Zarinara Alireza, Kamali Koorosh, Akhondi Mohammad Mahdi
Reproductive Biotechnology Research Centre, Avicenna Research Institute, Tehran, Iran.
Social Determinants of Health Research Center, Zanjan University of Medical Sciences, Zanjan, Iran.
J Family Reprod Health. 2021 Sep;15(3):179-185. doi: 10.18502/jfrh.v15i3.7136.
To analyze and compare four methods for estimating the chance of treatment success in infertile couples. In a retrospective cohort study, information on demographic and clinical features, including age, body mass index (BMI), duration of infertility, semen analysis, previous history of treatment and clinical examination of infertile couples were analyzed. Treatment success (childbearing) was calculated with four methods as live birth ratio, conditional probability and survival analysis (life table and Kaplan-Meyer method) and results are compared. The fertility ratio for the first treatment cycle was 29.72% which decreased to 23.13% by total treatment cycles. The success rate was 75.4%. With conditional probability calculation at the end of the five treatment cycles. With the life table method in a five-year period, the probability for live birth was 78% and by Kaplan-Meyer method 73.1% and the median of treatment time was 562 days. Calculation of infertility treatment success rate by only simple live birth ratio of childbearing couples is associated with underestimation. Using the conditional probability method reduces that underestimation, but it is not considered the censored cases in the treatments. It seems life table (as a proxy of survival analysis) presents the closest estimation to clinical facts with considering the repetition of the treatment cycle and the duration of treatment.
分析和比较四种评估不孕夫妇治疗成功几率的方法。在一项回顾性队列研究中，分析了有关人口统计学和临床特征的信息，包括年龄、体重指数（BMI）、不孕持续时间、精液分析、既往治疗史以及对不孕夫妇的临床检查。采用四种方法计算治疗成功（生育）情况，即活产率、条件概率以及生存分析（生命表法和Kaplan - Meyer法），并对结果进行比较。第一个治疗周期的生育率为29.72%，到整个治疗周期结束时降至23.13%。成功率为75.4%。在五个治疗周期结束时采用条件概率计算。采用生命表法在五年期间活产概率为78%，采用Kaplan - Meyer法为73.1%，治疗时间中位数为562天。仅通过生育夫妇的简单活产率来计算不孕治疗成功率会导致低估。使用条件概率法可减少这种低估，但它未考虑治疗中的截尾病例。考虑到治疗周期的重复性和治疗持续时间，生命表法（作为生存分析的替代方法）似乎给出了最接近临床实际情况的估计。
