maternal attachment is described in more detail below.
One of the most widely studied predictors of attachment is maternal depression (Field, 2010). The early disruptions in parenting that often accompany maternal depression may be an important mechanism that leads to impaired attachment in later infancy (Center on the Developing Child at Harvard University, 2009). Maternal depression has been positively associated with insecure attachment (Martins & Gaffan, 2000), especially when maternal depression is chronic and/or severe (Lyons-Ruth, Lyubchik, Wolfe, & Bronfman, 2002). Depression may undermine a mother's ability to interact in a responsive and sensitive manner with her baby, which may disrupt the development of secure attachments (Teti, Gelfand, Messinger, & Isabella, 1995). Specifically, research shows depressed mothers have two distinct negative interaction patterns with infants: intrusive, overstimulating behaviors and withdrawn, understimulating behaviors (Malphurs, Raag, Field, Pickens, & Pelaez-Nogueras, 1996).
While many studies have examined the effect of maternal depression on infant attachment, fewer studies have looked at its effect on maternal emotional attachment. One study of 112 middle-class primiparous women found that mothers with chronic depression (lasting 12 months or longer) reported increased hostility and anger toward their infants on the Postnatal Attachment Questionnaire than did briefly depressed or never depressed mothers (Cornish et al., 2006). However, to our knowledge, no study has examined the relation of depression during pregnancy and maternal postpartum attachment in Latina mothers.
Additional factors may contribute to maternal attachment besides maternal depression, including pregnancy intendedness. According to the National Survey of Family Growth, pregnancy intendedness refers to a pregnancy that is wanted at the time of conception or sooner, as well as whether or not contraception was being used (Chandra, Martinez, Mosher, Abma, & Jones, 2005). A study based on the National Longitudinal Survey of Youth found that unintended toddlers were more fearful and less positive compared to intended children, although unintended and intended children did not differ significantly in attachment (Baydar, 1995). Unintended pregnancies have also been associated with negative parenting interactions and childrearing practices (Gipson, Koenig, & Hindin, 2008). In their study of 140 Japanese women, Goto, Yasumura, Yabe, Anazawa, and Hashimoto (2005) found that mothers of unintended children were more likely to report parenting difficulty and feeling that they were abusing their child. Pregnancy intention may be associated with physical abuse (Gipson et al., 2008), and children who are abused or maltreated are more likely to be insecurely attached (Carlson et al., 2003).
Partner relationship satisfaction is another potential predictor of postpartum maternal attachment. A woman's relationship and satisfaction with her partner can change during the perinatal period, with possible implications for maternal attachment. High marital quality (Howes & Markman, 1989) and relationship satisfaction (Frosch, Mangelsdorf, & McHale, 2000) have been associated with improved security in child attachments. Results from a longitudinal study of 20 families with a toddler indicated that marital quality both before and after birth were related to child attachment security and functioning (Howes & Markman, 1989). In another longitudinal study of 57 families, marital conflict and interparental hostility predicted less secure preschooler–mother attachments, while positive marital engagement was associated with more secure child–father attachments (Frosch et al., 2000).
Most studies on predictors of maternal attachment, as reviewed above, have been conducted with middle-class European American samples. Given the vast changes in demographics in the U.S., a broader range of socio-demographic and ethnic groups should be included in this research. Specifically, there is a need for research on maternal attachment among Latinas. Latinos are the fastest growing population in the U.S. (Passel and Cohn, 2008, U.S. Census Bureau, 2007). Latinas typically have larger families and higher fertility rates than non-Latina European-American women (U.S. Department of Health and Human Services, Health Resources and Services Administration, 2001). In 2007, Latinas comprised 14% of the adult female population in the United States but accounted for 25% of all births to women aged 15–44 (Gonzales, 2008).
Cross-cultural research on attachment has shown that culture plays a significant role in parenting behaviors and infant attachment (van Ijzendoorn & Sagi-Schwartz, 2008). Preliminary research suggests that infant attachment may manifest itself differently in Latino populations compared with European-American populations (Zayas & Solari, 1994). For example, studies have found that the percentage of securely attached infants is slightly lower for Latina infants than for European-American infants (Fracasso, Busch-Rossnagel, & Fisher, 1994), particularly among recent immigrants from Central America and Mexico (Lieberman, Weston, & Pawl, 1991). Maternal behaviors associated with secure attachments may also differ in Latina populations compared to European-American populations. Among low-income Puerto Rican and Dominican mother–infant pairs, more frequent parental involvement and instruction with children was associated with increased secure attachment, contrasting with findings from European-American populations in which such parental interventions are viewed as intrusive and associated with insecure child attachment (Zayas & Solari, 1994).
Cultural values may also influence attachment. For example, one study found that both European-American and Puerto Rican mothers rated securely attached children as more agreeable than insecurely attached children, but European-American mothers valued the child's self-confidence and independence, whereas Puerto Rican mothers valued the child's obedience (Harwood & Miller, 1991). This research demonstrates the importance of considering cultural context when examining maternal behaviors and infant attachment.
In addition, predictors of attachment, such as pregnancy intendedness and relationship satisfaction, may differ somewhat in Latina as compared with European-American populations. Recent findings suggest length of time in the U.S. is an important predictor of Hispanic immigrant women's pregnancy intendedness, with very recent immigrants the most likely to report a birth was mistimed (Christensen, Perry, Le, & Ahmed, 2010). The potentially high rates of unintended pregnancy in this population may play an important role in the development of maternal attachment. There is also some evidence that partner satisfaction may vary between Anglo and Latino populations. Mexican American couples report higher marital quality and stability compared with African American couples, despite similar rates of socioeconomic disadvantage (Bulanda & Brown, 2007). The cultural value of familism, which emphasizes the family's key role in Latino culture, including family interconnectedness, support, commitment, and intergenerational cohesion (Steidel & Contreras, 2003), may enhance the importance of marriage and family among Latino populations (Bulanda and Brown, 2007, Contreras et al., 1996).
To date, we are not aware of any published, peer-reviewed studies examining predictors of early infant attachment among Latinas. The current study addressed this gap in the attachment literature by exploring the influence of maternal depressive symptoms, pregnancy intendedness, and partner satisfaction on early maternal attachment in a Latina sample. Participants were low-income Latinas enrolled in a preventive intervention designed to reduce perinatal depression (see Le, Perry, & Stuart, 2011). Using a longitudinal research design, we tested the following hypotheses: Hypothesis 1 Intervention effects: Women in the intervention group will have higher self-reported attachment to their newborns at 2 months postpartum compared to women in the usual care group. Hypothesis 2 Predictors of postpartum attachment: Hypothesis 2a Depression during pregnancy: Women with more depressive symptoms during pregnancy will report lower maternal attachment to their infants at 2 months postpartum than women with fewer depressive symptoms. Hypothesis 2b Pregnancy intention and feelings about pregnancy: Positive feelings about pregnancy and having an intended pregnancy will each be associated with higher maternal attachment at 2 months postpartum. Hypothesis 2c Relationship satisfaction: Higher satisfaction with one's partner during pregnancy will be related to higher maternal attachment at 2 months postpartum. Hypothesis 3 Multivariate analyses: Depressive symptoms during pregnancy will be the strongest predictor of maternal attachment at 2 months postpartum in a multi-variate regression model.
Section snippets
Participants
The study sample was recruited for a randomized, controlled trial conducted with low-income, predominantly immigrant, pregnant Latina women at high risk for depression. That trial, The Mothers and Babies (MB): Mood and Health Project, aimed to reduce the number of new cases of major depression through the use of a group cognitive-behavioral intervention course. The intervention was intended to teach women mood management skills to help alleviate depressive symptoms. While positive attachment
Description of the sample
As shown in Table 1, the women in the sample had a mean age of 25.4 years (SD = 4.6) and an average of less than 9 years of education (SD = 4.0). The majority of participants were married and living with a partner (64%), while 36% were single. Fifty-nine percent of participants’ partners were employed, and 90% of the households had an annual income under $30,000. Average parity for the women was 1 (SD = 1.1); 42% were primiparas, 32% had two children, and the remainder had three or more children.
Discussion
This study examined predictors of early attachment in a sample of low-income, predominantly immigrant Latina women at high risk for depression. Prenatal depressive symptoms, feelings about pregnancy and pregnancy intendedness, and partner satisfaction were assessed separately and in combination as possible predictors of maternal attachment. Prenatal depressive symptoms in the latter part of pregnancy emerged both in bivariate and multivariate analyses as the strongest predictor of early
Limitations and strengths of the current study
All of the measures used in this study are maternal self-report and may be subject to social desirability, recall, and other biases. In particular, women with higher levels of depressive symptoms may report more negative outcomes on other measures than women with lower levels of depressive symptoms, since depression may result in negative reporting biases (Fergusson et al., 1993, Richters, 1992, Youngstrom et al., 1999). For example, several authors have reported that depressed mothers are more
Conclusion
This study adds to our understanding of the development of maternal attachment among Latinas. Depressive symptoms during the latter part of pregnancy were found to be an important risk factor for lower maternal attachment. Screening and interventions during pregnancy are needed to identify women who are depressed or at risk for depression, rather than waiting until the postpartum period. Special attention should be focused on Latinas, who have higher levels of depression than the general
References (83)
A literature update on maternal–fetal attachment
Journal of Obstetric, Gynecologic, & Neonatal Nursing: Clinical Scholarship for the Care of Women, Childbearing Families, & Newborns
(2008)- et al.
Analysis of the psychometric properties of the Spanish version of the beck depression inventory in Argentina
Psychiatry Research
(1998) - et al.
Race-ethnic differences in marital quality and divorce
Social Science Research
(2007) Postpartum depression effects on early interactions, parenting, and safety practices: A review
Infant Behavior & Development
(2010)- et al.
Longitudinal study of maternal depressive symptoms and child well-being
Journal of the American Academy of Child & Adolescent Psychiatry S2-Journal of the American Academy of Child Psychiatry
(2001) - et al.
Maternal depression and infant temperament characteristics
Infant Behavior & Development
(2008) Prenatal and postnatal attachment: A modest correlation
Journal of Obstetric, Gynecologic, and Neonatal Nursing
(1996)- et al.
A meta-analytic study of predictors of maternal-fetal attachment
International Journal of Nursing Studies
(2009) - et al.
Influence of adolescent maternal characteristics on infant development
Infant Mental Health Journal
(2007) - et al.
Patterns of attachment: A psychological study of the strange situation
(1978)