Descriptive statistics for time to first hospitalisation for myocardial infarction (MI), fatal coronary heart disease or death by gender in the Framingham study and by treatment group for the Digoxin study
| Framingham heart study | Males | Females | All |
|---|---|---|---|
| 25th percentile of survival time (95% CI), years* | 14.4 (13.7 to 15.4) | 20.8 (20.0 to 21.8)* | 17.6 (16.9 to 18.1) |
| Restricted mean, years (over 24 years) | 18.9 (18.6 to 19.3) | 21.0 (20.8 to 21.2) | 20.1 (19.9 to 20.3) |
| Median follow-up time, years (censored observations only) | 24 (.,.) | 24 (.,.) | 24 (.,.)† |
| Median follow-up time, years (accounting for event times) | 24 (.,.) | 24 (.,.) | 24 (.,.)† |
| Digoxin study | Treatment | Placebo | All |
| 25th percentile of survival times (95% CI), years | 2.4 (2.2 to 2.5) | 2.4 (2.3 to 2.6) | 2.4 (2.3 to 2.5) |
| Restricted mean, years (over 4.9 years) | 3.7 (3.6 to 3.8) | 3.7 (3.6 to 3.7‡) | 3.7 (3.6 to 3.7‡) |
| Median follow-up time, years (censored observations only) | 3.6 (3.6‡ to 3.7) | 3.6 (3.6‡ to 3.7) | 3.6 (3.6‡ to 3.7) |
| Median follow-up time, years (accounting for event times) | 3.7 (3.7‡ to 3.7‡) | 3.7 (3.7‡ to 3.8) | 3.7 (3.7‡ to 3.7‡) |
*For the Framingham study it is not possible to estimate the median event time, because less than 50% of the study population is known to have had the event by the end of the observation period. Thus, we present the estimated time point at which 25% of the study population has the event, which is after 20.8 years among female participants with a 95% CI from 20.0 to 21.8 years.
†The distribution of follow-up times is unusual, because almost all participants with censored observations (2671 out of 2704) were observed for the maximum of 24 years. As a result it is not possible to estimate a 95% CI for the median follow-up times.
‡95% CI boundary is the same as estimate due to rounding.
CI, confidence interval.