DOI: 10.1002/jclp.

22946

RESEARCH ARTICLE

Collaborative Problem Solving reduces

children's emotional and behavioral difficulties
and parenting stress: Two key mechanisms

Georgina H. Heath1 | Christopher Fife‐Schaw2 | Lu Wang3,4 |

3,4
Christopher J. Eddy | Michael J. G. Hone | Alisha R. Pollastri3,4
5

1
Wandsworth Child and Adolescent Mental
Health Service, London, UK Abstract
2
School of Psychology, University of Surrey, Objectives: This study aimed to evaluate the effective-
Surrey, UK
3
ness of the Collaborative Problem Solving (CPS) ap-
Department of Psychiatry, Massachusetts
General Hospital, Boston, Massachusetts proach in home‐based family therapy and to explore two
4
Department of Psychiatry, Harvard Medical hypothesized mechanisms of change.
School, Boston, Massachusetts Method: Sixty‐seven families with children aged 3–12 years
5
Crossroads Children's Mental Health Centre, old completed a 12‐week home‐based CPS treatment
Ottawa, Ontario, Canada
program. Parent‐report measures were completed pre‐ and
Correspondence post‐intervention, including measures on parents’ fidelity of
Alisha R. Pollastri, Department of Psychiatry,
Massachusetts General Hospital, 151 Merrimac using CPS, parents’ empathy, children's executive function-
Street, Boston, MA 02114. ing, children's behavioral difficulties, and parenting stress.
Email: apollastri@[Link]
Results: There were significant reductions in children's
behavioral difficulties and parenting stress, and significant
improvements in children's executive functioning and
parents’ empathy. These improvements were greatest for
parents who had the greatest fidelity to CPS. Improvements
in children's executive functioning and parents’ empathy
mediated the relationship between parents’ CPS fidelity
and outcomes.
Conclusions: These results provide evidence that
home‐based family treatment with CPS may achieve positive
child and family outcomes by building children's executive
function skills and improving parents’ empathy.

KEYWORDS

Collaborative Problem Solving, emotional and behavioral

problems, empathy, executive functioning, parent stress

1226 | © 2020 Wiley Periodicals, Inc. [Link]/journal/jclp J. Clin. Psychol. 2020;76:1226–1240.

HEATH ET AL. | 1227

1 | INTRODUCTION

Collaborative Problem Solving (CPS) is an approach for understanding and reducing emotional and behavioral
difficulties among children and adolescents, and is frequently applied in family therapy to reduce parenting stress
and family dysfunction. The philosophy of the CPS approach is that difficulties such as oppositionality, defiance,
aggression, avoidance, and withdrawal are caused by the mismatch between an individual's neurocognitive skills
and the skills needed to handle a demand (Greene, 1998; Greene & Ablon, 2005). The neurocognitive skills needed
may be in executive functioning (including working memory, attention, emotion regulation, and cognitive flexibility),
or in nonexecutive areas such as language/communication or social thinking.
The practice of CPS includes two main phases: assessment and intervention. Specific activities within these two
phases can be completed by any adult caregiver, including a parent, teacher, or mental health provider. In the assessment
phase of the CPS approach, an adult caregiver identifies a target youth's lagging skills, and creates a list of expectations
and triggers that are difficult for the youth to handle. These expectations and triggers comprise a list of “problems to solve”
in the intervention phase of the approach. During the intervention phase, the youth and adult solve those problems
collaboratively, at times also adjusting demands to match the skill level of the youth (Ablon, 2019).
The CPS approach has been applied in a number of outpatient, inpatient, and school settings across North America
with documented benefits (for a review of outcomes across settings, see Pollastri, Epstein, Heath, & Ablon, 2013).
Evaluations of CPS in outpatient settings have found that CPS is effective for reducing oppositional behaviors, hyper-
activity/inattention, and parenting stress, and for improving relationships between parents and their children (Epstein &
Saltzman‐Benaiah, 2010; Greene et al., 2004; Johnson et al., 2012; Pollastri et al., 2013; Wolff, Greene, & Ollendick, 2008).
These studies suggest that CPS is effective with individual families and in parent groups, and a randomized controlled trial
suggested that CPS is at least as effective as a standard behavioral parent training model for improving youth symptoms
(Greene et al., 2004). However, while there is promising evidence for the efficacy of CPS, there has been little empirical
investigation into the mechanism of these effects. In this study, we begin to explore variables that are hypothesized to
drive the reductions in children's difficulties and parenting stress observed after treatment with CPS. We do this to gain a
greater understanding of the CPS approach, which can then drive ongoing improvements to the intervention. Additionally,
by determining which specific factors drive improved functioning in children and adolescents, we can reveal key com-
ponents required for healthy child and adolescent development.

1.1 | Child difficulties and parenting stress: Related outcomes

There is considerable evidence to suggest a bidirectional relationship between youth's emotional and behavioral
difficulties and parenting stress. Youths' emotional and behavioral difficulties heighten parenting stress (e.g.,
Solem, 2011), and in turn, parenting stress can lead to an increase in youths' difficulties (e.g., Krahé, 2015).
Parenting stress has been associated with parenting practices that are more punitive, irritable, and critical, which in
turn increases negative child–parent interactions and the likelihood that a child will develop conduct problems
(Pinderhughes, Dodge, Bates, Pettit, & Zelli, 2000). In addition, Kazdin and Whitley (2003) found that adding a
component targeting parenting stress to parent management training resulted in greater therapeutic change.
Therefore, it is predicted that in this sample of clinically referred youth, children's difficulties and parenting stress
will be closely related, and we include both as key outcome variables in this study.

1.2 | Hypothesized mechanisms of change

The two hypothesized mechanisms of change explored in this study come directly from the theory of change for the
CPS approach. Specifically, the theory of change suggests that use of CPS results in reductions in youths’ emotional
1228 | HEATH ET AL.

and behavioral difficulties and parenting stress via multiple pathways. In one pathway, the problem solving that
occurs between adult and child directly improves the youth's ability to meet the specific expectations that have
been problematic, and also indirectly improves his or her ability to meet future expectations by building necessary
neurocognitive skills. In the second pathway, increased empathy that occurs when adults understand that youths’
difficulties are a result of skill deficits rather than willful or manipulative behavior leads to a reduction in the adult's
use of punitive discipline and an improved youth/adult relationship, which then drive emotional and behavioral
improvements. Consistent with this theory of change, the two key mechanisms explored in this study are im-
provement in children's neurocognitive skills and improvement in parents’ empathy.
We focus on children's executive functions, which refers to a set of higher order cognitive functions including
planning, shifting from one idea to another, inhibiting responses, controlling emotions, and solving problems
(Miyake et al., 2000). Executive functions encompass many of the neurocognitive skills built through use of CPS.
The CPS process builds executive skills by scaffolding opportunities for the child to practice (a) regulating emotions
while expressing his or her concerns about an unmet expectation; (b) inhibiting impulses while hearing the adult's
concerns about the unmet expectation; (c) generating solutions that could address all concerns; (d) considering
outcomes of multiple solutions; and (e) flexibly moving off an original idea if a solution is found to be unsatisfactory.
There is strong evidence to suggest that executive function deficits are associated with challenging behaviors
(Ogilvie, Stewart, Chan, & Shum, 2011), including externalizing behavior (Poland, Monks, & Tsermentseli, 2015),
antisocial behavior (Sorge, Skilling, & Toplak, 2015), and aggression (Granvald & Marciszko, 2016). Executive
function deficits appear to play a causal role in predicting later challenging behaviors (Hughes & Ensor, 2008;
Martel et al., 2007), and interventions intended to improve children's executive function skills have had beneficial
effects in reducing children's emotional and behavioral problems (Bierman et al., 2008; Domitrovich, Cortes, &
Greenberg, 2007; Kam, Greenberg, & Kusche, 2004). One study found that improvements in executive functioning
fully mediated the relationship between intervention and reduction in child conduct problems (Piehler et al., 2013).
Thus, since evidence suggests that youth who exhibit emotional and behavioral difficulties may have executive
function deficits that prevent them from responding adaptively to a given situation, and since CPS targets the
growth of executive function skills, this study examines whether the CPS intervention helps with the growth in
children's executive function skills as predicted, and whether growth in executive function skills predicts behavior
change for children.
The second mechanism explored in this study is parents' empathy. Empathy is defined as the ability to
understand and share the feelings and thoughts of another (Snow, 2000). Parental empathy plays an im-
portant role in the development of children's prosocial behavior and adjustment (Eisenberg & Strayer, 1990).
Research has found associations between high parental empathy and children's prosocial behavior, including
children's empathy (Upshaw, Kaiser, & Sommerville, 2015). In contrast, low parental empathy has been
associated with increased child misconduct (Psychogiou, Daley, Thompson, & Sonuga‐Barke, 2008) and
bullying behaviors (Curtner‐Smith et al., 2006). Furthermore, research has shown that interventions targeted
at increasing parents’ empathy helped to reduce children's emotional and behavioral difficulties (Flory, 2004;
Havigurst et al., 2013; Van Zeijl et al., 2006).
When caregivers learn and begin to adhere to the CPS philosophy, their conventional belief that children's
misbehavior is intentional and goal‐directed is replaced by the belief that a child's misbehavior is a downstream
effect of the child's struggle to meet the expectations placed on them. In addition, in learning the process of CPS,
parents and other caregivers are taught to actively listen to the child's concerns, to empathize with and validate
these concerns, and to ensure that the child's concerns are addressed when generating solutions. Greene and Ablon
(2005) suggest that this change in attribution, and the collaborative approach itself, foster a greater sense of
empathy among caregivers. Ashworth, Tapsak, and Li (2012) proposed that increased parental empathy is the
primary mechanism of change in CPS. However, to date, there has been no study that has measured whether
empathy increases among adults using the CPS approach. Therefore, this study will explore whether the CPS
HEATH ET AL. | 1229

intervention increases parents’ empathy as predicted, and whether increasing parents’ empathy drives changes in
children's difficulties and corresponding parenting stress following the CPS intervention.

2 | METHOD

2.1 | Participants and procedure

Participants were recruited from a multiservice community mental health center in eastern Ontario, Canada. This
service agency provides treatment for children up to 12 years old who have severe emotional, behavioral, and
social difficulties, and their families. Children were typically referred by their parents or caregivers for treatment
but may have also been referred by a family physician. CPS home‐based therapy was offered at the center for
families in which children are struggling significantly. Utilizing the CPS approach, a trained caseworker visited the
family at home 2–3 times per week for up to 12 weeks and delivered treatment to the child and their family.
Caseworkers held a Child and Youth Worker diploma from a Canadian college (or had received comparable
education) and all had attended 32 hr of intensive CPS training (referred to as Tier 1 and Tier 2 training). Case-
workers received weekly supervision and case consultation by a registered psychotherapist who had received over
60 hr of training and supervision in CPS, and who was certified to practice, supervise, and train others in CPS.
Issues of fidelity and adherence to the CPS model were addressed during these supervision sessions, on a case‐by‐
case basis.
Caseworkers invited all families of qualifying children to participate in the study. To qualify, children must have
been referred for behavioral difficulties and scheduled to begin home‐based CPS treatment between January 2015
and January 2016 and must have had a verbal IQ at least in the low average range (per formal assessment or
assumed unless deficits were evident). Children diagnosed with severe Autism Spectrum Disorder were excluded
from the study. Families provided informed consent for participation.
As part of routine data collection, the service agency asked all primary caregivers to complete measures of
children's emotional and behavioral difficulties, children's executive functioning, and parenting stress pre‐ and
posttreatment. For children participating in this study, caregivers consented to having these clinical measures used
for study purposes, and completed two additional questionnaires: one to assess parents' empathy and one for
parenting variables central to the CPS approach, including adherence to the CPS philosophy of “skill not will,” and
ability to predict challenging behaviors before they arose. For measures collected in the clinic, data provided for
research included only subscale scores, thus we rely on published psychometric information and do not report
internal consistencies for the current study.
Between January 2015 and January 2016, 84 caregivers consented to participate in the study and completed
the preintervention measures, and 67 completed both pre‐ and post‐intervention measures. Of the 17 participants
with no postintervention measures, 10 were still receiving treatment at the end of the study period, four had not
completed the postintervention measures, and three had dropped out of treatment. There were no significant or
notable differences between the 67 participants who completed all tasks compared to the three caregivers who
dropped out of treatment, or the four caregivers who did not complete postintervention measures. Remaining
analyses are reported for the 67 participants who had completed measures at both timepoints.
Children in the sample ranged in age from 3 to 12 years (M = 7.11, SD = 2.44), and the majority were male
(76.1%). Sixty‐five families’ primary language was English; the other two spoke English fluently enough to parti-
cipate in the treatment program and the study. Thirty‐nine percent of caregivers were single parents, and the
majority of primary caregivers were mothers (83.9%). Household income was greater than $60,000 CAD per year
for 47.8% of families, and the main source of income was employment (64.2%). Forty‐seven primary caregivers
were European/Caucasian (70.1%), 11 were Native Canadian (16.4%), two were Native American (3.0%), one was
Black/African (1.5%), and six were unreported (9.0%). While particular mental health diagnoses were not collected
1230 | HEATH ET AL.

for study participants, per inclusion criteria, all youth had been referred for significant behavior concerns, with a
preintervention average Total Difficulties Score on the Strengths and Difficulties Questionnaire (as introduced
below) in the very high range (mean = 21.1).

2.2 | Measures

2.2.1 | Youth emotional and behavioral difficulties

The Strengths and Difficulties Questionnaire, Parent‐Report Version (SDQ; Goodman, 1997) is a commonly used
measure of children's emotional and behavioral difficulties. This measure is designed to be completed by parents of
children aged 2–17 years. It consists of 25 questions divided into five subscales: Emotional Symptoms, Conduct
Problems, Hyperactivity, Peer Relationship Problems, and Prosocial Behavior. Answers are provided using a three
point Likert scale (“not true,” “somewhat true,” or “very true”). For each subscale, scores range from 0 to 10, with a
higher score on each of the first four subscales indicating greater difficulties. Scores are summed from all scales
except the prosocial scale to generate a Total Difficulties score, which ranges from 0 to 40. Total Difficulties scores
equal to or over 17 (16 for 2–4 year‐olds) are considered clinically significant (90th percentile of the normative
sample). Parents also provided answers to the SDQ “impact supplement,” consisting of 5 questions about the degree
to which the difficulties impact the child's functioning in different settings. Impact scores range from 0 to 10. The SDQ
versions used in this study were the English (USA) versions for children 2–4 years old, 4–10 years old, and
11–17 years old, and we used the four‐fold system to classify scores in each scale and subscale as “close to average,”
“slightly raised,” “high,” and “very high” (Scoring of the Strengths and Difficulties Questionnaire). The SDQ has
demonstrated sound psychometric properties with a mean internal reliability coefficient of 0.73 (Goodman, 2001).

2.2.2 | Parenting stress

To measure parenting stress, primary caregivers completed the Parenting Stress Index—Short Form (PSI‐SF;
Abidin, 1995), which is a 36‐item questionnaire designed to measure three major domains of parenting stress: child
characteristics (subscale: Difficult Child), parent characteristics (subscale: Parental Distress), and parent/child inter-
action (subscale: Parent–Child Dysfunctional Interaction). Each subscale contains 12 statements rated on a five point
Likert scale (“strongly agree” to “strongly disagree”). Subscale scores range from 12 to 60, and are summed to form a
Total Stress score (ranging from 36 to 180). A higher score indicates more stress. Total Stress score equal to or above
101 on the short form is considered clinically significant (85th percentile of the normative sample). Internal reliability
for the Short Form subscales have been reported between 0.88 and 0.95 (Reitman, Currier, & Stickle, 2002).

2.2.3 | Child executive functioning

The parent‐report version of the Behavior Rating Inventory of Executive Function (BRIEF; Gioia, Isquith, Guy, &
Kenworthy, 2000) is an 86‐item questionnaire commonly used to assess children's executive functioning. On the
BRIEF, executive functioning is indicated by the following subscales: Inhibit, Shift, Emotional Control (these three
are summed to form a Behavioral Regulation Index), Initiate, Working Memory, Plan/Organize, Organization of
Materials, and Monitor (these five are summed to provide a Metacognition Index). All eight subscales are summed
to provide a Global Executive Composite (GEC) score as a measure of children's overall executive functioning. Item
responses use a three point Likert scale (“never a problem,” “sometimes a problem,” or “often a problem”). Raw
scores are converted to T scores based on population norms. T scores range from 0 to 100 with a mean of 50, and
HEATH ET AL. | 1231

higher scores indicate greater executive function deficits (>65 is classified as clinically significant). The BRIEF is
intended for children 5–18 years old, and was only administered to children aged five and above, thus fewer
participants completed the BRIEF pre‐ and post‐intervention (N = 49), and analyses involving the Global Executive
Composite scores from the BRIEF were limited to this subset of participants. The BRIEF has demonstrated sound
psychometric properties, with internal consistency ranging between 0.80 and 0.98 (Gioia et al., 2000).

2.2.4 | Parental empathy

Parental empathy was measured using a previously adapted version of the Interpersonal Reactivity Index (IRI;
Davis, 1983; Psychogiou et al., 2008). This adapted version of the IRI was reworded to measure child‐directed
parental empathy, as opposed to general empathy, and includes two subscales of Perspective Taking (e.g., “I try to
look at my child's side of disagreement before I make a decision”) and Emotional Concern (e.g., “I am often quite
touched by things that I see happen to my child”). Answers are provided using a 5‐point Likert scale (from “does not
describe me very well” to “describes me very well”). Subscale scores range from 0 to 28. Higher scores on the
Perspective Taking and Emotional Concern subscales have been associated with increased empathy in prior re-
search (Cliffordson, 2001), and thus a sum of these two subscales are used in the current study to represent
parents’ empathy. Internal consistency coefficients of both Perspective Taking and Emotional Concern subscales
were satisfactory or better in this study (ranging from 0.56 to 0.86). Reliability of the composite score was
calculated using stratified alpha (Cronbach, Schönemann, & McKie, 1965), and it was 0.77 and 0.86, based on
preintervention and postintervention data, respectively.

2.2.5 | Parents’ CPS fidelity

The CPS Adherence and Impact Measure, Parent Version (CPS‐AIM‐P, Wang et al., 2019) is an 11‐item parent‐report
measure of parents’ attitudes and abilities related to CPS. It includes three subscales: The Philosophy Scale assesses
caregivers’ adherence to the CPS Philosophy (e.g., disagreeing with “My child chooses to act out to get out of doing things
he/she doesn't like”); the Prediction Scale assesses caregivers' ability to predict their child's challenging behavior (e.g.,
disagreeing with “My child's behavior toward me is unpredictable”); and the Parental Stress Scale assesses the quality of
parent–child relationships (e.g., agreeing with “Dealing with my child drains my energy”). Items are scored on a 7‐point
Likert scale (from “strongly agree” to “strongly disagree”). A mean is calculated for each subscale, to produce a subscale
score ranging between 1 and 7, with a higher score representing greater adherence, greater ability, and less stress. The
internal reliabilities of the CPS‐AIM‐P subscales in this study were satisfactory, ranging from 0.73 to 0.82. Given that
scores on the Parental Stress subscale were correlated with the Total Stress score measured by the PSI (r = .65), and that
PSI measures the same construct more comprehensively, the Parental Stress subscale of CPS‐AIM‐P was omitted for the
purpose of this study. In this study, the Philosophy and the Prediction subscales were combined into a single score to
represent parent's fidelity to CPS. Reliability of the composite score was calculated using stratified alpha, and it was 0.84
and 0.82, based on preintervention and postintervention data, respectively.

2.3 | Analytic approach

Unless otherwise stated, all data were normally distributed with skew and kurtosis values between ±2 SD. Paired
t tests were used to assess whether children's difficulties, children's executive functioning, parents’ empathy,
parenting stress, and parents’ CPS fidelity changed significantly following home‐based CPS treatment. Cohen's d
was calculated for each t test for effect sizes. Next, residualized change scores for key variables were computed by
1232 | HEATH ET AL.

taking the residuals of each variable after regressing on the preintervention values for that variable, to remove the
impact of its baseline differences. All analyses were done in RStudio (2016; version 1.0.143).
To evaluate mechanisms of change, regression analyses using the residualized change score approach were
conducted to see whether growth in children's executive function skills, parental empathy, and parents’ CPS fidelity
accounted for variance in the two outcome variables post‐intervention, including children's emotional and behavioral
difficulties and parenting stress, controlling for baseline variances of these variables before the intervention. For
children's difficulties, residualized change scores of the SDQ was the dependent variable (DV), residualized change
scores of parents’ empathy and children's executive functioning were the mediators (M), and parents’ fidelity to CPS
was the independent variable (IV). For parenting stress, residualized change scores of the PSI was the DV, and the
rest were the same as above. It was hypothesized that the CPS intervention would directly impact parents’ CPS
fidelity, including their increased understanding of challenging behavior and ability to predict problems; such changes
would increase parents’ empathy toward their child's struggles, and help their child build necessary executive
function skills, and these improvements would eventually drive the changes in the child's difficulties and parenting
stress. That is, the effect of the CPS intervention is via parents’ CPS fidelity, and its impacts on the outcome variables
are mediated by parents’ empathy and child executive function skills. Formal mediation analyses were evaluated using
bootstrapped confidence intervals for the indirect effects (MacKinnon, 2008).
Preliminary analysis suggested that none of the demographic variables (age, sex, single parent or not, and
family income) significantly predicted change over time as covariates in the regression analysis (with residualized
changes in CPS fidelity as the IV) for either children's difficulties (SDQ) or parenting stress (PSI). Thus, remaining
analyses were conducted without demographic covariates.

3 | RESULTS

3.1 | Change over time

Table 1 presents results of paired t tests examining change in all relevant scales and subscales over time. Results of
paired t tests indicate that after the intervention, parents’ adherence to the CPS philosophy as well as their ability to
predict children's behavioral difficulties improved significantly, t(65) = 3.97, and t(65) = 4.06, p < .001, for each scale,
respectively, as measured by the CPS‐AIM‐P. This confirms the delivery and receipt of an intervention that was
specifically grounded in CPS.
Results indicate improvement in both outcome variables. First, there was a statistically significant reduction in
children's difficulties from pre‐ to post‐intervention as measured by the SDQ Total Difficulties score, mean dropped
from 21.10 or “very high,” to 17.99 or “high,” t(66) = 5.39, p < .001, with statistically significant improvements in all
subscales: Emotional Problems, Conduct Problems, Hyperactivity, Peer Problems, as well as Prosocial Behavior and
SDQ Impact score. Additionally, results indicated a statistically significant decrease in parenting stress, as measured
by the PSI Total Stress scores, t(63) = 5.12, p < .001, with reductions in all three subscales: Difficult Child, Parental
Distress, and Parent–Child Dysfunctional Interaction.
Since both the SDQ and PSI manuals designate clinical significance cut off scores based on normative samples,
we defined clinically significant improvement as changing from above to below these cut off scores for SDQ Total
Difficulties and PSI Total Stress, and clinically significant deterioration as changing from below to above the
corresponding cut off scores. At intake, 55 children were above the clinical cut off scores on SDQ Total Difficulties,
and 37 parents were above the clinical cut off scores on PSI Total stress. After the CPS intervention, 31% of all
children experienced clinically significant reductions in their SDQ Total Difficulties (21 out of 67), while only 3%
experienced clinically significant increases in SDQ Total Difficulties (2 out of 67). In terms of parenting stress, 28%
of all parents reported clinically significant reductions in PSI Total Stress (18 out of 64 with PSI Total Stress both
pre‐and post‐CPS‐intervention), and only 2% (1 out of 64) reported clinically significant increases.
HEATH ET AL. | 1233

T A B L E 1 Change in key variables from pre‐ to post‐intervention

Measure Composite—scale Mean pre (SD) Mean post (SD) Range Diff. t(df) p Cohen's d

SDQ Total Difficulties 21.1 (5.99) 17.99 (6.45) 0–40 −3.12 5.39(66) ** 0.658
Emotional Problems 4.52 (2.85) 3.84 (2.87) 0–10 −0.69 2.85(66) * 0.348
Conduct Problems 5.21 (1.94) 4.1 (1.96) 0–10 −1.1 4.86(66) ** 0.594
Hyperactivity 7.64 (2.49) 7.04 (2.43) 0–10 −0.6 2.63(66) * 0.321
Peer Problems 3.73 (2.35) 3 (2.18) 0–10 −0.73 3.52(66) ** 0.43
Prosocial 6.33 (1.85) 6.96 (1.92) 0–10 0.63 −2.51(66) * −0.307
Impact 5.25 (2.54) 3.58 (2.93) 0–10 −1.67 4.06(62) ** 0.512

PSI‐SF Total Stress 102.45 (19.39) 92 (20.59) 36–180 −10.45 5.12(63) ** 0.64
Parental Distress 31.69 (8.88) 27.89 (8.08) 12–60 −3.8 3.76(63) ** 0.47
Dysfunctional 29.28 (8.01) 26.46 (7.2) 12–60 −2.82 4.33(63) ** 0.541
Interaction
Difficult Child 41.48 (7.26) 37.65 (9.18) 12–60 −3.83 4.03(63) ** 0.504

BRIEF Global 71.62 (8.77) 67.92 (11.34) 0–100 −3.7 3.39(48) ** 0.484
Behavioral Regulation 74.47 (10.04) 69.35 (11.23) 0–100 −5.13 3.99(51) ** 0.553
Index
Metacognition Index 67.21 (10.3) 64.73 (12.1) 0–100 −2.48 2.59(48) * 0.37

IRI IRI Empathy 39.83 (7.45) 41.88 (8.26) 0–56 2.05 −2.18(64) * −0.271
Emotional Concern 23.26 (3.83) 23.3 (3.67) 0–28 0.05 0.03(64) n.s. 0.004
Perspective Taking 16.58 (5.22) 18.58 (5.62) 0–28 2 −3.23(64) * −0.4

CPS‐AIM‐P CPS Integrity Total 3.56 (1.34) 4.39 (1.16) 1–7 0.83 −4.65(65) ** −0.573
Philosophy 3.32 (1.53) 4.12 (1.34) 1–7 0.8 −3.97(65) ** −0.488
Prediction 3.88 (1.67) 4.76 (1.41) 1–7 0.88 −4.06(65) ** −0.499

Abbreviations: BRIEF, Behavior Rating Inventory of Executive Function; CPS‐AIM‐P, CPS Adherence and Impact Measure—
Parent Version; IRI, Interpersonal Reactivity Index; PSI‐SF, Parenting Stress Index–Short Form; SDQ, Strengths and
Difficulties Questionnaire.
**p <= .001; *p < .05; n.s.p > .05.

Results also indicated improvements in the two variables that were hypothesized as mediators: children's
executive function skills and parents’ empathy. Increase in parents’ empathy were driven by a statistically sig-
nificant increase in IRI Perspective Taking, t(64) = 3.23, p < .05, with no significant differences found on IRI Emo-
tional Concern. Children's executive function skills, as measured by the BRIEF Global Executive Composite,
improved significantly, t(48) = 3.39, p < .001, and this appeared to be primarily driven by a statistically significant
decrease in the Behavioral Regulation subscales of Inhibit, Shift, and Emotional Control. There was a smaller but
still statistically significant improvement from pre‐ to post‐intervention on the Metacognition Index.

3.2 | Relationships among key variables and their changes

Correlations between key variables pre‐ and post‐CPS‐intervention were detailed in Table 2. Specifically, post-
intervention, SDQ Total Difficulties and PSI Total Stress were significantly correlated. CPS‐AIM Fidelity was
significantly related to children's SDQ Total Difficulties, parents’ PSI Total Stress, parents’ IRI Empathy, and
children's BRIEF Global Executive Composite. Finally, children's SDQ Total Difficulties were significantly correlated
with children's BRIEF Global Executive Composite, but not with parents' IRI Empathy, while parents' PSI Total
Stress was significantly correlated with both children's BRIEF Global Executive Composite and parents’ IRI Em-
pathy postintervention.
1234 | HEATH ET AL.

T A B L E 2 Pearson's correlations between key variables pre‐ and post‐intervention

Preintervention Postintervention

SDQ PSI BRIEF IRI AIM SDQ PSI BRIEF IRI

Pre PSI 0.33

BRIEF 0.62* 0.4
IRI −0.17 −0.53* −0.21
AIM −0.35 −0.49* −0.24 0.14

Post SDQ 0.71* 0.3 0.54* −0.19 −0.22

PSI 0.31 0.67* 0.45* −0.37 −0.31 0.50*
BRIEF 0.42 0.22 0.67* −0.27 0.05 0.66* 0.61*
IRI −0.03 −0.35 −0.01 0.53* 0.08 −0.06 −0.51* −0.18
AIM −0.26 −0.28 −0.27 0.15 0.36 −0.45* −0.53* −0.45* 0.39*

Abbreviations: AIM , CPS Adherence and Impact Measure—Parent Version (CPS‐AIM‐P), Fidelity; BRIEF, Behavior Rating
Inventory of Executive Function, Global Executive Composite; IRI, Interpersonal Reactivity Index, Empathy; PSI, Parenting
Stress Index–Short Form, Total Stress; SDQ, Strengths and Difficulties Questionnaire, Total Difficulties.
*p < .05, with Bonferroni correction.

Next, to explore how changes in the variables related to one another, we calculated residualized change scores
by obtaining the residuals after regressing the value of each variable postintervention on its own value pre-
intervention, and examined the association among the residualized change scores for all the variables, including the
independent variable (parents’ CPS‐AIM Fidelity), the hypothesized mediators (parents’ IRI Empathy and children's
BRIEF Global Executive Composite), and the dependent outcome variables (children's SDQ Total Difficulties and
parents’ PSI Total Stress). Pearson's correlations indicated that, after removing the impact of baseline variances in
each variable, change in parents’ CPS‐AIM Fidelity was significantly correlated with change in children's BRIEF
Global Executive Composite score (r = −.46, p < .001) and change in parents’ IRI Empathy (r = −.39, p < .001). There
were significant correlations between change in children's SDQ Total Difficulty and change in PSI Total Stress
(r = .44, p < .001), and both were significantly correlated with change in parents’ CPS‐AIM Fidelity (r = −.43 and −.50,
respectively, p < .001). Furthermore, change in children's BRIEF Global Executive Composite was significantly
correlated with change in children's SDQ Total Difficulties and parents’ PSI Total Stress (r = .54 and .60,
respectively, p < 0.001). Finally, change in parents’ IRI Empathy was not associated with change in children's SDQ
Total Difficulty (r = −.01, p > 0.05), but was significantly correlated with change in parents’ PSI Total Stress
(r = −.42, p < .001).

3.3 | Predicting changes over time and identifying mechanisms of change

We next used regression models and the residualized change score approach to test the hypothesized mediations
and used bootstrapped confidence intervals to evaluate any indirect effects. Results of the direct and indirect
effects are detailed in Figure 1.
In terms of predicting change in children's SDQ Total Difficulties, results indicate that all predictors (change in
parents’ CPS‐AIM Fidelity, children's BRIEF Global Executive Composite, and parents’ IRI Empathy) accounted for
29% of variance in the change of children's difficulties. In particular, improvements in parents’ CPS‐AIM Fidelity
significantly predicted reductions in children's SDQ, coefficient = −1.79, t(65) = −3.82, p < .001; such effect was
mediated by the improvements in children's BRIEF Global Executive Composite, coefficient = 0.24, t(63) = 3.85,
p = .002. Parents’ empathy was not a significant mediator, coefficient = 0.12, t(63) = 1.74, p > .05.
HEATH ET AL. | 1235

F I G U R E 1 Results on hypothesized
mediation mechanisms (top panel: SDQ;
bottom panel: PSI; **p <= .001; *p < .05)
[Color figure can be viewed at
[Link]]

In terms of predicting change in parents’ PSI Total Stress, results indicated that all predictors together ac-
counted for 43% of variance in the change of parenting stress. In particular, improvements in parents’ CPS‐AIM
Fidelity significantly predicted reductions in PSI Total Stress, coefficient = −7.05, t(65) = −4.59, p < .001; this effect
was mediated by improvements in children's BRIEF Global Executive Composite, coefficient = 0.93, t(63) = 5.07,
p < .001, as well as growth in parents’ IRI Empathy, coefficient = −0.60, t(63) = −2.82, p = .006.
Finally, neither children's preintervention SDQ Total Difficulties or parents’ preintervention PSI Total Stress
predicted the residualized change in children's BRIEF Global Executive Composite or parents’ IRI Empathy.

4 | D I S C U S SI O N

Our study had two aims: First, to explore the effectiveness of CPS in home‐based family therapy, and second, to
explore possible mechanisms of change within the CPS model. This extends prior research on CPS outcomes across
settings and begins exploring not just what outcomes CPS can accomplish, but how those outcomes are accom-
plished. A major strength of the study is that participants were children and families referred to treatment in a
community mental health agency, and thus results should be generalizable to community settings, where CPS is
most often used.
This study found that treatment with CPS was associated with positive youth and family outcomes, and
extended prior research by providing evidence for improvements in factors not previously studied. First, consistent
with prior research, parents reported a reduction in children's behavioral difficulties following the CPS interven-
tion. In particular, reductions were notable for children's externalizing behavior (conduct problems) and the impact
of difficulties on family functioning, and parents also reported reductions in children's emotional distress, hyper-
activity, and peer difficulties, as well as improvements in children's prosocial behavior following the intervention.
These findings support the growing evidence suggesting reductions in children's behavioral difficulties associated
with CPS (Pollastri et al., 2013) and extend those outcomes to emotional and social domains. Of note, while
improvement in children's outcomes were limited to parent reports, which could raise questions regarding sub-
jective versus objective change, it is typically parents’ reports of problematic youth behavior that provide the
impetus for referral to family treatment. Thus, it could be argued that parents’ perception is an important target of
1236 | HEATH ET AL.

the intervention; treatment cannot be considered successful unless parents report subjective improvements in key
outcomes. Regardless, confirming symptom improvement via multiple reporters or from objective measures of
behavior should be one goal of future research.
Second, also in accordance with previous research (Epstein & Saltzman‐Benaiah, 2010; Greene et al., 2004;
Schaubman, Stetson, & Plog, 2011), parents reported a statistically significant decrease in parenting stress post-
intervention that was also clinically significant for many caregivers. Specifically, there was a significant reduction in
the level of distress caregivers experienced within the parenting role, a decrease in dysfunctional interactions
occurring between parent and child, and a significant decrease in parents’ report of child characteristics that make
parenting difficult. These changes were strongest for parents who reported the greatest change in fidelity to CPS,
including an ability to predict their children's behavioral difficulties and adherence toward CPS's “skill‐not‐will”
philosophy. These results paint a picture of increased wellness for children and parents that are engaged in
treatment with CPS, and further support the growing evidence of positive child and family outcomes after treat-
ment with CPS. These results also suggest that it is not just intensive family therapy more generally that influenced
these outcomes, but rather adherence to aspects of the CPS approach in particular, providing an avenue for future
research on how to increase parents’ CPS fidelity once they are engaged in therapy.
Improvements in the two outcome variables, children's difficulties and parenting stress, were closely correlated
and based on prior research are likely to be related bidirectionally. We speculate that the CPS approach reduces
parenting stress by helping parents to predict children's behavioral difficulties and respond proactively as opposed
to reactively. It may also be that parents who are less distressed are engaging in more productive problem solving,
and thus either making more accommodations that better allow their children to meet expectations, or helping
their children build the skills needed to meet those expectations. Alternatively, parents who are less stressed may
rate their children's difficulties as less troubling, even if there is no objective change in children's symptoms, and
this will need to be ruled out. As mentioned previously, though parents’ subjective experiences are a key target of
family therapy, adding objective measures of children's symptoms will be useful to answer this question, and this is
an area ripe for future research. These findings highlight the need to consider, and possibly target, parenting stress
in interventions aimed at reducing child behavioral difficulties.
The second aim of this study was to explore mechanisms that drive these changes. In sum, consistent with the
theory of change for CPS, we found that improvements in parents’ CPS fidelity predicted reduction in children's
difficulties and parenting stress, and such predictions were achieved and mediated by growth in children’ executive
function skills as well as parents’ empathy.
Our hypothesis that improving children's executive function skills would drive improvements in children's and
parents’ outcomes was based on prior research indicating that executive functioning is associated with children's
difficulties, and that other treatments that include a component of executive functioning training have led to
improvements in children's emotional and behavioral outcomes. Consistent with our expectations, there was a
significant improvement in children's executive functioning following family therapy with CPS, and particularly in
the domains of executive functioning related to behavior regulation (the ability to inhibit and control impulses and
unwanted behavior, think flexibly, and control emotions). Furthermore, improvements in children's executive
functioning was the greatest predictor of improvements in children's emotional and behavioral difficulties, and
improvements in children's executive functioning also predicted reductions in parenting stress. This suggests that
children's executive functioning is a viable target when trying to influence important child and family outcomes.
When youth are better able to engage executive function skills, they are more capable of inhibiting undesirable
responses, controlling their emotions, and flexibly considering multiple solutions to a problem. These abilities allow
them to meet adult expectations, thus reducing problematic behavior, increasing the frequency of prosocial be-
havior, and reducing their emotional distress and the stress their parents experience. This finding provides strong
confirmation of one of the central hypotheses regarding the theory of change for CPS, provides a deeper under-
standing of how the CPS intervention currently works, and provides a rich area for future exploration and possible
improvements to the intervention to maximize executive functioning skill growth and child and parent outcomes.
HEATH ET AL. | 1237

To our knowledge, there has been no prior empirical study of whether parents’ empathy increases post‐CPS‐
intervention. In this study, consistent with our hypotheses, parents reported an increased ability to take their
children's perspective after the intervention. Additionally, parents’ improved empathy predicted reductions in
parenting stress. Contrary to prior research, in this study, improved empathy was not directly associated with a
reduction in youth difficulties. It is possible that the relationship observed in prior research was mediated by a
reduction in parenting stress, and this warrants further research. In the case of CPS, we hypothesize that arming
parents with the understanding that their children are doing the best they can with the skills they have and
teaching parents a problem‐solving process during which they investigate their children's point‐of‐view may
improve parents’ ability to take their children's perspective, which makes them feel more competent and less
stressed, thus contributing to an improved relationship and, ultimately, improved child outcomes.
The results outlined above contribute to our growing understanding of CPS, family treatment, and youth
development, but should be interpreted with certain limitations in mind. First, as mentioned earlier, parents’
reports of children's symptoms are an important information source and target for treatment, however, they may
not provide a complete picture of children's functioning. Additionally, like all subjective report measures, results on
our measures may be influenced by a social desirability bias. Both of these issues could be remedied in future
studies by using more objective measures, for example observational measures of parents’ perspective‐taking
abilities and fidelity to CPS when interacting with their child, or performance tasks measuring children's executive
functioning. Finally, with no comparison group and no information on number of sessions, it is not unequivocally
clear that proportion of the observed changes in outcomes may have occurred due to maturation or the passing of
time, rather than specifically due to the intervention. However, because CPS fidelity was included in our model, we
are cautiously optimistic that the results are due to this specific intervention. Here, too, further research is
warranted.
Despite these limitations, there are numerous strengths and clinical implications of our study. These results
make a significant contribution to the growing literature on the CPS approach, and importantly, provide an
exploration of not just whether CPS works, but how. This is consistent with our field's growing awareness of the
importance of going beyond studies of mean change to clarify mechanisms of intervention effects to constantly
improve existing interventions (e.g., Center on the Developing Child, 2016).
These findings support the use of CPS as an effective treatment for reducing children's emotional and beha-
vioral difficulties and parenting stress. They also suggest that CPS is a promising intervention to help children
improve executive function skills with support from their caregivers, and that there can be downstream effects on
other outcomes. This will be particularly beneficial for families of children who are already receiving executive
functioning skill‐building in school but need additional support outside of school. Similarly, findings from this study
suggest that parents’ empathy may be a viable target for treatment. Family therapy interventions typically attempt
to decrease parenting stress via reduction of the primary cause of stress, traditional cognitive restructuring, or by
teaching coping strategies. If further research establishes a causal relationship, parent empathy may provide
another pathway through which clinicians can consider addressing parenting stress.
In answering some questions, the results of this study raises others, and thus provides directions for future
research. As mentioned earlier, these results could be extended by including comparison groups and by including
objective measures of the outcomes examined. It would also be useful to collect data longitudinally, to examine
whether, and for how long, changes persist after discharge from treatment. There are also variables that may
contribute to the theory of change for CPS and that could be tested, including children's language processing and
social skills, or parenting style. Furthermore, this study focused on the use of CPS within an outpatient home‐based
setting only. While the approach is very similar in different types of settings, to our knowledge, there has not yet
been research into these mechanisms of change in children's outcomes within school or inpatient/residential
settings.
Finally, despite mean changes across the sample, parents’ CPS fidelity and symptom improvements did not
occur for all children and parents; a minority of families did not respond to CPS as expected and hoped. This is not
1238 | HEATH ET AL.

unusual in intervention research, but it is frequently ignored. In fact, this variability provides an opportunity for
further research into moderators of treatment effects. Future research should include specific measurement of
other youth and family factors (such as diagnoses and medications at intake), which were not measured in this
study. Results from these investigations will promote a better understanding of the CPS approach so that the
intervention can either be modified for those groups who do not respond as well, or be targeted toward children
and families who will see the most benefit.
In conclusion, this study contributes to the current evidence regarding the effectiveness of the CPS approach in
a home‐based treatment setting, and explores variables that predict improvements in children's and parents’
outcomes when using CPS. Our findings indicate that children's executive functioning and parents’ empathy are
important for healthy child development and family functioning, and can be identified as targets for interventions
aimed at reducing children's emotional and behavioral difficulties in family therapy.

A C K N O W L E D GM E N T S
The home‐based family therapy featured in this study is Collaborative Problem Solving (CPS). CPS was originated
as a parenting approach by Greene (1998), and then was further developed and manualized for treatment providers
by Greene and Ablon (2005). Dr. Greene continues to develop his own version of the approach, now called
Collaborative and Proactive Solutions, which he differentiates from the CPS approach. As of this writing, Colla-
borative and Proactive Solutions and CPS are identified on the Blue Menu of Evidence‐Based Psychosocial In-
terventions for Youth ([Link] as belonging to the same family of
treatment approaches, which collectively are referred to as Cooperative Problem Solving approaches. The authors
acknowledge Dr. Greene's seminal contribution to this family of approaches.

OR CID
Lu Wang [Link]
Alisha R. Pollastri [Link]

REFERENC ES
Abidin, R. (1995). The Parenting Stress Index. Lutz, FL: Psychological Assessment Resources.
Ablon, J. S. (2019). What is collaborative problem solving and why use the approach? In A. R. Pollastri, J. S. Ablon & M. J. G.
Hone (Eds.), Collaborative problem solving: An evidence‐based approach to implementation and practice. Switzerland:
Springer Nature.
Ashworth, K., Tapsak, S., & Li, S. T. (2012). Collaborative problem solving: Is empathy the active ingredient? Graduate
Student Journal of Psychology, 14, 83–92.
Bierman, K. L., Domitrovich, C. E., Nix, R. L., Gest, S. D., Welsh, J. A., Greenberg, M. T., … Gill, S. (2008). Promoting academic
and social‐emotional school readiness: the head start redi program. Child Development, 79, 1802–1817.
Center on the Developing Child. (2016). From best practices to breakthrough impacts: A science‐based approach to
building a more promising future for young children and families. Retrieved from [Link]
[Link]
Cliffordson, C. (2001). Parents' judgments and students' self‐judgments of empathy: The structure of empathy and
agreement of judgments based on the interpersonal reactivity index (IRI). European Journal of Psychological Assessment,
17(1), 36–47.
Cronbach, L. J., Schönemann, P., & McKie, D. (1965). Alpha coefficients for Stratified‐Parallel Tests. Educational and
Psychological Measurement, 25(2), 291–312.
Curtner‐Smith, M. E., Culp, A. M., Culp, R., Scheib, C., Owen, K., Tilley, A., … Coleman, P. W. (2006). Mothers' parenting and
young economically disadvantaged children's relational and overt bullying. Journal of Child & Family Studies, 15,
177–189.
Davis, M. H. (1983). Measuring individual differences in empathy: Evidence for a multidimensional approach. Journal of
Personality and Social Psychology, 44, 113–126.
Domitrovich, C. E., Cortes, R. C., & Greenberg, M. T. (2007). Improving young children's social and emotional competence: A
randomized trial of the preschool “PATHS” curriculum. The Journal of Primary Prevention, 28, 67–91.
Eisenberg, N., & Strayer, J. (Eds.). (1990). Empathy and its development. Cambridge, UK: Cambridge University Press.
HEATH ET AL. | 1239

Epstein, T., & Saltzman‐Benaiah, J. (2010). Parenting children with disruptive behaviors: Evaluation of a collaborative
problem solving pilot program. Journal of Clinical Psychology Practice, 1, 27–40.
Flory, V. (2004). A novel clinical intervention for severe childhood depression and anxiety. Clinical Child Psychology and
Psychiatry, 9, 1359–1045.
Gioia, G. A., Isquith, P. K., Guy, S. C., & Kenworthy, L. (2000). Behavior rating inventory of executive function. Child
Neuropsychology, 6, 235–238.
Goodman, R. (1997). The strengths and difficulties questionnaire: A research note. Journal of Child Psychology and Psychiatry,
38, 581–586.
Goodman, R. (2001). Psychometric properties of the Strengths and Difficulties Questionnaire. Journal of the American
Academy of Child and Adolescent Psychiatry, 40, 1337–1345.
Granvald, V., & Marciszko, C. (2016). Relations between key executive functions and aggression in childhood. Child
Neuropsychology, 22, 537–555.
Greene, R. W. (1998). The explosive child: A new approach for understanding and parenting easily frustrated, chronically inflexible
children. New York, NY: Harper Collins.
Greene, R. W., & Ablon, J. S. (2005). Treating explosive kids: The collaborative problem solving approach. New York, NY:
Guildford.
Greene, R. W., Ablon, J. S., Goring, J. C., Raezer‐Blakely, L., Markey, J., Monuteaux, M. C., … Rabbit, S. (2004). Effectiveness
of collaborative problem solving in affectively dysregulated children with oppositional‐defiant disorder: Initial findings.
Journal of Consulting and Clinical Psychology, 72, 1157–1164.
Hughes, C., & Ensor, R. (2008). Does executive function matter for preschoolers' problem behaviors? Journal of Abnormal
Child Psychology, 36, 1–14.
Johnson, M., Östlund, S., Fransson, G., Landgren, M., Nasic, S., Kadesjö, B., … Fernell, E. (2012). Attention‐
deficit/hyperactivity disorder with oppositional defiant disorder in Swedish children—an open study of collaborative
problem solving. Acta Paediatrica, 101(6), 624–630.
Kam, C., Greenberg, M. T., & Kusche, C. A. (2004). Sustained effects of the PATHS curriculum on the social and
psychological adjustment of children in special education. Journal of Emotional and Behavioral Disorders, 12, 66–78.
Kazdin, A. E., & Whitley, M. K. (2003). Treatment of parental stress to enhance therapeutic change among children referred
for aggressive and antisocial behavior. Journal of Consulting and Clinical Psychology, 71, 504–515.
Krahé, B. G. (2015). Child aggression as a source and a consequence of parenting stress: a three‐wave longitudinal study.
Journal of Research on Adolescence, 25, 328–339.
MacKinnon, D. P. (2008). Introduction to statistical mediation analysis. New York, NY: Lawrence Erlbaum Associates.
Martel, M. M., Nigg, J. T., Wong, M. M., Fitzgerald, H. E., Jester, J. M., Puttler, L. I., … Zucker, R. A. (2007). Childhood and
adolescent resiliency, regulation, and executive functioning in relation to adolescent problems and competence in a
high‐risk sample. Development and Psychopathology, 19, 541–563.
Miyake, A., Friedman, N. P., Emerson, M. J., Witzki, A. H., Howerter, A., & Wager, T. D. (2000). The Unity and diversity of
executive functions and their contributions to complex “Frontal Lobe” tasks: A latent variable analysis. Cognitive
Psychology, 41, 49–100.
Ogilvie, J. M., Stewart, A. L., Chan, R. C., & Shum, D. H. (2011). Neuropsychological measures of executive function and
antisocial behavior: A meta‐analysis. Criminology, 49, 1063–1107.
Piehler, T. F., Bloomquist, M. L., August, G. J., Gewirtz, A. H., Lee, S. S., & Lee, W. S. C. (2013). Executive functioning as a
mediator of conduct problems prevention in children of homeless families residing in temporary supportive housing: A
parallel process latent growth modeling approach. Journal of Abnormal Child Psychology, 42, 681–692.
Pinderhughes, E. E., Dodge, K. A., Bates, J. A., Pettit, G. S., & Zelli, A. (2000). Discipline Responses: Influence of parents'
socioeconomic status, ethnicity, beliefs about parenting, stress, and cognitive‐emotional processes. Journal of Family
Psychology, 14, 380–400.
Poland, S. E., Monks, C. P., & Tsermentseli, S. (2015). Cool and hot executive function as predictors of aggression in early
childhood: Differentiating between the function and form of aggression. The British Journal of Developmental Psychology,
34, 181–197.
Pollastri, A. R., Epstein, L. D., Heath, G. H., & Ablon, J. S. (2013). The collaborative problem solving approach: Outcomes
across settings. Harvard Review of Psychiatry, 21, 188–199.
Psychogiou, L., Daley, D., Thompson, M. J., & Sonuga‐Barke, E. J. S. (2008). Parenting empathy: Associations with
dimensions of parent and child psychopathology. British Journal of Developmental Psychology, 26, 221–232.
Reitman, D., Currier, R. O., & Stickle, T. R. (2002). A critical evaluation of the Parenting Stress Index‐Short Form (PSI‐SF) in
a head start population. Journal of Clinical Child and Adolescent Psychology, 31, 384–392.
Schaubman, A., Stetson, E., & Plog, A. (2011). Reducing teacher stress by implementing collaborative problem solving in a
school setting. Social Work Journal, 35, 72–93.
1240 | HEATH ET AL.

Scoring of the Strengths & Difficulties Questionnaire. Scoring the SDQ. Retrieved from [Link]
sdqinfo/[Link]
Snow, N. E. (2000). Empathy. American Philosophical Quarterly, 37, 65–78.
Solem, M. M. (2011). Predicting parenting stress: Children's behavioral problems and parents' coping. Infant & Child
Development, 20, 162–180.
Sorge, G. B., Skilling, T. A., & Toplak, M. E. (2015). Intelligence, executive functions, and decision making as predictors of
antisocial behavior in an adolescent sample of justice‐involved youth and a community comparison group. Journal
o Behavioral Decision Making, 28, 477–490.
Upshaw, M. B., Kaiser, C. R., & Sommerville, J. A. (2015). Parents' empathic perspective taking and altruistic behavior
predicts infants' arousal to others' emotions. Frontiers in Psychology, 6, 1–11.
Wang, L., Albright, M., Katzenstein, T., Raftery‐Helmer, J. N., Epstein, L. D., Berkowitz, M., … Pollastri, A. R. (2019). The
collaborative problem solving adherence & impact measure for parents (CPS‐AIM‐P). Think: Kids, Massachusetts General
Hospital, Harvard Medical School. Retrieved from [Link]
Wolff, J. C., Greene, R. W., & Ollendick, T. H. (2008). Differential responses of children with varying degrees of reactive and
proactive aggression to two forms of psychosocial treatment. Child & Family Behavior Therapy, 30(1), 37–50.
Van Zeijl, J., Mesman, J., Van IJzendoorn, M. H., Bakermans‐Kranenburg, M. J., Juffer, F., Stolk, M. N., … Alink, L. R. (2006).
Attachment‐based intervention for enhancing sensitive discipline in mothers of 1‐ to 3‐year‐old children for
externalizing behavior problems: A randomized controlled trial. Journal of Consulting and Clinical Psychology, 74,
994–1005.

How to cite this article: Heath GH, Fife‐Schaw C, Wang L, Eddy CJ, Hone MJG, Pollastri AR. Collaborative
Problem Solving reduces children's emotional and behavioral difficulties and parenting stress: Two key
mechanisms. J. Clin. Psychol. 2020;76:1226–1240. [Link]
Copyright of Journal of Clinical Psychology is the property of John Wiley & Sons, Inc. and
its content may not be copied or emailed to multiple sites or posted to a listserv without the
copyright holder's express written permission. However, users may print, download, or email
articles for individual use.