The very premise of implementation science suggests that there is a science to putting evidence-based interventions into practice in real-world settings. Yet, unlike the natural and physical sciences, the field has no equivalent of a universal law that can explain how implementation unfolds across contexts despite the publication of over a hundred frameworks in the last two decades. Richard Feynman has famously called social science a ‘pseudoscience’ for its lack of invariant law to explain social phenomena. While reading John Gall’s The Systems Bible, I really enjoyed his humorous attempts in creating axioms and laws to explain systems thinking. In a similar spirit, this blog post provocatively explores the use of Newton’s laws of motion as a lens through which to understand implementation.
First Law: Implementation Inertia
An established routine practice tend to remain in practice, while new interventions tend to remain unimplemented, unless acted upon by sufficient forces of change.
Newton’s first law states that an object at rest remains at rest, while an object in motion remains in motion, unless acted upon by an external force. The same might be said of implementation. Promoting the uptake of evidence-based interventions often involves more than simply introducing something new. It requires changing or replacing existing workflows and practices. Behavioural economics teaches us that individuals are usually prone to status quo bias, with a tendency to prefer maintaining the current situation over change, even when potentially better alternatives are available. Established practices, routines, and ways of working exist for their own reasons and therefore have an inertia of their own. Viewed through this lens, Newton’s first law reminds us that change is inherently difficult. Implementation strategies must generate sufficient impetus to overcome such inertia, or current practices are likely to persist.
Second Law: Implementation ‘Force’
The likelihood of implementation success depends on the force of implementation efforts relative to the specific barriers to change.
Newton’s second law states that the acceleration of an object depends on its mass and the force applied to it. It is commonly expressed as F = ma, a formula many of us encountered in high school physics. In the context of implementation, acceleration can be thought of the pace of change or progress towards successful uptake of a new evidence-based intervention, force represents the strength and effectiveness of the efforts driving implementation, while mass corresponds to the magnitude of barriers standing in its way.
This law of motion has two core implications. First, greater force produces greater acceleration. Applied to implementation, the analogy suggests that stronger and more effective implementation efforts can increase the speed and likelihood of successful practice change. In research, we often think ‘if we build it, they will come’, assuming that evidence of an intervention’s effectiveness is adequate to encourage its adoption. The existence of implementation science as a field clearly indicates otherwise. Evidence-based interventions need to be accompanied by effective implementation strategies in place to facilitate uptake and help achieve positive implementation outcomes.
Second, the greater the mass, the more force is required to achieve the same acceleration. In implementation terms, the greater the barriers to change (i.e., inertia), the greater the implementation effort needed to overcome them. Deeply entrenched barriers within organisational structures and cultures, for example, will likely require more intensive efforts than surface-level barriers that can be readily addressed. Similarly, implementing a complex intervention may require greater efforts than introducing a relatively simple change. However, it is not just about more efforts but more importantly, choosing the right ones. Implementation strategies are not invariably effective per se. They only work and generate the ‘implementation force’ if they address the specific barriers at hand.
Third Law: Implementation Reaction
Every implementation effort to change a system generates a reaction from the system.
Newton’s third law states that for every action, there is an equal and opposite reaction, implying that forces always occur in mutual pairs. We experience such reaction forces in our everyday life. For example, when we walk, our foot pushes against the ground, and the ground pushes back with equal force, propelling us forward.
One of the most important concepts that I learnt from system thinking was that interventions are understood as events introduced into a system, which in turn generates its own system dynamics in response. Understanding implementation through the lens of Newton’s third law tells us that every implementation effort to change practice is likely to elicit a response from the system in which it is introduced, although unlikely to be an equal one. A new intervention may disrupt established routines, redistribute workloads, and alter roles and responsibilities, prompting individuals and organisations to respond accordingly. This differs from the construct of inertia in the first law, which represents the resistance arising from existing practice, while ‘reaction force’ refers specifically to the effects in response to imposing an implementation strategy or the new evidence-based practice itself.
Importantly, these ‘reaction forces’ are not necessarily bad. They are simply feedback from the system, which can inform how we recalibrate our implementation efforts or make adaptations to the evidence-based intervention. In her 1977 lecture, Donella Meadows wisely described the most basic feedback loop that governs our decision-making. We take action with the aim of changing the state of a system, while the existing state of the system in turn influences the decisions we go on to make. Likewise, I think implementation efforts are a perpetual iterative process, where we continuously learn from these reaction forces. Successful implementation is fundamentally aligning the fit between an evidence-based practice and the implementation context. Unfortunately, many interventions that show efficacy in controlled research settings are found to be poorly implemented in practice because they were not designed with implementation in mind. This law tells us that fit runs both ways. The reaction force is how the system informs us that the intervention, not just the setting, needs to change.
Is Implementation Science a Science?
Coming from a more positivist background, I have always wondered whether the field of Implementation Science meets the tenets that govern physical sciences. This blog post showcases an exploration of how we can apply Newton’s laws of motion to understand implementation. It is interesting to ponder whether the notion of establishing invariant laws is possible given the nuanced and complex nature of implementation. If that is not possible, can we still call it a science?