Response to Letters Regarding Article, “Television Viewing Time and Mortality: The Australian Diabetes, Obesity and Lifestyle Study (AusDiab)”

Abstract

We thank Drs Ding, Mekary, and Katz for their interest in our recent findings showing a positive association between television viewing time and mortality. 1 Drs Ding and Mekary helpfully bring to our attention the relevance of their isotemporal substitution model to our findings. Consistent with their perspective, we acknowledge that the basic partition model applied in our analyses has limitations with respect to not being able to account for total activity time. As a potential explanation of the relationship between television viewing time and mortality, their "displacement" hypothesis is intriguing. However, as noted in our article, we and others, including Mekary et al, 2 have previously shown television viewing time to be only weakly correlated with moderate to vigorous leisure-time exercise. Studies using objective measures of physical activity (mainly accelerometers) show clearly that only a small proportion of total waking hours is spent engaging in moderate to vigorous exercise (Ͻ5%), with the balance being distributed between sedentary time and light-intensity physical activity. 3 With this in mind, we speculated in our article that television viewing may displace opportunities to engage in lightintensity activities, which have been shown to be beneficially associated with cardiometabolic biomarkers, including 2-hour postchallenge blood glucose. However, capturing time spent in lightintensity physical activity with the use of self-reported measures has been notoriously problematic for physical activity epidemiology research and is also a problem for sedentary behavior studies. This is largely due to recall and measurement error. We certainly see scientific virtue in addressing this "displacement across the full 24-hour day" hypothesis as a potential explanation for the relationships that we have reported. The isotemporal substitution model could be considered in future studies to better understand such relationships. A significant limitation in doing so will be that self-reported measures of physical activity, including our own, have less than perfect reliability and validity, which inevitably will compromise what can be derived from these sophisticated models. The increasing adoption of objectively measured physical activity and sedentary time approaches may offer new insights into the relevance of the displacement hypothesis and help to minimize the residual bias that presently exists with self-reported measures.Dr Katz correctly points out (as we also have done in our article) that a baseline-only assessment of television viewing and leisuretime exercise and the inability to account for all other unmeasured or unknown confounding factors are limitations. However, these are inherent limitations that our findings have in common with a plethora of prospective epidemiological cohort studies, all showing significant relationships of exercise and physical activity with premature mortality. We are encouraged that since the publication of our findings, positive associations of television viewing with prema...

Screen-Based Entertainment Time, All-Cause Mortality, and Cardiovascular Events

“…Our all-cause mortality results are in agreement with a large study of Canadian adults who were followed up for 12 years, where the all-cause mortality HR for the highest category of nonrecreational (work, school, housework) daily sitting time ("almost all the time") was 1.54 (26). Another study among Ͼ8,000 Australian adults reported a very similar all-cause mortality HR (1.46) for Ն4 h of TV watching compared with the Ͻ2 h/day reference group (23). Both studies were also robust for adjustments or stratifications by sex, physical activity level, smoking, and BMI.…”

“…Although we found no evidence of a dose-response relationship, our analysis suggests that a threshold of Ն2 h/day of screen time might be linked to an increased risk for a CVD event. The Australian study (23) found that daily TV viewing times in excess of 4 h/day (but not 2 to 4 h/day) were associated with CVD death risk. We speculate that this disagreement occurs because our exposure variable was more inclusive than TV alone that was used in the Australian study and because our CVD outcome included nonfatal as well as fatal events.…”

Physical activity and cancer prevention: a systematic review of clinical trials

“…Increased time spent in sedentary behaviors has been associated with colorectal, endometrial, ovarian, and prostate cancer risk [79] as well as insulin resistance, elevated CRP [80] and all-cause mortality [81]. These associations are generally independent of time spent engaging in moderate-vigorous intensity physical activity [82].…”

Sleep duration and activity levels in Estonian and Swedish children and adolescents

“…In addition to the health benefits of being highly active, health risks are related to sedentary behaviours (Dunstan et al 2010;Hamilton et al 2007;Healy et al 2008). Sedentary behaviour or inactivity refers to activities that do not increase energy expenditure substantially above the resting level and includes activities such as sleeping, sitting, lying down, and watching television, and other forms of screen-based entertainment (Pate et al 2008).…”