What do natural experiments tell us about childhood obesity prevention approaches?

During the past decade, we’ve seen childhood obesity become one of our country’s most pressing public health and economic development issues. Government at all levels, and private companies, foundations, and organizations are pouring hundreds of millions of dollars into finding solutions.

In some respects, it’s a “throw the kitchen sink” approach. Let’s try a whole bunch of different things to improve nutrition/food access and increase physical activity, and we’ll make progress, right? This includes at the policy level. In fact, there is more interest now than 10 years ago for using public policy to change aspects of our environment that will help nudge us to making healthier decisions.

Minneapolis, MN
Minneapolis, MN

Sure, there are a variety of factors associated with childhood obesity, some environmental, some genetic, some individual behavior, some physiological. But when it comes to assessing what works and what doesn’t, public health researchers face two key challenges. First, it’s almost impossible to isolate these various factors. This complicates our task of trying to decipher which are causal and which are confounding. Second, well-designed, controlled scientific studies that assess the efficacy of interventions are difficult and costly. So, instead of relying on gold-standard randomized-controlled trials, we’re stuck with trying to interpret and extrapolate from observational, cross-sectional studies.

One alternative, which has it’s limitations, is a natural experiment. These studies don’t control for who receives the exposure of interest and who doesn’t. Okay, now in plain English. In real world scenarios, public policy interventions apply to everyone. There are no control and experimental groups. In this case, the only way to assess the potential impact of the policy is to compare outcomes before and after the policy was enacted. Or, you could compare between the population exposed to the policy and another that isn’t (for example, one state with another). These have considerable limitations, but they do offer some insights.

A recent review tried to assess natural experiments focused on addressing childhood obesity through environmental change. The researchers analyzed 37 studies from 2005-2013 that attempted to evaluate changes to the food and/or built environments and their impact on obesity outcomes. Studies focused either on improving diet and/or food access, or increasing physical activity.

Here’s what they found:

Regulatory approaches to improve the food environment showed greater impact than more voluntary approaches.

  • Studies that evaluated the impact of regulatory approaches to improve the food environment within schools and restaurants, such as trans fat bans, restrictions of sugary foods and beverages, or increased availability of fruits/vegetables reported more favorable results on consumer purchases and food consumption.
  • Studies of nutrition improvements to the federal WIC program found improvements in fruit and vegetable purchases and use of farmers markets. These improvements included food vouchers for low-income women and children to purchase more fruits, vegetables, whole grains, and lower fat milk, and payment system reform to allow these vouchers to be used at farmers markets.
  • Studies that assessed nutrition labeling did not find an impact on food purchasing or improving nutrition outcomes.

Active transportation interventions showed the strongest impact on physical activity-related outcomes.

  • Studies that looked at impacts on physical activity levels from improved access to greenspace and outdoor play/exercise equipment, including paths/trail, found positive results when follow-up was more than 6 months.
  • Active transportation interventions, such as those that help promote walking and biking, demonstrated positive results related to process outcomes, such as increased usage.


Though the study sheds some light on policy interventions related to childhood obesity prevention, there are some huge limitations.

  • I alluded to one with the active transportation studies. Most of these assess process outcomes, not obesity outcomes, such as changes in body mass index.
  • The second deals with follow-up time. For physical activity studies, “impacts were usually assessed within 1 year after implementation,” which meant anywhere from 2 to 14 months. Studies generally found more positive results with follow-up time greater than 6 months. This makes sense as it usually takes time for people to change their behavior based on an environmental change. However, studies haven’t typically assessed long-term usage past 1 year. Like adopting a new diet, weight loss is often observed in the short-term. But, after a year or more, many people slowly revert back to their previous weight before they began dieting. Does the same principle apply to physical activity?

Take Away

One of the main conclusions by the researchers from this study was that we still have a lot to learn when it comes to policy interventions that address childhood obesity. Many differ in their efficacy and as stated in the paper, “evidence is lacking on whether environmental and policy modifications are successful in maintaining healthy weight and/or reducing overweight.”


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s