Parental emotional adjustment as a primary target for parenting programs: a cross-sectional study

Abstract

Background

Parental emotional adjustment refers to the degree of distress associated with the parenting role and can be correlated with increased use of negative parenting practices. Parenting programs are a critical strategy globally to help ensure children reach their full developmental potential; however, their effects on parental mental health outcomes are inconsistent.

Methods

Cross-sectional self-report survey data from mothers, fathers, and caregivers (e.g., grandmother) from Afghanistan, Indonesia, Kyrgyzstan, Malaysia, and Uzbekistan were analysed to explore: (1) the relative contributions of parental adjustment on parenting practices as measured by the Parenting and Family Adjustment Scales, and (2) differences in parenting practices across diverse contexts. Descriptive statistics were used to characterise the sample. Multiple logistic regression was used to explore the relationship between parental adjustment and parenting generally and parental consistency, coercive parenting, positive encouragement, and behaviours supporting the parent-child relationship specifically.

Results

A total of 642 participants (mean age = 33.2 years; 79% female; 89.9% partnered) from Afghanistan (n = 111), Indonesia (n = 157), Kyrgyzstan (n = 118), Malaysia (n = 103), and Uzbekistan (n = 153) completed the survey. Results showed that parental adjustment was significantly positively associated with parenting practices, explaining 5.9% of the unique variance. Specifically, better parental adjustment was significantly positively associated with the use of positive parenting practices, including praise, attention, affection, and displays of pride, but not significantly associated with negative parenting practices such as spanking, shouting, and getting angry with children. Finally, significant differences in parental consistency, coercive parenting, positive encouragement, and parenting practices supporting the parent-child relationship were found across the five LMICs.

Conclusions

A robust literature demonstrates the importance of supporting parents’ mental health and emotional adjustment to improve both parent and child outcomes. Our results highlight that better emotional adjustment is associated with more positive parenting practices, suggesting that interventions that support a parent’s ability to cope with the parenting role will increase the use of responsive and nurturing childrearing practices. However, further research is now required to understand how best to integrate mental health-related content with other key intervention material (e.g., training in responsive caregiving), including with regards to the optimal ‘dose’ of each component, to optimise effectiveness.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40359-025-02646-8.

Background

Parenting and early childhood development

The United Nations Convention on the Rights of the Child emphasizes the significant influence of parenting on early childhood development stating, “Parents or, as the case may be, legal guardians, have the primary responsibility for the upbringing and development of the child. The best interests of the child will be their basic concern” [1]. Even in the context of adversity, nurturing care and safety provided by parents and families promote health and wellbeing in early childhood, improving functional and quality of life outcomes in adolescence and adulthood [2, 3]. Without this buffering caregiving effect, extensive research has demonstrated a linear relationship between early adversity and the likelihood of negative outcomes in learning, behaviour and physical and mental health [4–6]. Importantly, the benefits of quality parenting, characterized by nurturing care, responsiveness, and stimulation, for early childhood development are evident across low-, middle-, and high-income countries [2, 7].

Clinical mental health symptoms and parenting practices

There is a large body of literature demonstrating that poor mental health among parents, including specifically in relation to childrearing [8], negatively impacts child development and wellbeing and disrupts family functioning [9–11]. For example, maternal depression can negatively impact attachment security for infants [12]. This association may be due to reduced levels of responsiveness and emotional availability on the part of the mother [12]. A study conducted with 265 Ghanaian parents, including both mothers and fathers, found that parental depression was associated with more negative parenting practices, such as discouragement of negative emotions, physical abuse, and psychological abuse [13]. In turn, children were observed to exhibit more externalizing behaviours (e.g., fighting) and attention problems (e.g., easily distracted) [13]. A recent systematic review of 26 studies conducted in seven low- and middle-income countries (LMICs) confirmed the significant relationship between parental mental health problems and impairments in socioemotional development in children aged 24- to 59-months, with maternal depression being the most frequently studied mental health problem among parents [14]. Importantly, a recent mediation meta-analysis showed that parenting practices significantly mediated the relationship between maternal depression and childhood outcomes (e.g., social, emotional, and cognitive functioning) [15]. Though the effect size of the mediation model was small (r = 0.016), the analyses highlighted that increases in positive parenting and decreases in negative parenting or both acted as mediators in the association between maternal depression and functioning in early childhood [15].

Parenting stress and emotional adjustment

It is important to recognise that all adults can be vulnerable to stress and, in turn, emotional adjustment problems when they become parents as this reflects a major life change [16]. Parenting stress has been defined as “a set of processes that lead to aversive psychological and physiological reactions arising from attempts to adapt to the demands of parenthood” [17]. A myriad of factors dynamically influence the degree of stress experienced by parents, including parenting expectations, perceived demands, available resources, and parent-child and parent-family relationships. Parenting stress impacts parents differently; however, for some it manifests as adjustment problems. Parental adjustment relates to the mental health (i.e., the state of mental wellbeing) of a parent in the parenting role [18]. Parental adjustment problems, include sadness, worry, stress, diminished coping skills, and reduced life satisfaction. Studies have found that the prevalence of adjustment problems for parents with infants is as high as 33% amongst mothers and 17% amongst fathers [16, 19]. Research about parental adjustment is limited; however, it is known that subclinical symptoms of depression and anxiety can impair psychosocial functioning and disrupt family relationships, including between parents and children. Indeed, parental adjustment difficulties, such as worry and distress, put parents at greater risk of engaging in negative parenting practices [20]. For example, it has been shown that parental dysphoria, defined as mild depressed mood, was associated with reductions in focused engagement in enriching activities (e.g., reading, playing), but not childcare activities (e.g., feeding, bathing, dressing) [20]. Notably, adjustment problems are associated with the direct effects of parenthood as well as individual risk factors such as stressors (e.g., financial difficulties) [21] and pre-existing mental health problems [16].

Importantly, social support relationships and networks, including with partners, broader families, and communities, can be protective for parent’s mental health and emotional adjustment [22]. Emotional support has been defined as “the positive, potentially health promoting or stress-buffering, aspects of relationships such as instrumental aid, emotional caring or concern, and information” [23]. A qualitative study highlighted that South African child caregivers who felt more socially supported reported better mental health and, in turn, were less likely to use negative parenting practices (e.g., shouting) and more likely to engage in positive parenting practices (e.g., play) [24].

Parenting programs and interventions

The relationship between poor parental adjustment and parenting behaviours suggests that parents may benefit from interventions that provide guidance and support to help them adapt to the parenting role. This is likely to be of particular significance in LMICs where parents often confront multiple social determinants that put them at risk for parenting stress and adjustment problems, including socioeconomic disadvantage, a history of adverse childhood experiences, discrimination, inequality, and a poor-quality physical environment (e.g., air and noise pollution, poor housing quality) [25]. Parenting programs and interventions have been shown to be an effective way to educate parents about nurturing, responsive and stimulating childrearing practices, with outcomes showing improvements in parent-child interactions and early childhood development [26–29]. Importantly, evidence of effectiveness extends beyond high-income countries (HICs). Indeed, a large systematic review and meta-analysis of 111 studies identified significantly larger effects of parenting interventions with regards to parenting practices as well as child cognitive, language and motor development outcomes in LMICs relative to HICs [28]. Nevertheless, the effects of parenting programs and interventions specifically on parental mental health and adjustment are inconsistent in the literature. For example, evidence from a study conducted with Ugandan mothers indicates that parenting programs that specifically teach parents emotion regulation skills in addition to promoting responsive and stimulating caregiving result in significant reductions in depression symptoms among mothers [30]. Improvements in cognitive and language outcomes for children were also reported [30]. In contrast, a meta-analysis of 16 interventions targeting both parenting and parental mental health did not find significant reductions in depressive symptoms amongst female caregivers [31]. Furthermore, a systematic review of 11 studies concluded that there was insufficient evidence to indicate that parenting interventions are an effective way to improve symptoms of depression, anxiety, or stress among mothers [32].

Study aim

Parenting programs consistently emphasize the importance of increasing opportunities for early play and learning, including promoting increased parental responsiveness and participation in stimulating activities [28]. However, few interventions include content targeting parental mental health specifically [28]. This is particularly striking as our own research conducted in LMICs highlights that parents are interested in learning strategies to support their emotional adjustment to the parenting role [33]. Specifically, parents we spoke to recognised that higher levels of stress and worry can negatively impact their parenting practices (e.g., shouting) and, in turn, their children’s experiences of them as parents [33]. Furthermore, as shown in Textbox 1, qualitative data collected during our team’s evaluation of a digital parenting program implemented in diverse LMICs (described in more detail in the Methods section), emphasized a strong desire for support from parents and caregivers.

Textbox 1.

Qualitative data emphasizing the importance of supporting parental adjustment

“A parent is very vulnerable mentally because they have to be with their child 24 hours a day, and especially if our partner is not capable of understanding us because they are tired from work. Sometimes it shakes mothers… so it would be great if we can have a solution to help us young mothers feel a bit better if we are at the point where we are stressed out.”

Mother, Indonesia, 2022

“During the first two months of motherhood, I felt like I was very low because I didn’t really know how to manage this human, I mean, how to understand her. There was support from my mother, from my husband’s mother, but I don’t think it was something I need because they would not relate, they cannot relate. They do not know what it is. Maybe for the second baby it’s going to be different, but for now it is the way it is.”

Mother, Uzbekistan, 2023

“My point was that a really happy mother means a happy child. If mother is alright, her health and mental condition is fine, then the child will be alright as well. So really, we need to put more emphasis on mother’s wellbeing.”

Mother, Uzbekistan, 2023

In light of the above, this study sought to answer the research question: Should parental emotional adjustment be a primary target for parenting interventions? Cross-sectional quantitative data provided by parents and caregivers from Afghanistan, Indonesia, Kyrgyzstan, Malaysia, and Uzbekistan were used to: (1) measure the relative contributions of parental adjustment to parenting, including both positive (e.g., talking, hugging, and praising children) and negative (e.g., shouting, getting annoyed at, and smacking children) parenting behaviours; and (2) investigate differences in parenting based on country of residence, level of education, and employment status. The STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) Checklist for cross-sectional studies [34] is provided as Additional File 1.

Methods

Study context

The cross-sectional survey study reported in this paper was embedded within a larger prospective, mixed methods, multi-site study evaluating Minderoo Foundation’s Thrive by Five International Program. Thrive by Five aimed to empower parents and other caregivers (e.g., grandparents, community leaders) with evidence-based information to support the healthy development of children from birth to age five. The Youth Mental Health and Technology Team at the University of Sydney’s (Australia) Brain and Mind Centre was responsible for developing the content for the Program. The research team comprised of researchers with diverse disciplinary training as well as a broad range of cultural and ethnic backgrounds that traversed WEIRD (i.e., Western, Educated, Industrialised, Rich, Democratic) countries (Australia, England, and the United States) and LMICs (Bangladesh, Iran, India, and Pakistan) as well as Indigenous backgrounds.

The Thrive by Five content is underpinned by a scientific framework that highlights key neurobiological systems critical to early child development [35] and developed for each country in accordance with a standardized co-design protocol [36]. Regarding the latter, parents, other caregivers, and subject matter experts (e.g., clinical psychologists, early childhood educators, medical specialists, anthropologists, linguists) were key stakeholders throughout the research and development processes, providing invaluable insights to inform the program’s design and iterative refinement across diverse contexts. Specifically, all content was developed, refined, and ultimately finalised in consultation with child caregivers and subject matter experts through an extensive and iterative co-design process conducted in partnership with in-country partners and local champions. Translations of the content into local languages were also approached with considerable care and ample time was allocated for refinement based on expert feedback.

The co-designed content was distributed primarily via country-specific iterations of the Thrive by Five app available for free through local app stores as well as via other digital (e.g., WhatsApp chatbot) and non-digital methods (e.g., printed materials). The standardized protocol for the broader study evaluating Thrive by Five has been published elsewhere and primarily aimed to evaluate the impact of the co-designed Thrive by Five app and its content on parent and caregiver knowledge, behaviours, attitudes, and confidence, and the perceived strength of connection between a child and their parents, family, community, and culture [37]. Results from this multi-site evaluation study were fed back to in-country partners and on-the-ground subject matter experts to allow for the collaborative review and refinement of the interpretation of the findings. In this study, we are using cross-sectional survey data collected after participants had used the Thrive by Five app in five LMICs.

Participant recruitment

Participants for this study were recruited over two years (2022–2023). All participants were parents or caregivers (e.g., grandparents, aunt, uncle, nanny) who had used the Thrive by Five app and its content naturalistically; there were no minimum standards with regards to time spent using the app or engaging with the content. Inclusion criteria included being aged 18-years or older and self-identifying as a parent or caregiver of at least one child between birth and 5 years of age. Recruitment for the evaluation study was facilitated by the research sites in each respective country using their established networks and advertising mechanisms under the leadership of the Site Principal Investigator. As documented in the evaluation protocol and supported by the Central Limit Theorem [38], each site aimed to recruit a minimum of 100 participants [37]. Recruitment methods were chosen based on appropriateness for the site’s community and context and included emails, poster displays, paper-based and online internal news articles, handouts, and digital advertisements on social media. The research team actively worked with in-country partners to strategise how to improve representativeness of the sample and how these potential participants might be recruited.

Informed consent and data collection procedures

Whenever possible, a web-based Participant Information Sheet (PIS) was distributed to interested participants via REDCap [39, 40], a secure electronic data collection and management tool hosted at the University of Sydney, using a public link. However, to account for variability in access to the internet, computers, and smartphones, paper-based versions of the PIS were also available, distributed via post or by hand by the site. As detailed in the published evaluation protocol [37], the PIS was translated from English into local languages based on guidance from the site. All participants provided informed consent by submitting the completed survey prior to participating in this study.

After reading the PIS, the survey was made available to participants by the site via REDCap [39, 40] or on paper based on participant preference. Participants were invited to complete the survey between 2- and 24-weeks after first accessing the app and had one week to do so. The survey was translated from English into up to three local languages to enable participants to provide their responses in their preferred language. The survey was first translated by a professional translator. The translation was then reviewed and edited by a second translator, with any discrepancies or concerns discussed with the original translator. For all languages that have NAATI accreditation, the translations are NAATI accredited. NAATI is the national standards and certifying authority for translators and interpreters in Australia (https://www.naati.com.au/).

Self-report survey

The quantitative self-report survey was anonymous; however, for descriptive purposes, participants were asked to provide basic demographic information, including their age, gender, country of birth, language spoken at home, level of education, employment status, marital status, and number of children for whom they care. In relation to this study, participants were also asked to self-report how many times they had engaged with the Thrive by Five content during the week prior to completing the survey. No identifying information is collected in the Thrive by Five app; therefore, it is not possible to link app usage and self-report data.

Participants completed two standardized questionnaires, namely an adapted version of the Parenting and Family Adjustment Scales (A-PAFAS) [41] and the System Usability Scale (SUS) [42]. The original Parenting and Family Adjustment Scales is a 30-item validated measure of changes in parenting practices in response to parenting interventions [41]. This scale has been found to have good internal consistency and satisfactory construct and predictive validity [41] and has been used in multiple countries, including Afghanistan [43] and Indonesia [44]. It consists of two primary scales: Parenting and Family Adjustment. The 18-item Parenting scale is broken down into four subscales: parental consistency (5 items), coercive parenting (5 items), positive encouragement (3 items), and parent-child relationship (5 items). The parental consistency subscale enquires about the consistency with which the participant manages a child’s misbehaviour. Items on the coercive parenting subscale investigate a participant’s responses to the child when they misbehave, including shouting, getting angry, making them feel badly, spanking or smacking, arguing, or getting annoyed. The positive encouragement subscale measures the participants use of praise and rewards to encourage desired behaviours. Items on the parent-child relationship subscale enquire about behaviours relating to the strength of the bond between the parent and child, such as chatting or talking, giving the child hugs, kisses or cuddles, feeling proud of the child, enjoying spending time with the child, and having a good relationship with the child.

The 12-item Family Adjustment scale is comprised of three subscales: parental adjustment (5 items), family relationships (4 items), and parental teamwork (3 items). The parental teamwork subscale was not included in this study in an effort to avoid stigmatizing participants without partners. The parental adjustment subscale explores how well the participant is managing the emotional demands of being a caregiver, including feelings of stress, sadness, worry, and life satisfaction. Finally, the family relationships subscale examines how supported the participant feels by their family.

Items are customarily scored on a 4-point scale; however, for this study, the wording of the response options was adapted to inquire about self-reported changes in childrearing behaviours and to include a “neutral” response within the Likert scale (now 5-point scale) to allow an option for ‘no change’. As such, statements in the A-PAFAS were prefaced with ‘Since using Thrive by Five…’, and response options and associated scores included: Much less than I used to (0); Less than I used to (1); As much as I ever did (2); More than I used to (3); and Much more than I used to (4). For the Parenting scale and all subscales, items were summed. Higher scores indicate lower levels of parenting and family functioning skills. Using Cronbach’s alpha, the internal consistency of the coercive parenting (α = 0.784), positive encouragement (α = 0.784), and parent-child relationship (α = 0.961) subscales was found to be acceptable based on established standards [45]. The items comprising the parental consistency subscale were slightly less reliable (α = 0.695), which aligns with previous research [44, 46]. This will be reviewed in greater detail in the discussion.

The SUS is a standardized, 10-item 5-point Likert scale that is widely used to assess the perceived usability and acceptability of a wide range of products and systems, such as apps [42]. Previous research has highlighted the reliability and validity of the measure [47–49]. Additionally, the SUS has been translated into and found to be reliable in several languages, including Arabic [50], Bahasa Indonesia [51], and Malay [52]. The SUS was included in this study as app usability had the potential to impact engagement with the Thrive by Five content, thus influencing the degree of self-reported change in childrearing practices assessed by the A-PAFAS.

The SUS score is computed by converting the scale from 0 (poorest rating) to 4 (best rating) with adjustment for odd-numbered (subtract 1 from the raw score) and even-numbered (subtract the raw score from 5) questions. The adjusted score is summed and multiplied by 2.5 to get the standard SUS score. The final SUS score ranges from 0 to 100, with 0 being extremely poor usability and 100 being excellent usability. The internal consistency of the SUS was acceptable as measured by Cronbach’s alpha (α = 0.725).

Data analysis

IBM SPSS (Version 29) was used for all quantitative analyses. Across all analyses, the α level was < 0.05 unless otherwise specified. As informed consent was indicated by submitting a completed survey, there was no missing data. Descriptive statistics were used to summarize the sociodemographic characteristics, parenting practices, adjustment, and relationship variables, and system usability factors of the sample both as a group as well as by country. One-way analysis of variances (ANOVAs) and Bonferroni post-hoc comparisons were computed to examine differences in parenting based on country of residence, level of education, and employment status. Bivariate Pearson’s correlations were calculated to identify significant associations between demographics, parenting adjustment and family relationships, system usage and usability, and country of residence with parenting practices. As the parenting practices subscale of the A-PAFAS was not normally distributed for each level of the categorical variables, Mann Whitney U tests were used to investigate these potential associations. Those variables found to be significantly correlated with parenting practices were subsequently included in a regression model.

To determine the relative contributions of each variable to parenting practices, a multiple linear regression model was constructed. All assumptions of linear regression were met. All variables found to be significantly associated with parenting practices were entered into the model. Exploratory multiple linear regression models were constructed in the same fashion as described previously for each of the four subscales comprising the Parenting scale to examine the relative contribution of parental adjustment to each model.

Finally, an exploratory one-way multivariate analysis of variance (MANOVA) was computed to examine country-specific variation on the four subscales that comprise the Parenting scale of the A-PAFAS. Bonferroni post-hoc tests with a Bonferroni correction for multiple comparisons (α < 0.0125) were calculated. Pairwise comparisons were also examined, again corrected for multiple comparisons (α < 0.002).

Ethics considerations

This study has been approved by the University of Sydney Human Research Ethics Committee (HREC) (Project 2021/956). Where a country specific HREC exists (e.g., Indonesia, Kyrgyzstan, and Malaysia), a site-specific protocol and supporting documents were submitted for local ethics approval, prior to initiating the research at the identified site. The local ethics committees are the Research Ethics Committee at the University of Respati Indonesia, the Committee on Bioethics at the Global Research Institute (GlORI) Foundation, Kyrgyzstan, and the SEGi Research Ethics Committee, SEGi University Malaysia. There was not a local ethics committee operating in Afghanistan at the time of this research [53–55]. Additionally, there are no social science ethics standards or policies established in Uzbekistan at the national and institutional level [56]. Importantly, the Republic of Uzbekistan Law on Science and Scientific Activity 2019 states, “persons carrying out scientific activities are required to: not to harm human life and health, the environment; not to allow plagiarism, misappropriation of other people’s scientific developments, not to rely on false information; comply with intellectual property rights and scientific ethics standards” [57], all of which are considerations of the University of Sydney HREC approval. However, there are no guidelines with regards to seeking local ethics approval in Uzbekistan. In relation to the latter, in instances where the country did not have a governing ethics body (e.g., Afghanistan and Uzbekistan), a site-specific protocol and supporting documents were approved by the University of Sydney HREC. For each site, the Site Principal Investigator assisted in identifying and providing advice on the appropriate country specific HREC as well as with the preparation and submission of an ethics application as required. In accordance with ethics approval, PISs were available for and specific to each site.

Results

Participant sociodemographic, parenting, and system usability factors

A total of 642 participants (mean age = 33.2 years, range = 19–68; 79% female; 89.9% partnered) from Afghanistan (n = 111), Indonesia (n = 157), Kyrgyzstan (n = 118), Malaysia (n = 103), and Uzbekistan (n = 153) completed the evaluation survey. Summarized group-based demographics, parenting adjustment and family relationships variables, and system usage and usability factors are presented in Table 1. Notably, participants were well-educated, with the majority (56.1%) having completed tertiary education. Similarly, more than half of the participants (56.9%) were employed full- or part-time. Participants, on average, were providing care to between 2 and 3 children (mean = 2.6; SD = 1.6).

Table 1.

Sociodemographic, parenting adjustment and family relationships, and system usability factors and their associations with parenting practices

	Total sample (n = 642)	Association with A-PAFAS Parenting scale
Sociodemographics
Continuous Variables	Mean (SD)	r
Age, years	32.2 (7.2)	-0.069
Total number of children	2.6 (1.6)	0.04
Categorical Variables	n (%)	U
Gender: Female	508 (79.0)	26207.50**
Highest level of education:
Primary	16 (2.5)	4101
Secondary#	256 (39.9)	46,881
Tertiary^	361 (56.1)	47,821
Other!	9 (1.3)	2314
Current employment
Full or part-time	365 (56.9)	47,839
Home or carer duties	182 (28.2)	47,923
Other&	95 (14.8)	25,783
Marital Status: Partnered~	578 (89.9)	18,429
Adapted Parenting and Family Adjustment Scales	Mean (SD)	r
Parenting Adjustment	5.7 (3.7)	0.688**
Family Relationship	3.7 (3.5)	0.710**
System Use and Usability	Mean (SD)	r
App use, in the last week	4.4 (6.2)	0.221**
System Usability, total score	64.8 (12.6)	-0.258**
Country of Residence	n (%)	U
Afghanistan	111 (17.3)	20084.50**
Indonesia	157 (24.5)	14759.50**
Kyrgyzstan	118 (18.4)	26568.00**
Malaysia	153 (23.7)	20674.00**
Uzbekistan	103 (15.9)	26218.00**

#High School, Secondary and Diploma in the context of Afghanistan, Kyrgyzstan, and Uzbekistan

^Degree, Post-Graduate and Diploma in the context of Malaysia and Indonesia,

^!While given the option to provide additional information, no participants clarified their level of education after responding ‘Other’

^&Other employment = unemployed, unable to work, student

~Not partnered = divorced, separated, never married, widowed, other

^@The A-PAFAS Parenting scale is comprised of the Parental Consistency, Coercive Parenting, Positive Encouragement, and Parent-Child Relationships subscales

Our sample had a mean score of 22.6 (SD = 9.8, range 0 to 53) on the A-PAFAS Parenting scale. As noted previously, there are no established cut-off scores for this scale, but lower scores reflect a self-reported increase in the use of positive parenting practices. As shown in Table 1, Parenting had a significant positive association with parental adjustment, family relationships, frequency of app use, system usability, gender, and country of residence. In contrast, Parenting was not associated with participant age, number of children, marital status, level of education, or employment status.

Factors associated with parenting practices

A multiple linear regression equation examining significant predictors of Parenting was found to be significant (F(9,632) = 134.395, p = < 0.001) with a R² of 0.657. Parenting adjustment, family relationships, system usability, frequency of Thrive by Five use, parent gender, and country of residence were all entered in the model as they were significantly correlated with Parenting. As indicated in Table 2, the final model showed that parental adjustment and family relationships explained 5.9% and 4.9% of the unique variance in Parenting, respectively. Indeed, an additional 1-point increase (i.e., lower scores reflect a self-reported increase in positive parenting behaviours) on the family relationship or parental adjustment subscales was associated with 0.957 and 0.904 increase on the Parenting scale, respectively. Residence in Indonesia (3.2%) and Kyrgyzstan (1.0%) explained an additional 4.2% of the variance in the model. Finally, system usability (0.7%) and app usage (0.3%) each accounted for a small portion of the unique variance in the model.

Table 2.

Multiple linear regression model to evaluate predictors of parenting (n = 642)

				Correlations
	B	SE B	p	Partial	Part	Unique variance (%)
Family relationship	0.957	0.101	< 0.001	0.353	0.221	4.9
Parental adjustment	0.904	0.087	< 0.001	0.382	0.242	5.9
Female	-0.607	0.760	0.425
App use, in the last week	0.091	0.042	0.031	0.085	0.050	0.3
System Usability, total score	-0.076	0.020	< 0.001	-0.152	-0.090	0.7
Indonesia	-6.086	0.780	< 0.001	-0.296	-0.182	3.2
Uzbekistan	-1.302	0.790	0.100
Afghanistan	1.434	0.951	0.132
Kyrgyzstan	-3.444	0.786	< 0.001	-0.172	-0.102	1.0

Exploratory multiple linear regression models were also constructed to investigate significant contributors to each of the subscales comprising the Parenting scale. All four models were found to be significant, including for parental consistency (F(9,632) = 13.313, p < 0.001) with a R² of 0.159; coercive parenting (F(9,632) = 58.350, p < 0.001) with a R² of 0.454, positive encouragement (F(9,632) = 37.581, p < 0.001) with a R² of 0.349, and parent-child relationship (F(9,632) = 92.800, p < 0.001) with a R² of 0.569. The full details of the final models are provided in Additional File 2. In summary, parental adjustment was found to be a significant predictor of scores on the positive encouragement and parent-children relationship subscales, explaining 4.9% and 10.2% of the unique variance in each model, respectively.

Differences in parenting based on demographics and country of residences

One-way ANOVAs did not find significant differences in Parenting scores based on level of education (F(3,638) = 1.069, p = 0.362) or employment status (F(2,639) = 0.572, p = 0.565); however, Parenting differed significantly by country of residence (F(4,637) = 52.221, p < 0.001). Bonferroni post-hoc comparisons indicated that scores were significantly lower (i.e., reflecting a self-reported increase in the use of positive parenting practices) in Indonesia compared to all other countries. Participants from Kyrgyzstan reported significantly higher scores on the Parenting scale relative to Indonesian participants, but significantly lower scores compared to participants from Afghanistan, Malaysia, and Uzbekistan. Parenting scores did not significantly differ between Afghanistan, Uzbekistan, and Malaysia. Further details about the country-specific demographics, parenting practices, adjustment and relationship variables, and system usage and usability factors are presented in Table 3.

Table 3.

Sociodemographic, parenting behaviours, adjustment and relationships, and system usability factors by country of residence

	Total (n = 642)	Afghanistan (n = 111)	Indonesia (n = 157)	Kyrgyzstan (n = 118)	Malaysia (n = 103)	Uzbekistan (n = 153)
Sociodemographics
Continuous Variables	Mean (SD)
Age, years	32.2 (7.2)	30.4 (5.6)	33.3 (7.1)	30.1 (5.8)	35.4 (4.7)	32.1 (9.7)
Total number of children	2.6 (1.6)	2.9 (1.8)	2.1 (1.0)	2.6 (1.6)	2.7 (1.5)	2.4 (1.6)
Categorical Variables	n (%)
Gender: Female	508 (79.0)	23 (20.6)	147 (93.5)	112 (94.8)	92 (89.2)	134 (87.6)
Highest level of education:
Primary	16 (2.5)	2 (1.7)	4 (2.4)	3 (2.4)	5 (4.9)	2 (1.2)
Secondary#	256 (39.9)	16 (14.3)	60 (38.1)	35 (29.7)	29 (28.2)	116 (75.7)
Tertiary^	361 (56.1)	93 (83.8)	92 (58.6)	75 (63.6)	69 (67.0)	32 (20.8)
Other!	9 (1.3)	0 (0.0)	1 (0.01)	5 (4.1)	0 (0.0)	3 (2.0)
Current employment:
Full or part-time	365 (56.9)	96 (86.5)	88 (56.1)	45 (38.0)	76 (73.8)	60 (39.1)
Home or carer duties	182 (28.2)	12 (10.7)	34 (21.7)	58 (49.2)	23 (22.2)	55 (35.8)
Other&	95 (14.8)	3 (2.6)	35 (22.3)	15 (12.6)	4 (3.9)	38 (24.7)
Marital Status: Partnered~	578 (89.9)	106 (95.5)	141 (89.7)	113 (95.8)	98 (95.0)	120 (78.3)
Adapted Parenting and Family Adjustment Scales	Mean (SD)
Parenting Adjustment	5.7 (3.7)	6.5 (4.5)	4.3 (3.1)	5.3 (3.8)	6.3 (3.6)	6.5 (3.4)
Family Relationship	3.7 (3.5)	4.0 (4.0)	1.4 (2.5)	3.9 (3.3)	4.7 (3.1)	5.2 (3.0)
Parenting@	22.6 (9.8)	27.3 (9.5)	14.9 (7.0)	20.5 (7.3)	26.1 (9.8)	25.9 (8.9)
Parental Consistency	8.9 (2.7)	10.3 (2.1)	8.3 (2.6)	9.4 (2.5)	7.5 (2.7)	9.7 (2.4)
Coercive Parenting	5.9 (4.3)	9.3 (3.7)	3.1 (3.7)	4.3 (3.2)	58 (3.2)	8.3 (4.5)
Positive Encouragement	3.2 (2.6)	2.9 (3.0)	2.2 (2.0)	3.6 (2.1)	4.5 (2.5)	3.7 (2.4)
Parent-Child Relationships	4.0 (4.9)	4.6 (5.8)	1.5 (2.8)	3.2 (3.7)	8.3 (5.9)	4.3 (3.4)
System Use and Usability	Mean (SD)
App use, in the last week	4.4 (6.2)	4.0 (3.4)	1.7 (1.6)	2.6 (3.6)	3.2 (3.8)	9.9 (9.7)
System Usability, total score	64.8 (12.6)	66.1 (12.4)	63.4 (10.4)	67.8 (12.7)	64.8 (11.1)	63.4 (14.8)

#High School, Secondary and Diploma in the context of Afghanistan, Kyrgyzstan, and Uzbekistan

^Degree, Post-Graduate and Diploma in the context of Malaysia and Indonesia

^!While given the option to provide additional information, no participants clarified their level of education after responding ‘Other’

^&Other employment = unemployed, unable to work, student

~Not partnered = divorced, separated, never married, widowed, other

^@The A-PAFAS Parenting scale is comprised of the Parental Consistency, Coercive Parenting, Positive Encouragement, and Parent-Child Relationships subscales

Differences in parenting subscales based on country of residence

A one-way MANOVA indicated that the four A-PAFAS subscales comprising the Parenting scale differed significantly by country of residence (F(16, 1937) = 29.617, p < 0.001; Wilk’s Λ = 0.513, partial η² = 0.154). Specifically, at α < 0.0125, country of residence had a significant effect on parental consistency (F(4, 637) = 25.095; p < 0.001; partial η² = 0.136), coercive parenting (F(4, 637) = 65.103; p < 0.001; partial η² = 0.290), positive encouragement (F(4, 637) = 16.204; p < 0.001; partial η² = 0.092), and the parent-child relationship (F(4, 637) = 39.854; p < 0.001; partial η² = 0.200).

Pairwise comparisons by country are presented in Fig. 1. Regarding parental consistency, participants from Malaysia did not differ from those in Indonesia, but did report greater consistency in parenting practices (i.e., lower scores) relative to those from Afghanistan, Kyrgyzstan, and Uzbekistan. Indonesian participants also scored lower relative to respondents from Afghanistan and Uzbekistan, but not Kyrgyzstan. There were no significant differences observed in parental consistency between Afghanistan, Kyrgyzstan, and Uzbekistan.

Fig. 1.

Differences in parenting practices by country of residence

Figure 1 presents the estimated marginal means for each of the four subscales that comprise the Parenting subscale of the A-PAFAS based on country of residence.

Coercive parenting was reported to be significantly lower in Indonesia relative to Afghanistan, Malaysia, and Uzbekistan. While there was no difference found between scores among Indonesian and Kyrgyzstani participants, the latter group reported using significantly fewer coercive parenting practices relative to respondents from Afghanistan, Malaysia and Uzbekistan. Similarly, Malaysian participants reported significantly lower levels of coercive parenting compared to those from Afghanistan and Uzbekistan and no differences were found between the latter two groups.

As presented in Fig. 1, Indonesian participants reported significantly lower (i.e., better) scores on the positive encouragement subscale relative to Kyrgyzstan, Malaysia, and Uzbekistan. Participants from Afghanistan reported similar levels of positive encouragement relative to those from Indonesia, Kyrgyzstan and Uzbekistan, but significantly lower relative to respondents from Malaysia. No differences were found between Indonesia and Afghanistan or between Afghanistan, Kyrgyzstan, and Uzbekistan.

Malaysian participants reported significantly worse scores (i.e., higher) on the parent-child relationship subscale relative to respondents from all other countries. Indonesian participants scored significantly lower relative to respondents from Afghanistan, Malaysia, and Uzbekistan. No differences were observed between participants from Indonesia and Kyrgyzstan or from Afghanistan, Kyrgyzstan, and Uzbekistan.

Discussion

Our findings contribute to the evidence base demonstrating the relationship between parental adjustment and parenting practices. More specifically, across the five LMICs represented in this study, we found that better parental adjustment was associated with a self-reported increase in the use of positive forms of encouragement and engagement as well as behaviours to support stronger parent-child relationships. A previous meta-analysis found that positive psychological wellbeing amongst parents generally (i.e., not in relation to adapting to the parenting role), was associated with the use of more positive parenting practices [58]. However, ours is the first study to showcase the association specifically between better parental adjustment and greater use of positive parenting practices, such as praise, attention, affection, displays of pride, and increased time spent with children. This outcome underscores the importance of considering parental adjustment problems as a target for intervention, recognising that subclinical and/or transient symptoms of mental ill-health can influence parenting practices.

Parental adjustment was not found to be a significant predictor of the use of coercive parenting practices, such as spanking, shouting, and getting angry at children. These results contrast with previous studies in both HICs and LMICs that have emphasized the association between parental adjustment problems and poorer psychological health, and lower levels of parental involvement in children’s activities and the use of more negative parenting practices [58].

Family relationships were also associated with parenting practices in this study which is consistent with findings from earlier research [59]. Indeed, familial support has been shown to improve parental mental health and wellbeing [60]. To that end, the United Nation Children’s Fund ‘Caring for the Caregiver’ package emphasizes a relationship-centred approach to caregiving systems, promoting enhanced support for caregivers by partners and broader families [61]. In accordance with family systems theory, these findings emphasise the interconnectedness of individuals and relationships within families, including how these can influence parenting practices and, in turn, childhood outcomes [62].

Parental adjustment was not found to be significantly associated with the parental consistency subscale of the A-PAFAS Parenting scale. Notably, previous studies of the original PAFAS have identified concerns regarding the reliability and validity of this subscale for LMICs. For example, the results of confirmatory factor analyses conducted as part of validation studies in Indonesia and China found that parental consistency had low factor loadings. In each study, removal of two items (Item 3: “I follow through with a consequence when my child misbehaves”; Item 11: “I deal with my child’s misbehaviour the same way all the time”) improved the accuracy of the fit of the four-factor model. Additionally, the subscale was found to have poor internal reliability in both contexts [44, 46]. In this study, internal reliability was found to be slightly below acceptable limits (α = 0.695). Interviews with parents participating in the Chinese study referenced previously indicated that flexibility in parenting was valued over consistency, which may have influenced understanding of and responses to items on the parental consistency subscale [46]. Taken together, these findings suggest that the principles of parental consistency are not universally considered valuable or desirable across contexts. As such, any relationship between parental adjustment and parental consistency practices is likely to be significantly influenced by sociocultural and contextual factors. Indeed, our results showed that parental consistency differed significantly by country of residence; however, this finding needs to be interpreted cautiously.

System usability, as measured by the SUS, and self-reported app usage were also found to have a significant relationship with parenting practices. The International Standards Organization identified three key components necessary to achieve good usability, namely effectiveness, satisfaction, and efficiency [63]. In relation to the latter, it is noted that if users require a considerable amount of time to engage with or complete tasks in a system, usability is judged to be low. In this regard, it is important to consider the concept of ‘learnability’ as a component of usability [49]. Consistent with previous literature [49], our team’s results (unpublished) from the broader Thrive by Five evaluation referenced previously demonstrate that the SUS is multidimensional, comprised of usability and learnability subscales. Questions on the SUS contributing to the learnability subscale include: “I think I would likely need the support of a technical person to be able to use the system” (Item 4) and “I needed to learn a lot of things before I could get going with the system” (Item 10). Our findings suggest that parents and caregivers who require more time to acquire new information report lower scores on the learnability items, thus deflating the overall SUS score.

In the context of digital parenting programs such as Thrive by Five, poor learnability may indicate a wider gap between current parenting practices and those that are recommended within the program. Furthermore, learnability may be particularly important to consider in LMICs where users may require additional time to personally adapt content to their own culture, context, circumstances, and needs [64]. In the context of this study, lower usability and learnability scores could have interfered with parents’ ability to benefit from the Thrive by Five content, which may have been reflected by no or fewer self-reported changes in parenting practices as assessed by the A-PAFAS.

Parent gender was not found to be a significant predictor of parenting practices in this study. As the majority of participants (79%) were women, it may be that there was not sufficient power to detect a gender-based difference. Previous research conducted in a small Midwestern city in the United States showed that parental adjustment problems may manifest differently for mothers and fathers. For example, fathers who worried about their ability in the parenting role tended to have more difficulties in their parent-child relationships; however, similar worries were not associated with the same behaviours amongst mothers [65]. Overall, however, there is little investigation of the relationship between gender, emotional adjustment, and parenting practices. Importantly, it is recommended that future studies extend beyond mothers to build understanding of the effects of mental health and adjustment problems within the broader family system, including fathers and other caregivers (e.g., grandparents). This may be particularly relevant in multi-generational families common to LMICs in which childrearing responsibilities are shared more broadly, extending beyond the nuclear family.

Finally, our study highlights differences in parental adjustment and the use of positive and negative parenting practices across five culturally diverse LMICs. Overall, participants from Indonesia reported using more positive and fewer negative parenting practices relative to respondents from other countries. Little variability was observed with regards to the use of positive encouragement and behaviours to strengthen bonds between parent and child apart from Malaysian participants who tended to engage in these practices significantly less frequently than other respondents. Coercive parenting practices were significantly more common amongst participants from Afghanistan, Malaysia, and Uzbekistan relative to Indonesia and Kyrgyzstan. Finally, participants from Indonesia and Malaysia reported more consistency in their parenting practices compared to participants from other countries; however, it is important to consider this data in the context of the concerns raised earlier about the reliability and validity of the subscale.

Culture is one factor that could contribute to differences in parenting practices across countries of residence. Indeed, conceptualisations of and beliefs about early childhood development and parenting vary markedly across cultures and geographic areas [66, 67]. Whilst there are culture-common influences on parenting (e.g., the need to keep young children safe), culture-specific factors (e.g., social competencies and values) also shape how children are raised [68]. Parents in different cultures may use the same childrearing practice for the same purpose (e.g., child-directed speech to support language development) [69]. In other instances, the same parenting behaviour may be perceived differently across cultures (e.g., harsh physical treatment can be used in initiation rites or viewed as abusive) [68].

It is critical to recognize, however, that neither culture nor parenting practices are static. Indeed, they are both dynamically impacted by personal experiences, economic and social status, personal goals and priorities, family-level life stressors, and varying levels of social support [70–72]. Furthermore, there are a myriad of contextual factors that could influence parenting practices within and across cultures. For example, differing levels of access to and engagement with social media or digital health interventions, such as Thrive by Five, are likely to differentially shape how parents interact with and care for their children. In light of the above, we caution against overinterpreting the survey data presented in this study at the risk of presenting a biased narrative. The country-specific samples are not nationally representative, and the survey data does not contain sufficient demographic, social, contextual, or cultural details to enable a thorough investigation of common or contrasting parenting practices within and between cultural groups.

Practical implications

There has been a call for interventions to promote parental mental health in LMICs, including addressing context specific modifiable risk factors with the aim of optimizing early childhood development [14, 73]. Indeed, it is recognised in the global early childhood literature that enhanced support for parents along the mental health continuum will increase the ability of parents and families to protect their children from sociocultural and contextual barriers to development [4]. Furthermore, addressing parents’ emotional adjustment problems may, in turn, allow them to more fully engage in and benefit from broader interventions promoting nurturing and responsive childrearing practices [74]. Taken together, these findings suggest that content to support parents’ emotional wellbeing as they adjust to the parenting role may be a vital component of parenting programs with the aim of optimizing early childhood development outcomes.

Notably, a meta-analysis of multi-component interventions targeting both parenting practices and parental mental health broadly did not report significant reductions in depressive symptoms amongst caregivers [31]. The authors acknowledge that the meta-analysis may have been underpowered given the small number of studies included. Additionally, they note that the content of the multi-component interventions emphasised strategies for supporting children’s health and development, with only one or two sessions allocated to parental wellbeing which may be insufficient to achieve the desired effect [31]. As such, research is now required to explore how best to integrate and combine content related to emotional adjustment with other key intervention material (e.g., training in responsive caregiving), including with regards to the optimal ‘dose’ of each component. As part of this, it may be more effective to take a transdiagnostic approach (i.e., targeting symptoms that are common to but not diagnostic of multiple disorders) to parental mental health, rather than focus on specific diagnoses such as depression [74]. Indeed, this approach is likely to be better suited to understanding the impacts of emotional adjustment problems on parenting and, in turn, childhood outcomes. It will also be important to explore how to adapt interventions for diverse cultural contexts, and, in turn, optimally deliver such programs to parents with varying needs and circumstances [2].

Limitations

This study has some limitations that are important to note. The county-specific samples are not nationally representative and, therefore, caution needs to be used when interpreting the results as they may not be generalizable to the general population. While the primary measures used in this study, namely the PAFAS and SUS, have both been validated in cross-cultural contexts, the specific translations used in this study were not independently validated due to time constraints; however, NAATI certified translators were used whenever possible and all surveys underwent back translation, the gold standard for foreign language translation. Additionally, all survey translations were reviewed and approved by representatives from each research site.

The data used in this study are cross-sectional prohibiting conclusions about causality in relation to parental adjustment and parenting practices. It is important to note, however, that the questions in the A-PAFAS are worded to capture self-reported changes in parenting practices (i.e., Since using Thrive by Five…) as a result of engaging with the Thrive by Five childrearing content. Finally, this study did not include measures of child behaviour and physical and mental health, which are known to impact parental mental health and emotional adjustment [75].

Conclusions

Importantly, not all parents who struggle to adjust emotionally to the parenting role engage in negative parenting practices, particularly when familial and social supports are in place. Nevertheless, to optimize the development of children in their early years, it is imperative to support the caregivers who are caring for them [76]. Our results highlight that parental emotional adjustment influences the use of positive parenting practices in diverse LMICs. This finding suggests that interventions that support a parent’s ability to adjust to and cope with the parenting role could increase the use of responsive and nurturing childrearing practices. However, further research is now required to explore how best to deliver content to support parental adjustment as the potential benefits of embedding this material in parenting programs have not been realised to date.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

ANOVA

Analysis of variance

A-PAFAS

Adapted parenting and family adjustment scales

HIC

High income country

LMIC

Low- and middle-income country

MANOVA

Multivariate analysis of variance

NAATI

National accreditation authority for translators and interpreters

PAFAS

Parenting and family adjustment scales

PIS

Participant information sheet

SUS

System usability scale

UNICEF

United Nation’s children’s fund

Author contributions

Authors HL, YS, and IH were integral in securing funding to support the study. The impact evaluation study was designed by HL, YS, VL, and ME, with additional insights and support from QK. Scientific oversight and guidance were provided by IH to ensure all activities were conducted responsibly and in a culturally appropriate manner. LOP provided oversight and guidance with regards to the interpretation of the results in relation to the LMIC context. HL was responsible for all data analyses, with consultation provided by QK and GH. HL drafted the original manuscript, and all authors contributed to and have approved the final manuscript.

Funding

This research was conducted by the University of Sydney’s Brain and Mind Centre pursuant to an agreement between the University and Minderoo Foundation Limited (Minderoo). IBH is supported by a NHMRC L3 Investigator Grant (GNT2016346).

Data availability

The datasets analysed during the current study are not publicly available to protect the privacy and confidentiality of the participants but can be made available from the corresponding author on reasonable request and with appropriate ethics approval.

Declarations

Ethical considerations and consent to participate

This study has been approved by the University of Sydney Human Research Ethics Committee (HREC) (Project 2021/956). Where a country specific HREC exists (e.g., Indonesia, Kyrgyzstan, and Malaysia), a site-specific protocol and supporting documents were submitted for local ethics approval, prior to initiating the research at the identified site. The local ethics committees are the Research Ethics Committee at the University of Respati Indonesia, the Committee on Bioethics at the Global Research Institute (GlORI) Foundation, Kyrgyzstan, and the SEGi Research Ethics Committee, SEGi University Malaysia. There was not a local ethics committee operating in Afghanistan at the time of this research [53–55]. Additionally, there are no social science ethics standards or policies established in Uzbekistan at the national and institutional level [56]. Importantly, the Republic of Uzbekistan Law on Science and Scientific Activity 2019 states, “persons carrying out scientific activities are required to: not to harm human life and health, the environment; not to allow plagiarism, misappropriation of other people’s scientific developments, not to rely on false information; comply with intellectual property rights and scientific ethics standards” [57], all of which are considerations of the University of Sydney HREC approval. However, there are no guidelines with regards to seeking local ethics approval in Uzbekistan. In relation to the latter, in instances where the country did not have a governing ethics body (e.g., Afghanistan and Uzbekistan), a site-specific protocol and supporting documents were approved by the University of Sydney HREC. For each site, the Site Principal Investigator assisted in identifying and providing advice on the appropriate country specific HREC as well as with the preparation and submission of an ethics application as required. In accordance with ethics approval, PISs were available for and specific to each site. All participants provided informed consent prior to participating in the research.

Consent for publication

Not applicable.

Competing interests

Professor Ian Hickie is the Co-Director, Health and Policy at the Brain and Mind Centre (BMC) University of Sydney. The BMC operates an early-intervention youth service at Camperdown under contract to headspace. He is the Chief Scientific Advisor to, and a 3.2% equity shareholder in, InnoWell Pty Ltd which aims to transform mental health services through the use of innovative technologies. Victoria Loblay is a board member for Matana Foundation, a philanthropic organization that provides funding to programs for disadvantaged young people in Australia. She does not receive any financial benefit for this role.The other authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.