A few years ago, I was doing a summer internship in which I looked at health outcomes for hospitalized patients. I sat in an office and read about patients with issues like high blood pressure and cholesterol. At a certain point, I realized that the reports on their outcomes were interesting, but the real solution to the problems I was studying was happening outside my window. My window overlooked a park, where kids would run around all day until they were exhausted. And it got me thinking that if all kids were as active as those ones, there would a lot fewer reports for me to read.

So last year, I worked with several medical and law students to design a county-level childhood obesity prevention policy. The need for such programs is self-explanatory: More than one third of children in the U.S. are overweight or obese. By the time people reach adulthood, that proportion goes up to two thirds. By creating a team of both medical and law students, we hoped to come up with approaches that achieved the goal of improving health, and did so in a practical and implementable way.

Over the course of several months, we analyzed dozens of programs that have been used to bring down childhood obesity rates in various communities across the country. The programs ranged from well-known approaches (e.g. a soda tax or menu calorie counts) to some more obscure ones. My personal favorite was the walking school bus (WSB). Think about how your parents used to tell you that things were tougher in their day when they had to walk to school (in the snow, going uphill, barefoot, etc.). The goal of a WSB is to bring that world back. The catch is that parents/adults walk along a predetermined “bus” route, pick up kids along the way, and then walk them to school. Kids get a supervised walk that allows them to get some exercise every day.

Case studies, and one meta-analysis, suggest that WSBs are an effective way to increase the amount of exercise kids get. But odds are, you’ve never heard about them before. Neither have most school officials, local politicians, and others in a position to take action on childhood obesity. That’s because WSBs are not widely used. This realization led me to an interesting question: Which factors make a local program or intervention spread to other communities? What does it take to turn a single success story into a widespread strategy?

These are hardly new questions. Every business or non-profit that plans to scale up considers it. Atul Gawande, MD, attempted to figure out why certain medical interventions spread in a New Yorker article last year. Whether you’re talking about social programs, technology, or just an idea, the question remains. I don’t pretend to have the answer, but my work reviewing obesity prevention policies did lead me to a few conclusions about the spread of local programs.

First, success is necessary but not sufficient for a program’s spread. Just because it proves to be successful does not mean anyone else will adopt it. WSBs were one example. Granted, WSBs are not adaptable to every community — they require schools to be within walking distance and rely on good weather. But the same story is true for other approaches. For instance, joint-use agreements are a strategy where schools open up their facilities (e.g., outdoor fields, basketball courts, etc.) after school hours to give children and families access to recreational space. Despite a correlation between these agreements and better health outcomes, they remain in limited use in many of the communities where recreational space is most lacking.

So if success doesn’t lead to a program’s spread, what does? I believe one factor is the involvement and enthusiasm of multiple stakeholders, potentially including local government, businesses, school administrators, and involved community members. A second factor is the development of measurable and achievable goals. It is nearly impossible to see incremental changes in health outcomes, so programs designed to change health must establish metrics that can demonstrate progress.

The list of lessons from our survey of local programs goes on, but the biggest takeaway is clear. Problems in health care require not only a solution, but successful execution.

Akhilesh Pathipati is a medical student who blogs at Scope, where this article originally appeared.