Gravar-mail: Are we assuming too much with our statistical assumptions? Lessons learned from the ALTTO trial