Skip to main content
. 2020 Aug 21;25:42. doi: 10.1186/s12199-020-00880-7

Table 3.

Regression analysis between TV time, age, gender, and skin color/ethnicity with cardiometabolic risk in adolescents

β Cardiometabolic risk
CI (95%)
Model 1 (AIC, 6062.282; BIC, 6107.834)
 TV time 0.002 (0.001; 0.003)
 Age − 0.029 (− 0.125; 0.066)
Gender
  Male 1
  Female 0.010 (− 0.369; 0.388)
Skin color/ethnicity
  White 1
  Black − 1.252 (− 1.998; − 0.540)
  Brown/mulatto − 0.571 (− 1.139; − 0.003)
  Indigenous 1.044 (− 1.214; 3.302)
  Yellow − 1.252 (− 3.506; 1.002)
Model 2 (AIC, 6058.391; BIC, 6108.814)
 TV time 0.015 (0.005; 0.026)
 Age 0.109 (− 0.036; 0.255)
 TV time × age − 0.001 (− 0.002; − 0.001)
Model 3 (AIC, 6064.084; BIC, 6114.698)
 TV time 0.002 (0.001; 0.004)
Gender
  Male 1
  Female 0.113 (− 0.477; 0.702)
 TV time × gender (ref. male) 1
 TV time × gender − 0.001 (− 0.004; 0.003)
Model 4 (AIC, 6068.321; BIC, 6134.118)
 TV time 0.009 (− 0.011; 0.029)
Skin color/ethnicity
  White 1
  Black − 1.396 (− 2.497; − 0.296)
  Brown/mulatto − 0.687 (− 1.558; 0.184)
  Indigenous 2.541 (− 0.905; 5.986)
  Yellow − 2.783 (− 7.683; 2.116)
 TV time × white 1
 TV time × black 0.001 (− 0.004; 0.006)
 TV time × brown/mulatto 0.001 (− 0.003; 0.005)
 TV time × indigenous − 0.006 (− 0.017; 0.005)
 TV time × yellow 0.007 (− 0.013; 0.027)

CI confidence interval, AIC Akaike information criterion, BIC Bayesian information criterion, TV television, β linear regression