Evaluation of a web-enhanced parenting program for military families: After Deployment, Adaptive Parenting Tools/ADAPT

Transcription

1 Evaluation of a web-enhanced parenting program for military families: After Deployment, Adaptive Parenting Tools/ADAPT Friends of NIDA Congressional Briefing, Washington, DC; March 11 th, 2013 Abigail Gewirtz, Ph.D. Associate Professor Dept of Family Social Science & Institute Of Child Development, University of Minnesota

2 Acknowledgements ADAPT study funded by NIDA (R01 DA ; (associated funding from SAMHSA s National Child Traumatic Stress Network (SM56177); & NICHD R43 HD ) Co-investigators: Drs. Marion Forgatch, Dave DeGarmo, Melissa Polusny Research staff and facilitators Hundreds of National Guard and Reserve (NG/R) families National Guard command and communication staff Thanks to Friends of NIDA and the American Psychological Association for the invitation to present!

3 Background & Rationale Increase in attention to families of NG/R service members given data indicating: Higher levels of drug use, PTSD, other combatrelated adjustment problems among NG/R compared with Active Duty personnel (Jacobson et al., 2008 alcohol use; Milliken, Auchterlonie, and Hoge, PTSD, depression, interpersonal conflict) Posttraumatic distress predicts parenting challenges following deployment (Gewirtz, Polusny, Erbes, & DeGarmo, 2010)

4 Parenting practices mediate impact of family stressors on child adjustment Deployment is a family stressor Separations from family and children (MacDermid, 2006) Combat related stressors (Cozza et al., 2005) Reintegration is a key transition point Transitions offer special prevention opportunities Parenting practices mediate the impact of deployment stress on child outcomes robust association in range of populations between family stressors (risks including parental substance use, PTSD, other psychiatric illness, interpersonal conflict, as well as family transitions, socioeconomic stress) and parenting difficulties, and subsequent child maladjustment (e.g., Beardslee et al., 1983; Patterson, 1982, 1986; Belsky, 1984; Capaldi, 1991; Conger et al., 2002; Palmer, 2008)

5

6 An empirically supported intervention to reduce substance use and improve child adjustment: Parent Management Training Oregon (PMTO) Well-established family of parent training programs listed on multiple federal clearinghouses as best practice (e.g. Parenting Through Change). Largest group of parenting interventions on SAMHSA s NREPP) NIDA-funded studies have demonstrated the utility of PMTO in improving positive parenting and reducing coercion, thereby reducing child substance use and related risk behaviors PMTO (follow-up data to 9 years) demonstrated Reduced substance use & risk, improved child adjustment Improved parents mental health, substance use, couple functioning Improved SES higher income, education, occupation Positive parenting mediates the effects of family stress on children s adjustment

7 Univariate Effect Sizes Cohen s d Oregon Divorce Study: Cascading effects (Patterson, Forgatch & DeGarmo, 2010)

8 PMTO has a large-scale, validated implementation infrastructure Implemented nationwide in: Norway Netherlands Iceland Statewide in Michigan, Kansas, Implementation underway in NYC, Mexico City Independent effectiveness trials in Norway, Iceland; (e.g. Ogden & Amlund-Hagen, 2010) Validated fidelity model (higher observed fidelity predicts change in parenting; Forgatch & DeGarmo, 2011)

9 Fidelity To Outcome Theoretical Model Fidelity Change Parenting Change Child Behavior

10 Fidelity Effects on Pre/Post Treatment Change in Effective Parenting Norwegian Sample Fidelity Effects on Change in Observed Parenting.85** Baseline Maternal Parenting -.66** 2 R =.44 e1 Encouragement Sessions.71***.16* Post TX Change Maternal Parenting e3 e2 Discipline Sessions.74*** Intervention Fidelity.18* 2 R =.46.86** Baseline Father Parenting -.66** Post TX Change Father Parenting e4 Therapists = 114 Families = 287 chi-square =3.627, df = 6, P =.727, CFI = 1.000, cmindf =.604, rmsea =.000 tp <.10; *p <.05; **p <.01; ***p <.001 Forgatch, DeGarmo, Patterson (in prep)

11 H: Fidelity will Drift After Transfer G1 Fidelity Training Hypothesized Covariates: Agency Characteristics Therapist Characteristics Family Characteristics (Forgatch, DeGarmo, & Patterson, 2011) G2 Fidelity Training G3 Fidelity

12 Fidelity Does Not Drift After Transfer 3 Generations in Norway: FIMP Scores at Certification N= G 1 avgfg1 G2 G3 avgfg2 G1 G2 avgfg3 G3 (Forgatch, DeGarmo, & Patterson, 2011)

13 H: Training to Fidelity Can Be Replicated Norway G1 Michigan G N= Norway G1 FIMP 2001 Michigan G1 FIMP 2008 (Forgatch, DeGarmo, & Patterson, 2011)

14 Extending PMTO for military families: After Deployment Adaptive Parenting Tools

15 Effectiveness of a web-enhanced parenting program for military families 5 year NIDA-funded study ( ) 400 NG/R families will be recruited; each family followed for two years (recruitment started July 2011; 310 families already recruited) Random assignment to a parenting program (ADAPT; N=240) or parenting services-as-usual (web and print resources; N=160) Parents and teachers complete online questionnaires, and observational, self-report, and physiological data are gathered from families (parents & children) at baseline, 6, 12, and 24 months. Outcomes: parenting practices, couple adjustment, child substance use risk & related (behavior & emotional) problems, parent emotion regulation, parent emotion socialization

16 Modifications to PMTO for military families: ADAPT ADAPT is a 14-week long, web-enhanced, group-based program for troops returning from deployment who have at least one child aged 5-12yrs Weekly, provided in the community, 2hrs long, groups began Sept 2011 Online ADAPT is available to participants for 12 months Skills: teaching through encouragement, discipline, problem-solving, monitoring, positive involvement Modifications include: attention to emotion regulation in family communication (emotion socialization) Mindfulness training (to address experiential avoidance associated with combat stress symptoms) Emotion coaching (esp. responding to children s anxiety) Attention to military culture and values Emphasis on united parenting front (for two-parent families) Addressing common barriers to participation Web-component to increase involvement in group program by other caregivers, spouses, etc. Stand alone online ADAPT is under development

17

18 Study interim baseline data 376 people in 251 families in Reserve Units and Minnesota National Guard 158 men, 218 women 38 deployed mothers (20%); 153 deployed fathers Married: 8.9 years on average (SD = 5.3) Mean number of marriages: 1.2 Average number of children: 2.3 (SD = 1) Mean family income: $71,281 (SD = $34,761) Number of deployments (ever): 1-13 for men, 1-5 for women (mean = 3.4 for men, 1.3 for women) Number of deployments since 2001: 1-6 for men, 1-5 for women (mean = 1.8 for men, 1.4 for women) Deployment status by family Male deployed = 213 Female deployed = 22 Both deployed = 16

19 Preliminary intervention dosage data Cohorts 1 &2 first 6 ADAPT groups 42 families invited; 33 participated in at least one session (78%) Participating families were 85% currently married Participants were 48% fathers, 52% mothers 79% families attended at least 50% sessions (i.e. 7 out of 14 sessions) Home practice completion 48%; 55% of families used online tools Average general satisfaction 3.39 on a 0-4 scale

20 6-month outcome data - preliminary 87 participants (39 controls, 48 intervention) in 54 families with baseline and Time 2 (post intervention data at 8 months post-baseline) - 33 couples 21 single parents Outcomes (parent self-report only) Parenting practices (ineffective discipline; APQ) Couple satisfaction (DAS) Parenting self-efficacy (PLOC; time 2 only) Parent mindfulness (FFMQ) Child deviant peer association (DVPEER) 21

21 LEVEL-1 DESCRIPTIVE STATISTICS VARIABLE NAME N MEAN SD MINIMUM MAXIMUM MALE EDUCATIO EMPLOYME INCOME OVDEPLOY DEPLOY CGRADE APQPOS APQDIS APQMON AUDIT AU_HI AU_ME AU_LO AU_NO DAS FFMQOBS FFMQDES FFMQAWR FFMQJDG FFMQNRE DERS1N DERS1G DERS1I DERS1A DERS1S DERS1C DERS SCL SCLDEP PVIQ PEER DVPEER TOTPEER PSITOT PSISAT APQPOS APQDIS APQMON AUDIT AU_HI AU_ME AU_LO AU_NO DAS FFMQOBS FFMQDES FFMQAWR FFMQJDG FFMQNRE DERS2N DERS2G DERS2I DERS2A DERS2S DERS2C DERS SCL SCLDEP PVIQ PEER DVPEER TOTPEER PLOC PSITOT PSISAT HLM Data File: Models will appropriately adjust for couple nesting LEVEL-2 DESCRIPTIVE STATISTICS VARIABLE NAME N MEAN SD MINIMUM MAXIMUM ADAPT BOY EFADAPT EFBOY

22 Final estimation of fixed effects: Ineffective discipline (Alabama Parenting Questionnaire short form; APQ; Elgar et al., 2007) Fixed Effect Coefficient Standard error t-ratio Approx. d.f. p-value For INTRCPT1, β 0 INTRCPT2, γ <0.001 ADAPT, γ BOY, γ ZCGRADE, γ For MALE slope, β 1 INTRCPT2, γ For EDUCATIO slope, β 2 INTRCPT2, γ For INCOME slope, β 3 INTRCPT2, γ For OVDEPLOY slope, β 4 INTRCPT2, γ For APQDIS1 slope, β 5 INTRCPT2, γ <

23 APQDIS ADAPT 8.00 ADAPT = 0 ADAPT = APQDIS MALE 24

24 10.00 ADAPT = 0 ADAPT = APQDIS APQDIS1 25

25 Couple Adjustment; Abbreviated Dyadic Adjustment Scale (Spanier, 1976) Fixed Effect Coefficient Standard error t-ratio Approx. d.f. p-value For INTRCPT1, β 0 INTRCPT2, γ ADAPT, γ BOY, γ CGRADE, γ For MALE slope, β 1 INTRCPT2, γ For EDUCATIO slope, β 2 INTRCPT2, γ For INCOME slope, β 3 INTRCPT2, γ For OVDEPLOY slope, β 4 INTRCPT2, γ For DAS1 slope, β 5 INTRCPT2, γ <

26 3.75 ADAPT = 0 ADAPT = DAS MALE 27

27 Parenting Self-efficacy (PLOC, Campis et al, 1986); Final estimation of fixed effects (with robust standard errors) Fixed Effect Coefficient Standard error t-ratio Approx. d.f. p-value For INTRCPT1, β 0 INTRCPT2, γ <0.001 ADAPT, γ BOY, γ For MALE slope, β 1 INTRCPT2, γ For EDUCATIO slope, β 2 INTRCPT2, γ For INCOME slope, β 3 INTRCPT2, γ For OVDEPLOY slope, β 4 INTRCPT2, γ For CGRADE slope, β 5 INTRCPT2, γ

28 Mindfulness FFMQ describing feelings (Baer et al., 2008) Fixed Effect Coefficient Standard error t-ratio Approx. d.f. p-value For INTRCPT1, β 0 INTRCPT2, γ ADAPT, γ BOY, γ CGRADE, γ For MALE slope, β 1 INTRCPT2, γ For EDUCATIO slope, β 2 INTRCPT2, γ For INCOME slope, β 3 INTRCPT2, γ For OVDEPLOY slope, β 4 INTRCPT2, γ For FFMQDES1 slope, β 5 INTRCPT2, γ <

29 FFMQ acting with awareness scale Fixed Effect Coefficient Standard error t-ratio Approx. d.f. p-value For INTRCPT1, β 0 INTRCPT2, γ ADAPT, γ BOY, γ CGRADE, γ For MALE slope, β 1 INTRCPT2, γ For EDUCATIO slope, β 2 INTRCPT2, γ For INCOME slope, β 3 INTRCPT2, γ For OVDEPLOY slope, β 4 INTRCPT2, γ For FFMQAWR1 slope, β 5 INTRCPT2, γ <

30 Single Level Repeated Measures Pre Post MANOVA 31

31 Child deviant peer association Describing Friends, (Dishion et al., 1995) Fixed Effect Coefficient Standard error t-ratio Approx. d.f. p-value For INTRCPT1, β 0 INTRCPT2, γ ADAPT, γ BOY, γ CGRADE, γ For MALE slope, β 1 INTRCPT2, γ For EDUCATIO slope, β 2 INTRCPT2, γ For INCOME slope, β 3 INTRCPT2, γ For OVDEPLOY slope, β 4 INTRCPT2, γ ADAPT, γ For DVPEER1 slope, β 5 INTRCPT2, γ

32 0.950 ADAPT = 0 ADAPT = DVPEER OVDEPLOY 33

33 Multilevel HLM Models: Explained Variance and Effect Size for ADAPT R 2 r d ES Dyadic Adjustment (DAS-7) Medium Parental Locus of Control (PLOC) Small to Medium Poor Discipline (APQ) Medium to Large Describe Emotions (FFMQ) Small to Medium Active Awareness (FFMQ) Medium Cohen s d (.20 = Small,.50 = Medium,.80 = Large) 34

34 ADAPT results to date indicate: ADAPT intervention is feasible and acceptable At 8-month follow-up, controlling for baseline scores, those in the intervention group showed less poor discipline, better couple adjustment, increased mindfulness, and stronger parenting selfefficacy (T2 only), than those in the services-as-usual group. No main effects of treatment on reduction in association with deviant peers, but ADAPT condition buffered exposure to multiple deployments (i.e. reductions in deviant peer associations among those in the ADAPT condition with parents deployed more frequently) Gains were consistent across baseline scores and parent gender However, these scores only reflect first two cohorts of families to participate in ADAPT, and short follow-up time period

35 Thank you Abi Gewirtz: ADAPT s website: