UC Berkeley

The widespread prevalence of chronic diseases has become a healthcare crisis worldwide. The incidence of chronic diseases has dramatically increased despite large expenditures in their prevention. More effective and efficient public health solutions are needed. To take advantage of novel findings in genomics, metabolism and physiology, innovative translational and implementation methods are needed. Operationally, innovation is understood as to develop a solution that achieves equal or greater effectiveness with less resource(s) than the previous ones.

How can we reduce the level of uncertainty when aiming for innovation in public health?

The main assertion of this dissertation is that, holding constant resources and time, we are more likely to develop an innovative solution with an iterative problem solving process of multiple cycles of feedback than with a more systematic solving process with only one cycle. This dissertation proposes the use of design sciences methods and principles to guide trial-error iterations during the innovation process.

Design sciences encompass a wide variety of disciplines that are engaged in the creation and manufacturing of novel products, objects and solutions such as in architecture, engineering, computer science, and artificial intelligence. Design theorists have distilled a set of principles to practice “evidence-based” design, by analyzing how different design disciplines tackle problems. Three of the most important principles are to break down complex problems into smaller, functional modules, to prioritize functionality of a solution over theoretical or technical preferences and to build solutions iteratively and by progressive approximation.

Design theorists have also concluded that some problems cannot be fully understood before solution generation begins. This can be contrasted with the traditional, but weaker, approach of investigators who define problems using only findings of past studies and pre- design interventions to address those problems before an intervention project begins—for example in a grant application. This approach keeps the problem definition and the intervention design fixed even as more information becomes available that challenges the original problem statement and intervention design. Design science theory and methods explain the drawbacks of this approach and the value of using more iterative, forward- oriented design methods.

The prospect of successfully supporting people to improve their health through dietary and fasting strategies motivated the development of the Dietary Intervention Canvas framework described in this dissertation. This framework is intended to guide the design and development of innovative and more effective dietary interventions. It integrates principles from design, behavioral and nutritional sciences with a novel approach to conceptualize dietary interventions. It is located at the intersection of translating experimental findings into clinical and population level interventions. This framework was built by carefully analyzing dietary behaviors across different cultural settings and life course stages. It was also based on an analysis of the behavioral strategies utilized by people who succeeded in long-term dietary change and weight loss.

The Dietary Intervention Canvas was used to tackle a problem that currently has high levels of uncertainty: remission of type 2 diabetes. Novel physiological findings reveal that remission of type 2 diabetes mellitus may be achievable. However, such experimental findings needed to be translated into interventions that consider patient characteristics and the cultural milieu of the target population.

In this dissertation, the Dietary Intervention Canvas was applied to distill key elements of the vast literature regarding diabetes causality, identify potential mechanisms of remission, translate them into medical nutrition therapy specifications, and formulate dietary intervention concepts with promising cultural acceptability and feasibility. The intervention was focused on patients in semi-rural settings in Mexico. Literature review, interviews with key stakeholders including members of the target population, healthcare providers and primary caregivers led to the development of multiple intervention concepts and prototypes. The result was a set of specifications to build intervention components that support patients to make dietary, psychological and physiological changes to achieve remission of the disease.

The dissertation is organized as follows. The first paper provides a summary of the analysis that describes the behavioral foundations of this dissertation. It addresses the seemly paradoxical observation that people interested in improving their health through dietary changes fail to sustain those changes. The second paper describes in detail the Dietary Intervention Canvas framework. The third paper describes its application to develop an intervention prototype for remission of type 2 diabetes within the Mexican context.

This dissertation describes a new path forward in the critical effort to develop more successful interventions by drawing on design science theory and methods as well as knowledge from multiple disciplines. The Dietary Intervention Canvas is a novel and promising framework intended to help researchers and practitioners support people to achieve seemingly unreachable goals—such as the remission of type 2 diabetes. Further research is needed to apply the Canvas in diverse populations and evaluate the outcomes.