Symptom-based phenotypes and health-related quality of life in Spanish patients with long COVID: a latent class analysis
5 affiliations
- 1Primary Care Research Unit of Salamanca (APISAL), Salamanca Primary Care Management (SACyL), Institute of Biomedical Research of Salamanca (IBSAL)
- 2Castilla and León Health Service–SACYL, Gerencia Regional de Salud
- 3Research Network on Chronicity, Primary Care and Health Promotion (RICAPPS)
- 4Department of Medicine, University of Salamanca
- 5Department of Biomedical and Diagnostic Sciences, University of Salamanca
Abstract
Long COVID presents with heterogeneous, fluctuating symptoms that markedly reduce health-related quality of life (HRQol). This study aimed to identify symptom-based subgroups of patients with Long COVID using Latent Class Analysis (LCA) and to compare health-related quality of life across the identified classes. A cross-sectional study was conducted in Salamanca, Spain. Adults with Long COVID per WHO criteria were recruited from primary care and a specialized post-COVID unit. Sociodemographic, clinical, and symptom data ( ad hoc checklist) were collected; HRQoL was assessed using the SF-36. Latent Class Analysis identified symptom-based groups; model selection was based on BIC, entropy, and interpretability. Classes were compared on sociodemographic, clinical, and HRQoL variables. A total of 304 participants were included in the analysis (68.1% women; mean age 52.7 ± 11.9 years). Four classes were identified: Class 1 (28.9%) high multisystem symptom burden; Class 2 (20.1%) low symptom burden; Class 3 (25.3%) moderate neuropsychiatric and sleep-related symptoms; and Class 4 (25.7%) moderate physical-respiratory symptoms. HRQoL differed across classes in all SF-36 domains ( p < 0.001). Class 1 had the lowest scores [physical functioning 34.97 (95% CI: 32.56–37.38); vitality 28.98 (27.21–30.75); social functioning 26.31 (23.27–29.34); mental health 38.14 (35.61–40.67)], whilst Class 2 had the highest scores [52.36 (50.84–53.88); 46.85 (44.47–49.23); 47.07 (44.41–49.72); and 50.04 (48.04–52.04), respectively]. PCS and MCS scores were also lower in Class 1 [30.73 (28.50–32.96) and 35.07 (31.93–38.22), respectively] and higher in Class 2 [48.49 (46.32–50.65) and 48.49 (45.76–51.23), respectively]. Classes 3 and 4 showed intermediate patterns of quality of life, with lower scores in Class 3 compared with Class 4 in terms of social functioning, emotional role and mental health. LCA identified four distinct latent classes based on symptoms, which differ in symptom burden and predominant symptom patterns, and are associated with differences in health-related quality of life. These findings provide a better understanding of the clinical heterogeneity of long COVID. However, the classes are exploratory and sample-specific. The results need to be replicated in independent cohorts and in longitudinal studies.
Clinical trial registration: Clinicaltrials.gov , Identifier NCT05819840.
Related
- Greater and Clinically Meaningful Improvement in Health-Related Quality of Life After Device-Aided Therapy in Younger Patients Earlier Within Advanced Parkinson’s Disease: Real-World Evidence from the DATs-PD GETM Spanish Registry
- Cross-Cutting Symptom Burden and Health-Related Quality of Life in Hospitalized Patients With Chronic Illness: A Cross-Sectional Study
- Symptom burden, Psychosocial Distress, Resilience and Health-Related Quality of Life in Women with Metastatic Breast Cancer Receiving Contemporary Systemic Therapy
- The Relationship Between Symptom Experiences and Quality of Life in Patients on Hemodialysis: The Mediating Role of Positive Psychological Capital.
- Symptom network structure and its association with health-related quality of life in convalescent stroke survivors: a cross-sectional study
BibTeX
@article{dominguezmartin2026symptombased,
title = {{Symptom-based phenotypes and health-related quality of life in Spanish patients with long COVID: a latent class analysis}},
author = {Andrea Domínguez-Martín and Nuria Suárez-Moreno and Alicia Navarro-Cáceres and Silvia Arroyo-Romero and Olaya Tamayo-Morales and Susana González-Sánchez and Ana B. Castro-Rivero and Elena Navarro-Matías and Clara Domínguez-Martín and Andrea Sánchez-Moreno and Emiliano Rodríguez-Sánchez and Luis García-Ortiz and Cristina Lugones-Sánchez and Manuel A. Gómez-Marcos},
journal = {Frontiers in Public Health},
year = {2026},
month = nov,
doi = {10.3389/fpubh.2026.1972184},
url = {https://www.frontiersin.org/articles/10.3389/fpubh.2026.1972184/full},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Symptom-based phenotypes and health-related quality of life in Spanish patients with long COVID: a latent class analysis
AU - Andrea Domínguez-Martín
AU - Nuria Suárez-Moreno
AU - Alicia Navarro-Cáceres
AU - Silvia Arroyo-Romero
AU - Olaya Tamayo-Morales
AU - Susana González-Sánchez
AU - Ana B. Castro-Rivero
AU - Elena Navarro-Matías
AU - Clara Domínguez-Martín
AU - Andrea Sánchez-Moreno
AU - Emiliano Rodríguez-Sánchez
AU - Luis García-Ortiz
AU - Cristina Lugones-Sánchez
AU - Manuel A. Gómez-Marcos
PY - 2026
DA - 2026/11/05
JO - Frontiers in Public Health
DO - 10.3389/fpubh.2026.1972184
UR - https://www.frontiersin.org/articles/10.3389/fpubh.2026.1972184/full
AB - Long COVID presents with heterogeneous, fluctuating symptoms that markedly reduce health-related quality of life (HRQol). This study aimed to identify symptom-based subgroups of patients with Long COVID using Latent Class Analysis (LCA) and to compare health-related quality of life across the identified classes. A cross-sectional study was conducted in Salamanca, Spain. Adults with Long COVID per WHO criteria were recruited from primary care and a specialized post-COVID unit. Sociodemographic, clinical, and symptom data ( ad hoc checklist) were collected; HRQoL was assessed using the SF-36. Latent Class Analysis identified symptom-based groups; model selection was based on BIC, entropy, and interpretability. Classes were compared on sociodemographic, clinical, and HRQoL variables. A total of 304 participants were included in the analysis (68.1% women; mean age 52.7 ± 11.9 years). Four classes were identified: Class 1 (28.9%) high multisystem symptom burden; Class 2 (20.1%) low symptom burden; Class 3 (25.3%) moderate neuropsychiatric and sleep-related symptoms; and Class 4 (25.7%) moderate physical-respiratory symptoms. HRQoL differed across classes in all SF-36 domains ( p < 0.001). Class 1 had the lowest scores [physical functioning 34.97 (95% CI: 32.56–37.38); vitality 28.98 (27.21–30.75); social functioning 26.31 (23.27–29.34); mental health 38.14 (35.61–40.67)], whilst Class 2 had the highest scores [52.36 (50.84–53.88); 46.85 (44.47–49.23); 47.07 (44.41–49.72); and 50.04 (48.04–52.04), respectively]. PCS and MCS scores were also lower in Class 1 [30.73 (28.50–32.96) and 35.07 (31.93–38.22), respectively] and higher in Class 2 [48.49 (46.32–50.65) and 48.49 (45.76–51.23), respectively]. Classes 3 and 4 showed intermediate patterns of quality of life, with lower scores in Class 3 compared with Class 4 in terms of social functioning, emotional role and mental health. LCA identified four distinct latent classes based on symptoms, which differ in symptom burden and predominant symptom patterns, and are associated with differences in health-related quality of life. These findings provide a better understanding of the clinical heterogeneity of long COVID. However, the classes are exploratory and sample-specific. The results need to be replicated in independent cohorts and in longitudinal studies. Clinical trial registration: Clinicaltrials.gov , Identifier NCT05819840.
SN - 2296-2565
ER -