“…This was used by 15 (20%) of the studies. 21 , 22 , 28 , 33 , 40 , 48 , 51 , 61 , 71 , 72 , 75 , 83 , 86 , 88 , 90 Eight of these 15 papers treated PROMs as an outcome only and sought to understand what clinical factors affect PROM scores, and/or used regression to understand how early PROMs were associated with longer term PROMs 21 , 33 , 40 , 48 , 51 , 61 , 83 , 90 ; and 2 papers treated PROMs both as a factor in regression analysis and as an outcome. 28 , 86 A fourth design identified was time-to-event studies, which used survival analysis to investigate outcomes, although this was used in only 3 (4%) papers.…”