sharon sanders
Editor-in-Chief & President
Research Letter
Infectious Diseases
October 25, 2024
Hospitalization for COVID-19, Other Respiratory Infections, and Postacute Patient-Reported Symptoms
Yaqing Gao, MPH[SUP]1[/SUP]; Yunhe Wang, MSc[SUP]1[/SUP]; Li Chen, MB[SUP]2[/SUP]; et alJunqing Xie, DPhil[SUP]2[/SUP]; Daniel Prieto-Alhambra, PhD[SUP]2[/SUP]
Author Affiliations Article Information
Introduction
Chronic multisystem manifestations of SARS-CoV-2 infection (ie, post–COVID-19 condition [PCC]) are a specific type of postacute infection syndrome (PAIS) that can occur after other lower respiratory tract infections (LRTIs) but are often overlooked.[SUP]1[/SUP] Research on PAIS primarily captures it through inpatient diagnoses, missing milder cases and those not seeking care.[SUP]2[/SUP] We performed pairwise comparisons of patient-reported, multisystem, postacute symptoms among individuals hospitalized with COVID-19, other LRTIs, and those without LRTI hospitalizations during the same study period.
Methods
This cohort study used data from the UK Biobank (UKB)[SUP]3[/SUP] and received ethical approval from the North West Multi-center Research Ethics Committee. All participants provided written informed consent. Additional details on the UKB are shown in the eAppendix in Supplement 1. We followed STROBE reporting guidelines.
We categorized respondents into COVID-19 hospitalization, other LRTI hospitalization, and a reference group, those without any LRTI hospitalizations (eFigure in Supplement 1). We identified COVID-19 and non–COVID-19 LRTI hospitalizations that occurred between March 1, 2020, and the survey date using International Statistical Classification of Diseases and Related Health Problems, Tenth Revision codes; 399 respondents hospitalized within 12 weeks of their survey date were excluded to study postacute symptoms.[SUP]4[/SUP] The index date was the latest hospitalization date for COVID-19 or other LRTI groups and the survey date for the reference group.
We used overlap weighting to balance participants’ characteristics across the 3 groups,[SUP]5[/SUP] including age at index date, sex, socioeconomic and lifestyle factors measured at UKB recruitment, and pre–index date disease history (eTable in Supplement 1). Additional details on patient exclusion and our statistical methods are shown in are shown in the eAppendix in Supplement 1.
Results
In total, 191 710 eligible participants (mean [SD] age at index date, 69.2 [7.6] years; 109 523 female [57.1%]) were finally included: 1153 hospitalized with COVID-19, 1304 with other LRTIs, and 189 253 in the reference group. After the overlap weighting, all prespecified covariates were well balanced across the 3 groups, with standardized mean differences less than 0.1 (Table 1). Compared with the reference group, COVID-19 hospitalization was associated with higher risks of 23 of 45 symptoms (Table 2). These symptoms were observed across the ear, nose, and throat; respiratory; neurological; gastrointestinal; and musculoskeletal systems. The greatest risk was observed for ageusia (odds ratio [OR], 2.27; 95% CI, 1.87-2.75) and severe fatigue (OR, 2.18; 95% CI, 1.70-2.81). Other LRTI hospitalizations (vs reference group) were also associated with higher risks of 18 of 45 symptoms. The most prominent symptoms were related to the respiratory system (eg, pain on breathing, OR, 2.51; 95% CI, 1.99-3.17), with a modestly increased risk of gastrointestinal and neurological symptoms.
Compared with other LRTIs, COVID-19 hospitalizations were associated with an increased risk of 7 individual symptoms: anosmia (OR, 1.80; 95% CI, 1.34-2.42), ageusia (OR, 1.52; 95% CI, 1.13-2.03), postural tachycardia (OR, 1.60; 95% CI, 1.12-2.29), problem of thinking (OR, 1.36; 95% CI, 1.14-1.62), bone pain (OR, 1.33; 95% CI, 1.06-1.67), mild fatigue (OR, 1.19; 95% CI, 1.01-1.40), and severe fatigue (OR, 1.49; 95% CI, 1.02-2.17). Conversely, persistent chest cough was less common for COVID-19.
Discussion
This cohort study found that PAIS is not unique to COVID-19; it can also occur in people with other severe LRTIs. However, compared with other LRTIs, COVID-19 appeared to impose an extra burden of neurological, cognitive, and fatigue symptoms. These findings highlight the similarities and differences between PCC and PAIS triggered by other pathogens, which will inform tailored clinical management and offer mechanistic insights into these previously overlooked syndromes. Limitations include potential residual confounding, the healthier profile of UKB participants vs the general population, and the lack of repeated-measure data for symptom trajectories.
https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2825348
Infectious Diseases
October 25, 2024
Hospitalization for COVID-19, Other Respiratory Infections, and Postacute Patient-Reported Symptoms
Yaqing Gao, MPH[SUP]1[/SUP]; Yunhe Wang, MSc[SUP]1[/SUP]; Li Chen, MB[SUP]2[/SUP]; et alJunqing Xie, DPhil[SUP]2[/SUP]; Daniel Prieto-Alhambra, PhD[SUP]2[/SUP]
Author Affiliations Article Information
- [SUP]1[/SUP]Nuffield Department of Population Health, University of Oxford, Oxford, United Kingdom
- [SUP]2[/SUP]Centre for Statistics in Medicine and NIHR Biomedical Research Centre Oxford, NDORMS, University of Oxford, Oxford, United Kingdom
Introduction
Chronic multisystem manifestations of SARS-CoV-2 infection (ie, post–COVID-19 condition [PCC]) are a specific type of postacute infection syndrome (PAIS) that can occur after other lower respiratory tract infections (LRTIs) but are often overlooked.[SUP]1[/SUP] Research on PAIS primarily captures it through inpatient diagnoses, missing milder cases and those not seeking care.[SUP]2[/SUP] We performed pairwise comparisons of patient-reported, multisystem, postacute symptoms among individuals hospitalized with COVID-19, other LRTIs, and those without LRTI hospitalizations during the same study period.
Methods
This cohort study used data from the UK Biobank (UKB)[SUP]3[/SUP] and received ethical approval from the North West Multi-center Research Ethics Committee. All participants provided written informed consent. Additional details on the UKB are shown in the eAppendix in Supplement 1. We followed STROBE reporting guidelines.
We categorized respondents into COVID-19 hospitalization, other LRTI hospitalization, and a reference group, those without any LRTI hospitalizations (eFigure in Supplement 1). We identified COVID-19 and non–COVID-19 LRTI hospitalizations that occurred between March 1, 2020, and the survey date using International Statistical Classification of Diseases and Related Health Problems, Tenth Revision codes; 399 respondents hospitalized within 12 weeks of their survey date were excluded to study postacute symptoms.[SUP]4[/SUP] The index date was the latest hospitalization date for COVID-19 or other LRTI groups and the survey date for the reference group.
We used overlap weighting to balance participants’ characteristics across the 3 groups,[SUP]5[/SUP] including age at index date, sex, socioeconomic and lifestyle factors measured at UKB recruitment, and pre–index date disease history (eTable in Supplement 1). Additional details on patient exclusion and our statistical methods are shown in are shown in the eAppendix in Supplement 1.
Results
In total, 191 710 eligible participants (mean [SD] age at index date, 69.2 [7.6] years; 109 523 female [57.1%]) were finally included: 1153 hospitalized with COVID-19, 1304 with other LRTIs, and 189 253 in the reference group. After the overlap weighting, all prespecified covariates were well balanced across the 3 groups, with standardized mean differences less than 0.1 (Table 1). Compared with the reference group, COVID-19 hospitalization was associated with higher risks of 23 of 45 symptoms (Table 2). These symptoms were observed across the ear, nose, and throat; respiratory; neurological; gastrointestinal; and musculoskeletal systems. The greatest risk was observed for ageusia (odds ratio [OR], 2.27; 95% CI, 1.87-2.75) and severe fatigue (OR, 2.18; 95% CI, 1.70-2.81). Other LRTI hospitalizations (vs reference group) were also associated with higher risks of 18 of 45 symptoms. The most prominent symptoms were related to the respiratory system (eg, pain on breathing, OR, 2.51; 95% CI, 1.99-3.17), with a modestly increased risk of gastrointestinal and neurological symptoms.
Compared with other LRTIs, COVID-19 hospitalizations were associated with an increased risk of 7 individual symptoms: anosmia (OR, 1.80; 95% CI, 1.34-2.42), ageusia (OR, 1.52; 95% CI, 1.13-2.03), postural tachycardia (OR, 1.60; 95% CI, 1.12-2.29), problem of thinking (OR, 1.36; 95% CI, 1.14-1.62), bone pain (OR, 1.33; 95% CI, 1.06-1.67), mild fatigue (OR, 1.19; 95% CI, 1.01-1.40), and severe fatigue (OR, 1.49; 95% CI, 1.02-2.17). Conversely, persistent chest cough was less common for COVID-19.
Discussion
This cohort study found that PAIS is not unique to COVID-19; it can also occur in people with other severe LRTIs. However, compared with other LRTIs, COVID-19 appeared to impose an extra burden of neurological, cognitive, and fatigue symptoms. These findings highlight the similarities and differences between PCC and PAIS triggered by other pathogens, which will inform tailored clinical management and offer mechanistic insights into these previously overlooked syndromes. Limitations include potential residual confounding, the healthier profile of UKB participants vs the general population, and the lack of repeated-measure data for symptom trajectories.
https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2825348