`km` in `fit.model` output names time, mean, upper and lower 95% CI as `time, surv, lower, upper` `make.surv` output names these as `t, S, low, upp`. This makes applying a common function to both of these more difficult.