計量生物学
Online ISSN : 2185-6494
Print ISSN : 0918-4430
ISSN-L : 0918-4430
42 巻, 1 号
選択された号の論文の4件中1~4を表示しています
原著
  • 張 方紅, 青木 誠, 柿爪 智行
    2021 年 42 巻 1 号 p. 15-32
    発行日: 2021年
    公開日: 2022/04/22
    ジャーナル フリー

    After the implementations of the revised “Ministerial Ordinance on Good Post-marketing Study Practice for Drugs” in April 2018, the Pharmaceuticals and Medical Devices Agency requires to conduct more efficient and effective post-marketing study (PMS). To design a PMS, it is important to define a research question based on the information of clinical trials, the characteristics of target diseases and the product profile. Nevertheless, designing a PMS faces some key challenges: (1) it is difficult to specify a research question due to strong exploratory aspects; (2) the results of PMS are not so useful for practical pharmacovigilance activities despite spending a huge resource/cost and (3) the accumulated experience and knowledge from existing clinical trials are not utilized. It is potentially difficult to address those challenges by the frequentist approach, although it can be used in PMS. We propose a Bayesian approach to address these three key challenges for a single arm PMS. Under the Bayesian approach, we can set and answer a research question in a probabilistic way and then offer some added values to frequentist method. Moreover, we discuss the sample size calculation for the research question based on the Bayesian approach and practical issues of implementing the Bayesian approach.

  • 濱口 雄太, 松嶋 優貴, 野間 久史
    2021 年 42 巻 1 号 p. 33-54
    発行日: 2021年
    公開日: 2022/04/22
    ジャーナル フリー

    In evidence-based medicine, meta-analysis is a relevant method for research syntheses. Random-effects model has been a primary statistical tool for meta-analysis since it enables a quantitative evaluation of the treatment effect accounting for the between-studies heterogeneity. In practices of meta-analyses, some studies may have markedly different characteristics from the others and such “outlying” studies might yield misleading results. For this problem, although several frequentists’ methods to detect outlying studies have been developed, there has been no effective Bayesian method to detect outlying studies and to assess their influence. In this article, we proposed influence diagnostic methods for meta-analyses using four Carlin-Louis-type influence measures; (a) relative distance, (b) standardized residual, (c) Bayesian p-value, and (d) scale parameters in scale mixture models. We also demonstrated the practical effectiveness of these proposed methods through applications to four meta-analyses for a spinal manipulative therapy, renin angiotensin system inhibitors, a known history of gestational diabetes, and antenatal corticosteroids.

研究速報
  • 河津 優太, 土田 潤, 安藤 宗司, 平川 晃弘, 寒水 孝司
    2021 年 42 巻 1 号 p. 55-64
    発行日: 2021年
    公開日: 2022/04/22
    ジャーナル フリー

    Phase I oncology clinical trials are designed to evaluate the safety of test drug/s and to determine the recommended dose for subsequent trials. However, the method to determine the required number of participants (sample size) has not been sufficiently developed for these dose-finding studies. The usual approach involves time-consuming calculations using numerical experiments to establish the required sample size that will allow the determination of accurate recommended dose of the test drug/s. In this study, we propose a time-saving method to determine the sample size. This method uses an alternative index of the proportion to accurately select the recommended dose. In the numerical experiments, we compared the performances (sample size, proportion of correctly selected recommended dose, and calculation time for the determination of sample size) of the proposed method with those of the conventional method. The sample size and the proportion of correctly selected recommended dose, determined by the proposed method, were slightly different from those calculated using the conventional method. The calculation time of the proposed method was consistently shorter than that of the conventional method. These results suggest that the proposed method can be used to roughly estimate the sample size of phase I oncology studies.

feedback
Top