{"title":"不需要p值。“暴雷酒渣鼻:一种以抗炎为基础的治疗方法”分析决策评论","authors":"Yung Gonzaga, Ana Luísa Sampaio","doi":"10.1002/jvc2.533","DOIUrl":null,"url":null,"abstract":"<p>We read with interest the study by Handgretinger and colleagues entitled <i>Rosacea Fulminans: An Anti-inflammatory-based Therapeutic Approach</i>,<span><sup>1</sup></span> and although the topic addressed is of great clinical relevance, the presentation of the results, particularly concerning the choice of the statistical test, warrants discussion.</p><p>The study tracked the evolution of the skin condition in six6 patients diagnosed with Rosacea fulminans before and after treatment with azithromycin. According to the methodological section, a <i>t</i> test for independent samples was used to compare the before and after treatment. This choice seems inappropriate to us, as it overlooks some fundamental assumptions in selecting statistical tests that assess the association between two variables.<span><sup>2, 3</sup></span></p><p>The first issue concerns the independence between observations: The ‘before’ and ‘after’ measurements for the same individual are dependent, as they are related to the same subject. The independent samples <i>t</i> test assumes that the observations in each group are independent, which is not the case here.</p><p>The second issue concerns the nature of the variable being studied. The Clinical Erythema Assessment (CEA), for example, is an ordinal categorical variable. For instance, on a scale from 0 to 4, as in this case, the differences between ‘0 and 1’ may not represent the same quantitative change as a difference between ‘3 and 4.’ The <i>t</i> test assumes that the data are numbers, meaning that the differences between the values are consistent and quantifiable. For example, the difference between ages 25 and 20 is the same as the difference between 37 and 32, which is 5 years. The <i>t</i> test works with numerical variables, making it inappropriate for dealing with categorical variables, even if they are represented by numbers and have a natural order.</p><p>The third issue pertains to the necessity of a statistical test at all. In this regard, we would like to emphasize that the reason for writing this letter is not merely to criticize the choice of the statistical test, but to highlight that the value of a study like this should not rely on statistical significance. It is much more related to the descriptive and visual nature capable of creating a clear sensorial perception of clinical benefit and generating a relevant hypothesis to be explored more thoroughly in future studies.<span><sup>4</sup></span> In this regard, in our opinion, the study by Handgretinger and colleagues achieves the goal. It is a beautiful and illustrative case series that does not need <i>p</i> values to demonstrate its importance.</p><p>Yung Gonzaga conceived and wrote the article. Ana Luísa Sampaio wrote and reviewed the article.</p><p>The authors declare no conflict of interest.</p><p>Not applicable.</p>","PeriodicalId":94325,"journal":{"name":"JEADV clinical practice","volume":"3 5","pages":"1711-1712"},"PeriodicalIF":0.0000,"publicationDate":"2024-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/jvc2.533","citationCount":"0","resultStr":"{\"title\":\"No need for a p value. Comments on analytical decisions in “Rosacea fulminas: An anti-inflammatory-based therapeutic approach”\",\"authors\":\"Yung Gonzaga, Ana Luísa Sampaio\",\"doi\":\"10.1002/jvc2.533\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p>We read with interest the study by Handgretinger and colleagues entitled <i>Rosacea Fulminans: An Anti-inflammatory-based Therapeutic Approach</i>,<span><sup>1</sup></span> and although the topic addressed is of great clinical relevance, the presentation of the results, particularly concerning the choice of the statistical test, warrants discussion.</p><p>The study tracked the evolution of the skin condition in six6 patients diagnosed with Rosacea fulminans before and after treatment with azithromycin. According to the methodological section, a <i>t</i> test for independent samples was used to compare the before and after treatment. This choice seems inappropriate to us, as it overlooks some fundamental assumptions in selecting statistical tests that assess the association between two variables.<span><sup>2, 3</sup></span></p><p>The first issue concerns the independence between observations: The ‘before’ and ‘after’ measurements for the same individual are dependent, as they are related to the same subject. The independent samples <i>t</i> test assumes that the observations in each group are independent, which is not the case here.</p><p>The second issue concerns the nature of the variable being studied. The Clinical Erythema Assessment (CEA), for example, is an ordinal categorical variable. For instance, on a scale from 0 to 4, as in this case, the differences between ‘0 and 1’ may not represent the same quantitative change as a difference between ‘3 and 4.’ The <i>t</i> test assumes that the data are numbers, meaning that the differences between the values are consistent and quantifiable. For example, the difference between ages 25 and 20 is the same as the difference between 37 and 32, which is 5 years. The <i>t</i> test works with numerical variables, making it inappropriate for dealing with categorical variables, even if they are represented by numbers and have a natural order.</p><p>The third issue pertains to the necessity of a statistical test at all. In this regard, we would like to emphasize that the reason for writing this letter is not merely to criticize the choice of the statistical test, but to highlight that the value of a study like this should not rely on statistical significance. It is much more related to the descriptive and visual nature capable of creating a clear sensorial perception of clinical benefit and generating a relevant hypothesis to be explored more thoroughly in future studies.<span><sup>4</sup></span> In this regard, in our opinion, the study by Handgretinger and colleagues achieves the goal. It is a beautiful and illustrative case series that does not need <i>p</i> values to demonstrate its importance.</p><p>Yung Gonzaga conceived and wrote the article. Ana Luísa Sampaio wrote and reviewed the article.</p><p>The authors declare no conflict of interest.</p><p>Not applicable.</p>\",\"PeriodicalId\":94325,\"journal\":{\"name\":\"JEADV clinical practice\",\"volume\":\"3 5\",\"pages\":\"1711-1712\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-08-21\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://onlinelibrary.wiley.com/doi/epdf/10.1002/jvc2.533\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"JEADV clinical practice\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://onlinelibrary.wiley.com/doi/10.1002/jvc2.533\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"JEADV clinical practice","FirstCategoryId":"1085","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1002/jvc2.533","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
No need for a p value. Comments on analytical decisions in “Rosacea fulminas: An anti-inflammatory-based therapeutic approach”
We read with interest the study by Handgretinger and colleagues entitled Rosacea Fulminans: An Anti-inflammatory-based Therapeutic Approach,1 and although the topic addressed is of great clinical relevance, the presentation of the results, particularly concerning the choice of the statistical test, warrants discussion.
The study tracked the evolution of the skin condition in six6 patients diagnosed with Rosacea fulminans before and after treatment with azithromycin. According to the methodological section, a t test for independent samples was used to compare the before and after treatment. This choice seems inappropriate to us, as it overlooks some fundamental assumptions in selecting statistical tests that assess the association between two variables.2, 3
The first issue concerns the independence between observations: The ‘before’ and ‘after’ measurements for the same individual are dependent, as they are related to the same subject. The independent samples t test assumes that the observations in each group are independent, which is not the case here.
The second issue concerns the nature of the variable being studied. The Clinical Erythema Assessment (CEA), for example, is an ordinal categorical variable. For instance, on a scale from 0 to 4, as in this case, the differences between ‘0 and 1’ may not represent the same quantitative change as a difference between ‘3 and 4.’ The t test assumes that the data are numbers, meaning that the differences between the values are consistent and quantifiable. For example, the difference between ages 25 and 20 is the same as the difference between 37 and 32, which is 5 years. The t test works with numerical variables, making it inappropriate for dealing with categorical variables, even if they are represented by numbers and have a natural order.
The third issue pertains to the necessity of a statistical test at all. In this regard, we would like to emphasize that the reason for writing this letter is not merely to criticize the choice of the statistical test, but to highlight that the value of a study like this should not rely on statistical significance. It is much more related to the descriptive and visual nature capable of creating a clear sensorial perception of clinical benefit and generating a relevant hypothesis to be explored more thoroughly in future studies.4 In this regard, in our opinion, the study by Handgretinger and colleagues achieves the goal. It is a beautiful and illustrative case series that does not need p values to demonstrate its importance.
Yung Gonzaga conceived and wrote the article. Ana Luísa Sampaio wrote and reviewed the article.