Today’s post comes from Jonathan Kurka and Matthew Buman, discussing a recent paper that they published in the American Journal of Epidemiology. Below is a video of Dr Buman explaining the main findings of his study, which was recorded in the fall of 2012. More on Mr Kurka and Dr Buman can be found at the bottom of this post.
It seems there is never enough time in the day, and this is often the reason people tend not to exercise even if they know they should. There are only 24 hours in a day and all of our daily behaviors can be broken down into three basic categories. Sedentary behaviors include any activity in which you aren’t active, such as sitting while watching TV, sitting while at a computer at work, driving your car, or lying down without sleeping. Active behaviors include walking, jogging, exercising, or even performing household chores such as cleaning, cooking, and vacuuming. Sleeping behaviors include naps and your nightly rest period.
Note: in research, we often break activity behaviors up, based on intensity, into light intensity (LIPA; activities that don’t require a lot of effort, but are still active, such as walking and cleaning) and moderate-to-vigorous (MVPA; activities that really make you sweat!) physical activity.
Because the total time in one day equals the sum of sleep, sedentary behavior, LIPA, and MVPA, if you increase time in one behavior, you’ll have to decrease time in another. So if we were able to add a 30 minute jog (that’s MVPA) to our day, what other behavior would we be doing 30 minutes less of? Would that jog replace watching a TV show (sedentary)? Or cleaning the kitchen (LIPA)? Or perhaps going to bed 30 minutes earlier (sleep)?
We know for many years of previous research that MVPA is associated with markers of cardiovascular disease (e.g. waist circumference, high-density lipoprotein (HDL) cholesterol, low-density lipoprotein (LDL) cholesterol, triglycerides, insulin levels, glucose, and C-reactive protein). It turns out that sleep duration is also related to many of these same biomarkers. Most recently, we have come to realize that sedentary behavior, independent of MVPA, is uniquely associated with these biomarkers as well. But how do all these behaviors fit together? What are the trade-offs, one behavior for another? This was the focus of our analyses. We sought to explore the effects of re-allocating 30 minutes/day from one behavior to another on markers for cardiovascular disease.
To do this we conducted isotemporal substitution models using data from the National Health and Nutrition Examination Survey (NHANES). NHANES is a program of studies designed to assess the health and nutritional status of adults and children in the United States. The program includes an in-person home interview and a laboratory examination in a mobile examination center. By incorporating a complex sampling strategy and weighting system (details can be found on the Centers for Disease Control and Prevention website), a sample that is statistically representative of the entire U.S. population can be created. After removing ineligible persons, 2,185 remained with data for waist circumference and HDL and 923 remained with data for triglycerides and insulin.
We used a statistical technique called isotemporal substitution modeling. These types of models have more commonly been used in studies involving food intake, but we applied this technique to the context of daily behaviors. These models estimate the “substitution association” of reallocating 30 minutes/day from one behavior to another (e.g., reallocating 30 minutes/day of sedentary behavior to 30 minutes/day of MVPA). We also controlled for several other variables to make sure any results were due to the re-allocation of behaviors and not due to the sex, ethnicity, marital status, education, work status, ratio of family income to poverty level, smoking, depressive symptoms, intake of total energy, saturated fat, caffeine, and alcohol; general health rating; previous diagnosis of cancer or malignancy, diagnosis of CVD or diabetes, or current use of diabetic, antihypertensive, lipidemic, or other CVD medication.
What did we find?
We’ve provided a graph that highlights our findings. Briefly, a relative risk < 1 (a marker to the left of the vertical line) indicates that by reallocating 30 minutes/day of one behavior to the other reduced the value for that cardiovascular risk factor. A marker to the right of the vertical line indicates an increased value in that risk factor from the reallocation of 30 minutes/day of one behavior to the other.
It is pretty clear from this graph that by reallocating 30 minutes/day of either sleep, sedentary, or LIPA behaviors to MVPA decreases waist circumference risk (that’s a good thing, as larger waist circumference indicates obesity), increases HDL cholesterol risk (that’s a good thing, because HDL is the “good” cholesterol), decreases triglycerides (that’s a good thing because triglycerides increase risk for obesity), and decreases insulin risk (that’s a good thing, as this is an indicator for diabetes). All good things!
Another interesting finding is that it doesn’t take MVPA to improve health. By reallocating sedentary behavior to LIPA, there was a 1.9% reduction in triglycerides and a 1.4% reduction in insulin levels and reallocating sedentary behavior to sleep resulted in a 2.4% reduction in insulin.
What’s the take home message?