A Twenty-Year Look at Weight Gain


New Year’s is almost upon us, and for many it brings the usual resolutions – eating healthy, getting active, and ultimately losing weight. In this spirit, We talked to Arlene Hankinson, MD, instructor in preventive medicine at the Northwestern University Feinberg School of Medicine and lead author on a new study recently published in the Journal of the American Medical Association. It tracked the eating habits, physical activity and general health of more than 3,000 individuals over 20 years to gain insight into the development of heart disease over time. You may be surprised by some of the results.

Arlene Hankinson, MDArlene Hankinson, MDWhat were the basics of the study, and what did you find?
We started by looking at young adults between the ages of 18 and 30 and had them come back for repeated examinations (tracking physical activity, weight, and other health-related factors) over the next 20 years.

The people who had consistently high levels of physical activity over the twenty-year period gained the least amount of weight. For men, that translated into gaining about 22 pounds over the 20 years, while men with low physical activity gained about 28 pounds. For women the results were even more dramatic – women who maintained high physical activity gained 20 pounds over the 20 years, and women with low physical activity gained 33 pounds.

There were two things that were really interesting, just in those results. The first is that everybody gained weight, so having high physical activity didn’t eliminate the potential to gain weight over time. But, obviously people with high physical activity gained less. The other interesting thing was the gender difference. Women with high physical activity gained less weight than men with high physical activity compared to people in their same gender.

What do you think accounts for the gender difference?
It could be in part because we are measuring activity by self-report, meaning that people have to tell us what they do. Men are more likely to overestimate what they actually do in terms of physical activity, so it may have created a weaker association, because they’re inflating their true level of activity.

Another reason may be because men and women compensate differently in terms of their eating habits for their activity levels. You would expect people to eat more when they’re more active, but it may be that men eat even more than women with high activity, so it attenuates how much less weight is gained. But our diet data comes from self-report also, and women could underestimate how much they actually eat. So in the end, we don’t know. There could still be some true biological differences between men and women; it’s just that our study didn’t capture them. And there could be other behavioral differences that our study just didn’t capture as well.

What do you mean by “high activity”?
We used a questionnaire that measured about thirteen different kinds of activity, and we asked about how frequently the individual engaged in those activities, and for how long. Each activity also had an intensity scored associated with it. Using this information, we generated a score, and those in the top third of their gender based on their score were placed in the “high activity” category.

However, just being in the top third was not a big deal. It was staying in the top third over the twenty years that qualified the individual as maintaining a high level of activity. So you could be at a high activity level at any of the times you came back for your exam, but for our study, to be considered “high activity” you had to stay there. Only 11% of women and 12% of men did that.

What if you had low levels of activity when you were young and starting the study, but then improved your habits in middle age? Could you still end up in the high activity group?
No – our study was all about maintaining. You had to achieve it first, and then keep it up. The scenario you described would actually make you someone with “inconsistent activity,” and there is a large proportion of people who fell into this group. There were a lot of different scenarios, but the take-home message about inconsistent activity is that a lot a of people met it, and being inconsistent showed no difference in weight gain compared to people with consistently low activity.

That’s terrible!
It is. It suggests that activity cycling – continuous starting and stopping – may not be good. But the book isn’t closed on inconsistent activity. We don’t understand a lot about it. We still have to tease out the different ways you can have inconsistent activity – comparing starting out low and ending strong, and starting out strong and ending low.

Hopefully people will take home from this that it’s never too late to start, and that it’s more important for you to start something you can keep up, rather than shoot for some lofty goal that you can’t maintain.

It is New Year’s time, so a lot of goal setting will be happening.
I would just like for people to be as encouraged and realistic as possible. It’s more about small changes than big dramatic changes, and incorporating physical activity into your everyday life.

What were some of the challenges in keeping up with the same subjects over 20 years?
When you have a study that starts off with participants who are so young and in a period of change in their lives, there’s a lot of movement. There’s lost follow-up because people go away to college, or join the military, or their family moves. So trying to keep track of who originated from where and where they [are presently] is a big thing.

But CARDIA did a great job of keeping track of people, and one way they did this was by sending a birthday card every year. It was a way to re-engage people, let them know that we cared about them, and let them know that we’re still here. It really engendered good will, so people wanted to come back (for follow-ups).

Also, people who participate in these studies, by and large, want to help. They’re invested in it. We did lose around 30% [of our subjects], but a retention rate of 70% over a twenty-year period is a great statistic.

These findings are just one part of the larger Coronary Artery Risk Development in Young Adults (CARDIA) Study, funded by the National Institutes of Health. What work is left to be done?
This is really the best time for CARDIA, because this year we’re finishing up the year 25 examinations, so we’ll have five more years of data. Now that [our subjects] are in middle age, they’re developing a lot of the cardiovascular complications, and the whole purpose of the study is to look at how different risk factors evolve. So now we can actually see associations that we couldn’t see before. CARDIA lets us go back to the very beginning, and see how things change over time, and how heart disease develops.


Add new comment

This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.