Multiple Mediation Analysis of a Task-shared Psychosocial Intervention for Perinatal Anxiety: Exploratory Findings from a Randomized Controlled Trial in Pakistan

Abstract

Background:

The “Healthy Mother Healthy Baby” (HMHB) study is a phase three, single-blind randomized clinical trial conducted at Holy Family Hospital (HFH) in association with Rawalpindi Medical University (RMU). We aimed to examine the mediators of a specialized psychosocial approach based on Cognitive Behavioural Therapy principles, targeting pregnant women experiencing anxiety. The HMHB intervention was effective in treating perinatal anxiety symptoms and preventing future depressive episodes.

Methods:

The trial randomized participants into two arms: the HMHB intervention or Enhanced Usual Care (EUC), following World Health Organization guidelines. The HMHB intervention comprised strategies to strengthen social support networks, improving mother-baby bonding and strategies to deal with interpersonal conflicts, economic challenges, and societal gender preferences using stress management techniques and culturally resonant illustrations. Participants underwent rigorous data collection at three pivotal timepoints: baseline, third trimester, and 6-weeks postnatal. The primary outcome was anxiety symptom severity scores using the Hospital Anxiety and Depression Scale (HADS) at 6-weeks post-childbirth. Four potential mediators – social support, behavioural activation, perceived stress, and pregnancy experience – were assessed in the third trimester of pregnancy.

Results:

A total of 1200 participants were randomized to the HMHB and EUC arms. In the six-week follow-up time point, 379 participants remained in the EUC group, and 387 continued in the HMHB group. Post-intervention, HMHB participants displayed significant improvements in postnatal anxiety and depression scores. Mediation analyses revealed social support and pregnancy hassles as significant mediators of the intervention’s effect on postnatal anxiety outcomes, while only social support emerged as a significant mediator for depression outcomes.

Conclusion:

The HMHB intervention showed promising results in improving anxiety and depression scores among pregnant women. Significant mediation effects suggest the importance of targeting social support and managing pregnancy-related hassles for optimal intervention effectiveness.

Background

Pregnancy, while often being a period of joy and anticipation, is marked by profound physical and psychological changes (Waqas et al., 2015). For many expectant mothers, the weight of these changes, along with external factors, can trigger or intensify feelings of anxiety and depression (Gelaye et al., 2016). Acknowledging the intertwined relationship between the mental well-being of the mother and the overall health of both mother and child, the importance of timely interventions becomes paramount (Fisher et al., 2012). In regions like Pakistan, where cultural, societal, and economic pressures exert additional layers of stress, the need for region-specific, evidence-based interventions is even more pronounced (Naveed et al., 2018). Recent studies elucidate several region and culture specific factors which predispose Pakistani women to developing perinatal anxiety. For instance, culturally, there is a strong preference for male offspring rooted in the patriarchal nature of Pakistani society; socially, factors such as low social support, harassment, marital problems, and limited autonomy in decision-making also contribute significantly to mental health issues (Naveed et al., 2018; Waqas et al., 2015; Waqas et al., 2020). Economically, poor household income, unplanned pregnancies, and poor access to quality obstetric and gynaecological care also exacerbate stress levels (Naveed et al., 2018; Waqas et al., 2015; Waqas et al., 2020). The intersection of these determinants creates a complex environment where perinatal mental health is critically impacted.

Despite this pressing need, there remains an absence of preventive interventions specifically tailored for antenatal anxiety in these contexts. This gap is especially concerning given the evident high prevalence of antenatal anxiety and the limited resources allocated for its mitigation (Atif et al., 2019; Atif et al., 2023; Fisher et al., 2012; Kakuma et al., 2011). Historically, antenatal depression has overshadowed the research landscape on CMD during pregnancy. However, it’s critical to recognize the frequent co-occurrence of anxiety and depression in the antenatal phase (Waqas and Rahman, 2021). The ramifications of antenatal anxiety, even when experienced as early as the second trimester, can significantly impact child health (Gelaye et al., 2016). Such observations are particularly pertinent in settings like Pakistan, where a staggering 47% of births are categorized as small-for-gestational age (SGA) (Broekman et al., 2014).

This paper delves into the novel approach of our “Happy Mother – Healthy Baby” intervention (HMHB), which leverages non-specialist providers to administer anxiety-mitigating strategies rooted in cognitive–behavioural therapy (CBT) (Atif et al., 2019; Atif et al., 2023; Surkan et al., 2020). The notable underrepresentation of CBT in preventive initiatives, specifically for anxiety across both high-income and LMICs is perplexing, especially in light of its evidence-base. With the nexus between clinical and subthreshold antenatal anxiety and postnatal depression being established (Waqas et al., 2023a; Waqas and Rahman, 2021; Waqas et al., 2015), our anxiety-centred intervention offers the dual potential of alleviating antenatal anxiety and staving off postnatal depression (Surkan et al., 2023).

HMHB, tailored to the unique psychosocial fabric of the Pakistani context, was developed to treat symptoms of perinatal anxiety and prevent postnatal depression among expectant mothers (Atif et al., 2019; Atif et al., 2023; Surkan et al., 2020). The present study seeks to delve deeper into this domain, exploring the potential pathways that might modulate the efficacy of HMHB. Guided by qualitative insights and a review of existing evidence-based practices (Elias et al., 2024; Rahman et al., 2018; Waqas et al., 2023b; Waqas et al., 2022a), we created the “Happy Mother, Healthy Baby” (HMHB) intervention, rooted in cognitive behaviour therapy principles (Atif et al., 2019). The intervention includes several evidence-based components: building an empathetic relationship, challenging negative thoughts, stress management, relaxation, behavioural activation, problem-solving, and family involvement. These components were applied through a structured three-step approach: 1) identifying unhealthy or unhelpful thoughts and behaviours, 2) replacing them with healthier alternatives, and 3) practicing these healthier thoughts and behaviours. Delivered by non-specialist providers, the intervention utilized culturally relevant illustrations and examples to engage women in helpful behaviours (Atif et al., 2019; Atif et al., 2023). Feedback from the non-specialist providers indicated that the intervention was acceptable, feasible, and perceived positively by the women who participated (Atif et al., 2019; Atif et al., 2023).

The evidence for above noted components of CBT based interventions is robust (Singla et al., 2017; Singla et al., 2021). Previous evidence synthesis exercises show that multicomponent psychosocial interventions for perinatal mental disorders largely comprise of psychoeducation, cognitive and behavioural, interpersonal, birth preparation and caregiver skills (Waqas et al., 2023b). However, previous literature comprising trials and evidence synthesis exercises largely focus on perinatal depression. And there is a paucity of data on trials for strategies tackling perinatal anxiety, as noted by the World Health Organization (Rahman et al., 2018).

The present study addresses the paucity of data by investigating four specific constructs hypothesized to drive the intervention effects of the HMHB intervention: social support, behavioural activation, perceived stress, and improvement in pregnancy experiences. This investigation is grounded in mediation analysis within a randomized controlled trial framework, aiming to evaluate the theoretical foundations of HMHB and elucidate the ‘black box’ of relevant treatment factors. This line of inquiry in important as it improve our understanding of how psychosocial interventions work, and assist clinicians and researchers in developing more effective interventions.

Methods

Study design

The HMHB was a phase three, two-armed, single-blind individual randomized clinical trial. Details on the development of intervention and the study design for the associated RCT are provided elsewhere (Atif et al., 2019; Atif et al., 2023; Surkan et al., 2020). Briefly, this study was conducted at the outpatient Obstetrics and Gynaecology Department of Holy Family Hospital (HFH), a public tertiary care facility associated with Rawalpindi Medical University (RMU). The study design and its protocols were ethically approved by the Johns Hopkins Bloomberg School of Health, the Human Development Research Foundation Ethics Committee, the RMU Institutional Research Forum, and the Global Mental Health Data Safety and Monitoring Board affiliated with the US National Institute of Mental Health.

A trained assessment team, based in the outpatient department, introduced the HMHB study to eligible pregnant women and their families presenting for antenatal care in the Obstetrics and Gynaecology Department. After receiving informed consent, the assessment team screened the pregnant women following a three-phased approach. Firstly, basic demographic and gestational health were assessed against our eligibility criteria: at ≤22 gestational weeks of pregnancy, age ≥18 years old, resides ≤20km from HFH and understands Urdu. Thereafter, women were assessed for further eligibility to participate in the trial where presence of any significant health conditions, learning disabilities, psychiatric disorders, and other acute conditions served as contraindications to participation. At level 3, the research team evaluated women for anxiety and depressive symptoms using the Hospital Anxiety and Depression scale, followed by confirming the absence of a current Major Depressive Episode (MDE) diagnosis using The Structured Clinical Interview for DSM-5 (SCID-5).

Utilizing randomized sequences, participants were assigned to study arms, a strategy that incorporated varying block sizes. The integrity of the study’s single-blind nature was fiercely protected. The allocation status was concealed from the assessment team, and principal investigators. Both the intervention and assessment teams operated in isolation, stationed at distinct locations within the HFH Obstetrics and Gynaecology Department.

Interventions

Happy Mother - Healthy Baby:

The HMHB intervention was a specialized psychosocial approach, tailored for expectant women navigating anxiety during their early to mid-stages of pregnancy. This intervention was grounded in Cognitive Behavioural Therapy (CBT) principles and was borrowed from the foundational tenets of the Thinking Healthy Program (Rahman et al., 2020). This latter program was a validated intervention targeted at mothers grappling with perinatal depression. Its fundamental components encompassed psychoeducation, empathetic engagement, challenging maladaptive thoughts, activating positive behaviours, integrating familial support, and effective problem-solving techniques (Surkan et al., 2020).

In the preparatory phase preceding the trial, qualitative research was employed to pinpoint specific risk factors associated with antenatal anxiety. Notable factors identified were adverse life events, interpersonal conflicts, insufficient support systems, economic challenges, inadequate healthcare access, societal preferences concerning the child’s gender, complications during pregnancy, and apprehensions about childbirth (Atif et al., 2019). Many of these identified risks mirrored findings from other Pakistani research studies (Naveed et al., 2018; Waqas et al., 2023a; Waqas et al., 2015). The HMHB intervention was meticulously designed to address these specific risks. It incorporated techniques for managing stress, such as regulated breathing exercises. A significant feature was the use of culturally resonant illustrations, previously used in the THP, aiding in facilitating guided explorations, activating desired behaviours, stress alleviation, and effectively conveying pivotal health-centric messages (Atif et al., 2019; Surkan et al., 2020).

The core structure of this intervention consisted of five consecutive weekly individual sessions during early to mid-pregnancy. These sessions were enriched by accompanying take-home exercises. Further enhancing the intervention’s robustness was an additional therapy booster session. Depending on each participant’s unique psychological needs and her enrolment timing within the study, reinforcing booster sessions (that did not introduce additional content) were also incorporated prior to the third-trimester final core session. Whenever feasible, these intervention sessions were synchronized with the participants’ standard hospital visits, ensuring streamlined care and convenience. The variable number of booster sessions was a testament to the program’s flexibility, catering to the distinct needs of every expectant mother. All the booster sessions were manualized a priori and provided according to treatment response measured using the HADS scale. Details regarding the development of the intervention and structure of sessions can be read elsewhere (Atif et al., 2019).

Enhanced Usual Care:

Participants randomized to the control group received what was termed Enhanced Usual Care (EUC). Following guidelines from the World Health Organization (WHO), the EUC protocol called for women to receive eight antenatal visits. This frequency aimed to ensure a positive pregnancy experience. Each of these visits was comprehensive, covering health status evaluations, routine examinations consistent with the pregnancy stage, and open forums for addressing any potential concerns. A noteworthy enhancement to the standard care was the additional training the medical staff of HFH underwent, focusing on depression (Surkan et al., 2020). This initiative ensured a heightened sensitivity and understanding of maternal mental health issues. Other tangible enhancements included reimbursements for transportation, free obstetric ultrasounds as well as receive reminders for appointments and facilitated care without waiting in line.

Measures

A detailed battery of clinical and psychosocial assessments was performed with randomized participants at four timepoints: at baseline, during the third trimester, at childbirth, and finally at 6-weeks postnatal in the hospital. Data collection efforts, especially during the baseline and third trimester visits, were harmonized with patients scheduled antenatal visits and the therapists’ availability. Birth outcome data were extracted from medical records.

Outcome

Our study outcome is anxiety and depressive symptom severity scores on the anxiety and depression subscale of Hospital Anxiety and Depression Scale (HADS) (Zigmond and Snaith, 1983) at 6 weeks post-childbirth, as assessed by independent evaluators who were blind to treatment status. This variable was selected over the trials’ primary outcome of remission status because HADS offers a continuous score, which provides more variability in our regression analyses.

Prior to analysis, we also conducted detailed psychometric analyses for the HADS scale. Psychometric analyses using the trial data revealed that HADS had good internal consistency and validity. Overall, the scale demonstrated good internal consistency with a Cronbach’s alpha value of 0.90 among all items. Both the anxiety subscale (α = 0.83) and depression subscale (α = 0.83) also yielded good internal consistency. Exploratory factor analysis revealed that the two-factor solution exhibited satisfactory factor loadings for each item. In the case of the anxiety dimension, factor loadings varied between 0.32 and 0. For the depression dimension, factor loadings ranged from 0.33 to 0.73. Confirmatory factor analysis revealed that this bifactor model was fit, yielding excellent measures of goodness of fit: CFI of 0.95, TLI of 0.94, NFI of 0.95, GFI of 0.96, and PCMIC/DF of 33.11.

Potential mediators

Four separate scales were used to assess the four a priori mediators measured in the third trimester of pregnancy.

Social support:

The Multidimensional Scale of Perceived Social Support (MSPSS) was used to assess the trial participants’ sense of support from three primary sources: family members, friends, and a significant partner (Sharif et al., 2021). Comprising 12 questions, the MSPSS is both widely recognized and rigorously validated in Pakistan (Sharif et al., 2021).

Behavioural activation:

The PREMIUM Abbreviated Activation Scale (PAAS) is a concise tool comprising five items. It was initially crafted and implemented in a separate trial named the Healthy Activity Program (Singla et al., 2021). The PAAS captures five elements related to behavioural activation, which is an essential treatment component in CBT based interventions. It measures the trial participants’ self-reported involvement in diverse activities, like how frequently she took part in various tasks and her efficacy in achieving set objectives. The scale also evaluates associated joy (e.g., ‘Did you participate in pleasurable activities?’) and a sense of accomplishment (‘Were you satisfied with the scope and nature of your activities?’). One of the questions is framed inversely, asking, ‘Did you often dwell on your issues repeatedly?’. Each item is scored from 0, indicating ‘not at all’, to 5, meaning ‘fully agree’, leading to a cumulative score out of 25 (Manos et al., 2011; Singla et al., 2021).

Perceived stress scale:

The Perceived Stress Scale (PSS) was utilized to gauge perceived stress levels. This scale evaluates the extent to which individuals perceive situations in their lives as stressful. It focuses on factors like unpredictability, perceived loss of control, and feelings of being overwhelmed. Additionally, the PSS poses straightforward questions about current stress experiences. The PSS queries participants about their emotions and thoughts over the previous month, asking them to reflect on the frequency of specific feelings (Lee, 2012). This scale demonstrated acceptable internal consistency in the present study sample (Cronbach’s alpha 0.77).

Pregnancy experience scale:

The PES-Brief scale, comprising twenty items, was developed to gauge both positive (“uplifts”) and negative (“hassles”) encounters during pregnancy. Half of the items capture uplifts while the other half focus on hassles. Responses are gathered using a four-point Likert scale, where 3 signifies “A great deal” and 0 means “not at all”. Each item on the PES-Brief exclusively assesses either a hassle or an uplift (Zaidi et al., 2022). Total scores for each subscale were determined by summing up the responses related to the frequency of experiences as indicated on the PES. For current analyses, the hassles subscale of the PES was hypothesized as a potential mediator, because it was an essential target for cognitive restructuring exercises during therapy sessions. The PES-Brief scale has been validated previously in a large sample of Pakistani pregnant women with mild anxiety (Zaidi et al., 2022).

Statistical analysis

This study is a secondary mediation analysis embedded within a randomized controlled trial of the Healthy Mother - Healthy Baby intervention. Multiple mediation analysis was conducted in MPlus 8.1 because it allows testing for multiple mediation paths simultaneously, offers built-in bootstrapping capabilities and ease in handling non-normal outcome data (Muthén and Muthén, 1998–2017). Prior to running of mediation analysis, assumption checks were performed to ensure assumptions for regression analysis (normality, heteroscedasticity, multicollinearity, and independence of observations) were met. We followed Baron & Kenny’s four step approach (MacKinnon et al., 2007; Valente et al., 2016), to ensure assumptions for mediation analyses were met. For this, a series of regression analysis were performed for each mediating pathway to demonstrate that there are significant effects of the independent variable on the proposed mediator (X → M) and of the proposed mediator on the outcome (M → Y), adjusted for the independent variable. Statistical significance was set at P < 0.05. Each of these regression models were controlled for baseline anxiety and depressive symptom scores, participants age, education levels and socioeconomic conditions. We also controlled the path models for baseline scores on the mediators.

After the assumptions for mediation analysis were met, MPLUS was used to run multiple mediation analysis. The significance and effect size for the mediating pathways were assessed using the model indirect command in MPlus 8.1. All mediators were tested in parallel to one another in structural equation model (Figures 1 and 2). Bootstrapped (with 10000 replications) bias-corrected estimate of the 95% confidence interval (CI) for each indirect effect were tested in the mediation analysis (Mackinnon et al., 2004). These estimates are robust to violations of distributional assumptions (1). We also estimated the contribution of each potential mediator on the total effect by dividing each mediating effect by the total effect.

Figure 1:

Multiple mediation analysis exhibiting indirect effects of four mediators on postnatal symptoms of anxiety

Figure 2:

Multiple mediation analysis exhibiting indirect effects of four mediators on postnatal symptoms of depression

Results

A total of 379 participants were randomized to the EUC and 387 to HMHB. On average, participants in both the arms were aged 25.29 (4.65) years, reported a monthly household income of PKR 20,977.83 (SD= 13,730. 48). A higher proportion was educated between the middle to the matriculation (grade 10) level (46.8%) (Table 1). No differences were found between participants who remained versus those who dropped out in both the arms; similarly, there were no differences between treatment and control conditions.

Table 1:

Baseline characteristics of intervention recipients

Characteristics	Control	Intervention
Maternal age	25.45 (4.61)	25.13 (4.70)
Maternal education	2.62 (1.55)	2.55 (1.60)
Household income	20940.65 (13818.48)	21014.24 (13661.54)
Multidimensional Scale of Perceived Social Support (MSPSS)	39.46 (10.23)	38.29 (11.16)
Patient activation (PAAS)	10.87 (2.95)	10.97 (2.64)
Pregnancy Experience Scale (Hassles)	20.43 (7.95)	20.43 (7.95)
Pregnancy Experience Scale (Uplifts)	16.41 (5.98)	16.98 (6.03)
Marital relationship	16.58 (5.14)	16.05 (5.46)
Perceived Stress Scale	20.32 (2.47)	19.93 (2.57)
Depressive symptoms	6.61 (2.63)	6.77 (2.91)
Anxiety symptoms	11.08 (1.88)	11.02 (2.04)

Descriptive frequencies and t-tests of potential mediating variables and clinical outcome are presented in Table 2. Post-intervention, participants randomized to the HMHB reported significant improvement in the postnatal anxiety scores (Cohen’s d= −0.94, 95% CI: −0.77 to −1.10 and postnatal depression scores (d= −0.96, 95% CI: −0.80 to −1.13). Findings regarding effectiveness of the intervention can be read elsewhere (Surkan et al., 2023).

Table 2:

Correlation between mediators (third trimester) and outcomes measured postnatally

	Pregnancy experience (Hassles)	Stress management	Social support	Patient activation
Pregnancy experiences (Hassles frequency)	-	0.1751	−0.3761	−0.3261
Stress	0.1751	-	−0.0062	0.0302
Social support	−.03761	−0.0062	-	0.3151
Patient activation	−0.3261	0.0302	0.3151	-
Postnatal anxiety symptoms	0.2771	0.0072	−0.2651	−0.1961
Postnatal depressive symptoms	0.2761	0.0382	−0.2851	−0.2021

¹denotes p < 0.001,

²denotes p > 0.05

Women enrolled in the HMHB intervention reported significantly improved social support scores (d= 0.43, 95% CI: 0.27 to 0.59), patient activation (d= 0.36, 95% CI: −0.21 to 0.52), stress management (d= −0.16, 95% CI: −0.31 to −0.003) and pregnancy hassles (d= −0.71, 95% CI: −0.87 to −0.55) and pregnancy uplifts (d= 0.54, 95% CI: 0.38 to 0.70). All variables met the assumptions for mediational analyses based on Baron and Kenny’s four-step approach, except perceived stress, which yielded significant associations with the outcome but not with the intervention arm (p = 0.05). There was no evidence of multicollinearity between the variables (VIF < 3).

Mediation analyses

For the outcome of anxiety symptom severity, we found both that social support (a × b= −0.05, 95% CI: −0.097 to −0.019) and pregnancy hassles (a × b= −0.064, 95% CI: −0.14 to −0.012) exerted significant mediation effects, suggesting their independent roles in partly mediating the effects of the HMHB intervention on postnatal anxiety symptoms. This was not the case for the hypothesized mediators of patient activation (a × b= 0.02, 95% CI: −0.06 to 0.001), and stress management (a × b= 0.01, 95% CI: −0.002 to 0.03). In individual mediation analysis, social support accounted for 8.6% and pregnancy hassles accounted for 11.8% of the total intervention effects.

For the outcome of depression symptom severity, only social support (a × b= −0.07, 95% CI: −0.12 to −0.03) emerged as a significant mediator. While patient activation (a × b= −0.02, 95% CI: −0.05 to 0.003), stress management (a × b= 0, −0.01 to 0.01) and pregnancy hassles (a × b= −0.05, 95% CI: −0.11 to 0.01) exerted non-significant indirect effects. In individual mediation analysis, social support accounted for 9.55% of total intervention effects. Figures 1 and 2 present the diagrammatic representation of these mediation analyses.

Discussion

The findings from the current study shed valuable light on the complex interplay of psychosocial factors influencing postnatal anxiety and depressive symptoms among Pakistani perinatal women. Through a comprehensive analysis, two significant mediators emerged, indicating that improvement in social support and reduced pregnancy hassles partially account for the therapeutic effects of the HMHB on postnatal anxiety and preventive effects on postnatal depression.

Our exploration into the mechanisms underpinning the HMHB intervention emphasizes the centrality of social support in psychosocial interventions. In the landscape of psychotherapeutic interventions for perinatal depression, the role of social support, identified in prior research as a non-specific yet pivotal component, is consistently underscored (Grigaitytė and Söderberg, 2021; Malik et al., 2023; Sharif et al., 2021; Singla et al., 2021; Waqas et al., 2015). The pivotal role of social support as a mediator aligns with existing literature which suggests that a robust social network can act as a buffer against postnatal psychological distress (Cohen and Wills, 1985). This study corroborated the well-documented notion that when expectant women and new mothers perceive higher levels of support from their families, friends, and partners, they are better equipped to navigate the myriad challenges of pregnancy and motherhood (Naveed et al., 2018; Sharif et al., 2021; Waqas et al., 2015). In a cultural context like Pakistan, where family and societal ties hold immense significance, amplifying social support can be an effective strategy in alleviating maternal anxiety and depression symptoms (Malik et al., 2023; Naveed et al., 2018; Sharif et al., 2021; Singla et al., 2021). Furthermore, the significant mediation effect of social support on postnatal depression underscores its overarching influence on maternal mental well-being.

These conclusions align with our qualitative findings from anxious pregnant women at HFH (Nazir et al., 2022; Rowther et al., 2020). Nazir et al., revealed that women’s social environments in Pakistan both restrict pregnancy-related agency and contribute to antenatal anxiety. Women’s self-advocacy during pregnancy is often limited to indirect means, such as appealing to husbands or returning to their natal home. The level of autonomy during pregnancy is contingent on the decision-making context, with peer and family support being critical for maternal mental health. Rowther et al., complement these insights, showing that many pregnant women experience physical and social isolation within joint family systems. This isolation, exacerbated by societal expectations and norms about pregnancy, is recognized by healthcare providers as a factor contributing to increased anxiety symptoms, reduced access to care, and poorer health behaviours (Rowther et al., 2020).

Interestingly, the results pertaining to pregnancy hassles provide a nuanced understanding of the role of the pregnancy experience. While past research has recognized the implications of negative pregnancy experiences on postnatal mood (Waqas et al., 2023a), our findings emphasize that interventions like HMHB can equip mothers to better manage these hassles, thereby reducing their impact on postnatal anxiety outcomes (Atif et al., 2019; Atif et al., 2023; Surkan et al., 2020). This reinforces the importance of holistic antenatal care that addresses not just physiological, but also psychological and experiential aspects of pregnancy.

Conversely, the non-significant mediation effects of patient behavioural activation and stress management merit consideration. While both constructs are theoretically linked to psychological well-being (Afaq et al., 2021; Cohen and Wills, 1985; Hopko et al., 2006; Malik et al., 2021; Manos et al., 2011; Singla et al., 2017; Singla et al., 2021; Trombello et al., 2017), their absence as significant mediators in this study might suggest that other prevailing factors, potentially unique to the Pakistani cultural and socio-economic landscape, could overshadow their influence. It’s also plausible that the tools used to measure these constructs may not have captured their essence in the specific context of this population. It is to be noted that the intervention itself led to improvements in behavioural activation and stress management; however, the effect sizes were modest. Therefore, the intervention effects mediated by these constructs may have been weak, requiring higher statistical powers for definitive analyses. More research is needed to further explore above suggestions, as there is scant literature on mediators of psychosocial intervention for perinatal anxiety to draw comparisons at this stage as noted in previous systematic reviews (Elias et al., 2024; Rahman et al., 2018; Waqas et al., 2023b; Waqas et al., 2022a).

Considering the strong evidence that highlights the crucial roles of social support and pregnancy experience, it is essential for future interventions aimed at addressing perinatal depression and anxiety in LMICs to incorporate these elements. Not only does it bridge the gap between intervention and outcomes, but its multifaceted nature also offers a flexible, culturally adaptable avenue for refining psychosocial interventions (Waqas et al., 2023b; Waqas et al., 2022b). We recommend refining and tailoring CBT based psychosocial interventions—ensuring they resonate with the specific cultural and socio-economic contexts of our target population.

Limitations

Despite the rigorous methodology employed and valuable insights obtained, several limitations of this study should be acknowledged. One prominent limitation is the potential existence of more intricate mediational pathways that were not explored in this study. For instance, other important mediators like empathy and the quality of the patient-therapist relationship, which have been hinted at in the literature as essential in such therapies, were not considered (Singla et al., 2017). Their exclusion might have led to an incomplete picture of the entire spectrum of mediational influences. Additionally, mediation studies specifically tailored to psychosocial therapies remain in their nascent stages (Cuijpers et al., 2019; Huibers et al., 2021). The field could benefit from more advanced techniques to accurately delineate mediational mechanisms. For example, ecological momentary assessments, which track real-time psychological experiences and environmental interactions, can offer a richer, dynamic perspective on how interventions like HMHB produce their therapeutic effects (Huibers et al., 2021). Utilizing such sophisticated methods would pave the way for a more comprehensive understanding of the complex interplay of factors influencing therapeutic outcomes in psychosocial interventions. The inherent constraints of self-report measures, potential biases in reporting, and the cultural specificity of the findings, may affect the internal or external validity of these findings. Furthermore, while the study presents promising avenues through which HMHB impacts maternal mental health, a deeper exploration into other potential mediators could provide a more comprehensive understanding.

Conclusion

This study accentuates the importance of targeted psychosocial interventions like HMHB in the Pakistani maternal health landscape. By revealing the pivotal roles of social support and pregnancy experience in modulating postnatal anxiety and depression, it offers valuable insights for clinicians, researchers, and policymakers. For holistic maternal care, integrating such evidence-based interventions, cognizant of cultural and societal nuances, will be crucial. Future research can further explore the intricate web of mediators and expand on the promising foundation laid by this study.