Challenge phases Main content User comments entryt's statistics and author info Related themes, inspirations and entries Share entry Challenge activity feed Footer links

The Challenge


How might we improve the way citizens and governments interact? Read the challenge brief


Fitter City: Aggregating data locally to battle obesity

Fitter City allows those collecting information about their diet and exercise to contribute to a local aggregator, and for changes to be transparently mapped to public resources and campaigns.

It has become clear that one of the most substantial public health threats faced by the United States is the obesity epidemic. Meeting that challenge will require a combination of public health communication,design-based solutions, and other forms of research and activism. A number of cities have begun to focus more heavily on how to make their public spaces healthier,and what kinds of policies can help to encourage this. Some of these have attracted national attention: mayor Bloomberg’s requirements on transfats, menu labeling, and, most recently, soft drink sizes, for example. But cities pay attention to things like the American College of Sports Medicine’s annual report on healthiest metropolitan areas, and aim to be on that list.

At the same time, the “quantified self” movement has provided many Americans with the ability to track their own behaviors in some detail, and share this selectively with others. They record what they eat, what they weigh, how many steps they take, and a number of other metrics of their behavior and health. Sites like My Fitness Pal have over 30 million individual contributors. The gathered data are already fairly easily available, with time-stamp and often geographical tagging. In fact, APIs make many of these systems reasonably interoperable, with FitBit being able to gather data from MyFitness Pal, Microsoft HealthVault drawing from both, and dozens of other sites sharing data, usually on an individual basis. Very little large-scale aggregation is currently occurring. Even on the sites themselves, geographical aggregation tends to be clumsy or absent.

Good government means transparency about what efforts are being made and what effects these efforts have. Many local governments are experimenting with public policy, and while that policy may be data-driven, it is not always easy for citizens to access or influence that data. Fitter City provides a platform that allows for citizens to contribute their own metrics, and to include information about public policy effects. Does a new park in your neighborhood contribute to more steps taken? Does mandates calorie counts on menus or a new farmers’ market change the make-up of the local diet?

The data will draw on crowdsourced quantitative measures many are already engaged in, as well as news, announcements, and other qualitative data that can be layered onto our visualizations. Data will be coded for time and geographic region (using postal zip codes). While we will focus on drawing data first from the New York and Phoenix metropolitan areas,the platform will be available throughout the US.

Our approach leverages already existing technologies to provide for a crowd-sourced aggregation and transparency. By knowing what works, and where it works, not only can local citizens affect public policy(and be aware of local government initiatives), but other localities can benefit from the demonstrated results.


Rather than revise the above statement, we have opted to try to address some of the questions raised in the refinement process directly. Consider this a bit of a self-interview:

Q: Who asked for this site?

A: This is very much a case of giving people what they don’t already know they want. Obesity is a public health crisis is the United States and around the world, and has a deep negative impact not just on individual lives but on communities and our polity. It is also a complex and embedded problem that is difficult for media organizations to cover and for activists to coalesce around. Creating healthier places to live is an outcome that can be achieved with better policy designed and implemented in collaborations between citizens, local governments, and other local organizations.

Achieving this requires transparency. Over the last few years government agencies and health organizations have increased surveillance of public health, and there have been a number of exciting initiatives, especially in the largest cities. But this is a problem that cannot be solved top-down. It requires communities to come together around public policy that works on a local level. Citizens participating in gathering data in real-time about behavior and outcomes, and discussing this within a policy-oriented framework, represents not only our best chance for a broad intervention in this important area, but also a potential participatory framework for addressing other public policy challenges.

This is not a case of build it and they will come, but a recognition that there is an untapped potential for leveraging existing efforts and desires. By bringing together citizens own data with public policy efforts and providing a mechanism for continual monitoring of healthy communities, we can help to bring attention to the issue, and provide the space and some of the means for gaining traction.

What we propose is a pilot--the first iteration and a starting point for this process. One of the reasons we think it should exist is that no one knows they need it. And we think that the solution that makes the most sense is directly related to this Knight News Challenge: networking citizens and government.

Q: But aren’t people already tracking this stuff?

A: Yes, they are, and that’s precisely what we hope to leverage and redeploy in a useful way. Unfortunately, much of this data is not assembled in a way that makes it useful to policy makers, health care professionals, or concerned citizens.

Self-tracking, particularly when it comes to fitness and diet, is becoming extraordinarily popular. And many of these systems provide public or private APIs that allow for the exchange of data. As a result, many people are keeping fairly detailed logs, often making use of personal instrumentation, that far exceed the ability of ex post facto surveys for public health surveillance. Many of these sites offer some way to localize results and create online communities and networks of individuals, but the focus remains on personal changes and not on changes to the community environment. As a result, it can be difficult to make use of the localized features. (FitBit, for example, provides city-level discussion boards, many of which are filled with comments from people who express relief in finally figuring out how to find the board.) And it can be extraordinarily difficult to aggregate data by time and place from multiple sources.

Governments and non-governmental organizations also provide tracking of health within localities. This most frequently draws on telephone surveys of residents and of providers. One of the more recognized examples here is the CDC’s Behavioral Risk Factor Surveillance System (BRFSS), and their Environmental Public Health Tracking Network represents a well-funded effort to combine health outcomes reporting with environmental indicators at local levels. But this is just one of hundred of surveillance projects around the world that attempt to measure public health outcomes.

Much of this data is collected via surveys on a periodic (rarely more frequently than annual) basis. Again, we have no interest in competing against such data collection, but hope to include it as one of many sources of data in a space that allows for local citizens and governments to easily track and test interventions.

We are not reinventing the wheel, we’re just trying to build an axle.

Q: Sounds good. What will this look like?

A: We want to make something that works, something that can grow, and something that can inspire others to contribute. As such we are focusing on three things.

First, we want to make sure we are able to collect and effectively compare multiple sources of data. That means building a lightweight and modular infrastructure that can effectively draw from existing sources of both user-created data and more traditional sources of health surveillance data. We also want to be able to link this by time and place to more qualitative sources of news and information about community efforts to improve healthy behaviors.

Second, we want to make the interface as simple as possible for users to access and make sense of. We love some of the data-driven tools as much as anyone, and neat tools that let you combine and graph bizarre combinations of variables. Heck, give us dirty data and an hour with R and we’ll be as pleased as punch! But the average user needs this to just work. We will focus on an interface design that is thoroughly user-tested and provides a simple space for you to monitor your own activity as well as that of your community, to explore how your community differs from others, and to see how changes in policies and resources correlates to reported changes in behaviors and (eventual) health outcomes.

Finally, we want this to be a simple system that is built for expansion. We believe that this will be both popular and useful, and that by building the pilot, there will be a great proliferation in requests for, and interest in, expansion. And, we hope, that interest will come with resources that will make the project healthy and sustainable in the long run. To that end, we will design for both scalability and for modular expandability.

Q: But how do you do that, and how do other people draw on your work.

A: We are aware that the proposed project is not without risk. It may fail. But we hope that the work will provide a foundation for future projects whether it is a runaway success from the outset or eventually fails spectacularly. (Oh, and by the way, we think it will be a smashing success.) One of the ways to ensure that the project represents a learning opportunity even if it fails to meet its initial goals is to model the values we are trying to imbue the site with: transparency and participation.

The best way to do that is to proceed as openly as possible. That means the obvious things (Github is our friend, and we will share code, data, artifacts, and conversations widely), but also the hard work of being as inclusive of local stakeholders as possible. The only way we can succeed globally is to succeed locally, and the only way to do that is to be as inclusive as we can, drawing on the goodwill and contributions of the many, many dedicated people already working on this problem in the local community.

That local community is the greater Phoenix area, and while our focus in the pilot will be on developing resources and outreach where we live, from the outset the site will serve every community with citizens interested in contributing.

Q: Who is this we?

A: I (Alex Halavais) am lucky to have a wonderful cadre of interested colleagues and students to draw on in my own interdisciplinary college, at Arizona State University at large, and at various institutions and organizations throughout the Phoenix area.

This includes a large group at ASU and Mayo Clinic that have recently formed an initiative around innovative obesity solutions. My own experience crosses both technical development skills with a thorough grounding in the social sciences.

In many ways, just as the project itself aims to leverage existing resources to provide a new resource, I am hoping that funding provided by a grant from Knight will allow us to focus the efforts of a number of people locally who are already interested in this topic and in these kinds of projects.

Q: Where will the money be spent?

A: What is proposed does not require any breakthroughs of hard problems in technology, nor substantial organizational hurdles. It requires good design, community involvement, and sound project management. In turn, those things require some money to reach fruition.

In particular, we need to pay a designer and a developer, and we need resources for helping to facilitate meetings among stakeholders, both for assisting in design (charrettes, testing) and for assessment. Finally, we will need to put resources into a communications campaign locally to draw initial users to the site. Our requested amount to accomplish this is $205,000.

In each of these cases, we hope to draw on students and faculty of the university and integrate the design and research processes with our own courses and labs.

What is your project? [1 sentence max]

How great would it be to have one website that provided a gathering point for city-by-city diet and exercise, and its relation to data-driven municipal public policy?

Where are you located?

Phoenix, Arizona, USA

How did you hear about the contest?

  1. Word of mouth
Alexander's profile photo
entry submitted by: Alexander Halavais
March 09, 2013, 05:15PM
225 views 2 comments 15 applause Applaud


Join the conversation and post a comment.

April 01, 2013, 10:19PM
Since you mentioned MyFitnessPal, have you talked with their team about this? It would be really interesting to actually pull in data from their users, rather than having to recreate the userbase from scratch.
Alexander Halavais's reply to Nick Such's comment
April 02, 2013, 08:35PM
Thanks for the question.

No I haven't yet. I guess part of the issue here is that as a MFP user, I would be fine with my data being used that way, but I'm not sure that is true of all MFP users.

The second issue is though rebuilding the user base at a new site is not ideal, it does mean that those who have shared their data are also more directly involved in its use. I think if we are talking about citizens communicating with government openly, it's important to have their attention and voice come along with their data.

The way the project is currently envisioned is that users would permit FitterCity to pull the data from MFP and a range of other sites with either open APIs, private APIs that see value in the project (e.g., MFP), or some other means of permitted transfer of data.


Forgot my password?

New user? Sign up!