A Study of the Effects of TV Viewing

Does the number of hours of TV viewing as a child correlate with (i.e., predict) adult health outcomes, such as BMI, aerobic power, blood cholesterol level, or blood pressure? RJ Hancox and colleagues addressed this question in the study described in the abstract below.

Association between child and adolescent television viewing and adult health: a longitudinal birth cohort study. Hancox RJ, et al.: Lancet. 2004;64(9430):257-62.

Abstract

"BACKGROUND: Watching television in childhood and adolescence has been linked to adverse health indicators including obesity, poor fitness, smoking, and raised cholesterol. However, there have been no longitudinal studies of childhood viewing and adult health. We explored these associations in a birth cohort followed up to age 26 years.

METHODS: We assessed approximately 1000 unselected individuals born in Dunedin, New Zealand, in 1972-73 at regular intervals up to age 26 years. We used regression analysis to investigate the associations between earlier television viewing and body-mass index, cardiorespiratory fitness (maximum aerobic power assessed by a submaximal cycling test), serum cholesterol, smoking status, and blood pressure at age 26 years.

FINDINGS: Average weeknight viewing between ages 5 and 15 years was associated with higher body-mass indices (p=0.0013), lower cardiorespiratory fitness (p=0.0003), increased cigarette smoking (p<0.0001), and raised serum cholesterol (p=0.0037). Childhood and adolescent viewing had no significant association with blood pressure. These associations persisted after adjustment for potential confounding factors such as childhood socioeconomic status, body-mass index at age 5 years, parental body-mass index, parental smoking, and physical activity at age 15 years. In 26-year-olds, population-attributable fractions indicate that 17% of overweight, 15% of raised serum cholesterol, 17% of smoking, and 15% of poor fitness can be attributed to watching television for more than 2 hr. a day during childhood and adolescence.

INTERPRETATION: Television viewing in childhood and adolescence is associated with overweight, poor fitness, smoking, and raised cholesterol in adulthood. Excessive viewing might have long-lasting adverse effects on health."

Notice that the exposure (hours of TV viewing) and the outcomes in this study are both continuous measures. Chi-square tests could have been used if the investigators had categorized TV viewing (e.g., none, little, moderate, high) and also categorized the outcomes (e.g., obese or not), but they used continuous variables for both the exposure of interest and for the various outcomes, so we will need a different tool for assessing these data.