The scientific method is fundamentally about establishing a hypothesis about a given intervention and testing it through quantifiable means. One very good example comes from the High Value Health Care Collaborative in the US. The Collaborative, which comprises premier research hospitals and health care organizations, proposed to test the efficacy of shared decision making in health treatments: for any given treatment, the doctor would engage the patient in a deep discussion about what the patient really wants and they jointly arrive at a decision on the application of the treatment. The Collaborative has tested this decision making method on various types of treatments, e.g. knee replacement surgery. One indicator it has used to demonstrate the efficacy of the method in terms of bringing down costs is the reduction in the length of stay of patients in the hospital for a given treatment. For knee replacement surgery for instance, the number of days declined significantly across health care regions in the United States. Using this type of evidence, they have been able to argue that shared decision making, which is basically an approach to address the asymmetry of information, is an intervention that improves the delivery of health care.
Shared decision making is an implementation intervention as opposed to a policy intervention. Improving delivery means improving implementation. Implementation in turn requires experimenting with possible interventions and, to learn whether an implementation intervention works or not, one must have indicators and data. In science of delivery, the “science” part is basically about rigorous hypothesis testing of implementation interventions.