Do LUTS Predict Mortality? An Analysis Using Random Forest Algorithms

IF 3.6 3区医学

Clinical Interventions in Aging Pub Date : 2024-02-12 DOI:10.2147/cia.s432368

Jonne Åkerla, Jaakko Nevalainen, Jori S Pesonen, Antti Pöyhönen, Juha Koskimäki, Jukka Häkkinen, Teuvo LJ Tammela, Anssi Auvinen

{"title":"Do LUTS Predict Mortality? An Analysis Using Random Forest Algorithms","authors":"Jonne Åkerla, Jaakko Nevalainen, Jori S Pesonen, Antti Pöyhönen, Juha Koskimäki, Jukka Häkkinen, Teuvo LJ Tammela, Anssi Auvinen","doi":"10.2147/cia.s432368","DOIUrl":null,"url":null,"abstract":"Purpose: To evaluate a random forest (RF) algorithm of lower urinary tract symptoms (LUTS) as a predictor of all-cause mortality in a population-based cohort. Materials and Methods: A population-based cohort of 3143 men born in 1924, 1934, and 1944 was evaluated using a mailed questionnaire including the Danish Prostatic Symptom Score (DAN-PSS-1) to assess LUTS as well as questions on medical conditions and behavioral and sociodemographic factors. Surveys were repeated in 1994, 1999, 2004, 2009 and 2015. The cohort was followed-up for vital status until the end of 2018. RF uses an ensemble of classification trees for prediction with a good flexibility and without overfitting. RF algorithms were developed to predict the five-year mortality using LUTS, demographic, medical, and behavioral factors alone and in combinations. Results: A total of 2663 men were included in the study, of whom 917 (34%) died during follow-up (median follow-up time 15.0 years). The LUTS-based RF algorithm showed an area under the curve (AUC) 0.60 (95% CI 0.52– 0.69) for five-year mortality. An expanded RF algorithm, including LUTS, medical history, and behavioral and sociodemographic factors, yielded an AUC 0.73 (0.65– 0.81), while an algorithm excluding LUTS yielded an AUC 0.71 (0.62– 0.78). Conclusion: An exploratory RF algorithm using LUTS can predict all-cause mortality with acceptable discrimination at the group level. In clinical practice, it is unlikely that LUTS will improve the accuracy to predict death if the patient’s background is well known. Keywords: lower urinary tract symptoms, mortality, machine learning, cohort studies ","PeriodicalId":10417,"journal":{"name":"Clinical Interventions in Aging","volume":"82 1","pages":""},"PeriodicalIF":3.6000,"publicationDate":"2024-02-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Clinical Interventions in Aging","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.2147/cia.s432368","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 0

Abstract

Purpose: To evaluate a random forest (RF) algorithm of lower urinary tract symptoms (LUTS) as a predictor of all-cause mortality in a population-based cohort.
Materials and Methods: A population-based cohort of 3143 men born in 1924, 1934, and 1944 was evaluated using a mailed questionnaire including the Danish Prostatic Symptom Score (DAN-PSS-1) to assess LUTS as well as questions on medical conditions and behavioral and sociodemographic factors. Surveys were repeated in 1994, 1999, 2004, 2009 and 2015. The cohort was followed-up for vital status until the end of 2018. RF uses an ensemble of classification trees for prediction with a good flexibility and without overfitting. RF algorithms were developed to predict the five-year mortality using LUTS, demographic, medical, and behavioral factors alone and in combinations.
Results: A total of 2663 men were included in the study, of whom 917 (34%) died during follow-up (median follow-up time 15.0 years). The LUTS-based RF algorithm showed an area under the curve (AUC) 0.60 (95% CI 0.52– 0.69) for five-year mortality. An expanded RF algorithm, including LUTS, medical history, and behavioral and sociodemographic factors, yielded an AUC 0.73 (0.65– 0.81), while an algorithm excluding LUTS yielded an AUC 0.71 (0.62– 0.78).
Conclusion: An exploratory RF algorithm using LUTS can predict all-cause mortality with acceptable discrimination at the group level. In clinical practice, it is unlikely that LUTS will improve the accuracy to predict death if the patient’s background is well known.

Keywords: lower urinary tract symptoms, mortality, machine learning, cohort studies

查看原文本刊更多论文

LUTS 能预测死亡率吗？使用随机森林算法进行分析

目的：评估下尿路症状（LUTS）作为全因死亡率预测指标的随机森林（RF）算法：使用邮寄问卷对 3143 名分别出生于 1924 年、1934 年和 1944 年的男性进行了评估，问卷包括丹麦前列腺症状评分（DAN-PSS-1）以评估下尿路症状，以及有关医疗状况、行为和社会人口因素的问题。1994年、1999年、2004年、2009年和2015年重复进行了调查。直到 2018 年年底，对该队列的生命状态进行了随访。RF 使用分类树集合进行预测，具有良好的灵活性，不会出现过度拟合。我们开发了 RF 算法，利用 LUTS、人口、医疗和行为因素单独或组合预测五年死亡率：研究共纳入了 2663 名男性，其中 917 人（34%）在随访期间死亡（中位随访时间为 15.0 年）。基于LUTS的RF算法显示，五年死亡率的曲线下面积（AUC）为0.60（95% CI 0.52-0.69）。包括 LUTS、病史、行为和社会人口因素在内的扩展 RF 算法的 AUC 为 0.73（0.65- 0.81），而不包括 LUTS 的算法的 AUC 为 0.71（0.62- 0.78）：结论：使用 LUTS 的探索性 RF 算法可以预测全因死亡率，在组别水平上具有可接受的区分度。在临床实践中，如果对患者的背景了如指掌，下尿路症状不太可能提高预测死亡的准确性。

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分钟内获得全文求助全文

来源期刊

Clinical Interventions in Aging GERIATRICS & GERONTOLOGY-

CiteScore

6.20

自引率

2.80%

发文量

193

期刊介绍： Clinical Interventions in Aging, is an online, peer reviewed, open access journal focusing on concise rapid reporting of original research and reviews in aging. Special attention will be given to papers reporting on actual or potential clinical applications leading to improved prevention or treatment of disease or a greater understanding of pathological processes that result from maladaptive changes in the body associated with aging. This journal is directed at a wide array of scientists, engineers, pharmacists, pharmacologists and clinical specialists wishing to maintain an up to date knowledge of this exciting and emerging field.