Telomere length and survival
Both the binomial regression model and the Cox time-dependent
proportional hazard model indicated that adult RTL is positively
correlated with survival. In the binomial model, corrected z-RTL was
statistically significantly related to survival to the next year, with a
slope of 0.39 (Table 3, Fig. 1). Age also had a statistically
significant quadratic relationship with survival, with early-life and
late-life survival being lower than mid-life (Table 3, Fig. 2). There
was no difference in survival between the sexes (Table 3).
Similarly, the Cox model showed a statistically significant and negative
relationship between corrected z-RTL and mortality. Corrected z-RTL had
a negative coefficient of -0.15 on survival, meaning that for every unit
increase in corrected z-RTL the hazard ratio is multiplied by a factor
of 0.86, i.e. a 14% decrease in mortality (Table 4). Age showed also a
quadratic, U-shaped effect on mortality (Table 4). There was no
significant effect of sex (Table 4).