TABLE 1

Number of Deaths That Would Require Trial Suspension and Review of Recruitment Relative to Infants Included

No. of Infants Included in the StudyStopping Rule for No. of Deaths
1Continuea
2Continuea
33
43
54
64
74
85
95
105
115
126
136
146
  • a Based on 25% mortality in the cooled cohort, it is not unlikely that the first 2 consecutive infants might die (P = .06). If the third infant also dies, the probability that this is a random event is very small (P = .016); thus, the study should stop and review after 3 deaths.