The covariates used in the analyses were 1) whether an estimated death rate was used to convert a point estimate to a hazard ratio (yes or no); 2) whether standard error was estimated (yes or no); 3) the proportion of respondents who were male; 4) the mean age of the sample at baseline, divided by 10; 5) the age range of the sample at baseline, divided by 10; 6) the age of the publication (i.e., years elapsed since publication), divided by 10; 7) the age of the study (i.e., years elapsed since the collection of baseline data), divided by 10; 8) the duration of the baseline period, in years; 9) the number of years elapsed between the end of baseline and the beginning of follow-up; 10) the maximum follow-up duration, in years; 11) a series of interaction terms between gender, mean age, follow-up duration, and study age; 12) geographic region; 13) sample size, log transformed; 14) a series of variables that indicated the level of statistical adjustment; 15) a subjective quality rating (range, 1–3); and 16) the composite scale of study quality (continuous; range, 0–10). For descriptive statistics for select variables, see Table 1.