Anastasia Shikhaleva

Background: Post-COVID-19 condition (PCC) affects many survivors, with evidence of sex-specific differences in prevalence and symptom profiles. However, few prediction studies have examined whether sex-stratified models improve prediction or generalize across sexes. This study aimed primarily to develop and compare sex-stratified machine learning models for PCC prediction using routinely available baseline variables, and secondarily to assess cross-sex generalizability and adversarial robustness. Methods: We analyzed a prospective longitudinal cohort of 1006 adults hospitalized with COVID-19 at Sechenov University Hospital Network (Moscow, Russia). Demographics, smoking status, and pre-existing comorbidities were extracted from medical records, and PCC status was assessed at 6-month follow-up. Machine learning models—including classical algorithms and graph-based neural networks—were trained separately for males and females. Cross-sex validation evaluated generalizability, variable importance aided interpretation, and adversarial perturbations assessed model robustness. Results: PCC prevalence was higher in females (53.9%) than males (39.1%). Overall predictive performance was modest across all models, with AUC values ranging approximately 0.50–0.61. Graph-based models achieved the highest discrimination, with the best AUC reaching approximately 0.61, while classical approaches provided limited predictive value. Cross-sex validation showed minor asymmetry: models trained on male data performed slightly better on female cases than vice versa. Adversarial testing revealed sensitivity of all models to input perturbations. Conclusions: Demographics and comorbidities alone provide insufficient information for reliable PCC prediction. Modest sex-specific differences in model generalizability suggest distinct, sex-associated PCC phenotypes, but richer multimodal data—including clinical biomarkers, wearable-derived measures, and patient-reported outcomes—will be required to develop clinically useful and equitable predictive models. Sex-stratified approaches should be considered in future post-viral syndrome prediction studies.

Background: The COVID-19 pandemic has led to widespread long-term complications, known as post-COVID conditions (PCC), particularly affecting vulnerable populations such as cancer patients. This study aims to predict the incidence of PCC in hospitalised cancer patients using the data from a longitudinal cohort study conducted in four major university hospitals in Moscow, Russia. Methods: Clinical data have been collected during the acute phase and follow-ups at 6 and 12 months post-discharge. A total of 49 clinical features were evaluated, and machine learning classifiers including logistic regression, random forest, support vector machine (SVM), k-nearest neighbours (KNN), and neural network were applied to predict PCC. Results: Model performance was assessed using the area under the receiver operating characteristic curve (AUC), sensitivity, and specificity. KNN demonstrated the highest predictive performance, with an AUC of 0.80, sensitivity of 0.73, and specificity of 0.69. Severe COVID-19 and pre-existing comorbidities were significant predictors of PCC. Conclusions: Machine learning models, particularly KNN, showed some promise in predicting PCC in cancer patients, offering the potential for early intervention and personalised care. These findings emphasise the importance of long-term monitoring for cancer patients recovering from COVID-19 to mitigate PCC impact.

Despite the impressive progress in diagnosis and management of acute COVID-19, data regarding the consequences of this infection are just emerging. The World Health Organization has proposed the term post-COVID condition (PCC) to describe the wide range of sequelae of acute COVID-19. With more than 200 million confirmed cases of COVID-19, PCC may develop into a major problem for many years to come for the millions of COVID-19 survivors worldwide. Few studies were conducted in primary care, and very few studies have focused exclusively on children and adolescents.Objective. To review existing data on PCC. Analysis of manuscripts published in peer-reviewed journals and clinical protocols. PCC is characterized by a wide range of systemic, cardio-pulmonary, gastrointestinal, neurological, and psychosocial symptoms.Conclusion. Although PCC prevalence is difficult to estimate due to methodological limitations of the existing studies, there is no doubt that this problem is a significant healthcare burden. There is a need for further observational and interventional studies to establish optimal PCC prevention and management strategies.

Background The long-term sequelae of coronavirus disease 2019 (Covid-19) in children remain poorly characterised. This study aimed to assess long-term outcomes in children previously hospitalised with Covid-19 and associated risk factors. Methods This is a prospective cohort study of children (≤18 years old) admitted with confirmed Covid-19 to Z.A. Bashlyaeva Children’s Municipal Clinical Hospital in Moscow, Russia. Children admitted to the hospital during the first wave of the pandemic, between April 2, 2020 and August 26, 2020, were included. Telephone interview using the International Severe Acute Respiratory and emerging Infection Consortium (ISARIC) Covid-19 Health and Wellbeing paediatric follow up survey. Persistent symptoms (>5 months) were further categorised by system(s) involved. Findings Overall, 518 of 853 (61%) of eligible children were available for the follow-up assessment and included in the study. Median age was 10.4 years (IQR, 3–15.2) and 270 (52.1%) were girls; median follow-up since hospital discharge was 256 (223-271) days. At the time of the follow-up interview 126 (24.3%) participants reported persistent symptoms among which fatigue (53, 10.7%), sleep disturbance (36, 6.9%,) and sensory problems (29, 5.6%) were the most common. Multiple symptoms were experienced by 44 (8.4%) participants. Risk factors for persistent symptoms were: age “6-11 years” (odds ratio 2.74 (95% confidence interval 1.37 to 5.75) and “12-18 years” (2.68, 1.41 to 5.4), and a history of allergic diseases (1.67, 1.04 to 2.67). Interpretation A quarter of children experienced persistent symptoms months after hospitalization with acute covid-19 infection, with almost one in ten experiencing multi-system involvement. Older age and allergic diseases were associated with higher risk of persistent symptoms at follow-up. Our findings highlight the need for replication and further investigation of potential mechanisms as well as clinical support to improve long term outcomes in children. Funding None. Evidence before this study Evidence suggests that Covid-19 may result in short- and long-term consequences to health. Studies in children and adolescents are limited and available evidence is scarce. We searched Embase for publications from inception to April, 25, 2021, using the following phrases or combinations of phrases “post-covid condition” or “post-covid syndrome” or “covid sequalae” or “post-acute covid” or “long covid” or “long hauler” with “pediatric*” or “paediatric*” or “child*” or “infant*” or “newborn*” or “toddler*” or “neonate*” or “neonatal” or “adolescent*” or “teen*”. We found small case series and small cohort studies looking at Covid-19 consequences in children. No large cohort studies of previously hospitalised children, assessing symptom duration, categorisation or attempting multivariable analyses to identify independent risk factors for long Covid development were identified. Added value of this study To our knowledge, this is the largest cohort study with the longest follow-up since hospital discharge of previously hospitalised children. We found that even months after discharge from the hospital, approximately a quarter of children experience persistent symptoms with one in ten having multi-system involvement. Older age and allergic diseases are associated with Covid-19 consequences. Parents of some children report emotional and behavioural changes in their children after Covid-19. Implications of all the available evidence Our findings highlight the need for continued global research of Covid-19 consequences in the paediatric population. Older children admitted to the hospital should be carefully monitored upon discharge. Large, controlled studies aiming to identify risk groups and potential intervention strategies are required to fill knowledge gaps.

Background The long-term sequalae of COVID-19 remain poorly characterised. In this study, we aimed to assess long-standing symptoms (LS) (symptoms lasting from the time of discharge) in previously hospitalised patients with COVID-19 and assess associated risk factors. Methods This is a longitudinal cohort study of adults (≥18 years of age) with clinically diagnosed or laboratory-confirmed COVID-19 admitted to Sechenov University Hospital Network in Moscow, Russia. Data were collected from patients discharged between April 8 and July 10, 2020. Participants were interviewed via telephone using Tier 1 ISARIC Long-term Follow-up Study CRF and the WHO CRF for Post COVID conditions. Reported symptoms were further categorised based on the system(s) involved. Additional information on dyspnoea, quality of life and fatigue was collected using validated instruments. Multivariable logistic regressions were performed to investigate risk factors for development of LS categories. Findings Overall, 2,649 of 4,755 patients discharged from the hospitals were available for the follow-up and included in the study. The median age of the patients was 56 years (IQR, 46–66) and 1,353 (51.1%) were women. The median follow-up time since hospital discharge was 217.5 (200.4-235.5) days. At the time of the follow-up interview 1247 (47.1%) participants reported LS. Fatigue (21.2%, 551/2599), shortness of breath (14.5%, 378/2614) and forgetfulness (9.1%, 237/2597) were the most common LS reported. Chronic fatigue (25%, 658/2593) and respiratory (17.2% 451/2616) were the most common LS categories. with reporting of multi-system involvement (MSI) less common (11.3%; 299). Female sex was associated with LS categories of chronic fatigue with an odds ratio of 1.67 (95% confidence interval 1.39 to 2.02), neurological (2.03, 1.60 to 2.58), mood and behaviour (1.83, 1.41 to 2.40), dermatological (3.26, 2.36 to 4.57), gastrointestinal (2.50, 1.64 to 3.89), sensory (1.73, 2.06 to 2.89) and respiratory (1.31, 1.06 to 1.62). Pre-existing asthma was associated with neurological (1.95, 1.25 to 2.98) and mood and behavioural changes (2.02, 1.24 to 3.18) and chronic pulmonary disease was associated with chronic fatigue (1.68, 1.21 to 2.32). Interpretation 6 to 8 months after acute infection episode almost a half of patients experience symptoms lasting since hospital discharge. One in ten individuals experiences MSI. Female sex is the main risk factor for majority of the LS categories. chronic pulmonary disease is associated with a higher risk of chronic fatigue development, and asthma with neurological and mood and behaviour changes. Individuals with LS and MSI should be the main target for future research and intervention strategies. Funding This study is supported by Russian Fund for Basic Research and UK Embassy in Moscow. The ISARIC work is supported by grants from: the NIHR Health Protection Research Unit (HPRU) in Emerging and Zoonotic Infections at University of Liverpool in partnership with Public Health England (PHE), in collaboration with Liverpool School of Tropical Medicine and the University of Oxford [award 200907], Wellcome Trust and Department for International Development [215091/Z/18/Z], and the Bill and Melinda Gates Foundation [OPP1209135], EU Platform for European Preparedness Against (Re-) emerging Epidemics (PREPARE) [FP7 project 602525] This research was funded in part, by the Wellcome Trust. The views expressed are those of the authors and not necessarily those of the DID, NIHR, Wellcome Trust or PHE. Evidence before this study Evidence suggests that COVID-19 may result in short- and long-term consequences to health. Most studies do not provide definitive answers due to a combination of short follow-up (2-3 months), small sample size, and use of non-standardised tools. There is a need to study the longer-term health consequences of previously hospitalised patients with COVID-19 infection and to identify risk factors for sequalae. Added value of this study To our knowledge, this is the largest cohort study (n=2,649) with the longest follow-up since hospital discharge (6-8 months) of previously hospitalised adult patients. We found that 6-8 months after discharge from the hospital, around a half (47.1%) of patients reported at least one long-standing symptom since discharge. Once categories of symptoms were assessed, chronic fatigue and respiratory problems were the most frequent clusters of long-standing symptoms in our patients. Of those patients having long-term symptoms, a smaller proportion (11.3%) had multisystem involvement, with three or more categories of long-standing symptoms present. Although most patients developed symptoms since discharge, a smaller number of individuals experienced symptom beginning symptom appearing weeks or months after the acute phase. Female sex was a predictor for most of the symptom categories at the time of the follow-up interview, with chronic pulmonary disease associated with chronic fatigue-related symptoms, and asthma with a higher risk of neurological symptoms, mood and behaviour problems. Implications of all the available evidence The majority of patients experienced long-lasting symptoms 6 to 8 months after hospital discharge and almost half reported at least one long-standing symptom, with chronic fatigue and respiratory problems being the most frequent. A smaller number reported multisystem impacts with three or more long-standing categories present at follow-up. A higher risk was found for women, for chronic pulmonary disease with chronic fatigue, and neurological symptoms and mood and behaviour problems with asthma. Patterns of the symptom development following COVID-19 should be further investigated in future research.