ABSTRACT
Background
Evidence gaps remain regarding the influence of perinatal depression on mother–child engagement and child social–emotional development.
Objectives
We assessed relationships between perinatal depression, mother–child engagement and child social–emotional development among Kenyan mother–child pairs.
Methods
Mother–child pairs attending maternal‐child health services in four sites in Western Kenya were followed from pregnancy through early childhood. Study nurses serially assessed perinatal depression (pregnancy, 6 weeks, and 9 months postpartum with the Center for Epidemiologic Studies Depression scale, CESD‐10 scores ≥ 10), mother–child engagement activities (6‐monthly, 24 to 60 months post‐delivery with UNICEF Multiple Indicator Cluster Surveys) and child social–emotional delay (6‐monthly, 30 to 60 months post‐delivery with Ages and Stages Questionnaires). We estimated prevalence and correlates of low mother–child engagement and social–emotional delay.
Results
Among 884 mothers, the median age was 26 years (IQR 22.0, 30.3), 91.6% were married, and 36.8% had perinatal depression. High mother–child engagement (≥ 4 activities in the prior 3 days) ranged from 27.1% to 94.1% from 24 to 60 months post‐delivery. The frequency of child social–emotional delay ranged from 26.6% at 30 months to 4.4% at 60 months. Low mother–child engagement at any point (< 4 activities) was more common among women with perinatal depression (adjusted relative risk [RR] 1.20, 95% confidence interval [CI] 1.08, 1.33) and was associated with twice the risk of child social–emotional delay (RR 2.22, 95% CI 1.78, 2.77). Mothers who reported adverse childhood experiences (ACES) (RR 1.08, 95% CI 1.04, 1.11) and intimate partner violence (IPV) (RR 1.28, 95% CI 1.11, 1.47) interacted less frequently with their children than women without these experiences.
Conclusion
In this cohort of Kenyan mother–child pairs followed from pregnancy through childhood, perinatal depressive symptoms were associated with lower mother–child engagement, which was associated with double the risk of child social–emotional delay.
Keywords: early childhood development, Kenya, mother–child engagement, perinatal depression, social–emotional development
1. Background
Globally, depression in the perinatal period is associated with long‐lasting consequences for a range of adverse mother and child health outcomes [1]. Potential adverse outcomes include preterm birth and small‐for‐gestational age and extend to poor mother–child bonding, child developmental challenges and even adolescent depression and anxiety [1]. Existing evidence from high‐income countries supports a relationship between maternal mental distress and child social–emotional development, with a proposed mechanism linking them through mother–child attachment [2, 3]. As such, maternal psychological distress is linked to child mental health challenges—a leading cause of youth morbidity and mortality globally [4].
Perinatal depression is more common in low‐ and middle‐income countries (LMICs) in the African Region compared to high‐income countries (> 25% [5, 6] vs. 10% [7]). Other factors that diminish psychosocial well‐being are more common during the perinatal period in low‐resource settings, such as financial stress, intimate partner violence and HIV risk [8]. Early childhood development in this region lags behind higher‐resourced settings; social–emotional delay is particularly high in East Africa (31%) [9]. The impact of maternal mental disorders and child social–emotional delays is compounded by the low availability of psychological support services in the region. Gaps in maternal‐child health could be alleviated by understanding linked maternal‐child psychological outcomes and addressing them through mental and developmental health services integrated into routine care.
Few studies evaluate relationships between perinatal mental health and later child developmental outcomes. One cross‐sectional study among mother–child pairs in Kenya and Zambia found an association between maternal engagement activities and child developmental scores [10]. A few studies in South Africa have longitudinally followed mother‐infant pairs into early (2–5 years) and even late (10 years) childhood, identifying associations between maternal depression and suboptimal child physical growth and behavioural challenges [11, 12]. However, these studies did not evaluate behavioural interactions between mothers and children in early childhood or serially assess child social–emotional development to fully characterise the linked nature of mother–child psychological outcomes in this formative period.
To inform early childhood development programs in East Africa, we conducted a longitudinal analysis among Kenyan mother‐infant pairs to estimate the extent of mother–child engagement and risk of child social–emotional delay at multiple time points in early childhood (2–5 years old). We further explored relationships between perinatal mental health and later mother–child behavioural engagement, as well as the impact on child social–emotional development.
2. Methods
2.1. Cohort Selection
This longitudinal analysis utilised data from the PrEP Implementation for Mothers in Antenatal Care (PrIMA) Study and the PrIMA Extension Study (PrIMA‐X study). The PrIMA study was a cluster randomised controlled trial evaluating two models for offering pre‐exposure prophylaxis for HIV prevention to Kenyan pregnant HIV‐negative women receiving maternal child health (MCH) care in 20 public sector health facilities in Western Kenya. Between January 2018 and July 2019, women attending MCH were screened and enrolled. Eligible women were pregnant, HIV‐uninfected, ≥ 15 years old and were able to provide consent. The PrIMA‐X study enrolled PrIMA participants in a follow‐on longitudinal observational safety evaluation of PrEP use during and after pregnancy. HIV‐negative women aged ≥ 15 years who had completed the PrIMA study and were receiving MCH services (postnatal care or child immunisations) in four of the PrIMA sites were recruited to participate in the PrIMA‐X study between October 2020 and July 2022. Mother–child dyads were followed up to 60 months after birth. Data from pregnancy through 9 months postpartum were leveraged from the PrIMA study [13]. Our analysis includes data collected from January 2018 (PrIMA study) through October 2024 (PrIMA‐X study). We involved people with related lived experience via a community advisory board to inform study design and implementation.
2.2. Procedures and Measures
Questionnaires were administered by trained study nurses in Kiswahili, Dhuluo or English using tablet‐based REDCap surveys. We collected information about demographics, pregnancy history, partner characteristics, psychosocial factors, mother‐infant engagement activities, and child social–emotional development. Household crowding—a marker of socioeconomic status—was defined as the ratio of people per room greater than the median (≥ 2 people/room) [14]. We evaluated maternal adverse childhood experiences using the 43‐item World Health Organisation Adverse Childhood Experiences International Questionnaire (ACES‐IQ) with the binary scoring approach [15]. ACES‐IQ scores range from 0–13; higher scores indicate more adverse experiences during childhood, and dichotomised high ACES as scores ≥ 6. To evaluate maternal social support during pregnancy, we used the 18‐item Medical Outcomes Study Social Support Survey (MOS‐SSS), in which higher scores indicate greater social support (range: 18–90) [16]. We dichotomised this variable with a cut‐point of ≥ 72 for low social support [17]. We used the 4‐item Hurt, Insult, Threaten, Scream Scale to assess experience of intimate partner violence (IPV) during pregnancy, with a cut‐point of ≥ 10 to define IPV (range: 4–20) [18].
2.3. Exposure
Experience of depressive symptoms was collected at enrolment during pregnancy, 6 weeks, and 9 months postpartum using the 10‐item Center for Epidemiologic Studies Depression Scale (CESD‐10). We used a validated cut‐point of 10 or greater to define moderate‐to‐severe depressive symptoms, hereafter called ‘depression’ (CESD‐10 score ≥ 10) [19]. The CESD‐10 is commonly applied in large‐scale epidemiologic studies globally and was validated for use among perinatal African populations [20] and in Kenya [21]. Women were categorised with perinatal depression if they had CESD‐10 scores ≥ 10 at least once measured at the following visits: pregnancy (enrolment), 6 weeks, 9 months postpartum.
2.4. Outcomes
Mother–child engagement (MCE) activities were assessed from 24 to 60 months post‐delivery using UNICEF Multiple Indicator Cluster Survey Version‐6 items from the questionnaire for children under five [22]. Mothers reported whether they had engaged in activities with their child in the past 3 days (reading, storytelling, singing, going outside, playing and naming/counting/drawing). Engaging in < 4 activities in the past 3 days was associated with low MCE, consistent with UNICEF reporting [23]. Child social–emotional developmental delay (SED) was assessed using the Ages and Stages Questionnaire: Social–Emotional Second Edition (ASQ: SE‐2), which asks mothers to report on the frequency of a series of child behaviours over time. We used standard cutpoints for social–emotional delay warranting referral (30‐month: ≥ 85; 36‐month: ≥ 105; 48‐month: ≥ 85; 60‐month: ≥ 95) [24].
We selected the ASQ: SE‐2 to measure child social–emotional development in the PrIMA study for multiple reasons. This tool is increasingly used in large‐scale epidemiologic studies globally; thus, our use of ASQ: SE‐2 improves the comparability of our findings with other settings. As a self‐reported measure based on a parent's own experience with the child, the burden of administration is low. This facilitates efficient screening in busy public sector MCH clinics. A validation study among South African and Zambian children supports the feasibility and psychometric properties of the Ages and Stages Questionnaire in the region [25]. The Kenyan Ministry of Health does not currently maintain policy guidelines specifying a measurement tool for child social–emotional development for use in MCH settings; we will align future studies with tools specified in Kenyan policies.
Our primary research questions were (a) whether and to what extent perinatal depression is associated with mother–child engagement activities among older children (24–60 months post‐delivery) and (b) whether and to what extent mother–child engagement activities were associated with infant social–emotional development in this age group. We hypothesised that women who had experienced perinatal depression would demonstrate lower engagement with their children in the period from 24 to 60 months post‐delivery. We hypothesised that children would have an elevated risk for social–emotional developmental delay if they had experienced lower mother–child engagement at an earlier age. We further hypothesised that the strength of the relationship between mother–child engagement and infant social–emotional development would differ across levels of a mother's prior experience of perinatal depression.
2.5. Statistical Analysis
Prevalence of low MCE was calculated at each 6‐monthly study visit from 24 months to 60 months post‐delivery. Prevalence of child SED was calculated for each ASQ:SE‐2 milestone timepoint from 30–60 months post‐delivery. The average number of MCE activities and ASQ: SE‐2 scores were also calculated at each visit (median, interquartile range [IQR]). Demographic characteristics were evaluated using descriptive statistics (overall and stratified by perinatal depression).
To identify correlates of low MCE from 24–60 months post‐delivery, we fit univariable models in which the correlates of interest served as independent variables and low MCE from 24–60 months post‐delivery as the dependent variable. We estimated model parameters using robust Poisson regression based on the generalised estimating equations methods (independent correlation structure, robust standard errors and clustered by participant). Correlates of the number of mother–child interaction activities were also evaluated using similar estimation methods (Gaussian family, identity link). In multivariable models, variables hypothesised a priori to confound relationships between maternal factors and low MCE (maternal age [years], marital status [married/living with partner vs. not married/living with partner], household crowding [> 2 people per room versus ≤ 2 people per room]), and gestational age in weeks at enrolment in PrIMA were included. We used similar estimation methods to identify correlates of SED and ASQ: SE‐2 score from 30–60 months post‐delivery (SED: Poisson family, log link; ASQ: SE‐2 score: Gaussian family, identity link). When evaluating the relationship between mother–child engagement and child social–emotional development, we used time‐lagged variables to examine associations between mother–child engagement measured at one MCH visit and child social–emotional development evaluated at the next 6‐monthly MCH visit, ensuring appropriate temporal ordering.
To evaluate potential effect modification by perinatal depression in the relationship between MCE and child SED on the multiplicative scale, models with an interaction term for perinatal depression and mother–child engagement (low MCE, number of mother–child activities) were fit. We also evaluated interaction on the additive scale using the relative excess risk due to interaction (RERI) with 95% confidence intervals estimated using 2000 non‐parametric bootstrap resamples.
2.6. Missing Data
For depressive symptom scale data, item‐level scores were imputed using multiple imputation with chained equations for participants missing data in < 5 out of 10 depressive symptom scale items (12.7% 114/892) [26]. We used 50 imputed datasets and incorporated item‐level depressive symptoms and characteristics associated with missingness (maternal age, marital status and multiparity) [27]. This approach retains the efficiency of statistical estimates [28]. We did not analyse scores for participants missing over half of the scale items. We report the denominators for each time point in tables and figures, indicating the number of participants who reported data for that visit.
3. Results
3.1. Participant Characteristics
Overall, 99.1% (884/892) of mothers enrolled in the PrIMA/PrIMA‐X study had child SED data for at least one visit between 30 to 60 months. Median maternal age during pregnancy was 26 years (interquartile range [IQR]: 22.0–30.3), the majority of mothers were married, and median years of educational attainment were 10.0 (IQR: 8.0, 12.0). Slightly over half of the children were female, and about one in ten were preterm births (Table 1). Over a quarter of mothers experienced high adverse childhood experiences (ACES score ≥ 6), the most frequent of which was the death of one or both parents, which occurred during the woman's childhood. One in ten women reported intimate partner violence in the 2 weeks prior to their pregnancy enrolment visit. A quarter of mothers reported moderate‐to‐severe depressive symptoms during pregnancy; slightly over a third reported depression at least once from pregnancy through 9 months post‐delivery. Among the women included in this analysis (n = 10, 1.1%), participants were lost to follow‐up before the 60‐month post‐delivery visit, and 0.2% (n = 2) participants voluntarily exited the study prior to the 60‐month post‐delivery visit.
TABLE 1.
Characteristics of Kenyan mother–child pairs (N = 884).
| Overall (N = 884) (*) | Perinatal MSD (CESD‐10 ≥ 10) N = 325 | No perinatal MSD (CESD‐10 < 10) N = 559 | |
|---|---|---|---|
| n (%) or Median (IQR) | n (%) or Median (IQR) | n (%) or Median (IQR) | |
| Demographic characteristics | |||
| Age (years) | 25.9 (22.0, 30.3) | 26.2 (22.6, 30.2) | 25.4 (21.8, 30.4) |
| Adolescents and young women (≤ 24 years) | 551 (62.4) | 213 (65.7) | 338 (60.5) |
| Gestational age (enrolment, weeks) | 24.0 (19.0, 28.0) | 24.0 (19.0, 28.0) | 24.0 (18.0, 28.0) |
| Married or living with a partner | 805 (91.6) | 299 (92.3) | 506 (91.2) |
| Completed education (years) | 10.0 (8.0, 12.0) | 10.0 (8.0, 12.0) | 11.0 (8.0, 13.0) |
| Completed education (< 12 years) | 500 (58.3) | 188 (60.8) | 312 (56.8) |
| Regularly employed | 152 (17.3) | 55 (17.0) | 97 (17.5) |
| Financial support from partner a | 862 (99.2) | 316 (98.8) | 546 (99.5) |
| Household crowding (≥ 2 people/room) | 395 (44.8) | 144 (44.4) | 251 (45.1) |
| Multiparous | 729 (82.6) | 282 (86.8) | 447 (80.1) |
| Psychosocial characteristics | |||
| Adverse childhood experiences score b | 4.0 (2.0, 6.0) | 4.0 (2.0, 6.0) | 4.0 (2.0, 6.0) |
| High adverse childhood experiences b | 245 (28.7) | 108 (34.7) | 137 (25.2) |
| Social support score c | 76 (67, 87) | 74.0 (62.5, 84.0) | 77.0 (69.5, 90.0) |
| Low social support c | 271 (31.4) | 118 (36.9) | 153 (28.1) |
| Intimate partner violence d | 96 (10.9) | 61 (18.8) | 35 (6.3) |
| CESD‐10 score e | 4.0 (2.0, 10.0) | 12.0 (8.0, 15.0) | 3.0 (0.0, 5.0) |
| Child characteristics | |||
| Infant sex (female) | 438 (51.8) | 153 (49.8) | 285 (52.9) |
| Preterm birth | 96 (10.9) | 40 (12.5) | 56 (10.1) |
Note: (*)We had < 5% missingness for each variable.
Among those with a current partner.
We evaluated adverse childhood experiences using the 43‐item Adverse Childhood Experiences International Questionnaire (ACES‐IQ) with the binary scoring method (absolute range: 0–13; higher scores indicate higher ACES); defining high ACES as scores of 6 and above (high ACES: ACES‐IQ score ≥ 6 = ‘Yes’, ACES‐IQ score < 6 = ‘No’)|.
We evaluated social support at enrolment in pregnancy using the 18‐item Medical Outcomes Study social support score (MOS‐SSS), defining low social support as scores below 72 (Low social support: MOS‐SSS score < 72 = ‘Yes’, MOS‐SSS score ≥ 72 = ‘No’).
We evaluated intimate partner violence at enrolment in pregnancy using the 4‐item Hurt, Insult, Threaten, and Scream scale (HITS), defining intimate partner violence as scores of 10 and above (IPV: HITS score ≥ 10 = ‘Yes’, HITS score < 10 = ‘No’).
We evaluated moderate‐to‐severe depressive symptoms in the perinatal period using the 10‐item Center for Epidemiologic Studies Depression Scale (CESD‐10), defining MSD as scores of 10 and above (MSD: CESD‐10 score ≥ 10 = ‘Yes’, CESD‐10 score < 10 = ‘No’).
3.2. Prevalence and Correlates of Parent‐To‐Child Engagement
High mother–child engagement (≥ 4 MCE activities in the prior 3 days) ranged from 27.1% to 94.0% during the period from 24 months to 60 months post‐delivery (Figure 1). Singing to a child and taking him/her outside were the most frequent types of mother–child activities; telling him/her a story was least common, yet still consistently high (> 50% starting at 30 months). Engagement was lowest at 24 months post‐delivery (mean: 1.6 activities, standard deviation: 2.4), and mothers reached peak engagement with children by 48 months post‐delivery (mean: 5.5 activities, SD: 1.3).
FIGURE 1.
Mother–child engagement from 24–60 months post‐delivery among Kenyan mother–child pairs. *The mean (standard deviation) for the number of mother–child engagement activities at each time point is shown.
Mothers with less than 12 years of education were more likely to report low MCE and fewer engagement activities than those with higher educational attainment (Table 2). Mother–child pairs in crowded households were more likely to have low MCE and fewer MCE activities in the prior week compared to their counterparts in uncrowded households.
TABLE 2.
Associations between maternal factors and low mother–child engagement (24 months—60 months post‐delivery) N = 883 women, N = 4882 visits.
| Low mother–child engagement (< 4 activities past 3 days vs. ≥ 4 activities) | Number of mother–child engagement activities (past 3 days) | |||
|---|---|---|---|---|
| Univariable analysis (1) | Multivariable analysis (2) | Univariable analysis (3) | Multivariable analysis (4) | |
| RR (95% CI) | aRR (95% CI) | β (95% CI) | aβ (95% CI) | |
| Demographic characteristics | ||||
| Age (years) | 1.01 (1.00, 1.02) | 1.01 (0.99, 1.01) | −0.01 (−0.02, 0.001) | −0.01 (−0.02, 0.004) |
| Adolescents and young women (≤ 24 years) | 1.05 (0.95, 1.17) | 1.02 (0.92, 1.14) | −0.08 (−0.21, 0.06) | −0.04 (−0.18, 0.09) |
| Gestational age (enrolment, weeks) | 1.00 (0.99, 1.01) | 1.00 (1.00, 1.01) | −0.002 (−0.01, −0.01) | −0.002 (−0.01, 0.01) |
| Married or living with a partner | 0.92 (0.77, 1.10) | 0.92 (0.76, 1.09) | 0.09 (−0.15, 0.33) | 0.10 (−0.14, 0.34) |
| Completed education (< 12 years) | 1.32 (1.19, 1.48) | 1.31 (1.16, 1.47) | −0.32 (−0.45, −0.19) | −0.30 (−0.44, −0.16) |
| Regularly employed | 0.97 (0.85, 1.11) | 1.00 (0.87, 1.16) | 0.08 (−0.09, 0.24) | 0.05 (−0.13, 0.22) |
| Financial support from partner a | 1.12 (0.55, 2.29) | 1.18 (0.57, 2.43) | −0.19 (−1.09, 0.70) | −0.23 (−1.14, 0.64) |
| Household crowding (≥ 2 people/room) | 1.20 (1.08, 1.33) | 1.19 (1.07, 1.32) | −0.20 (−0.34, −0.07) | −0.17 (−0.31, −0.04) |
| Multiparous | 1.12 (0.97, 1.29) | 1.09 (0.93, 1.28) | −0.18 (−0.34, −0.02) | −0.14 (−0.33, 0.05) |
| Psychosocial characteristics | ||||
| Adverse childhood experiences score b | 1.09 (1.06, 1.13) | 1.08 (1.04, 1.11) | −0.13 (−0.18, −0.08) | −0.10 (−0.15, −0.05) |
| High adverse childhood experiences b | 1.06 (0.95, 1.19) | 1.05 (0.94, 1.19) | −0.08 (−0.23, 0.07) | −0.06 (−0.21, 0.08) |
| Low social support c | 1.09 (0.98, 1.22) | 1.08 (0.97, 1.21) | −0.13 (−0.28, −0.01) | −0.12 (−0.26, 0.02) |
| Intimate partner violence d | 1.29 (1.12, 1.49) | 1.28 (1.11, 1.47) | −0.37 (−0.59, −0.15) | −0.36 (−0.59, −0.14) |
| Perinatal depression e | 1.21 (1.09, 1.34) | 1.20 (1.08, 1.33) | −0.26 (−0.40, −0.13) | −0.26 (−0.39, −0.12) |
| Child characteristics | ||||
| Infant sex (male) | 0.99 (0.89, 1.11) | 1.01 (0.90, 1.11) | −0.01 (−0.12, 0.15) | 0.01 (−0.12, 0.14) |
| Preterm birth | 1.09 (0.92, 1.31) | 1.09 (0.92, 1.30) | −0.11 (−0.34, 0.14) | −0.10 (−0.34, 0.14) |
Note: (1) Relative risks from univariable generalised estimating equation with Poisson family, log link, independent correlation structure, clustered by participant, with outcome variable: low mother–child engagement (low MCE: < 4 activities prior 3 days vs. ≥ 4 activities). (2) Adjusted relative risks from multivariable generalised estimating equation with Poisson family, log link, independent correlation structure, clustered by participant with outcome variable: low mother–child engagement (low MCE: < 4 activities prior 3 days vs. ≥ 4 activities); enrolment adjustment variables: Maternal age (years), Married/living with a partner (Y/N), household crowding (Y/N), gestational age (weeks). (3) Adjusted relative risks from multivariable generalised estimating equation with Gaussian family, identity link, independent correlation structure, clustered by participant with outcome variable: number of mother–child engagement activities (prior 3 days); adjustment variables: Age (years), Married/living with a partner (Y/N), household crowding (Y/N).
Among those with a current partner.
We evaluated adverse childhood experiences using the 43‐item Adverse Childhood Experiences International Questionnaire (ACES‐IQ) with the frequency scoring method; defining high ACES as scores of 6 and above (high ACES: ACES‐IQ score ≥ 6 = ‘Yes’, ACES‐IQ score < 6 = ‘No’)|.
We evaluated social support at enrolment during pregnancy using the 18‐item Medical Outcomes Study social support score (MOS‐SSS), defining low social support as scores below 72 (Low social support: MOS‐SSS score < 72 = ‘Yes’, MOS‐SSS score ≥ 72 = ‘No’).
We evaluated intimate partner violence at enrolment during pregnancy using the 4‐item Hurt, Insult, Threaten, and Scream scale (HITS), defining intimate partner violence as scores of 10 and above (IPV: HITS score ≥ 10 = ‘Yes’, HITS score < 10 = ‘No’).
We evaluated moderate‐to‐severe depressive symptoms during the perinatal period using the 10‐item Center for Epidemiologic Studies Depression Scale (CESD‐10), defining MSD as scores of 10 and above (MSD: CESD‐10 score ≥ 10 = ‘Yes’, CESD‐10 score < 10 = ‘No’).
Multiple psychosocial factors were associated with low MCE and the number of MCE activities. Mothers reporting a higher frequency of adverse experiences in childhood were more likely to demonstrate low MCE and fewer MCE activities. Mothers reporting intimate partner violence were over 25% more likely to have low MCE and also reported fewer MCE activities than those not reporting IPV. When mothers experienced perinatal depression, they were about 20% more likely to engage in lower MCE activities from 24–60 months post‐delivery compared to those who did not report perinatal depression. We did not identify differences in MCE by parity, child gender, or preterm birth status.
3.3. Prevalence and Correlates of Social–Emotional Development Delay
Child SED (ASQ: SE‐2 scores ≥ age‐specific threshold) ranged from 4.4% to 26.6% during the period from 30 to 60 months (Figure 2). SED was more frequent in earlier childhood (30 months post‐delivery) compared to later childhood (60 months post‐delivery).
FIGURE 2.
Social–emotional developmental delay from 30–60 months post‐delivery among Kenyan children. *Scores indicating referral for further assessment of social–emotional development are needed: 30‐month scores ≥ 85; 36‐month scores ≥ 105; 48‐month scores ≥ 85; 60‐month scores ≥ 95. The median (interquartile range) for the social–emotional development score is shown.
Associations with both SED (delay) and with the SED score were assessed. The risk of SED was higher among children whose mothers were married or cohabiting and among those in crowded households, compared with those without these conditions (Table 3). However, we did not identify a relationship between maternal marital status or household crowding and social–emotional developmental scores. No relationships between maternal educational attainment or maternal ACES and child SED were found. However, SED scores were higher (worse) among children whose mothers had < 12 years of education compared to those with higher education.
TABLE 3.
Associations between maternal factors and child social–emotional development (30 months—60 months post‐delivery) N = 877 women; N = 2322 visits.
| Social–emotional developmental delay (SED) vs. Non‐SED | Social–emotional development score | |||
|---|---|---|---|---|
| Univariable analysis (1) | Multivariable analysis (2) | Univariable analysis (3) | Multivariable analysis (4) | |
| RR (95% CI) | aRR (95% CI) | β (95% CI) | aβ (95% CI) | |
| Demographic characteristics | ||||
| Age (years) | 1.00 (0.98, 1.02) | 0.99 (0.97, 1.02) | 0.03 (−0.18, 0.24) | −0.03 (−0.25, 0.18) |
| Adolescents and young women (≤ 24 years) | 0.94 (0.74, 1.19) | 0.85 (0.66, 1.09) | 0.02 (−2.52, 2.56) | −0.72 (−3.33, 1.89) |
| Gestational age (enrolment, weeks) | 1.00 (0.98, 1.01) | 1.00 (0.97, 1.02) | −0.05 (−0.20, 0.10) | −0.04 (−0.21, 0.10) |
| Married or living with a partner | 1.56 (0.98, 2.49) | 1.61 (1.00, 2.60) | 3.83 (−0.40, 8.05) | 3.76 (−0.51, 8.02) |
| Completed education (< 12 years) | 1.14 (0.90, 1.45) | 1.03 (0.81, 1.32) | 2.38 (−0.12, 4.88) | 2.06 (−0.46, 4.57) |
| Regularly employed | 0.99 (0.73, 1.34) | 1.06 (0.77, 1.44) | 1.60 (−1.81, 5.01) | 1.94 (−1.68, 5.56) |
| Financial support from partner a | 0.77 (0.36, 1.68) | 0.60 (0.27, 1.25) | −4.41 (−15.21, 6.39) | −6.95 (−17.63, 3.72) |
| Household crowding (≥ 2 people/room) | 1.34 (1.07, 1.68) | 1.38 (1.09, 1.74) | 1.43 (−1.16, 4.03) | 1.71 (−0.91, 4.33) |
| Multiparous | 1.27 (0.91, 1.78) | 1.17 (0.79, 1.73) | 1.28 (−1.83, 4.38) | 0.56 (−2.92, 4.05) |
| Psychosocial characteristics | ||||
| Adverse childhood experiences b | 1.04 (0.97, 1.11) | 1.04 (0.97, 1.12) | 0.56 (−0.18, 1.31) | 0.69 (−0.07, 1.45) |
| High adverse childhood experiences b | 0.82 (0.63, 1.06) | 0.83 (0.63, 1.08) | 1.80 (−0.73, 4.33) | 2.09 (−0.49, 4.68) |
| Low social support c | 1.11 (0.88, 1.42) | 1.12 (0.88, 1.43) | 0.45 (−2.32, 3.21) | 0.80 (−1.99, 3.60) |
| Intimate partner violence d | 1.45 (1.06, 1.99) | 1.35 (0.97, 1.88) | 4.96 (1.31, 8.61) | 4.43 (0.66, 8.21) |
| Perinatal depression e | 0.78 (0.61, 0.99) | 0.77 (0.61, 0.98) | 0.95 (−1.45, 3.34) | 1.01 (−1.39, 3.41) |
| Child characteristics | ||||
| Infant sex (male) | 1.06 (0.84, 1.33) | 1.06 (0.84, 1.34) | 1.70 (−0.80, 4.19) | 1.82 (−0.68, 4.31) |
| Preterm birth | 0.85 (0.57, 1.27) | 0.87 (0.59, 1.30) | −0.90 (−5.58, 3.80) | −0.73 (−5.42, 3.96) |
| Mother–child engagement (prior visit) | ||||
| Low MCE (< 4 activities per prior 3‐days) f | 2.26 (1.82, 2.82) | 2.22 (1.78, 2.77) | 6.99 (3.42, 10.56) | 6.99 (3.45, 10.53) |
| Zero MCE activities (prior 3 days) f | 2.80 (2.22, 3.54) | 2.76 (2.18, 3.48) | 12.81 (7.87, 17.64) | 12.96 (8.18, 17.73) |
Note: (1) Relative risks from univariable generalised estimating equation with Poisson family, log link, independent correlation structure, clustered by participant, with outcome variable: social–emotional delay (SED: ASQ:SE‐2 score ≥ age‐specific threshold vs. ASQ:SE‐2 score<age‐specific threshold). (2) Adjusted relative risks from multivariable generalised estimating equation with Poisson family, log link, independent correlation structure, clustered by participant with outcome variable: social–emotional delay (SED: ASQ:SE‐2 score≥age‐specific threshold vs. ASQ:SE‐2 score<age‐specific threshold); enrolment adjustment variables: Maternal age (years), Married/living with a partner (Y/N), household crowding (Y/N), gestational age (weeks). (3) Relative risks from univariable generalised estimating equation with Gaussian family, identity link, independent correlation structure, clustered by participant, with outcome variable: Social–emotional development score (ASQ:SE‐2). (4) Adjusted relative risks from multivariable generalised estimating equation with Gaussian family, identity link, independent correlation structure, clustered by participant with outcome variable: Social–emotional development score (ASQ:SE‐2); adjustment variables: Age (years), Married/living with a partner (Y/N), household crowding (Y/N).
Among those with a current partner.
We evaluated adverse childhood experiences using the 43‐item Adverse Childhood Experiences International Questionnaire (ACES‐IQ) with the frequency scoring method; defining high ACES as scores of 6 and above (high ACES: ACES‐IQ score ≥ 6 = ‘Yes’, ACES‐IQ score < 6 = ‘No’)|.
We evaluated social support at enrolment in pregnancy using the 18‐item Medical Outcomes Study social support score (MOS‐SSS), defining low social support as scores below 72 (Low social support: MOS‐SSS score < 72 = ‘Yes’, MOS‐SSS score ≥ 72 = ‘No’).
We evaluated intimate partner violence at enrolment in pregnancy using the 4‐item Hurt, Insult, Threaten, and Scream scale (HITS), defining intimate partner violence as scores of 10 and above (IPV: HITS score ≥ 10 = ‘Yes’, HITS score < 10 = ‘No’).
We evaluated moderate‐to‐severe depressive symptoms during the perinatal period using the 10‐item Center for Epidemiologic Studies Depression Scale (CESD‐10), defining MSD as scores of 10 and above (MSD: CESD‐10 score ≥ 10 = ‘Yes’, CESD‐10 score < 10 = ‘No’).
We evaluated mother–child engagement (MCE) activities using the UNICEF Multiple Indicator Cluster Survey (Version‐6) with the parent–child interaction items, defining low MCE as scores of < 4 activities in the prior 3 days vs. ≥ 4 activities.
Children of mothers who experienced IPV during pregnancy were over 30% more likely to exhibit SED compared to those not reporting IPV. Similarly, SED scores were higher among children whose mothers reported IPV during pregnancy compared to those not reporting IPV. SED was less likely among children whose mothers reported perinatal depression compared to those without perinatal depression. However, no differences in SED scores among children by maternal perinatal depression status were found.
Exposure to low MCE at the preceding post‐delivery visit was associated with twice the risk of child SED at the subsequent visit, compared to exposure to high MCE at the preceding visit. Correspondingly, exposure to low MCE at the prior visit was associated with SED scores that were nearly 7 points higher on average at the subsequent visit than exposure to high MCE. Among children with zero MCE activities at the prior visit, the risk of SED was nearly three times higher, and SED scores were, on average, 12 points higher than among children with some MCE activities. No differences in risk of SED or SED scores by other maternal demographic or psychosocial factors or by infant sex or preterm birth were found.
3.4. Moderating Effect of Perinatal Depression
Potential differences in the relationship between low MCE and risk of SED stratified by maternal perinatal depression were evaluated. We identified a difference in the level of effect of low MCE on SED by perinatal depression status using the relative excess risk due to interaction (RERI). The combined exposure to low MCE and perinatal depression was associated with a synergistic effect on the risk of SED (RERI 0.28, 95% CI −0.49, 0.95). Among children whose mothers experienced perinatal depression, being exposed to low MCE was associated with triple the risk of SED compared to those exposed to high MCE when evaluated on the multiplicative scale (aRR 3.09, 95% CI 2.11, 4.56). Among children whose mothers did not experience perinatal depression, being exposed to low MCE was associated with twice the risk of SED compared to those exposed to high MCE (aRR 1.98, 95% CI 1.51, 2.60). We did not find differences in the relationship between MCE level and SED score when stratified by perinatal depression.
4. Comment
4.1. Principal Findings
In our study of a cohort of Kenyan mother‐infant pairs followed from pregnancy through up to 5 years, exposure to low maternal engagement was associated with twice the risk of social–emotional delay compared with exposure to higher engagement. Like prior studies from high‐income country settings [29], mothers in Kenya with perinatal depression were less likely to engage with their children than those without depression symptoms. Child social–emotional delay was associated with multiple maternal psychosocial factors among this cohort, including experience of intimate partner violence and low mother–child engagement, which aligns with a growing body of evidence highlighting the negative impact of a mother's adverse childhood experiences on her interactions with her children, including in LMICs [30].
4.2. Strengths of the Study
Prospective data from a large cohort of pregnant women enabled rigorous evaluation of relationships among multiple psychosocial factors and child social–emotional development at multiple time points spanning pregnancy through 60 months post‐delivery.
4.3. Limitations of the Data
All measures were self‐reported by mothers and not confirmed by observation, making them subject to recall and social desirability bias, as well as to potential bias related to mental health status. However, the inclusion of self‐reported measures rather than resource‐intensive observational assessments enabled our team to feasibly collect the first longitudinal data from pregnancy through 5 years among a large cohort of mother‐infant pairs in an understudied setting. Use of the MICS items for mother–child engagement facilitates comparability across global UNICEF studies. This study is nested in an ongoing longitudinal study among mother‐infant pairs followed for up to 60 months. Thus, the participants may differ systematically from women and mothers who are not inclined to participate in a research study for an extended period; this may limit the generalisability of our results.
Results are subject to residual confounding. Our team deliberately chose not to pursue mediation analyses since our aim was to assess relationships of interest and evaluate effect measure modification. Our aim was not to understand a potential mechanism of the impact of perinatal depression through mother–child engagement on infant social–emotional development. Further, the presence of exposure‐mediation interaction introduces challenges in traditional mediation analysis methods for measurement and interpretability [31]. Future research could explore this potential mechanism among mother–child pairs in East Africa.
4.4. Interpretation
Maternal peripartum depression was associated with lower maternal‐child engagement, which, in turn, was associated with child socio‐emotional delay. Despite these associations and counter to our hypotheses based on prior studies, we found that perinatal depression was associated with lower (as opposed to higher) risk of child social–emotional delay. It is unclear why we found a negative association with delay. It is possible that there was differential reporting bias regarding child social–emotional delay between women with and without peripartum depression. Additionally, peripartum depression appeared to synergistically modify the association between maternal engagement and child social–emotional delay. Further, in East African contexts, the perinatal period is characterised by familial and communal support [32], which may overcome deficits in child engagement that typically link perinatal depression and child social–emotional delay. Evidence linking mother–child engagement activities with child social–emotional development is strong in high‐income settings [3]. We found that low engagement was associated with twice the risk of child social–emotional delay among these Kenyan mother–child pairs. In cases with extremely low MCE (zero mother–child engagement activities in the prior 3 days), the risk of social–emotional delay at the subsequent milestone was three times higher. This suggests a potential dose–response relationship where the benefits to child social–emotional development increase with more frequent mother–child engagement.
Less frequent mother–child engagement (e.g., playing together) was associated with increased risk of child social–emotional delay, which is high in the African Region [9]. These findings highlight the need to integrate mental health services within healthcare settings frequented by Kenyan mother–child pairs. Parenting interventions in LMIC settings that offer psychosocial support, psychoeducation, and parenting/family skills training have been shown to effectively improve several mother–child outcomes [33]. Evidence‐based psychological interventions, such as cognitive behaviour therapy or problem‐solving skills training, are effective in preventing and treating perinatal depression [34]. In lower‐resource settings, both psychological and parenting interventions have been delivered by peer or lay counsellors, demonstrating a feasible approach to offering needed support services [34]. Integrating low‐intensity, evidence‐based perinatal mental health and parenting interventions into well‐attended healthcare settings like antenatal and postnatal care could have long‐lasting benefits for linked mother–child mental health outcomes.
5. Conclusions
In this large cohort of Kenyan mother–child pairs followed from pregnancy through 5 years of early childhood, low mother–child engagement was more common among women who experienced perinatal depression and low engagement was associated with twice the risk of child social–emotional delay. Mothers reporting ACEs, household crowding, and intimate partner violence interacted less frequently with their children. Having a mother who experienced IPV was associated with poorer social–emotional development. Interventions that address perinatal depression and encourage developmentally appropriate mother–child engagement should be integrated into antenatal and postnatal health services and prioritised for those reporting adverse childhood experiences and intimate partner violence during pregnancy.
Author Contributions
Conceptualisation: A.L., J.K., G.J.‐S., J.P., formal analysis: A.L., M.M., S.W., L.G., J.S., J.P., funding acquisition: J.K., J.P., G.J.‐S., investigation: A.L., J.K., J.C.D., L.G., M.M., F.A., B.O., N.N., G.J.‐S., J.P., methodology: B.A.R., J.K., G.J.‐S., J.P., project administration: J.K., J.C.D., L.G., F.A., S.W., B.O., N.N., resources: J.K., G.J.‐S., J.P., supervision: J.K., F.A., G.J.‐S., J.P., visualisation: G.J.‐S., writing – original draft: A.L., G.J.‐S., J.P., writing – review and editing: A.L., J.K., F.A., J.C.D., L.G., M.M., N.N., S.W., J.S., B.A.R., G.J.‐S., J.P.
Funding
This study was funded by the Eunice Kennedy Shriver National Institute of Child Health and Human Development (R01‐HD100201) with support from the University of Washington/Fred Hutch Center for AIDS Research (P30‐AI027757) and National Institute of Mental Health (K01MH137324‐A1). AL was additionally supported by NICHD (F32‐HD108857). JP was additionally supported by NICHD (R01‐HD108041) and the National Institute of Nursing Research (R01‐NR019220).
Ethics Statement
Ethical approval was obtained from the University of Washington Institutional Review Board (STUDY00008710) and the Kenyatta National Hospital, University of Nairobi Ethics and Research Committee (KNH‐UoN ERC: P921/11/2019). The study received approval from the Siaya and Homabay county Departments of Health and facility administrators. All participants provided written informed consent.
Conflicts of Interest
This paper represents the opinions of the authors and is not meant to represent the position or opinions of organisations, nor the official position of any staff members. Dr. Larsen reports receiving NIH grants during the conduct of the study. Dr. John‐Stewart reports grants from NIH, CDC and Thrasher Foundation, and personal fees from UpToDate and UW, outside the submitted work. Dr. Kinuthia reports receiving NIH grants during the study. Dr. Pintye reports receiving NIH grants during the study. All other authors have nothing to report.
Acknowledgements
We thank the PrIMA Extension study team nurses. We gratefully acknowledge the women who participated in this study for their time and contributions. We also thank the Kenyan Ministry of Health, Homa Bay and Siaya Counties, and the facility heads and in‐charges for their collaboration.
Larsen A., Kinuthia J., Abuna F., et al., “Perinatal Depression, Maternal Engagement and Child Social–Emotional Development: A Cohort Study,” Paediatric and Perinatal Epidemiology 40, no. 5 (2026): 577–587, 10.1111/ppe.70140.
Editor's note: A Commentary based on this article appears on pages 588‐590.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
- 1. Stein A., Pearson R. M., Goodman S. H., et al., “Effects of Perinatal Mental Disorders on the Fetus and Child,” Lancet 384 (2014): 1800–1819. [DOI] [PubMed] [Google Scholar]
- 2. Madigan S., Oatley H., Racine N., et al., “A Meta‐Analysis of Maternal Prenatal Depression and Anxiety on Child Socioemotional Development,” Journal of the American Academy of Child and Adolescent Psychiatry 57 (2018): 645–657. [DOI] [PubMed] [Google Scholar]
- 3. De Waal N., Boekhorst M. G. B. M., Nyklíček I., and Pop V. J. M., Maternal‐Infant Bonding and Partner Support During Pregnancy and Postpartum: Associations With Early Child Social‐Emotional Development 72 (2023): 871, 10.1016/j.infbeh.2023.101871. [DOI] [PubMed] [Google Scholar]
- 4. Kieling C., Baker‐Henningham H., Belfer M., et al., “Child and Adolescent Mental Health Worldwide: Evidence for Action,” Lancet 378 (2011): 1515–1525. [DOI] [PubMed] [Google Scholar]
- 5. Dadi A. F., Wolde H. F., Baraki A. G., and Akalu T. Y., “Epidemiology of Antenatal Depression in Africa: A Systematic Review and Meta‐Analysis,” BMC Pregnancy and Childbirth 20 (2020): 20, 10.1186/S12884-020-02929-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Dadi A. F., Akalu T. Y., Baraki A. G., and Wolde H. F., “Epidemiology of Postnatal Depression and Its Associated Factors in Africa: A Systematic Review and Meta‐Analysis,” PLoS One 15 (2020): 15, 10.1371/journal.pone.0231940. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Fisher J., Cabral de Mello M., Patel V., et al., “Prevalence and Determinants of Common Perinatal Mental Disorders in Women in Low‐ and Lower‐Middle‐Income Countries: A Systematic Review,” Bulletin of the World Health Organization 90 (2012): 139–149. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Gelaye B., Rondon M., Araya R., and Williams M., “Epidemiology of Maternal Depression, Risk Factors, and Child Outcomes in Low‐Income and Middle‐Income Countries,” Lancet Psychiatry 3 (2016): 973–982. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Gil J. D. C., Ewerling F., Ferreira L. Z., and Barros A. J. D., “Early Childhood Suspected Developmental Delay in 63 Low‐ and Middle‐Income Countries: Large Within‐ and Between‐Country Inequalities Documented Using National Health Surveys,” Journal of Global Health 10 (2020): 10, 10.7189/JOGH.10.010427. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Onyango S., Kitsao‐Wekulo P., Langat N., et al., “Maternal Stimulation and Early Child Development in Sub‐Saharan Africa: Evidence From Kenya and Zambia,” BMC Public Health 23 (2023): 1–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Verkuijl N. E., Richter L., Norris S. A., Stein A., Avan B., and Ramchandani P. G., “Postnatal Depressive Symptoms and Child Psychological Development at 10 Years: A Prospective Study of Longitudinal Data From the South African Birth to Twenty Cohort,” Lancet Psychiatry 1 (2014): 454–460. [DOI] [PubMed] [Google Scholar]
- 12. Gordon S., Rotheram‐Fuller E., Rezvan P., Stewart J., Christodoulou J., and Tomlinson M., “Maternal Depressed Mood and Child Development Over the First Five Years of Life in South Africa,” Journal of Affective Disorders 294 (2021): 346–356. [DOI] [PubMed] [Google Scholar]
- 13. Dettinger J. C., Kinuthia J., Pintye J., et al., “PrEP Implementation for Mothers in Antenatal Care (PrIMA): Study Protocol of a Cluster Randomised Trial,” BMJ Open 9 (2019): 9, 10.1136/bmjopen-2018-025122. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Melki I. S., Beydoun H. A., Khogali M., Tamim H., and Yunis K. A., “National Collaborative Perinatal Neonatal Network (NCPNN). Household Crowding Index: A Correlate of Socioeconomic Status and Inter‐Pregnancy Spacing in an Urban Setting,” Journal of Epidemiology and Community Health 2004, no. 58 (1978): 476–480. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. World Health Organization , Adverse Childhood Experiences International Questionnaire (ACE‐IQ) (Switzerland, 2018). [Google Scholar]
- 16. Sherbourne C. D. and Stewart A. L., “The MOS Social Support Survey,” Social Science & Medicine 32 (1991): 705–714. [DOI] [PubMed] [Google Scholar]
- 17. Gold K. J., Spangenberg K., Wobil P., and Schwenk T. L., “Depression and Risk Factors for Depression Among Mothers of Sick Infants in Kumasi, Ghana,” International Journal of Gynecology & Obstetrics 120 (2013): 228–231. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Sherin K. M., Sinacore J. M., Li X. Q., Zitter R. E., and Shakil A., “Hits: A Short Domestic Violence Screening Tool for Use in a Family Practice Setting,” Family Medicine 30 (1998): 508–512. [PubMed] [Google Scholar]
- 19. Andresen E. M., Malmgren J. A., Carter W. B., and Patrick D. L., “Screening for Depression in Well Older Adults: Evaluation of a Short Form of the CES‐D (Center for Epidemiologic Studies Depression Scale),” American Journal of Preventive Medicine 10 (1994): 77–84. [PubMed] [Google Scholar]
- 20. Baron E. C., Davies T., and Lund C., “Validation of the 10‐Item Centre for Epidemiological Studies Depression Scale (CES‐D‐10) in Zulu, Xhosa and Afrikaans Populations in South Africa,” BMC Psychiatry 17 (2017): 16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Kilburn K., Prencipe L., Hjelm L., Peterman A., Handa S., and Palermo T., “Examination of Performance of the Center for Epidemiologic Studies Depression Scale Short Form 10 Among African Youth in Poor, Rural Households,” BMC Psychiatry 18 (2018): 201. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. UNICEF , Multiple Indicator Cluster Survey Questionnaire for Children Under Five (Round 6) (Bureau of Statistics, 2018). [Google Scholar]
- 23. UNICEF , Development of the Early Childhood Development Index in MICS Surveys (UNICEF, 2017). [Google Scholar]
- 24. Squires J., Bricker D., and Twombly E., Ages & Stages Questionnaires: Social‐Emotional, Second Edition (ASQ:SE‐2): A Parent‐Completed Child Monitoring System for Social‐Emotional Behaviors, (Brookes Publishing, 2015). [Google Scholar]
- 25. van Heerden A., Hsiao C., Matafwali B., Louw J., and Richter L., “Support for the Feasibility of the Ages and Stages Questionnaire as a Developmental Screening Tool: A Cross‐Sectional Study of South African and Zambian Children Aged 2–60 Months,” BMC Pediatrics 17 (2017): 1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Royston P. and White I. R., “Multiple Imputation by Chained Equations (MICE): Implementation in Stata,” Journal of Statistical Software 45 (2011): 1–20. [Google Scholar]
- 27. Bono C., Ried L. D., Kimberlin C., and Vogel B., “Missing Data on the Center for Epidemiologic Studies Depression Scale: A Comparison of 4 Imputation Techniques,” Research in Social and Administrative Pharmacy 3 (2007): 1–27. [DOI] [PubMed] [Google Scholar]
- 28. White I. R., Royston P., and Wood A. M., “Multiple Imputation Using Chained Equations: Issues and Guidance for Practice,” Statistics in Medicine 30 (2011): 377–399. [DOI] [PubMed] [Google Scholar]
- 29. Martucci M., Aceti F., Giacchetti N., and Sogos C., “The Mother‐Baby Bond: A Systematic Review About Perinatal Depression and Child Developmental Disorders,” Rivista di Psichiatria 56 (2021): 223–236. [DOI] [PubMed] [Google Scholar]
- 30. Imran N., Liaqat S., Bodla Z. H., Zeshan M., and Naveed S., “Impact of Adverse Childhood Experiences on the Mother‐Infant Social‐Emotional Well‐Being in Lowand Middle‐Income Countries,” Psychiatric Annals 51 (2021): 316–321. [Google Scholar]
- 31. Richiardi L., Bellocco R., and Zugna D., “Mediation Analysis in Epidemiology: Methods, Interpretation and Bias,” International Journal of Epidemiology 42 (2013): 1511–1519. [DOI] [PubMed] [Google Scholar]
- 32. Musie M. R., Anokwuru R. A., Ngunyulu R. N., and Lukhele S., “African Indigenous Beliefs and Practices During Pregnancy, Birth and After Birth,” in Working With Indigenous Knowledge: Strategies for Health Professionals, ed. Mulaudzi F. M., and Lebese R. T. (2022), 85–106. [PubMed] [Google Scholar]
- 33. Pedersen G. A., Smallegange E., Coetzee A., et al., “A Systematic Review of the Evidence for Family and Parenting Interventions in Low‐ and Middle‐Income Countries: Child and Youth Mental Health Outcomes,” Journal of Child and Family Studies 28 (2019): 2036–2055. [Google Scholar]
- 34. Chowdhary N., Sikander S., Atif N., et al., “The Content and Delivery of Psychological Interventions for Perinatal Depression by Non‐Specialist Health Workers in Low and Middle Income Countries: A Systematic Review,” Best Practice & Research. Clinical Obstetrics & Gynaecology 28 (2014): 113–133. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.