Guidelines to understand and compute the number needed to treat.

IF 6.6 2区医学 Q1 PSYCHIATRY Evidence Based Mental Health Pub Date : 2021-11-01 DOI:10.1136/ebmental-2020-300232

Valentin Vancak, Yair Goldberg, Stephen Z Levine

{"title":"Guidelines to understand and compute the number needed to treat.","authors":"Valentin Vancak, Yair Goldberg, Stephen Z Levine","doi":"10.1136/ebmental-2020-300232","DOIUrl":null,"url":null,"abstract":"Objective: We aim to explain the unadjusted, adjusted and marginal number needed to treat (NNT) and provide software for clinicians to compute them.Methods: The NNT is an efficacy index that is commonly used in randomised clinical trials. The NNT is the average number of patients needed to treat to obtain one successful outcome (ie, response) due to treatment. We developed the nntcalc R package for desktop use and extended it to a user-friendly web application. We provided users with a user-friendly step-by-step guide. The application calculates the NNT for various models with and without explanatory variables. The implemented models for the adjusted NNT are linear regression and analysis of variance (ANOVA), logistic regression, Kaplan-Meier and Cox regression. If no explanatory variables are available, one can compute the unadjusted Laupacis et al's NNT, Kraemer and Kupfer's NNT and the Furukawa and Leucht's NNT. All NNT estimators are computed with their associated appropriate 95% confidence intervals. All calculations are in R and are replicable.Results: The application provides the user with an easy-to-use web application to compute the NNT in different settings and models. We illustrate the use of the application from examples in schizophrenia research based on the Positive and Negative Syndrome Scale. The application is available from https://nntcalc.iem.technion.ac.il. The output is given in a journal compatible text format, which users can copy and paste or download in a comma-separated values format.Conclusion: This application will help researchers and clinicians assess the efficacy of treatment and consequently improve the quality and accuracy of decisions.","PeriodicalId":12233,"journal":{"name":"Evidence Based Mental Health","volume":"24 4","pages":"131-136"},"PeriodicalIF":6.6000,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10231569/pdf/ebmental-2020-300232.pdf","citationCount":"10","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Evidence Based Mental Health","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1136/ebmental-2020-300232","RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"PSYCHIATRY","Score":null,"Total":0}

引用次数: 10

Abstract

Objective: We aim to explain the unadjusted, adjusted and marginal number needed to treat (NNT) and provide software for clinicians to compute them.

Methods: The NNT is an efficacy index that is commonly used in randomised clinical trials. The NNT is the average number of patients needed to treat to obtain one successful outcome (ie, response) due to treatment. We developed the nntcalc R package for desktop use and extended it to a user-friendly web application. We provided users with a user-friendly step-by-step guide. The application calculates the NNT for various models with and without explanatory variables. The implemented models for the adjusted NNT are linear regression and analysis of variance (ANOVA), logistic regression, Kaplan-Meier and Cox regression. If no explanatory variables are available, one can compute the unadjusted Laupacis et al's NNT, Kraemer and Kupfer's NNT and the Furukawa and Leucht's NNT. All NNT estimators are computed with their associated appropriate 95% confidence intervals. All calculations are in R and are replicable.

Results: The application provides the user with an easy-to-use web application to compute the NNT in different settings and models. We illustrate the use of the application from examples in schizophrenia research based on the Positive and Negative Syndrome Scale. The application is available from https://nntcalc.iem.technion.ac.il. The output is given in a journal compatible text format, which users can copy and paste or download in a comma-separated values format.

Conclusion: This application will help researchers and clinicians assess the efficacy of treatment and consequently improve the quality and accuracy of decisions.

查看原文

微信好友朋友圈 QQ好友复制链接

本刊更多论文

了解和计算治疗所需数量的指南。

目的:我们旨在解释未调整数、调整数和治疗所需的边缘数(NNT)，并为临床医生提供计算它们的软件。方法:NNT是随机临床试验中常用的疗效指标。NNT是指因治疗而获得一个成功结果(即反应)所需治疗的患者平均人数。我们开发了用于桌面的nntcalc R包，并将其扩展为一个用户友好的web应用程序。我们为用户提供了一个用户友好的分步指南。应用程序计算带有和不带有解释变量的各种模型的NNT。对调整后的NNT实施的模型是线性回归和方差分析(ANOVA)、逻辑回归、Kaplan-Meier和Cox回归。如果没有可用的解释变量，可以计算未经调整的Laupacis等人的NNT, Kraemer和Kupfer的NNT以及Furukawa和Leucht的NNT。所有的NNT估计量都是用相应的95%置信区间来计算的。所有的计算都是用R进行的，并且是可复制的。结果:该应用程序为用户提供了一个易于使用的web应用程序，可以在不同的设置和模型下计算NNT。我们举例说明应用程序在精神分裂症研究基于积极和消极综合症量表。该应用程序可从https://nntcalc.iem.technion.ac.il获得。输出以日志兼容的文本格式给出，用户可以复制粘贴或以逗号分隔值格式下载。结论:该应用程序将帮助研究人员和临床医生评估治疗效果，从而提高决策的质量和准确性。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文去求助

来源期刊

Evidence Based Mental Health PSYCHIATRY-

CiteScore

18.10

自引率

7.70%

发文量

期刊介绍： Evidence-Based Mental Health alerts clinicians to important advances in treatment, diagnosis, aetiology, prognosis, continuing education, economic evaluation and qualitative research in mental health. Published by the British Psychological Society, the Royal College of Psychiatrists and the BMJ Publishing Group the journal surveys a wide range of international medical journals applying strict criteria for the quality and validity of research. Clinicians assess the relevance of the best studies and the key details of these essential studies are presented in a succinct, informative abstract with an expert commentary on its clinical application.Evidence-Based Mental Health is a multidisciplinary, quarterly publication.