Impact of changes in television viewing time and physical activity on longevity: a prospective cohort study

BACKGROUND Television viewing is a highly prevalent sedentary behavior among older adults, yet the mortality risks associated with hours of daily viewing over many years and whether increasing or decreasing viewing time affects mortality is unclear. This study examined: 1) the long-term association between mortality and daily viewing time; 2) the influence of reducing and increasing in television viewing time on longevity and 3) combined effects of television viewing and moderate-to-vigorous physical activity (MVPA) on longevity. METHODS Participants included 165,087 adults in the NIH-AARP Diet and Health (aged 50-71 yrs) who completed questionnaires at two-time-points (Time 1: 1994-1996, and Time 2: 2004-2006) and were followed until death or December 31, 2011. Multivariable-adjusted Cox proportional hazards regression was used to estimate Hazard Ratios and 95% confidence intervals (CI) with self-reported television viewing and MVPA and all-cause mortality. RESULTS Over 6.6 years of follow-up, there were 20,104 deaths. Compared to adults who watched < 3 h/day of television at both time points, mortality risk was 28% greater (CI:1.21,1.34) those who watched 5+ h/day at both time-points. Decreasing television viewing from 5 + h/day to 3-4 h/d was associated with a 15% reduction in mortality risk (CI:0.80, 0.91) and decreasing to <3 h/day resulted in an 12% lower risk (CI:0.79, 0.97). Conversely, adults who increased their viewing time to 3-4 h/day had an 17% greater mortality risk (CI:1.10, 1.24) and those who increased to 5+ h/day had a 45% greater risk (CI:1.32, 1.58), compared to those who consistently watched <3 h/day. The lowest mortality risk was observed in those who were consistently active and watched < 3 h/day of television. CONCLUSIONS We confirm that prolonged television viewing time was associated with greater mortality in older adults and demonstrate for the first time that individuals who reduced the amount of time they spent watching television had lower mortality. Our findings provide new evidence to support behavioral interventions that seek to reduce sedentary television viewing in favor of more physically active pursuits, preferably MVPA. Given the high prevalence of physical inactivity and prolonged television viewing in older adults, favorable changes in these two modifiable behaviors could have substantial public health impact. TRIAL REGISTRATION ClinicalTrials.gov number, NCT00340015.