image description

New study: This is how you “keep it off”

Just in time for your “Christmas in July,” a new study published in the prestigious journal JAMA (free full text here) tells us how we should eat to help “keep it off” or maintain our energy expenditure in a healthy way. As many people can attest, (temporarily) losing weight can seem easier than maintaining that weight loss. This is because the body acclimates to weight loss by reducing its energy expenditure (kinda like conserving water in times of drought). This new study looked at how different types of diets affect energy expenditure after a period of weight loss. So, let’s get into the details.

What did they do? They recruited 21 folks who were overweight or obese

Study design (Fig 1 from the paper). Click on the figure for larger view.

(mean BMI of 34.4) and between 18-40 years old. They had them go through a 4 week baseline period followed by a 12 week weight loss program to achieve at least a 10% weight loss. Following the weight loss, there was 4 week weight stabilization period before participants tried three types of diets that were matched on caloric content for four weeks each in a randomized order. The main outcome measure was resting energy expenditure, with secondary measures of total energy expenditure, hormone levels, and metabolic markers.

So what did these three maintenance diets look like? The three diets included a low-fat diet (what most people will tell you to eat – whole grains and lean meats); a low glycemic index diet (avoiding foods that spike blood sugar levels and insulin release); and a very low carbohydrate diet. This figure shows the major differences between the three diets using data from the paper.

Differences between the three diets (adapted from Table 1 of the paper). Click on the figure for larger view.

What did they find? First, as expected, losing weight resulted in reductions in resting and total energy expenditure (see below). However, eating the very low carb diet resulted in the smallest decreases in energy expenditure from the pre-weight loss baseline. And (this is almost too good to be true), these differences in energy expenditure were not due to differences in physical activity. In fact, while on the very low carb diet, folks had higher energy expenditure in spite of engaging in less physical activity!

Energy expenditure and physical activity (adapted from Table 3 of the paper). Click on the figure for a larger view.

Second, what happened to all of the markers that we might worry about when one is eating the very low carb diet that is high in saturated fat and cholesterol and low in fiber? Well, the results below show that insulin sensitivity was best on the very low carb diet, HDL (good) cholesterol was best on the very low carb diet, triglycerides were best on the very low carb diet, and systolic blood pressure was not best, but markedly improved relative to pre-weight loss on the very low carb diet. So far, so good…

Biomarkers on the three different diets (adapted from Table 3 of the paper). Click on the figure for a larger view.

But of course, nothing is perfect and the authors note that C-reactive protein (a marker of inflammation) tended to be higher with the very low carb diet and urinary cortisol (a marker of stress) was higher on the very low carb diet. Nonetheless, leptin levels were quite improved and the authors note that participant ratings of hunger and well-being did not differ between the three diets.

Additional biomarkers on the three diets (adapted from Table 3 of the paper). Click on the figure for a larger view.

So, what do we take away from this? This study has several strengths in that it included an initial weight loss phase with baseline assessments, the presentation of the diets was randomized, and the study used a crossover or within-subjects design so each person could serve as his/her own control. However, there are also several limitations, which include the relatively small sample size (N=21) and the relatively short duration of the test diets (4 weeks). It would be interesting to see longer-term results in a larger sample of people, but it is also always easier to design a perfect study than to conduct one.

As for the conclusions, we’ll leave it to the authors:

“In conclusion, our study demonstrates that

  • commonly consumed diets can affect metabolism and components of the metabolic syndrome in markedly different ways during weightloss maintenance, independent of energy content.
  • The low-fat diet produced changes in energy expenditure and serum leptin that would predict weight regain. In addition, this conventionally recommended diet had unfavorable effects on most of the metabolic syndrome components studied herein.
  • In contrast, the very low carbohydrate diet had the most beneficial effects on energy expenditure and several metabolic syndrome components, but this restrictive regimen may increase cortisol excretion and CRP.
  • The low–glycemic index diet appears to have qualitatively similar, although smaller, metabolic benefits to the very low-carbohydrate diet, possibly without the deleterious effects on physiological stress and chronic inflammation.
  • These findings suggest that a strategy to reduce glycemic load rather than dietary fat may be advantageous for weight-loss maintenance and cardiovascular disease prevention.
  • Ultimately, successful weight-loss maintenance will require behavioral and environmental interventions to facilitate long-term dietary adherence.”

And, that’s what we do here at Dan’s Plan. Cheers to these authors for a really great and important study.

Have a Happy (and safe) Fourth of July folks!

Reinforcement in action and good public policies

Earlier this week, we wrote about behavioral principles including positive reinforcement, negative reinforcement, and punishment. Today, we bring you two examples of folks putting those principles into action.

1) The first example comes from my own town of Bloomington, IN. When we first moved to Bloomington, we were surprised on garbage day when we saw that everyone’s trash had been picked up except ours! Not quite sure whether this was some sort of a hazing for the new folks (we do live in a college town, after all), we asked the neighbors what was going on. Their response, we didn’t use a trash sticker.

In Bloomington, each 35 gallon or 40 pound can, bag, or piece of trash must have

A sheet of trash stickers from the city of Bloomington, IN

a $2 trash sticker attached to it. The stickers (see right) can be purchased at any grocery or hardware store. As such, Bloomington residents are essentially punished for generating garbage.

Now, we’ve talked about some of the limitations of punishment and those limitations apply here as well. Some people will try to evade the punisher by dropping off their garbage in dumpsters around town. However, the trash stickers are only part of the story…

The city also picks up recycling “for free.” (note: there is obviously a cost for picking up recycling, but the resident does not endure the “pain of paying” with each bin put out to the curb). The beauty of the policies is realized when they are paired together. Garbage generation is punished at the same time that recycling is reinforced. Together, the two policies set up a strong incentive for individuals to “convert” as much waste as possible from trash that must be paid for at the curb to recycling that is picked up for free. Very clever (and effective) indeed (note: if you look carefully, these contigencies are actually stated on the stickers – another smart design!).

2) The second example comes from the city of Washington DC (although similar programs exist in California, Maine, Massachusetts, and Rhode Island). In DC, the Unity Health Care Upper Cardozo Clinic distributes “fruit and vegetable prescriptions,” which provide $1 per family member per day to be spent at Farmers Markets ($112 per month for a family of four). This clearly helps reduce the cost of buying fresh and local produce, and is a good incentive in its own right. Folks who have studied the programs have reported that:

  • 66% of people ate more fruits and vegetables as a result of the program and 38% improved their body mass index
  • The program also brought new customers to farmers markets. More than half of families that received fruit-and-vegetable prescriptions had never, or rarely, been to a farmers market

So, think about how providing these vouchers or reinforcers contrasts with

Example of a farmers market voucher

punitive or punishing policies such as taxes or bans (on soda pop, etc.). It is not to say that those policies are not effective at reducing the consumption of whatever is taxed or prohibited, but when implemented alone, they do not necessarily result in folks engaging in alternative behaviors such as buying fresh, local produce.

And it doesn’t end there. There were other effects associated with providing incentives for folks to go to the market. From the Washington Post article: “Anecdotally, the program is inspiring families to embrace other aspects of a healthful lifestyle, Lambke says. One family got friends to join the program. Another tried out a new walking trail. ‘The best quote was a kid who came back to me and said that the thing he liked about the program was that his parents played with him more,’ Lambke says. ‘Oh, man. That was awesome.’”

But is it cost-effective? In Lambke’s opinion, “They come back once a month. It’s not a huge amount of money. In the broad spectrum, when we think about how much we spend on Lipitor every year, it is a cheap, cheap intervention. And arguably more effective.”

I couldn’t have said it better myself – have fun at your farmers markets this weekend folks!

A brief primer on Behaviorism – don’t be afraid…

This past month I was shocked when I received the June issue of the Atlantic in the mail and the cover read, “THE END OF TEMPTATION: How the creepy science of behavior modification is reshaping our desires.” WTF?! The end of temptation? The creepy science of behavior modification? Reshaping our desires? I am a behaviorist and I am used to people not fully understanding many behavioral principles and much of Skinner’s work, but the Atlantic? Really?

Then I started to read the article (actually titled, “The Perfected Self”) by David H. Freedman and I started to calm down a bit (overzealous marketers must have been responsible for the cover). The piece actually provides a fairly balanced representation of Skinner’s work. However, there are still a few things that we should review – so here goes.

First, the introduction to the article states, “Skinner’s ideas are making an unlikely comeback today, powered by smartphone apps that are transforming us into thinner, richer, all-around-better versions of ourselves. The only thing we have to give up? Free will.” Damn. Really? Give up free will? This is an unfortunate repetition of a common misunderstanding of Skinner’s work. So, don’t worry folks, you can apply behavioral modification techniques without giving up free will – in fact this type of work has been very well-developed to help thousands of people with autism lead more fulfilling lives.

So, why are people so scared about their free will? It might be part semantics and part fear. The semantics part comes from the fact that most behaviorists talk about behavior “coming under the control of stimuli.” So, if I want to teach my dog to sit by showing her a treat, she might not initially sit all the time. However, if I repeatedly show her a treat (present the stimulus) and then give her the treat when she sits (reinforce the behavior); and importantly, not give her the treat when she doesn’t sit, over time she will (and indeed has) develop the habit of sitting when I show her a treat. As behaviorists we would say that her sitting behavior has come under stimulus control (of the treat). Now bear in mind that she has not relinquished any “free will.” In fact, she will often not sit for a treat if there are alternative reinforcers available, such as when we are outside and there is a squirrel nearby…

Which leads us to our second point regarding choice, free will, and fear - what should we really be concerned about here? Skinner’s work and behaviorism was perceived as threatening because people incorrectly thought that the demonstrations of how at external stimuli influence our behaviors meant that we no longer had any choice or that we had given up our control or free will. This is not true, and in fact, a better understanding of such behavioral principles allows us to use them to our advantage (see examples of this in the fascinating work being done by Brian Wansink and colleagues). Moreover, this irrational fear of how we might use behavioral techniques to our advantage ignores the fact that major corporations and food marketers have already developed similar techniques to get us to eat their garbage. Think for a second about all of the advertising, rewards programs, coupons, rebates, product placement, celebrity endorsements, etc. etc. that you see and hear every day. Have you ever thought about how many stimuli that you encounter on a daily basis that have been specifically designed to persuade you into behaving in a certain way? Conversely, do you want to design your own environments and your own stimuli to encourage you to do the things that you would actually like to do? If so, welcome to the science of behavior.

Lastly, let’s cover some of the basic principles of behavior modification so we have a common vocabulary and can avoid any confusion at the outset.

Operant (or Skinnerian) conditioning involves: 1) the presentation of a stimulus (could be a treat like the above example, an advertisement, driving by a fast food restaurant, or just walking in the door after work); 2) followed by a behavior; 3) followed by a consequence (another stimulus). This is distinct from classical or Pavlovian conditioning in which there is no consequence of the behavior or action. In Pavlovian conditioning, an conditioned stimulus (e.g., a bell ringing) is paired with an unconditioned stimulus (e.g., food presentation) such that an unconditioned response to the food (i.e., a dog salivating) comes to be associated with the bell ringing. When this happens, the salivation is now said to be a conditioned response to the bell and classical or Pavlovian conditioning has occurred. To get an idea of the potential strength and duration of classical conditioning, ask a former smoker if there are any sights, smells, or places that still make them want to light up a cigarette and see what they say. Make sure to note the (likely fond) look on their face as they recall those associations.

But, back to operant conditioning. There are three major ways in which operant conditioning occurs: positive reinforcement, negative reinforcement, and punishment.

Positive reinforcement occurs when the stimulus presented after the behavior increases the likelihood that the behavior will occur again. An example of this is the presentation of the treat to the dog after she sits. Giving the treat right after she sits increases the likelihood that she will sit again when I show her another treat. Praise might also serve as a positive reinforcer in this regard.

Negative reinforcement occurs when the removal of a stimulus after the behavior increases the likelihood that the behavior will occur again. So, if I

This embarrassed dog would like to know why she has to wear these beads

was a person who made my dog wear a silly outfit that she would prefer not to wear (because such behavior is embarrassing to the dog and to the owner), but I removed the outfit whenever she sat down, over time this would increase the likelihood that she would sit down (because dogs, like humans, prefer not to look silly). Note that reinforcement (positive and negative) always involves the increased likelihood that a behavior will occur, which is not to be confused with…

Punishment occurs when the presentation or removal of a stimulus decreases the likelihood that a behavior will occur again. Scolding or yelling at a dog would represent a punishing stimulus if it followed a particular behavior and decreased the likelihood that the behavior occurred again. Anyone who is familiar with the law (or children for that matter) should have a basic understanding of punishment and why it is not a very effective method for behavior modification. This is the conclusion that Skinner came to by the end of his career and it is largely due to the fact that individuals subjected to punishment often simply try to avoid the punishing stimulus (e.g., don’t get caught, sneak out when grounded) as opposed to not engaging in the behavior that a person is trying to discourage. Punishment is not a good way to effectively change behavior for the long-term.

So, we hope to have provided the basics of behavior modification and operant conditioning here. The science of behavior deserves a resurgence and we here at Dan’s Plan will champion that effort. Don’t fear behavioral modification. By learning about these principles and putting them into practice, you won’t have to give up any rights or your free will, but you might learn a lot about how stimuli in your environment shape your behavior.

Currently viewing page 10 of 100First8910111213203040Last