Scaling Up Care for Orphans in Tanzania: A Task-sharing Approach to - - PowerPoint PPT Presentation

Shannon Dorsey1, Karen O’Donnell2, Kate Whetten2; Wenfeng Gong3; Dafrosa Itemba4, & Rachel Manongi5

1University of Washington,; 2Duke University; 3Johns Hopkins

University; 4Tanzania Women Research Foundation; 5Kilimanjaro Christian Medical Centre

Scaling Up Care for Orphans in Tanzania: A Task-sharing Approach to Mental Health Treatment

NIMH R34 MH081764; 2009-2012 NIMH R01 MH96633; 2012-2017

Acknowledgements

 NIMH  USAID Victims of Torture Fund

(funded training development)

 Lui Mfangavo (coordinator, interviewer)  Simon Chudy (lay counselor)  Suzan Kitomari (lay counselor)  Bibiana Gali (lay counselor)  Leonia Rugalabamu (lay counselor)  Karthik Balasubramanian  Lillian Chinganyana  Wenfeng Gong

Acknowledgements

 Implementation Research Institute  Dr. Dorsey is an investigator with the Implementation

Research Institute (IRI), at the George Warren Brown School

• f Social Work, Washington University in St. Louis; through

an award from the National Institute of Mental Health (R25 MH080916-01A2) and the Department of Veterans Affairs, Health Services Research & Development Service, Quality Enhancement Research Initiative (QUERI).

Orphans in Low and Middle Income Countries

 Estimated 143 million

 16.6 million of these to HIV/AIDS

 “More than the loss of a parent1…”

 High rates of exposure to potentially traumatic events  Higher rate trauma exposure associated with mental health

problems

1Whetten, Ostermann, Whetten, O’Donnell & Thielman, 2011;

Cluver, Fincham & Seedat, 2009

Problems of Orphans: Sadness and Grief

During development: “need help with children’s sadness”

Low and Middle Income Countries

 Significant mental health treatment gap (over 75%)1

 Scarcity: Few MH professionals2  Inequity: MH services within and across countries2

Lower Middle Income Low Income

1Kohn, Saxena, Levav, & Saraceno, 2004; 2Saraceno et al., 2007; WHO, 2008; WHO, 2009

Treatment Gap: Children

For each child with need…

.16% receive treatment

Saraceno et al., 2007

LMIC: Addressing Treatment Gap

Mental

health professionals

Implementation Strategy: “Task Shifting/Sharing Approach”

Lay Counselors— Little or No Mental Health Training

Patel, 2009

Randomized Trials: Evidence for Evidence-based MH Interventions in LMIC

 Uganda1

 Adults  Adolescents  Internally displaced

persons

 India3

 Adults  Depression/Anxiety

 Pakistan2

 Perinatal women with

depression

 Iraq, Thailand (displaced

Burmese)4

 Adults  Torture, systematic violence

1Bolton et al., 2003, 2007; 2Rahman, Malik, Sikander, Roberts, & Creed, 2008; 3Patel et al., 2010 ; 4Bolton et al., results forthcoming (Dorsey, involved)

Modifications: Not to Core Components

Patel, Chowdhary, Rahman, & Verdeli, 2011 HOW training is conducted Local idioms and stories Simplifying terms, avoiding “clinical” terms: depression Supervision Situating within local context

TF-CBT Intervention

Pa Pamoja Tuna naweza

Trauma-focused Cognitive Behavioral Therapy (TF-CBT)

Moshi, Tanzania

TF-CBT Feasibility Study (NIMH; Duke & UW; Whetten; Dorsey, Co-I) Orphans: Traumatic Stress and Grief-focused TF-CBT Randomized Trial (NIMH; Duke & UW; Dorsey & Whetten) Orphans: Traumatic Stress and Grief-focused Murray, Consultant TF-CBT Randomized Trial (DCOF; Johns Hopkins; Murray & Bolton) TF-CBT Randomized Trial (NICHD; Johns Hopkins) Murray & Bolton HIV-prevention focus Dorsey: Faculty TF-CBT Feasibility Study (NIMH; Johns Hopkins; Murray) Sexual abuse

Trauma-focused Cognitive Behavioral Therapy

Sessions 1-4 5-8 9-12 Psychoeducation Trauma Narrative Conjoint Parent Parenting Skills Development and Child Sessions Processing Relaxation Enhancing In-vivo Gradual Safety and Affective Exposure Future Expression and Development Regulation Cognitive Coping

Entire Process is Desensitization/Exposure Baseline Assessment

http://tfcbt.musc.edu/

TF-CBT Internationally

 TF-CBTWeb registrants from over 70 countries  Cohen, Mannarino & Deblinger (2006) translated into Dutch, Korean,

Mandarin, & German

 Currently being translated into Japanese and Polish

 China, Japan, Singapore  Norway, Germany, Sweden, the Netherlands  Zambia, Tanzania, Cambodia, Indonesia  South Korea (Introduction)

http://ctg.musc.edu/

Qualitative Study (DIME Procedures1): Orphan Problems in Tanzania

Qualitative Study with Orphans and Guardians

 Many needs related to education, food, clothing, and shelter  Mental health problems still in the running: mentioned by guardians AND

children

JHU Applied Mental Health Research group; Bolton, 2001

Qualitative Study (DIME Procedures): Orphan Problems in Tanzania

 Tabia Mbaya (Bad Behavior)  Unyanyasaji (humiliation/treated badly)  Kuathirika kisaikologia (psychological problems)  Kutopendwa (not feeling/being loved)  Msongo wa mawazo (stress; overthinking)

TF-CBT with Orphans in Tanzania

Feasibility study

Can lay counselors deliver the intervention with fidelity? Is the model acceptable with children/adolescents, guardians, and counselors? ID any needed modifications for local delivery * guardian support

Partnership with Local CBO

 Tanzania Women Research Foundation

Study Design

 N = 64 children (7-13)

 Half urban; half rural

Assessment

 Pre-treatment  Post-treatment  3-month follow-up  12-month follow-up

TF-CBT Intervention

Focus Group Feedback

Single sex groups Divide by age: 7-10; 11-13 Tea and snacks to start

TF-CBT Intervention

 12 groups; Child and guardian

 TF-CBT components AND four grief-specific components

 3 individual visits for individual Imaginal Exposure

 At home or community location (e.g., school)  12-session agenda: Collaboratively developed with lay counselors

post-training

 Revised iteratively

Adherence Monitoring: Groups

Monitoring Individual Child and Guardian Response

Current Study Status

 Completed all 8 groups (N = 64)  All end of treatment and 3-months post-treatment data

collected

 One-year follow up data still to be collected for groups 7 & 8

 High levels of satisfaction with the group, reported in

guardian exit interviews

 Good attendance

Completed Pilot; 8 groups (N = 64) All data collected (12 mo. to be analyzed) High satisfaction: guardian exit interviews Good attendance

PTSD Outcomes

5 10 15 20 25 30 35 40 45 50 Baseline End of Treatment 3 Month FU PTSD Symptoms Caregiver Child

Panel Regression: BL-ET: Caregiver β = 11.19, p < .001; Child β = 15.38, p < .001

Traumatic Grief Outcomes

5 10 15 20 25 30 Baseline End of Treatment 3-month FU Child

Panel Regression: BL-ET: Child β = 8.46, p < .001

Summary and Implications

 TF-CBT: feasible and acceptable

 Guardians requested additional group meetings

 Lay Counselors: High fidelity  Loved the intervention  Outcomes appear promising

Limitations

 Small sample size  No control group

TF-CBT: RCT in Eastern Africa

 Moshi, Tanzania  Bungoma, Kenya  RCT: TF-CBT compared to usual care (UC) supports  Lay Counselors from NIMH R34 assist with training of new

providers in both sites

 Provide supervision, under supervision themselves

12 counselors (6 in each country) 20 groups in each country (320 youth: TF-CBT; 320 UC)

Training Model: Murray, Dorsey et al., 2011

Apprenticeship Model: Lay Counselors

Murray, Dorsey, et al., 2011

Apprenticeship Model: Lay Counselors

Murray, Dorsey, et al., 2011

R34 Counselors:

Partners

Outcomes of Interest

Clinical Outcomes

 PTSD Symptoms  Traumatic Grief  Depression  Child Functioning

Implementation Outcomes

 Fidelity (BRs; self-reported; via coded audiotapes)  Child/Guardian attendance  Acceptability  Provider Knowledge  Supervisory Relationship

Outcomes of Interest

Clinical Outcomes

 PTSD Symptoms  Traumatic Grief  Depression  Child Functioning

Implementation Outcomes

 Fidelity (BRs; self-reported; via coded audiotapes)  Child/Guardian attendance  Acceptability  Provider Knowledge  Supervisory Relationship

To Date…

 October 2012: Training  Administered Modified

Practice Attitudes Scale

 Independent ratings of overall counselor fidelity post-

training

 Similar pattern ratings across counselors; with “expert”

 Knowledge pre and post-test

Possibilities for Broad Scale Up…

Remote Behavioral Rehearsal and Supervision

Lay Counselors as Co-Trainers & Supervisors

Adherence

“Step Sheets” guide practice Adherence =Following the recipe

Lay Counselors as Co-Trainers & Supervisors

Competence: List developed collaboratively: “What do You SAY in feedback?” Spicing it up: Flexibility within Fidelity

Task-shifting/ Task-sharing

“…however, the biggest barrier to scaling up may be the perception held by mental health specialists about the risks of non- specialist health workers delivering [psychological treatments]…”

Patel et al., 2011, p. 527

NEXT Goal: Cascading Implementation Conditions

Apprenticeship Model: Expert Trainers; Local Supervisors,

closely supervised themselves by experts

Train-the-trainer Model: Training AND Supervision by Local

Supervisors

• Experts on site for training: Observation, feedback, & coaching
• Provide less intense ongoing supervision

Local Responsibility Model

• Experts consult on training remotely
• Limited supervision

Shannon Dorsey dorsey2@uw.edu

Thank you.