HHS Public Access

Author manuscript
J Child Psychol Psychiatry. Author manuscript; available in PMC 2022 January 21.
Author Manuscript

Published in final edited form as:

J Child Psychol Psychiatry. 2019 August ; 60(8): 848–856. doi:10.1111/jcpp.13052.

Exploring the Promise of Assessing Dynamic Characteristics of

the Family for Predicting Adolescent Risk Outcomes
Gregory M. Fosco1, Hio Wa Mak1, Amanda Ramos1, Emily LoBraico1, Melissa Lippold2
1The Pennsylvania State University
2University of North Carolina, Chapel Hill
Author Manuscript

Abstract
Background: Family-based assessments of risk factors for adolescent emotional, behavioral, and
substance use problems can be used to identify adolescents who are at risk and intervene before
problems cause clinically significant impairment. Expanding traditional methods for assessing
risk, this study evaluates whether lability, referring to the degree to which parent-adolescent
relationships and parenting fluctuate from day to day, might offer additional value to assessment
protocols aimed at identifying precursor risk factors.

Methods: This study sampled 151 adolescents and caregivers, collecting data at a baseline
assessment, a 21-day daily diary protocol, and a 12-month follow-up assessment. Daily diary
data were used to calculate within-family lability scores in parenting practices, parent-adolescent
connectedness, and parent-adolescent conflict.
Author Manuscript

Results: Regression analyses evaluated whether lability predicted adolescent’s depression,

anxiety, antisocial behavior (ASB), drunkenness, and marijuana use at 12-month follow-up.
Lability in parent-adolescent connectedness, accounting for baseline levels, gender, age, and initial
levels of outcomes, was associated with risk for depression, anxiety, ASB, drunkenness, and
marijuana use. Lability in parenting practices also was associated with risk for depression, anxiety,
and drunkenness. Baseline levels moderated some of these effects. Parent-adolescent conflict
lability was only associated with depression.

Conclusions: These findings provide evidence for substantial value added when including
dynamic assessments of family lability in predicting long-term adolescent risk outcomes and call
for integration of dynamic methods into assessment practices.
Author Manuscript

Keywords
Family Risk Assessment; Adolescent Psychopathology Risk; Adolescent Substance Use Risk;
Parent-Adolescent Relationships; Parenting Practices

This paper is currently In Press at the Journal of Child Psychology and Psychiatry.
Conflict of interest: The authors have no conflicts of interest to declare.
 Fosco et al. Page 2

Introduction
Author Manuscript

Developmentally, adolescents exhibit a notable increase in the prevalence of emotional

(e.g., depression, anxiety) and behavioral (e.g., antisocial behavior) mental health problems,
as well as engaging in heavy drinking or illicit drug use (Merikangas et al., 2010),
underscoring adolescence as a period of vulnerability. Recent trends in rising rates of service
utilization (Collishaw, 2015) indicate promising improvements in availability and perceived
acceptability of mental health services; however, psychiatric problems are costly to the
individual experiencing impairment, to educational and social settings, and to society at
large (Greenberg & Lippold, 2013). From a public health standpoint, it is more humane,
efficient, and effective to identify and intervene with at-risk adolescents early, before
onset of clinically significant impairment (Catalano et al., 2012). Such risk identification
calls for a shift away from symptom screening (e.g., depression screeners) to a focus
on precursor risk factors that foreshadow clinically-significant problems (Burns & Rapee,
Author Manuscript

2016). To effectively identify adolescents at risk for developing problems, it is critical to

fully characterize the nature and influence of precursor risk factors.

Family functioning plays a central role in adolescent well-being, with effects that persist into
adulthood (Fosco, Caruthers, & Dishion, 2012). Positive family relationships and effective
parenting are well-established protective factors, while family environments characterized
by poor relationships, harsh parenting, and high conflict undermine adolescent mental health
and elevate risk for engaging in antisocial behavior (ASB) and substance use (Greenberg
& Lippold, 2013; Repetti, Taylor, & Seeman, 2002). Perhaps equally important is the
accumulating evidence that these family factors are malleable and amenable to family-
centered preventive interventions (Redmond et al., 2009). Thus, accurate assessments of
malleable family risk and protective factors can identify adolescents at risk and guide
Author Manuscript

intervention resources to ameliorate precursor risk before problems become entrenched.

Three parent-adolescent factors have garnered considerable empirical support to have

associations with mental health problems, problem behavior, and substance use in
adolescence. These factors include: a) effective parenting practices, referring to warmth,
praise for good behavior, consistency, and parental monitoring, b) parent-adolescent
connectedness, referring to relationships characterized by trust, communication, and
emotional closeness, and c) parent-adolescent conflict referring to disagreements and/or
hostility. All three factors are consistently associated with these key outcomes in cross-
sectional and longitudinal studies (Ackard, Neumark-Sztainer, Story, & Perry, 2006;
Duncan, Duncan, Biglan, & Ary, 1998; Laird, Pettit, Bates, & Dodge, 2003; Van Ryzin,
Fosco, & Dishion, 2012; Weymouth, Buehler, Zhou, & Henson, 2016).
Author Manuscript

We argue that it is important to distinguish between traditional approaches that assess levels
of family risk or protective factors and dynamic approaches that assess lability – the extent
to which risk and protective factors fluctuate within families. Traditional, global assessments
capturing levels of family risk or protective factors, such as single-occasion surveys or
observations, represent the vast majority of extant basic research to date. These assessments
offer information about overall perceptions of how close and connected an adolescent feels
toward his/her parent, or the degree to which parents use effective parenting practices.

J Child Psychol Psychiatry. Author manuscript; available in PMC 2022 January 21.
 Fosco et al. Page 3

Dynamic characteristics of the family, such as lability in relationships or parenting over

Author Manuscript

time, offer additional information about risk that is not captured in global assessments (Ram
& Gerstorf, 2009). Lability refers to the degree to which parenting practices, parent-child
connectedness, and parent-child conflict fluctuate around their average levels over time
(Lippold, Fosco, Ram, & Feinberg, 2016). As an illustration, two families may report
identical levels of overall relationship quality in a global assessment, but one family may
have the similar scores each day (low lability), while the other fluctuates from extreme high
to extreme low scores across days (high lability). Although indistinguishable in terms of
their levels of relationship quality, adolescents in these two families have fundamentally
different relationship dynamics at home. An assessment of lability would be necessary
to differentiate risk in these families that would otherwise go undetected in traditional
assessments.

High lability may be an additional risk factor for maladjustment. From a deviance
Author Manuscript

perspective, lability in parenting practices or connectedness would reflect day-to-day “gaps”

in parental involvement that provide unsupervised opportunities for adolescents to engage
in problem behavior or substance use (Ary et al., 1999). From a bonding perspective,
lability in parent-adolescent relationships may undermine adolescents’ sense of security
and stability in their relationships with their parents, which is a foundational context for
identity, social, and behavioral development (Ackard et al., 2006; Rohner, 2004). Lability
in connectedness may erode adolescents’ confidence that they can rely on parental support
when they experience difficulties, and they may even experience periods of disconnect
as rejection. Additionally, lability in parenting, connectedness, and conflict would reflect
intermittent periods of hostility or deficient nurturance that are stressful for youth; over
time, this stress accumulates to undermine physical and mental health (Repetti et al., 2002).
These three perspectives converge around the common theme that lability in parenting and
Author Manuscript

parent-adolescent relationships should signal risk for poorer long-term adolescent outcomes.

Emerging evidence highlights lability as a unique and robust predictor of adolescent mental
health and substance use outcomes, even when accounting for levels of those family factors.
High lability in parenting from year to year is a risk factor for substance use, delinquency,
and depression, even when accounting for average levels of parenting during that time
(Lippold et al., 2016; Lippold, Hussong, Fosco, & Ram, 2018). At a daily timescale, high
lability in parental warmth was associated with diminished youth well-being, accounting
for average levels across days (Lippold, Davis, Lawson, & McHale, 2016). The current
study evaluates whether assessments of day-to-day lability in parenting practices, parent-
adolescent connectedness, and parent-adolescent conflict provide additional predictive value
– uniquely or in combination with traditional global assessments of these constructs–
Author Manuscript

in predicting adolescent risk for depression, anxiety, ASB, and substance misuse. Such
information could guide improvements to standard practices in risk screening.

The Current Study

Intensive longitudinal methods, such as daily diary methods, have unique strengths
in capturing aspects of family life that may get lost in broad-stroke, single-occasion
assessments (Bolger & Laurenceau, 2013). Daily diary designs offer several advantages

J Child Psychol Psychiatry. Author manuscript; available in PMC 2022 January 21.
 Fosco et al. Page 4

over traditional assessments because they minimize recall bias by shortening the reporting
Author Manuscript

timeframe, are more sensitive to capturing low-intensity experiences, and offer ecological
validity by assessing family members in their natural environment (Shiffman, Stone, &
Hufford, 2008). We leveraged a 21-day daily diary design to assess lability in parenting
practices, parent-adolescent connectedness, and parent-adolescent conflict. These lability
indicators were then evaluated as predictors of adolescent well-being alongside baseline
global assessments of these constructs (akin to a traditional screening tool). Our guiding
research questions were: a) does lability contribute unique predictive value for adolescent
depressive symptoms, anxiety symptoms, ASB, and substance use one year later? and b)
is lability associated with these outcomes at all levels of baseline family functioning (i.e.,
baseline level as a moderator of lability and outcomes)? It is possible that lability is only a
risk in certain families (e.g., in the context of moderate to low relationship quality).
Author Manuscript

Methods
Sample
Participants were 151 families of 9th and 10th grade adolescents who participated in part
of the larger Penn State Family Life Optimizing Well-being (FLOW) study, which was
approved by the University Institutional Review Board. Families were recruited through
high schools and family referrals to take part in a longitudinal study that included a daily
diary assessment. Families were eligible for the study if they met the following criteria:
adolescents lived in one, two-caregiver household continuously, Internet access and means to
complete the daily surveys at home, English fluency, the participating adolescent was in 9th
or 10th grade, and both parent and adolescent agreed to participate (via consent and assent).
The adolescents (61.5% female) were between the ages of 13 and 16 years old (M = 14.60,
SD = 0.83) and majority Caucasian (83.4%). Caregivers (95.6% female) were on average
Author Manuscript

43.4 years of age (Range 30–61), majority Caucasian (90.1%), with the majority of parents
being married (88.7%), having a median income of $70,000-$79,999, and at least a high
school degree (96.6%). In this specific study, families who participated completed baseline
assessments and then were assigned to complete daily assessments across 21 days and a
follow-up at 12 months. Daily questionnaires took approximately 5 minutes to complete and
were sent out nightly at 7 p.m. and access to links were available until 9 a.m. the following
morning.

Attrition—Of the 151 families, 10 youth did not complete the 12-month assessment.
Comparisons of demographic (e.g., sex, age, family income), baseline family factors (e.g.,
parent-child relationship), and baseline adolescent factors (e.g., anxiety, ASB) revealed only
two predictors of attrition: younger parents (t(141) = −1.98, p = .05) and low child anxiety
Author Manuscript

(t(32.40) = - 7.16, p < .001) were slightly more likely to drop out of the study. The final
analytic sample for this study was 141 adolescents and their parents.

Measures
Baseline Risk Assessment.—Parents reported on their parenting practices at baseline
using the General Child Management scale (Spoth, Redmond, & Shin, 1998). The 13 items
were rated from 1 (Never) to 5 (Always), capturing parental monitoring, positive discipline,

J Child Psychol Psychiatry. Author manuscript; available in PMC 2022 January 21.
 Fosco et al. Page 5

consistency, and harsh parenting (reverse-coded) and had good reliability (α = .72). Parent-
Author Manuscript

adolescent connectedness was reported by parents and adolescents. Parents completed

the 10-item closeness subscale of the Parent Child Relationship Scale (Hetherington &
Clingempeel, 1992). Items were rated from 1 (Not at all) to 5 (Extremely) and had high
reliability (α = .90). Adolescents rated 10 items from the communication and trust subscales
of the Inventory of Parent and Peer Attachment (Armsden & Greenberg, 1987). Items were
rated from 1 (Completely Untrue) to 5 (Completely True) and had high reliability (α = .91).
Parent-adolescent conflict was reported by parents and adolescents. Parents completed the
3-item recurring conflict scale and the 2-item negative affective quality scale. These 5 items
were rated from 1 (Never) to 5 (Always) and had good reliability (α = .76). Adolescents
completed 5 items from the Behavioral Affective Rating Scale (Conger, 1989), rated from 1
(Almost Never) to 5 (Almost Always) scale with good reliability (α = .85). See Appendix A
for a list of items in these scales.
Author Manuscript

Daily Measures for Lability.—The baseline risk measures were mirrored in daily
assessments, adapted as needed to fit a daily timescale. All items were rated on a 10-point
sliding scale (with 0.1 increments). Parents reported on daily parenting practices, including
5 items that tap into warmth, praise, consistency, and effective limit setting. This scale
exhibited good between-person reliability across diary days (R1F = .83). Parents and
adolescents rated 4 items assessing parent-adolescent connectedness on a daily basis. A
sample item included “I felt close and connected to my [parent/child] today”. Reliability
was good for parent (R1F = .90) and adolescent reports (R1F = .95). Finally, parents
and adolescents rated two items assessing parent-adolescent conflict (e.g., “How much
tension was there between you and your child?”). Reliability was good for parents (R1F =
.57) and adolescents (R1F = .77). After missing data were removed, lability scores were
computed using mean squared successive differences (MSSD), which captures the extent of
Author Manuscript

individuals’ change from one day to the next across all days (temporal instability) and is
less vulnerable to trends in the data (Jahng, Wood, & Trull, 2008). This resulted in three
variables: parenting lability, connectedness lability, and conflict lability.

Adolescent Outcomes.—All outcomes measures were assessed by adolescent report at

baseline and 12-month follow-up. Depressive symptoms were assessed using the 10-item
depression subscale of the Revised Child Anxiety and Depression Scale (RCADS; Ebesutani
et al., 2012). Adolescents reported how often things happened to them by indicating Never
(1), Sometimes (2), Often (3), Always (4) to items such as: “I feel sad or empty.” Reliability
was good at baseline and follow-up (α = .91, .92). Anxiety symptoms were assessed using
the 7-item Generalized Anxiety Disorder – 7 scale (GAD-7; Spitzer, Kroenke, Williams, &
Lowe, 2006). Adolescents rated how often in the last two weeks they experienced symptoms
Author Manuscript

(e.g., “feeling nervous, anxious, or on edge”), from Not at All (1), Several Days (2),
More Than Half the Days (3), Nearly Everyday (4). Reliability was good at baseline and
follow-up (α = .90, .94). ASB was assessed using the 10-item Antisocial Behavior Scale
(Dishion & Cavanaugh, 2003). Adolescents rated past-month frequency of behaviors such
as “intentionally hit or threatened to hit someone” from Never (1), Once or Twice (2), 3–5
Times (3), 6–10 Times (4), 11–20 Times (5), More Than 20 Times (6). This scale had
good reliability at baseline and follow-up (α = .87 and .95). Substance use was assessed as

J Child Psychol Psychiatry. Author manuscript; available in PMC 2022 January 21.
 Fosco et al. Page 6

past-month frequency of being drunk and smoking marijuana, using a single item for each:
Author Manuscript

“How many times did you [get drunk/smoke marijuana] in the past month?”

Results
Descriptive statistics and bivariate correlations among study variables are presented in Table
1. Correlations indicated that baseline levels of parenting and connectedness were only
modestly correlated with lability scores (r’s = −.18 to −.27). However, baseline levels of
conflict and conflict lability were more highly correlated (r’s = .42 to .47), yet still distinct
constructs.

Table 2 presents three sets of hierarchical regression analyses in which 12-month outcomes
were regressed on baseline levels of family variables and lability scores. Baseline scores of
outcomes, age, and gender were added as covariates. In a second step, an interaction term
Author Manuscript

for baseline levels and lability was entered in the equation. Before analysis, all predictors
and outcomes were standardized. In order to minimize risk of false discovery rate due to
multiple statistical tests, p-value calculations were adjusted using the Benjamini-Hochberg
(1995) approach.

The first set of analyses focused on parenting practices. All models, except that predicting
marijuana use, accounted for statistically significant variance in the outcomes. Baseline
parenting practices were not associated with any of the outcomes. Higher parenting lability
was associated with increases in adolescents’ depressive symptoms (β = .29), anxiety
symptoms (β = .18) and drunkenness (β = .29) at 12-month follow-up. Neither baseline
parenting nor parenting lability were associated with ASB. None of the interaction terms for
baseline parenting*lability were statistically significant.
Author Manuscript

The second set of analyses focused on parent-adolescent connectedness, computed

separately using adolescent and parent reports of connectedness. Baseline levels of
connectedness were not associated with any of the five outcomes. However, connectedness
lability was associated with increases in depressive symptoms (β = .27/.24), ASB (β =
.21/.33), and increases in the frequency of drunkenness (β = .31/.25) in models in which
both adolescent and parent reported connectedness, respectively. Connectedness lability
was associated with increases in anxiety symptoms (β = .17) and marijuana use (β =
.33) in models using adolescent, but not parent, report of connectedness. In the second
step, three baseline*lability interactions (via adolescent report) emerged as significant for
connectedness, when predicting ASB, drunkenness and marijuana use. These interactions
were probed in two ways (see Figure 1 for an example). We examined simple slopes for
lability at high (+1 SD) and low (−1 SD) values of baseline connectedness. Connectedness
Author Manuscript

lability was associated with increases in ASB (β = .47), drunkenness (β = .52), and
marijuana use (β = .65) at relatively high (but not low) levels of baseline connectedness.
Specificity is offered in calculations of the region of significance, such that connectedness
lability was associated with increases in ASB when baseline connectedness was −0.27 SD
or higher (values were standardized), with drunkenness when baseline connectedness was
−0.70 SD or higher, and with marijuana use when baseline connectedness was greater than
−0.49 SD. These findings suggested minimum levels of connectedness needed for lability

J Child Psychol Psychiatry. Author manuscript; available in PMC 2022 January 21.
 Fosco et al. Page 7

to serve as a risk factor for these three outcomes were below the sample average. However,
Author Manuscript

at extremely low levels of baseline connectedness, lability was negatively associated with
marijuana use (−2.33 SD; 1.4% of the sample).

The third set of analyses focused on parent-adolescent conflict using adolescent and parent
report of conflict. Of the 10 models run, only one finding emerged as statistically significant.
Parent report of conflict lability was associated with increases in adolescent depressive
symptoms (β = .34).

As a robustness test of lability as a predictor of risk outcomes, regressions were recomputed

by adding individual mean scores (iMean) of family variables across the 21 days of the study
to the models. When including both baseline and iMean scores, only two findings for lability
changed: parenting lability and connectedness lability became statistically non-significant
predictors of anxiety. Ten of the 12 coefficients remained statistically significant, suggesting
Author Manuscript

that the pattern of results was robust, relative to the baseline ratings and relative to the
average ratings (across days) of family functioning.

Discussion
This study evaluated whether there is added predictive value gained by incorporating
assessments of family lability to traditional (single-assessment) screening approaches when
predicting adolescent depression, anxiety, ASB, and substance use. Lability in parenting
practices and parent-adolescent connectedness exhibited robust predictive value for most
of these outcomes. However, parent-adolescent conflict and conflict lability were not
significant predictors in this sample. Overall, this study underscores the promise of
integrating dynamic assessments of family lability into traditional screening approaches as a
means of strengthening precursor risk screening in adolescents.
Author Manuscript

Parenting lability was a risk factor for adolescent depression, anxiety, and drunkenness,
regardless of baseline levels of parenting. Vacillations in the quality of parenting from day
to day reflect a family context in which there are days when family rules are upheld and
good behavior is praised and other days when limits are not enforced and good behavior
may go unacknowledged. From this view, adolescents may experience parenting lability
as intermittently rejecting, unpredictable, and stressful. Over time, these experiences may
undermine adolescents’ sense of security in the family and/or may accumulate to overwhelm
their coping resources, undermining their mental health, as indicated by increases in
depression and anxiety (Repetti et al., 2002; Rohner, 2004). In addition, the link between
parenting lability and increases in drunkenness are consistent with a deviance model of
risk in which intermittent periods of diminished parenting may also create opportunities for
Author Manuscript

adolescents to engage in deviant behavior (Ary et al., 1999).

High levels of lability in parent-adolescent connectedness was a consistent predictor of

most outcomes in this study. However, nuance in these findings emerged when considering
baseline levels of connectedness as a moderator of lability. Baseline connectedness
moderated connectedness lability when predicting ASB, drunkenness, and marijuana use,

J Child Psychol Psychiatry. Author manuscript; available in PMC 2022 January 21.
 Fosco et al. Page 8

but not when predicting depression or anxiety outcomes. These findings may indicate that
Author Manuscript

there are different underlying risk processes at work for these outcomes.

From a deviance perspective, our results indicate that “gaps” from day to day in
connectedness are a risk factor for ASB and substance use, in the context of a close overall
relationship. During periods of disconnect, adolescents may not seek parental guidance or
advice when navigating developmental challenges (e.g., peer substance use), or may even
seek connection with a deviant peer group that amplifies their risk for ASB and substance
use outcomes (Ackard et al., 2006; Dishion et al., 2004). However, in families with low
general connectedness, lability would not create “gaps” in supervision or guidance, thereby
not conferring risk for deviant behaviors.

However, connectedness lability is a risk factor for depression and anxiety, regardless of
baseline relationship levels. These relationships, characterized by days of feeling close
Author Manuscript

and connected and days of feeling disconnected may reflect a context of intermittent and
unpredictable nurturance, leading to accumulating stress that erodes adolescents’ mental
health over time (Repetti et al., 2002). Another interpretation may be that disrupted security
in the parent-adolescent relationship may elicit emotional distress over time (Rohner, 2004).

Findings for parent-adolescent conflict were surprising: neither level nor lability were
associated with adolescent outcomes (except depression), which is inconsistent with past
work. This raises questions about whether these measures were not capturing risk processes
sufficiently, or if this sample had unusually low levels of conflict. Regardless, we advise
caution when interpreting these null results until systematic replication is conducted.
However, the pattern of correlations among baseline and lability scores for conflict,
compared to parenting and connectedness, was noteworthy. Specifically, conflict lability
Author Manuscript

was considerably more highly correlated with the baseline level than the others, perhaps
shedding light onto conflict as a process. Distinguishing between level and lability in
conflict may be less appropriate. By definition, conflict may be an episodic process, such
that families with higher conflict may experience more conflict events (e.g., a “spiky”
line), whereas parenting and connectedness may reflect a continuous (rather than episodic)
process, characterized by smooth lines across days. Research characterizing couple conflicts
underscores the importance of capturing frequency and intensity of conflict in the home
(Fosco, DeBoard, & Grych, 2007). Alternatively, the high correlation between lability and
level of conflict may also indicate a “floor effect” limiting the measurement of lability.
Future work using higher-risk samples, should explore these issues.

This study has implications for family risk assessment and/or screening, family-based
interventions, and basic family science. First and foremost, the current findings call for
Author Manuscript

more work incorporating dynamic characteristics of the family—such as lability—into

developmental models of risk and protective processes. Our findings underscore the
importance of assessing lability in parenting and parent-adolescent connectedness when
predicting adolescent depression, anxiety, ASB, and substance use. Traditional family risk
assessments may benefit from integrating dynamic assessments of lability to capture a more
comprehensive picture of adolescent risk for long-term problems. With the ever-increasing
ubiquity of mobile technology, there is an opportunity for brief and efficient daily surveys to

J Child Psychol Psychiatry. Author manuscript; available in PMC 2022 January 21.
 Fosco et al. Page 9

be integrated into family risk assessments. This calls for the development of accessible and
Author Manuscript

engaging assessments with added potential for real-time feedback delivery to families.

This study also holds implications for family-based interventions. Intervention evaluations
should incorporate dynamic assessments of family functioning to better understand the
change processes underlying their effects. Currently, little is known about whether
interventions can reduce lability in these processes. Systematic work is needed to a)
evaluate whether existing interventions effectively reduce lability in these relationships and
b) develop new intervention content or components that can bolster intervention effects
regarding lability. An important first step is to incorporate daily diary designs in pre-post
assessments of family-based interventions.

Finally, the current study points to new gaps in family theory that await explicit evaluation.
We identified several perspectives – deviance models, parent-adolescent bonding, and
Author Manuscript

stressor accumulation – that may explain associations between family lability and adolescent
outcomes. However, the next step for this basic family research is to better explicate which
of these processes explain the effects presented in this study, or refine theories to better
explicate the underlying processes.

There were several limitations to this study. First, this sample was relatively low-risk.
Future work should replicate these findings in a sample of families with a greater range
of risk across family conflict, relationship quality, and adolescent risk. Moreover, greater
racial, socioeconomic, and family structure diversity would provide a more generalizable
test of lability in family functioning as a precursor risk factor for adolescent mental
health problems and substance misuse. Future work should also include other domains of
family functioning that may inform family risk assessments, such as family-level cohesion
Author Manuscript

and sibling relationship quality. Finally, other dynamic characteristics, such as rigidity
or reactivity in family functioning, may also offer promise as precursor risk factors for
adolescent problem outcomes (Ram & Gerstorf, 2009).

In conclusion, this study provides compelling preliminary evidence for the importance of
incorporating dynamic characteristics of the family – in this case, lability in parenting and
parent-adolescent connectedness – into family risk assessments for adolescent depression,
anxiety, antisocial behavior, and substance misuse. Beyond traditional approaches to family
screening that focus on the presence or absence of risk and protective factors (i.e., levels
of family functioning), understanding the degree to which families exhibit lability in
functioning can provide unique insight into adolescent risk. These findings call for expanded
assessments of family functioning and enhanced evaluation of family-based interventions.
Author Manuscript

Supplementary Material
Refer to Web version on PubMed Central for supplementary material.

Acknowledgments
This study was supported by the Karl R. and Diane Wendle Fink Early Career Professorship for the Study of
Families. Authors were supported by several funding sources: the Penn State Social Science Research Institute
(Fosco), the National Institute on Drug Abuse (Mak: T32 DA017629 and P50 DA039838; LoBraico: T32

J Child Psychol Psychiatry. Author manuscript; available in PMC 2022 January 21.
 Fosco et al. Page 10

DA017629; Lippold: R03 DA038685), and Institute of Education Sciences (R305B090007) and the John Templeton
Foundation (GF13361–152622) (Ramos). The content is solely the responsibility of the authors and does not
Author Manuscript

necessarily represent the official views of the NIDA, NIH, IES, or the John Templeton Foundation.

References
Ackard DM, Neumark-Sztainer D, Story M, & Perry C (2006). Parent-child connectedness and
behavioral and emotional health among adolescents. American Journal of Preventive Medicine,
30(1), 59–66. doi:10.1016/j.amepre.2005.09.013 [PubMed: 16414425]
Armsden GC, & Greenberg MT (1987). The inventory of parent and peer attachment: Individual
differences and their relationship to psychological well-being in adolescence. Journal of Youth &
Adolescence, 16(5), 427–454. doi:10.1007/BF02202939 [PubMed: 24277469]
Ary DV, Duncan TE, Biglan a, Metzler CW, Noell JW, & Smolkowski K. (1999). Development of
adolescent problem behavior. Journal of Abnormal Child Psychology, 27(2), 141–50. [PubMed:
10400060]
Benjamini Y, & Hochberg Y (1995). Controlling the False Discovery Rate : A Practical and Powerful
Approach to Multiple Testing. Journal of the Royal Statistical Society, 57(1), 289–300.
Author Manuscript

Bolger N, & Laurenceau J-P (2013). Intensive Longitudinal Methods: An Introduction to Diary and
Experience Sampling Research Guilford Press.
Burns JR, & Rapee RM (2016). Psychological Assessment Screening for Mental Health Risk in
High Schools : The Development of the Youth RADAR Screening for Mental Health Risk in High
Schools : The Development of the Youth RADAR. Psychological Assessment, 28, 1220–1231.
[PubMed: 26595219]
Catalano RF, Fagan AA, Gavin LE, Greenberg MT, Irwin CE, Ross DA, & Shek DTL (2012).
Worldwide application of prevention science in adolescent health. The Lancet, 379(9826), 1653–
1664. doi:10.1016/s0140-6736(12)60238-4
Collishaw S (2015). Annual research review: Secular trends in child and adolescent mental health.
Journal of Child Psychology and Psychiatry and Allied Disciplines, 56(3), 370–393. doi:10.1111/
jcpp.12372
Conger RD (1989). Iowa Youth and Families Project, Wave A. In Report prepared for Iowa State
University Ames, IA: Institute for Social and Behavioral Research.
Author Manuscript

Dishion TJ, Nelson SE, & Bullock BM (2004). Premature Adolescent Autonomy: Parent
Disengagement and Deviant Peer Process in the Amplification of Problem Behavior. Journal of
Adolescence, 27, 515–530. [PubMed: 15475044]
Duncan SC, Duncan TE, Biglan a, & Ary D (1998). Contributions of the social context to the
development of adolescent substance use: a multivariate latent growth modeling approach. Drug
and Alcohol Dependence, 50(1), 57–71. [PubMed: 9589273]
Ebesutani C, Reise SP, Chorpita BF, Ale C, Regan J, Young J, … Weisz JR (2012). The Revised Child
Anxiety and Depression Scale-Short Version: Scale Reduction via Exploratory Bifactor Modeling
of the Broad Anxiety Factor. Psychological Assessment, 24(4), 833–845. doi:10.1037/a0027283
[PubMed: 22329531]
Fosco GM, Caruthers AS, & Dishion TJ (2012). A six-year predictive test of adolescent family
relationship quality and effortful control pathways to emerging adult social and emotional health.
Journal of Family Psychology, 26(4), 565–575. doi:10.1037/a0028873 [PubMed: 22709261]
Fosco GM, DeBoard RL, & Grych JH (2007). Making sensen of family violence: Implications
of children’s appraisals of interparental aggression for their short- and long-term functioning.
Author Manuscript

European Psychologist, 12(1), 6–16. doi:10.1027/1016-9040.12.1.6

Greenberg MT, & Lippold MA (2013). Promoting Healthy Outcomes Among Youth with Multiple
Risks: Innovative Approaches. Annual Review of Public Health, 34, 253–270. doi:10.1146/
annurev-publhealth-031811-124619
Hetherington EM, & Clingempeel WG (1992). Coping With Marital Transitions : a Family Systems
Perspective, (May)
Laird RD, Pettit GS, Bates JE, & Dodge KA (2003). Parents’ Monitoring-Relevant Knowledge
and Adolescents’ Delinquent Behavior: Evidence of Correlated Developmental Changes and
Reciprocal Influences. Child Development, 74, 752–768. [PubMed: 12795388]

J Child Psychol Psychiatry. Author manuscript; available in PMC 2022 January 21.
 Fosco et al. Page 11

Lippold MA, Davis KD, Lawson KM, & McHale SM (2016). Day-to-day Consistency in Positive
Parent–Child Interactions and Youth Well-Being. Journal of Child and Family Studies, 25, 3584–
Author Manuscript

3592. doi:10.1007/s10826-016-0502-x [PubMed: 28736495]

Lippold MA, Fosco GM, Ram N, & Feinberg ME (2016). Knowledge lability: Within-person changes
in parental knowledge and their associations with adolescent problem behavior. Prevention
Science, 17. doi:10.1007/s11121-015-0604-5
Lippold MA, Hussong A, Fosco GM, & Ram N (2018). Lability in the parent’s hostility and
warmth toward their adolescent: Linkages to youth delinquency and substance use. Developmental
Psychology, 54, 348–361. doi:10.1037/dev0000415 [PubMed: 29154647]
Merikangas KR, He JP, Burstein M, Swanson SA, Avenevoli S, Cui L, … Swendsen J (2010). Lifetime
prevalence of mental disorders in U.S. adolescents: Results from the national comorbidity survey
replication-adolescent supplement (NCS-A). Journal of the American Academy of Child and
Adolescent Psychiatry, 49, 980–989. doi:10.1016/j.jaac.2010.05.017 [PubMed: 20855043]
Ram N, & Gerstorf D (2009). Time-structured and net intraindividual variability: tools for examining
the development of dynamic characteristics and processes. Psychology and Aging, 24, 778–91.
doi:10.1037/a0017915 [PubMed: 20025395]
Author Manuscript

Redmond C, Spoth RL, Shin C, Schainker LM, Greenberg MT, & Feinberg M (2009). Long-
term protective factor outcomes of evidence-based interventions implemented by community
teams through a community-university partnership. Journal of Primary Prevention, 30, 513–530.
doi:10.1007/s10935-009-0189-5
Repetti RL, Taylor SE, & Seeman TE (2002). Risky families: Family social environments
and the mental and physical health of offspring. Psychological Bulletin, 128, 330–366.
doi:10.1037/0033-2909.128.2.330 [PubMed: 11931522]
Rohner RP (2004). The parental “acceptance-rejection syndrome”: Universal correlates of perceived
rejection. American Psychologist, 59, 830–840. doi:10.1037/0003-066X.59.8.830
Shiffman S, Stone AA, & Hufford MR (2008). Ecological Momentary Assessment. Annual Review of
Clinical Psychology, 4, 1–32. doi:10.1002/9781118384404.ch20
Spitzer R, Kroenke K, Williams J, & Lowe B (2006). A brief measure for assessing generalized
anxiety disorder. Arch Inern Med, 166, 1092–7. doi:10.1001/archinte.166.10.1092
Spoth R, Redmond C, & Shin C (1998). Direct and indirect latent-variable parenting
outcomes of two universal family-focused preventive interventions: extending a public
Author Manuscript

health-oriented research base. Journal of Consulting and Clinical Psychology, 66, 385–399.
doi:10.1037/0022-006X.66.2.385 [PubMed: 9583342]
Van Ryzin MJ, Fosco GM, & Dishion TJ (2012). Family and peer predictors of substance use from
early adolescence to early adulthood: An 11-year prospective analysis. Addictive Behaviors, 37,
1314–1324. doi:10.1016/j.addbeh.2012.06.020 [PubMed: 22958864]
Weymouth BB, Buehler C, Zhou N, & Henson RA (2016). A meta-analysis of parent-adolescent
conflict: Disagreement, hostility, and youth maladjustment. Journal of Family Theory and Review,
8(1), 95–112. doi:10.1111/jftr.12126
Author Manuscript

J Child Psychol Psychiatry. Author manuscript; available in PMC 2022 January 21.
 Fosco et al. Page 12

Key Points
Author Manuscript

• Identifying precursor family risk factors is critical for preventing serious

mental health and substance use problems in adolescents.

• The current state of the field relies on single-assessment risk assessments,

which may offer incomplete predictive information.

• Our findings, using daily diary methods, reveal that lability (day-to-day
variability) in parenting practices and parent-adolescent connectedness
were robust predictors (accounting for traditional screening measures) of
adolescent mental health problems, antisocial behavior, and substance use.

• Future work should integrate dynamic methods into family risk assessments.
Author Manuscript
Author Manuscript
Author Manuscript

J Child Psychol Psychiatry. Author manuscript; available in PMC 2022 January 21.
 Fosco et al. Page 13
Author Manuscript
Author Manuscript

Figure 1.
Example plot of baseline connectedness*connectedness lability in relation to adolescent
antisocial behavior. 1a presents plotted simple slopes for connectedness lability and ASB
at 1 SD above and below the mean level of baseline connectedness. 1b presents the region
of significance for connectedness lability at different levels of baseline connectedness in
relation to ASB. Simple slopes are significant outside the lower and upper bounds (dotted
vertical lines).
Author Manuscript
Author Manuscript

J Child Psychol Psychiatry. Author manuscript; available in PMC 2022 January 21.
 Author Manuscript Author Manuscript Author Manuscript Author Manuscript

Table 1.

Means, Standard Deviation, and Bivariate Correlations.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Fosco et al.

1. Depressive Symptoms --

2. Anxiety Symptoms .69* --

3. ASB .20* .31* --

4. Drunkenness .03 .15 .30* --

5. Marijuana Use −.04 .08 .30* .63* --

6. Baseline Parenting (P) −.27* −.20* −.15 −.10 −.02 --

7. Baseline Connect (A) −.23* −.10 −.10 −.04 .05 .16 --

8. Baseline Connect (P) −.25* −.14 −.15 −.08 −.03 .53* .39* --

9. Baseline Conflict (A) .34* .28* .12 .04 .00 −.14 −.34* −.15 --

10. Baseline Conflict (P) .29* .20* .06 .10 .06 −.49* −.26* −.41* .44* --

11. Parenting Lability (P) .33* .21* .27* .34* .13 −.19* −.07 −.23* .16 .21* --

12. Connect Lability (A) .38* .25* .26* .30* .32* −.12 −.18* −.21* .34* .24* .43* --

13. Connect Lability (P) .30* .17* .40* .29* .18* −.21* −.06 −.27* .23* .20* .72* .51* --

14. Conflict Lability (A) .28* .10 .17* .03 .06 −.14 −.10 −.05 .42* .24* .31* .58* .41* --

15. Conflict Lability (P) .35* .19* .18* .15 −.04 −.28* −.22* −.22* .32* .47* .52* .26* .44* .39* --
16. Age .03 −.07 .01 .01 −.01 −.07 .10 −.09 −.12 −.04 .04 .07 −.04 .05 .02 --

17. Male Gender −.14 −.17* .28* .16 .01 −.11 −.02 −.20* −.03 .17* .27* .05 .20* .06 .17* .11 --

M 1.54 1.58 1.18 0.12 0.17 4.11 4.22 4.16 1.69 2.09 1.90 2.64 2.41 4.60 4.76 14.74 0.38

J Child Psychol Psychiatry. Author manuscript; available in PMC 2022 January 21.
SD 0.63 0.77 0.55 0.61 0.92 0.42 0.68 0.60 0.70 0.57 1.82 4.40 2.62 6.67 5.48 0.72 0.49

Note:
*
p < .05. Parenting = Parenting practices, Connect = Parent-Adolescent Connectedness, Conflict = Parent-Adolescent Conflict, A = adolescent report of family variable, P = parent report of family variable.
Page 14
 Author Manuscript Author Manuscript Author Manuscript Author Manuscript

Table 2.

Hierarchical Regressions on Family Variables Predicting Adolescent Adjustment Outcomes.

Depressive Symptoms Anxiety Symptoms Antisocial Behaviors Drunkenness Marijuana Use

Fosco et al.

Parenting Practices β β β β β
Step 1 (R2) (.37**) (.23**) (.15**) (.19**) (.02 ns)
Baseline Parenting Practices P −.16 −.15 −.08 −.02 --

Parenting Lability P .29** .18* .17 .29** --

Outcome at Baseline P .40** .32** .15 .24** --

Age .00 −.10 −.04 −.04 --

Gender −.39* −.31 .40 .16 --

Step 2 (R2) (.39**) (.23**) (.16**) (.20**) (.03 ns)

Baseline*Lability P −.16 −.05 −.11 −.12 --

Connectedness

Step 1 (R2) (.33**/.33**) (.20**/.21**) (.16**/.22**) (.20**/.17**) (.11*/.04 ns)

Baseline Connectedness A/P −.06/−.09 −.02/−.12 −.03/−.02 .01/−.01 .12/--

Connectedness Lability A/P .27**/.24** .17*/.12 .21*/.33** .31**/.25** .33**/--

Outcome at Baseline A/P .40**/.42** .33**/.35** .13/.11 .28**/.25** .03/--

Age .00/.02 −.10/−.09 −.05/−.01 −.06/−.02 −.06/--

Gender −.22/−.33 −.19/−.26 .50*/.39 .30/.21 .02/--

Step 2 (R2) (.35**/.36**) (.20**/.23**) (.21**/.22**) (.24**/.18**) (.20**/.04 ns)

Baseline*Lability A/P .14/−.21 .04/−.20 .24**/−.06 .20*/−.13 .32**/--

J Child Psychol Psychiatry. Author manuscript; available in PMC 2022 January 21.
Conflict

Step 1 (R2) (.31**/.38**) (.19**/.20**) (.13**/.13**) (.11*/.12**) (.01ns /.01ns)

Baseline Conflict A/P .12/.03 .15/.05 .05/−.07 .03/.02 −−/−−

Conflict Lability A/P .16/.34** −.03/.14 .10/.15 .00/.10 −−/−−

Outcome at Baseline A/P .42**/.46** .33**/.34** .15/.16 .27**/.27** −−/−−

Age .02/.01 −.07/−.09 −.03/−.04 −.04/−.04 −−/−−

Gender −.21/−.31 −.16/−.23 .51*/.48* .32/.28 −−/−−

Page 15
 Author Manuscript Author Manuscript Author Manuscript Author Manuscript

Depressive Symptoms Anxiety Symptoms Antisocial Behaviors Drunkenness Marijuana Use

Step 2 (R2) (.32**/.38**) (.20**/.20**) (.14**/.14**) (.11*/.16**) (.01ns /.02ns)

Baseline*Lability A/P −.10/−.01 −.09/−.03 −.08/−.15 .03/.22 −−/−−
Fosco et al.

Note: β = Standardized Coefficients (unstandardized regression coefficients were used for the categorical variable Gender). Baseline*Lability reflects the interaction term between baseline family variable
and the lability score. A = adolescent report of family variable, P = parent report of family variable.
*
p < .05,
**
p < .01.

J Child Psychol Psychiatry. Author manuscript; available in PMC 2022 January 21.
Page 16