Scholarlink

Background <p>Long COVID (coronavirus disease) poses a substantial challenge to individual and global public health. In the context of China’s Omicron wave, key aspects such as its long-impact on non-hospitalized adults, the role of reinfection and other pathogenic infections during recovery require updated evidence. Therefore, the aim of this study is to delineate factors associated with long COVID and create a robust predictive model for primary-care to identify individuals at high risk of long COVID, utilizing data collected during the Omicron wave.</p> Methods <p>This study employed an online survey to assess SARS-CoV-2 infection status and track long COVID symptoms 9 to 15 months post-initial infection in 2099 participants. Associated factors of long COVID were identified through univariable and multivariable logistic regression analyses. The long COVID risk prediction model was visualized using a nomogram, and its performance was evaluated using the area under the curve (AUC) and a calibration curve.</p> Results <p>An analysis of 1,408 valid responses revealed that 35.9% of participants reported persistent long COVID symptoms 9 to 15 months after their initial SARS-CoV-2 infection. During the rehabilitation period, SARS-CoV-2 re-infection and other respiratory infections were reported in 32.4% and 34.3% of cases, respectively. The most prevalent long COVID symptoms were as follows: fatigue (16.2%), cough (9.0%), decreased activity tolerance (7.3%), shortness of breath (6.0%), expectoration (5.8%), and forgetfulness (5.7%). Multivariable analysis identified the factors independently associated with long COVID: female gender, absence of comorbidities, frequency of SARS-CoV-2 infections, history of other respiratory infections, and the presence of seven acute symptoms. Using logistic regression analysis, we developed a nomogram for long COVID prediction, which achieved an AUC of 0.731. Additional subgroup analysis revealed that participants with either SARS-CoV-2 reinfection or other pathogenic respiratory infections were more likely to report both a higher number of long COVID symptoms and increased symptom severity.</p> Conclusions <p>In summary, to mitigate the risk of long COVID, it is critical to prevent SARS-CoV-2 reinfection and other respiratory infections during the post-infection period. We developed a user-friendly nomogram model with satisfactory predictive performance to evaluate the risk of long COVID among COVID-19 patients.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Associated factors and predictive nomogram of long COVID: a cross-sectional study in China

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年3月12日

查看全文
  • Zhijin Zhang,
  • Yi Xue,
  • Leyi Gao,
  • Lujia Guan,
  • Haifan Zhang,
  • Haomiao Ma,
  • Xuyan Li,
  • Jieqiong Li,
  • Huqin Yang,
  • Zhaohui Tong

摘要

Background

Long COVID (coronavirus disease) poses a substantial challenge to individual and global public health. In the context of China’s Omicron wave, key aspects such as its long-impact on non-hospitalized adults, the role of reinfection and other pathogenic infections during recovery require updated evidence. Therefore, the aim of this study is to delineate factors associated with long COVID and create a robust predictive model for primary-care to identify individuals at high risk of long COVID, utilizing data collected during the Omicron wave.

Methods

This study employed an online survey to assess SARS-CoV-2 infection status and track long COVID symptoms 9 to 15 months post-initial infection in 2099 participants. Associated factors of long COVID were identified through univariable and multivariable logistic regression analyses. The long COVID risk prediction model was visualized using a nomogram, and its performance was evaluated using the area under the curve (AUC) and a calibration curve.

Results

An analysis of 1,408 valid responses revealed that 35.9% of participants reported persistent long COVID symptoms 9 to 15 months after their initial SARS-CoV-2 infection. During the rehabilitation period, SARS-CoV-2 re-infection and other respiratory infections were reported in 32.4% and 34.3% of cases, respectively. The most prevalent long COVID symptoms were as follows: fatigue (16.2%), cough (9.0%), decreased activity tolerance (7.3%), shortness of breath (6.0%), expectoration (5.8%), and forgetfulness (5.7%). Multivariable analysis identified the factors independently associated with long COVID: female gender, absence of comorbidities, frequency of SARS-CoV-2 infections, history of other respiratory infections, and the presence of seven acute symptoms. Using logistic regression analysis, we developed a nomogram for long COVID prediction, which achieved an AUC of 0.731. Additional subgroup analysis revealed that participants with either SARS-CoV-2 reinfection or other pathogenic respiratory infections were more likely to report both a higher number of long COVID symptoms and increased symptom severity.

Conclusions

In summary, to mitigate the risk of long COVID, it is critical to prevent SARS-CoV-2 reinfection and other respiratory infections during the post-infection period. We developed a user-friendly nomogram model with satisfactory predictive performance to evaluate the risk of long COVID among COVID-19 patients.