Seeing Is Believing: Using Skypeto Improve Diabetes Outcomesin YouthDiabetes Care 2015;38:1427–1434 | DOI: 10.2337/dc14-2469

OBJECTIVE

The objective of this study was to compare the relative effectiveness of twomodes of delivering Behavioral Family Systems Therapy for Diabetes (BFST-D)to improve adherence and glycemic control among adolescents with type 1 di-abeteswith suboptimal glycemic control (HbA1c‡9.0% [‡74.9mmol/mol]): face toface in clinic (Clinic) and Internet videoconferencing (Skype) conditions.

RESEARCH DESIGN AND METHODS

Adolescents aged 12 to 18 years and at least one adult caregiver were randomizedto receive BFST-D via the Clinic or Skype condition. Participants completed up to10 therapy sessions within a 12-week period. Changes in youth- and parent-reported adherence and glycemic control were compared before and after theintervention and at follow-up assessment.

RESULTS

Using an intent-to-treat analytic approach, no significant between-group differ-ences were identified between the before, after, and follow-up assessments.Groups were collapsed to examine the overall effects of BFST-D on adherenceand glycemic control. Results identified that statistically significant improvementsin adherence and glycemic control occurred from before to after the intervention;improvements were maintained at 3-month follow-up.

CONCLUSIONS

Delivery of BFST-D via Internet-based videoconferencing is viable for addressingnonadherence and suboptimal glycemic control in adolescents with type 1 diabe-tes, potentially reducing important barriers to care for youth and families.

The intensivemanagement of type 1 diabetes has beenwell established as critical tooptimizing long-term health outcomes (1). Modern diabetes care is difficult toaccomplish, however, because it is often complex and demanding for youth andfamilies. Achieving optimal management of type 1 diabetes is difficult at any age (2)but particularly during adolescence, as youth assume increasing responsibility fortheir care (3–5). While numerous factors likely contribute to adherence difficultiesduring this critical developmental period, family functioning is an important pre-dictor of adherence and glycemic control (6–8). Specifically, family conflict, parent–adolescent communication, and family problem-solving have been associated withdiabetes outcomes during adolescence (9,10).Given the role of family interactions, interventions to address family functioning

during this developmental period have been well tested. Behavioral Family Systems

1Harold Schnitzer Diabetes Health Center,Portland, OR2Institute on Development & Disability, Divisionof Psychology, Department of Pediatrics, OregonHealth & Science University, Portland, OR

Corresponding author: Michael A. Harris,harrismi@ohsu.edu.

Received 23 October 2014 and accepted 3 May2015.

Clinical trial reg. no. NCT02274103, clinicaltrials.gov.

© 2015 by the American Diabetes Association.Readers may use this article as long as the workis properly cited, the use is educational and notfor profit, and the work is not altered.

See accompanying article, p. 1435.

Michael A. Harris,1,2 Kurt A. Freeman,2 and

Danny C. Duke1,2

Diabetes Care Volume 38, August 2015 1427

CLIN

CARE/ED

UCATIO

N/N

UTR

ITION/PSYC

HOSO

CIAL

Therapy (BFST) is a well-supported inter-vention designed to improve family func-tioning and adherence in youth withdiabetes. BFST is a structured,manualizedintervention that includes four primarycomponents: problem-solving, communi-cation skills, cognitive restructuring, andfamily systems interventions. BFST im-proves family communication andproblem-solving when compared withstandard medical care (11–13). A modi-fied version, BFST for Diabetes (BFST-D),which specifically targets optimizing di-abetes care, significantly improves glyce-mic control, treatment adherence, anddiabetes-related family conflict amongadolescents with type 1 diabetes (14,15).Further, a home-based version of BFST-Dhas resulted in statistically and clinicallymeaningful improvements in family func-tioning and glycemic control (16,17), thusdemonstrating that modified deliverymethods can be used to improve accessto care.While growing evidence supports the

benefits of interventions such as BFST-Dto assist with adherence in pediatricchronic health conditions, access towell-trained providers remains a barrierto implementation. Specialty behavioralhealth care often is concentrated inlarge urban areas and associated withuniversity clinics and/or academic med-ical health centers. Rural settings pres-ent particular challenges to accessingbehavioral health services because ofshortages of providers. While primarycare providers in rural settings often at-tempt to address their patients’ behav-ioral health needs, limitations in time,expertise, and reimbursement for suchservices (18,19) are barriers to address-ing the often complex behavioral healthand family dynamic needs of youth withpoorly controlled diabetes. Accessibilitylimitations in rural and underservedcommunities often require families totravel considerable distances to receiveservices or forgo mental health servicesaltogether.Although effective, home-based de-

livery of BFST-D is unlikely to be a prac-tical or sustainable solution to reducingbarriers to care, particularly for patientsin rural areas. Thus additional methodsof delivering evidence-based treatmentare needed. The use of technology todelivery behavioral health care is onemechanism for increasing access toservices.

Telemental health care, or the deliveryof psychological or behavioral health carevia communication networks, has beendeployed to overcome a shortage ofand/or uneven distribution of behavioralhealth resources and infrastructure (20).Importantly, recent technological ad-vancements such as the increasing avail-ability and rapid growth in the adoptionof household Internet among low-incomeand rural families (21,22) has createdopportunities to provide services usingtelemental health. In particular, Inter-net-based videoconferencing may be aneffective method of overcoming ob-stacles to care by increasing the availabil-ity of expert or specialized services toaddress regional shortages of qualifiedproviders (20,23,24). Use of Internet-based videoconferencing platforms offersseveral pragmatic advantages, includingbeing available at no or low cost to fam-ilies, allowing for real-time audio and vi-sual interaction between patients andproviders, and facilitating delivery of treat-ment in the family home, which decreasethe direct costs and time associated withtreatment.

The use of telemental health is receiv-ing increased attention (25,26). Moststudies, however, have been prelimi-nary or descriptions of feasibility andusage rather than well-controlled ran-domized clinical trials (25). While inter-ventions that use Internet technologyare beginning to include youth with di-abetes (26,27), to date reports have fo-cused exclusively on feasibility andenrollment outcomes. Well-controlledrandomized clinical trials examiningthe effectiveness of behavioral healthinterventions using teleconferencingtechnology for youth with poorly con-trolled diabetes (i.e., HbA1c $9.0%[$74.9 mmol/mol]) are absent in thepublished literature.

We report the results of a clinical trialcomparing the delivery of BFST-D to ado-lescents with poorly controlled diabetesand their caregiver(s) randomized to re-ceive the intervention via conventionalconditions (Clinic) or videoconference(Skype). Because BFST-D has been shownto improve family functioning, adher-ence, and glycemic control, our primaryinterest was whether outcomes woulddiffer depending on delivery method.Thus, our first hypothesis was that BFST-Ddelivered via videoconferencing (Skype)would yield outcomes that were not

significantly different than conventionalclinic-based delivery. Second, regardlessof condition, we hypothesized that BFST-Dwould significantly improve regimen ad-herence and glycemic control in youthwith poorly controlled type 1 diabetes.

RESEARCH DESIGN AND METHODS

ParticipantsInclusion criteria at enrollmentrequired a diagnosis of type 1 diabetesof at least 1 year of duration, age 12 to 19years, and suboptimal glycemic control.Suboptimal glycemic control was opera-tionalized as HbA1c $9.0% ($74.9mmol/mol). Adolescents needed to re-side with and be accompanied by a pri-mary caretaker, intend to reside with theprimary caretaker for the duration of thestudy (7 months), have no history ofmental retardation or other mentalhealth condition that would precludethe completion of study measures, andnot have an uncontrolled medical condi-tion (e.g., cystic fibrosis) that wouldconfound assessment of adherence todiabetes care recommendations.

A total of 138 individuals and familieswere approached; 9 youth were deter-mined ineligible and 39 declined partic-ipation. Thus, 90 youth and at least onelegal guardian completed the baselineassessment. No knownmortality threatswere identified as a result of attrition.Total dropout rate was 21%. Figure 1summarizes participants’ progressthrough the trial recruitment and as-sessment stages in the format recom-mended by the Consolidated Standardsof Reporting Trials (28).

The mean age of participating youthwas 15.04 years (SD 1.79 years), and55% were male (Table 1). Most care-givers were mothers (76.7%), while21.1% were fathers and 2.2% grand-mothers. Forty percent of youth werereported to reside with both parents.The sample included the following pro-portion of families in each Hollingsheadsocioeconomic stratum: lower, 5.6%;lower middle, 8.0%; middle, 41.0%; up-per middle, 37.4%; and upper, 8.0%. Theracial/ethnic distribution of youth partic-ipants was 87.8% white, 4.4% Hispanic/Latin American, 1.1% Native American,1.1% Hawaiian/Pacific Islander, and 5.6%reported bi- or multiethnic. Youth andcaregivers who declined to participate(n = 39) did not significantly differ fromparticipants on age, duration of diabetes,

1428 Skype for Parent–Teen Diabetes Conflict Diabetes Care Volume 38, August 2015

socioeconomic status, sex, race/ethnicity,family composition, and parental maritalstatus. Further, youth and caregivers whodid not complete outcome assessmentsat all assessment points did not signifi-cantly differ on the same demographicvariables from those who completed theoutcome assessment at all points.

Measures

Demographic Information

A demographic information form wascompleted by caregivers to gather med-ical information and the demographic

data necessary to calculate the Hollings-head index of social status (29). Youthand/or families who declined to partici-pate completed a modified demo-graphic information form to allow forcomparison with participants.

Measurement of Adherence

The Diabetes Self-Management Profile(DSMP) (30) is a 25-item semistructuredinterview with a possible score of 84;higher total scores indicate better adher-ence. Administration takes approximately15–20min. Questions assess five areas of

diabetes management, including insulinadministration and dose adjustment,blood glucose monitoring, exercise, diet,and hypo- and hyperglycemia manage-ment. Analogous caregiver and youthforms were used in this study. Previousresearch (30) demonstrated good psy-chometric properties for the DSMP. Forthe current study, internal consistencywas acceptable and consistent across as-sessments for youthmeasures (before in-tervention: a = 0.75; after intervention:a = 0.81; follow-up, a = 0.76) and care-giver (before intervention,a = 0.81; afterintervention: a = 0.82; follow-up: a =0.84).

Measure of Glycemic Control

Glycemic control was evaluated via anHbA1c assay using a Seimens/BayerDCA 2000 analyzer, which provided anestimate of blood glucose concentra-tions over the preceding 2–3 months(31). Glycemic control was estimatedby an HbA1c assay conducted concurrentwith (or within 4 weeks of) the before,after, and follow-up assessments. Theequivalence of DCA 2000 and referencelaboratory measurements has been es-tablished previously (32).

ProceduresPrimary recruitment (n = 85) occurredduring regularly scheduled visits to atertiary diabetes clinic at a large teach-ing hospital located in the U.S. PacificNorthwest. Several participants (n = 5)were identified via flyers and recom-mendations provided by other diabetesclinics at local hospitals and were re-ferred to the primary recruitment cen-ter to confirm eligibility and completeconsent/assent procedures. Participants’HbA1c from their most recent clinic visitbefore recruitment was used to pre-screen for potential eligibility. Potentialparticipants whowere contacted throughflyers placed at other regional hospital-based diabetes centers were screenedfor major inclusion criteria before theirstudy intake visit.

Informed consent was obtained fol-lowing institutional review board guide-lines, and caregivers(s) and adolescentssigned institutionally approved in-formed consent and assent forms be-fore participation. After obtainingconsent/assent, participants were ran-domized to either Skype (n = 46) or Clinic(n = 44) treatment conditions using ablock design. Participants in each

Figure 1—Consolidated Standards of Reporting Trials statement of youth recruitment.

Table 1—Adolescent participant demographic information

Clinic Skype

Sex, n (%)Female 18 (51.4) 17 (48.6)Male 26 (47.3) 29 (52.7)

Age (years) 15.04 (1.79) 14.94 (1.77)

Diabetes duration 6.51 (3.24) 6.56 (3.77)

Baseline HbA1c, % (mmol/mol)* 11.10 (2.14) (97.8 [18.86]) 11.15 (1.73) (98.4 [15.27])

Hollingshead index 37.02 (11.90) 37.64 (9.93)

Data are mean (SD) unless otherwise noted. *Before the intervention.

care.diabetesjournals.org Harris and Associates 1429

condition received up to 10 sessions ofBFST-D that lasted 1–1.5 h each. Thera-pists and families communicated to estab-lish regular meeting times, and a researchassistant provided reminder calls for up-coming appointments approximately 2days before the appointment. When par-ticipantsmissed appointments, a researchassistant attempted to reschedule untilthe appointment occurred or until a 3-month time window for receiving treat-ment elapsed. All participants continuedto receive regular medical care by theirdiabetes care providers during the courseof the study.Data on adherence and glycemic con-

trol were gathered at three time points:pre- (within 4 weeks preceding the firsttreatment session), post- (within 4 weeksof either the last treatment session or atthe end of the 3-monthwindow for treat-ment), and at follow-up (3 months aftertreatment). Blood samples necessary forconducting HbA1c assays were collectedconcurrent with questionnaire adminis-tration and as a normal part of the partic-ipants’ routine clinic visits, when possible.Trained research personnel unfamiliarwith the participants completed all inter-views and administered all standardizedoutcomemeasures, with the exception ofHbA1c, which was measured by clinicnursing staff.

Participation Incentives, Travel

Reimbursement, and Equipment

Participants were providedmonetary in-centives to promote completion ofstudy-related tasks. Youth received $90for each evaluation completed. Eachfamily received $20 per treatment ses-sion completed, regardless of treatmentcondition. Families who traveled morethan 50miles round trip for study-relatedpurposes were eligible to receive $0.55per mile up to a maximum of $250 overthe course of the study. Families random-ized to the Skype condition were loaned alaptop computer with a webcam andmobile hotspot (or funds to pay for In-ternet access), if needed. Families werefree to use their personal home com-puter and Internet access rather thanservices or equipment supplied by thestudy, if preferred.

TreatmentBFST-Dwas delivered by clinicianswith amaster’s or doctoral psychology de-gree trained in the intervention and su-pervised by the first author (M.A.H.);

each clinician provided treatment toparticipants in both conditions. BFST-Dconsisted of four primary components:training in problem-solving, communi-cation training, cognitive restructuring(dispelling strong beliefs), and familytherapy approaches. Problem-solvingincluded training families to use a for-mal problem-solving approach with dis-crete steps consisting of defining theproblem; generating solutions; groupdecision making; planning, implement-ing, and monitoring chosen solution(s);and revisiting, renegotiating, and refin-ing ineffective solutions. Communica-tion training included instructions,feedback, modeling, and rehearsal of ap-proaches to address common parent/ad-olescent communicationerrors. Cognitiverestructuring included addressing and al-tering family members’ irrational beliefs,attitudes, and attributions that could im-pede effective parent–adolescent inter-actions. Family therapy interventionsincluded both functional and structuralapproaches to target maladaptive or in-effective family system patterns andcharacteristics (e.g., weak parental co-alitions and cross-generational coali-tions) that could impede effectiveproblem-solving and communication.

Families received an intervention tai-lored to meet their needs as determinedby baseline assessment information andongoing observations and interactionsduring treatment. Sessions consisted ofdiscussions about family problem-solvingand conflict resolution. Therapists weredirective and active, providing instruc-tions and feedback, and modeling andguiding the rehearsal of skills. Behavioralassignments were provided at each ses-sion, and progress was reviewed duringthe next session. Each session includeddelivery of didactic information and em-phasized teaching the family to apply in-dependently the targeted skills in thereal-life/home setting. A focus on diabe-tes care was integrated throughout (e.g.,problem-solving completion of blood glu-cose checks, effective communication re-garding difficulties with adherence).

Data Management and StatisticalAnalysis PlanData were entered into a computerizeddata file located on a secure local net-work. Data were checked for normality,outliers, or data entry errors beforeanalyses. An intent-to-treat analytic

approach was used. Multiple imputa-tion procedures using pooled means(five) were used to estimate and re-place missing data (33–35). Initial sta-tistical analyses compared differences indemographic and outcome variablesamong treatment groups at baseline. As-sociations among variables were exam-ined to identify potential covariates thatmay differ by treatment condition.Repeated-measuresANOVAwasconductedto examinewithin-subject changes acrossassessments (before, after, and follow-up) and between-subject effects foreach treatment group (Clinic and Skype).Effect sizes were estimated by calculatingCohen d.

RESULTS

No statistically significant between-groupdifferences at baselinewere identified forpatient age, duration of diabetes, socio-economic status, sex, race/ethnicity, fam-ily composition, or parentalmarital status(Table 1). The groups did not differ signif-icantly at baseline with respect to HbA1cor DSMP scores (parents and youth).

Hypothesis 1: Between-GroupDifferencesA Levene test of equality of error vari-ances was conducted for measures of ad-herence and glycemic control and wasnonsignificant for group assignment(Clinic and Skype) across the three assess-ments (before, after, and follow-up),which suggested group variability wasequal. A repeated-measures ANOVAdemonstrated no between-group differ-ences in treatment effects for adherenceand glycemic control (F(1) = 0.09; P =0.77) (Table 2). In addition, groups didnot significantly differ on the mean num-ber (SD) of sessions completed (Clinic =6.82 [3.39], Skype = 5.84 [3.25]). Themean number of BFST-D sessions com-pleted was significantly associated withHbA1c after therapy (r = 20.37; P ,0.001) and at the follow-up assessment(r = 20.35; P , 0.001). Dropout rateswere significantly different across treat-ment conditions (Clinic: n = 9; Skype: n =17), with a significantly greater dropoutrate among those in the Skype condition(x2(1) = 10.88; P , 0.001; Fig. 1).

Hypothesis 2: Total Sample AnalysesBecause effects of treatment on adher-ence and glycemic control did not differbased on treatment delivery modality(i.e., Clinic versus Skype), we collapsed

1430 Skype for Parent–Teen Diabetes Conflict Diabetes Care Volume 38, August 2015

groups to test our secondary hy-potheses that BFST-D would result inimproved adherence and glycemic con-trol. First, a within-subjects repeated-measures ANOVA was conducted tocompare the effects of BFST-D on parent-reported adherence (P-DSMP) acrossthe before, after, and follow-up assess-ments. The Mauchly test indicated thatthe assumption of sphericity was notviolated (x2(2) = 3.50; P = 0.17). A sig-nificant main effect for P-DSMP acrossassessments was identified (F(2) =18.96; P , 0.001; h2

p = 0.176). Giventhe significant main effect, pairwisecomparisons were conducted that iden-tified significant differences (P, 0.001)in P-DSMP scores from before (mean[SD] 47.56 [12.78]) to after the inter-vention (53.22 [12.64]) (d = 0.45), andfrom the assessment before the inter-vention (47.56 [12.78]) to the follow-upassessment (50.94 [12.38]) (d = 0.27). Inaddition, significant differences (P =0.009) were identified from the assess-ment after the intervention (53.22[12.64]) to the follow-up assessment(50.94 [12.38]) (d = 0.18). Results indi-cated that significant improvementsoccurred in parent-reported adherence(P-DSMP scores) from before to afterthe intervention and that a portion ofthe improvements were lost at follow-up; however, significant improvementswere maintained at 3-month follow-upcompared with baseline (Fig. 2).Next, a repeated-measures ANOVAwas

conducted to compare the effects ofBFST-D on youth-reported adherence(Y-DSMP) across the assessments beforeand after and at follow-up. The Mauchlytest indicated that the assumption of sphe-ricity was not violated (x2(2) = 1.27; P =0.53). A significant main effect for Y-DSMPacross assessments was identified (F(2) =13.26; P , 0.001; h2

p = 0.131). Pairwisecomparisons identified significant differ-ences (P, 0.001) in Y-DSMP scores fromthe assessment before (52.23 [11.03]) to

that after (57.06 [11.10]) (d = 0.44), andfromtheassessmentbefore (52.23 [11.03])to that at follow-up (56.57 [11.30]) (d =0.39). Results indicate that significant im-provements in Y-DSMP scores occurredfrom before to after the intervention andthat the improvements were maintainedat the 3-month follow-up (Fig. 2).

Finally, a repeated-measures ANOVAwas conducted to compare the effectsof BFST-D on glycemic control (HbA1c)across assessments before, after, andat follow-up. A Mauchly test indicatedthat the assumption of sphericity wasnot violated (x2(2) = 4.21; P = 0.12). Asignificant main effect for HbA1c acrossassessments was identified (F(2) =12.95; P, 0.001; h2

p = 0.127). Pairwisecomparisons identified significant dif-ferences (P , 0.01) in HbA1c scoresfrom the assessment before (11.13[1.71]) to that after (10.42 [1.85]) (d =0.40), and from the assessment before(11.13 [1.71]) to that at follow-up (10.46[1.88]) (d = 0.37). Results indicate that

significant improvements in glycemiccontrol (HbA1c) occurred from beforeto after the intervention and that im-provements were maintained at the3-month follow-up (Fig. 3).

CONCLUSIONS

These results demonstrate that BFST-Dcan be effective when delivered via con-ventional methods (i.e., Clinic) or via In-ternet-based videoconferencing (i.e.,Skype). Regardless of method, signifi-cant improvements in youth- and parent-reported adherence and glycemiccontrol occurred. Further, there wasno difference in treatment effect acrossconditions. Although effect sizes wererelatively small, previous research sug-gests small improvements have impor-tant positive health implications andmay reduce risk of future complications(36). Our outcomes suggest the inter-vention achieved an overall margin ofsuperiority (e.g., the magnitude of dif-ference has practical importance),

Table 2—Means (SDs) across assessments

Group

Skype (n = 46) Clinic (n = 44)

Before intervention After intervention Follow-up Before intervention After intervention Follow-up

HbA1c% 11.15 (1.73) 10.40 (1.66) 10.61 (1.83) 11.10 (2.14) 10.45 (2.05) 10.31 (1.95)mmol/mol 98.4 (15.27) 90.2 (14.40) 92.5 (15.95) 97.8 (18.86) 90.7 (19.53) 89.2 (16.87)

P-DSMP 48.12 (13.55) 52.84 (13.49) 50.09 (12.91) 46.01 (12.05) 53.61 (11.83) 51.83 (11.88)

Y-DSMP 52.64 (11.59) 56.10 (13.30) 55.55 (12.81) 51.81 (11.53) 58.07 (8.25) 57.66 (9.50)

Figure 2—Mean (SD) reported adherence.

care.diabetesjournals.org Harris and Associates 1431

providing further evidence regardingthe utility of BFST-D. The use of an intent-to-treat approach to analyses, whichyields conservative estimates comparedwith conventional analytic methods,bolsters the evidence for the treat-ment’s effectiveness.The results add further evidence re-

garding the effectiveness of BFST-Dregardless of delivery format. The pri-marily skills-based nature of the BFST-Dlikely contributes to its adaptabilityacross delivery formats. Importantly,parents and youth reported strongwork-ing alliances regardless of whether theyreceived BFST-D via the clinic or video-conferencing (37). As such, videoconfer-encing seems to be an effective andacceptable method for delivering thistreatment.Using technology to deliver special-

ized behavioral health care such asBFST-D has several potential advantagesover traditional outpatient psychother-apy, including reducing burden to pa-tients (e.g., travel time) and improvingaccessibility to specialty providers. Ourfindings support establishing videocon-ferencing to provide behavioral healthservices to adolescents and their fami-lies who reside significant distancesfrom regional diabetes centers and forwhom travel time and/or costs maybe overly burdensome. A variety of vid-eoconferencing platforms are freely

available. Further, growing accessibilityof high-speed Internet services in ruralcommunities (21,22) has made the useof videoconferencing an increasinglyachievable means of delivering behav-ioral health.

Study limitations include participantsbeing racially and ethnically homoge-nous and predominately white. Futureefforts should explore the use of video-conferencing technology with minorityyouth with type 1 diabetes. Other limi-tations include almost all of the studyparticipants receiving their diabetescare from one of three primary diabetesclinics within the same large metropoli-tan area. Patients who live remote fromlarge metropolitan areas often receivediabetes care from local providers whohave fewer resources (e.g., limited pedi-atric training, no certified diabetes edu-cators on staff). Thus, care provided toyouth with diabetes who access care inlarger cities may be systematically dif-ferent from that provided to youth inrural settings.

Greater attrition occurred amongthose in the Skype condition. Reasonsfor this were not systematically investi-gated but may relate to the convenienceand ease of videoconferencing lesseningpersonal investment in the appointmentand overall process. Future studiesshould formally assess reasons for par-ticipant attrition. In addition, given the

lack of a waitlist or mock treatment con-dition, the possibility that treatment ef-fects were partially the result ofregression to the mean is a limitationthat should be considered in future stud-ies. Finally, adolescents were providedcompensation for the time andeffort nec-essary to complete questionnaires andparticipate in the study. Whether this un-duly affected patient retention should beconsidered in future studies and in effortsto translate the intervention to clinicalsettings. Given the focus of this study(i.e., to test delivery format) and thebody of previous research demonstrat-ing the effectiveness of BFST-D (11–17),we believe these findings are an impor-tant addition to the extant literature de-spite the limitations.

While the current findings support us-ing freely available, Internet-based tech-nology to deliver behavioral health careto youth with diabetes and their fami-lies, important ethical, legal, and prac-tice implications must be consideredbefore doing so. These issues havebeen nicely summarized by Richardsonet al. (38) and include integrating tech-nology into practice procedures, cross-state licensure or collaboration withcertified personnel, privacy and security,standard-of-care issues (e.g., emergencyprotocols, determining responsibilitiesof involved staff, liability for risks ofabandonment or negligence in the faceof equipment failure), management ofreimbursement regulations for tele-health and traditional services, and infra-structure management. Thus, adoptingtelemental health services requires care-ful attention to multiple issues and ismore complex than simply using elec-tronic technologies to deliver care. Fur-ther, while access to high-speed Internetis increasing in the U.S. (21), families inrural settings and/or of low economicmeans may still have limited access.Thus, disparities in access to BFST-D orother similar interventions may remaineven if telehealth is implemented.

Current findings raise several impor-tant questions for future research. First,although this study demonstrated thatimproved outcomes were sustained atthe 3-month follow-up, future studiesshould examine the durability of im-provements over a longer time period.Second, this study focused on adoles-cents known to have poor adherenceand glycemic control because BFST-D is

Figure 3—Mean (SD) glycemic control.

1432 Skype for Parent–Teen Diabetes Conflict Diabetes Care Volume 38, August 2015

designed to effectively respond to suchchallenges. Third, future research maywish to explore whether traditional(Clinic) BFST-D may be used to preventnonadherence challenges by targetingpreadolescents and their families. Suchefforts may extend understanding ofhow to effectively improve health out-comes for this population. Fourth, BFST-Dis an established intervention, but itsapplication through teleconferencingtechnology is unique. A disassemblystudy to examine the contributions ofeach of the four primary componentsmay inform modification of BFST-D andlead to a more streamlined intervention.Fifth, outcomes for youth with poor gly-cemic control were clinically meaningfuland encouraging. However, most adoles-cents in the study remained in subop-timal glycemic control. Future workshould further examine means of opti-mizing intervention outcomes to achievegreater improvements in glycemic con-trol. For example, future researchersmay wish to examine the intensive ap-plication of active intervention compo-nents (based on the fourth pointabove) and/or the influence of the inter-vention provided over an extended pe-riod of time to determine whether moreintense treatment or longer treatmentduration results in improved adherenceand glycemic control. As a related issue,evaluation of how BFST-D affects qualityof life for youth with diabetes may en-hance the clinical utility of outcomes,even if optimal glycemic control is notachieved. To support this, previous re-search has shown that diabetes-specificparent–youth conflict predicts youth-reported quality of life (39).In summary, these findings support the

effectiveness of delivering BFST-D usingvideoconferencing technology. Further,the results add to growing evidence thatBFST-D can be effectively delivered in var-ied ways. Thus, while future research willlikely extend our understanding of family-based interventions to address healthoutcomes for youth with diabetes, avail-able evidence suggests that identifyingpersonnel and resources necessary to in-tegrate BFST-D into clinical practice canbe an important and effective componentof diabetes care.

Acknowledgments. The authors thank theadolescents and their families who participated

in the study. In addition, the authors thank thediabetes care team at Harold Schnitzer DiabetesHealth Center at Oregon Health & Science Uni-versity for their support and assistance in con-ducting the study, with special acknowledgmentto Beth Hirschfield, Bruce Boston, and LisaShimomaeda.Funding. The research project was funded byan American Diabetes Association grant (no.1-10-CT-21, M.A.H.).Duality of Interest. No potential conflicts ofinterest relevant to this article were reported.Author Contributions.M.A.H. conceptualizedand designed the study, procured and managedgrant funding, trained interventionists, pro-vided direct intervention, and wrote the man-uscript. K.A.F. conceptualized and designed thestudy, procured and managed grant funding,managed personnel, and wrote the manuscript.D.C.D. provided direct intervention, managedpersonnel and database, analyzed data, andwrote the manuscript. M.A.H. is the guarantorof this work and, as such, had full access to allthe data in the study and takes responsibility forthe integrity of the data and the accuracy of thedata analysis.Prior Presentation. Aspects of this study werepresented in abstract form at the 74th ScientificSessions of the American Diabetes Association,San Francisco, CA, 13–17 June 2014; the 73rdScientific Sessions of the American Diabetes As-sociation, Chicago, IL, 21–25 June 2013; the72nd Scientific Sessions of the American Diabe-tes Association, Philadelphia, PA, 8–12 June2012; and the 71st Scientific Sessions of theAmerican Diabetes Association, San Diego, CA,24–28 June 2011.

References1. Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions andComplications (DCCT/EDIC) Research Group;Nathan DM, Zinman B, Cleary PA, et al. Modern-day clinical course of type 1 diabetes mellitusafter 30 years’ duration: the Diabetes Controland Complications Trial/Epidemiology of Diabe-tes Interventions and Complications and Pitts-burgh Epidemiology of Diabetes ComplicationsExperience (1983–2005). Arch Intern Med 2009;169:1307–13162. American Diabetes Association. Standards ofmedical care for patients with diabetesmellitus.Diabetes Care 2002;25:213–2293. Cohen DM, Lumley MA, Naar-King S,Partridge T, Cakan N. Child behavior problemsand family functioning as predictors of adher-ence and glycemic control in economically dis-advantaged children with type 1 diabetes:a prospective study. J Pediatr Psychol 2004;29:171–1844. Hanson CL, Henggeler SW, Harris MA, et al.Contributions of sibling relations to the adapta-tion of youths with insulin-dependent diabetesmellitus. J Consult Clin Psychol 1992;60:104–1125. La Greca AM, AuslanderWF, Greco P, SpetterD, Fisher EB Jr, Santiago JV. I get by with a littlehelp from my family and friends: adolescents’support for diabetes care. J Pediatr Psychol1995;20:449–4766. Wysocki T, Hough BS, Ward KM, Green LB.Diabetes mellitus in the transition to adulthood:

adjustment, self-care, and health status. J DevBehav Pediatr 1992;13:194–2017. Duke DC, Geffken GR, Lewin AB, Williams LB,Storch EA, Silverstein JH. Glycemic control inyouth with type 1 diabetes: family predictorsand mediators. J Pediatr Psychol 2008;33:719–7278. Wysocki T, Harris MA, Buckloh LM, et al. Ef-fects of behavioral family systems therapy fordiabetes on adolescents’ family relationships,treatment adherence, and metabolic control. JPediatr Psychol 2006;31:928–9389. Bobrow ES, AvRuskin TW, Siller J. Mother-daughter interaction and adherence to diabetesregimens. Diabetes Care 1985;8:146–15110. Miller-Johnson S, Emery RE, Marvin RS,Clarke W, Lovinger R, Martin M. Parent-childrelationships and the management of insulin-dependent diabetes mellitus. J Consult Clin Psy-chol 1994;62:603–61011. Wysocki T, Miller K, Greco P, et al. Behaviortherapy for families of adolescents with diabe-tes: effects on directly observed family interac-tions. Behav Ther 1999;30:507–52512. Wysocki T, Harris MA, Greco P, et al. Ran-domized, controlled trial of behavior therapy forfamilies of adolescents with insulin-dependentdiabetes mellitus. J Pediatr Psychol 2000;25:23–3313. Wysocki T, Greco P, Harris MA, Bubb J,White NH. Behavior therapy for families of ado-lescents with diabetes: maintenance of treat-ment effects. Diabetes Care 2001;24:441–44614. Wysocki T, Harris MA, Buckloh LM, et al.Randomized trial of Behavioral Family SystemsTherapy for Diabetes: maintenance of effects ondiabetes outcomes in adolescents. DiabetesCare 2007;30:555–56015. Wysocki T, HarrisMA, Buckloh LM, et al. Ran-domized, controlled trial of Behavioral Family Sys-tems Therapy for Diabetes: maintenance andgeneralization of effects on parent-adolescentcommunication. Behav Ther 2008;39:33–4616. Harris MA, Mertlich D. Piloting home-basedbehavioral family systems therapy for adoles-cents with poorly controlled diabetes. ChildHealth Care 2003;32:65–7917. Harris MA, Freeman KA, Beers M. Familytherapy for adolescents with poorly controlleddiabetes: initial test of clinical significance. J Pe-diatr Psychol 2009;34:1097–110718. Cooper S, Valleley RJ, Polaha J, Begeny J,Evans JH. Running out of time: physician man-agement of behavioral health concerns in ruralpediatric primary care. Pediatrics 2006;118:e132–e13819. Meadows T, Valleley R, Haack MK, ThorsonR, Evans J. Physician “costs” in providing behav-ioral health in primary care. Clin Pediatr (Phila)2011;50:447–45520. Sood S, Mbarika V, Jugoo S, et al. Whatis telemedicine? A collection of 104 peer-reviewed perspectives and theoretical underpin-nings. Telemed J E Health 2007;13:573–59021. Horrigan JB. Home broadband adoption2009 [article online]. Washington, DC, Pew In-ternet and American Life Project, 2009. Avail-able from http://www.pewinternet.org/Reports/2009/10-Home-Broadband-Adoption-2009.aspx. Accessed 11 May 201522. Palermo TM, Wilson AC. eHealth applica-tions in pediatric psychology. In: Handbook of

care.diabetesjournals.org Harris and Associates 1433

Pediatric Psychology, 4th ed. RobertsMC, SteeleRG, Eds. New York, Guilford, 2009, p. 227–23723. Duncan AB, Velasquez SE, Nelson E. Usingvideoconferencing to provide psychologicalservices to rural children and adolescents: a re-view and case example. J Clin Child Adolesc Psy-chol 2014;43:115–12724. FlaumM. Telemental health as a solution tothe widening gap in between supply and de-mand for mental health services. In: TelementalHealth: Clinical, Technical, and AdministrativeFoundations for Evidence-Based Practice.Meyers K, Turvey CL, Eds. New York, Elsevier,2013, p. 11–2525. Roine R, Ohinmaa A, Hailey D. Assessingtelemedicine: a systematic review of the litera-ture. CMAJ 2001;165:765–77126. Whittemore R, Grey M, Faulkner M, Jaser S,Deamater A, Murphy K. Recruitment and partici-pation of youth in an Internet psycho-educationalintervention study. Diabetes 2009;58:A48627. Mulvaney SA, Rothman RL, Lybarger C,Wallston KA. Modeling problem solving withan Internet-based self-management interven-tion for adolescents with type 1 diabetes. Dia-betes 2009;58:A482

28. Moher D, Schulz KF, Altman DG; CONSORTGROUP (Consolidated Standards of ReportingTrials). The CONSORT statement: revised rec-ommendations for improving the quality of re-ports of parallel-group randomized trials. AnnIntern Med 2001;134:657–66229. Hollingshead AB. Four Factor Index of SocialStatus. New Haven, Yale University Departmentof Sociology, 1975.30. Harris MA, Wysocki T, Sadler M, et al. Val-idation of a structured interview for the assess-ment of diabetes self-management. DiabetesCare 2000;23:1301–130431. American Diabetes Association. Standardsof medical care in diabetesd2014. DiabetesCare 2014;37(Suppl. 1):S14–S8032. Tamborlane WV, Kollman C, Steffes MW,et al.; Diabetes Research in Children Network(DirecNet) Study Group. Comparison of finger-stick hemoglobin A1c levels assayed by DCA2000 with the DCCT/EDIC central laboratory as-say: results of a Diabetes Research in ChildrenNetwork (DirecNet) Study. Pediatr Diabetes2005;6:13–1633. Rubin DB. Multiple Imputation for Nonre-sponse in Surveys. New York, John Wiley andSons, 1987

34. Schafer JL, Graham JW. Missing data: ourview of the state of the art. Psychol Methods2002;7:147–17735. Schafer JL. Analysis of Incomplete Multivar-iate Data. Boca Raton, CRC Press, 199736. Diabetes Control and Complications TrialResearch Group. Effect of intensive diabetestreatment on the development and progressionof long-term complications in adolescents withinsulin-dependent diabetes mellitus: DiabetesControl and Complications Trial. J Pediatr1994;125:177–18837. Freeman KA, Duke DC, Harris MA. Behavioralhealth care for adolescentswith poorly controlleddiabetes via Skype: does working alliance remainintact? J Diabetes Sci Tech 2013;7:727–73538. Richardson LK, Frueh BC, Grubaugh AL,Egede L, Elhai JD. Current directions in video-conferencing tele-mental health research. ClinPsychol (New York) 2009;16:323–33839. Laffel LM, Connell A, Vangsness L, Goebel-Fabbri A, Mansfield A, Anderson BJ. Generalquality of life in youth with type 1 diabetes: re-lationship to patient management and diabetes-specific family conflict. Diabetes Care 2003;26:3067–3073

1434 Skype for Parent–Teen Diabetes Conflict Diabetes Care Volume 38, August 2015