Time-series analysis of the 2013 to 2024 detection rate of carbapenem-resistant Klebsiella pneumoniae in a tertiary hospital.

IF 2.4 3区医学 Q2 INFECTIOUS DISEASES

American journal of infection control Pub Date : 2025-08-06 DOI:10.1016/j.ajic.2025.08.002

Bing Zhang, Lei Li, Yi-Ping Mao

{"title":"Time-series analysis of the 2013 to 2024 detection rate of carbapenem-resistant Klebsiella pneumoniae in a tertiary hospital.","authors":"Bing Zhang, Lei Li, Yi-Ping Mao","doi":"10.1016/j.ajic.2025.08.002","DOIUrl":null,"url":null,"abstract":"Background: The detection rate of carbapenem-resistant Klebsiella pneumoniae (CRKP) is on a rapid rise. The seasonal autoregressive integrated moving average (SARIMA) model is extensively employed in predicting the infectious diseases incidence. We hypothesized CRKP detection rates showed seasonal patterns.Methods: Data on the annual and monthly detection rates of CRKP for all inpatients in the affiliated hospital of Xuzhou Medical University from 2013 to 2024 were collected retrospectively. Annual data were used to identify the annual trends in CRKP detection rates. Monthly data from 2013 to 2023 were used as the training set to establish the SARIMA model, and monthly data from 2024 were used for testing to evaluate the model's predictive performance.Results: CRKP had an annual detection rate of 32% on average, with a linear decrease during the COVID-19 pandemic (2020-2022) (χ2= 44.652, P < .001). The time-series of monthly CRKP detection rates displayed long-term fluctuations and seasonal patterns, with the SARIMA (0,1,1)(0,1,1)12 model providing the best fit (BIC = 4.504, stationary R² = 0.371), and showing a reasonable prediction accuracy (MAPE = 29.26%). High-incidence periods for CRKP were observed in January, February, and May of each year.Conclusions: The SARIMA model holds value in CRKP detection rate prediction, but the prediction performance needs to be improved.","PeriodicalId":7621,"journal":{"name":"American journal of infection control","volume":" ","pages":""},"PeriodicalIF":2.4000,"publicationDate":"2025-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"American journal of infection control","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1016/j.ajic.2025.08.002","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"INFECTIOUS DISEASES","Score":null,"Total":0}

引用次数: 0

Abstract

Background: The detection rate of carbapenem-resistant Klebsiella pneumoniae (CRKP) is on a rapid rise. The seasonal autoregressive integrated moving average (SARIMA) model is extensively employed in predicting the infectious diseases incidence. We hypothesized CRKP detection rates showed seasonal patterns.

Methods: Data on the annual and monthly detection rates of CRKP for all inpatients in the affiliated hospital of Xuzhou Medical University from 2013 to 2024 were collected retrospectively. Annual data were used to identify the annual trends in CRKP detection rates. Monthly data from 2013 to 2023 were used as the training set to establish the SARIMA model, and monthly data from 2024 were used for testing to evaluate the model's predictive performance.

Results: CRKP had an annual detection rate of 32% on average, with a linear decrease during the COVID-19 pandemic (2020-2022) (χ²= 44.652, P < .001). The time-series of monthly CRKP detection rates displayed long-term fluctuations and seasonal patterns, with the SARIMA (0,1,1)(0,1,1)₁₂ model providing the best fit (BIC = 4.504, stationary R² = 0.371), and showing a reasonable prediction accuracy (MAPE = 29.26%). High-incidence periods for CRKP were observed in January, February, and May of each year.

Conclusions: The SARIMA model holds value in CRKP detection rate prediction, but the prediction performance needs to be improved.

查看原文本刊更多论文

某三级医院2013-2024年耐碳青霉烯类肺炎克雷伯菌检出率时间序列分析

背景：耐碳青霉烯肺炎克雷伯菌（CRKP）在中国的检出率呈快速上升趋势。季节性自回归综合移动平均（SARIMA）模型被广泛应用于传染病的发病率预测。我们假设CRKP检出率具有季节性模式。方法：回顾性收集2013 - 2024年徐州医科大学附属医院所有住院患者CRKP年、月检出率数据。使用年度数据确定CRKP检出率的年度趋势。使用2013 - 2023年的月度数据作为训练集建立SARIMA模型，并使用2024年的月度数据进行测试，以评估模型的预测性能。结果：CRKP的年平均检出率为32%，在2019冠状病毒病大流行（2020-2022）期间呈线性下降趋势(χ2=44.652， P12模型拟合最佳（BIC=4.504，平稳R²=0.371），预测准确率合理（MAPE=29.26%）。CRKP的高发期为每年的1月、2月和5月。结论：SARIMA模型在预测CRKP检出率方面具有一定的应用价值，但预测效果有待提高。

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

American journal of infection control 医学-传染病学

CiteScore

7.40

自引率

4.10%

发文量

479

审稿时长

24 days

期刊介绍： AJIC covers key topics and issues in infection control and epidemiology. Infection control professionals, including physicians, nurses, and epidemiologists, rely on AJIC for peer-reviewed articles covering clinical topics as well as original research. As the official publication of the Association for Professionals in Infection Control and Epidemiology (APIC)