Maternal Stress and Bonding in the COVID-19 Pandemic: Mothers’ Relationship Quality With the Partner as a Resilience Factor

Abstract

Mothers with young children were particularly susceptible to elevated stress levels during the COVID-19 pandemic. This study examined dynamic, reciprocal, longitudinal relationships between maternal perceived stress, perceived relationship quality with the partner, and mother-to-child bonding. The sample consisted of n = 1,394 mothers of children aged 0 to 3 years who completed online questionnaires at three time points between May 2020 to May 2022. We analyzed a Random Intercept Cross-Lagged Panel Model with maternal stress, relationship quality with the partner, and mother-to-child bonding, enabling the separation of stable between-person differences and dynamic within-person processes. At the between-person level, lower partner relationship quality was associated with higher average stress (r = −.20) and impaired mother-to-child bonding (r = −.29), while stress and impaired bonding were strongly correlated (r = .56). At the within-person level, increases in partner relationship quality predicted intraindividual decreases in stress (β_T1→T2 = −.19; β_T2→T3 = −.21) and, at trend level, decreases in impaired bonding (β_T1→T2 = −.15; β_T2→T3 = −.16). The results demonstrate the interrelation of stable components of maternal stress, relationship quality, and bonding. Additionally, they highlight the dynamic role of the maternal perceived relationship quality with the partner in shaping maternal stress and bonding, positioning relationship quality as a resilience (or risk) factor for maternal well-being during crises. An exploratory model including sociodemographic and pandemic-related variables further showed that factors such as maternal education, income, family size, and work changes significantly shaped baseline stress and relational functioning. Hence, support services should focus on strengthening the relationship with the partner while being tailored to families’ structural contexts to mitigate stress, promote bonding, and prevent adverse consequences for mothers, children, and families.

Keywords

perceived stress parental relationship mother-child bonding COVID-19 pandemic family resilience

Introduction

While becoming a parent is often accompanied by positive emotions and a sense of meaning, this transition can also be associated with heightened perceived stress, negative emotions (e.g., worry, anxiety), changes in the couple relationship dynamic, sleep disturbance, and financial strain (Nelson et al., 2014). These challenges are particularly pronounced in the early postpartum period, when parents rely strongly on social support (Saur & Dos Santos, 2021). During the COVID-19 pandemic, however, access to social support was substantially reduced, leaving families to rely primarily on intrafamilial relationships (Buechel et al., 2022; Gao et al., 2022). Within this context, two central relational domains are especially relevant: the perceived quality of the partner relationship, where applicable, and the parent’s emotional connection to their child, commonly conceptualized as bonding (Hahlweg, 2016; Le Bas et al., 2020). Importantly, analyses based on an earlier dataset from the same study addressed in the present work indicate that mothers, in particular, experienced heightened psychological burden, including elevated levels of perceived stress, during the pandemic (Reck et al., 2025; Woll-Weber et al., 2025). Consequently, for mothers in the early postpartum years during the pandemic, both partner relationship quality and mother–child bonding may have functioned as sources of resilience or as additional stressors, depending on their quality. Understanding how maternal stress and intrafamilial relationship processes unfold is therefore critical, especially in crisis contexts.

The Interplay Between Maternal Perceived Stress and Partner Relationship Quality

Numerous studies have shown a link between mothers’ perceived stress and their perceived relationship quality with their partner: Some suggest pathways leading from perceived stress to problems in the parental relationship, including reduced relationship quality (Milek et al., 2017), while others argue for a pathway from partner relationship quality to maternal stress (Martin & Brock, 2023). During the pandemic, the prevalence of elevated stress levels among postpartum women was markedly higher than in pre-pandemic periods (Gao et al., 2022; Suárez-Rico et al., 2021). Furthermore, couples often experienced increased relationship problems and lower relationship quality with the partner compared to pre-pandemic times (Pietromonaco & Overall, 2022; Vigl et al., 2022). In line with pre-pandemic studies, several longitudinal studies during the pandemic found an association between high perceived stress and poor partner relationship quality (Balzarini et al., 2023; Overall et al., 2022). It seems reasonable to posit that pandemic-related stress resulted in negative processes within the couple relationship, such as maladaptive coping with increasing challenges, which could then have impacted partner relationship quality, as suggested by Pietromonaco and Overall (2021). Conversely, circumstances during the pandemic (e.g., lockdowns) and resulting struggles in the partner relationship (e.g., more conflicts) may have contributed to increased stress and mental health problems for mothers (Daugherty et al., 2022; Xie et al., 2022). For instance, results from a cross-lagged panel model (CLPM) in a sample of pregnant women during the pandemic, showed that maternal perceived partner relationship quality predicted later depression and anxiety symptoms, but not vice versa (Xie et al., 2022).

The Link Between Maternal Stress and Mother–To-Child Bonding

Maternal bonding, defined as the mother’s emotional connection to her child, is crucial for healthy child development (Le Bas et al., 2020) and has been repeatedly linked to maternal stress (O'Dea et al., 2023). Similar to difficulties in the relationship between the mother and her partner, it is plausible that elevated maternal stress may negatively affect the mother-to-child bond, while bonding impairments and the resulting negative emotions (e.g., guilt, shame) may, in turn, increase maternal perceived stress (de Cock et al., 2017; Reck et al., 2016; Yazdanimehr et al., 2023). A significantly higher proportion of mothers giving birth during the pandemic reported impairments in the mother–child bond compared with those who gave birth before the pandemic (Fernandes et al., 2021; Mousavi et al., 2021). In addition, studies during the pandemic indicate a positive association between mothers’ perceived stress and impaired bonding (Fernandes et al., 2021; Power et al., 2024). Despite the potential for reciprocal associations between the two constructs, most studies conducted both before and during the pandemic have primarily examined the effects of perceived stress on impaired bonding (O'Dea et al., 2023). In a prior study based on the same sample used in this paper, Reck et al. (2025) analyzed the first two of the three measurement points available at that time using a CLPM and found that impaired bonding predicted subsequent maternal perceived stress during the pandemic, whereas perceived stress did not predict later bonding impairments. These findings highlight the need for approaches capable of detecting not only potential longitudinal effects of stress on bonding, but also effects in the opposite direction.

Partner Relationship Quality and the Mother–Child Bond: Interlinked Intrafamilial Relationships

The mother’s perceived relationship with the partner and maternal bonding are highly interrelated and should therefore not be treated as independent levels within the family system (Cox & Paley, 2003; Tichelman et al., 2019). Poor partner relationship quality seems to be connected to lower-quality parenting practices (Chen et al., 2022; Roos et al., 2021), with potential long-term effects on the parent-child relationship and offspring development (Claussen et al., 2022). On the other hand, mothers’ perceived support by their partner, a crucial aspect of overall relationship quality, may be a protective factor for impaired bonding (Cuijlits et al., 2019). Although studies suggest that both of these intrafamilial relationships were challenged during the COVID-19 pandemic, longitudinal research examining their interrelation during this period remains scarce.

Interconnected Family Dynamics Between Maternal Stress, Partner Relationship Quality, and Bonding: Research Gaps and Study Aims

Although several studies conducted before and during the pandemic indicate longitudinal associations between maternal stress, relationship quality with the partner, and bonding (Balzarini et al., 2023; Power et al., 2024; Tichelman et al., 2019), several important gaps in the literature remain. First, previous studies have often examined only one direction of potential longitudinal effects between two of these constructs, without considering the reverse direction or possible bidirectional dynamics. On the one hand, elevated perceived stress and psychological burden may have led to lower partner relationship quality and impaired bonding (Liu et al., 2022; Overall et al., 2022). On the other hand, high-quality intrafamilial relationships may have served to mitigate perceived stress while negative experiences in intrafamilial relationships potentially increased stress levels and their adverse consequences (Mousavi et al., 2021; Pauly et al., 2022; Reck et al., 2025; Tepeli Temiz & Elsharnouby, 2022). Additionally, mothers’ relationship quality with their partner may be closely linked to the quality of the mother–child bond (Tichelman et al., 2019). Moreover, reciprocal dynamics between these processes may contribute to vicious cycles of heightened stress and deteriorating relationship quality within the family system, particularly when spending prolonged periods together at home due to pandemic-related restrictions. Therefore, examining maternal stress, partner relationship quality, and mother–child bonding within a longitudinal reciprocal framework appears crucial for a more comprehensive understanding of their interrelations during the pandemic.

Second, previous studies examining maternal perceived stress, partner relationship quality, and bonding have rarely focused on intraindividual changes in these constructs over time. While initial studies suggest that intraindividual changes in maternal stress are associated with changes in partner relationship quality, most have only examined their co-occurrence or investigated dynamic processes in one direction (Pauly et al., 2022; Rusu et al., 2020; Tepeli Temiz & Bilican, 2024). Studies examining such dynamic effects involving maternal bonding – or all three constructs simultaneously – remain particularly scarce, and, to the best of the authors’ knowledge, the present study is the first to investigate all three constructs jointly in a longitudinal framework in the context of the COVID-19 pandemic. To better capture these complex and dynamic intrafamilial processes, longitudinal analyses should distinguish between stable between-person differences and within-person fluctuations. Failure to separate these levels of analysis, as occurs in traditional cross-lagged panel models (CLPMs), may lead to misleading conclusions about developmental or causal processes. Advanced methodological approaches such as the Random-Intercept Cross-Lagged Panel Model (RI-CLPM; Hamaker et al., 2015) allow researchers to disentangle between- and within-person effects. While such distinctions are generally important in longitudinal research, they may be particularly relevant in the context of the COVID-19 pandemic, where rapidly changing circumstances and regulations likely contributed to substantial within-person variation in family dynamics and parental stress. A better understanding of these family processes during the COVID-19 pandemic offers valuable insights into the most effective ways of providing support to families during such exceptional times.

Addressing these research gaps, we hypothesized and tested reciprocal longitudinal associations between mothers’ perceived stress, their perceived relationship quality with their partner, and maternal bonding within a model that separates stable between-person differences and dynamic within-person fluctuations. To this end, an RI-CLPM was applied to longitudinal data from a German community sample across three time points during the COVID-19 pandemic.

Method

Participants and Procedure

The data for this study was assessed as part of the longitudinal online survey “CoviFam”, which is a COVID-19-specific subproject of a larger longitudinal observational study (“COMPARE_interaction”). The online survey link was offered to a community sample of parents with children between 0 and 3 years in Germany, Austria, and Switzerland to assess their psychosocial well-being during the COVID-19 pandemic. Parents were recruited via medical institutions (e.g., birth clinics, pediatricians, gynecologists), professional regional and national networks (e.g., midwifes, nurses), and different social media networks (i.e., Instagram, Twitter, Facebook). Data was collected from May 2020 to May 2022. The first measurement point (T1) spanned from May to November 2020. The online survey link was reopened and participants of the first wave were contacted again in February and March 2021 for a second measurement point (T2) and in February until May 2022 for a third measurement point (T3), when regulations were stricter again (see Figure 1 for the severity of regulations). Participants provided their informed consent online. The study was approved by the ethics committee of the medical faculty, Ruprecht-Karls-University, Heidelberg, in agreement with the Ludwig Maximilian University, Munich (vote: S-446/2017).

Figure 1.

Covid-19 situation in Germany from January 2020 to May 2022. Trajectories of cases, deaths, and government response during the COVID-19 pandemic in Germany. Confirmed deaths are shown cumulatively. Grey areas represent the study measurement times. Figure based on the Oxford COVID-19 Government Response Tracker (https://github.com/OxCGRT/covid-policy-tracker/blob/master/documentation/index_methodology.md)

Both mothers and fathers were asked to participate in the online survey (T1: total N = 1,936 between May and November 2020). In this sample, n = 1,744 participants identified as female, n = 185 as male, n = 2 as diverse (non-binary/other), and n = 2 participants did not want to indicate a gender. Due to the small number of participants identifying as male, diverse or who did not report a gender, and the resulting problems thereof for multigroup analyses, only participants identifying as female were included. For this study, given that restrictions and regulations regarding the pandemic were quite different between countries, participants living in Austria and Switzerland were excluded (n = 91) to investigate a more homogeneous sample. Further, n = 42 participants who reported to be single at all time points, and participants with children older than 3 years at T1 (n = 196), who misunderstood the target age group of the study, were excluded. This resulted in n_T1 = 1,415, n_T2 = 593, and n_T3 = 340. During data preparation, discrepancies in key stable sociodemographic information (e.g., birthdays of parent or child) within some cases were discovered. This revealed that, in these cases, the surveys were not consistently filled out by the same parent (e.g., T1 by the mother and T2 by the father) or for the same study child longitudinally (e.g., T1 for their first child, and T2 for their second child). Parents with inconsistent data were contacted again to clarify and identify inconsistencies stemming from simple typos. For the remaining cases, that were either proven to be filled out incorrectly or could not be clarified, data for the respective time point were set to missing data. For instance, in cases when participants reported the same birthday for their study child at T1 and T3, but a different birthday at T2, data from T2 was deleted. This resulted in the deletion of data from n = 21 cases at T1, n = 49 cases at T2, and n = 35 cases at T3. Accordingly, data were available from n_T1 = 1,394, n_T2 = 544, and n_T3 = 305 (see Table 1 for demographic information). All available data were included in the RI-CLPM analysis using Full Information Maximum Likelihood (FIML) estimation, which allows cases with missing data, including dropouts, to be used in the analysis.

Table 1.

Demographic Information of the Sample (n_T1 = 1,394, n_T2 = 544, n_T3 = 305)

Variables	T1Mean (SD)	T2Mean (SD)	T3Mean (SD)
Maternal age (in years)	33.21 (4.26)	34.35 (4.28)	35.28 (4.30)
Child age (in months)	17.57 (9.47)	23.31 (8.99)	35.36 (9.33)
Number of children in household	1.56 (0.74)	1.64 (0.82)	1.77 (0.78)
Gestational age (in weeks)	39.76 (1.88)	39.78 (1.79)	39.90 (1.51)

	n (%)	n (%)	n (%)
Child sex assigned at birth (T1) - Female - Male	675 (48.42) 719 (51.58)	265 (50.67) 258 (49.33)	145 (50.88) 140 (49.12)
Household income (monthly in Euro, T1) - 0–1,000 - 1,000–2,000 - 2,000–3,000 - 3,000–5,000 - > 5,000	12 (0.86) 106 (7.60) 337 (24.18) 666 (47.78) 268 (19.23)	5 (0.96) 30 (5.74) 117 (22.37) 250 (47.80) 118 (22.56)	2 (0.70) 18 (6.32) 60 (21.05) 134 (47.02) 69 (24.21)
Change in household income due to the pandemic (T1) - No - Yes	1,035 (74.25) 359 (25.75)	398 (76.10) 125 (23.90)	221 (77.54) 64 (22.46)
Maternal education (T1) - (Technical) university entrance qualification - High secondary qualification - Low secondary qualification - Left school without qualification	1,187 (85.15) 188 (13.49) 18 (1.29) 1 (0.07)	486 (89.34) 54 (9.93) 4 (0.74) -	271 (88.85) 31 (10.16) 3 (9.84) -
Maternal work status - Parental leave - Employed - Public servant - University student - Self-employed - Housewife - Apprenticeship - Unemployed - Other	665 (47.70) 435 (31.21) 73 (5.24) 51 (3.66) 43 (3.08) 46 (3.30) 9 (0.65) 8 (0.57) 64 (4.59)	183 (33.64) 218 (40.07) 44 (8.09) 25 (4.60) 27 (4.96) 16 (2.94) 2 (0.37) 8 (1.47) 21 (3.86)	71 (23.27) 154 (50.49) 26 (8.52) 14 (4.59) 19 (6.23) 5 (1.64) 2 (0.66) 3 (0.98) 11 (3.60)
Change in work situation due to the pandemic (e.g., reduced hours, remote work, T1) - No - Yes	824 (59.11) 570 (40.89)	286 (54.68) 237 (45.32)	157 (55.09) 128 (44.91)
Maternal family status - Married (living together) - In a relationship (living together) - In a relationship (not living together) - Single - Divorced - Widowed	1,069 (76.69) 307 (22.02) 15 (1.08) - 2 (0.14) 1 (0.07)	437 (80.33) 102 (18.75) 4 (0.74) 1 (0.18) - -	258 (84.59) 40 (13.11) 4 (1.31) 1 (0.33) 2 (0.66) -

Note. T1, T2, and T3 represent the first, second, and third measurement point, respectively. Where totals deviate from the full sample size, this is due to missing information not provided by participants.

Regarding attrition analysis, first, T1 data of n = 544 mothers who participated at both T1 and T2 (and optionally also T3) were compared to T1 data of mothers who only participated at T1 (n = 818): The groups did not differ significantly on most sociodemographic and outcome variables at T1: Differences were only found for (1) mothers’ age (t = −4.83, p < .001), with mothers in the longitudinal sample being slightly older (M_{T1andT2/T1toT3} = 33.88 years, M_onlyT1 = 32.73 years, d = 0.27). The difference for (2) perceived stress was negligible in effect size with mothers in the longitudinal sample reporting slightly lower values (t = 2.52, p = .012; d = 0.14; M_{T1andT2/T1toT3} = 18.04, M_onlyT1 = 19.04). For (3) school degree, (4) monthly net income, (5) family status, and (6) changes in occupation due to the pandemic, the significant differences were negligible in effect size (Wilcoxon’s W = 19,259, p < .001, Cliff’s delta = −0.12; W = 194,584, p = .003, Cliff’s delta = −0.09; χ² (4) = 9.38, p = .032, Cramer’s V = 0.08; χ² (1) = 6.97, p = .008, Cramer’s V = 0.07, respectively). The first step of the attrition analysis indicates no meaningful differences between mothers who continued after T1 and those who dropped out after T1, with respect to the assessed sociodemographic and outcome measures. The second step of the attrition analysis also indicated no meaningful differences between mothers who continued after T2 (n = 232) and those who dropped out after T2 (n = 292), with respect to the assessed sociodemographic and outcome measures at T2: The groups did not differ significantly on any sociodemographic or outcome variables, except that children of mothers who took part at all three time points were slightly younger, with a negligible effect size (Wilcoxon’s W = 37,294, p = .047, d = 0.18, M_T1toT3 = 22.66 months, M_T1toT2 = 24.27 months).

Measures

Multiple constructs of family context and general demographics were assessed in the aforementioned larger online survey. For this study, maternal perceived stress, relationship quality with the partner, and impaired bonding were the constructs of interest.

Perceived Stress

Mothers’ self-reported perceived stress was assessed using the German version of the Perceived Stress Scale (PSS-10) as published by Reis et al. (2019) (original version by S. Cohen et al. (1983)). The instrument contains 10 items (e.g., “In the last month, how often have you been upset because of something that happened unexpectedly?”). Items are answered on a 5-point Likert scale (0/“never” to 4/“very often”). Higher sum scores, ranging from 0 to 40, indicate higher perceived stress. The internal consistencies of the scale were equal or above 0.88 at all time points (Cronbach’s α = 0.88 at T1, α = 0.90 at T2, α = 0.90 at T3).

Relationship Quality With the Partner

For the assessment of mothers’ perceived relationship quality with their partner, the abridged version of the German partnership questionnaire “Partnerschaftsfragebogen” (PFB-K, Hahlweg, 2016) was applied. Nine of its 10 items describe the partner’s potential behavior within a relationship (e.g., “He/She tells me that he/she likes me”) or mutual/joint couple behavior (e.g., “We talk to each other for at least half an hour in the evening”). Participants were asked to indicate on a 4-point Likert scale from 0 to 3 (“never/very rarely” to “very often”) how often such behavior occurs in their relationships. A 10th item assessed overall couple relationship satisfaction and was rated on a 6-point scale from 0 to 5 (“very unhappy” to “very happy”). This 10th item is not part of the sum score and was therefore excluded from further analysis. The PFB-K consists of three sub-scales (i.e., disagreement, affectionateness, and communication). An overall sum score was calculated as follows: [(9-disagreement) + affectionateness + communication] (Hahlweg, 2016). Higher overall sum scores, ranging from 0 to 27, indicate higher relationship quality. The internal consistencies of the scale were equal or above 0.85 at all time points (Cronbach’s α = 0.85 at T1, α = 0.87 at T2, α = 0.88 at T3).

Maternal Impaired Bonding

To assess maternal impaired bonding, the German version of the Postpartum Bonding Questionnaire (PBQ-16, Reck et al., 2006), originally developed by Brockington et al. (2001), was applied. The items (e.g., “I feel distant from my baby”) are scored on a 6-point Likert scale (0 for “always” to 5 for “never”). Higher sum scores, ranging from 0 to 80, represent more impaired bonding, hence lower maternal bonding quality. The internal consistencies of the scale were equal or above 0.85 at all time points (Cronbach’s α = 0.85 at T1, α = 0.87 at T2, α = 0.86 at T3). It should be noted that the PBQ-16 has been validated only for use during the first year postpartum. Given that our sample included mothers of children aged 0–3 years, we applied a slightly adapted version in which the term “baby” was replaced by the more general term “child.” Based on expert consensus, this adaptation was deemed the best available proxy for assessing the mother–child relationship across the full age range studied, as alternative validated instruments are only applicable for toddlerhood onward. To address potential age-related differences, child age was therefore included as a control variable in an exploratory model.

Statistical Analyses

All analyses were conducted in R, version 4.5.1 (R Core Team, 2025) using RStudio, version 2025.5.1.513 (Posit team, 2025) applying the lavaan package (Rosseel, 2012) as the primary tool for structural equation modeling. In preliminary analyses, paired t-tests with Cohen’s d (J. Cohen, 1988) were used to examine changes in the main outcome measures across time points. Dynamic longitudinal relationships between the model variables (i.e., perceived stress, relationship quality, and impaired bonding) were investigated by applying an RI-CLPM over three measurement points (T1, T2, T3). This approach enabled the separation of stable trait-like between-person differences (captured in a random intercept for each model variable) and dynamic intraindividual within-person processes (captured by a latent factor for each model variable for each time point). Parameter estimates were obtained by a maximum-likelihood estimation with robust estimators of model fit (MLR), robust (Huber-White) standard errors, and a scaled test statistic that is (asymptotically) equal to the Yuan-Bentler test statistic, as it is robust against the violation of normal distribution (Rosseel, 2020). A full-information maximum-likelihood (FIML) approach was applied to compensate for missing data. At the item level, however, no missing data occurred in the computation of sum scores, as all items of the main questionnaires were mandatory in the online survey, in contrast to sociodemographic variables that were not compulsory and could therefore contain missing values.

The random intercept of each variable was identified by its observed score, with factor loadings constrained to 1. Error variances for the observed scores were constrained to zero, ensuring that all variance was captured by the latent factor structures. Model parsimony was evaluated by comparing an RI-CLPM with freely estimated stability and cross-lagged paths to one with paths constrained to be equal over time, testing whether the stability and cross-lagged effects from T1 to T2 were equivalent to those from T2 to T3. To evaluate the quality of model fits, a range of fit indices was inspected, including the comparative fit index (CFI), the Tucker Lewis Index (TLI; also called the non-normed fit index), the root-mean-square error of approximation (RMSEA), and the standardized root-mean-square residual (SRMR). In line with Hu and Bentler (1999) and Little and Kline (2016), cut-off values for good model fit were CFI >.90, TLI >.90, SRMR <.09, and RMSEA <.08. Cut-offs for excellent model fit were CFI >.95, TLI >.95, SRMR <.08, and RMSEA <.06. As common model fit indices may be biased due to the large numbers of parameters and variables, additional reliance is placed on the unbiased standardized root-mean-square error of approximation (uSRMR; Shi et al., 2018; Ximénez et al., 2022). Intraclass correlations (ICCs) were computed for all three model variables to examine the proportion of variance attributable to stable between-person differences versus within-person processes.

Transparency and Openness

The study was pre-registered prior to analyses (see https://aspredicted.org/DHR_OLH). Originally, a traditional CLPM was preregistered because only two of the three measurement points were available at time of preregistration. However, with the third measurement point available, enabling the computation of an RI-CLPM, we decided to deviate from the preregistration, given the major methodological benefits of the RI-CLPM over traditional CLPMs (Hamaker et al., 2015). Neither the model variables nor the research question on their longitudinal associations were modified.

Results

Preliminary Analyses Regarding Maternal Perceived Stress, Partner Relationship Quality, and Impaired Bonding

Means and standard deviations for the PSS-10, PFB-K, and PBQ-16 can be seen in Table 2. Table 2 also summarizes the results of the paired t-tests, which revealed a significant difference for all measures between time points (ps < .001, small to medium effect sizes), except for the comparison of partner relationship quality between T2 and T3.

Table 2.

Mean, Standard Deviations, and t-Test Results of the Perceived Stress Scale (PSS-10), Partnership Questionnaire (PFB-K), and the Postpartum Bonding Questionnaire (PBQ-16) at the Three Measurement Times

Measure	T1	T2	T3	T1 to T2		T2 to T3		T1 to T3
Measure	M (SD)	M (SD)	M (SD)	p	d	p	d	p	d
Perceived stress	18.64 (7.00)	21.18 (7.13)	19.71 (7.22)	<.001	0.51	<.001	0.29	<.001	0.22
Partner relationship quality	18.03 (5.23)	17.52 (5.52)	17.23 (5.57)	<.001	0.20	.053	0.12	<.001	0.25
Impaired bonding	11.87 (6.93)	13.94 (7.91)	15.95 (7.47)	<.001	0.34	<.001	0.74	<.001	0.42

Note. Higher impaired bonding scores (PBQ-16) represent a more impaired bonding, i.e., lower bonding quality. d = Effect size Cohen’s d, representing the size of the difference between measurement times.

Correlations between the three main variables of interest across all measurement points are presented in Table 3. All three main variables were significantly intercorrelated at each time point, with absolute values ranging from .10 to .47. Significant correlations greater than .10 between the main variables and the sociodemographic or pandemic-related variables are presented in Table 4. Perceived stress was associated with child age, number of children in the household, maternal education level, household income, and change in the work situation. Perceived relationship quality with the partner was associated with maternal age, child age, and number of children in the household. Impaired bonding was associated with child age, maternal education level, and change in the work situation.

Table 3.

Pearson Correlations of Perceived Stress, Partner Relationship Quality, and Impaired Bonding at the Three Measurement Times

	[1]	[2]	[3]	[4]	[5]	[6]	[7]	[8]
[1] Perceived stress T1	-
[2] Perceived stress T2	.64*** [0.59, 0.69]	-
[3] Perceived stress T3	.55*** [0.46, 0.63]	.59*** [0.5, 0.67]	-
[4] Partner relationship quality T1	−.32*** [−0.37, −0.27],	−.26*** [−0.34, −0.18]	−.27*** [−0.37, −0.16]	-
[5] Partner relationship quality T2	−.28*** [−0.36, −0.2]	−.32*** [−0.39, −0.24]	−.28*** [−0.39, −0.16]	.81*** [0.78, 0.84]	-
[6] Partner relationship quality T3	−.15* [−0.26, −0.03]	−.20** [−0.32, −0.08]	−.29*** [−0.39, −0.18]	.70*** [0.63, 0.76]	.78*** [0.73, 0.82]	-
[7] Impaired bonding T1	.41*** [0.37, 0.45]	.38*** [0.3, 0.45]	.35*** [0.24, 0.45]	−.23*** [−0.28, −0.18]	−.18*** [−0.26, −0.1]	−.10^† [−0.21, 0.02]	-
[8] Impaired bonding T2	.37*** [0.29, 0.44]	.47*** [0.4, 0.53]	.37*** [0.26, 0.47]	−.24*** [−0.32, −0.16]	−.27*** [−0.35, −0.19]	−.18** [−0.3, −0.06]	.75*** [0.71, 0.79]	-
[9] Impaired bonding T3	.33*** [0.22, 0.43]	.33*** [0.22, 0.44]	.47*** [0.38, 0.55]	−.18** [−0.29, −0.06]	−.18** [−0.3, −0.06]	−.21*** [−0.32, −0.1]	.66*** [0.59, 0.72]	.71*** [0.64, 0.77]

Note. Values in square brackets indicate the 95% confidence interval for each correlation. Notably, higher impaired bonding scores (PBQ-16) represent a more impaired bonding, i.e. lower bonding quality.

†p < .10, *p < .05, **p < .01, ***p < .001.

Table 4.

Significant Correlations Above .10 Between Main and Sociodemographic/Pandemic-Related Variables

	Maternal age	Child age	Nr of children	Education	Income	Change in work situation
Perceived stress T1	-	T1: .12***	T1: .11***	T1: −.12***	T1: −.18***	T1: .10***
Perceived stress T2	-	T2: .13**	-	T3: −.22***	T2: −.13**	T2: .18***
Perceived stress T3	-	-	-	-	T3: −.18**	-
Partner relationship quality T1	T1: −.12***		T1: −.11***	-		-
Partner relationship quality T3	-	-	-	-	-	T1: −.13*
Impaired bonding T1	-	T1: .26***	-	-	-	T1:.12***
Impaired bonding T2	-	T2:.17***	-	T2: .14**	-	T2: .11*
Impaired bonding T3	-	T3: .18**	-	-	-	-

Note. Higher impaired bonding scores (PBQ-16) represent a more impaired bonding, i.e., lower bonding quality. All remaining correlations can be obtained from the published dataset and R script. *p < .05, **p < .01, ***p < .001.

Between- and Within-Person Effects Between Maternal Perceived Stress, Partner Relationship Quality, and Impaired Bonding (RI-CLPM)

First, the unconstrained RI-CLPM with freely estimated stability and cross-lagged paths was compared to a model with paths constrained to be equal over time. Both models exhibited excellent global fit according to commonly used fit indices (free model: χ² (3) = 2.36, p = .501 CFI = 1.000, TLI = 1.006, RMSEA = 0.000 (90% CI: 0.000-0.095), SRMR = 0.011, uSRMR = 0.011; constrained model: χ² (18) = 15.48, p = .629, CFI = 1.000, TLI = 1.004, RMSEA = 0.000 (90% CI: 0.000-0.044), SRMR = 0.023, uSRMR = 0.023). Model fit was evaluated using robust (scaled) χ² and fit indices to account for non-normality. The constrained model was retained because the equality constraints did not significantly worsen fit, Δχ² (15) = 13.10, p = .59, and because it showed lower AIC (41,753.340 vs. 41,768.907) and BIC (41,942.516 vs. 42,036.906) values than the free model, indicating greater parsimony. Although the TLI exceeded 1 in both models – suggesting slight over-fitting – the constrained model’s TLI was closer to 1, indicating a lesser risk of overfitting. The selected constrained model with standardized coefficients is displayed in Figure 2. Note that model parameters were constrained to be time-invariant in their unstandardized form; however, the standardized coefficients varied across time points due to differences in the relative variances of the variables at each time point.

Figure 2.

Results of the random-intercept cross-lagged panel model. Outer gray area labeled “between-person differences” incorporates the random intercepts connected by non-directional paths. Inner area labeled “within-person processes” represents the interindividual changes connected by directional paths at each time point. Dashed lines indicate statistically non-significant effects, solid lines statistically significant effects. BON = impaired bonding, REL = partner relationship quality, PS = perceived stress. ***p < .001; **p < .01; *p < .05; ^†p < .1

Intraclass correlations (ICCs) indicated that 50.4% of the variance in perceived stress (ICC_{PerceivedStress} = 0.504), 66.3% of the variance in partner relationship quality (ICC_{RelationshipQuality} = 0.663), and 77.6% of the variance in impaired bonding (ICC_{ImpairedBonding} = 0.776) were attributable to stable between-person differences. The remaining 49.6%, 33.7%, and 22.4% of the variance, respectively, reflected within-person fluctuations over time.

At the between-person level, all three random intercepts were significantly associated (see Figure 2): Mothers with generally higher relationship quality with their partner reported lower overall levels of perceived stress and impaired bonding, while higher average levels of perceived stress were linked to generally more impaired bonding.

Regarding within-person processes, autoregressive effects were significant for all three variables between all time points with standardized coefficients ranging from β = 0.20 to β = .45 (see Figure 2). Furthermore, all within-person cross-sectional correlations (range: │r│ = .18–.35) were statistically significant, except for the association between impaired bonding at T1 and both perceived stress (p = .17) and relationship quality with the partner (p = .06) at T1. These effects suggest that when relationship quality improves, perceived stress and impaired bonding decrease simultaneously, while increases in perceived stress and impaired bonding co-occur (at T2 and T3).

Most interestingly, a significant cross-lagged effect (β_T1→T2 = −.19; β_T2→T3 = −.21; p = .03) of perceived relationship quality with the partner on perceived stress was found, indicating that intraindividual increases in relationship quality, relative to the mother’s “typical” average or trait, predicted subsequent decreases in perceived stress. Conversely, intraindividual decreases in relationship quality were associated with subsequent increases in perceived stress (see Figure 2). Furthermore, the cross-lagged effect of perceived relationship quality on impaired bonding was marginal (β_T1→T2 = −.15; β_T2→T3 = −.16; p = .055), suggesting that intraindividual increases in relationship quality may predict decreases in impaired bonding, and vice versa. Based on the benchmark values provided by Orth et al. (2024), these cross-lagged effects are classified as large in magnitude (β > .12). All remaining cross-lagged effects were nonsignificant.

Finally, two robustness checks were conducted. First, comparing the model based on the full sample, including all available participants (n = 1,394), with the model based on a restricted sample, including only participants with data from at least two time points (n = 596), yielded nearly identical results and parameter estimates. The only minor difference was that the autoregressive effect of perceived stress was significant in the full sample (p = .048) but fell just below the significance threshold in the restricted sample (p = .052).

Second, an exploratory RI-CLPM was performed entering sociodemographic/pandemic-related variables as control variables on the between-person level. The specification of predictive pathways was based on the prior correlation analysis (see Table 4). Hence, child age, number of children in the household, maternal education, household income, and change in work situation due to the pandemic at T1 were added as predictors of the random intercept of perceived stress. For the random intercept of perceived relationship quality with the partner, maternal age and the number of children in the household at T1 were entered as predictors. Lastly, child age, maternal education, and change in work situation due to the pandemic at T1 were included as predictors of the random intercept of impaired bonding. Regarding the model fit, the Chi² test was significant, χ² (62) = 90.995, p = .010, which is common in large samples. However, other fit indices indicated excellent fit (CFI = 0.991, TLI = 0.988, RMSEA = 0.019 (90% CI: 0.009-0.026), SRMR = 0.031, uSRMR = 0.052). Following the same pattern as in the main model, all three random intercepts were significantly related (see Table 5). Furthermore, all sociodemographic control variables significantly predicted the respective random intercepts, with the exception of changes in work situation due to the pandemic, which did not significantly predict the random intercept of impaired bonding (see Table 5 for all regression results). Since control variables were added at the between-person level, no substantial changes to within-effects were expected. In fact, the within-person cross-lagged effect of relationship quality on perceived stress, which was significant in the main model, remained virtually unchanged and large in magnitude (β > .12; Orth et al., 2024), but narrowly missed conventional significance in the controlled model (p = .051). This suggests that while the effect is likely present, it is estimated with reduced precision in the controlled model, most likely due to increased standard errors following the inclusion of control variables. Consistent with the main RI-CLPM, the within-person cross-lagged effect of relationship quality on impaired bonding was marginal (p = .057) and large in size.

Table 5.

Results of the Exploratory RI-CLPM With Sociodemographic and Pandemic-Related Control Variables on the Between-Person Level

Paths	Estimate (β)	SE	p
Between-person effects
RI_PS ↔ RI_REL	−0.290	2.097	.005
RI_PS ↔ RI_BON	0.567	2.130	<.001
RI_REL ↔ RI_BON	−0.173	1.801	.017
RI_PS ← - Child age - Number of children in household - Maternal education - Income - Changes in work situation due to pandemic	0.115 0.128 −0.117 −0.163 0.113	0.018 0.215 0.221 0.182 0.347	.001 <.001 .001 <.001 .001
RI_BON ← - Child age - Maternal education - Changes in work situation due to pandemic	0.255 0.074 0.058	0.020 0.229 0.372	<.001 .010 .053
RI_REL ← - Maternal age - Number of children in household	−0.118 −0.074	0.033 0.205	<.001 .037
Within-person effects
Autoregressions: W_PS T1 → W_PS T2 W_REL T1 → W_REL T2 W_BON T1 → W_BON T2 W_PS T2 → W_PS T3 W_REL T2 → W_REL T3 W_BON T2 → W_BON T3	0.194 0.389 0.208 0.206 0.436 0.319	0.101 0.452 0.123 0.101 0.114 0.123	.039 <.001 .007 .039 <.001 .007
Cross-lagged effects: W_PS T1 → W_REL T2 W_PS T1 → W_BON T2 W_REL T1 → W_PS T2 W_REL T1 → W_BON T2 W_BON T1 → W_PS_T2 W_BON T1 → W_REL T2 W_PS T2 → W_REL T3 W_PS T2 → W_BON T3 W_REL T2 → W_PS T3 W_REL T2 → W_BON T3 W_BON T2 → W_PS_T3 W_BON T2 → W_REL T3	−0.076 −0.012 −0.162 −0.142 0.029 −0.039 −0.080 −0.012 −0.185 −0.158 0.032 −0.060	0.049 0.066 0.144 0.126 0.097 0.050 0.049 0.066 0.144 0.126 0.097 0.050	.257 .852 .051 .057 .731 .402 .257 .852 .051 .057 .731 .402

Note. RI = Random Intercept, CLPM = Cross-lagged Panel Model, SE = Standard Error, PS = Perceived Stress, REL = Partner Relationship Quality, BON = Impaired Bonding (Higher impaired bonding scores (PBQ-16) represent a more impaired bonding, i.e., lower bonding quality.), W = Latent Within-Person Factor.

Discussion

This study examined the dynamic longitudinal reciprocal relations between perceived stress, relationship quality with the partner, and impaired mother-to-child bonding in mothers with children between 0–3 years of age during the COVID-19 pandemic from May 2020 to May 2022. By applying an RI-CLPM, effects at the between- and within-person level could be separated, allowing the disentanglement of stable interindividual differences from dynamic intraindividual processes and thus providing a more precise understanding of the associations under investigation.

A significant link between all three random intercepts – describing stable between-person traits – was revealed: While perceived stress and impaired bonding were positively linked, relationship quality with the partner was negatively associated with both perceived stress and impaired bonding. At the within-person level, intraindividual increases in relationship quality predicted subsequent decreases in perceived stress. Further, trend-level evidence points to decreasing impaired bonding following increases in perceived relationship quality with the partner.

The results regarding the random intercepts indicate that mothers who, on average, experience a lower relationship quality with their partner also tended to report higher perceived stress and impaired bonding, and mothers with generally higher perceived stress typically reported more impaired bonding. Notably, these effects remained significant and almost equal in size when controlling for sociodemographic or pandemic-related factors. These results are in line with previous studies that identified associations between perceived stress, partner relationship quality, and maternal bonding before (Martin & Brock, 2023; Tolja et al., 2020) and during the COVID-19 pandemic (Mousavi et al., 2021; Overall et al., 2022; Pietromonaco & Overall, 2022; Power et al., 2024; Xie et al., 2022). Importantly, the correlations between random intercepts do not imply a temporal direction; rather, they suggest that enduring individual differences in one construct are associated with enduring individual differences in the other, possibly due to reciprocal long-term influences or shared underlying factors. While all random intercepts were significantly correlated, the correlation between those of perceived stress and impaired bonding was particularly large. At the same time, perceived stress and impaired bonding were not connected via cross-lagged paths at a within-person level. This suggests that their link is mainly rooted in the association of their stable, trait-like components, rather than in intraindividual changes, potentially reflecting enduring personal or contextual factors that influence both perceived stress and bonding.

At the within-person level, intraindividual fluctuations in mothers’ perceived relationship quality with the partner predicted fluctuations in perceived stress: On the one hand, the direction of effects indicates that increases in relationship quality predicted decreased perceived stress, suggesting a protective effect of high-quality partner relationships during the COVID-19 pandemic. On the other hand, the results can also be interpreted in reverse with intraindividual decreases in perceived relationship quality (relative to the mothers’ own typical average) being associated with subsequent increases in perceived stress. These results are in line with previous studies identifying a link between intraindividual changes in partner relationship quality and stress: In a pre-pandemic study, greater fluctuations in partner relationship quality over time were associated with poorer individual well-being (i.e., increased psychological distress and reduced life satisfaction), but not vice versa (Whitton et al., 2014). Similarly, findings from a traditional CLPM in a sample of pregnant women during the COVID-19 pandemic showed that maternal perceived partner relationship quality predicted later symptoms of depression and anxiety, whereas the reverse effects were not significant (Xie et al., 2022).

Furthermore, Pauly et al. (2022) found that changes in perceived stress and relationship quality from pre-to during the pandemic were significantly correlated with each other. However, several previous studies only investigated effects of fluctuating stress on the partner relationship without considering reciprocal effects (Rusu et al., 2020; Tepeli Temiz & Bilican, 2024). Our results, framed within a reciprocal-effects model, show that intraindividual fluctuations in relationship quality influenced subsequent perceived stress, but not the other way around. By accounting for participants’ stable levels, these findings provide important new insights into the dynamic processes unfolding during the COVID-19 pandemic. It must be noted that this effect was only marginal in the exploratory RI-CLPM that included sociodemographic control variables (p = .051). Importantly, the effect sizes remained similar in magnitude, suggesting that the shift in statistical significance was likely due to the influence of the control variables on the estimation and standard errors of the within-person paths. This is consistent with the notion that statistical significance is sensitive to model complexity, whereas the underlying substantive process appears robust. At the same time, caution is warranted, as the effect narrowly failed to reach the conventional significance threshold, raising the possibility that it may be sample-specific or less stable than suggested. Future studies are therefore needed to corroborate the robustness of these findings.

Further, the results indicated a potential effect such that when mothers reported higher (or lower) partner relationship quality than usual, this tended to predict subsequent decreases (or increases) in impaired bonding. This potential effect was only marginal (p = .055), underscoring the need for further replication in future studies. Nevertheless, this finding is in line with previous studies that found positive longitudinal associations between partner relationship quality – particularly regarding partner support – and maternal bonding, with subsequent effects on infant and child development (Cuijlits et al., 2019; de Waal et al., 2023).

These results underscore that, for mothers, the relationship with the partner functioned as a resilience or risk factor during the pandemic, not only at a stable between-person level but also in terms of within-person dynamics. In particular, fluctuations in the perceived security and quality of the relationship appear to be crucial for maternal stress levels and, potentially, for maternal bonding. In line with this, previous studies emphasize the relevance of considering both general levels and dynamic changes in couple relationship functioning during the pandemic: While couples with a generally better dyadic coping have been found to report fewer fights and less disagreement during the pandemic, initially stronger dyadic coping did not buffer against increases in conflict (Lee et al., 2022). Hence, challenges posed by the COVID-19 pandemic often increased the risk for elevated conflict even for couples with previously high relationship quality (Lee et al., 2022). Because of the distinctive difficulties presented by the pandemic, such as limited access to childcare and social support systems or changes in one’s job/financial situation, parents frequently struggled to cope with internal and external stressors as a couple (Lin et al., 2025), potentially even if their coping mechanisms had been effective in the past. This underscores the need for multifaceted support for families during crises like the COVID-19 pandemic, including structural (e.g. financial support in case of reduced income) and psychological support.

The latter is supported by results from the exploratory model in which sociodemographic and pandemic-related variables were included. The exploratory model highlights that stable individual differences in perceived stress, relationship quality with the partner, and bonding were partly shaped by demographic and pandemic-related factors (see Table 5). For perceived stress, older child age, more children in the household, and changes in the work situation due to the pandemic were associated with greater stress, whereas higher maternal education and higher income were linked to lower stress, which is in line with similar studies during the pandemic (Brown et al., 2020; Cusinato et al., 2020). For impaired bonding, older child age and higher maternal education were linked to more impaired bonding, with work changes showing only a marginal effect. However, an important limitation must be noted: the PBQ-16 was validated for children aged 0–1 year. Thus, the association with child age potentially reflects reduced validity of the PBQ-16 for older children, rather than a genuine age effect on bonding. In contrast, the association between higher maternal education and more impaired postpartum bonding appears more robust, as similar findings have been reported in previous studies as well (Dubber et al., 2015; Lehnig et al., 2019; Mazúchová et al., 2021). Finally, perceived relationship quality with the partner was negatively predicted by maternal age and the number of children in the household, indicating that older mothers and larger families reported lower relationship quality. This is in line with previous research showing that older parental age is associated with reduced relationship quality with the partner - an effect that may be partly explained by the general decline in partner relationship quality over the course of long-term relationships (Jadva et al., 2022; van Laningham et al., 2001). Likewise, a higher number of children in the household has repeatedly been linked to lower partner relationship quality, likely due to increased parenting demands, elevated stress, and diminished opportunities for couple interaction (Ahlborg et al., 2009; Saheba & Mansuri, 2020). Taken together, family structure, educational background, and economic stability appear to shape the baseline levels from which intraindividual fluctuations unfold, underscoring the importance of considering these contextual factors when interpreting stress and relational dynamics during the COVID-19 pandemic.

Overall, this study’s results suggest that couple-centered approaches strengthening the relationship with the partner are promising for preventing high maternal stress, and possibly also impaired bonding, with potential positive long-term effects for family well-being. Previous studies have demonstrated positive effects of couple-based prevention and intervention programs on the partner relationship, but also on individual and family well-being (Bradbury & Bodenmann, 2020; Kanter & Schramm, 2018; Kysely et al., 2021; Lavner et al., 2020). Based on this study’s findings in a relatively low-risk community sample with comparatively high socioeconomic status, the observed effects underscore the importance of a public health approach to relationship-focused support. It is therefore highly plausible that easily accessible large-scale interventions aimed at strengthening partner relationships may be even more critical in the general population, particularly among families with lower socioeconomic resources. First evidence for the effectiveness of brief online programs targeting different aspects of relationship quality and functioning with the partner (e.g., closeness, conflict) during the COVID-19 pandemic underscores possibilities for cost-effective, low-threshold programs during such crises (Prime et al., 2024; Rodriguez et al., 2021; Tsai et al., 2020).

In terms of strengths and limitations, to the best of the authors’ knowledge, this was the first study to examine longitudinal reciprocal effects among perceived stress, perceived relationship quality with the partner, and impaired bonding in mothers during the COVID-19 pandemic. In addition, the large sample size and longitudinal design enabled the use of advanced statistical modeling separating between-person differences and within-person processes. Regarding limitations, the high dropout rate in this study must be considered. However, comprehensive attrition analyses and robustness checks indicated very few significant differences with negligible to small effect sizes regarding the main variables and sociodemographics, suggesting that dropout is unlikely to have introduced bias into our results. Model indices further indicated near-perfect fit, raising over-fitting concerns and calling for replication with lower dropout rates or more parsimonious models. Another limitation of this study is that it was conducted in a relatively low-risk sample with predominantly higher socioeconomic status, which may restrict the generalizability of the findings to more socioeconomically diverse populations. Moreover, we did not collect data on participants’ ethnicity, which represents an additional limitation and potential source of bias. In addition, the measure used to assess maternal bonding (PBQ-16) has not been validated for toddlers and preschool-aged children and should therefore be regarded as a proxy to assess the mother-child relationship; however, child age was included as a control variable in an exploratory model, which did not alter the main results. Furthermore, this study relied exclusively on maternal reports, as only a very small number of fathers or other primary caregivers participated in the survey. This reliance on a single informant may have introduced a single-informant bias into our results. Future studies would benefit from recruiting more fathers and other caregivers, as well as including further data sources, such as observations of parent-child or parental interaction, and ecological momentary assessments.

Conclusion

This study’s results underscore the interrelation of maternal perceived stress, perceived relationship quality with the partner, and mother-to-child bonding, and highlight the dynamic relevance of the relationship quality with the partner as a risk or resilience factor for maternal stress and, potentially, for bonding during the COVID-19 pandemic. These processes, however, do not occur in isolation: Sociodemographic factors such as maternal education, income, family size, and pandemic-related work changes significantly shaped baseline levels of stress and relational functioning. Consequently, support services should focus on strengthening the relationship with the partner early and tailor interventions to families’ structural contexts to mitigate stress, promote bonding, and prevent long-term adverse effects on maternal and child well-being.

Footnotes

Acknowledgements

We would like to thank all volunteers who participated in the online survey and all colleagues who contributed to this study.

ORCID iD

Christian F. J. Woll-Weber

Ethical Considerations

The study was approved by the ethics committee of the medical faculty, Ruprecht-Karls-University, Heidelberg, in agreement with the Ludwig Maximilian University, Munich (vote: S-446/2017). Written informed consent was obtained from participants to participate in the study.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the University of Mannheim’s Graduate School of Economic and Social Sciences, Center for Doctoral Studies in Social and Behavioral Sciences (CDSS).

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The datasets, analysis scripts, and study materials are openly available in the PsychArchives repository at https://doi.org/10.23668/psycharchives.22158, and https://doi.org/10.23668/psycharchives.22159, Woll and Stürmlinger (2026).

Open Science Statement

Pre-Registration: The study design, hypotheses, and analysis plan were pre-registered on the platform “AsPredicted” prior to data analysis (see ).

References

Ahlborg

Misvaer

Möller

(2009). Perception of marital quality by parents with small children: A follow-up study when the firstborn is 4 years old. Journal of Family Nursing, 15(2), 237–263. https://doi.org/10.1177/1074840709334925

Balzarini

R. N.

Muise

Zoppolat

Di Bartolomeo

Rodrigues

D. L.

Alonso-Ferres

Urganci

Debrot

Bock Pichayayothin

Dharma

Chi

Karremans

J. C.

Schoebi

Slatcher

R. B.

(2023). Love in the time of COVID: Perceived partner responsiveness buffers people from lower relationship quality associated with COVID-related stressors. Social Psychological and Personality Science, 14(3), 342–355. https://doi.org/10.1177/19485506221094437

Bradbury

T. N.

Bodenmann

(2020). Interventions for couples. Annual Review of Clinical Psychology, 16(16), 99–123. https://doi.org/10.1146/annurev-clinpsy-071519-020546

Brockington

Oates

George

Turner

Vostanis

Sullivan

Loh

Murdoch

(2001). A screening questionnaire for mother-infant bonding disorders. Archives of Women's Mental Health, 3(4), 133–140. https://doi.org/10.1007/s007370170010

Brown

S. M.

Doom

J. R.

Lechuga-Peña

Watamura

S. E.

Koppels

(2020). Stress and parenting during the global COVID-19 pandemic. Child Abuse & Neglect, 110, Article 104699. https://doi.org/10.1016/j.chiabu.2020.104699

Buechel

Nehring

Seifert

Eber

Behrends

Mall

Friedmann

(2022). A cross-sectional investigation of psychosocial stress factors in German families with children aged 0-3 years during the COVID-19 pandemic: Initial results of the CoronabaBY study. Child and Adolescent Psychiatry and Mental Health, 16(1), 37. https://doi.org/10.1186/s13034-022-00464-z

Chen

M.-T.

Lin

G.-X.

Chang

Y.-P.

(2022). Good partners, good parents: Parent relationship satisfaction predicts parenting efficacy for young children. Applied Research in Quality of Life, 17(4), 2405–2422. https://doi.org/10.1007/s11482-022-10043-1

Claussen

A. H.

Holbrook

J. R.

Hutchins

H. J.

Robinson

L. R.

Bloomfield

Meng

Bitsko

R. H.

O’Masta

Cerles

Maher

Rush

Kaminski

J. W.

(2022). All in the family? A systematic review and meta-analysis of parenting and family environment as risk factors for attention-deficit/hyperactivity disorder (ADHD) in children. Prevention Science, 25(S2), 1–23. https://doi.org/10.1007/s11121-022-01358-4

Cohen

(1988). Statistical power analysis for the behavioral sciences (2nd ed.). Lawrence Erlbaum Associates, Publishers.

10.

Cohen

Kamarck

Mermelstein

(1983). A global measure of perceived stress. Journal of Health and Social Behavior, 24(4), 385–396. https://doi.org/10.2307/2136404. https://www-jstor-org-443.web.bisu.edu.cn/stable/2136404

11.

Cox

M. J.

Paley

(2003). Understanding families as systems. Current Directions in Psychological Science, 12(5), 193–196. https://doi.org/10.1111/1467-8721.01259. https://www-jstor-org-443.web.bisu.edu.cn/stable/20182875

12.

Cuijlits

van de Wetering

A. P.

Endendijk

J. J.

van Baar

A. L.

Potharst

E. S.

Pop

V. J. M.

(2019). Risk and protective factors for pre- and postnatal bonding. Infant Mental Health Journal, 40(6), 768–785. https://doi.org/10.1002/imhj.21811

13.

Cusinato

Iannattone

Spoto

Poli

Moretti

Gatta

Miscioscia

(2020). Stress, resilience, and well-being in Italian children and their parents during the COVID-19 pandemic. International Journal of Environmental Research and Public Health, 17(22), 8297. https://doi.org/10.3390/ijerph17228297

14.

Daugherty

J. C.

Bueso-Izquierdo

Lara-Cinisomo

Lozano-Ruiz

Caparros-Gonzalez

R. A.

(2022). Partner relationship quality, social support and maternal stress during pregnancy and the first COVID-19 lockdown. Journal of Psychosomatic Obstetrics and Gynaecology, 43(4), 563–573. https://doi.org/10.1080/0167482X.2022.2101446

15.

Dubber

Reck

Müller

Gawlik

(2015). Postpartum bonding: The role of perinatal depression, anxiety and maternal-fetal bonding during pregnancy. Archives of Women's Mental Health, 18(2), 187–195. https://doi.org/10.1007/s00737-014-0445-4

16.

de Cock

E. S. A.

Henrichs

Klimstra

T. A.

Janneke

B. M.

Maas

Vreeswijk

C. M. J. M.

Meeus

W. H. J.

van Bakel

H. J. A.

(2017). Longitudinal associations between parental bonding. Parenting Stress, and Executive Functioning in Toddlerhood, 26(6), 1723–1733. https://doi.org/10.1007/s10826-017-0679-7

17.

de Waal

Boekhorst

M. G. B. M.

Nyklíček

Pop

V. J. M.

(2023). Maternal-infant bonding and partner support during pregnancy and postpartum: Associations with early child social-emotional development. Infant Behavior and Development, 72, Article 101871. https://doi.org/10.1016/j.infbeh.2023.101871

18.

Fernandes

D. V.

Canavarro

M. C.

Moreira

(2021). Postpartum during COVID-19 pandemic: Portuguese mothers’ mental health, mindful parenting, and mother-infant bonding. Journal of Clinical Psychology, 77(9), 1997–2010. https://doi.org/10.1002/jclp.23130

19.

Gao

Zhang

Liu

Zhang

Wang

Liu

Xie

Yin

Yue

(2022). Psychological health status in postpartum women during COVID-19 pandemic: A systematic review and meta-analysis. Journal of Affective Disorders, 319, 99–111. https://doi.org/10.1016/j.jad.2022.08.107

20.

Hahlweg

(2016). Fragebogen zur Partnerschaftsdiagnostik (2nd, newly normed ed.). Hogrefe.

21.

Hamaker

E. L.

Kuiper

R. M.

Grasman

R. P. P. P.

(2015). A critique of the cross-lagged panel model. Psychological Methods, 20(1), 102–116. https://doi.org/10.1037/a0038889

22.

Bentler

P. M.

(1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1–55. https://doi.org/10.1080/10705519909540118

23.

Jadva

Lysons

Imrie

Golombok

(2022). An exploration of parental age in relation to parents’ psychological health, child adjustment and experiences of being an older parent in families formed through egg donation. Reproductive BioMedicine Online, 45(2), 401–409. https://doi.org/10.1016/j.rbmo.2022.03.029

24.

Kanter

J. B.

Schramm

D. G.

(2018). Brief interventions for couples: An integrative review. Family Relations, 67(2), 211–226. https://doi.org/10.1111/fare.12298

25.

Kysely

Bishop

Kane

R. T.

McDevitt

de Palma

Rooney

(2021). Couples therapy delivered through videoconferencing: Effects on relationship outcomes, mental health and the therapeutic alliance. Frontiers in Psychology, 12, Article 773030. https://doi.org/10.3389/fpsyg.2021.773030

26.

Lavner

J. A.

Barton

A. W.

Beach

S. R. H.

(2020). Direct and indirect effects of a couple-focused preventive intervention on children’s outcomes: A randomized controlled trial with African American families. Journal of Consulting and Clinical Psychology, 88(8), 696–707. https://doi.org/10.1037/ccp0000589

27.

Le Bas

G. A.

Youssef

G. J.

Macdonald

J. A.

Rossen

Teague

S. J.

Kothe

E. J.

McIntosh

J. E.

Olsson

C. A.

Hutchinson

D. M.

(2020). The role of antenatal and postnatal maternal bonding in infant development: A systematic review and meta‐analysis. Social Development, 29(1), 3–20. https://doi.org/10.1111/sode.12392

28.

Lee

S. J.

Ward

K. P.

Rodriguez

C. M.

(2022). Longitudinal analysis of short-term changes in relationship conflict during COVID-19: A risk and resilience perspective. Journal of Interpersonal Violence, 37(15-16), NP14239–NP14261. https://doi.org/10.1177/08862605211006359

29.

Lehnig

Nagl

Stepan

Wagner

Kersting

(2019). Associations of postpartum mother-infant bonding with maternal childhood maltreatment and postpartum mental health: A cross-sectional study. BMC Pregnancy and Childbirth, 19(1), 1–10. https://doi.org/10.1186/s12884-019-2426-0

30.

Lin

Soejima

Zhang

Kitao

(2025). Impact of COVID-19 on well-being in child-rearing families: A systematic review of longitudinal studies. Social Sciences, 14(5), 312. https://doi.org/10.3390/socsci14050312

31.

Little

T. D.

Kline

R. B.

(Eds.), (2016). Methodology in the social sciences. Principles and practice of structural equation modeling (4th ed.). Guilford publications.

32.

Liu

C. H.

Hyun

Mittal

Erdei

(2022). Psychological risks to mother-infant bonding during the COVID-19 pandemic. Pediatric Research, 91(4), 853–861. https://doi.org/10.1038/s41390-021-01751-9

33.

Martin

R. C. B.

Brock

R. L.

(2023). The importance of high-quality partner support for reducing stress during pregnancy and postpartum bonding impairments. Archives of Women's Mental Health, 26(2), 201–209. https://doi.org/10.1007/s00737-023-01299-z

34.

Mazúchová

Kelčíková

Maskalová

Malinovská

Grendár

(2021). Mother-infant bonding and its associated factors during postpartum period. Kontakto, 23(2), 126–132. https://doi.org/10.32725/KONT.2021.018

35.

Milek

Randall

A. K.

Nussbeck

F. W.

Breitenstein

C. J.

Bodenmann

(2017). Deleterious effects of stress on time spent together and parents’ relationship satisfaction. Journal of Couple & Relationship Therapy, 16(3), 210–231. https://doi.org/10.1080/15332691.2016.1238799

36.

Mousavi

Z. S.

Alizadeh

Aghaie

Kahrobaie

Goli

(2021). Mother-infant attachment behaviors and related psychological factors in covid-19 pandemic: A review study. Social Determinants of Health, 7, 1–12. https://doi.org/10.22037/sdh.v7i1.35722

37.

Nelson

S. K.

Kushlev

Lyubomirsky

(2014). The pains and pleasures of parenting: When, why, and how is parenthood associated with more or less well-being? Psychological Bulletin, 140(3), 846–895. https://doi.org/10.1037/a0035444

38.

O’Dea

G. A.

Youssef

G. J.

Hagg

L. J.

Francis

L. M.

Spry

E. A.

Rossen

Smith

Teague

S. J.

Mansour

Booth

Davies

Hutchinson

Macdonald

J. A.

(2023). Associations between maternal psychological distress and mother-infant bonding: A systematic review and meta-analysis. Archives of Women’s Mental Health, 26(4), 441–452. https://doi.org/10.1007/s00737-023-01332-1

39.

Orth

Meier

L. L.

Bühler

J. L.

Dapp

L. C.

Krauss

Messerli

Robins

R. W.

(2024). Effect size guidelines for cross-lagged effects. Psychological Methods, 29(2), 421–433. https://doi.org/10.1037/met0000499

40.

Overall

N. C.

Chang

V. T.

Pietromonaco

P. R.

Low

R. S. T.

Henderson

A. M. E.

(2022). Partners’ attachment insecurity and stress predict poorer relationship functioning during COVID-19 quarantines. Social Psychological and Personality Science, 13(1), 285–298. https://doi.org/10.1177/1948550621992973

41.

Pauly

Lüscher

Berli

Scholz

(2022). Dynamic associations between stress and relationship functioning in the wake of COVID-19: Longitudinal data from the German family panel (pairfam). Journal of Social and Personal Relationships, 39(11), 3183–3203. https://doi.org/10.1177/02654075221092360

42.

Pietromonaco

P. R.

Overall

N. C.

(2021). Applying relationship science to evaluate how the COVID-19 pandemic may impact couples’ relationships. American Psychologist, 76(3), 438–450. https://doi.org/10.1037/amp0000714

43.

Pietromonaco

P. R.

Overall

N. C.

(2022). Implications of social isolation, separation, and loss during the COVID-19 pandemic for couples’ relationships. Current Opinion in Psychology, 43, 189–194. https://doi.org/10.1016/j.copsyc.2021.07.014

44.

Posit team . (2025). RStudio: Integrated development environment for R. Posit Software, PBC. [Computer software]. https://www.posit.co/

45.

Power

Weise

Mack

J. T.

Karl

Garthus-Niegel

(2024). Does parental mental health mediate the association between parents’ perceived stress and parent-infant bonding during the early COVID-19 pandemic? Early Human Development, 189, Article 105931. https://doi.org/10.1016/j.earlhumdev.2023.105931

46.

Prime

Muise

Jambon

Koven

Markwell

Wade

(2024). A pilot randomized controlled trial of a brief conflict reappraisal intervention for community couples with young children. Cognitive Therapy and Research, 48(5), 895–909. https://doi.org/10.1007/s10608-024-10485-x

47.

R Core Team . (2025). R: A language and environment for statistical computing. https://www.R-project.org/

48.

Reck

Klier

C. M.

Pabst

Stehle

Steffenelli

Struben

Backenstrass

(2006). The German version of the postpartum bonding instrument: Psychometric properties and association with postpartum depression. Archives of Women’s Mental Health, 9(5), 265–271. https://doi.org/10.1007/s00737-006-0144-x

49.

Reck

von Tettenborn

Woll-Weber

C. F. J.

Küçükakyüz

S. M.

Megele

M. J.

Kaubisch

Marx

A. K. G.

Hagl

Nonnenmacher

Müller

Zietlow

A.-L.

(2025). The longitudinal relationship between maternal bonding and mental health during the COVID-19 pandemic. Journal of Reproductive and Infant Psychology, 1–14. https://doi.org/10.1080/02646838.2025.2539220

50.

Reck

Zietlow

A.-L.

Müller

Dubber

(2016). Perceived parenting stress in the course of postpartum depression: The buffering effect of maternal bonding. Archives of Women's Mental Health, 19(3), 473–482. https://doi.org/10.1007/s00737-015-0590-4

51.

Reis

Lehr

Heber

Ebert

D. D.

(2019). The German version of the perceived stress scale (PSS-10): Evaluation of dimensionality, validity, and measurement invariance with exploratory and confirmatory bifactor modeling. Assessment, 26(7), 1246–1259. https://doi.org/10.1177/1073191117715731

52.

Rodriguez

L. M.

Stewart

S. H.

Neighbors

(2021). Effects of a brief web-based interpersonal conflict cognitive reappraisal expressive-writing intervention on changes in romantic conflict during COVID-19 quarantine. Couple & Family Psychology, 10(3), 212–222. https://doi.org/10.1037/cfp0000173

53.

Roos

L. E.

Salisbury

Penner-Goeke

Cameron

E. E.

Protudjer

J. L. P.

Giuliano

Afifi

T. O.

Reynolds

(2021). Supporting families to protect child health: Parenting quality and household needs during the COVID-19 pandemic. PLoS One, 16(5), Article e0251720. https://doi.org/10.1371/journal.pone.0251720

54.

Rosseel

(2012). Lavaan: An R package for structural equation modeling. Journal of Statistical Software, 48(2), 1–36. https://doi.org/10.18637/jss.v048.i02

55.

Rosseel

(2020). Small sample solutions for structural equation modeling. In van de Schoot

Miočević

(Eds.), Small sample size solutions (pp. 226–238). Routledge.

56.

Rusu

P. P.

Nussbeck

F. W.

Leuchtmann

Bodenmann

(2020). Stress, dyadic coping, and relationship satisfaction: A longitudinal study disentangling timely stable from yearly fluctuations. PLoS One, 15(4), Article e0231133. https://doi.org/10.1371/journal.pone.0231133

57.

Saheba

K. A.

Mansuri

(2020). Marital adjustment of the number of children and couples. International Journal of Indian Psychology, 8(1), 1–6. https://doi.org/10.25215/0801.032

58.

Saur

A. M.

Dos Santos

M. A.

(2021). Risk factors associated with stress symptoms during pregnancy and postpartum: Integrative literature review. Women & Health, 61(7), 651–667. https://doi.org/10.1080/03630242.2021.1954132

59.

Shi

Maydeu-Olivares

DiStefano

(2018). The relationship between the standardized root mean square residual and model misspecification in factor analysis models. Multivariate Behavioral Research, 53(5), 676–694. https://doi.org/10.1080/00273171.2018.1476221

60.

Suárez-Rico

B. V.

Estrada-Gutierrez

Sánchez-Martínez

Perichart-Perera

Rodríguez-Hernández

González-Leyva

Osorio-Valencia

Cardona-Pérez

Helguera-Repetto

A. C.

Espino y Sosa

Solis-Paredes

Reyes-Muñoz

(2021). Prevalence of depression, anxiety, and perceived stress in postpartum Mexican women during the COVID-19 lockdown. International Journal of Environmental Research and Public Health, 18(9), 4627. https://doi.org/10.3390/ijerph18094627

61.

Tepeli Temiz

Bilican

F. I.

(2024). Daily external stress, internal stress, dyadic coping, and relationship functioning of Turkish couples during the COVID-19: A daily diary study. Stress and Health: Journal of the International Society for the Investigation of Stress, 40(5), Article e3455. https://doi.org/10.1002/smi.3455

62.

Tepeli Temiz

Elsharnouby

(2022). Relationship satisfaction and well-being during the COVID-19 pandemic: Examining the associations with interpersonal emotion regulation strategies. Cognitive Therapy and Research, 46(5), 902–915. https://doi.org/10.1007/s10608-022-10317-w

63.

Tichelman

Westerneng

Witteveen

A. B.

van Baar

A. L.

van der Horst

H. E.

de Jonge

Berger

M. Y.

Schellevis

F. G.

Burger

Peters

L. L.

(2019). Correlates of prenatal and postnatal mother-to-infant bonding quality: A systematic review. PLoS One, 14(9), Article e0222998. https://doi.org/10.1371/journal.pone.0222998

64.

Tolja

Nakić Radoš

Anđelinović

(2020). The role of maternal mental health, infant temperament, and couple’s relationship quality for mother-infant bonding. Journal of Reproductive and Infant Psychology, 38(4), 395–407. https://doi.org/10.1080/02646838.2020.1733503

65.

Tsai

Hardebeck

Ramos

F. P.

Turlove

Nordal-Jonsson

Vongdala

Zhang

Kohlenberg

R. J.

(2020). Helping couples connect during the COVID-19 pandemic: A pilot randomised controlled trial of an awareness, courage, and love intervention. Applied Psychology. Health and Well-Being, 12(4), 1140–1156. https://doi.org/10.1111/aphw.12241

66.

van Laningham

Johnson

D. R.

Amato

(2001). Marital happiness, marital duration, and the U-Shaped curve: Evidence from a five-wave panel study. Social Forces, 79(4), 1313–1341. https://doi.org/10.1353/sof.2001.0055

67.

Vigl

Strauss

Talamini

Zentner

(2022). Relationship satisfaction in the early stages of the COVID-19 pandemic: A cross-national examination of situational, dispositional, and relationship factors. PLoS One, 17(3), Article e0264511. https://doi.org/10.1371/journal.pone.0264511

68.

Whitton

S. W.

Rhoades

G. K.

Whisman

M. A.

(2014). Fluctuation in relationship quality over time and individual well-being: Main, mediated, and moderated effects. Personality and Social Psychology Bulletin, 40(7), 858–871. https://doi.org/10.1177/0146167214528988

69.

Woll-Weber

C. F. J.

Reck

Marx

A. K. G.

Küçükakyüz

S. M.

Müller

von Tettenborn

Nonnenmacher

Zietlow

A.-L.

(2025). Mothers of young children during the COVID-19 pandemic: Depressive symptoms and perceived stress. Psychopathology, 58(4), 1–14. https://doi.org/10.1159/000543715

70.

Woll

C. F. J.

Stürmlinger

(2026). Code for the article “Maternal stress and bonding in the COVID-19 pandemic: Mothers’ relationship quality with the partner as a resilience factor”. PsychArchives. https://doi.org/10.23668/psycharchives.22158.

71.

Woll

C. F. J.

Stürmlinger

(2026). Data set for “Maternal stress and bonding in the COVID-19 pandemic: Mothers’ relationship quality with the partner as a resilience factor” [Dataset]. PsychArchives. https://doi.org/10.23668/psycharchives.22159.

72.

Xie

E. B.

Rioux

Madsen

J. W.

Lebel

Giesbrecht

G. F.

Tomfohr-Madsen

(2022). Romantic relationship quality and mental health in pregnancy during the Covid-19 pandemic. Journal of Social and Clinical Psychology, 41(5), 463–490. https://doi.org/10.1521/jscp.2022.41.5.463

73.

Ximénez

Maydeu-Olivares

Shi

Revuelta

(2022). Assessing cutoff values of SEM fit indices: Advantages of the unbiased SRMR index and its cutoff criterion based on communality. Structural Equation Modeling: A Multidisciplinary Journal, 29(3), 368–380. https://doi.org/10.1080/10705511.2021.1992596

74.

Yazdanimehr

Aflakseir

Sarafraz

Taghavi

(2023). The structural model of mother-infant bonding in the first pregnancy based on the mother’s attachment style and parenting style: The mediating role of mentalization and shame. BMC Psychology, 11(1), 396. https://doi.org/10.1186/s40359-023-01436-4