The role of mother-adolescent relationship quality in moderating the effect of adolescent anxiety on psychosocial functioning

ABSTRACT

Objective

The effects of anxiety on adolescents’ psychosocial outcomes are well established, but little consideration has been given to the potential influence of the parent-adolescent relationship in moderating these effects. This study examined the moderating role of parent-adolescent connectedness and hostility in the association between anxiety and adolescent psychosocial functioning (measured by positive development [PD] and oppositional defiant behaviour [ODB]) within a community sample of mothers of adolescents.

Method

Participants were 723 Australian mothers (M age = 44.05 years, SD = 5.97) of adolescents aged 11 to 17 years (M = 14.32 years, SD = 5.97; 49% male). Participants completed an online survey comprising measures of parent-adolescent relationships, parenting practices, parental psychological distress, and adolescent anxiety and psychosocial functioning.

Results

Consistent with the first hypothesis, results from hierarchical regression analyses revealed that adolescent anxiety, connectedness, and hostility were independent predictors of PD and ODB. Inconsistent with predictions, parent-reported anxiety had a stronger, negative association with PD when mothers viewed the relationship with their adolescents as more connected and less hostile. Neither parent-adolescent connectedness nor hostility moderated the association between maternal reported adolescent anxiety and ODB.

Conclusions

Further longitudinal research is needed to understand how the parent-adolescent relationship context might affect outcomes and inform family-based prevention and intervention efforts for at-risk youth with anxiety symptomatology.

KEY POINTS

What is already known about this topic:

1. Anxiety disorders are the most common mental health problems affecting adolescents.

2. Anxiety can have wide-reaching implications for adolescents, affecting their behavioural functioning and the development of positive developmental skills and competencies.

3. The quality of the mother-adolescent relationship plays a central role in influencing psychosocial outcomes for adolescents generally, but little is known how this important relationship influences psychosocial outcomes when an adolescent is experiencing anxiety.

What this topic adds:

1. This research extended previous research that focused on young people with diagnosed anxiety disorders by showing that even among a community sample, anxiety is associated with negative outcomes for adolescents’ behavioural functioning and positive development.

2. The study also emphasised the importance of considering adolescent anxiety and its effects on a young person’s psychosocial functioning within the context of the mother-adolescent relationship.

3. Specifically, the study provided additional evidence for the strong influence of parent-adolescent hostility on adolescent oppositional behaviour, while highlighting that interactions between parents and adolescents when responding to and coping with anxiety play a complex role in adolescent positive development.

KEYWORDS:

• Parent-adolescent relationship
• parenting
• anxiety
• adolescent psychosocial functioning
• youth
• positive development

Introduction

Anxiety disorders are the most common mental health problems affecting children and adolescents, with up to 20% of young people in Western countries estimated to meet diagnostic criteria for an anxiety disorder (Koydemir & Essau, 2018). The symptoms of anxiety are associated with impairment in a young person’s everyday functioning (Hammerness et al., 2007; Kendall et al., 2010), resulting in negative consequences for academic and psychosocial functioning across adolescence and into adulthood (de Lijster et al., 2018; Morin et al., 2011). Developmentally, anxiety can interfere with the development of the positive competencies and strengths needed for adolescents to successfully transition into adulthood (Raudino et al., 2013). One recent systematic review indicated that anxious adolescents report poorer positive developmental outcomes compared to their non-anxious peers, including lower social competence and assertiveness, more interpersonal problems and school refusal and worse school performance (de Lijster et al., 2018). Adolescent anxiety can also affect behavioural functioning as it can co-occur with oppositional defiant behaviour and conduct behaviour problems (Kendall et al., 2010). Adolescent anxiety symptoms, such as avoidance of feared situations, may manifest as defiant or oppositional behaviour at home and school, and anxious adolescents with comorbid oppositional defiant behaviour have higher levels of negative emotionality, reactive aggression and emotional dysregulation, which in turn exacerbates displays of oppositional defiant behaviour (Bubier & Drabick, 2009; Drabick et al., 2010).

The parent-adolescent relationship is central to an adolescent’s psychosocial functioning broadly (Laursen & Collins, 2009), and in the development and maintenance of anxiety disorders. Patterns of interactions between parents and children characterised by overinvolvement, control, intrusiveness, rejection, and hostility have been identified as key contributors to the development and maintenance of anxiety in children and adolescents (Drake & Ginsburg, 2012; Emerson et al., 2019). Studies that examined the role of mothers have found that adolescent perceptions of the quality of their relationship with their mothers is associated with anxiety symptoms in adolescence (van Eijck et al., 2012), and higher levels of maternal anxious parenting (i.e., over-protective behaviour and overt expressions of worry) perceived by adolescents significantly predicts the maintenance of adolescent anxiety symptoms over time (Rapee, 2009). The parent-adolescent relationship is also known to be central to the behavioural functioning and positive development of typically developing adolescents (Laursen & Collins, 2009). Yet, the extent to which the quality of the parent-adolescent relationship impacts the level of oppositional behaviour and positive development among adolescents experiencing anxiety has received little attention. Thus, research is needed to better understand the associations between the parent-adolescent relationship, adolescent anxiety, and psychosocial outcomes for adolescents, and how dimensions of the parent-adolescent relationship interact with adolescent anxiety in influencing psychosocial outcomes for adolescents.

The influence of the parent-adolescent relationship on adolescent psychosocial functioning

Theory and research on the parent-adolescent relationship (e.g., Barber et al., 2005; Rohner, 2016) centres on two key dimensions that predict and explain the influence of the parent-adolescent relationship on development and wellbeing: connectedness and hostility. Connectedness refers to the emotional connection between parents and children, and the extent to which children feel loved, cared for, supported, and accepted by their parents (Boutelle et al., 2009; K. Burke et al., 2021). Parent-adolescent connectedness has been found to play a protective role in adolescent mental health, with cross-sectional (Foster et al., 2017) and longitudinal (Boutelle et al., 2009; Parker & Benson, 2004) research finding that connectedness is related to lower levels of adolescent depression, suicidality, and oppositional behaviour and conduct problems. In addition, longitudinal research indicates that parent-adolescent connectedness facilitates positive outcomes for young people, including greater self-worth and self-esteem (Boutelle et al., 2009; McAdams et al., 2017; Parker & Benson, 2004), prosocial behaviour (van der Storm et al., 2022) and self-control (Li et al., 2019). Connected parent-adolescent relationships also support young people as they transition into emerging adulthood. In an Australian longitudinal study of 1,158 young people, the quality of adolescents’ relationship with parents measured in late childhood and adolescence significantly predicted greater positive development in emerging adulthood (O’Connor et al., 2020). Importantly, in meta-analyses that have examined the independent influence of mother versus father relationships, this body of evidence suggests that connectedness with mothers and fathers has a positive influence on adolescent outcomes (Li et al., 2019; van der Storm et al., 2022)

In comparison, hostility refers to the extent to which the verbal and physical interactions and emotional tone in the parent-adolescent relationship is characterised by negativity, criticism, and rejection (K. Burke et al., 2021). Both mother-adolescent and father-adolescent hostility has a significant impact on a young person’s psychosocial functioning and has found to be associated with oppositional behaviour and conduct problems, poor peer relationships and poor psychological adjustment (Khaleque, 2017; Pinquart, 2017; Weymouth et al., 2016). Canadian cross-sectional research with over 2,000 young adolescents found that adolescents who report feeling rejected by their parents are more likely to engage in suicidal ideation and suicide attempts compared to those who are rejected in peer relationships (Fotti et al., 2006), while Irish cross-sectional research indicated that adolescent-reported maternal and paternal criticism and rejection are associated with higher levels of emotional distress (Kenny et al., 2013). These cross-sectional findings are supported by longitudinal research, which has found that when an adolescent feels unloved and rejected by their family, their poor psychological adjustment (e.g., conduct problems, depression, anxiety, suicidal behaviours, and illicit drug use) is significantly more likely to continue into adulthood (Moffitt et al., 2002; Raudino et al., 2013). Consistent with the meta-analytic research on parent-adolescent connectedness, both mother-adolescent and father-adolescent hostility has a negative impact on adolescent psychosocial functioning (Khaleque, 2017; Pinquart, 2017; Weymouth et al., 2016).

The influence of the parent-adolescent relationship in the context of adolescent anxiety

Overall, there is a large body of research demonstrating the central role of the parent-adolescent relationship among typically developing adolescents. However, there is little research on the role of this relationship in the context of adolescent anxiety. Given the association between parent-adolescent connectedness and hostility and adolescent positive development and oppositional behavioural problems, these factors are likely to have a similar influence on the adjustment of adolescents experiencing elevated anxiety symptoms.

Several studies have provided insight into the complex associations among adolescent anxiety, the parent-adolescent relationship, and adolescent outcomes. Research on treatment responsiveness indicates that the parent-adolescent relationship plays an important role in outcomes for young people with anxiety. Two studies examining family-related moderators of outcomes following Cognitive Behaviour Therapy (CBT) found that lower family cohesion (Victor et al., 2006) and greater family dysfunction and parental hostility towards their child (Crawford & Manassis, 2001) were associated with a less favourable treatment response. Two other studies involving community samples of adolescents also provide support for complex associations among adolescent anxiety, the parent-adolescent relationship, and adolescent outcomes. In a U.S. longitudinal study following young people from age 12–13 years until they were 19–20 years, Rueter et al. (1999) found that, among young adolescents showing elevated, yet subthreshold, anxiety symptoms, frequent and persistent parent-adolescent conflict increased the risk of having anxiety disorder by late adolescence. A more recent study found that levels of anxiety among typically developing older adolescents was a moderator between mother-adolescent communication and adolescent coping, with open mother-adolescent communication positively predicting growth in coping skills only among adolescents with low anxiety (Simpson et al., 2020).

In summary, the research suggests that the parent-adolescent relationship influences psychosocial outcomes for community adolescents, and for those experiencing subclinical and clinical levels of anxiety. Based on the research, it is argued that parent-adolescent hostility and connectedness has a direct influence on adolescent psychosocial adjustment, as well as an influence on the relationship between anxiety and psychosocial functioning. The theorized moderation model suggests that parent-adolescent relationship quality alters the strength of the relationship between adolescent anxiety and adolescent psychosocial functioning, in that the quality of the relationship with their parents may potentially change the trajectory of both positive and negative outcomes for anxious adolescents. For these adolescents, it may be that the parent-adolescent relationship becomes even more important than for typically developing adolescents, such that high connectedness and low hostility has a stronger influence on positive development and the severity of oppositional defiant behaviour.

The present study

This study aimed to a) re-examine the influence of the parent-adolescent relationship on adolescent psychosocial functioning and b) extend understanding of the nature of these associations to explore how anxiety might influence adolescent psychosocial functioning in the context of either connected or hostile parent-adolescent relationships. The present study employed an online survey of community-residing mothers to evaluate the role of parent-adolescent relationships and adolescent anxiety in the positive development (PD) and oppositional defiant behaviour (ODB) of adolescents aged 11 to 17 years. While both female and male parents and caregivers were invited to take part in this study, only female parents, the vast majority of whom were mothers, completed the survey. Thus, given recent meta-analyses indicate that mothers’ and fathers’ relationship behaviour is similarly related to adolescent psychosocial adjustment (e.g., van der Storm et al., 2022; Weymouth et al., 2016), mother reports of the parent-adolescent relationship were used in the current study.

Firstly, the study aimed to investigate the independent influence of mother-adolescent connectedness and hostility on PD and ODB. It was hypothesised that (1) maternal reported connectedness and hostility would uniquely predict PD and ODB after controlling for sociodemographic factors and other parenting constructs known to play a central role in adolescent outcomes (i.e., parental psychological distress, parenting practices). Parental psychological distress was considered an important covariate in this context, given consistent findings that parental mental health problems can negatively affect a parent’s capacity to respond sensitively and appropriately in interactions with their child (Drake & Ginsburg, 2012; England, 2009). In addition, higher levels of parental anxiety have been found to be associated with more negative perceptions of the family environment (Hughes et al., 2008).

Secondly, the study aimed to examine the moderating role of connectedness and hostility in the mother-adolescent relationship on the association between adolescent anxiety and the outcomes of PD and ODB in separate moderation analyses. It was predicted that (2) maternal reported adolescent anxiety would be more weakly associated with PD at higher levels of connectedness and lower levels of hostility (i.e., more connected and less hostile relationships will buffer the negative influence of anxiety on positive developmental outcomes), and (3) maternal reported adolescent anxiety would be more strongly associated with ODB at lower levels of connectedness and higher levels of hostility (i.e., anxious adolescents with less connected and more hostile relationships with their mothers are likely to display more oppositional behaviour).

Method

Participants and recruitment

Participants were 723 female parents of adolescents aged 11 to 17 years (M = 14.32 years, SD = 5.97). Participants were recruited via a community outreach approach in the Australian states of New South Wales, Victoria, and Queensland. This involved paid (50.2%) and organic posts (26.7%) on social media advertising, as well as advertising in school newsletters and through community organizations. Participants were recruited during an 8-month period from May to December 2018, and the majority of respondents were female.

While recruitment was targeted at all parents of adolescents, most participants that responded to the survey were female caregivers (biological mothers N = 705; other female caregivers N = 18). Mothers ranged in age from 28 to 69 years (M = 44.05 years, SD = 5.97; see Table 1). Most mothers reported that they were born in Australia (83.1%) and were from an Australian cultural background (75.8%), but there was also representation from European, New Zealand and Indigenous Australian backgrounds. A third of the participants reported a vocational certificate or apprenticeship as their highest level of education, with approximately half having an undergraduate or postgraduate university degree. Half of the participants reported their adolescent lived in a two-parent household, with a quarter in single-parent households, and the remainder in step- or blended families. In terms of financial circumstances, most participants reported being able to meet essential expenses in the past 12 months (62.7%), had enough money to afford the food they wanted (43.3%), and did not worry about food running out (63.6%).

Table 1. Descriptive statistics and bivariate correlations with outcome variables (N = 723).

Sociodemographic Variable	M	SD	Correlations (Pearson’s r)
			AFS PD	AFS ODB
Parent’s age (years)	44.05	5.97	.02	−.12**
Adolescent’s reported age (years)	14.32	1.81	.02	−.21**
Financial stress	.00	.85	−.14**	.19**
	N	%	AFS PD	AFS ODB
Adolescent gender			.20**	−.07
Male	355	49.1
Female	366	50.6
Did not answer	2	0.3
Adolescent SCARED score in the clinical range (≥25)	257	35.5
Parent country of birth			.07	−.05
Australia	601	83.1
Other	119	16.5
Parent cultural background			.09*	−.07
Australian	548	75.8
Indigenous Australian	23	3.2
New Zealander	26	3.6
European (Northwestern, Southern and Eastern European)	102	14.1
African (North African, Sub-Saharan and Middle Eastern)	12	1.7
Asian (Northeastern, Southern and Central Asian)	4	0.6
People of the Americas	8	1.1
Parent educational level			.09*	−.08*
Some secondary school, up to completion	157	21.7
Technical college or trade certificate/apprenticeship	224	31.0
Undergraduate university degree	183	25.3
Postgraduate university degree	158	21.9
Parent employment type			−.12**	.14**
Full-time	260	36.0
Part-time or casual	283	39.1
Part- or full-time student	34	4.7
Home duties	108	14.9
Not in paid employment	38	5.3
Family structure			−.15**	.12**
Original two-parent family	399	55.2
Step or blended family	116	16.0
Sole parent family	202	27.9
Other	6	0.8
Language most frequently spoken at home			−.01	−.03
English only	685	94.7
English and other language	29	4.0
Mainly other language/Did not answer	9	1.2

SCARED = Screen for Child Anxiety Related Emotional Disorders; AFS = Adolescent Functioning Scale; PD = Positive Development; ODB = Oppositional Defiant Behaviour.

*p < .05, **p < .01 (2-tailed).

Measures

Sociodemographic information

Detailed demographic information was collected from respondents regarding their own and their adolescent’s background (see Table 1). Financial status was assessed using four items regarding the participant’s ability to meet essential household expenses and worries about having enough food and money. These were regrouped to form a single variable, “Financial Stress”, by averaging the standardized scores of the four items.

Adolescent anxiety

Symptoms of anxiety disorders in adolescents were assessed using the 41-item parent-report version of the Screen for Child Anxiety Related Emotional Disorders (SCARED; Birmaher et al., 1997). The measure assessed a range of anxiety symptoms across the categories of panic disorder, generalized anxiety disorder, separation anxiety disorder, social anxiety disorder, and school avoidance, including items such as, When my child feels frightened, it is hard for them to breathe and People tell me that my child worries too much. Items were rated on a 3-point Likert scale ranging from 0 (Not True or Hardly Ever True) to 2 (Very True or Often True) and summed to provide a total score. Support for the factor structure and construct validity of the SCARED was provided in a meta-analysis of 25 studies from 7 countries (Hale et al., 2011). Internal consistency in this study for the total score was α = 0.96.

Adolescent psychosocial functioning

Adolescent psychosocial functioning was measured using the 9-item Positive Development (PD) and 8-item Oppositional Defiant Behaviour (ODB) subscales from the Adolescent Functioning Scale (AFS; Dittman et al., 2016). The PD subscale (α = 0.83) related to effective decision-making, resilience, self-regulation, and positive engagement with others (e.g., My teenager talks about their views, ideas and needs appropriately). The ODB subscale (α = 0.94) assessed disruptive behaviours such as physical and verbal aggression, disobedience, and argumentativeness (e.g., My teenager hurts me or others (e.g., hits, pushes, kicks)). Participants rated each statement based on how true it was of their adolescent’s behaviour over the preceding 4 weeks on a 6-point scale ranging from 0 (Not at All True) to 5 (True Most of the Time). A mean score for each subscale was created by averaging the item scores for each subscale, where higher scores indicated higher levels of the behaviour being assessed. Convergent and discriminant validity for the AFS, along with evidence for the factor structure of the scale, was established and confirmed in two samples of Australian parents of adolescents aged 11 to 18 years (Dittman et al., 2016, 2022).

Parent-adolescent relationship quality

The quality of the mother-adolescent relationship was assessed using the Parent-Adolescent Relationship Scale (PARS; K. Burke et al., 2021). The PARS is a 15-item measure that assesses the quality of the relationship between parents and their adolescent children. This study used the 6-item Connectedness subscale (α = 0.92) to assess warmth, acceptance and support in the relationship (e.g., During stressful times in my teenager’s life, I check if they are okay) and the 5-item Hostility subscale (α = 0.75) to measure negativity and criticism in the relationship (e.g., I criticize my teenager). Participants rated their responses on the PARS on a 6-point scale ranging from 0 (Not At All True) to 3 (Nearly Always or Always True). Support for the construct validity of the PARS was provided through confirmatory factor analyses, along with expected correlations with existing parenting measures, in a sample of Australian parents of adolescents (K. Burke et al., 2021).

Parenting practices

Parenting practices were measured using the short form of the Alabama Parenting Questionnaire (APQ-9; Elgar et al., 2007). The 9-item APQ comprises three 3-item subscales: Positive Parenting (α = 0.86; e.g., You compliment your teenager when they do something well), Inconsistent Discipline (α = 0.73; e.g., You threaten to punish your teenager and then do not actually punish them) and Poor Supervision (α = 0.77; e.g., Your child is out with friends you don’t know). Participants rated their responses to several statements on a 5-point Likert-type scale ranging from 1 (Never) to 5 (Always). The three items contributing to each subscale were summed to provide the subscale scores, with higher scores corresponding to higher frequency use of that parenting practice in the home. Evidence for the validity of the APQ-9 has been shown through confirmation of its three-factor structure, along with evidence that it discriminates between parents of children with and without disruptive behaviour disorders (Elgar et al., 2007).

Parental psychological distress

The short form of the Depression Anxiety and Stress Scales (DASS-21; Lovibond & Lovibond, 1995) was used to assess parental psychological distress. Participants rated each statement on a 4-point Likert-type scale ranging from 0 (Did Not Apply to Me at All) to 3 (Applied To Me Very Much, Or Most Of The Time) to indicate how much the statement applied to them over the preceding week. Example items included: I found it difficult to relax and I felt down-hearted and blue. The three 7-item subscales were summed to create a total score (α = 0.95), which prior research has demonstrated construct validity in the form of support for the factor structure of the scale (Henry & Crawford, 2005).

Procedure

The study was a cross-sectional, anonymous online survey. Institutional ethics approval was obtained from the authors’ Human Research Ethics Committee (approval #2014001087). Gatekeeper approval was obtained from state education departments to advertise the study in schools. Participants provided informed consent online before being given access to the survey. The survey was hosted on the online survey platform, Qualtrics, with all promotional materials directing participants to access the survey via a dedicated study website.

Data analysis

Total scale scores were computed with appropriate items reversed, and mean scale scores were used for the PARS and AFS. Correlational analyses explored the bivariate associations between all predictor and criterion variables. Hierarchical multiple regression analyses on PD and ODB scores were then conducted to examine the independent contributions of mother-adolescent connectedness and hostility to adolescent psychosocial functioning. Variables significant at the bivariate level were included in the regression models.

Moderation analyses were conducted to test the contribution of the interaction between adolescent anxiety and mother-adolescent relationship quality (connectedness and hostility) in predicting PD and ODB using model 1 of the PROCESS tool for SPSS version 3.5 (Hayes, 2022). Bootstrapped data including 5000 re-samples were plotted and inspected to assess the conditional effect of adolescent anxiety on PD and ODB under the constructs of connectedness and hostility. If the interaction term indicated significance (based on F statistic and R² change), the moderating effect of the interaction term was inspected via a simple slopes analysis.

Results

Data inclusion

Qualtrics recorded 876 survey responses. However, 143 responses were excluded because they were not valid attempts at the survey (i.e., participants had completed no more than the initial demographic items), or the adolescent was outside the target age range. Ten respondents (eight who identified as male and two who did not specify gender) were excluded as they represented a very small proportion of the total and removed so that analyses focused on female caregivers. The final N was 723 female caregivers. The data was checked for missing values and errors prior to conducting preliminary analyses. One parent response that was missing more than 50% of data was removed. All scales had <1% missing data. Little’s (1988) test indicated that data was missing completely at random, $\mathit{X}$² (22) = 0.00, p = 1.00. The Expectation-Maximization (EM) imputation method was used to estimate item-level missing data.

Data checking

Analyses to assess normality of data revealed skewness on SCARED Total, DASS-21, PARS Connectedness and APQ-9 Poor Supervision. Transformations showed no noticeable improvements in the distribution of the scales. Hence, the original scales were retained. Regression diagnostics ensured no violation of the assumptions of normality, linearity, multicollinearity, and homoscedasticity. One parent response was a multivariate outlier, but its removal did not reduce skewness and the response was retained.

Bivariate correlations

Pearson’s r was reported for categorical variables as no differences were found between parametric and non-parametric correlations. PD and ODB demonstrated several small to moderate significant correlations with socio-demographic variables (see Table 1). Higher levels of PD were significantly associated with parents being from a cultural background other than Australian, higher education levels, full-time work, less financial stress, reporting on a female adolescent and being from an original, two-parent family. Parents who were younger, with lower educational levels, less likely to be working full-time, under greater financial stress, with younger adolescents and living in non-original family households reported higher ODB.

Correlations between predictor variables and PD and ODB revealed significant small to large relationships in expected directions (see Table 2). Hostility was strongly correlated with ODB and moderately with PD, while connectedness was moderately correlated with ODB and PD. Higher hostility and lower connectedness in the mother-adolescent relationship was associated with lower PD and higher ODB. Moderate associations were found for adolescent anxiety and parental psychological distress such that higher levels related to lower PD and higher ODB. Small to moderate correlations suggested that greater use of ineffective parenting practices were associated with lower levels of PD and higher ODB.

Table 2. Bivariate correlations among predictor and criterion variables (N = 723).

	M	SD	1	2	3	4	5	6	7	8	9
SCARED Total	21.82	17.49	–
DASS-21 Total	30.56	25.04	.41**	–
APQ-9 Positive Parenting	12.94	1.89	−.01	−.08*	–
APQ-9 Inconsistent Discipline	6.52	2.42	.15**	.25**	−.12**	–
APQ-9 Poor Supervision	4.47	2.08	.09*	.19**	−.22**	.26**	–
PARS Connectedness	4.30	.80	−.03	−.14**	.56**	−.13**	−.25**	–
PARS Hostility	1.53	.77	.09*	.28**	−.27**	.48**	.26**	−.26**	–
AFS PD	2.80	1.03	−.28**	−.30**	.27**	−.31**	−.35**	.37**	−.42**	–
AFS ODB	1.77	1.14	.30**	.36**	−.22**	.46**	.29**	−.27**	.67**	−.50**	–

SCARED = Screen for Child Anxiety Related Emotional Disorders; DASS-21 = Depression Anxiety Stress Scales; APQ-9 = Alabama Parenting Questionnaire; PARS = Parent-Adolescent Relationship Scale; AFS = Adolescent Functioning Scale; PD = Positive Development; ODB = Oppositional Defiant Behaviour.

*p < .05, **p < .01.

Influence of mother-adolescent connectedness and hostility on adolescent psychosocial functioning

Two hierarchical multiple regressions were conducted to evaluate which of the components of mother-adolescent relationship quality, connectedness and hostility, were independent predictors of PD and ODB after controlling for sociodemographic factors, adolescent anxiety, parental psychological distress, and parenting practices (i.e., positive parenting, inconsistent discipline, poor supervision). Factors more distally related to adolescent functioning were entered in earlier blocks of the model; sociodemographic variables were entered at Step 1; SCARED total score at Step 2; DASS-21 total score at Step 3; APQ-9 subscale scores at Step 4; and PARS subscale scores at Step 5.

In the model for PD, the sociodemographic variables entered at Step 1 were parent cultural background, educational level, employment status, adolescent gender, family structure and financial stress (see Table 3). Adding adolescent anxiety at Step 2 contributed 9.1% of the variance in PD, while the inclusion of parental psychological distress at Step 3 added 2.5% of the variance in PD. At Step 4, parenting practices explained an additional 13.4% of the variance in adolescent PD. Finally, at Step 5, with all variables considered simultaneously, connectedness and hostility explained an additional and significant 7.4% of the variance in PD. At the final step, adolescent gender, family structure, adolescent anxiety, poor supervision, hostility, and connectedness were unique predictors. The full model was significant and explained 40.7% of variance in adolescent PD, F(13, 706) = 37.24, p < .001.

Table 3. Results of hierarchical regression analysis predicting positive development (N=723).

	Step 1			Step 2			Step 3			Step 4			Step 5
Variable	B	SE	β	B	SE	β	B	SE	β	B	SE	β	B	SE	β
Parent cultural background	.04	.02	.07*	.04	.02	.07*	.04	.02	.07*	.04	.02	.06*	.03	.02	.05
Parent educational level	.03	.04	.04	.04	.04	.04	.03	.04	.03	.06	.03	.06	.05	.03	.05
Parent employment status	−.06	.03	−.07	−.03	.03	−.03	−.02	.03	−.03	−.02	.03	−.03	−.00	.03	.00
Adolescent gender	.43	.07	.21***	.58	.07	.28***	.54	.07	.26***	.47	.07	.23***	.46	.06	.23***
Family structure	−.09	.03	−.12**	−.08	.03	−.10**	−.07	.03	−.10**	−.06	.03	−.08*	−.06	.02	−.07*
Financial stress	−.07	.05	−.06	−.01	.05	.01	.05	.05	.05	.05	.05	.04	.03	.04	.03
SCARED Total score				−.02	.00	−.32***	−.02	.00	−.25***	−.01	.00	−.24***	−.02	.00	−.25***
DASS-21 Total score							−.01	.00	−.19***	.00	.00	−.10**	−.00	.00	−.04
APQ-9 Positive Parenting										.11	.02	.20***	.02	.02	.04
APQ-9 Inconsistent Discipline										−.06	.01	−.15***	−.02	.01	−.05
APQ-9 Poor Supervision										−.10	.02	−.20***	−.01	.02	−.16***
PARS Connectedness													.29	.05	.22***
PARS Hostility													−.31	.05	−.23***
	R^2 full model = 8.3%, F(6, 713) = 10.76, p < .001			R^2 full model = 17.4%, R^2 change = .09, F change (1, 712) = 78.00, p < .001			R^2 full model = 19.8%, R^2 change = .03, F change (1, 711) = 22.05, p < .001			R^2 full model = 33.3%, R^2 change = .13, F change (3, 708) = 47.44, p < .001			R^2 full model = 40.7%, R^2 change = .07, F change (2, 706) = 44.17, p < .001

SCARED = Screen for Child Anxiety Related Emotional Disorders; DASS-21 = Depression Anxiety Stress Scales; APQ-9 = Alabama Parenting Practices, short form; PARS = Parent-Adolescent Relationship Scale.

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

In the model for ODB (see Table 4), the sociodemographic variables entered at Step 1 were parent age, educational level and employment status, adolescent age, family structure and financial stress. Step 1 explained 8.6% variance. At Step 2, adolescent anxiety explained an additional 6.8% of the variance in ODB. At Step 3, the inclusion of parental psychological distress contributed an additional 5.2% of the variance in ODB, while 17.6% of variance was added with the inclusion of parenting practices at Step 4. Finally, at Step 5, mother-adolescent connectedness and hostility explained an additional 19.0% of the variance in ODB. The full model explained 57.2% of the variance in OBD, F(13, 702) = 72.23, p < .001. Hostility made the greatest contribution to the model, with adolescent age and anxiety, inconsistent discipline, poor supervision and connectedness also contributing unique variance.

Table 4. Results of hierarchical regression analysis predicting oppositional defiant behaviour (N=723).

	Step 1			Step 2			Step 3			Step 4			Step 5
Variable	B	SE	β	B	SE	β	B	SE	β	B	SE	β	B	SE	β
Parent age	−.01	.01	−.07	−.02	.01	−.09*	−.01	.01	−.07*	−.01	.01	−.04	−.00	.01	−.01
Parent educational level	−.00	.04	−.00	.00	.04	.00	.01	.04	.01	−.02	.04	−.02	−.01	.03	−.01
Parent employment status	.06	.04	.07	.04	.04	.04	.03	.04	.03	.03	.03	.03	.01	.03	.01
Adolescent age	−.12	.02	−.18***	−.11	.02	−.17***	−.11	.02	−.18***	−.12	.02	−.20***	−.10	.02	−.16***
Family structure	.05	.03	.06	.04	.03	.04	.03	.03	.04	.02	.03	.02	.02	.02	.03
Financial stress	.17	.06	.13**	.12	.06	.09*	.02	.06	.02	.02	.05	.02	.07	.04	.05
SCARED Total score				.02	.00	.27***	.01	.00	.18***	.01	.00	.16***	.01	.00	.19***
DASS-21 Total score							.01	.00	.27***	.01	.00	.15***	.00	.00	.05
APQ-9 Positive Parenting										−.08	.02	−.13***	.00	.02	.01
APQ-9 Inconsistent Discipline										.14	.02	.31***	.05	.01	.11***
APQ-9 Poor Supervision										.10	.02	.18***	.06	.02	.11***
PARS Connectedness													−.13	.04	−.09**
PARS Hostility													.76	.05	.52***
	R^2 full model = 8.6%, F(6, 709) = 11.14, p < .001			R^2 full model = 15.4%, R^2 change = .07, F change (1, 708) = 56.97, p < .001			R^2 full model = 20.6%, R^2 change = .05, F change (1, 707) = 46.17, p < .001			R^2 full model = 38.2%, R^2 change = .18, F change (3, 704) = 66.76, p < .001			R^2 full model = 57.2%, R^2 change = .19, F change (2, 702) = 156.28, p < .001

SCARED = Screen for Child Anxiety Related Emotional Disorders; DASS-21 = Depression Anxiety Stress Scales, 21 items; APQ-9 = Alabama Parenting Practices, short form; PARS = Parent-Adolescent Relationship Scale.

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

Overall, the results from hierarchical multiple regression analyses showed that adolescent anxiety, connectedness, and hostility were independent predictors of both PD and ODB after controlling for sociodemographic factors, parental psychological distress and parenting practices.

Moderating Role of Mother-Adolescent Connectedness and Hostility on the Relationship Between Adolescent Anxiety and Adolescent Psychosocial Functioning

The interaction effects of adolescent anxiety and the mother-adolescent relationship on PD and ODB were tested in four separate moderation models: 1) anxiety X connectedness on PD; 2) anxiety X hostility on PD; 3) anxiety X connectedness on OBD and 4) anxiety X hostility on ODB (see Table 5). Each moderation regression model included the sociodemographic and parenting variables found to be significant covariates, adolescent anxiety, and the relevant parent-adolescent relationship variable (i.e., connectedness or hostility) entered separately, and finally, an interaction term representing the product of anxiety and each parent-adolescent relationship variable.

Table 5. Results from regression analysis examining the moderating effect of connectedness and hostility on adolescent anxiety by adolescent psychosocial functioning (PD and ODB; N=723).

Outcome Variable	Predictor		Coeff. (B)	SE	t	p
PD	Connectedness	Constant	.61	.40	1.53	.128
		Anxiety	.02	.01	1.93	.054
		Connectedness	.49	.07	6.82	.000
		Anxiety x Connectedness	−.01	.00	−3.33	.001
	Hostility	Constant	2.76	.33	8.42	.000
		Anxiety	−.03	.00	−7.17	.000
		Hostility	−.51	.06	−7.88	.000
		Anxiety x Hostility	.01	.00	3.97	.000
ODB	Connectedness	Constant	3.37	.52	6.50	.000
		Anxiety	.01	.01	0.97	.333
		Connectedness	−.19	.08	−2.56	.011
		Anxiety x Connectedness	−.00	.00	−0.06	.955
	Hostility	Constant	1.35	.40	3.40	.001
		Anxiety	.01	.00	3.14	.002
		Hostility	.76	.06	12.50	.000
		Anxiety x Hostility	.00	.01	0.24	.813

PD = Positive Development; ODB = Oppositional Defiant Behaviour.

In the first PD model, the interaction between anxiety and connectedness explained a significant increase in variance in PD, after controlling for significant covariates (R² change = .01, F(1, 703) = 7.97, p < .01). Simple slopes for the association between anxiety and PD were tested for low (16^th percentile), moderate (50^th percentile), and high (84^th percentile) levels of connectedness. The simple slopes revealed a significant negative association between anxiety and positive development, but anxiety was more strongly related to positive development at high levels of connectedness (B = −.02, SE = .00, t = −7.77, p < .001) than for moderate (B = −.02, SE = .00, t = −7.89, p < .001) or low (B = −.01, SE = .00, t = −2.48, p < .05) levels of connectedness (see Figure 1). The direction was unexpected, as this result indicated that parent-reported anxiety was more strongly associated with positive development when connectedness was high.

Figure 1. Impact of parent-adolescent connectedness and hostility on the association between adolescent anxiety and positive development.

Similarly, the interaction between anxiety and hostility explained a significant increase in variance in PD (R² change = .01, F(1, 703) = 13.40, p < .01). Simple slopes testing for the association between anxiety and PD at low (16^th percentile), moderate (50^th percentile), and high (84^th percentile) levels of hostility revealed a significant positive association between anxiety and positive development. Anxiety was more strongly related to positive development at low levels of hostility (B = −.020, SE = .00, t = −8.18, p < .001) than for moderate (B = −.015, SE = .00, t = −7.91, p < .001) or high (B = −.007, SE = .00, t = −2.85, p < .01) levels of hostility (see Figure 1). Again, the direction of this effect was unanticipated. This result suggests that parent-reported anxiety was more strongly associated with positive development when hostility was low.

Neither the interaction between anxiety and connectedness nor the interaction between anxiety and hostility contributed significant variance to the prediction of ODB. Thus, the mother-adolescent relationship did not moderate the relationship between anxiety and ODB.

Discussion

The present study examined the influence of parent-adolescent connectedness and hostility on adolescent positive development (PD) and oppositional defiant behaviour (ODB), within the context of adolescent anxiety and maternal perceptions of the parent-adolescent relationship. Consistent with the first hypothesis, the quality of the mother-adolescent relationship made an important contribution to the prediction of adolescent psychosocial functioning. In multivariate analyses, connectedness and hostility each uniquely predicted PD and ODB after controlling for a comprehensive set of variables including parental psychological distress, parenting practices and adolescent anxiety. As expected, mother-adolescent relationships characterized by lower hostility and higher connectedness were related to better PD and lower ODB among adolescents. These findings add to those of previous research that has found that warm and supportive parent-adolescent relationships protect adolescents against conduct problems and risk-taking behaviour (Foster et al., 2017; Markham et al., 2003), and also promote self-esteem, positive development and prosocial behaviour (Boutelle et al., 2009; O’Connor et al., 2020; van der Storm et al., 2022). The findings were also consistent with research that has highlighted the negative impact of parent-adolescent relationships that are high in criticism, conflict and negativity on adolescent behavioural and emotional functioning (Moffitt et al., 2002; Raudino et al., 2013; Weymouth et al., 2016). Hostility was a particularly strong predictor of ODB, underscoring how these types of negative interactions between parents and their adolescents may lead to defiant, disruptive, and aggressive behaviour in adolescents (Weymouth et al., 2016). Overall, the findings highlighted the importance of considering adolescent wellbeing within the context of family relationships.

As expected, results from multivariate analyses showed that adolescent anxiety was associated with adolescent psychosocial functioning. Higher levels of parent-reported adolescent anxiety were a unique predictor of both lower adolescent PD and higher ODB, even after controlling for parental psychological distress, parenting practices, and mother-adolescent relationship quality. Around a third of parents (35.5%) in this sample reported that their adolescent was in the clinical range on the SCARED, suggesting that there is a high level of anxiety symptoms among adolescents in the community. Thus, this pattern of findings extends previous research focusing on young people with diagnosed anxiety disorders (de Lijster et al., 2018), and suggests that, even among a community sample of parents of adolescents, anxiety has a negative impact on adolescents’ behavioural functioning and positive development. Poorer adolescent psychosocial functioning was also independently predicted by ineffective parenting practices including inconsistent discipline and poor supervision. The role of poor supervision was noteworthy, as it remained a unique predictor of both PD and ODB after controlling for mother-adolescent relationship quality. Poor supervision measured by the APQ-9 refers to parental knowledge of adolescent behaviour (e.g., the friends they are with or where they are) and has been found to be associated with oppositional behaviour and conduct problems (e.g., J. D. Burke et al., 2008; Laird et al., 2003).

The second hypothesis, which related to the potential buffering influence of positive mother-adolescent relationships on the association between adolescent anxiety and PD, was not supported. While mother-adolescent connectedness and hostility both played a moderating role in the association between adolescent anxiety and PD, the moderation effect occurred in a direction opposite to what was expected. The association between parent-reported adolescent anxiety and PD was stronger, rather than weaker, at higher levels of connectedness and lower levels of hostility. Put another way, and bearing in mind the cross-sectional nature of the study, greater mother-adolescent connectedness and lower mother-adolescent hostility exacerbated the negative influence of adolescent anxiety on positive development.

This unexpected finding may be explained by the notion that parental reports of anxiety are inextricably linked to the quality of the mother-adolescent relationship and the extent to which the adolescent shares their thoughts, feelings, and activities with their parent. Parents of adolescents experiencing anxiety may be more closely involved in their adolescent’s life to protect or assist their child to manage their feelings. The unintended consequence of this protectiveness and support may be to reduce opportunities for autonomy and development of skills associated with positive development and to reinforce anxiety-related behaviours such as reassurance-seeking. It is important to acknowledge that the present study was based on mother-report only, which is unlikely to provide a complete account of the relationship between parents and adolescents, particularly in the context of reported discrepancies in parent and adolescent ratings of this relationship (Kapetanovic & Boson, 2022). Future research is needed that assesses parent and adolescents’ perspectives of the parent-adolescent relationship to better understand if greater levels of connectedness (as reported by parents) are interpreted by adolescents as overprotective and overcontrolling, and therefore not supportive of positive development.

Nonetheless, this argument is consistent with previous research. For example, Simpson et al. (2020) found that adolescents who engaged in more reassurance-seeking from parents and who received more parental support also reported greater anxiety. They suggested that parents who encourage their adolescent to turn to them for support, reassurance and assistance with problem-solving may inadvertently remove the young person from the anxiety-provoking situations and lead them to rely on reassurance-seeking behaviour. The McLeod et al. (2007) meta-analysis on the association between parenting and child anxiety demonstrated the potential unintended consequences of parental concern and protection. Results showed that the parenting dimension of parental control had a stronger association with childhood anxiety than parental rejection, with this effect holding for mother and father behaviour. It is possible that excessive parental control could impact children’s development of self-efficacy and elevated anxiety, which in turn could lead to parental responses and behaviours which are more controlling. Taken together with prior research, this finding highlighted that interactions between parents and adolescents when responding to and coping with anxiety play a complex role in adolescent psychosocial functioning.

The third hypothesis was not supported. Neither connectedness nor hostility moderated the relationship between anxiety and ODB. For ODB, hostility was found to be a far stronger predictor than anxiety or connectedness in the multivariate analyses, which may have diminished its effect as a moderator. The lack of a moderating effect on the association between anxiety and ODB may also be due to shared risk factors between these frequently co-occurring conditions, including child temperament, social information processing and parent-child processes. The finding is consistent with a prior longitudinal study with adolescents that found parental warmth and harsh discipline moderated the effect of life stress on depression, but not on conduct problems (Wagner et al., 1996). The reciprocal processes involved in how parent-adolescent relationships interact with adolescent anxiety to influence behaviour outcomes remain a question for future research.

There were several limitations in the present study that should be acknowledged. Firstly, the study was cross-sectional and therefore causality of relationships, particularly in relation to the moderation models, cannot be inferred. Secondly, the sample was limited to predominantly female biological parents. The perspectives of fathers, a commonly under-represented group in child and adolescent psychology research (Phares et al., 2005), as well as other caregiver types, were not accounted for due to the small number of responses from these groups. The study is also limited by the lack of adolescent report, which is important given that many symptoms of anxiety are subjective and internalized, and because it meant that collateral information on mother-adolescent relationship quality was not obtained. Given that the processes examined in this study are bidirectional, including the association between adolescent anxiety symptoms and parent-adolescent relationship quality (van Eijck et al., 2012), a better understanding of the perspectives of adolescents is particularly important. Finally, this study is limited by its measurement of parenting practices often related to disruptive behaviours in adolescence, rather than to parenting constructs that are more specifically associated with the development and maintenance of anxiety, such as anxious modelling, control, and overinvolvement (Drake & Ginsburg, 2012). Overall, longitudinal research involving mother, father and adolescent reports and a more diverse sample of participants is needed to better understand how the quality of the parent-adolescent relationship may influence outcomes for young people experiencing anxiety.

In summary, the present study examined several parenting influences on adolescent psychosocial functioning and although many of the hypothesized associations had been documented in prior research, this was one of few studies that reviewed all these parenting variables in one model. Findings identified aspects of the parent-adolescent relationship as unique contributors to adolescent psychosocial functioning and potentially modifiable targets for intervention to improve adolescent wellbeing. Given the high prevalence of adolescent anxiety and increasing recognition of parental involvement in the aetiology of child and adolescent anxiety, further research is needed that assesses how the parent-adolescent relationship may either hinder or enhance outcomes for young people with anxiety.

Author contributions

All authors contributed equally to the conceptualization, drafting and revising of the manuscript. All authors read and approved the final manuscript.

Compliance with ethical standards

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional research committee and national ethical standards for research, and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Ethics approvals were obtained from the authors’ institutional HREC (The University of Queensland) and from the Education Departments of the Australian states of NSW, Queensland and Victoria.

Informed consent

Informed consent was obtained from all participants included in the study.

Disclosure statement

Drs Burke and Dittman hold Honorary Appointments with the Parenting and Family Support Centre (PFSC), School of Psychology at the University of Queensland. The PFSC is partly funded by royalties stemming from published resources of the Triple P – Positive Parenting Program, which is developed and owned by The University of Queensland (UQ). Royalties are also distributed to the Faculty of Health and Behavioral Sciences at UQ and contributory authors of published Triple P resources. Triple P International (TPI) Pty Ltd is a private company licenced by Uniquest Pty Ltd on behalf of UQ, to publish and disseminate Triple P worldwide. The authors of this report have no share or ownership of TPI. Dr Dittman is an author on a Triple P Program and receives royalties from TPI.

Data availability statement

Data associated with this project is stored securely by The University of Queensland, Australia. Requests for access to the data can be made in writing to the corresponding author.

Additional information

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.