Day | No. of subjects remaining at risk | No. of cases of therapeutic failure | No. of subjects withdrawn for any reason | Interval risk^{b} | Cumulative risk^{c} |
---|---|---|---|---|---|

0 | 34 | 0 | 0 | 0 | 0 |

2 | 34 | 2 | 1 | 0.0597 | 0.0597 |

4 | 31 | 0 | 0 | 0 | 0.0597 |

7 | 31 | 3 | 1 | 0.0984 | 0.1522 |

11 | 27 | 5 | 2 | 0.1923 | 0.3152 |

14 | 20 | 4 | 0 | 0.2000 | 0.4522 |

18 | 16 | 2 | 0 | 0.1250 | 0.5207 |

21 | 14 | 0 | 1 | 0 | 0.5207 |

28 | 13 | 3 | 2 | 0.2500 | 0.6405 |

↵

*a*Reprinted from reference 15 with permission of the publisher.↵

*b*Interval risk was calculated as follows:*i*= [*N*− (*w*/2)]^{−1}, where*N*is the number of subjects remaining at risk,*i*is the number of cases of therapeutic failure, and*w*is the number of subjects withdrawn for any reason.↵

*c*The cumulative incidence of therapeutic failure (CR_{n}) was calculated as follows: CR_{n}= 1 − [(1 − IR_{n}) × (1 − CF_{n}_{− 1})], where IR is interval risk,*n*is the day of the test, and*n*− 1 is the prior interval. To calculate the CR for day 11, first estimate the interval risk as follows: 5[27 − (2/2)]^{−1}= 0.1923 and, finally, 1 − [(1 − 0.1923) × (1 − 0.1522) = 0.3152. Thus, the 11-day cumulative incidence (or risk) of therapeutic failure was 31.5%.