Journal of Consulting and Clinical Psychology © 2016 American Psychological Association

2017, Vol. 85, No. 2, 178 –186 0022-006X/17/$12.00 http://dx.doi.org/10.1037/ccp0000155

BRIEF REPORT

Internet-Delivered, Family-Based Treatment for Early-Onset OCD:

A Pilot Randomized Trial

Jonathan S. Comer and Jami M. Furr Caroline E. Kerns

Florida International University Boston University
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Elizabeth Miguel and Stefany Coxe R. Meredith Elkins and Aubrey L. Carpenter
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Florida International University Boston University

Danielle Cornacchio Christine E. Cooper-Vince

Florida International University Massachusetts General Hospital and Harvard University

Mariah DeSerisy Tommy Chou and Amanda L. Sanchez

Fordham University Florida International University

Muniya Khanna Martin E. Franklin

Children’s Hospital of Philadelphia and University of University of Pennsylvania
Pennsylvania

Abbe M. Garcia and Jennifer B. Freeman

Brown University

Objective: Despite advances in supported treatments for early onset obsessive– compulsive disorder
(OCD), progress has been constrained by regionally limited expertise in pediatric OCD. Videotelecon-
ferencing (VTC) methods have proved useful for extending the reach of services for older individuals,
but no randomized clinical trials (RCTs) have evaluated VTC for treating early onset OCD. Method:
RCT comparing VTC-delivered family based cognitive– behavioral therapy (FB-CBT) versus clinic-
based FB-CBT in the treatment of children ages 4 – 8 with OCD (N ⫽ 22). Pretreatment, posttreatment,
and 6-month follow-up assessments included mother-/therapist-reports and independent evaluations
masked to treatment condition. Primary analyses focused on treatment retention, engagement and
satisfaction. Hierarchical linear modeling preliminarily evaluated the effects of time, treatment condition,
and their interactions. “Excellent response” was defined as a 1 or 2 on the Clinical Global Impressions-
Improvement Scale. Results: Treatment retention, engagement, alliance and satisfaction were high across
conditions. Symptom trajectories and family accommodation across both conditions showed outcomes
improving from baseline to posttreatment, and continuing through follow-up. At posttreatment, 72.7% of
Internet cases and 60% of Clinic cases showed “excellent response,” and at follow-up 80% of Internet
cases and 66.7% of Clinic cases showed “excellent response.” Significant condition differences were not

This article was published Online First November 21, 2016. Hospital of Philadelphia, and Department of Psychiatry, University of
Jonathan S. Comer and Jami M. Furr, Department of Psychology, Pennsylvania; Martin E. Franklin, Department of Psychiatry, University of
Florida International University; Caroline E. Kerns, Department of Psy- Pennsylvania; Abbe M. Garcia and Jennifer B. Freeman, Department
chological and Brain Sciences, Boston University; Elizabeth Miguel and of Psychiatry and Human Behavior, Brown University.
Stefany Coxe, Department of Psychology, Florida International University; This work was principally supported by the International OCD Foun-
R. Meredith Elkins and Aubrey L. Carpenter, Department of Psychological dation (IOCDF), as well by the NIH (K23 MH090247). Abbe M. Garcia
and Brain Sciences, Boston University; Danielle Cornacchio, Department and Jennifer B. Freeman receive book royalties from Oxford University
of Psychology, Florida International University; Christine E. Cooper- Press.
Vince, Massachusetts General Hospital, and Department of Psychiatry, Correspondence concerning this article should be addressed to Jonathan
Harvard University; Mariah DeSerisy, Department of Psychology, Ford- S. Comer, Center for Children and Families, Mental Health Interventions
ham University; Tommy Chou and Amanda L. Sanchez, Department of and Technology, Department of Psychology, Florida International Univer-
Psychology, Florida International University; Muniya Khanna, Children’s sity, 11200 SW 8th Street, Miami, FL 33199. E-mail: [email protected]

178
 INTERNET-DELIVERED EARLY OCD TREATMENT 179

found across outcomes. Conclusions: VTC methods may offer solutions to overcoming traditional
barriers to care for early onset OCD by extending the reach of real-time expert services regardless of
children’s geographic proximity to quality care.

What is the public health significance of this article?

Despite advances in supported treatments for early onset obsessive– compulsive disorder (OCD),
progress has been constrained by regionally limited expertise in pediatric OCD and local mental
health workforce shortages. This proof-of-concept pilot study suggests videoteleconferencing meth-
ods could be useful for overcoming traditional barriers to care for early onset OCD by extending the
reach of real-time expert services.

Keywords: behavioral telehealth, videoteleconferencing, OCD, early intervention, early childhood

This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Although research on treating pediatric OCD has historically youth. Such findings speak to the incremental benefits of VTC care
focused on OCD in middle childhood and adolescence (e.g., Bar- relative to expectations of future treatment, but cannot speak to
rett, Healy-Farrell, & March, 2004; Pediatric OCD Treatment comparisons with empirically supported standards of care. To date,
Study (POTS) Team, 2004; Piacentini, Bergman, Jacobs, Mc- there have been no controlled evaluations of VTC methods for
Cracken, & Kretchman, 2002), recent years have witnessed sub- treating early onset OCD. In one case series, Comer and colleagues
stantial advances in the development of family based cognitive– (2014) remotely treated five young children with OCD between
behavioral therapy (FB-CBT) methods for redressing the problems the ages of 4 and 8 with VTC-delivered FB-CBT. This real-time,
of early onset OCD (i.e., onset ⬍ age 9; Freeman et al., 2008, Internet-based interactive VTC format offered a comparable quan-
2014). For example, in a recent randomized controlled trial (RCT), tity of therapist contact to standard FB-CBT; all five families
Freeman and colleagues (2014) found their FB-CBT program to be completed a full treatment course, all showed symptom improve-
superior to family based relaxation training (reaction time [RT]) in ments, all showed at least partial diagnostic response, and all
the treatment of 5– 8-year-olds with OCD, with 72% of FB-CBT- treated children’s mothers characterized the quality of services
treated youth showing excellent response versus only 41% of received as “excellent.”
RT-treated youth. The present pilot RCT is the first randomized evaluation of
Despite advances, obstacles to quality care remain, including Internet-delivered treatment versus supported clinic-based treat-
limited numbers of professionals trained in early onset OCD, poor ment for early onset OCD. Given the early success of VTC
quality of available services, long waitlists, stigma, and transpor- methods for treating older youth with OC-spectrum disorders
tation issues (see Comer & Barlow, 2014; Comer et al., 2014). (e.g.,Himle et al., 2012; Storch et al., 2011) as well as the results
Technological innovations may offer useful solutions to overcom- of Comer and colleagues (2014), we hypothesized that VTC-
ing traditional barriers to effective care—particularly for low base delivered FB-CBT in this pilot trial would yield strong engage-
rate conditions with regionally limited specialty care options, such ment (as measured via high treatment completion), therapeutic
as early onset OCD (Comer, 2015; Comer & Barlow, 2014; Comer alliance, and satisfaction. In secondary analyses, we hypothesized
et al., 2014; Kazdin & Blase, 2011). Notably, videoteleconferenc- that VTC-delivered FB-CBT would yield significant symptom and
ing (VTC) methods have shown success in overcoming geograph- diagnostic improvements across time, and that condition differ-
ical obstacles to quality care in the treatment of a range of child ences between VTC-delivered and clinic-based FB-CBT would
not be supported, as evidenced by similar outcome trajectories and
internalizing and externalizing problems (e.g., Comer et al., 2015;
rates of response. Given strong links between pediatric OCD and
Duncan, Velasquez, & Nelson, 2014; Jones et al., 2014; Nelson &
patterns of family accommodation (e.g., Caporino et al., 2012),
Patton., 2016) by extending the reach of expert services, address-
exploratory analyses also examined the extent to which the two
ing regional workforce shortages, and overcoming issues of trans-
treatment conditions yielded improvements in parental accommo-
portation, stigma, and quality of care even in regions that do offer
dation.
clinic-based services (see Comer et al., 2014; Crum & Comer,
2016; Duncan et al., 2014). Patient populations can participate in
real-time services conducted by experts, regardless of their geo- Method
graphic proximity to an expert mental health facility. Moreover,
leveraging technology to deliver interventions in natural settings
Participants
(e.g., homes) may extend the ecological validity of treatments, as
services can be offered in the very settings where many problems Table 1 presents study eligibility criteria and baseline charac-
occur. teristics of study participations (N ⫽ 22).
Despite promise, remote VTC strategies for pediatric OCD have
received only limited empirical attention. In the only controlled
Measures
evaluation, Storch and colleagues (2011) found 81% of VTC-
treated youth in middle childhood and adolescence with OCD met Outcome measures included have all demonstrated favorable
posttreatment responder criteria, compared to only 13% of waitlist reliability and validity in previous work. Specific details of psy-
 180 COMER ET AL.

Table 1
Eligibility Criteria, Demographic, and Clinical Characteristics of Sample (N ⫽ 22)

Eligibility Criteria
Inclusion Exclusion

1. Child between ages of 4 and 8 years 1. Child has emotional or behavioral

problems more impairing than
OCD
2. Child meets diagnostic criteria for a principal 2. Child receiving medication or
diagnosis of OCD at least 3 months in duration psychotherapy to manage
emotional or behavioral problems
3. Child and participating parent(s) English speaking 3. Child has pediatric acute-onset
neuropsychiatric syndrome (PANS)
4. Family home equipped with a computing device 4. History of severe physical or
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

mental impairments (e.g.,

This document is copyrighted by the American Psychological Association or one of its allied publishers.

intellectual disability, deafness,

blindness, developmental delay) in
child or participating parent(s)
5. Child is ward of the state

Baseline demographic and clinical characteristics of sample (N ⫽ 22)

M (SD) % (N)
Child age, in years 6.65 (1.3)
Gender
Male 59.1 (13)
Female 40.9 (9)
Race/ethnicity
Caucasian/Non-Hispanic/Latino 91.0 (20)
Hispanic/Latino 4.5 (1)
Biracial 4.5 (1)
Maternal age, in years 38.1 (5.1)
Maternal education
Completed college 90.9 (20)
Did not complete college 9.0 (2)
Paternal age, in years 39.6 (4.0)
Paternal education
Completed college 68.2 (15)
Did not complete college 31.8 (7)
Biological parents’ marital status
Married 81.8 (18)
Not married 18.2 (4)
Annual household incomea
⬍$50,000 18.2 (4)
$50,000–$100,000 9.1 (2)
⬎$100,000 63.6 (14)
# child diagnoses 1.7 (.8)
Child comorbid diagnoses
Generalized anxiety disorder 22.7 (5)
Separation anxiety disorder 18.2 (4)
Social anxiety disorder 18.2 (4)
ADHD 13.6 (3)
Specific phobia 9.1 (2)
Note. ADHD ⫽ Attention deficit hyperactivity disorder.
a
2 families did not report annual household income.

chometric support for each measure are available from the corre- from 1 (Never) to 7 (Always) to characterize their perceptions of
sponding author by request. the affective bond between the client and therapist and the extent
Treatment satisfaction and therapeutic alliance. The Client of their agreement about the goals and tasks of treatment. We
Satisfaction Questionnaire (Larsen, Attkisson, Hargreaves, & included posttreatment mother- and therapist-WAI total reports
Nguyen, 1979) is a generic 8-item assessment of consumer satis- (␣Mother ⫽ .90 and ␣Therapist ⫽ .70 in present sample).
faction with services received (␣ ⫽ .86 in present sample). Post- Diagnostic, symptom, severity, and impairment outcomes.
treatment mother-reports were included in the present study. The The Anxiety Disorders Interview Schedule for Children and Par-
Working Alliance Inventory (WAI; Horvath, 1994) is a 36-item ents for DSM–IV (ADIS-IV-C/P; Silverman & Albano, 1996) is a
assessment of perceptions of the quality of therapeutic rapport and semistructured diagnostic interview that assesses child psychopa-
collaboration in treatment. Respondents rate each item on a scale thology in accordance with DSM–IV. We administered the ADIS-P
 INTERNET-DELIVERED EARLY OCD TREATMENT 181

(parent version) for all children. The ADIS-C (child version) was children’s primary dependence on the family system and with
also administered for youth ages 7– 8, and parent and child diag- appreciation for the restricted cognitive and capacities character-
nostic profiles were integrated using the “or” rule (see Comer & istic of early childhood. Parents are trained as coaches for their
Kendall, 2004). DSM–IV diagnoses are assigned and clinical se- children, ensuring out-of-session adherence and motivation; paren-
verity ratings (CSRs) range from 0 to 8 (CSR ⱖ4 denotes full tal accommodation of child symptoms is an explicit treatment
diagnostic criteria met). The Children’s Yale-Brown Obsessive- focus. Parents are taught to use differential attention, modeling,
Compulsive Scale (CY-BOCS; Scahill et al., 1997) is a 10-item, and scaffolding techniques to manage child symptoms. Children
semistructured, clinician-rated interview. CY-BOCS total score learn to externalize their symptoms by “bossing back” OCD.
ranges include “subclinical” (0 –7), “mild” (8 –15), “moderate” Therapists work with parents and children to create a fear hierar-
(16 –23), “severe” (24 –31), and “extreme” (32– 40; ␣ ⫽ .78 in chy that guides graduated E/RP tasks.
present sample). The Clinical Global Impression-Severity and Internet-delivered FB-CBT. Internet-delivered FB-CBT fol-
Improvement Scales (CGI-S/I) is the most widely used clinician- lows the same 14-week Freeman and Garcia (2009) program, but
rated measure of treatment-related changes in functioning (Guy & uses a VTC platform to allow therapists to remotely deliver real-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Bonato, 1970). CGI-S rates illness severity on a 7-point scale, time treatment to families directly in their homes. Internet-
This document is copyrighted by the American Psychological Association or one of its allied publishers.

ranging from 1 (“normal”) to 7 (“among the most severely ill delivered FB-CBT offers a comparable quantity of therapist time
patients”); CGI-I rates clinical improvement on a 7-point scale, as in clinic-based FB-CBT. Ethical and administrative consider-
ranging from 1 (“very much improved”) to 7 (“very much worse”). ations for use of VTC to deliver treatment are considered else-
CGI-I scores of 1 or 2 reflect “excellent response.” The Children’s where (Crum & Comer, 2016; Kramer, Kinn, & Mishkind, 2015),
Global Assessment Scale (CGAS; Shaffer et al., 1983) is a widely as are details on specific hardware and equipment used in the
used clinician-rated measure of overall child disturbance in func- present work (Comer et al., 2014). Our work relied on easy-to-use
tioning (range: 0 –100; lower scores ⫽ greater impairment). web conferencing appliances that are compliant with existing
Family accommodation. The Family Accommodation Scale– practice standards for VTC-delivered treatment. In lieu of collab-
Parent Report (FAS-PR; Flessner et al., 2009) assesses caregiver orative in-room activities that are central to clinic-based FB-CBT,
reports of family member participation in children’s OCD rituals, we used a series of interactive computer games to enhance chil-
including the facilitation of child avoidance, modification of fam- dren’s understanding of treatment concepts (see Comer et al.,
ily/parent routines in response to child OCD symptoms, and direct 2014). Further details of how treatment was adapted for VTC can
involvement in child compulsions (␣ ⫽ .81 in present sample). be found elsewhere (Comer et al., 2014).

Treatment Procedure
FB-CBT (Freeman & Garcia, 2009). Is a 14-week clinic- The recruitment, treatment, and follow-up assessment phases
based program drawing on supported CBT approaches used with were 3/5/12–7/30/15, 4/5/12– 8/4/15, and 2/1/13–2/24/16, respec-
older youth. FB-CBT contains exposure and response prevention tively. Procedures were approved by the Boston University IRB.
(E/RP) modifications tailored specifically for developmental com- IEs and therapists (N ⫽ 7) were masters-level trainees in clinical
patibility with children ages 4 – 8, with an awareness of young psychology. Participating families were recruited from families

Figure 1. Flow of participants across study phases.

 182 COMER ET AL.

Table 2
Engagement and Treatment Satisfaction Across Treatment Conditions

Internet-delivered FB-CBT Clinic-based FB-CBT

Variable M SD N % M SD N % Significance Test

# cases who completed full treatment course 10 90.9 10 90.9

Mean # sessions/case that started on timea 5.27 4.5 5.45 3.75 t(20) ⫽ ⫺.10, p ⫽ .92
Working allianceb
Mother-report 223.45 34.8 238.60 18.2 t(19) ⫽ ⫺1.23, p ⫽ .23
Therapist-report 226.10 32.9 233.00 17.4 t(18) ⫽ ⫺.59, p ⫽ .57
Treatment satisfaction (mother-report)b 28.55 4.5 30.50 2.0 t(19) ⫽ ⫺1.27, p ⫽ .22
Note. FB-CBT ⫽ Family-Based Cognitive-Behavioral Therapy (Freeman & Garcia, 2009).
a b
Starting “on time” was defined as session starting within 10 minutes of scheduled start time. Assessed at posttreatment.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

seeking services at the Center for Anxiety and Related Disorders FB-CBT family dropped out after Session 1 and one Clinic FB-
(CARD) at Boston University. An independent evaluator (IE) CBT family dropped out after Session 10. One hundred percent of
collected informed consent and then conducted a baseline ADIS families participated in the posttreatment assessment, and 86.4%
and CY-BOCS. CGAS and CGI scores were subsequently gener- of families participated in the 6-month follow-up assessment (see
ated by the IE. Families meeting eligibility were then evenly Figure 1). There were no significant differences between families
randomized using a two-digit random number generator to receive who did and did not participate in the follow-up assessment.
either clinic-based or VTC-based FB-CBT. Treatment was pro- Missing value analysis showed missingness on outcome variables
vided at no cost. Families participated in 12 sessions across 14 was not related to condition, previous waves of the same variable,
weeks. VTC families were provided with a temporary equipment or demographic variables, a missingness pattern consistent with
kit (⬃$200; see Comer et al., 2014), although most families missing at random (MAR). HLM mixed-effects models in the
independently possessed sufficient equipment. Therapist and IE present analysis utilized data from the intent-to-treat sample, ac-
training procedures, as well as security and confidentiality proce- counting for missing data using maximum likelihood (ML) esti-
dures for VTC treatment, are presented elsewhere (Comer et al., mation, which produces unbiased estimates when data are MAR.
2014). Interrater agreement on categorical codes (i.e., CGI-I/S, Families did not differ across conditions on any baseline demo-
OCD diagnosis) was high (i.e., ⬎80% interrater agreement after graphic or clinical variables. Mothers participated in treatment for
training and ⬎80% interrater agreement on study cases). The same 95.5% (N ⫽ 21) of cases. Fathers participated in treatment for
team of therapists provided treatment in both conditions. Treat-
68.2% of cases.
ment integrity checklists were completed on 10% of sessions;
Table 2 presents data across conditions regarding treatment
treatment integrity was high (94%) and did not significantly differ
engagement and satisfaction. Significant differences were not
across the conditions. At posttreatment and again at 6-month
found across conditions with regard to treatment retention or
follow-up, families met with IEs masked to treatment condition
session tardiness. All but one family in each condition completed
who conducted ADIS and CY-BOCS interviews and generated
a full course of treatment, and families in both conditions started
CGAS and CGI scores. Parents completed baseline, posttreatment,
less than half of their sessions on time (i.e., within 10 min of
and 6-month follow-up forms via an online survey application.
Families received $45 for completing each assessment point. schedule). Mother- and therapist-report alliance was very high
across conditions. Mothers reported very high treatment satisfac-
tion across conditions.
Results
HLM using ML estimation was used to model the nonindepen-
Treatment retention was high, with 90.9% of families complet- dence due to nesting of repeated observations (level 1) within
ing the full course of treatment of 12 sessions. One Internet participants (level 2). For each HLM, treatment condition and the

Table 3
Clinical Outcomes Across Assessment Points, by Condition

Internet-delivered FB-CBT Clinic-based FB-CBT

6 month follow-up 6 month follow-up
Outcome Pre M (SD) Post M (SD) M (SD) Pre M (SD) Post M (SD) M (SD)

Child OCD symptoms

CY-BOCS total 22.9 (4.1) 14.9 (7.3) 11.8 (9.5) 23.2 (3.2) 14.2 (7.8) 10.7 (8.3)
OCD CSR 5.1 (.8) 3.4 (1.2) 2.4 (2.6) 4.9 (.9) 3.1 (1.3) 2.4 (2.7)
Child global severity and functioning
CGI-S 4.9 (.7) 3.2 (1.5) 2.6 (2.5) 4.6 (.9) 3.3 (1.6) 2.8 (1.6)
CGAS 48.0 (8.0) 61.4 (12.0) 66.6 (15.9) 54.6 (9.5) 62.2 (15.0) 65.1 (14.3)
Family accommodation 29.5 (7.8) 19.5 (9.7) 15.6 (14.2) 21.1 (6.7) 14.8 (6.9) 12.4 (4.3)
Note. Means reflect model estimated means (not raw means); FB-CBT ⫽ Family-Based Cognitive-Behavioral Therapy (Freeman & Garcia, 2009);
CY-BOCS ⫽ Child Yale-Brown Obsessive-Compulsive Scale; CSR ⫽ Clinical Severity Rating; CGI-S ⫽ Clinical Global Impression-Severity Scale.
 INTERNET-DELIVERED EARLY OCD TREATMENT 183

Table 4
Results of Mixed-Effects Models Examining the Effects of Condition, Time, and
Their Interactions

Condition ⫻
Time
Condition Time Interaction
Outcome b p b p b p

Child OCD symptoms

CY-BOCS Total .3 .91 ⫺4.7 ⬍.0001 ⫺.6 .73
OCD CSR ⫺.2 .65 ⫺1.0 ⬍.0001 .1 .93
Child global severity and functioning
CGI-S ⫺.4 .53 ⫺1.0 ⬍.0001 .2 .45
CGAS 6.6 .21 7.8 ⬍.0001 ⫺3.4 .15
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Family accommodation ⫺6.3 .19 ⫺5.8 ⬍.0001 2.1 .22

This document is copyrighted by the American Psychological Association or one of its allied publishers.

Note. CY-BOCS ⫽ Child Yale-Brown Obsessive-Compulsive Scale; CSR ⫽ Clinical Severity Rating; CGI-
S ⫽ Clinical Global Impression-Severity Scale.

natural log (ln) of time were fixed effect predictors; the Group ⫻ and related disorders across the life span (Herbst et al., 2012;
ln(time) interaction was also included. A random intercept allows Himle et al., 2012; Storch et al., 2011; Wootton, 2016), and add to
individuals to vary in their mean outcome value. A random slope a broader literature documenting the potential role of new tech-
with respect to time was omitted because of the small sample size. nologies for meaningfully expanding the reach of supported men-
Table 3 presents model estimated means for all clinical outcomes, tal health care (Comer, 2015; Comer & Barlow, 2014; Comer et
by treatment condition, across the assessments points. Table 4 al., 2014; Kazdin & Blase, 2011). Taken together, the present
presents the results of analyses examining the effects of time, findings speak to the overall feasibility and acceptability of
condition, and their interactions on clinical outcomes. For every Internet-delivered FB-CBT, and to its preliminary efficacy.
clinical outcome there was a significant effect of time across
subjects (see Table 4). Across outcomes, there were no significant
interaction effects of Time ⫻ Condition. Figure 2 presents the Internet-delivered FB-CBT
symptom trajectory, by condition, for CY-BOCS scores; the gen-
Clinic-based FB-CBT
eral shape of change depicted is comparable for all outcomes.
Table 5 presents data on the clinical significance of outcomes 25
across time. Within-subjects effect sizes from pre-to-post and from
pre-to-follow-up were mostly large in magnitude. Between 60%
and 80% of participants showed clinically significant responder 20
statuses as assessed via the CGI-I (score of 1 or 2), the CY-BOCS
(“subclinical”/“mild” range; i.e., score ⱕ15), and the ADIS-C/P
(no OCD diagnoses) at posttreatment and at follow-up. Differences
CY-BOCS Score

between conditions on clinical significance responder statuses 15

were nonsignificant (see Table 5).

Discussion 10

The present findings provide the first empirical support from a

controlled trial evaluating the potential utility of leveraging VTC
technology to remotely deliver real-time treatment for early onset 5
OCD. Internet-delivered FB-CBT showed strong engagement and
satisfaction, with roughly 90% of Internet-treated youth complet-
ing a full course of treatment, mothers and therapists both report-
0
ing high therapeutic alliance, and mothers reporting very high 0 2 4 6 8 10
satisfaction with services received. At posttreatment, roughly three
fourths of Internet cases showed an “excellent response,” and at Months
follow-up four-fifths showed an “excellent response.” Differences
Figure 2. Trajectories of change across months, by treatment condition,
between treatment formats across time were not supported, and
for scores on the Children’s Yale-Brown Obsessive-Compulsive Scale
response rates among Internet-treated youth were also roughly (CY-BOCS). The average timing of treatment completion was 4.52 months
comparable to those found in previous work evaluating clinic- postbaseline assessment; the average timing of the follow-up evaluation
based FB-CBT (Freeman et al., 2008, 2014). These findings build was 9.79 months postbaseline assessment. The general quadratic shape of
on the small but growing body of work supporting the promising change depicted here for CY-BOCS scores is comparable for all other
role of VTC formats for broadening the delivery of CBT for OCD clinical outcomes assessed.
 184 COMER ET AL.

Table 5
Clinical Significance of Outcomes, by Treatment Condition

Within-subjects effect sizes (d) across clinical outcomes, by treatment condition

Internet-delivered
FB-CBT Clinic-based FB-CBT
Pre vs. Pre vs. Pre vs. Pre vs.
Outcome Post Follow-up Post Follow-up

Child OCD symptoms

CY-BOCS Total 1.53 1.27 1.64 1.45
OCD CSR 1.67 1.06 1.40 .80
Child global severity and functioning
CGI-S 1.59 1.16 1.09 .83
CGAS ⫺1.29 ⫺1.43 ⫺.73 ⫺.77
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Family accommodation 1.38 1.03 1.41 1.38

This document is copyrighted by the American Psychological Association or one of its allied publishers.

Between-subjects effect sizes (d) across clinical outcomes

Variable At Post At Follow-up

Child OCD symptoms

CY-BOCS Total .09 .12
OCD CSR .24 .00
Child global severity and functioning
CGI-S ⫺.06 ⫺.01
CGAS ⫺.06 .10
Family accommodation .56 .31

Responder statuses, by treatment condition

Posttreatment 6-month follow-up
Responder status Internet % Clinic % Significance test Internet % Clinic % Significance test

Excellent responder1
72.7 60.0 ␹ (1, N ⫽ 22) ⫽ .38, p ⫽ .66
2
80.0 66.7 ␹ (1, N ⫽ 17) ⫽ .43, p ⫽ .63
2

“Sub-clinical”/“mild” OCD severity2 72.7 60.0 ␹2(1, N ⫽ 22) ⫽ .38, p ⫽ .54 55.6 75.0 ␹2(1, N ⫽ 17) ⫽ .70, p ⫽ .40
No OCD diagnosis3 63.3 60.0 ␹2(1, N ⫽ 22) ⫽ .03, p ⫽ .86 70.0 77.8 ␹2(1, N ⫽ 17) ⫽ .15, p ⫽ .70
Note. FB-CBT ⫽ Family-Based Cognitive-Behavioral Therapy (Freeman & Garcia, 2009); CY-BOCS ⫽ Child Yale-Brown Obsessive-Compulsive
Scale; CSR ⫽ Clinical Severity Rating; CGI-S ⫽ Clinical Global Impression-Severity Scale; CGI-I ⫽ Clinical Global Impression-Improvement Scale
1
Score of 1 (“very much improved”) or 2 (“much improved”) on CGI-I. 2 ⱕ 15 on CY-BOCS. 3 As determined via the ADIS-IV-C/P.

Although dissemination and implementation efforts have al- delivering expert services for severe conditions that may not
ready had an appreciable impact on mental health services, given readily lend themselves to traditional dissemination methods.
the great diversity of mental health problems, broad training efforts Several limitations merit comment. First, the sample size in this
of regional generalist providers alone may not be sufficient to pilot RCT may have been underpowered to detect some differ-
adequately address the tremendous prevalence and burden of child ences, although all within-group analyses did yield significant
disorders. As Comer and Barlow (2014) note, the considerable results. Future work with larger samples is needed to replicate
resources required for quality dissemination and implementation these promising results and to evaluate mechanisms of treatment
may preclude large-scale competency training in the treatment of response, including potential mediators and moderators. Second,
low base-rate disorders, such as pediatric OCD, in order to prior- the present sample was predominantly urban, Caucasian, and of
itize the most common conditions affecting the general population. relatively high socioeconomic advantage. Future research
Moreover, treatment innovations that are too complex do not get must improve efforts to recruit affected families from more
routinely incorporated (or incorporated with fidelity) into frontline diverse backgrounds. Third, participants were recruited from a
practice (Rogers, 2003). Given the broad diversity of training and metropolitan-based clinic and as such findings may not generalize
educational backgrounds across the mental health workforce, spe- to more remote communities with limited access to computers,
cialized/complex treatment methods for low base-rate disorders technological literacy, and/or clinic-based services. Indeed, the
(such as E/RP for pediatric OCD) may not readily lend themselves present findings may speak to the generic potential of Internet-
to broad dissemination and implementation efforts. Indeed, “put- delivered treatments, particularly with regard to issues of ecolog-
ting all of our eggs in the dissemination basket” and in the broad ical validity associated with treating families in their homes and
training of a generalist mental health workforce (see Comer & other natural settings, but cannot speak to whether VTC methods
Barlow, 2014) may not adequately ensure appropriate and acces- can effectively treat rural youth, economically depressed youth, or
sible care for young children with low base-rate problems such as other populations underserved by quality clinic-based services.
OCD that require complex treatments. With continued support, Encouragingly, as of 2015, over 50% of rural households and over
VTC-delivered E/RP may prove to be a useful solution for broadly 60% of households earning $20,000 –$40,000 already have house-
 INTERNET-DELIVERED EARLY OCD TREATMENT 185

hold broadband Internet connections (Pew Research Center, 2015). Comer, J. S., Furr, J. M., Cooper-Vince, C. E., Kerns, C. E., Chan, P. T.,
As we approach household Internet connectivity for all, regardless Edson, A. L., . . . Freeman, J. B. (2014). Internet-delivered, family-based
of geography and income, proof-of-concept trials such as this are treatment for early-onset OCD: A preliminary case series. Journal of
essential to evaluating the preliminary merits of Internet-delivered Clinical Child and Adolescent Psychology, 43, 74 – 87. http://dx.doi.org/
treatments, but future efforts are needed to examine treatment 10.1080/15374416.2013.855127
Comer, J. S., Furr, J. M., Cooper-Vince, C., Madigan, R. J., Chow, C.,
responses in families living in underserved regions.
Chan, P., . . . Eyberg, S. M. (2015). Rationale and considerations for the
Moreover, comparing two active treatments in the absence of a
Internet-based delivery of parent-child interaction therapy. Cognitive
“no treatment” condition precludes consideration of the extent to and Behavioral Practice, 22, 302–316. http://dx.doi.org/10.1016/j.cbpra
which effects may simply reflect the passage of time. However, .2014.07.003
previous pediatric OCD trials utilizing waitlist controls (e.g., Bar- Comer, J. S., & Kendall, P. C. (2004). A symptom-level examination of
rett et al., 2004) suggest that the simple passage of time does not parent-child agreement in the diagnosis of anxious youths. Journal of the
yield significant improvements. In addition, the present study American Academy of Child & Adolescent Psychiatry, 43, 878 – 886.
evaluated remote delivery of a treatment protocol designed to treat http://dx.doi.org/10.1097/01.chi.0000125092.35109.c5
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

early child OCD, and as such we only included youth ages 4 to 8. Crum, K. I., & Comer, J. S. (2016). Using synchronous videoconferencing
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Future research should examine treatment outcomes across a wider to deliver family-based mental healthcare. Journal of Child and Adoles-
age range of children with OCD, and to consider whether Internet- cent Psychopharmacology, 26, 229 –234. http://dx.doi.org/10.1089/cap
delivered treatment is comparably effective across different age .2015.0012
groups. Finally, it is possible that the in-session interactive com- Duncan, A. B., Velasquez, S. E., & Nelson, E. L. (2014). Using videocon-
puter games (designed for Internet-delivered FB-CBT to enhance ferencing to provide psychological services to rural children and ado-
lescents: A review and case example. Journal of Clinical Child and
children’s understanding of treatment concepts in the way in-office
Adolescent Psychology, 43, 115–127. http://dx.doi.org/10.1080/
collaborative art projects do in clinic-based treatment) resulted in
15374416.2013.836452
differential treatment dosages across conditions. Flessner, C. A., Sapyta, J., Garcia, A., Freeman, J. B., Franklin, M. E., Foa,
Despite limitations, the present study offers the first randomized E., & March, J. (2009). Examining the psychometric properties of the
evaluation of VTC methods for providing treatment to young Family Accommodation Scale-Parent-Report (FAS-PR). Journal of Psy-
children with early onset OCD, and provides support for the chopathology and Behavioral Assessment, 31, 38 – 46.
acceptability and preliminary efficacy of Internet-delivered FB- Freeman, J. B., & Garcia, A. M. (2009). Family based treatment for young
CBT. Amid a backdrop of workforce shortages in quality mental children with OCD: Therapist guide. New York, NY: Oxford University
health care and regional limitations in pediatric OCD expertise, the Press.
present findings suggest VTC methods may be a promising vehicle Freeman, J. B., Garcia, A. M., Coyne, L., Ale, C., Przeworski, A., Himle,
for meaningfully improving the accessibility of needed care. Cur- M., . . . Leonard, H. L. (2008). Early childhood OCD: Preliminary
rently, expert care in pediatric OCD tends to cluster around major findings from a family-based cognitive-behavioral approach. Journal of
metropolitan regions and academic hubs. As Comer and Barlow the American Academy of Child & Adolescent Psychiatry, 47, 593– 602.
http://dx.doi.org/10.1097/CHI.0b013e31816765f9
(2014) note, in the coming years specialty mental health “clinics”
Freeman, J., Sapyta, J., Garcia, A., Compton, S., Khanna, M., Flessner, C.,
may be housed online, rather than geographically bound, and
. . . Franklin, M. (2014). Family-based treatment of early childhood
systematically deliver quality treatments for low base-rate disor- obsessive-compulsive disorder: The Pediatric Obsessive-Compulsive
ders that are not easily disseminated. Specialty OCD centers may Disorder Treatment Study for Young Children (POTS Jr)—a random-
do well to incorporate VTC strategies to broaden their patient ized clinical trial. Journal of the American Medical Association Psychi-
catchment areas and extend evidence-based care to a greater pro- atry, 71, 689 – 698. http://dx.doi.org/10.1001/jamapsychiatry.2014.170
portion of affected families. Guy, W., & Bonato, R. R. (Eds.). (1970). Clinical global impressions.
Chevy Chase, MD: NIMH.
Herbst, N., Voderholzer, U., Stelzer, N., Knaevelsrud, C., Hertenstein, E.,
References Schlegl, S., . . . Külz, A. K. (2012). The potential of telemental health
applications for obsessive-compulsive disorder. Clinical Psychology Re-
Barrett, P., Healy-Farrell, L., & March, J. S. (2004). Cognitive-behavioral
family treatment of childhood obsessive-compulsive disorder: A con- view, 32, 454 – 466. http://dx.doi.org/10.1016/j.cpr.2012.04.005
trolled trial. Journal of the American Academy of Child & Adolescent Himle, M. B., Freitag, M., Walther, M., Franklin, S. A., Ely, L., & Woods,
Psychiatry, 43, 46 – 62. http://dx.doi.org/10.1097/00004583-200401000- D. W. (2012). A randomized pilot trial comparing videoconference
00014 versus face-to-face delivery of behavior therapy for childhood tic dis-
Caporino, N. E., Morgan, J., Beckstead, J., Phares, V., Murphy, T. K., & orders. Behaviour Research and Therapy, 50, 565–570. http://dx.doi
Storch, E. A. (2012). A structural equation analysis of family accom- .org/10.1016/j.brat.2012.05.009
modation in pediatric obsessive-compulsive disorder. Journal of Abnor- Horvath, A. O. (1994). Empirical validation of Bordin’s pan theoretical
mal Child Psychology, 40, 133–143. http://dx.doi.org/10.1007/s10802- model of the alliance: The Working Alliance Inventory perspective. In
011-9549-8 A. O. Horvath & L. S. Greenberg (Eds.), The working alliance: Theory,
Comer, J. S. (2015). Introduction to the special section: Applying new research and practice (pp. 109 –130). New York, NY: Wiley.
technologies to extend the scope and accessibility of mental health care. Jones, D. J., Forehand, R., Cuellar, J., Parent, J., Honeycutt, A., Khavjou,
Cognitive and Behavioral Practice, 22, 253–257. http://dx.doi.org/10 O., . . . Newey, G. A. (2014). Technology-enhanced program for child
.1016/j.cbpra.2015.04.002 disruptive behavior disorders: Development and pilot randomized con-
Comer, J. S., & Barlow, D. H. (2014). The occasional case against broad trol trial. Journal of Clinical Child and Adolescent Psychology, 43,
dissemination and implementation: Retaining a role for specialty care in 88 –101. http://dx.doi.org/10.1080/15374416.2013.822308
the delivery of psychological treatments. American Psychologist, 69, Kazdin, A. E., & Blase, S. L. (2011). Rebooting psychotherapy research
1–18. http://dx.doi.org/10.1037/a0033582 and practice to reduce the burden of mental illness. Perspectives
 186 COMER ET AL.

on Psychological Science, 6, 21–37. http://dx.doi.org/10.1177/ Rogers, E. M. (2003). Diffusion of innovations (5th ed.). New York, NY:
1745691610393527 Free Press.
Kramer, G. M., Kinn, J. T., & Mishkind, M. C. (2015). Legal, regulatory, Scahill, L., Riddle, M. A., McSwiggin-Hardin, M., Ort, S. I., King, R. A.,
and risk management issues in the use of technology to delivery mental Goodman, W. K., . . . Leckman, J. F. (1997). Children’s Yale-Brown
health care. Cognitive and Behavioral Practice, 22, 258 –268. http://dx Obsessive Compulsive Scale: Reliability and validity. Journal of the
.doi.org/10.1016/j.cbpra.2014.04.008 American Academy of Child & Adolescent Psychiatry, 36, 844 – 852.
Larsen, D. L., Attkisson, C. C., Hargreaves, W. A., & Nguyen, T. D. http://dx.doi.org/10.1097/00004583-199706000-00023
(1979). Assessment of client/patient satisfaction: Development of a Shaffer, D., Gould, M. S., Brasic, J., Ambrosini, P., Fisher, P., Bird, H., &
general scale. Evaluation and Program Planning, 2, 197–207. http://dx Aluwahlia, S. (1983). A Children’s Global Assessment Scale (CGAS).
.doi.org/10.1016/0149-7189(79)90094-6 Archives of General Psychiatry, 40, 1228 –1231. http://dx.doi.org/10
Nelson, E. L., & Patton, S. (2016). Using videoconferencing to deliver .1001/archpsyc.1983.01790100074010
individual therapy and pediatric psychology interventions with children Silverman, W. K., & Albano, A. M. (1996). The Anxiety Disorders Inter-
and adolescents. Journal of Child and Adolescent Psychopharmacology, view Schedule for Children for DSM–IV: Child and parent versions. San
26, 212–220. http://dx.doi.org/10.1089/cap.2015.0021 Antonio, TX: Psychological Corporation.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Pediatric OCD Treatment Study (POTS) Team. (2004). Cognitive-behavior Storch, E. A., Caporino, N. E., Morgan, J. R., Lewin, A. B., Rojas, A.,
Brauer, L., . . . Murphy, T. K. (2011). Preliminary investigation of
This document is copyrighted by the American Psychological Association or one of its allied publishers.

therapy, sertraline, and their combination for children and adolescents

with obsessive-compulsive disorder: The Pediatric OCD Treatment web-camera delivered cognitive-behavioral therapy for youth with
Study (POTS) randomized controlled trial. Journal of the American obsessive-compulsive disorder. Psychiatry Research, 189, 407– 412.
Medical Association, 292, 1969 –1976. http://dx.doi.org/10.1001/jama http://dx.doi.org/10.1016/j.psychres.2011.05.047
.292.16.1969 Wootton, B. M. (2016). Remote cognitive-behavior therapy for obsessive-
Pew Research Center. (2015). Home broadband 2015. Retrieved from compulsive symptoms: A meta-analysis. Clinical Psychology Review,
http://www.pewinternet.org/2015/12/21/2015/Home-Broadband-2015/ 43, 103–113. http://dx.doi.org/10.1016/j.cpr.2015.10.001
Piacentini, J., Bergman, R. L., Jacobs, C., McCracken, J. T., & Kretchman,
J. (2002). Open trial of cognitive behavior therapy for childhood Received January 12, 2016
obsessive-compulsive disorder. Journal of Anxiety Disorders, 16, 207– Revision received August 20, 2016
219. http://dx.doi.org/10.1016/S0887-6185(02)00096-8 Accepted August 25, 2016 䡲

Additional Journal Information

Copyright and Permission: Those who wish to reuse APA- Reprints: Authors may order reprints of their articles from the
copyrighted material in a non-APA publication must secure from printer when they receive proofs.
APA written permission to reproduce a journal article in full or Single Issues, Back Issues, and Back Volumes: For information
journal text of more than 800 cumulative words or more than 3 tables regarding single issues, back issues, or back volumes, visit www
and/or figures. APA normally grants permission contingent on per- .apa.org/pubs/journals/subscriptions.aspx or write to Order De-
mission of the author, inclusion of the APA copyright notice on the partment, American Psychological Association, 750 First Street,
first page of reproduced material, and payment of a fee of $25 per NE, Washington, DC 20002-4242; call 202-336-5600 or 800-374-
page. Libraries are permitted to photocopy beyond the limits of U.S. 2721.
copyright law: (1) post-1977 articles, provided the per-copy fee in the
code for this journal (0022-006X/17/$12.00) is paid through the Subscription Claims Information: A claim form to assist mem-
Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA bers, institutions, and nonmember individuals who have a problem
01923; (2) pre-1978 articles, provided that the per-copy fee stated in with their subscription is available at http://forms.apa.org/
the Publishers’ Fee List is paid through the CopyrightClearance journals/subclaim/; or call 1-800-374-2721.
Center. Formore information along with a permission request form go Change of Address: Send change of address notice and a recent
to: www.apa.org/about/contact/copyright/index.aspx mailing label to the attention of Subscriptions Department,
APA, 30 days prior to the actual change of address. APA will not
Disclaimer: APA and the Editors of the Journal of Consulting and replace undelivered copies resulting from address changes; jour-
Clinical Psychology® assume no responsibility for statements and nals will be forwarded only if subscribers notify the local post
opinions advanced by the authors of its articles. office in writing that they will guarantee periodicals forwarding
postage.
Electronic access: Individuals subscribers to this journal
have automatic access to all issues of the journal in APA Journal Staff: Rosemarie Sokol-Chang, PhD, Publisher,
thePsycARTICLES® full-text database. See http://www.apa.org/ APA Journals; Shelby E. Jenkins, Journal Production Manager;
pubs/journals.ccp and click on View Table of Contents. Jodi Ashcraft, Director, Advertising Sales and Exhibits.