When trying to lose weight, which is more important:

diet or exercise?

A national survey found that a "vast majority" of Americans

trying to control their weight

believed that food and beverage consumption

and physical activity were equally important.

Seven out of ten went with equally important,

about two out of ten thought exercise was more important,

and only about one out of ten chose diet.

The vast majority of Americans are wrong.

It's easy to understand how people might think

diet and exercise play equal roles.

After all, our weight is determined

by the balance of calories in and calories out.

What people may not understand

about this energy balance equation

is we have much more power over the "calories in" side.

In fact, on a day-to-day basis we have full control,

we could choose to eat zero calories

or 10,000 calories.

But most of the "calories out"

tend to be outside of our control.

Unlike wild animals who tend to burn

most of their calories on activity,

about 60 percent of our daily calories

are used up just to keep us alive,

what's called our resting, or basal, metabolic rate,

thanks in part to our energy-intensive brains.

Even if you stayed in bed all day,

you'd still burn more than 1,000 calories

just to fuel the basics like thinking, breathing,

keeping your heart pumping.

In contrast, even most "active" people

accrue less than two hours of exercise a week,

which may average out to be less than 100 calories

burned off each day.

That's only about 5 percent

of the total daily energy expenditure,

the "calories out" side of the equation.

So, the 2,000 calories we may take in every day from our diet

can exert 20 times more influence than exercise

over our weight destiny.

Most people believe that exercise is "very effective"

as a way to lose weight,

but this has been referred to as a myth

in the scientific literature.

In fact, it's been labeled

one of the most common misconceptions

in the field of obesity.

Yet virtually all formal weight-loss guidelines

include some sort of physical activity recommendation.

Can you outrun a bad diet?

Let's see what the science says.

Population studies have certainly found

strong correlations between physical inactivity and obesity.

But does a sedentary lifestyle lead to obesity,

or does obesity lead to a sedentary lifestyle?

It probably works a little bit in both directions.

To prove cause and effect,

and also quantify the relationship,

you really have to put it to the test.

Dozens of randomized controlled trials

involving thousands of participants

have been published on the effects of exercise

on weight loss.

Physical activity was not found to be an effective strategy.

For example, if you look at the studies

that tried using exercise alone to induce weight loss,

over an average of about five months,

people only lost about three pounds.

When you put all the studies together,

it looks like it took around six weeks of exercising

to get people to lose a single pound.

That was exercise alone, though.

What about as an adjunct to diet?

If you randomize people

into a diet and exercise intervention

versus just the diet alone,

the added exercise group does do better,

but the difference in weight loss

only averaged about two pounds.

The studies lasted between 3 and 12 months,

and all of that extra prescribed exercise

only seemed to translate into a few pounds.

The two-pound difference was statistically significant,

which means we're pretty sure it was a real effect,

but losing two pounds over a year's time

can hardly be considered clinically significant.

As a general rule, researchers like to see

at least a five- or six-pound drop.

In a meta-analysis of 18 randomized controlled trials

lasting a minimum of six months,

the diet-plus-exercise group

failed to significantly beat out the diet-only group at all.

There appeared to be no long-term benefit

to encouraging people to add exercise

to their weight-loss regimen.

What is going on?

Maybe exercise is just better

at preventing people from regaining weight.

No.

The vast majority of randomized controlled trials

examining weight-loss maintenance

also failed to show an exercise benefit.

Part of the problem is compliance.

It's one thing to tell people

to adhere to an exercise regimen;

it's another thing for them to actually do it.

When the same randomized controlled trials

were re-analyzed to exclude people

who flouted the instructions,

and analysis was limited to just those

who actually put in the time and sweat,

a clear advantage to exercise emerges.

Exercise only works if you actually do it,

though one reason people may become rapidly disillusioned

with their new gym membership

is their gross overestimation

of the capacity of exercise to burn off extra calories.

A slice of pizza has about 300 calories.

That converts into an hour of brisk walking per slice,

an hour a slice!

How many kids are jogging two hours a day

to burn off their happy meals?

Who's got time to climb 50 flights of stairs

to take care of the calories in just one Oreo cookie?

That's one of the reasons

that what we put in our mouths is most important.

Public health researchers have been experimenting

with including labeling on junk food.

Which label is more informative?

This one or these?

Still want that Toblerone if it means you have to walk

two hours?

Or cookies you'd have to skip rope

for 81 minutes for?

I think I'd run an hour just to avoid having to eat

shrimp-flavored potato chips.

Labeling fast food menus with little pictograms

of exercising stick figures was found to help nudge people

towards lower calorie options.

Seeing that the decision to supersize your fries

would mean walking an extra three miles that day

or that choosing the chicken salad

over the garden salad could mean having to run

nearly three miles,

people are more likely to make the healthier choice.