From Randomized Controlled Trials to RealLife: Models for Moving Dementia Caregiver Intervention Res

1
From Randomized Controlled Trials to Real-Life
Models for Moving Dementia Caregiver
Intervention Research to Community and Home

• Laura N. Gitlin, Ph.D.
• Director, Center for Applied Research on Aging
  and Health
• Thomas Jefferson University, Philadelphia
• (Supported by funds from NIA/NINR 5 U01 AG13265
• And AOA 90CG257)

2
Overview of Presentation

• What is the evidence?
• Moving from randomized trials to practice
• 3 types of randomized trials
• Implications for translation
• Translational steps
• Advancing Translational Work

3
WHAT IS THE EVIDENCE?
4
Key Developments in Dementia Caregiver Research

• Descriptive research prevalence and
  characteristics of caregiving
• Measurement burden scales assessing problem
  behaviors among AD patients

1985 to present
1990s to present

• Stress Health Process Models applied to
  caregiving and refined

1985-1995

• Modest success achieved with individual based
  interventions
• Economic incentive programs have limited impact

1995 to present

• Psychiatric and physical morbidity effects
  identified

1997-1999

• Role conflict and other secondary stressors
  identified
• Economic value and costs of caregiving

1997 to present

• Multi-site randomized interventions trials (e.g.
  REACH) developed and tested

1999

• Caregiver risks of poor health and mortality
  identified

• Multi-component interventions achieve clinically
  significant outcomes

2000

• Proliferation of caregiver interventions tested
  using RCT and found to be effective
• Translation of intervention strategies to
  practice settings

2001 to present
5
(No Transcript)
6
Summary from Meta-analyses

• Small but significant benefits derived on a wide
  range of outcomes
• Skill enhancement
• Burden/depression
• CR symptom reduction/NH placement
• Multi-component interventions more effective
• Tailoring to individual participants important
• Women benefit more than men
• Few tested studies with diverse caregivers

7
Research-Practice Gap

• Caregiver assessment not integrated within
  existing services
• Interventions with known efficacy have not been
  translated/integrated into
• Aging network of services
• National Family Caregiver Programs
• Existing health services (e.g., home care,
  hospital discharge planning)
• Families continue to be underserved and do not
  receive proven interventions

8
From Randomized Trial to Practice
9
Phases of RCTs

• Phase I (safety, feasibility,
  acceptability)
• Phase II (preliminary effect size,
  side effects,
• 
  dosing)
• Phase III (efficacy of new treatment
  compared to standard)
• Phase IV (application in clinical setting
  long-term safety ,fidelity)
• Phase V
  (sustainability)

Forward Translation
Reverse Translation
10
Translational Efforts

• Forward Translation
• Hierarchical system converging toward clinical
  practice
• Research chain starting with experimental and
  theoretical models converging or leading to
  clinical practice
• Glasgow et al (RE-AIM model) Need to expand
  assessment of interventions beyond efficacy

11
3 Models

• Pure top down
• Not service setting or profession specific
• Not all treatment elements may have translation
  potential
• Funding mechanism for sustainability unclear
• REACH II Multi-component
• (Annals of Internal Medicine, 2006)
• ________________________________________
• Hybrid
• Service setting and profession specific
• Potential for reimbursement under current
  Medicare Part B guidelines
• REACH I Environmental Skill-building Program
  (ESP)
• (Gitlin et al., TG, 2001, 2003, 2005)
• _______________________________________
• Embedded
• Tested within adoption setting by staff
• Cost absorbed by setting
• Adult Day Service Plus (ADS Plus)
• (Gitlin et al., TG, 2006)

12
Study Highlights
13
Study Highlights
14
Study Highlights
15
Pure Form (REACH II) Pros/Cons

• Strong research design possible
• Allows testing of new, innovative approaches
  without concern for setting or professional
  boundaries
• Multi-site design high in internal and external
  validity
• Treatment implementation data allows for
  component and dose-response analysis
• Tested intervention may be too complex for
  real-world conditions
• Entire intervention approach may not translate
  easily
• Cost-efficiency questionable
• Who, what, when and where to place intervention
  needs to be evaluated

16
Hybrid (Philadelphia Site REACH I) Pros/Cons

• Strong research design possible
• Intervention components are service ready
• Some ecological validity
• Profession-based
• May limit generalizability to settings with
  limited access to interventionists
• Affordability of adoption can become issue
• Dependent in part on whims of funding mechanisms
  and service structures for which it was designed

17
Embedded (Adult Day Plus) Pros/Cons

• Intervention is service ready
• High ecological validity
• Able to evaluate provider adoption and CG
  acceptability within targeted setting
• Outcomes may be confounded by site/practice
  characteristics
• Best used if able to build on proven
  interventions

18
RE-AIM ModelTranslational Capacity of 3 Models
19
Phase I Translational Steps
20
of Translational steps High Low
Pure REACH II
Embedded ADS Plus
Hybrid ESP
21
Phase II Translational Steps

• 1. Site Development
• Assess staff needs (e.g, hire of
  interventionists)
• Prepare site (e.g., importance of evidence-based
  programs, introduction to intervention)
• Establish referral mechanism, intake forms and
  billing/reimbursement procedures for intervention
• Develop marketing materials and plan for rollout

22
Phase II Translational Steps

• 2. Refine Intervention/Service Program
• Refine eligibility criteria
• Identify core domains to evaluate outcomes
• Refine session by session protocols
• Refine treatment manuals and package for site
  usability
• Identify treatment fidelity approach

23
Phase II Translational Steps

• 3. Training
• Establish certification criteria for training
  staff/interventionists
• Refine training manuals and materials
• Implement training and evaluate uptake
• 4. Implementation and Evaluation
• Ongoing monitoring of fidelity
• Evaluation of participant benefit
• Booster training if necessary
• On-going identification of lessons learned/costs
  etc.

24
Site Adoption Considerations

• Required resources for adoption
• Integration in existing structures to reduce cost
  and enhance sustainability
• Buy-in by site personnel
• Importance of using evidence-based programs
• Immutable and mutable aspects of intervention
  need to be identified (e.g., what can site change
  and what must be kept in tact when implementing a
  proven intervention)
• Treatment Fidelity development of monitoring
  forms and quality assurance to assure integrity
  of implementation

25
Conclusions

• 3 models of existing caregiver interventions
• Each are theory based, attain scientific
  integrity, proven effectiveness
• 3 models represent different translational needs
  and challenges
• Each model has different relationship to
  translational effort
• Each model presents pros/cons for science and
  translation

26
Conclusions

• Translational steps depend on type of RCT
• E.g., Embedded model has fewer and different set
  of translational steps than Pure Model
• If primary goal is immediate translation
  hybrid/embedded models may be preferred
• Able to test simultaneously efficacy and
  translational issues
• Very high ecological validity in that end users
  are part of intervention development and testing
• More rapid transition to real-world settings
• May be more efficient
• Cost, adoption capacity are easily tested as part
  of efficacy model

27
Implications

• Identifying translational steps for RCTs may
  expedite research-practice integration
• No one successful translational strategy
• Each model has pros/cons
• Designing interventions based on current
  reimbursement mechanisms and service structures
  may serve immediate needs but can hinder science
  and long-term benefits for families
• Need to balance translational approaches with
  basic science needs

28
Implications

• Moving forward with caregiver intervention
  studies
• Design of Phase III trials should identify
  upfront translational steps
• Need for more hybrid and embedded models
• Need for collaboration with potential adoption
  sites in developing study design/intervention
• Combine efficacy and effectiveness in one phase
• Funding mechanisms for translational research not
  well developed
• NIMH, NCI, NIH Roadmap
• Cost of translation needs to be considered