In Search of Synergy: Multilevel Implementation Interventions - - PowerPoint PPT Presentation
In Search of Synergy: Multilevel Implementation Interventions - - PowerPoint PPT Presentation
In Search of Synergy: Multilevel Implementation Interventions Bryan J. Weiner, Ph.D. You think because you understand 'one' you must also understand 'two', because one and one make two. But you must also understand 'and'. ~ Rumi GOT
You think because you understand 'one' you must also understand 'two', because one and one make two. But you must also understand 'and'. ~ Rumi
GOT SYNERGY?
- Implementation facilitators and barriers operate at
multiple levels of influence
- Yet, multi-level implementation interventions often don’t
work better than single-level ones Why?
- Selected implementation interventions do not target key
determinants of the problem
- Interdependence of key determinants of the problem are
poorly understood
THE CAUSAL LOGIC OF MULTILEVEL INTERVENTION
We explored how interventions at different levels could be combined to produce complementary or synergistic effects …using a causal modeling framework to develop potential combinations We concentrated on two forms
- f causal interdependence:
mediation and moderation …and identified 5 potentially useful strategies for combining interventions at different levels
MEDIATION Mediator: A variable that explains (or accounts for) the relationship between the dependent variable and the independent variable.
5
IV M DV
Full Mediation
MEDIATION Mediator: A variable that explains (or accounts for) the relationship between the dependent variable and the independent variable
IV M DV
Partial Mediation
MEDIATION Mediator: a variable that explains (or accounts for) the relationship between the dependent variable and the independent variable
IV M DV
Parallel Mediation
M
MEDIATION
8
IV M DV
Serial Mediation
M
Mediator: A variable that explains (or accounts for) the relationship between the dependent variable and the independent variable.
MODERATION Moderator: A variable that changes the strength of an effect or relationship between two variables.
IV W DV
MODERATION Moderator: A variable that changes the strength of an effect or relationship between two variables.
IV W DV W
Multiple Moderation
MODERATION Moderator: A variable that changes the strength of an effect or relationship between two variables.
IV W DV W
Moderated moderation
COMBINATIONS Moderated mediation: Mediating process or pathway is intensified or attenuated by individual difference or contextual condition.
IV M DV W W
SOCIAL ECOLOGICAL MODEL Levels as Settings*
* Or social units
SOCIAL ECOLOGICAL MODEL Levels as Determinants
Example: Possible Barriers to or Determinants of Colorectal Cancer Screening
Level of Influence Possible Barrier/Determinant Intervention Strategy
Intrapersonal
- Patient/provider knowledge, attitudes, beliefs
- Patient structural barriers (e.g., time, money)
- Inadequate cues to action
- Education, training, navigation
- Decision support
- Reminders
Interpersonal
- Poor physician-patient communication
- Inadequate social/spousal support
- Professional (social) norms about screening
- Communication
- Shared decision making
- Social support
- Opinion leaders
Organizational
- Poor care coordination, inefficient processes
- Inadequate staffing, inadequate data systems
- Quality improvement
- Service/staffing changes
- Role revisions, registries
Community
- Low community awareness/demand
- Inadequate supply of diagnostic colonoscopy
services
- Social marketing campaign
- Assisted transportation services
Macro Policy
- Limited or lack of insurance coverage
- Inadequate incentives to coordinate care
- Insurance coverage changes
- Bundled payment policies
- 1. ACCUMULATION STRATEGY
Legend Box: intervention and level of influence (in parentheses) Diamond: mediator Oval: outcome
Public reporting* (organizational) Op Opinion Leader (I (Interpersonal) Outreach visit (Intrapersonal)) Physician’s motivation Colorectal Cancer Screening *For expositional purposes, the diagram depicts interventions rather than determinants as causes. To keep the concept simple, a single mediating pathway is presented.
- 2. AMPLIFICATION STRATEGY
Public reporting* (Organizational) Opinion leader (Interpersonal) Audit and feedback (Intrapersonal) Physician’s motivation Colorectal Cancer Screening
- 3. FACILITATION STRATEGY
Public reporting* (Organizational) Opinion leader (Interpersonal) Clinical reminder (Intrapersonal) Physician’s motivation Colorectal Cancer Screening
- 4. CASCADE STRATEGY
Office systems
Colorectal Cancer Screening
Quality improvement (Organizational) Physician motivation Outreach visit (Intrapersonal) Manager motivation Advocacy at state level (Macro Policy)
- 5. CONVERGENCE STRATEGY
Public reporting* (Organizational) Opinion leader (Interpersonal) Patient education (Intrapersonal) Physician motivation Colorectal Cancer Screening Patient motivation Provider- patient interaction
AMPLIFICATION STRATEGY
Clinician’s motivation Participation in Training Incentives (Intrapersonal) Social norms Opinion leader (Interpersonal)
WHAT’S NEEDED
> Multilevel theories that explain how determinants at multiple levels interact to produce health and other
- utcomes