Testosterone Moderates the Influence of Attachment Anxiety on Quality of Antenatal Bonding Among Fathers but Not Mothers Exposed to Contextual Risk

Purpose:  The parent-infant relationship begins during pregnancy (Slade, et al., 2009) and is linked to post-natal infant developmental outcomes. Anxious adult attachment style is associated with poorer post-natal bonding for both fathers and mothers (Luz, et. al., 2017) and lower testosterone has been associated with parenting quality in fathers but not in mothers (Gray, et. al., 2002).  However, less is known about the factors that may influence prenatal bonding– especially for fathers.  The primary aims of this study were to examine the influence of anxious adult attachment and testosterone levels on quality of antenatal bonding in mothers and fathers separately, and to examine whether testosterone moderated the influence of anxious adult attachment style on antenatal bonding.  

Methods:  Data were collected longitudinally to examine the influence of bio-psycho-social factors on early parenting processes in expectant mothers and fathers exposed to psychosocial stressors such as poverty and violence. Inclusion criteria were a medically uncomplicated singleton pregnancy and participation of both parents. Fifty-one expectant mothers and 51 expectant fathers from a large, metropolitan area participated. The Experiences in Close Relationships Scale (Fraley, et. al., 2000) assessed level of adult attachment anxiety. Covariates included age, race, and level of depression.  Depression was assessed using the Edinburgh Postnatal Depression Scale (Cox, Holden & Sagovsky, 1987). Quality of antenatal bonding was measured using the Antenatal Attachment Scale (Condon, 1993).  Salivary testosterone was collected at three time points, once before and twice following a simulated baby cry protocol.

 

Results:   The mean age of expectant mothers was 25.75 years.  Thirty identified as African American and 25 were unemployed. Expectant fathers’ average age was 27.19 years.  Thirty identified as African American and 17 were unemployed. Linear regression was used to examine associations between the direction of testosterone change (increase or decrease) during the baby cry protocol (BCP), attachment anxiety and quality of prenatal attachment. For fathers, the final regression model was significant at F(6, 51)=3.49, p=.01, and R²=.32.  Increased testosterone levels were associated with decreased quality of antenatal bonding (B=-2.12, SE=.94, p<.05). The interaction term was significant (B=-2.06, p<.05); for fathers whose testosterone decreased during the BCP, increases in adult anxious attachment were associated with increases in quality of antenatal bonding. Among expectant mothers, increased testosterone levels during BCP were associated with increases in antenatal bonding quality F(6,51)=2.55, p<.05, and R²=.26.  However, adult attachment anxiety was not associated with bonding quality and there was no testosterone/attachment interaction effect for mothers. For fathers, greater depression was associated with lower bonding quality (B=-4.79, p<.01), but not for mothers.

Implications:  These data increase our understanding of the biological and psychological factors that influence antenatal bonding for both mothers and fathers, and highlight the ways in which these associations may be different for fathers.  Findings inform the development of father specific interventions that social workers can utilize to improve pre-natal and post-natal bonding for fathers and infants, thereby maximizing the known positive effects of sensitive early fathering on young children (Cabrera & Tamis-LeMonda, 2013).