Associations between formal social support in early postpartum, parenting self-efficacy, and parent-infant bonding: A cross-sectional survey

TL;DR

This study finds that informal support and reduced depressive symptoms are key to improving parental self-efficacy and bonding during the postpartum period.

Contribution

The study provides new insights into how formal and informal support, along with depressive symptoms, affect parenting outcomes in early postpartum.

Findings

Higher satisfaction with informal support was significantly linked to maternal self-efficacy.

Depressive symptoms were consistently associated with lower parental self-efficacy and bonding.

Non-exclusive breastfeeding was linked to higher maternal bonding difficulties.

Abstract

Formal social support from healthcare professionals is critical in facilitating the transition to parenthood and may contribute to parental self-efficacy. However, its association with parent-infant bonding remains unclear, and evidence to guide clinical practice is scarce, especially during the postpartum hospital stay.

This study aimed to (a) explore associations between formal social support during the postpartum stay, parenting self-efficacy, and parent-infant bonding; and (b) investigate the role of depressive symptoms and other confounding factors.

A cross-sectional survey was conducted from April 2023 to August 2024 in five hospitals in French-speaking Switzerland.

420 first-time parents (298 mothers and 122 partners).

First-time mothers and co-parents completed a cross-sectional survey. Variables were measured using the Hospital Anxiety and Depression Scale, the Birth Satisfaction Scale-Revised, the FOCUS-Postpartum Scale, the Perinatal Infant Care Social Support Scale, the Parenting Expectations Survey, and the Mother-Infant Bonding Scale. Descriptive statistics, correlation analyses, and regression analyses were performed.

A total of 298 mothers and 122 partners answered the survey. Parents reported low satisfaction with the information received during the postpartum hospital stay. In bivariate analyses, formal social support was positively associated with maternal self-efficacy (rho = 0.18, p = 0.004) but not with partners’ self-efficacy (rho= -0.07, p = 0.45) or parent-infant bonding. After adjustment for confounding factors, the association between formal social support and maternal self-efficacy was no longer significant. In the maternal regression model, satisfaction with informal general support was significantly associated with maternal self-efficacy (β = 0.62, p = 0.002). Non-exclusive breastfeeding was associated with higher maternal bonding difficulties (β = 0.34, p = 0.008). Partners with a secondary education level reported lower parenting self-efficacy (β = -35.49, p = 0.005) and lower bonding quality (β = 0.7, p = 0.019) than those with an apprenticeship education. Depressive symptoms were negatively associated with parental self-efficacy (mothers β = -2.44, p = 0.015; partners β = -4.57, p < 0.001) and bonding (mothers β = 0.08, p < 0.001; partners β = 0.07, p = 0.002).

Higher satisfaction with informal support was linked to maternal self-efficacy, while depressive symptoms consistently undermined both parental self-efficacy and bonding for both parents. Together with parents’ low satisfaction with information received, these findings underscore the need for tailored support and the importance of clear and consistent communication from healthcare professionals for both mothers and partners.

Full text

What is already known

• •Formal social support is crucial during the early postpartum period.
• •Evidence linking formal support to parenting self-efficacy and parent-infant bonding is limited.

Alt-text: Unlabelled box dummy alt text What this paper adds

• •Provides new evidence from the very early postpartum period, including both mothers and partners.
• •Distinguishes the respective roles of formal and informal social support in relation to parental self-efficacy and bonding.
• •Emphasises the role of depressive symptoms and contextual factors in parental self-efficacy and parent–infant bonding.

Alt-text: Unlabelled box dummy alt text

Background

1

Becoming a parent for the first time marks a significant life transition, leading to notable social and emotional adjustments (Finlayson et al., 2020; Schobinger et al., 2022). The early postpartum, lasting from the second to the seventh day after birth, is critical for developing parenting confidence (World Health Organisation, 2010). During this period, first-time parents are at higher risk of acute stress, difficulties with infant bonding, and potential mental health complications (Schobinger et al., 2020), and report lower parenting self-efficacy than multiparous parents (Liyana Amin et al., 2018; Samdan et al., 2022), possibly due to the absence of enactive mastery experience –prior successful experiences in infant care– which is key role for self-efficacy (Saether et al., 2023).

Parenting self-efficacy and bonding are central indicators of parental adjustment (Zheng and Gao, 2023; de Waal et al., 2025). Parenting self-efficacy refers to the belief parents hold in their capacity to care for their infant (de Montigny and Lacharité, 2005; Zheng and Gao, 2023). Four primary sources contribute to the development of self-efficacy beliefs: 1) past success experiences; 2) indirect experiences through observing others; 3) verbal encouragement; and 4) physiological and affective states, where positive emotions boost self-efficacy and negative emotions diminish it (Bandura, 1997; Fierloos et al., 2023). Early postpartum period is also when the parent-infant bond forms, encompassing emotional involvement and internal feelings toward the infant (Nakić Radoš et al., 2024; Toivo et al., 2023). Supporting parenting self-efficacy and bonding in the early postpartum may therefore foster a positive transition to parenthood.

Social support is an important protective factor during this transition. Social support is a dynamic, reciprocal process that enables individuals to navigate difficult or stressful situations more effectively, thereby improving the health and well-being of the recipient (Langford et al., 1997; Ekström-Bergström et al., 2022). It can be informal (support from family and close ones) or formal (support from healthcare professionals) (Leahy-Warren et al., 2019). Whilst the benefits of informal support are well described in the literature, evidence regarding formal support is limited, despite its importance during early postpartum. Formal support differs in nature and purpose from informal support. Informal support, typically provided by family and friends, is influenced by the nature of the relational and emotional relationship, whereas formal support from healthcare professionals is considered an intervention to support person-and family-centred care (Voorhees, 2023; Leahy-Warren, 2014). Professional mainly provide informational and appraisal support (e.g., guidance on infant care, reassurance), while informal support is mostly emotional or instrumental (Leahy-Warren, 2014, Francis Xavier et al., 2024). First-time parents often identify their partner as their primary source of emotional support (Francis Xavier et al., 2024; Leahy-Warren et al., 2019). However, first-time fathers tend to rely on formal support for expert guidance (Francis Xavier et al., 2024). This distinction highlights the unique role healthcare professionals can play in enhancing parents’ confidence and skills. Supportive behaviours include actions or interpersonal skills demonstrated by healthcare professionals to assist first-time parents in their transition to parenthood (Schobinger et al., 2025b, Francis Xavier et al., 2024). Healthcare professionals may use a wide range of behaviours to help first-time parents, such as providing information on infant care, yet, when given inappropriately (e.g., contradictory information), it can causes unnecessary stress to parents (Schobinger et al., 2024).

Adequate formal support reduces stress and risk of mental health disorders (World Health Organisation, 2022; Finlayson et al., 2020; Baldwin et al., 2019). When parents receive adequate informal support they develop a sense of security, fostering parenting self-efficacy (Fang et al., 2021). Informal social support is correlated with parenting self-efficacy (Fierloos et al., 2023; Zheng et al., 2018). However, only one study investigated if formal social support from professionals was associated with parenting self-efficacy and found no association (Salonen et al., 2009). In a recent scoping review, Sæther et al. (2023) explored first-time parents' experiences related to parental self-efficacy and highlighted the crucial role of professional support. First-time parents valued emotionally responsive and individualised support, trusting relationships, and continuity of care. They appreciated when professionals were respectful, non-judgmental, and took time to listen to their needs and goals. Fathers emphasised the importance of hands-on experience and disliked passive, overly formal instruction (Saether et al., 2023). These findings suggest that formal support may enhance parenting self-efficacy when it meets the specific needs of new parents. Several studies show that informal social support is positively associated with parent-infant bonding (Stuijfzand et al., 2020; Martin and Brock, 2023), but, so far, there is no evidence for formal social support and bonding.

When parents perceive a lack of formal social support or perceive it as inadequate, they are at higher risk of developing mental health problems (Ayers et al., 2019; Cirino and Knapp, 2019), and an increased risk of seeking emergency care for their newborn within the first two weeks after discharge (Barimani et al., 2014). Fathers and other co-parents may also be affected by a lack of formal social support, experiencing negative emotions and psychological distress (Kothari et al., 2022; Eddy et al., 2019). Despite growing recognition of the father's role, most research still focuses on mothers, and paternal experiences remain underexplored. Qualitative work such as Hodgson et al. (2021) has found that first-time fathers often feel “present but invisible” in perinatal services, citing poor communication and a lack of father-specific support. These findings are echoed in a Swiss qualitative study, where first-time fathers reported distinct needs, such as being included in infant care and receiving reassurance (Schobinger et al., 2022). Similarly, a scoping review identified that a non-inclusive hospital environment and healthcare professionals’ inattentiveness to paternal needs were inhibiting factors for first-time fathers’ parenting self-efficacy (Saether et al., 2023). Such findings underscore the importance of including fathers in studies on formal support, as their perceptions and needs may differ from those of mothers and are essential for developing inclusive, equitable care practices.

Beyond social support, psychosocial, obstetrical, and infant-related factors influence parenting self-efficacy and/or parent-infant bonding. Parental mental health and stress impact first-time parents’ parenting self-efficacy (Fang et al., 2021; Saether et al., 2023) and parent-infant bonding (Ngai and Lam, 2023). Obstetrical factors, such as mode of birth can influence birth satisfaction (Falk et al., 2019), which in turn has been associated with parenting self-efficacy (Brand et al., 2025) and bonding quality (Döblin et al., 2023; Nakić Radoš et al., 2024). Caesarean section and early discharge may enhance first-time fathers’ parenting self-efficacy (Saether et al., 2023). Infant factors may challenge parents' confidence and bonding (Azmoude et al., 2015; Nakić Radoš et al., 2024), especially for first-time parents (Saether et al., 2023). Parental education may shape expectations and skills relevant to early caregiving (Donithen and Schoppe-Sullivan, 2022; Fang et al., 2021). Given these findings, accounting for these factors is essential to fully understand the relationship between formal social support, parenting self-efficacy, and bonding.

In summary, formal social support is an important protective factor in the postpartum period, especially for first-time parents. While research on informal or mixed sources of support has found positive associations with parenting self-efficacy and bonding, evidence on formal social support during the postpartum stay remains scarce. The only study specifically examining formal support found no association, highlighting the need to better understand how supportive behaviours from healthcare professionals may contribute to parenting self-efficacy and bonding in the transition to parenthood.

The first objective of this study was thus to investigate relationships between formal social support and maternal self-efficacy and mother-infant bonding, as well as paternal self-efficacy and father-infant bonding. Based on the literature, we hypothesised that (H1) formal social support would be positively associated with perceived parenting self-efficacy, and (H2) negatively associated with parent-infant bonding difficulties.

The second objective was to identify which psychosocial, obstetrical, and infant factors may be associated with these relationships, for mothers and partners separately. We hypothesised that (H3) psychosocial, obstetric, and infant factors would be associated with perceived parental self-efficacy and bonding for mothers and partners.

Methods

2

Design

2.1

This study is part of the "SUPERSTAR" project, which includes (1) the development and psychometric validation of a new instrument assessing formal social support (Schobinger et al., 2025a- under review), and (2) the present cross-sectional study, which uses the newly developed instrument to investigate associations between formal social support and psychosocial outcomes in mothers and partners. Participants were recruited from postpartum wards of five hospitals in French-speaking Switzerland (one university hospital and four regional hospitals). All public hospitals in the canton of Vaud were invited to participate; four agreed to take part, while one declined. The selection of these hospitals was purposive, aiming to represent the regional population and to facilitate recruitment within a reasonable geographic area for the research team. Recruitment occurred from April 2023 to July 2024. The research team approached participants during their postpartum stay, provided study information, and offered a leaflet with key information and a QR code linking to an anonymous online survey. To foster recruitment leaflets about the study were placed in the waiting rooms of gynaecologists and midwives and social media (LinkedIn, Instagram, and Facebook) was also used to promote the study.

Participants and setting

2.2

Participants were invited to complete the questionnaire within the first month postpartum if they were first-time parents (over 18 years of age), gave birth to a healthy full-term newborn (≥37 gestational weeks), and had been present during birth (for partners). Furthermore, they needed to have been present during the postpartum stay at the hospital. Partners were considered present during the postpartum stay if they stayed overnight or were present at any time during the day at the hospital. Exclusion criteria included parents who were non-French speakers, whose infants were admitted to the Neonatal Intensive Care Unit (NICU), or experienced stillbirth, or whose infant died within 24 h post-delivery, as the study focused on a low-risk population.

Partners were invited to participate if present during birth and the postpartum stay. To maximise inclusion, midwives were asked whether the partner was a first-time parent, especially in cases where the mother had multiple children, but the partner might be experiencing parenthood for the first time. In such cases, the partner was invited to participate. Conversely, partners who already had children from previous relationships were excluded to maintain focus on first-time parenthood.

In Switzerland, health insurance is mandatory, and 98 % of women give birth in hospitals, with caesarean sections accounting for 33 % of births (Office fédéral de la statistique, 2019). The typical postpartum length of hospital stay is 2 to 3 days after a vaginal birth and 3–4 days after a caesarean section. Parents receive written information on self-care, newborn care and community resources, including midwifery services. Mothers and newborns are entitled to 16 postpartum midwife visits within the first eight weeks.

Sample size

2.3

Sample size was based on the findings of previous studies reporting medium-sized correlations between informal social support and parenting self-efficacy (Leahy-Warren et al., 2012; Shorey et al., 2014) and small correlations with parental bonding (Stuijfzand et al., 2020). G*Power analysis for bivariate correlation (two-tailed), assuming a small-to-moderate effect size (r = 0.20, α = 0.05) and 0.80 power, estimated a sample of 200 mothers and 200 partners.

For multiple regression testing H3 (moderate effect size (f² = 0.15, α = 0.05), power = 0.80, 20 predictors) 160 participants were required. The larger bivariate correlation estimate was retained as the final target sample. Adding 15 % for potential participant dropout and questionnaires with no usable data, the total target sample size was 460 parents (230 mothers, 230 partners).

Data collection

2.4

Participants completed an online anonymous survey on their experience of formal social support received during their postpartum hospital stay. Parents were invited to complete the survey at any time within the first month postpartum to allow flexibility. Mothers and partners participated individually; with couple participation encouraged. The completion time of the survey was between 20 and 30 min.

Instruments

2.5

Sociodemographic and birth-related variables included gestity, complications during pregnancy and in the immediate postpartum, mode of delivery, infant’s birth weight and gestational age, infant feeding at discharge, and postpartum duration stay.

Social support received from healthcare professionals

2.5.1

Formal social support parents received during the postpartum stay was measured using the 35-item FOCUS-Postpartum Scale, developed and validated with postpartum women. Participants rated healthcare professionals’ social support behaviours from 0 (did not meet their needs at all) to 4 (completely met their needs), providing a possible total score ranging from 0 to 140, where higher scores indicate a greater extent to which parents’ support needs were met. It includes an overall factor and four dimensions: “optimising parents’ learning” (13 items, ranging 0 to 52), “providing respectful care” (12 items, ranging 0 to 48), “caring with empathy” (5 items, ranging 0 to 20), and “admitting and orientating parents” (5 items, ranging 0 to 20) (Schobinger et al., 2025a- under review). The scale was developed following a rigorous multi-phase process aligned with COSMIN guidelines. Item generation was based on a literature review and qualitative interviews with postpartum mothers and fathers, followed by item selection and evaluation through a Delphi survey involving 90 first-time parents. The item pool was further refined with input from perinatal experts and parents. The scale demonstrated strong internal consistency in the initial validation study with postpartum mothers with α = 0.97, and its structural validity was supported by exploratory and confirmatory factor analyses (Schobinger et al., 2025a- under review). In the current study sample including both mothers and fathers/partners, the scale showed excellent reliability, with α = 0.97.

Social support in the perinatal period

2.5.2

Social support from healthcare professionals and informal support from family was measured using the Perinatal Infant Care Social Support instrument, a validated measure of postpartum social support for mothers (Leahy-Warren et al., 2019). Functional support is measured using 19 items rated on a 4-point Likert scale. This questionnaire was translated into French and culturally adapted according to Wild et al. (2005), with α = 0.92 in this study.

Satisfaction with formal and informal social support

2.5.3

Satisfaction with formal social support was assessed using five questions on a 100-mm visual analogue scale, ranging from 0 (extremely dissatisfied) to 100 (extremely satisfied) (Voutilainen et al., 2016). One question measured overall satisfaction, while the others addressed informational, instrumental, appraisal, and emotional support. Internal consistency for the five items was excellent (α = 0.93).

Exploratory factor analysis supported the structural validity of the scale. The suitability of the data for factor analysis was confirmed by a Kaiser-Meyer-Olkin measure of sampling adequacy of 0.867. Bartlett’s test of sphericity was significant (χ² = 1691.27, df = 10, p < 0.001), and the determinant of the correlation matrix (0.010) indicated no issue of multicollinearity. The exploratory factor analysis revealed a unidimensional structure: (first factor eigenvalue = 3.78, explaining 76 % of the variance, all loadings > 0.85). Although a composite score was possible, each item represents a distinct theoretical dimension of formal support. Therefore, we chose to analyse them individually to reflect the nuanced aspects of parents’ satisfaction. Scores were categorised as “not satisfied” (<50), “moderately satisfied” (50–79) and “satisfied.” (≥80), aligning with previous research, where ≥80 % is commonly interpreted as “satisfied or very satisfied” (Beshir et al., 2022; Vogel et al., 2019).

Satisfaction with informal social support was measured using a response format similar to formal support (five questions, each representing a distinct aspect of support). Exploratory analyses supported a single-factor structure (first factor eigenvalue = 2.52; loadings ranging from 0.46 to 0.79), and good internal consistency (α = 0.81).

General psychological distress

2.5.4

General psychological distress of parents was assessed using French version of the Hospital Anxiety and Depression Scale (Boini and Langevin, 2020). Anxiety and depression were measured with the subscales. The 14 self-reported items are rated on a 4-point Likert scale, with higher score indicating greater severity (range from 0 to 21 for each dimension) (Zigmond and Snaith, 1983). The Hospital Anxiety Depression Scale has good psychometric properties (Zigmond and Snaith, 1983). For this study, α was 0.84 for general distress, 0.76 for anxiety, and 0.78 for depression.

Birth satisfaction

2.5.5

Subjective birth experience was assessed using the 10-item Birth Satisfaction Scale-Revised, which measures postpartum birth satisfaction in mothers (Hollins Martin and Martin, 2014). Items are rated on a 4-point Likert scale; total scores range from 0 to 40, with a higher score indicating greater satisfaction. The scale has a good internal consistency (α = 0.79) and has been translated and validated in French (Devita et al., 2026). In our study, internal consistency was α = 0.75.

Perceived parental self-efficacy

2.5.6

The French version of the Parenting Expectations Survey was used to assess parenting self-efficacy in the early postpartum (de Montigny, 2002). This 25-item self-report scale, with responses rated on a 10-point Likert scale (total score range 0 to 250), has been validated in French for women at one month postpartum (de Montigny and Lacharité, 2002; Razurel et al., 2017). In this study α = 0.92 for mothers and α = 0.88 for partners.

Parent-infant bonding

2.5.7

The 8-item Mother-to-Infant Bonding scale assesses maternal feelings and the bonding quality on a 4-point Likert scale, with higher scores indicating problematic bonding (Taylor et al., 2005). It has fair reliability (α = 0.71), was validated in French for mothers (Bienfait et al., 2017), and used with both parents in a Swiss study (α = 0.77) (Horsch et al., 2017). In this study, Cronbach’s α was 0.67.

Data analysis

2.6

We used t-tests to compare sociodemographic differences between mothers and partners, and logistic regression to assess differences between complete and incomplete responses. Due to non-normal distribution of some continuous variables (e.g., support received, satisfaction, and bonding scores), differences between mothers and partners were assessed using the Wilcoxon rank-sum test (Mann–Whitney U test). For categorical variables (e.g., satisfaction yes/no), group differences were analysed using chi-square tests or Fisher’s exact tests when cell counts were small. Although not directly related to the main hypotheses, these comparative analyses aimed to provide a more detailed characterisation of the sample and to assess potential response bias, thereby supporting the internal validity and generalisability of the findings and are thus succinctly presented in the results section.

To test hypothesis one (H1) and two (H2), Spearman correlations coefficients were calculated between formal social support and parenting self-efficacy and bonding, due to non-normal distributions. Associations between FOCUS-Postpartum scores and Parenting Expectations Survey, as well as between FOCUS-Postpartum scores and Mother-Infant Bonding Scale, were analysed using a linear regression, adjusting for sociodemographic, obstetric, and infant-related factors, testing for hypothesis three (H3). These are referred to as the “parenting self-efficacy model” and the “bonding model” for mothers and partners, respectively. Covariates were selected a priori based on the literature and their relevance for each outcome. Regression assumptions (linearity, independence, homoscedasticity, normality of residuals) and variance inflation factors were examined. The main variable of interest in this study was the level of formal support received from healthcare professionals, measured with the FOCUS-postpartum scale. Satisfaction with formal social support was also assessed; however, it was highly correlated with the formal support score (overall satisfaction rho=0.72, p < 0.001; satisfaction with formal informational support rho=0.76, p < 0.001; satisfaction with formal instrumental support rho=0.79, p < 0.001; satisfaction with formal appraisal support rho=0.75, p < 0.001; satisfaction with formal emotional support rho=0.69, p < 0.001). After removing satisfaction with formal social support variables, variance inflation factors for other variables were below commonly accepted thresholds, indicating no problematic multicollinearity. Breusch–Pagan test indicated heteroskedasticity for mothers’ parenting self-efficacy model. Consequently, estimates for this model were reported with robust standard errors. For the parenting self-efficacy model in partners, all assumptions were met. Although satisfaction with informal and formal social support was categorised for descriptive purposes (to report the proportion of participants who were satisfied, moderately satisfied, or not satisfied), it was treated as a continuous variable in the regression analyses.

As the parent-infant bonding scores were zero-inflated (mothers: 36.3 %; partners: 31.9 %) and right-skewed, consistent with a mixture of “no bonding difficulties” (zeros) and “bonding difficulties” (positive values), we employed a hurdle-gamma regression approach combining: a logistic component (modelling zero vs. non-zero responses), which was specified as intercept-only and a Gamma General Linear Model (modelling the truncated positive values, accounting for skewness). Whilst the logistic component was included to account for the data structure, our primary analytical focus was on the Gamma Generalised Linear Model, which provided interpretable estimates of associations with bonding difficulties. The logistic model results are not reported, as they did not yield additional insights relevant to the study objectives.

Model diagnostics were performed using the Diagnostics for Hierarchical Regression Models package, which provides simulation-based residual checks for non-Gaussian models, including tests for uniformity and dispersion. To evaluate the robustness of the primary Gamma Generalised Linear Model results, sensitivity analyses were conducted by fitting alternative models (e.g., Tweedie, Negative Binomial, Log-normal) to the subset of positive Mother-Infant Bonding scores. Model performance was compared using the Akaike Information Criterion and simulation-based diagnostics for residual uniformity and dispersion, implemented via the Diagnostics for Hierarchical Regression Models package. Lower Akaike Information Criterion values indicate better overall fit, while uniformity and dispersion tests assess the adequacy of residual distribution and variance structure (ideal: small D and large p-value for uniformity; dispersion close to 1 with large p-value). These comparisons allowed us to determine whether alternative specifications provided a substantially better fit than the Gamma Generalised Linear Model. The Gamma Generalised Linear Model remained the most parsimonious and theoretically justified model across comparisons (see Table 1). A Log-normal model was also fitted but excluded due to theoretical mismatches with the data’s skewness.Table 1. Sensitivity analysis results.Table 1 dummy alt textParticipantsModelAkaike Information CriteriaUniformityDispersionMothersGamma455.92D = 0.099, p = 0.146φ=1.221, p = 0.256Tweedie460.68D = 0.102, p = 0.126φ=1.043, p = 0.624Negative binomial508.98D = 0.09, p = 0.223φ=1.031, p = 0.76PartnersGamma229.6D = 0.105, p = 0.496φ=0.93, p = 0.968Tweedie237.14D = 0.093, p = 0.65φ=0.614, p = 0.184Negative binomial249.98D = 0.136, p = 0.176φ=0.486, p < 0.001

Finally, statistical power for all regression models (parenting self-efficacy and bonding models for mothers and partners) was estimated using bootstrap resampling based on the observed effect sizes (5000 iterations). Pairwise comparisons between all levels of categorical predictors (e.g., education, mode of birth) were conducted to provide additional descriptive detail and to examine potential differences across levels of categorical covariates, using post hoc tests with Bonferroni correction to control the family-wise error rate. Effect sizes were reported as mean differences (95 % confidence intervals) between adjusted group means, along with adjusted p-values. All statistical analyses were performed using R (R Core Team, 2024) and Stata version 18 (StataCorp LLC, College Station, TX, USA). A two-tailed p-value of < 0.05 was considered statistically significant.

Mothers and partners were analysed separately to account for their distinct experiences of formal support during postpartum hospitalisation, and the individual-level nature of parenting self-efficacy and bonding. This approach avoided assuming equivalence across parents and prevented partners’ data from being overshadowed by the larger maternal sample. Since most participants responded individually rather than as couples, dyadic analyses were not feasible. For descriptive comparisons, participants were treated as independent units, as pairing information was unavailable.

Handling of missing data

2.6.1

Fifty-seven participants who did not respond to any of the items of the Parenting Expectations Survey or the Mother-Infant Bonding Scale were excluded from the analyses. Missing data patterns of these questionnaires were examined using Little’s missing completely at random test. For the Parenting Expectations Survey, data were missing completely at random (χ²(737) = 774.58, p = 0.164), with 1.3 % missing data in the remaining dataset. For the Mother-Infant Bonding Scale, missing completely at random assumption was violated (χ²(34) = 88.3, p < 0.001), with 0.3 % missing data remaining. Little's missing completely at random tests were performed separately for Parenting Expectations Survey and the Mother-Infant Bonding Scale as they measured distinct constructs with potentially different missingness mechanisms and as these were the only questionnaires with missing data requiring imputation. For completeness, a pooled test of all imputed variables from both questionnaires confirmed the data were missing completely at random, (χ²(1073) = 1032.56, p = 0.808), with 36 missingness patterns. Given the very low level of missingness and absence of systematic bias, missing values for both questionnaires were imputed using predictive mean matching method from the mice package. This method is under the missing at random assumption, which preserves the original data distribution, handles ordinal data well, and reduces the risk of introducing bias compared to other imputation methods (Van Buuren, 2018). Five imputed datasets were initially generated; however, the imputation was performed solely to obtain complete questionnaire scores for these primary outcomes, and only a single completed dataset was used for subsequent analyses. Descriptive statistics pre- and post-imputation were nearly identical (mean difference < 0.1 % of the SD), indicating negligible impact. Given the minimal missingness, formal sensitivity analyses for missing not at random were deemed unnecessary (White et al., 2011; Van Buuren, 2018).

Ethical considerations

2.7

This study was approved by the Human Research and Ethics Committee of the Canton of Vaud, Switzerland (Project No 2021-00762). Written informed consent was not necessary, as the study was categorised as anonymous, and completion of the online survey implied consent.

Results

3

Sample characteristics

3.1

A total of 420 participants (298 mothers (71 %) and 122 (29 %) partners) took part in this study and 363 (256 mothers and 107 partners) completed the parenting self-efficacy and bonding questionnaires. The majority were Swiss (65.6 %), had a university degree (68.6 %), and lived together (75 %). The majority gave birth at term (66.7 %) and in a university hospital (60.9 %). Postpartum complications were mainly haemorrhage (n = 17, 58.6 %) and severe perineal tear (n = 33, 8 %) (see Table 2).Table 2. Sample characteristics.Table 2 dummy alt textMothers (N = 298)Partners (N = 122)P-value95 % Confidence intervalAge (years; mean, SD)32.3 ± 4.633.2 ± 4.30.06 a−1.86 0.04Time elapsed between birth and questionnaire completion (days; mean, SD)13 ± 7.3511.31 ± 7.350.002 b−0.31 −0.06Postpartum length of stay (days; mean, SD)3.37 ± 1.193.27± 1.240.46 b−0.16 0.07Nationality0.71 c Swiss185 (64.2 %)82 (68.9 %) French37 (12.8 %)19 (16 %) Italian14 (4.9 %)6 (5 %) Portuguese17 (5.9 %)2 (1.7 %) Belgian8 (2.8 %)3 (2.5 %) Spanish8 (2.8 %)2 (1.7 %)Origin0.18 c Europe267 (90.5 %)114 (95.8 %) Latine America6 (2 %)3 (2.5 %) Africa8 (2.7 %) North America4 (1.7 %)2 (1.7 %) Asia6 (2 %)Civil status0.56 c Married/in a relationship217 (73.3 %)95 (77.9 %) Single75 (25.3 %)25 (20.5 %) Divorced4 (1.4 %)2 (1.6 %)Previous pregnancy0.008 c None233 (78.4 %)115 (94.3 %) One48 (16.2 %)6 (4.9 %) Two9 (3 %)1 (0.8 %)Previous perinatal loss0.18 c No231 (77.8 %)102 (83.6 %) Yes66 (22.2 %)20 (16.4 %)Prenatal class attendance0.73 c Yes250 (85.6 %)106 (86.9 %) No42 (14.4 %)16 (13.1 %)Complication during pregnancy0.23 c No250 (86.2 %)98 (81.7 %) Yes40 (13.8 %)22 (18.3 %)Mode of birth0.95 c Spontaneous vaginal birth162 (55.7 %)66 (55.5 %) Instrumental forceps32 (11 %)14 (11.8 %) Instrumental vacuum assisted33 (11.3 %)16 (13.4 %) Planned Caesarean section22 (7.6 %)8 (6.7 %) Emergency Caesarean section42 (14.4 %)15 (12.6 %)Postpartum complications0.01 c No258 (89 %)96 (80 %) Yes31 (11 %)24 (20 %)Infant sex0.33 c Girl144 (49.8 %)65 (55.1 %) Boy145 (50.2 %)53 (44.9 %)Infant feeding at discharge0.64 c Breast227 (76.2 %)96 (80 %) Bottle24 (8 %)7 (5.8 %) Both47 (15.8 %)17 (14.2 %)Gestational age at birth (weeks)39.63 ± 1.3239.57 ± 1.420.56 b−0.16 0.09Birth weight (grams)3317 ± 4363304 ± 3770.77 a−0.76 103.08General psychological distress (Median; IQR)111 (9)7 (6) Anxiety6 (5)5 (4)<0.001 b−0.49 −0.25 Depression4 (4.5)2 (4)0.0001 b−0.37 −0.231IQR = Interquartile Range. Mean and SD are reported for normally distributed variables; Median and IQR are reported for non-normally distributed variables.at-test.b: ranksum test .c: Chi-square test.

Most partners (n = 83, 68.6 %) stayed overnight during the postpartum stay.

No differences between participants who completed the Parenting Expectations Survey and bonding questionnaires and those who did not were found.

Formal support received from healthcare professionals and satisfaction

3.2

Formal support median scores were 109 (IQR = 40.5) for mothers and 115 (IQR = 32) for partners, with no statistically significant difference (p = 0.066).

Most participants were satisfied with the support received. Among mothers, 72 % were satisfied with general formal support and partners’ reported satisfaction was 81.8 %. The detailed categories of satisfaction are shown in Table 3.Table 3. Formal support received and Satisfaction.Table 3: dummy alt textMothers Median (IQR)Partners Median (IQR)P-valueranksum test95 % Confidence IntervalSupport receivedOptimising parents’ learning42 (17)44 (12)0.29^a^−0.05 0.19Providing respectful care36 (17.5)39 (13)0.07^a^−0.008 0.24Caring with empathy18 (5)19 (3.5)0.06^a^−0.001 0.24Admitting and orientating parents15 (7)15 (6)0.13^a^−0.24 0.22**Satisfaction (N (**%))**satisfied motherssatisfied partnersP-valuechi2 testSatisfaction with general formal support191 (72.6 %)90 (81.8 %)0.16^b^−3.67 11.06Satisfaction with formal informational support152 (57.1 %)69 (62.1 %)0.36^b^- 3.39 7.51Satisfaction with formal instrumental support165 (62.1 %)80 (72.7 %)0.14^b^- 3.19 11.77Satisfaction with formal appraisal support160 (60.6 %)72 (65.5 %)0.62^b^- 4.38 6.28Satisfaction with formal emotional support195 (74.1 %)91 (81.9 %)0.26^b^- 3.75 9.15IQR: interquartile range: ^a:^ ranksum test ^b:^ chi2 test.

Parenting self-efficacy and bonding

3.3

Mothers’ median parenting self-efficacy was 188 (IQR=42) and bonding 1 (IQR=2), while partners’ parenting self-efficacy was 184.5 (IQR=41) and bonding 1 (IQR=3) (see Table 4).Table 4. Description of main outcomes.Table 4: dummy alt textMothersPartnersP-value ranksum test95 % Confidence intervalParenting self-efficacy (median; IQR)188 ; 42184.5 ; 410.5−0.17 0.08Bonding (median; IQR)1 ; 21 ; 30.19−0.04 0.21Satisfaction of birth experience (median; IQR)30; 930; 90.84−0.14 0.11IQR: interquartile range.

Associations between formal support and parenting self-efficacy and parent-infant bonding

3.4

Formal support was positively associated with maternal parenting self-efficacy (rho = 0.18, p = 0.002) but not with partners’ parenting self-efficacy (rho = −0.07, p = 0.45). Furthermore, none of the subscales correlated with partners’ parenting self-efficacy.

Formal social support received during the postpartum stay was not associated with mother-infant (rho = −0.002, p = 0.97) or father-infant bonding (rho = −0.07, p = 0.42). (See Table 5)Table 5. Associations between formal support and parenting self-efficacy and bonding scores.Table 5: dummy alt textMaternal parenting self-efficacyMother-infant bondingPartners’ parenting self-efficacyFather-infant bondingFormal support0.18****−0.002−0.070.07Optimising parents’ learning0.20****−0.004−0.070.11Provision of respectful care0.17****−0.02−0.040.09Admitting and orientating parents0.14***−0.010.06−0.06Caring with empathy0.12−0.07−0.160.12Rho values that are in bold are significant: Significance levels: *p <0 .05, **p < 0.01.

Factors associated with maternal parenting self-efficacy and mother infant-bonding

3.5

The linear regression model including psychological and sociodemographic covariates explained 47.3 % of the variance in maternal parenting self-efficacy (R^2^ = 0.4730) [X^2^(25) =132.65; p < 0.001]. Higher depressive symptoms were significantly associated with lower maternal parenting self-efficacy (β = −2.447, p = 0.015), while greater satisfaction with informal support was associated with higher maternal parenting self-efficacy (β = 0.623, p = 0.002). Maternal anxiety and satisfaction with instrumental informal support showed marginal associations (see Table 6). Among the significant predictors of maternal parenting self-efficacy, depressive symptoms showed the strongest association.Table 6. Multiple regression model for maternal parenting self-efficacy and Gamma Generalised Linear Model regression for mother infant-bonding.Table 6: dummy alt text(N = 208)CoefficientRobust Std. errtp95 %CIMaternal Parenting self-efficacy aFormal social support0.0770.0771.010.312−0.074 0.228Age−0.2460.475−0.520.605−1.177 0.685Education Secondary−8.316.412−1.300.195−20.878 4.258 Tertiary−7.6775.295−1.450.147−18.055 2.701Depressive symptoms−2.4471.003−2.440.015−4.413 −0.481Anxiety symptoms−1.5660.941−1.660.096−3.41 0.278Birth preparation class attendance No−1.2345.419−0.230.82−11.855 9.387Complication during pregnancy None1.8556.1580.300.763−10.215 13.925Mode of birth Forceps assisted delivery−0.127.439−0.020.987−14.7 14.46 Vacuum assisted delivery−7.88217.41−1.060.287−22.406 6.642 Planned caesarean section10.0787.0611.430.154−3.762 23.918 Emergency caesarean section7.7754.9721.560.118−1.97 17.52Birth satisfaction0.3990.3431.160.245−0.273 1.071Postpartum complication None3.0697.780.390.693−12.18 18.318Postpartum length of stay0.0441.6330.030.978−3.157 3.245Time elapsed between birth and questionnaire completion0.2130.2890.740.465−0.353 0.779Infant sex Boy−1.9894.269−0.470.641−10.356 6.378Infant birth weight−4.3795.995−0.730.465−16.129 7.371Feeding mode Mixed or formula feeding6.4755.7361.130.259−4.768 17.718Perinatal Infant Care Social support score0.5860.4221.390.165−0.241 1.413Satisfaction with informal general support0.6230.2033.070.0020.225 1.021Satisfaction with informal informational support0.060.1050.570.568−0.146 0.226Satisfaction with informal instrumental support−0.2250.136−1.650.099−0.492 0.042Satisfaction with informal appraisal support0.1530.210.730.466−0.259 0.565Satisfaction with informal emotional support−0.1890.15−1.250.21−0.492 . 0.042**(N****=134)coefficientStd. errorzp95 %CI****Mother infant bonding**bFormal social support0.0030.0021.270.203−0.002 0.008Age0.0360.0152.440.0150.007 0.065Education Secondary0.0540.2280.240.811−0.392 0.509 Tertiary0.1550.1680.920.357−0.183 0.482Depressive symptoms0.0830.0194.31<0.0010.045 0.122Anxiety symptoms0.0030.0180.160.875−0.032 0.038Complication during pregnancy None−0.0060.154−0.040.969−0.317 0.292Mode of birth Forceps assisted delivery−0.0720.15−0.480.632−0.364 0.229 Vacuum assisted delivery−0.120.19−0.630.528−0.485 0.268 Planned caesarean section−0.2920.244−1.200.23−0.758 0.207 Emergency caesarean section−0.0720.2−0.360.718−0.458 0.334Birth satisfaction−0.0090.009−1.020.307−0.028 0.009Postpartum complication None−0.0180.152−0.120.908−0.328 0.274Postpartum length of stay−0.0250.049−0.510.612−0.121 0.073Time elapsed between birth and questionnaire completion−0.0040.008−0.510.611−0.019 0.012Infant sex Boy0.0050.1140.040.964−0.221 0.229Infant birth weight0.0040.130.030.973−0.252 0.263Feeding mode Mixed or formula feeding0.3440.1312. 460.0080.09 0.606Perinatal Infant Care Social support score−0.0090.008−1.070.285−0.024 0.007Satisfaction with informal general support0.0090.0051.610.107−0.002 0.019Satisfaction with informal informational support−0.0060.003−1.840.066−0.013 0Satisfaction with informal instrumental support−0.0030.005−0.690.493−0.012 0.006Satisfaction with informal appraisal support0.0080.0041.820.068−0.001 0.016Satisfaction with informal emotional support−0.0050.004−1.350.177−0.013 0.002aModel fit: Adjusted R² = 0.473, X^2^(25) = 123.648, p < 0.001.bModel fit: R² = 0.521.

In the hurdle-gamma regression model, 52.1 % of the variance in mother-infant bonding scores was explained (see Table 6). Depressive symptoms were associated with greater mother-infant bonding difficulties (β = 0.083, p < 0.001), corresponding to an 8.7 % increase in bonding difficulties per unit increase in depressive symptoms. Older maternal age was also associated with more bonding difficulties (β = 0.036, p = 0.015). Mothers who exclusively breastfed reported better infant bonding than those using mixed feeding, as indicated by lower Mother-Infant Bonding scores (β = 0.344, p = 0.008). Satisfaction with informal informational and appraisal support, were only marginally significant (p = 0.066 and 0.068) and underpowered (power = 0.41– 0.48). Among significant predictors, infant feeding mode showed one of the strongest associations with bonding difficulties, with mothers who exclusively breastfed reporting fewer difficulties than those using mixed feeding.

Factors associated with partners’ parenting self-efficacy and father-infant bonding.

3.6

The linear regression model including psychological and sociodemographic covariates, yielded a model fit of F(25, 67) = 2.157; p = 0.007 (see Table 7) for partners’ parenting self-efficacy. It explained 44.6 % of the variance of partners’ parenting self-efficacy (R^2^ = 0.446). Higher depressive symptoms were significantly associated with lower partners’ perceived self-efficacy (β = −4.569, p < 0.001), representing the strongest continuous predictor. Partners with secondary education reported lower self-efficacy compared to those with apprenticeship (β = −35.491, p < 0.005) (Table 7). Post-hoc comparisons (Table 8) using estimated marginal means with Bonferroni adjustment showed a significant trend in which partners with “secondary” education had a lower expected mean of parenting-self efficacy score than the “apprenticeship” group (estimated marginal means difference = 35.5, p = 0.014). Length of postpartum stay showed a marginal association. Post-hoc power analysis indicated that the observed effect sizes were underpowered in our sample.Table 7. Multiple regression model for partners’ parenting self-efficacy and Gamma Generalised Linear Model regression father-infant bonding.Table 7: dummy alt text(N = 93)CoefficientStd. errtp95 % CIpowerPartners’ parenting self-efficacyaFormal social support−0.1510.124−1.210.231−0.394 0.0920.296Age0.3740.7370.510.614−1.071 1.8190.078Education Secondary−35.49112.169−2.920.005−59.342 −11.640.704 Tertiary−10.5687.578−1.390.168−25.421 4.2850.342Depressive symptoms−4.5690.994−4.60>0.001−6.517 −2.6210.974Anxiety symptoms−0.0681.048−0.060.948−2.122 1.9860.116Birth preparation class attendance No9.7029.0461.070.287−8.028 27.4320.144Complication during pregnancy None2.6417.5310.350.727−12.12 17.4020.196Mode of birth Forceps assisted delivery1.41310.0050.140.888−18.197 21.0230.048 Vacuum assisted delivery−1.99.46−0.200.841−20.442 16.6420.092 Planned caesarean section16.41610.81.520.133−4.752 37.5840.338 Emergency caesarean section9.6869.6071.010.317−9.144 . 28.5160.208Birth satisfaction−0.0360.608−0.060.953−1.228 1.1560.054Postpartum complication None11.1038.4581.310.194−5.475 27.6810.262Postpartum length of stay4.3392.4361.780.079−0.436 9.1140.428Time elapsed between birth and questionnaire completion−0.2120.419−0.510.614−1.033 0.6090.124Overnight stay No−6.916.172−1.120.267−19.007 5.1870.162Infant sex Boy−2.3815.622−0.420.673−13.4 8.6380.106Infant birth weight−0.9567.455−0.130.898−15.568 13.6560.07Feeding mode Mixed or formula feeding7.4758.3630.890.375−8.916 23.8660.176Satisfaction with general informal support0.0050.2930.020.988−0.569 0.5790.092Satisfaction with informal informational support0.2960.1831.620.11−0.063 0.6550.396Satisfaction with informal instrumental support−0.1070.218−0.490.627−0.534 0.320.106Satisfaction with informal appraisal support−0.390.194−2.010.049−0.77 −0.010.514Satisfaction with informal emotional support0.5340.2642.020.0480.017 1.0510.562N (63)coefficientStd. error****zp95 %CIpowerFather-infant-bondingbFormal social support0.0060.0041.730.084−0.001 0.0130.53Age−0.0060.019−0.290.771−0.044 0.0330.28Education Secondary0.70.2972.350.0190.12 1.3060.68 Tertiary0.1640.1810.910.363−0.202 0.5190.38Depressive symptoms0.0720.0243.050.0020.025 0.1190.66Anxiety symptoms−0.0290.028−1.040.299−0.084 0.0270.25Complication during pregnancy None−0.2870.199−1.440.149−0.688 0.1040.38Mode of birth Forceps assisted delivery−0.410.245−1.670.095−0.894 0.0840.31 Vacuum assisted delivery0.1360.2390.570.569−0.332 0.620.43 Planned caesarean section0.3040.31.010.311−0.269 0.9280.276 Emergency caesarean section−0.2160.273−0.790.427−0.743 0.3450.212Birth satisfaction0.0190.0161.170.24−0.013 0.0510.42Postpartum complication None−0.0870.221−0.390.694−0.534 0.3450.19Postpartum length of stay−0.010.068−0.150.879−0.146 0.1250.18Time elapsed between birth and questionnaire completion0.0080.0110.690.488−0.014 0.010.45Infant sex Boy0.2520.1381.830.067−0.023 0.5250.5Infant birth weight0.4240.1972.160.0310.034 0.8170.45Feeding mode Mixed or formula feeding0.1210.210.580.565−0.293 0.550.23Satisfaction with general informal support0.0080.0071.070.286−0.007 0.0230.4Satisfaction with informal informational support−0.0030.006−0.610.545−0.015 0.0080.23Satisfaction with informal instrumental support0.010.0061.730.084−0.002 0.0210.62Satisfaction with informal appraisal support0.0010.0050.200.84−0.01 0.0120.4Satisfaction with informal emotional support−0.0180.007−2.530.011−0.033 −0.0040.64aModel fit: Adjusted R² = 0.446, F(25, 67) = 2.157, p = 0.007.bModel fit: R² = 0.5770.Table 8. Partner’s self-efficacy and father-infant bonding estimated marginal means per educational level.Table 8: dummy alt textPartners’ self-efficacyFather-infant bondingEducation LevelEstimated marginal meanStandard Error95 % CIEstimated marginal meanStandard Error95 % CIApprenticeship1997.31185 2140.8090.1760.463 1.155Secondary16411.40141 1861.5080.2840.952 2.064Tertiary1897.14174 2030.9730.1360.706 1.24Results are averaged over the levels of infant sex, mode of infant feeding, complication during pregnancy, mode of birth, postpartum complication. Results are given on the log (not the response) scale.

In the hurdle-gamma regression model, 57.7 % of the variance of father-infant bonding was explained (R^2^ = 0.5770). Infant birth weight emerged as the strongest predictor of father-infant bonding difficulties, with higher birth weight associated with more difficulties (β = 0.424, p = 0.031). Depressive symptoms were significantly associated with more father-infant bonding difficulties (β = 0.072, p < 0.002). Partners with secondary education had more bonding difficulties compared to the “apprenticeship” reference group (β = 0.700, p = 0.019). Mode of birth, infant sex, and satisfaction with informal instrumental support showed marginal associations (p-values ranging from 0.067 to 0.095). Post-hoc power analysis indicated that these effect sizes were underpowered in our sample (see Table 7). Overall model-wide power across all variables was low, with a mean of 39.6 %.

Discussion

4

This study provides new insights into how formal social support provided during the early postpartum hospital stay relates to parenting self-efficacy and bonding in mothers and partners. Overall, parents reported moderate to high levels of perceived formal social support and satisfaction, although satisfaction with the information provided was lower. Formal support was not significantly associated with parenting self-efficacy or parent-infant bonding for either mothers or partners after adjusting for covariates.

Bonding difficulties were relatively low, consistent with previous studies (Stuijfzand et al., 2020; Toivo et al., 2023). Parenting self-efficacy scores were slightly lower than those reported previously (Courtois and Wendland, 2023; Abdollahi et al., 2016). These differences may stem from the timing of the assessment, as most studies measured parenting self-efficacy approximately three months postpartum, whereas our data were collected within the first month, a period of transition to independent caregiving. Evidence suggests parenting self-efficacy declines from the second day after birth to six weeks postpartum (Zheng and Gao, 2023), underscoring the need for continuity of care between hospital and home settings. Healthcare professionals should enhance discharge preparation by ensuring timely post-discharge midwife visits and clear communication channels for parental support (Scroggins et al., 2024). Moreover, the cited studies were conducted in different countries (France, Iran and Canada), which vary in terms of cultural expectations, parental roles, and the organisation of postpartum care. Such contextual differences may also help explain the observed variations in parenting self-efficacy scores.

The majority of mothers and partners reported high satisfaction with formal social support, in line with other European studies (Barimani et al., 2014; Malouf et al., 2019). However, satisfaction with the information received during the postpartum stay was lower. Parents often experience information overload (Loudon et al., 2016; Stoodley et al., 2025) and perceive inconsistent information as unsettling (Schobinger et al., 2022). In McLeish’s study (2021), conflicting advice led some mothers to question professionals’ competencies. Another study in America focusing on mothers’ perspective on supportive professional behaviours and improvement related to the preparation to discharge, found that mothers preferred concise written materials to aid information retention (Scroggins et al., 2024). Providing consistent, prioritised information using diverse formats may therefore improve parental understanding and retention of information (Barimani and Vikström, 2015).

Parents reported moderate to relatively high levels of perceived formal social support responding to their needs; however, wide interquartile ranges indicate that although some parents perceive adequate formal support, others may have perceived it as insufficient or not meeting their needs. Therefore, tailored approaches are essential to ensure all parents receive sufficient support that effectively addresses their needs during the postpartum period (Schobinger et al., 2024).

Relationship between formal social support, parenting self-efficacy, and bonding

4.1

Bivariate analyses found a positive association between formal social support and maternal parenting self-efficacy, but not for partners, partially supporting hypothesis one (H1). This association did not remain significant after adjustment for psychosocial, obstetric, and infant-related factors. Formal social support was only measured in one study, which found no association with parenting self-efficacy (Salonen et al., 2009). In our study, the lack of association for partners may reflect their limited presence on the ward and societal norms positioning them as secondary caregivers (Schobinger et al., 2022). Healthcare professionals need to develop strategies to actively involve partners in postpartum care, including encouraging skin-to-skin contact, and offering tailored information and support that acknowledge partners’ specific needs and caregiving role.

No association was found between formal social support received during the postpartum hospital stay and parent-infant bonding, refuting hypothesis two (H2). Only one previous study reported an association; however, formal social support extended beyond the postpartum stay and was assessed using a different instrument (Wells et al., 2024). One possible explanation for the lack of association in our study is the relatively short duration of the postpartum stay, which may not have been sufficient for formal social support to significantly impact parent-infant bonding, as bonding is a gradual process (Nakić Radoš et al., 2024).

In addition to these findings, in our regression model, maternal parenting self-efficacy was positively associated with the satisfaction with general informal support, emphasising the role of informal social support and its association with parenting self-efficacy (Fang et al., 2021). For partners, satisfaction with informal emotional support was positively associated with paternal parenting self-efficacy, suggesting that informal emotional support plays a vital role for partners in developing confidence in parenting. This finding is consistent with previous studies, reporting that first-time fathers often rely on their partner or close relatives for emotional support, which enhances their parenting self-efficacy (Francis Xavier et al., 2024; Saether et al., 2023).

Conversely, partners who reported higher satisfaction with informal appraisal support tended to have lower parenting self-efficacy. One possible explanation is that actively seeking or relying on evaluative feedback may reflect underlying uncertainty in their parenting abilities or stress. This aligns with previous research showing that fathers with lower self-confidence or higher stress exhibit reduced parenting self-efficacy (Fang et al., 2021). Although partners in this study reported being satisfied with such feedback, its effect may sometimes be negative if the support is not delivered appropriately, highlighting the importance of considering not only the availability but also the quality and context of social support (Saether et al., 2023).

The non-significant association between formal social support and maternal parenting self-efficacy in the multivariate analysis may indicate that the relationship is more complex than a direct effect. For example, parenting self-efficacy might be influenced indirectly through other variables such as satisfaction with informal support or mental health status, suggesting possible mediation effects. Because our analyses did not test for mediation explicitly, future research using mediation analysis or structural equation modelling could better clarify these pathways. Overall, our findings support hypothesis three (H3), highlighting the important role of informal social support in parenting self-efficacy, particularly for mothers and, to some extent partners.

Other factors associated with parenting self-efficacy and parent-infant bonding

4.2

Beyond social support, other factors were associated with parenting self-efficacy. Partners with secondary education reported significantly lower parenting self-efficacy than those with apprenticeship training. While previous study suggests that higher education may be associated with lower paternal self-efficacy due to elevated expectations (Donithen and Schoppe-Sullivan, 2022), this was not observed among university-educated partners in our study. Their potentially higher standards may have been offset by greater socioeconomic stability or more flexible work–family arrangements. In contrast, secondary education may provide neither the practical orientation of apprenticeship nor the resources of university, leaving partners more vulnerable to reduced self-efficacy. In Switzerland, apprenticeship programmes are particularly practice-oriented, which may foster transferable problem-solving skills and confidence relevant to early caregiving. Nonetheless, this interpretation should be considered exploratory, as a systematic review concluded that the evidence linking education and parenting self-efficacy is mixed (Fang et al., 2021). Future research should explore whether hands-on caregiving experience, occupational self-concept, or peer support networks mediate this effect.

Father-infant bonding was also linked to education: partners with secondary education reported more bonding difficulties compared to those with apprenticeship training. Evidence regarding the association between parental education and bonding remains inconsistent, with some studies reporting greater involvement in infant care, among more educated fathers which could potentially facilitate bonding (Aslan et al., 2017), whereas others report no association (Bieleninik et al., 2021). These findings suggest that education may influence bonding indirectly, potentially through parenting self-efficacy. Indeed, higher parenting self-efficacy has consistently been associated with better parent–infant bonding (Qi et al., 2025; Liu et al., 2022).

In addition, father-infant bonding was associated with newborn weight, with partners having heavier infants reporting more bonding difficulties. Although this contrasts with findings form preterm samples (Taing et al., 2020). In our study with term infants, a different mechanism may be at play. A systematic review has identified birth weight as a factor influencing infant temperament (Takegata et al., 2021). Indeed, heavier infants may be more likely to exhibit a difficult temperament (Niegel et al., 2007), which may challenge fathers’ parenting self-efficacy and reduce their confidence in meeting their child’s needs, thereby affecting bonding (Qi et al., 2025). This interpretation remains speculative and warrants further investigation.

Mother-infant bonding was associated with maternal age and infant feeding mode. Older maternal age was linked to more bonding difficulties, consistent with findings from a systematic review reporting negative associations between maternal age and bonding quality (Tichelman et al., 2019). However, a more recent study found that younger maternal age predicted poorer bonding (Doyle et al., 2023). This suggests the relationship between maternal age and bonding is complex and may depend on other contextual factors. Our results may be explained by older mothers experiencing more fatigue, slower recovery, or greater competing responsibilities, which could reduce opportunities for early bonding with their infant (La Rosa et al., 2025). Mothers who did not exclusively breastfeed also reported more bonding difficulties. While evidence on the relationship between breastfeeding and bonding remains complex and debated (Bicking Kinsey and Hupcey, 2013; Doyle et al., 2023), exclusive breastfeeding may facilitate bonding (Nakić Radoš et al., 2024), through increased physical contact (skin-to-skin), inducing oxytocin release, a hormone involved in bonding (Modak et al., 2023). Breastfeeding may also facilitate early parent–infant interactions, thereby enhancing maternal self-efficacy, which in turn is positively related to bonding (Modak et al., 2023; Qi et al., 2025). Conversely, mothers who do not exclusively breastfeed may experience feelings of inadequacy or guilt, particularly if they intended to breastfeed, which can negatively influence their perception of self-efficacy (Russell et al., 2021) and, consequently the quality of their bond.

Finally depressive symptoms were significantly associated with both lower parenting self-efficacy and parent-infant bonding, consistent with previous research (Fang et al., 2021; Albanese et al., 2019). Depression symptoms may impair emotional availability and parental engagement (Döblin et al., 2023; Ngai and Lam, 2023), highlighting the importance of early identification of parental distress. Training healthcare professionals to recognise signs of parental distress and providing appropriate emotional support or referrals to mental health services should be prioritised.

Overall, our findings support hypothesis three (H3), highlighting the important role of informal social support in enhancing parenting self-efficacy, especially among mothers and, to some extent, partners. Additionally, several other factors—such as education, infant feeding mode and depressive symptoms—significantly influence parenting self-efficacy and parent-infant bonding. These results emphasise the complex interplay between these variables and call for tailored formal support that addresses parents’ specific needs during the postpartum period.

Strengths and limitations

4.3

Strengths include the inclusion of both parents and the use of validated questionnaires. However, several limitations should be noted. Fewer partners participated than mothers, and post-hoc power analyses indicated that some regression estimates were underpowered, particularly in partners’ analyses, potentially limiting the detection of associations. Nevertheless, several significant associations were identified between informal support, depressive symptoms, and parenting self-efficacy and bonding, highlighting meaningful relationships despite these limitations. Additionally, most participants were Swiss, highly educated, living in couples, and gave birth in a university hospital, which may limit generalisability to more diverse populations. Potential non-independence between mothers and partners from the same couple could not be accounted for, as pairing information was unavailable. Inherent to cross-sectional design, causal inferences cannot be drawn. Future studies with larger samples of fathers/partners and prospective designs are needed to explore these relationships. Another limitation relates to the analytical approach. Although multivariate regressions were conducted, mediation effects were not examined. Future studies should apply mediation analyses or structural equation modelling to disentangle these complex relationships.

Conclusions

5

This study provides new insights into the association between formal support received during the postpartum stay, parenting self-efficacy, and parent-infant bonding, showing that formal social support was not related to these outcomes, even after accounting for confounding factors. Satisfaction with information received from healthcare professionals was lower than with other types of support. Factors such as depression symptoms and satisfaction with informal social support emerged as key correlates of parenting self-efficacy and parent-infant bonding, emphasising the importance of mental health and informal networks in early parenthood.

Ethics approval and consent to participate

This study was approved by the Human Research and Ethics Committee of the Canton of Vaud, Switzerland (Project No 2021-00762). Written informed consent was not necessary, as the study was categorised as anonymous, and parents were deemed to consent by completing the online survey.

Consent for publication

Not applicable.

Availability of data and materials

The data collected for this study cannot be shared at this time as further analysis is still ongoing, particularly related to the validation of a tool that is currently in preparation. Once the analysis is completed, the data will be deposited in a relevant data repository and made available for citation and linking in the article.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

CRediT authorship contribution statement

Elisabeth Schobinger: Writing – review & editing, Writing – original draft, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Alain Lacroix: Writing – review & editing, Supervision, Software, Formal analysis, Data curation. Anne-Sylvie Ramelet: Writing – review & editing, Validation, Supervision, Methodology, Conceptualization. Antje Horsch: Writing – review & editing, Validation, Supervision, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.