From evidence to action: what it takes to speed adoption in health care
From evidence to action: what it takes to speed adoption in health care
Together with BluePath Health, we recently brought together a small group of cross-sector health leaders in New York for an intimate dinner salon to tackle a shared challenge: why adoption in healthcare still moves too slowly, even when innovation is strong. The conversation centered on what decision-ready evidence really looks like, and how tools like clinical simulation and data sandboxes can help teams learn faster, reduce risk earlier, and move from evidence to action.
Health care does not suffer from a lack of innovation. It suffers from a lack of decision-ready insights at the moment decisions are made.
We convened a small, cross-sector salon in NYC to focus on a simple question: Why is adoption still slow, even when innovative solutions are promising? Leaders from across health care shared a common frustration: traditional evidence often arrives too late, fails to answer implementation questions, or does not meaningfully reduce risk when real decisions must be made.
There is growing demand for something different - rapid, robust, practical evidence generated in weeks or months, not years. Insights that help organizations decide whether to try, refine, pause, or expand an approach in real time. To ground the discussion, we explored clinical simulation and data sandboxes as concrete tools that can support faster learning and earlier risk reduction.
Why adoption is hard to accelerate
The challenge is not a shortage of ideas or pilots. It is the complexity of real-world change. Many solutions are not designed around how people actually work, whether its patients, clinicians, frontline staff or administrative teams. Workforce burnout often limits the capacity for change. Risk aversion reflects real regulatory, financial, and reputational constraints. Reimbursement and incentives often fail to support adoption, even when something works.
Data is rarely the problem; useful data at the right time is. Evidence often arrives after key decisions are already made.
Participants also highlighted an important distinction: different decisions require different levels of certainty. Scaling and policy decisions demand high confidence. But improvement and early adoption decisions need faster, “good-enough” insight. When all decisions are held to the same evidentiary standard, learning slows and innovation stalls.
From more evidence to better decisions
This points to a shift in focus: the goal is not producing more evidence, but supporting better decisions as they happen with the right evidence.
Adoption moves when solutions meet real needs, incentives reinforce the right behaviours, and teams can test, adapt, and improve. Without these conditions, even strong ideas struggle to take hold.
How simulation and sandboxes help
Clinical simulation and data sandboxes offer a practical way forward. They allow organizations to test ideas safely, surface operational issues early, and learn before real-world implementation. They support iteration rather than one-time evaluation and help match rigor to the decision being made. Most importantly, they can help leaders decide what to do next, before committing significant time, resources, and political capital.
The opportunity
There is a clear opportunity to shift learning earlier in the adoption process. Simulation and sandboxes make it possible to test assumptions and workflows before rollout, reducing risk upfront instead of discovering problems late. This approach complements traditional evaluation: rapid learning for improvement and adoption, rigorous methods for scaling and policy. Together, they create a faster, more confident path from innovation to impact.
Where this leaves us
Adoption remains one of the hardest parts of innovation in health care. Speeding it up requires changing how decisions are supported and how learning happens during adoption and implementation, not just generating more studies.
Simulation and data sandboxes offer one practical pathway to earlier testing, reduced risk, and faster movement from idea to impact. They help shift learning upstream, where change is easier and consequences are smaller.
This conversation is ongoing, and the opportunity to build better decision-support systems for adoption is only growing.
Get in touch for more information. You can find us at hello@provahealth.com.
About us
Prova Health helps innovators in digital health, AI, and life sciences transform healthcare for the better by providing world-class clinical, academic, and health system expertise. We support evidence generation, market access, and capability-building activities to help innovative technologies achieve real-world impact. Our team has decades of experience working across clinical practice, healthcare technology, policy, academia, and management consulting in over 30 countries. We work closely with a global network of healthcare professionals, academics, regulators, policymakers, and industry leaders to understand clinical practice, policy environments, and adoption pathways across major healthcare systems. Prova Health partners with organisations ranging from early-stage innovators to global healthcare and life sciences companies, combining strategic insight with practical delivery to translate innovation into impact.
BluePath Health is a women-owned small business that works with public agencies and health system partners to develop and implement solutions that support sustainable community health improvement. Our work focuses on delivering timely, analytically robust, and decision-relevant insight across Medicaid, Medicare, telehealth, behavioural health, health information exchange, and other cross-sector health initiatives. We serve as the operational partner for Connecting for Better Health, a non-profit coalition focused on advancing data sharing to improve health outcomes, including through the Community Sandbox, a secure environment that simulates real-world data sharing using synthetic or de-identified data to test workflows, governance, and interoperability before live implementation.