“…Of the 80 RCTs, 75 reported the generation of random sequences, whereas the remainder mentioned only random assignments. Thirty-three studies ( 34 – 36 , 39 , 41 , 43 – 47 , 53 – 56 , 59 – 61 , 64 , 69 , 71 , 72 , 75 , 77 , 80 , 86 , 89 , 96 , 98 , 105 – 107 ) using random number tables, 21 ( 42 , 49 , 50 , 63 , 65 , 66 , 70 , 74 , 78 , 82 , 88 , 90 , 92 – 95 , 100 , 101 , 103 , 104 , 109 ) using computer allocation randomization, 15 ( 40 , 51 , 57 , 58 , 67 , 73 , 79 , 81 , 83 , 91 , 97 , 99 , 108 , 110 , 111 ) using the envelope method, and one ( 102 ) study using the lottery method were all rated as low risk. Three ( 32 , 33 , 76 ) studies randomly selected patients according to different treatment methods, one ( 87 ) according to the patient’s wishes, and one ( 84 ) according to the order of admission, all of which were rated as high risk.…”