check bug(?) Lee et al. (2018) outcome

[gabrielodom]

I think these "all missing" patterns should have a failure/relapse time at week 1, not 3.

> CTNote::outcomesCTN0094[c("who", "usePatternUDS", "lee2018_rel_time", "lee2018_rel_event")]
    who                        usePatternUDS lee2018_rel_time lee2018_rel_event
1     1                      ooooooooooooooo                3                 1
2     2                      *---oo-o-o-o+oo               12                 1
3     3              o-ooo-ooooooooooooooooo                7                 1
4     4             -------------------o-o-o               21                 0
5     5                      ooooooooooooooo                3                 1
6     6                      *oooooooooooooo                3                 1