<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>topic Sample size re-estimation after interim analysis in Statistical Procedures</title>
    <link>https://communities.sas.com/t5/Statistical-Procedures/Sample-size-re-estimation-after-interim-analysis/m-p/931878#M46458</link>
    <description>&lt;P&gt;Hi,&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;My question concerns statistical theory and not SAS per se. However, there are a lot of qualified statisticians on this forum, so I hope no one takes offence in me posting this here&amp;nbsp;&lt;span class="lia-unicode-emoji" title=":smiling_face_with_smiling_eyes:"&gt;😊&lt;/span&gt;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&lt;STRONG&gt;Background:&lt;BR /&gt;&lt;/STRONG&gt;I am working with a clinical study that includes one interim analysis ¾-way into the trial in addition to the final analysis at the end. At the interim analysis (using data from 60 subjects), the primary endpoint variable was tested with a hypothesis test at a very low alpha of 0.1%, conserving the rest of the alpha (4.9%) to the final analysis (the O’brien-Fleming method). The study protocol stipulates that the study should be stopped in case the interim analysis is significant. Otherwise, the sample size should be re-estimated based on the interim analysis findings. The target number of subjects at the end of trial is currently 80, as specified in the protocol.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&lt;STRONG&gt;Result:&lt;/STRONG&gt;&lt;BR /&gt;The result of the interim hypothesis test yielded a p-value of around 0.017. Hence, a very promising result, although, not significant at the predefined alpha for the interim analysis.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&lt;STRONG&gt;Questions:&lt;/STRONG&gt;&lt;BR /&gt;Can the study be stopped now? I assume no, since we did not meet the 0.1% alpha level in the interim analysis. However, clearly, if the study was stopped, significance would be attained, since there is 4.9% alpha remaining. Can there be grounds to let the study continue, but not all the way to 80 subjects? I.e. this would amount to a decrease in sample size as a result of the sample size re-estimation. However, it is not clear to me how that can be done. I could perform a conditional power calculation, but to my knowledge this only serves to answer the question of the original target sample size would be enough or if it needs to be increased. I assume I cannot use this framework to &lt;EM&gt;decrease &lt;/EM&gt;the sample size?&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Thanks in advance for any input.&lt;/P&gt;</description>
    <pubDate>Wed, 12 Jun 2024 10:59:04 GMT</pubDate>
    <dc:creator>JoakimE</dc:creator>
    <dc:date>2024-06-12T10:59:04Z</dc:date>
    <item>
      <title>Sample size re-estimation after interim analysis</title>
      <link>https://communities.sas.com/t5/Statistical-Procedures/Sample-size-re-estimation-after-interim-analysis/m-p/931878#M46458</link>
      <description>&lt;P&gt;Hi,&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;My question concerns statistical theory and not SAS per se. However, there are a lot of qualified statisticians on this forum, so I hope no one takes offence in me posting this here&amp;nbsp;&lt;span class="lia-unicode-emoji" title=":smiling_face_with_smiling_eyes:"&gt;😊&lt;/span&gt;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&lt;STRONG&gt;Background:&lt;BR /&gt;&lt;/STRONG&gt;I am working with a clinical study that includes one interim analysis ¾-way into the trial in addition to the final analysis at the end. At the interim analysis (using data from 60 subjects), the primary endpoint variable was tested with a hypothesis test at a very low alpha of 0.1%, conserving the rest of the alpha (4.9%) to the final analysis (the O’brien-Fleming method). The study protocol stipulates that the study should be stopped in case the interim analysis is significant. Otherwise, the sample size should be re-estimated based on the interim analysis findings. The target number of subjects at the end of trial is currently 80, as specified in the protocol.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&lt;STRONG&gt;Result:&lt;/STRONG&gt;&lt;BR /&gt;The result of the interim hypothesis test yielded a p-value of around 0.017. Hence, a very promising result, although, not significant at the predefined alpha for the interim analysis.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&lt;STRONG&gt;Questions:&lt;/STRONG&gt;&lt;BR /&gt;Can the study be stopped now? I assume no, since we did not meet the 0.1% alpha level in the interim analysis. However, clearly, if the study was stopped, significance would be attained, since there is 4.9% alpha remaining. Can there be grounds to let the study continue, but not all the way to 80 subjects? I.e. this would amount to a decrease in sample size as a result of the sample size re-estimation. However, it is not clear to me how that can be done. I could perform a conditional power calculation, but to my knowledge this only serves to answer the question of the original target sample size would be enough or if it needs to be increased. I assume I cannot use this framework to &lt;EM&gt;decrease &lt;/EM&gt;the sample size?&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Thanks in advance for any input.&lt;/P&gt;</description>
      <pubDate>Wed, 12 Jun 2024 10:59:04 GMT</pubDate>
      <guid>https://communities.sas.com/t5/Statistical-Procedures/Sample-size-re-estimation-after-interim-analysis/m-p/931878#M46458</guid>
      <dc:creator>JoakimE</dc:creator>
      <dc:date>2024-06-12T10:59:04Z</dc:date>
    </item>
  </channel>
</rss>

