Maternal Inner Strength Mediates the Longitudinal Association Between Family Emotional Expression and Fear of Childbirth: A Cross-Lagged Panel Mediation Analysis

Abstract

Purpose

Fear of childbirth (FOC) increases maternal and perinatal adverse health risks, and family factors are core predictors of FOC. This study aimed to clarify the mediating pathway through which familial emotional expression (FEE) is associated with FOC via pregnant women’s inner strength (IS), so as to supply empirical basis for antenatal psychological risk management.

Methods

A longitudinal survey was performed among pregnant women from the Department of Obstetrics and Gynecology of a tertiary Grade A hospital in Shenyang from September 2023 to August 2024. A total of 136 participants completed questionnaires in early, middle and late pregnancy, including the Childbirth Attitude Questionnaire (CAQ), Family Environment Scale (FES) and Inner Strength Scale (ISS). Two sets of longitudinal analyses were performed on this sample: latent class growth modeling (LCGM) was conducted via Mplus 8.3 to identify heterogeneous developmental trajectories of FOC; separately, cross-lagged panel mediation analyses were carried out with SPSS 26 and Amos 26.0 using the full-sample raw longitudinal data to examine the mediating mechanism among FEE, IS, and FOC.

Results

Three distinct FOC developmental trajectories were identified: “Low FOC stable group” (11.4%), “Moderate FOC stable group” (33.5%), and “High FOC elevated group” (55.1%). FEE was longitudinally associated with IS (β=0.691, P<0.001), while IS was longitudinally associated with FOC in a negative direction (β=-0.552, P<0.001). No significant association was observed between FEE and FOC (β=0.079, P=0.293). The total effect of FEE on FOC was −0.303 (P<0.05), the direct effect was 0.079 (P>0.05), and the indirect effect was −0.382 (P<0.05), proving that the effect of FEE on FOC operates through IS.

Conclusions

FOC presents heterogeneous developmental trajectories across different gestational trimesters. FEE is indirectly associated with lower FOC levels by promoting pregnant women’s IS. The findings provide empirical evidence for understanding the developmental characteristics and psychological influencing factors of FOC, and can inform targeted, individualized antenatal psychological screening and intervention strategies to alleviate childbirth fear and improve maternal mental health in clinical obstetric care.

Introduction

Pregnant women undergoing physiological changes commonly face psychological adaptation challenges, with fear of childbirth (FOC) being particularly prominent.¹ FOC is defined as a distressing emotional experience arising from perceived inadequacy in coping with labor-related challenges or adverse delivery expectations.² Its global incidence ranges from 3.7% to 43% (pooled prevalence: ~14%) among 853,988 pregnant women.³ Chinese pregnant women exhibit significantly higher rates, with a Meta-analysis of 8815 samples across 18 regions reporting a pooled incidence of 72.1%.⁴ This disparity reflects variations in regional cultural norms, healthcare resource distribution, and assessment methodologies. Fear of childbirth (FOC) impairs maternal, perinatal and fetal-offspring outcomes. It raises epidural/spinal analgesia need (70.2% vs 67.1%), autonomic dysfunction and gestational hypertension risk.^5,6 Severe FOC correlates with more healthcare use, longer hospitalization, prenatal depression and childbirth avoidance.^7–10 Higher FOC severity increases cesarean rates (high-risk cesarean preference) and prolonged labor (8 h vs 6.64 h).^11–13 FOC-related chronic stress disturbs uterine contractions, aggravates fear-tension-pain cycle, elevates emergency cesarean and delayed lactation risks.^14,15 Elevated catecholamines reduce placental oxygenation, increasing fetal distress, dystocia, intrauterine asphyxia risks and damaging offspring neuro-cognitive-emotional health.^16,17 FOC in women varies across different stages of pregnancy, and its developmental trajectory exhibits heterogeneity as pregnancy progresses.¹⁸ FOC during pregnancy demonstrates significant dynamic evolutionary characteristics, with its changing patterns intricately linked to the pregnancy cycle and associated physiological and psychological changes.¹⁹

Family factors are central determinants of FOC. Bowen Family Systems Theory (BFST) posits the family as a dynamic, interconnected holistic system, where intrafamilial interactions profoundly shape individual behaviors and mental health.²⁰ Studies confirm pregnant women’s psychological-physical well-being is not merely individual, but directly shaped by spousal attitudes and mother-in-law-daughter-in-law dynamics.^21,22 The family system comprises three core dimensions: cohesion, adaptability, and communication.²³ Focusing on its emotional-expression function, this study defines familial emotional expression as the process whereby family members provide emotional support verbally and behaviorally. Positive practices including empathetic communication and collaborative decision-making foster pregnant women’s psychological security, reflecting BFST’s core principles of “interdependence” and “constructive communication patterns”.

Studies have confirmed that high levels of family emotional support can alleviate pregnant women’s excessive preoccupation with childbirth risks.²⁴ A partner’s proactive responsiveness has been shown to significantly reduce the impact of childbirth-related fear on postpartum depression.²⁵ Poor family communication patterns, particularly empathy deficits in dialogue or conflict avoidance strategies, may intensify expectant mothers’ hypervigilance toward perinatal risk perception.²⁶ Studies have shown that familial emotional expression (external emotional resources such as spousal support and family care) serves as a critical factor in alleviating childbirth-related anxiety, though its precise mechanisms of action remain incompletely understood.

Recent psychological research highlights maternal inner strengths (eg, resilience, self-efficacy) as critical mediators linking external support to internal psychological states.^27,28 Such strengths cover stress cognitive reframing, emotion regulation, and perceived control over medical decisions,²⁹ rooted in Psychological Capital Theory which underscores the stress-buffering function of psychological resources;³⁰ for example, highly resilient women may view childbirth as growth rather than threat. As a core BFST construct, Differentiation of self (DoS) describes the capacity to sustain autonomy while retaining family emotional bonds,²⁰ and it predicts better psychological health, marital quality, physical health and intergenerational relationships.²⁰ Within BFST, maternal inner strength corresponds to DoS: maintaining rational-emotional balance amid family interactions. Unresolved family emotional conflicts reduce differentiation and aggravate childbirth-related fear; by contrast, individuals with high DoS can activate internal resources to mitigate anxiety.^31,32 This conceptual synthesis underscores the dual protective effects of inner psychological capital and systemic emotional regulation on perinatal mental health. Accordingly, as an external systemic support, familial emotional expression can negatively regulate fear of childbirth (FOC) by enhancing pregnant women’s inner strengths (ie, psychological capital and differentiation‑of‑self level). It forms a complete pathway of “external family support‑individual psychological empowerment‑alleviated FOC”, highlighting the synergistic value of systemic family emotional regulation and the mobilization of individual psychological resources for perinatal mental health management.

Previous cross‑sectional studies can only capture static associations among familial emotional expression, inner strengths and FOC at a single time point, and thus fail to clarify the dynamic effects of family‑based empowerment and the transformation of individual inner strengths on FOC. Yet perinatal women’s psychological states and family interaction patterns undergo ongoing shifts, and the onset, course, and resolution of FOC unfold as a temporally ordered process. Longitudinal research employing multi‑wave follow‑up assessments enables rigorous testing of the temporal logic embedded within the mediating pathway “familial emotional expression → inner strengths activation → FOC shift”.

In summary, family factors exert indirect, rather than direct, effects on pregnant women’s FOC, with maternal internal psychological states functioning as mediators. Adaptive familial emotional expression may trigger pregnant women’s psychological regulatory potential, foster psychological resource accumulation, and mobilize self‑efficacy and differentiation‑of‑self capacities, which in turn shapes responses to childbirth‑related stress and modulates FOC manifestations. Accordingly, this study advances its core hypothesis: familial emotional expression operates through the activation of maternal inner strengths to reshape the functional correlates of fear of childbirth. This mediating pathway illuminates the dynamic mechanism through which external family support translates into individual psychological capital, offering theoretical grounding for the development of family‑individual coordinated perinatal mental‑health intervention frameworks.

Materials and Methods

Study Design and Sample Selection

A longitudinal prospective study was performed in the Department of Obstetrics and Gynecology of a tertiary Grade-A hospital in Shenyang, between September 2023 and August 2024. Pregnant women were assessed at three time-points: early pregnancy (T1, 12±2 weeks of gestation), mid-pregnancy (T2, 24±2 weeks of gestation), and late pregnancy (T3, 36±2 weeks of gestation). At baseline T1, 152 participants were recruited. Among them, 144 completed the T2 assessment, with 8 participants lost from T1 to T2 (attrition rate = 5.26%). Ultimately, 136 participants finished all three-wave surveys including T3, corresponding to a cumulative attrition rate of 10.53% relative to the baseline sample. Reasons for dropout included voluntary withdrawal, transfer to other hospitals for antenatal care, loss-to-follow-up, and pregnancy termination during the observation period. Only participants who completed questionnaires at all three time-points were included in the final analytical sample (n = 136); cases with incomplete follow-up were excluded via list-wise deletion. Multiple imputation was adopted to handle item-level missing values within the final 136-participant dataset.

This study employed a convenience sampling method, selecting pregnant women who received prenatal care and delivered. Based on the literature review, a total of 15 independent variables were enrolled in this study. Conventionally, the sample size for influencing-factor analysis should be at least 5–10 times the number of independent variables.³³ After accounting for an estimated 20% participant attrition rate, the calculated required sample size ranged from 90 to 180 participants. Empirical methodological guidelines indicate that 120–150 participants can yield stable outputs for cross-lagged panel models given reliable measurements, parsimonious model specification, and moderate-to-strong effect sizes.³⁴ Latent class growth modeling (LCGM) has comparable sample-size requirements to structural equation modeling (SEM), and 100–150 participants is widely regarded as the minimum threshold for robust parameter estimation.³⁵ The final analytical sample of 136 in the present study falls within this empirical range.

The inclusion criteria: Pregnancy confirmed by ultrasound, age ≥20 years, regular prenatal care at the study hospital, provision of informed consent, and voluntary participation. The exclusion criteria: Pregnancy complications (eg, gestational diabetes and hypertensive disorders of pregnancy), congenital developmental abnormalities, and a history of mental disorders. Elimination criteria: Participants who withdrew consent or were lost to follow-up and pregnancy termination during follow-up.

Measurement

Measurement of Demographic Characteristics

A self-designed questionnaire was used, including age, education level (“high school or below”, “undergraduate or above”), occupation (“company employees”, “public institutions/civil servants”, “individual/temporary worker”, “full time wife”), per capita family income (“<4000CNY”, “4000-8000CNY”, “>8000CNY”), first pregnancy (“yes”, “no”), expected mode of delivery (“natural labour”, “caesarean birth”, “anything”), and the relationship between mother-in-law and daughter-in-law (“bad”, “not too bad”, “good”).

Measurement of Fear of Childbirth

We used the Childbirth Attitude Questionnaire (CAQ) developed by Lowe et al³⁶ to measure FOC among pregnant women. The CAQ consists of 16 items divided into four dimensions: fetal health, self-control, pain and injury during childbirth, and medical care. The answer is scored on a 1–4 scale, with a total score of 16–64, which is positively correlated with FOC level. The Cronbach’s alpha of the scale in this study was 0.845, 0.896, and 0.902 for the three periods.

Measurement of Family Emotional Expression

The Family Environment Scale (FES) includes 10 different characteristics of family environment, and this study selected the dimension of family emotional expression from it (9 items: 2, 12, 22, 32, 42, 52, 62, 72, 82).³⁷ The answer was divided into “Yes” and “No”, with a score of one for “Yes” and two for “No”, and the total score range of each dimension was 9–18. A higher score represents better emotional expression of the family. The Cronbach’s alpha of the scale in this study was 0.812, 0.804, and 0.845 for the three periods.

Measurement of Inner Strength

The Inner Strength Scale (ISS) was developed by Lundman et al and applied in all ranges of people, containing 20 items and 4 dimensions, composed of connectedness, creativity, firmness and flexibility.³⁸ The scale is scored by six‐point Likert scale, from 1 “completely disagree” to 6 “completely agree”, with a total score ranging from 20 to 120, where higher score represents stronger inner strength. The Cronbach’s alpha of the scale in this study was 0.811, 0.923, and 0.912 for the three periods.

Statistical Analysis

Using Mplus 8.3 software, a latent class growth model (LCGM) was established. The evaluation criteria for the optimal model fit included the Akaike Information Criterion (AIC), a standard metric for assessing statistical model fit; Bayesian Information Criterion (BIC) and adjusted BIC, which account for the degree of misclassification error; entropy, which measures classification accuracy with a range of 0–1; and Lo-Mendell-Rubin (LMR) test and Bootstrap Likelihood Ratio Test (BLRT) to determine whether model fit improved with an increase in the number of classes. Lower AIC, BIC, and adjusted BIC values indicate better model fit.³⁷ Entropy values closer to 1 reflect higher classification accuracy.³⁹ A significant P-value (<0.05) for LMR and BLRT suggests improved fit with additional classes.⁴⁰ SPSS 26.0 was used for univariate and correlation analyses. Amos 26.0 software was employed to test the mediation model, with the bias-corrected percentile bootstrap method applied to verify the significance of the mediation effects. Model fit quality was evaluated using the root mean square error of approximation and goodness-of-fit indices. As the study involved three-wave longitudinal survey data, a cross-lagged panel model was constructed.⁴¹ For the cross-lagged panel mediation model, FEE, IS and FOC were measured at T1, T2 and T3. Autoregressive paths of each variable across consecutive time-points were estimated to represent temporal stability. Cross-lagged paths were constructed to examine directional prospective effects. Within-wave residual correlations were released at each time-point to capture contemporaneous associations. Statistical significance was defined as P<0.05.

Results

Latent Classes of the LCGM for FOC

A total of 136 pregnant women participated in this study. As shown in Table 1, we set 1–5 classes to analyze the LCGM of the FOC. The LMR values of Class 5 were not statistically significant (P>0.05), and there were smaller sample sizes in Class 4, which were excluded from this study. The Entropy values of classes 2 and 3 were both greater than 0.80, indicating good classification accuracy of the model. The AIC, BIC, and aBIC values of Class 3 were all lower than those of Class 2, and the sample representativeness was better than that of Class 2. Therefore, in this study, class 3 FOC were selected as the optimal model.

Table 1.

Fitting Results of LCGM for FOC

Class	AIC	BIC	aBIC	Entropy	LMR(P)	BLRT(P)	Probability (%)
1	3113.541	3128.105	3112.287
2	3036.243	3059.545	3034.237	0.829	0.0066	<0.001	58.8/41.2
3	3029.076	3061.116	3026.318	0.808	0.0353	0.0050	11.0/55.9/33.1
4	3023.710	3064.487	3020.199	0.826	0.0451	0.0305	31.6/5.1/12.5/50.7
5	3026.726	3076.241	3022.462	0.827	0.4744	0.5950	47.1/2.9/28.7/13.2/8.1

Development Trajectory of the Latent Classes for FOC

Table 2 shows the mean values and variations of the intercepts and slopes for the three classes. Class-1 had the lowest initial level of FOC and no upward trend, named the “Low FOC stable group”, comprising 15 participants (11.4%). Class-2 has the highest initial level of FOC and shows an upward trend, named the “High FOC elevated group”, comprising 75 participants (55.1%). The initial level of FOC in Class-3 is moderate with no upward trend, and it is named the “Moderate FOC stable group”, accounting for 46 participants (33.5%). Figure 1 shows the developmental trajectory of the FOC in the three classes at different time points.

Table 2.

The Fitting Index of Three Classes

Class	Index	Mean	SE	t	P
Class-1	Intercept	23.033	2.334	9.876	<0.001
	Slope	0.670	1.162	0.576	0.564
Class-2	Intercept	38.326	1.585	24.185	<0.001
	Slope	4.415	0.930	4.747	<0.001
Class-3	Intercept	32.466	1.945	16.688	<0.001
	Slope	0.317	0.876	0.362	0.717

Figure 1.

Development trajectory model diagram of FOC.

Notes: T1, First trimester; T2, Second trimester; T3, Late trimester.

Abbreviations: FEE, Family Emotional Expression; IS, Inner Strength; FOC, Fear of Childbirth.

Univariate Analysis of Demographic Characteristics

Table 3 shows that pregnant women with higher education levels, full-time wives, average monthly income exceeding 8000, and good mother-in-law daughter-in-law relationships had lower levels of FOC (P<0.05).

Table 3.

The Impact of Demographic Characteristics on FOC (T3, N=136)

Variables	N(%)	Mean (SD)	F/t	P
Age			0.998	0.320
20~29	94	38.85±9.72
30~	42	40.62±9.09
Educational level			11.579	0.001
High school or below	42	43.40±9.94
Undergraduate or above	94	37.61±8.82
Occupation			3.018	0.032
Company employees	82	39.52±9.97a
Public institutions/civil servants	32	41.65±8.01a
Individual/Temporary Worker	13	38.77±9.72ab
Full time wife	9	31.11±6.14b
Per capita monthly income			4.543	0.012
<4000	15	42.40±8.35a
4000–8000	77	40.75±8.68a
>8000	44	36.00±10.54b
First pregnancy			3.659	0.058
Yes	92	38.32±9.01
No	44	41.64±10.30
Expected mode of delivery			1.924	0.150
Natural labour	78	38.03±9.40
Caesarean birth	38	41.34±9.29
Anything	20	41.05±10.07
The Relationship between Mother-in-law and Daughter-in-law			3.233	0.003
Bad	14	42.43±8.04a
Not too bad	44	42.68±9.10a
Good	78	37.00±9.41b

Notes: Values with different single superscript letters (a, b) are significantly different between groups. Values sharing the same single superscript letter show no significant group difference. Values marked with both superscripts a and b (ab) are not significantly different from the group marked a, and also not significantly different from the group marked b.

Correlation Analysis Between Variables

Figure 2 shows that T1 FEE was positively correlated with T2 IS (P<0.01) and negatively correlated with T3 FOC (P<0.01). T2 IS was negatively correlated with T3 FOC (P<0.01).

Figure 2.

Heatmap of correlations between variables across different time points.

Notes: T1, First trimester; T2, Second trimester; T3, Late trimester.

Abbreviations: FEE, Familial Emotional Expression; IS, Inner Strength; FOC, Fear of Childbirth.

Mediating Effect Test

This cross‑lagged path model involved three measurement time points (T1: early pregnancy, T2: mid‑pregnancy, T3: late pregnancy). Analyses strictly followed temporal ordering, such that variables at earlier time points predicted variables at subsequent time points. Autoregressive paths for all variables were simultaneously included in the model to account for the temporal stability of FEE, IS, and FOC across time. Residual correlations among variables within each time‑point were also estimated to statistically adjust for concurrent associations at the same assessment wave. The core hypothesized indirect mediating chain was specified as: T1 FEE → T2 IS → T3 FOC. All path coefficients, standard errors, and significance results are presented in Table 4 and the path diagram (Figure 3). The goodness‑of‑fit test of the four models showed that the relative chi‑square (chi square/degree of freedom) of each model was less than 3, and the root mean square approximation error (RMSEA) of each model was less than 0.08. The Goodness of Fit Index (GFI), Comparative Fit Index (CFI), Incremental Fit Index (IFI), Normed Fit Index (NFI), and Tucker–Lewis Index (TLI) of each model were higher than 0.90. Therefore, the model used in this study has a good and appropriate fit. Table 4 shows that T1 FEE was longitudinally associated with T2 IS, T2 IS was longitudinally associated with T3 FOC, and no significant association was observed between T1 FEE and T3 FOC. The preliminary explanation shows that IS plays a mediating role between FEE and FOC. Figure 3 shows the longitudinal mediating model.

Table 4.

Route Analysis Between FEE, IS and FOC at Different Time Periods

Path			Path Coefficient	SE	Standardization Path Coefficient	C.R.	P
T1 FEE	→	T2 FEE	−0.047	0.096	−0.042	−0.483	0.629
T1 IS	→	T2 IS	0.126	0.054	0.143	2.342	0.019
T1 FOC	→	T2 FOC	0.091	0.060	0.110	1.519	0.129
T1 FEE	→	T2 IS	6.606	0.583	0.691	11.326	<0.001
T1 IS	→	T2 FOC	−0.245	0.034	−0.527	−7.266	<0.001
T2 FEE	→	T3 FEE	−0.029	0.070	−0.035	−0.412	0.681
T2 IS	→	T3 IS	−0.144	0.060	−0.201	−2.415	0.016
T2 FOC	→	T3 FOC	0.462	0.046	0.554	10.159	<0.001
T2 FEE	→	T3 IS	−0.931	0.509	−0.152	−1.829	0.067
T2 IS	→	T3 FOC	−0.244	0.033	−0.552	−7.330	<0.001
T1 FEE	→	T3 FOC	0.333	0.371	0.079	1.052	0.293

Notes: Paths in bold represent the primary mediating pathways. Variables with statistical significance in Table 3 were added as a control variable in the SEM; T1, First trimester; T2, Second trimester; T3, Late trimester.

Abbreviations: FEE, Familial Emotional Expression; IS, Inner Strength; FOC, Fear of Childbirth.

Figure 3.

Longitudinal mediation model with cross lag of is between FEE and FOC.

Notes: Numbers in bold represent the primary mediating pathways. *P<0.05,^**P<0.01; T1, First trimester; T2, Second trimester; T3, Late trimester.

Abbreviations: FEE, Familial Emotional Expression; IS, Inner Strength; FOC, Fear of Childbirth.

As shown in Table 5, the total effect of T1 FEE →T3 FOC was −0.303 (P<0.05). After adding T2 IS, the direct effect of T1 FEE →T3 FOC was 0.079 (P>0.05). The indirect effect of T1 FEE →T3 FOC was −0.382 (P<0.05), indicating the effect of FEE on FOC operates through IS.

Table 5.

Bootstrap Analysis of Mediating Effect Significance Testing Between FEE and FOC

Mediation Effect Influence Path	Path Effect Value	P	95% CI
Total effect	−0.303	0.010	−0.423~-0.189
Direct effect	0.079	0.352	−0.082~0.188
Indirect effect	−0.382	0.010	−0.477~-0.264

Discussion

The Development Trajectory of FOC

The development trajectory of FOC in this study is divided into three categories, namely “Low FOC stable group”, “Moderate FOC stable group” and “High FOC elevated group”. Low-level FOC is characterized by mild anxiety that does not significantly disrupt daily life or pregnancy decisions. Moderate FOC involves higher anxiety levels, which may interfere with sleep or prenatal care adherence. High-level FOC severely affects physical and mental health, often leading to the avoidance of prenatal checkups, strong demands for cesarean delivery, or depressive tendencies. Notably, only the “High FOC elevated group” exhibited a statistically significant increase in fear severity as gestational age advanced. The above-mentioned grouping characteristics represent direct findings of the present study; the subsequent mechanistic interpretations regarding changes in fear of childbirth during pregnancy are mainly derived from previously published evidence. Generally, FOC tends to escalate with pregnancy progression, particularly among those lacking an understanding of the birthing process.⁴² In late pregnancy, physical changes (eg, contractions, frequent fetal movements) and concerns about unexpected medical risks (eg, unplanned cesarean delivery) may further amplify anxiety. Studies indicate that individuals with traumatic birth histories, anxiety/depression tendencies, or inadequate social support experience faster fear escalation during this period.^24,43,44 These women often develop catastrophic thinking by fixating on delivery risks, exacerbating fear, and triggering behaviors such as sleep disturbances or prenatal care avoidance, thereby creating a self-perpetuating “fear-stress-fear amplification” cycle. Clinically, early screening of high-risk groups (eg, psychologically vulnerable or younger mothers) and multidisciplinary support (obstetric, psychological, and educational interventions) are critical for improving maternal and neonatal outcomes. Combined with the heterogeneous trajectories of fear of childbirth identified in the present study and insights from existing literature, early identification and stratified assessment for psychologically high-risk pregnant women may provide potential references for future multidisciplinary obstetric support programmes, with an aim to optimize maternal-infant-related outcomes.

The Mechanism Underlying the Association Between Familial Emotional Expression and FOC

This study showed that the effect of familial emotional expression on FOC operates through inner strength. This result is consistent with our research hypothesis: the association between family emotional expression and fear of childbirth may be indirectly established through pregnant women’s psychological status. Previous literature has indicated that emotional expression from family members (eg, partners, parents), such as encouragement and empathy, may help create a stable psychological environment for pregnant women.⁴⁵ Such social support may mitigate feelings of isolation among pregnant women confronted with unknown childbirth-related risks and reduce anxiety-driven catastrophic thinking.⁴⁶ Through positive feedback, such as affirming the pregnant woman’s physical capabilities and emphasizing the value of her maternal role families foster a constructive perception of childbirth.⁴⁷ Emotional engagement from family members also motivates pregnant women to proactively learn about childbirth and develop personalized birth plans. The above collaborative behaviors may help improve pregnant women’s decision-making ability. By constructing a cognition of “shared risk bearing”, these behaviors may also alleviate pregnant women’s concerns regarding medical interventions (eg, intrapartum conversion from vaginal delivery to cesarean section).²⁴ In summary, based on the mediation model results of the present study, family emotional expression may be correlated with enhanced individual inner resilience in pregnant women. Individual inner strength originates from the potential of the mind, spirit, and self-awareness, manifesting as the capacity to maintain stability, integrate the self, and sustain growth amid stress, conflict, or challenges.²⁹ Its core lies in achieving dynamic equilibrium with external environments through self-perception and mobilization of internal resources. Inner strength enables self-recovery during setbacks, preventing fixation on negative emotions.^48,49 Previous literature has described the cognitive-behavioral pathways of inner strength. Through self-reliance and optimism, pregnant individuals reframe childbirth as a controllable physiological process. For instance, positive expectations for childbirth outcomes may reduce excessive worries about unknown risks and alleviate fear; confidence in one’s own capabilities may strengthen the will to confront adversities and mitigate psychological stress responses.^50,51 Proactive behaviors such as acquiring childbirth knowledge (eg, attending prenatal classes, consulting doctors) reduce excessive concerns about fetal health and role transitions. Inner strength promotes effective resource utilization at the behavioral level by reshaping cognition and enhancing coping capacity. Pregnant individuals with strong inner resilience tend to adopt proactive coping strategies rather than avoidance, forming a dynamic buffer against fear.⁵² This process interacts with maternal traits and social support systems to create an adaptive mechanism for childbirth-related stress. Based on the mediation model of the present study, we propose a theoretical pathway: family emotional support may correlate with pregnant women’s inner psychological resilience, which is further associated with the level of childbirth fear. This theoretical process implies that external social support may be transformed into individual internal psychological resources, which could further regulate childbirth fear through combined physiological and psychological pathways.

Implication

Based on the findings from this observational study, the following represents theoretical conceptual ideas rather-than empirically validated intervention protocols. Changes in the physical and mental state of pregnant women are closely associated with deep-rooted family emotional support. These results may inform the conceptualization of a trinity support framework of “physical participation, emotional resonance, and practical safeguards” for pregnant women. First, an exclusive emotional-connection mechanism for pregnancy might be considered: for example, family members learn about gestational hormonal changes in advance and adopt the “three-step emotional soothing method”. On this basis, family communication patterns could be optimized: pregnant women practice the “Specific Needs Expression Method”, while family members learn how to “decode pregnancy-related needs”. Pregnancy-oriented emotional rituals may be designed, such as weekly family-rotated gratitude expression towards pregnant women, and inviting older family members to share personal pregnancy experiences. A model of “experience inheritance and modern adaptation” might help mitigate intergenerational conflicts around parenting and sustain ongoing family support. In terms of prenatal-care companionship, partners could take part in key prenatal visits; after consultations, emotional affirmation, such as hugs, may validate pregnant women’s efforts. A symbolic “prenatal-result interpretation ritual” could be adopted, for instance organizing ultrasound images in a dedicated album with hand-written blessings. The present findings also offer theoretical references for a biopsychosocial-oriented framework aiming to foster pregnant women’s inner resources, without confirming its practical efficacy. First, physical-oriented empowerment concepts could include phased exercise plans: low-intensity activities such as water yoga or walking in the second trimester, paired with Kegel exercises and breathing-regulation training in late pregnancy; nutritional management ideas, such as keeping emotion-food association logs, co-developing anti-fatigue meal plans with partners to strengthen self-directed nutritional management. Second, cognitive-restructuring-related ideas cover mindfulness practices including fetal-connection meditation (daily belly-touching communication) and contraction-acceptance imagery training; potential cognitive-intervention approaches, such as wearing motivational-engraved wristbands for positive psychological cues, and maintaining “fear-transformation diaries” to reframe anxiety into actionable plans. Finally, social-support-building concepts include joint learning of childbirth-science animations by partners; symbolic rituals such as “pregnancy-achievement medals” (eg awarding a “Glucose Tolerance Warrior” badge after completing the OGTT test). For informed decision-making, visualized childbirth-preference checklists and 3D-anatomy-based science education may facilitate the shift from passive medical reception to active patient engagement.

Strengths and Limitations

This study adopted a longitudinal design, which provides better temporal sequencing of variables compared with cross-sectional studies. However, as an observational investigation, it cannot establish strict causal relationships. Rather, the present findings only reveal potential associative pathways between family emotional expression and FOC. To the best of our knowledge, this study is the first to preliminarily explore the underlying correlational mechanism of familial emotional factors on FOC trajectories during pregnancy. Using LCGM, this study identified heterogeneous developmental trends and their stability throughout gestation. Several limitations should be acknowledged. First, the single-centre convenience sampling restricted the sample representativeness. Given the unbalanced maternal and child healthcare resources and diverse socioeconomic and cultural contexts across China, the generalizability of the current findings to other regions should be interpreted cautiously. Second, although the longitudinal trajectory was captured, the relatively small sample size, particularly for the three-class LCGM classification, may affect the stability of subgroup trajectory distribution. Third, this study only focused on core familial and psychological factors associated with FOC, while other potential influencing factors were not included. Future studies with more comprehensive variable inclusion and larger multi-centre samples are warranted to further verify and extend the present conclusions.

Conclusions

Fear of childbirth (FOC) presents heterogeneous levels across gestation, with the High FOC elevated group showing progressive growth in childbirth fear as pregnancy advances. Familial emotional expression indirectly associated with FOC through its relationship with pregnant women’s inner strength. They offer potential targets and theoretical references for the subsequent development of family-oriented prenatal psychological intervention programmes and stratified psychological risk management. When interpreted in the context of this single-centre convenience-sampling design, the present findings are most applicable to pregnant women receiving prenatal care in large-city tertiary hospitals.

Funding Statement

This study did not receive any funding.

Abbreviations

FOC, Fear of Childbirth; FEE, Familial Emotional Expression; IS, Inner Strength; LCGM, Latent Class Growth Model; AIC, Akaike Information Criterion; BIC, Bayesian Information Criterion; LMR, Lo-Mendell-Rubin Test; BLRT, Bootstrap Likelihood Ratio Test.

Data Sharing Statement

The datasets generated and/or analyzed during the current study are not publicly available because the data also form part of an ongoing study, but are available from the corresponding author upon reasonable request.

Ethical Approval

This study was reviewed and approved by the Ethics Committee of General Hospital of the Northern Theater Command. This study was conducted in accordance with the principles of the Declaration of Helsinki. All participants were requested to read and sign an informed consent form before the start of the study.

Consent for Publication

This was deemed not applicable.

Author Contributions

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis, and interpretation, or in all these areas, took part in drafting, revising, or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Disclosure

The authors report no conflicts of interest in this work.