Public health has never been short on evidence. What it has consistently struggled with is turning that evidence into action.

That tension was front and center at the 2025 European Implementation Event (EIE2025), where researchers, policymakers, and practitioners gathered to confront a hard truth: implementation is not just difficult, it is fundamentally complex.

Over the past two decades, implementation science has grown rapidly. But growth has exposed a critical gap. Much of the field has been built on simplified models that do not fully reflect the messy realities of health systems, political environments, and community contexts.

The central question of the conference is the simple one: How do we balance scientific rigor with real-world relevance? What emerged was a set of shifts that could reshape public health moving forward.

The Big Idea: Complexity Is the Starting Point

A recurring theme across the conference was that complexity should no longer be treated as an excuse for failure. Instead, it must become the foundation for how we design, test, and scale interventions. Speakers pointed to the growing influence of:

• Political instability and austerity
• Climate change and global health emergencies
• Misinformation and declining trust
• Fragmented health and social systems

These forces actively shape whether interventions succeed or fail. Implementation science must evolve from controlled, linear thinking to dynamic, systems-based approaches.

What This Means for Public Health Research

1. From Controlled Trials to Real-World Learning Systems

There is a growing push toward learning health systems in which data, implementation, and evaluation occur simultaneously in real-world settings. For example, embedded trials using routine data (such as national audits) are being used to test interventions while they are already in practice. This represents a shift from:

• “Test, then implement”
  to
• “Implement, learn, and adapt in real time”

For public health researchers, this means:

• Designing studies that are flexible and iterative
• Embedding evaluation within systems rather than outside them
• Accepting trade-offs between control and relevance

2. The Rise of AI and Data-Driven Implementation

Another major theme was the integration of data science and artificial intelligence into implementation work. New platforms are being developed to:

• Automatically extract implementation barriers and strategies
• Map findings to frameworks like CFIR or behavior change models
• Accelerate evidence synthesis and decision-making

Early findings suggest these systems can achieve 70–80% accuracy in identifying key implementation concepts, dramatically reducing the time required for analysis. For public health, this could be transformative. Instead of waiting months or years for systematic reviews, practitioners could access near-real-time insights into what, where, and why something works (we’re actively working on this). But this also introduces new research challenges:

• How do we validate AI-generated insights?
• How do we ensure equity in algorithm-driven systems?
• Who governs these tools?

3. Rethinking “Impact”: Who Gets to Define Success?

A subtle but important shift discussed at the conference was the idea that impact is not universal. What researchers consider success may differ from:

• What policymakers prioritize
• What practitioners can realistically deliver
• What communities actually value

This creates tension and opportunity. Future research will need to:

• Co-define outcomes with stakeholders
• Incorporate qualitative and community-defined measures
• Move beyond narrow clinical or efficiency metrics

In other words, impact must become negotiated, not assumed.

What This Means for Public Health Policy

1. Policy Is No Longer Peripheral to Implementation

Historically, implementation science has focused more on programs and interventions than on policy itself. That is changing. Policy itself is an active driver of implementation success or failure. This includes:

• Regulatory environments
• Funding structures
• Political priorities
• Administrative processes

For policymakers, this means implementation science offers a new lens:

• Not just “What policy should we adopt?”
• But “How will this policy actually work in practice?”

2. The Challenge of Evidence Use in Policy Environments

Despite decades of work on evidence-based policymaking, the conference made clear that evidence does not naturally translate into policy decisions. Barriers include:

• Competing political agendas
• Time constraints and urgency
• Mismatched timelines between research and policy cycles
• Limited capacity for evidence synthesis

One emerging solution is knowledge mobilization organizations, which act as intermediaries between researchers and policymakers. These groups:

• Translate evidence into actionable insights
• Provide rapid reviews
• Support real-time decision-making

For public health systems, investing in these intermediaries may be just as important as funding research itself.

3. Climate Change as a Policy Implementation Stress Test

One of the most compelling discussions at EIE2025 focused on climate-related health policies. Researchers found that implementation is heavily influenced by how organizations perceive urgency:

• Acute events (like heatwaves) accelerate action
• Long-term risks often fail to sustain momentum

This creates a paradox:

• The most serious threats are often the hardest to act on

For policymakers, this highlights the need to:

• Align short-term incentives with long-term goals
• Build structures that maintain urgency over time
• Address workforce burnout linked to sustained crisis response

What This Means for Public Health Practice

1. Co-Creation Is Becoming the New Standard

One of the clearest practical shifts is the move toward co-design and co-creation. Instead of developing interventions in isolation, successful implementation now involves:

• Patients
• Practitioners
• Community members
• System stakeholders

In one example, adapting a medication adherence program required identifying 63 contextual factors and working through them collaboratively to tailor the intervention (yikes!). Interventions that are not co-designed are unlikely to be sustained.

2. Implementation Requires Infrastructure, Not Just Ideas

Another major theme was the lack of implementation infrastructure. Even when evidence exists, many systems lack:

• Dedicated roles for implementation
• Training programs
• Funding streams
• Data systems to support monitoring

Without this infrastructure, even the best interventions fail to scale. For public health organizations, this suggests a shift in investment priorities:

• From programs → to systems that support programs
• From short-term projects → to long-term capacity building

3. Economic Evaluation Is Missing and Critically Needed

One of the most striking gaps identified at the conference is the lack of economic evaluation in implementation science. Despite the high costs of implementation strategies, economic considerations are rarely integrated into decision-making. This creates real-world problems:

• Policymakers cannot compare strategies effectively
• Practitioners lack information on cost-efficiency
• Scaling decisions are made without financial clarity

Future practice will need to embed:

• Cost-effectiveness analysis
• Resource allocation modeling
• Stakeholder-informed economic frameworks

4. Bridging the Gap Between Frontline Reality and Scientific Models

Perhaps the most important practical insight is that implementation fails when it does not reflect frontline realities. This includes:

• Workflow constraints
• Staffing limitations
• Organizational culture
• Competing priorities

The conference emphasized the need for tools and guidance that are:

• Simple
• Adaptable
• Context-sensitive

Not every setting needs a complex framework. Sometimes, what practitioners need most is clear, actionable guidance that fits their environment.

The Bottom Line: A Field in Transition

Implementation science is moving:

• From linear → to systems thinking
• From researcher-driven → to stakeholder-driven
• From static evidence → to dynamic learning systems
• From isolated interventions → to integrated policy and practice ecosystems

For public health professionals, this is both a challenge and an opportunity. The challenge is to embrace complexity without becoming paralyzed by it. The opportunity is to finally close the gap between what we know and what we do.

What Comes Next for Public Health Leaders?

Impact will not come from better evidence alone. It will come from better implementation. That means:

• Designing research that works in the real world
• Building policies that anticipate complexity
• Creating systems that support sustained change

Public health has long been defined by its ability to respond to crises. The next era may be defined by its ability to implement solutions at scale, in context, and over time.