Original research••

How do clinicians use post-COVID syndrome diagnosis? Analysis of clinical features in a Swedish COVID-19 cohort with 18 months’ follow-up: a national observational cohort and matched cohort study

•,,,,,,,,,,.

...

Abstract

Introduction SARS-CoV-2 infection causes acute COVID-19 and may result in post-COVID syndrome (PCS). We aimed to investigate how clinicians diagnose PCS and identify associated clinical and demographic characteristics.

Methods We analysed multiregistry data of all SARS-CoV-2 test-positive individuals in Sweden (n=1 057 174) between 1 February 2020 and 25 May 2021. We described clinical characteristics that prompt PCS diagnosis in outpatient and inpatient settings. In total, there were 6389 individuals with a hospital inpatient or outpatient diagnosis for PCS. To understand symptomatology, we examined individuals diagnosed with PCS at least 3 months after COVID-19 onset (n=6389) and assessed factors associated with PCS diagnosis.

Results Mechanical ventilation correlated with PCS (OR 114.7, 95% CI 105.1 to 125.3) compared with no outpatient/inpatient contact during initial COVID-19. Dyspnoea (13.4%), malaise/fatigue (8%) and abnormal pulmonary diagnostic imaging findings (4.3%) were the most common features linked to PCS. We compared clinical features of PCS with matched controls (COVID-19 negative, n=23 795) and COVID-19 severity-matched patients (COVID-19 positive, n=25 556). Hypertension associated with PCS cohort (26.61%) than in COVID-19-negative (OR 17.16, 95% CI 15.23 to 19.3) and COVID-19-positive (OR 9.25, 95% CI 8.41 to 10.16) controls, although most individuals received this diagnosis before COVID-19. Dyspnoea was the second most common feature in the PCS cohort (17.2%), and new to the majority compared with COVID-19-negative (OR 54.16, 95% CI 42.86 to 68.45) and COVID-19-positive (OR 18.7, 95% CI 16.21 to 21.57) controls.

Conclusions Our findings highlight factors Swedish physicians associate with PCS.

What is already known on this topic

Post-COVID syndrome (PCS) is a recognised consequence of SARS-CoV-2 infection, but the specific clinical and demographic characteristics associated with its diagnosis are not well understood.

What this study adds

This study, based on the total population of laboratory-verified SARS-CoV-2-infected individuals with data from multiple nationwide registries in Sweden, reveals that mechanical ventilation during initial COVID-19 treatment is strongly associated with PCS. The most common clinical features linked to PCS are dyspnoea, malaise/fatigue and abnormal pulmonary diagnostic imaging findings. Hypertension and dyspnoea were found to be more common in the PCS cohort than in both COVID-19-negative and COVID-19-positive controls, though the majority had been diagnosed with hypertension previously. Dyspnoea was new to the majority of PCS individuals.

How this study might affect research, practice or policy

These findings provide valuable insights for clinicians diagnosing PCS and could inform future research into its causes and treatment. They may also influence policy decisions related to the care and management of patients recovering from COVID-19.

Introduction

Infection with SARS-CoV-2 resulting in COVID-19 exhibits a wide range of symptoms including respiratory and pulmonary symptoms during the acute phase of disease.1–3 The key characteristics include remarkably variable individual health responses from those who are asymptomatic or have mild symptoms to those with life-threatening severe disease or even death.1 2 An increasing number of patients suffer from disabling long-term consequences, post-COVID syndrome (PCS, also known as Long COVID, post-acute sequelae of COVID-19 or post-acute COVID-19 syndrome). Furthermore, the symptomatology of PCS varies between patients creating a challenge both for disease diagnosis, management and public health overview of the extent of the problem.

The global consensus on how PCS should be diagnosed and treated is still developing based on the efforts from the WHO, patient organisations and the Centre of Disease Control in providing guidelines and International Classification of Diseases (ICD) codes for diagnosis. PCS is thought to affect multiple organs,4–6 and potentially initial severity of COVID-19 could be a risk factor for PCS.6–8 Our understanding of the magnitude of how much vaccines protect against PCS is evolving.9 10 Breakthrough infections following vaccination and reinfection with SARS-CoV-2 also associate with long-term organ sequelae.9 11

The symptoms attributed to PCS are still being characterised, and there could be several different clinical phenotypes; however, dyspnoea and fatigue seem to be central symptomatic components of PCS.5 6 12–15 Post-intensive care syndrome (PICS) refers to a set of physical, cognitive and psychological impairments that survivors of intensive care often experience, with issues like muscle weakness, memory loss and depression.16 This could be similar to issues found in PCS.

To understand how clinicians attribute symptoms and PCS diagnosis, we examined risk factors and associated clinical features in patients who received a PCS ICD code in a total population of COVID-19 cohort in Sweden. To avoid classifying symptomatology due to PICS, COVID-19 patients with PCS were compared with disease severity-matched COVID-19 patients with no PCS diagnosis.

Methods

Definition of PCS

In September 2020, the WHO introduced ICD 10th version (ICD-10) code (U08.9 and U09.9) for PCS to distinguish between acute and late effects of COVID-19 infection. In Sweden, Z86.1A was implemented from 1 June 2020 to identify individuals with ‘COVID-19 in the personal history’, but was discontinued from 1 January 2021.17 The definition of post-COVID-19 condition by the WHO was: ‘Post-COVID-19 condition occurs in individuals with a history of probable or confirmed SARS-CoV-2 infection, usually 3 months from the onset of COVID-19 with symptoms that last for at least 2 months and cannot be explained by an alternative diagnosis’.18 Based on this definition, we defined PCS as the presence of the ICD-10 diagnostic codes Z86.1A, U08.9 and U09.9 that occurred at least 3 months post-COVID-19 in visits to outpatient and inpatient clinics (online supplemental file 1).

Source of data and study population

COVID-19 is a notifiable disease in Sweden and all laboratory-verified test-positive individuals are registered in the centralised communicable disease surveillance network (SmiNet) administered by the Swedish Public Health Agency, thereby including the spectrum of COVID-19 severity from those who do not display symptoms to those with fatal COVID-19, and has been described in previous studies.19–22 The personal identity numbers (PINs) from individuals with a laboratory-verified test-positive SARS-CoV-2 infection were extracted from SmiNet, and cross-linked with the following nationwide registries: Longitudinal Integrated Database for Health Insurance and Labour Market Studies registry (socioeconomic variables); Patient, Cancer, Death and Intensive Care Registries that are administered by Statistics Sweden, the Swedish National Board of Health and Welfare and the Swedish Intensive Care registry, respectively. Only COVID-19 patients who survived until at least 3 months after COVID-19 were included. Primary care data are not registered centrally and are therefore not accessible to perform a national study.

To identify clinical features clinicians associate with PCS (using diagnoses as proxy) and account for diseases prevalent in the background population/COVID-19 cohort not interpreted to be related to PCS by Swedish physicians in an outpatient/inpatient setting, we compared ICD-10 codes registered at healthcare visits at least 3 months after COVID-19 regardless of cause of visit for individuals with PCS; with ICD-10 codes registered at healthcare visits 3 months after index date or COVID-19 date for four control individuals matched on age, sex and county of residence that were either COVID-19 negative (no report of laboratory-verified positive SARS-CoV-2 test in SmiNet) or COVID-19 positive. The COVID-19-positive control individuals were also matched according to disease severity to ensure PICS did not factor into our study as a potential explanation for clinical features of PCS. The index date for COVID-19-negative control individuals was the corresponding COVID-19 date for the matched case, and they were identified by Statistics Sweden, and the four COVID-19-positive control individuals were identified from the COVID-19 cohort using R V.4.1.0 (R Core Team, 2021).

The data were pseudonymised by converting all PINs to a study ID.

Factors associating with receiving a PCS diagnosis by Swedish physicians

Potential factors under study included sex, age, disease severity, education, income and comorbidities. Age was grouped into 0–19, 20–39, 40–59, 60–79 and 80+ years. Disease severity was categorised into ‘no contact with outpatient or inpatient clinics’, ‘contact with outpatient clinic’, ‘hospitalisation’, ‘non-invasive ventilation and high-flow oxygen’, ‘intensive care’ and ‘mechanical ventilation’. Highest level of education was defined as primary, secondary or tertiary school and annual disposable income was categorised into quintiles. Comorbidities were summarised by the weighted Charlson Comorbidity Index (wCCI) and calculated using the algorithm designed specifically for Swedish registries,23 and based on ICD codes registered in the Patient and Cancer Registries from 1 January 1997 up to 1 month before the COVID-19/index date.

Clinical features physicians associate with PCS

The clinic and geographical region were characterised for the outpatient and inpatient clinics that recorded PCS diagnoses. Associated ICD-10 codes registered at PCS visits were used as a proxy to describe clinical features of PCS, and only ICD-10 codes where at least 1% of the cohort received these are included. The ICD-10 Z-chapter (factors influencing health status and contact with health services) was excluded from all analyses. Whether the diagnoses were new registrations or registered prior to COVID-19 was determined by searching historically for all visits to healthcare from 1 January 1997 to 1 month prior to COVID-19 (online supplemental file 1).

Statistical analysis

Potential factors associating with receiving a PCS diagnosis (sex, age, education, income, wCCI and CCI groups) were presented by descriptive statistics. The association between risk factors and PCS in the nationwide COVID-19 cohort was analysed using logistic regression adding variables in a stepwise process. Model 0 was a univariable model; model 1 included disease severity, age and sex; followed by addition of wCCI (model 2); addition of education (model 3); addition of income but not education (model 4) and finally addition of all variables (model 5). To identify which comorbidity groups were associated with PCS, a model including all variables (model 5) and individual CCI comorbidity groups were included.

Conditional logistic regression was used to identify diagnostic codes that associate with PCS compared with the matched COVID-19-negative and COVID-19-positive control individuals. The original p values were adjusted by false discovery rate and associations below 5% were considered significant. Outcomes from logistic regression models were presented with ORs and 95% CIs. Outcomes with p<0.05 were considered statistically significant. Missing data were included as a category of their own. Statistical analyses were carried out using R V.4.1.0 (R Core Team, 2021).

Patient and public involvement

The manuscript was circulated among a Swedish PCS patient group comprising of physicians and the Patient-Led Research Collaborative for comments and feedback, which were incorporated into the manuscript.

Results

A total of 16 151 (1.5%) received a PCS diagnosis (online supplemental file 1). To understand the symptomatology how medical doctors use the novel ICD-10 codes for PCS, we examined the population who had received a diagnosis (first or recurrent) at least 3 months after the COVID-19 date. There were 6389 (0.61%) individuals with a healthcare visit where a PCS diagnostic code was registered from 1 044 820 COVID-19 patients 3 months post-COVID-19 (figure 1 and table 1). A majority of patients who were diagnosed with PCS had been hospitalised, or had needed intensive care, or non-invasive and mechanical ventilation (61%), compared with 4.4% in the nationwide COVID-19 cohort (table 1). However, 33% of individuals with PCS diagnosis had not had contact with outpatient or inpatient clinics during the acute phase of initial COVID-19. Internal medicine (24.7%) and infection medicine (14.2%) outpatient and inpatient clinics registered the majority of PCS diagnostic codes in Sweden (online supplemental file 1). Geographically, 39.9% of all PCS visits were administered at Stockholm county council (online supplemental file 1), a county that accounts for 23% of the Swedish population (Statistics Sweden, 2021). The distribution of comorbidities according to the Charlson comorbidity grouping for the COVID-19 cohort, PCS individuals and matched control individuals is shown in online supplemental file 1.

Request permission

Figure 1

Flow chart of the COVID-19 cohort and control individuals, and distribution of post-COVID syndrome (PCS) codes used.

Table 1

Clinical and sociodemographic characteristics of the COVID-19 cohort and matched control individuals

Factors associating with receiving a PCS diagnosis by physicians in the COVID-19 cohort

We estimated risk factors for PCS diagnosis first in the full cohort of COVID-19-positive individuals prior to matching for disease severity or other demographic factors. The strongest risk factor for PCS diagnosis was needing mechanical ventilation during the acute phase of COVID-19 (adjusted OR (model 5 (adjusted for age, sex, education and comorbidities) 114.7, 95% (CI) 105.1 to 125.3) (figure 2 and online supplemental file 1). In model 5 including all potential risk factors, there was an approximately 25% and 61% higher risk of being diagnosed with PCS for the age groups 60–79 and 80+ years, respectively, compared with the reference group (20–39 years). Men had a lower risk of receiving a PCS diagnosis compared with women (0.78 (0.74 to 0.82)). Having only secondary and primary education compared with tertiary education associated with a lower risk of PCS diagnosis, whereas income did not significantly associate with PCS diagnosis (figure 2). Higher wCCI associated with PCS diagnosis, and specifically having had peripheral vascular disease, cerebrovascular disease, chronic obstructive pulmonary disorder, rheumatic disease, diabetes with complications, renal disease, ascites and malignancy associated significantly with an increased risk of receiving a PCS diagnosis (online supplemental file 1).

Request permission

Figure 2

Multivariable logistic regression of association between potential risk factors (COVID-19 severity, age, sex, comorbidities, education, income) during the acute phase and receiving a PCS diagnosis during visits to outpatient and inpatient clinics at least 3 months after COVID-19 onset. All COVID-19-positive individuals are included in this analysis. The multivariable logistic regression includes all variables in the figure (ie, model 5 from online supplemental file 1). AOR, adjusted OR, PCS, post-COVID syndrome; wCCI weighted Charlson Comorbidity Index.

Clinical features prompting diagnosis with PCS by physicians in an outpatient/inpatient setting

The most frequent symptoms and signs diagnostic codes (ICD-10 R-chapter) for PCS-associated healthcare visits were dyspnoea, malaise/fatigue, abnormal findings on pulmonary diagnostic imaging and chest pain (table 2 and online supplemental file 1). When including all diagnoses, the most frequent diagnosis was hypertension; however, the majority of the individuals had this diagnosis registered before COVID-19 (online supplemental file 1).

Table 2

Symptomatology associated with post-COVID syndrome (PCS) diagnosis

Diagnoses most likely associated with PCS symptomatology as seen by the clinicians, and with fewer than 10% having these diagnoses previously, were subjective mild cognitive disorder, post-viral fatigue syndrome, abnormal results of pulmonary function studies and diagnostic imaging, tachycardia, sleep disorder, malaise and fatigue, pulmonary embolism and skin paraesthesia (online supplemental file 1).

Matched cohort study comparing PCS patients with matched COVID-19-negative and COVID-19-positive control individuals

To identify clinical features of the PCS cohort, we extended the analysis to include all visits to hospital outpatient and inpatient clinics regardless of cause, instead of only visits due to PCS, at least 3 months after COVID-19 index date (online supplemental file 1). A matched cohort study was performed comparing these individuals without COVID-19 and those with COVID-19.

We then compared the overall healthcare usage between PCS and control cohorts. The most frequent diagnosis for PCS individuals compared with matched control individuals was hypertension (26.6% vs 3.1% in matched COVID-19-negative controls and 4.8% in matched COVID-19-positive controls) (table 3 and online supplemental file 1). However, 73.5% of the PCS individuals had received this diagnosis prior to COVID-19, indicating hypertension was not a new diagnosis following COVID-19. The second most frequent ICD-10 code was dyspnoea (17.2% in PCS vs 0.4% in matched COVID-19-negative controls and 1.1% in matched COVID-19-positive controls). Dyspnoea was a new diagnosis for 76% of the PCS individuals who received this diagnosis after COVID-19.

Table 3

Matched cohort study identifying diagnoses associated with the post-COVID syndrome (PCS) patients compared with matched COVID-19-negative and disease severity-matched COVID-19-positive control individuals

The diagnoses that were present at a significantly higher proportion in the PCS cohort compared with the COVID-19-negative/positive matched control individuals in the matched cohort study were listed according to percentage of PCS individuals who had not received this diagnosis prior to SARS-CoV-2 infection (online supplemental file 1).

Discussion

Currently, the pathophysiology and understanding of mechanisms of PCS are evolving including persistent infection, autoimmunity or dysfunction of blood clotting.19 20 24 The risk factors and clinical diagnostic features in COVID-19 patients with long-term sequelae that result in a PCS diagnosis have remained elusive, and most of the literature on PCS is based on follow-up studies of hospitalised patients25 26 and questionnaire data.7 10 12 27 28 A recent study was published using healthcare data covering 25% of the Israeli population29; however, no studies have been performed on a total country’s population including all individuals who have a laboratory-verified positive test for SARS-CoV-2 in the country regardless of initial COVID-19 severity. This approach allows us to attribute diagnosis of PCS and related symptomatology, and describe the characteristics of physicians who use the PCS diagnosis with regard to clinical specialty and geography.

In the total population of Sweden, we examined all laboratory-verified test-positive SARS-CoV-2 individuals and discovered that the initial COVID-19 severity was the strongest associating factor for physicians assigning a PCS diagnosis in an outpatient/inpatient setting. We found that Swedish physicians during the first and second waves of the pandemic primarily associated dyspnoea and fatigue among other symptoms with PCS. In addition, we identified several diagnoses that were present prior to COVID-19, and thereby not likely associated with clinical features of PCS.

It is noteworthy that the PCS code is novel and likely used differently across practices and geographical location. This is also supported by our results and in our study, we see the highest usage in the Stockholm metropolitan area. Furthermore, there likely is not yet a harmonised national standard across different locations on how to use ICD-10 codes for PCS. Our study provides insight into how these novel codes are used in Sweden and the symptomatology that prompts healthcare providers and medical doctors to assign the code. Yet, the novel ICD-10 codes likely do not capture diagnoses for all PCS patients.

The main strength of our study is the use of the entire population of laboratory-verified SARS-CoV-2 test-positive individuals in Sweden (approximately 1 million) during 1 February 2020–25 May 2021 with historical data from visits to outpatient/inpatient clinics from 1997. This approach allows population-based estimates and offers a clinical perspective, which is complementary to self-reported symptoms in questionnaire-based studies, with the added benefit of identifying the specific diagnoses that were new following COVID-19 onset. The Nordic countries are unique in having PINs for every individual in the population, with Sweden having a numerically larger population, providing the infrastructure that is needed for our study.30

A weakness is the lack of primary care data, which are not registered centrally in Sweden; and additionally, the access to outpatient/inpatient clinics and follow-up was not standardised. The PCS incidence is most likely conceivably larger than what is represented in the data, due to initial physician lack of awareness of PCS and associated novel ICD-10 codes for diagnosis. Therefore, we examined how clinicians attribute the code to patients who received the PCS diagnosis. This is made obvious in our study where the proportion of individuals diagnosed with PCS by a physician in an outpatient/inpatient setting during the early pandemic phases was 0.61%. This is considerably lower than reported in other studies ranging from 14.1%,7 12.5%,28 6.2%15 to 3.3% of the UK population according to the UK Office for National Statistics.31

The initial COVID-19 severity was a strong associating factor for physicians assigning a PCS diagnosis to patients in our study. This is corroborated in earlier reports.6–8 Another associating factor for physicians assigning a PCS diagnosis in our study was female sex, as found in other studies.15 25 26 31 32

Dyspnoea, malaise/fatigue/post-viral fatigue syndrome and abnormal pulmonary imaging/findings were some of the most prominent clinical features that physicians associated with PCS. These clinical features were also enhanced in the COVID-19 patients with PCS, even when compared with COVID-19 patients without a PCS diagnosis but with the same initial severity of disease. The majority of the PCS patients had not received these diagnoses prior to COVID-19 onset. Since we compare intensive care unit (ICU) patients with PCS with ICU patients without PCS, our findings cannot be attributed to PICS only.16 The PCS cohort has a characteristic profile compared with the remaining COVID-19 cohort. Although most of the literature on PCS is based on follow-up studies of hospitalised patients,25 26 our findings are in line with the diverse symptomatology previously reported by questionnaire data,7 12 27 28 and recently also with electronic health record data from Veteran’s affairs data.6 9 Furthermore, systematic reviews and meta-analyses identify fatigue and dyspnoea as common long-term symptoms.12–14

In our study, pulmonary embolism was present in 5% of the PCS cohort, which was a new diagnosis to the majority; and the OR was 15-fold higher when compared with disease severity-matched COVID-19-positive control individuals. In a previous study, we found that the risk of pulmonary embolism was highest for COVID-19 patients in need of intensive care, and still significantly increased in the time period of 3–6 months after COVID-19 indicating a potential association between pulmonary embolism and dyspnoea for this particular patient group.19 This indicates that the PCS cohort has a distinct symptomatic/disease profile as seen by the clinicians from the remaining COVID-19 cohort.

While the mechanisms of PCS are emerging, there are several potential likely disease mechanisms.24 For example, COVID-19 has been demonstrated to damage vasculature.33 Consequently, defects in blood supply and tissue oxygenation may contribute to more severe clinical picture particularly in patients where pulmonary damage through severe disease has occurred, and in patients who have dysfunction of peripheral vasculature. Furthermore, independent of disease severity, the dysfunction of the autonomic nervous system may be caused by autoimmunity, epitope spreading or direct persistent infection.34 35 Interestingly, in our study, 2% of individuals had SARS-CoV-2 detected in the same visit where they received a PCS diagnosis by a physician, raising the possibility of reinfection or persistent infection. Similarly, the clinical picture in PCS may relate to novel infections or opportunistic secondary infections such as those caused by Echerichia coli or other bacterial infections as we observe are enhanced in the PCS cohort in our study, or reactivation of latent pathogens including Epstein-Barr or varicella virus.24 34 35

In summary, we found that Swedish physicians associated the following with PCS: respiratory dysfunction, neurological and cognitive disorders, and cardiac and circulatory dysfunction. Initial COVID-19 severity was the strongest associating factor for receiving a PCS diagnosis by a physician. More studies are required to further characterise this emerging syndrome, and particularly in a primary care setting.

Correction notice: This article has been corrected since it was first published. The open access licence has been updated to CC BY. 30th July 2024.
Contributors: HMO—study design, data interpretation, funding acquisition, literature review, visualisation, writing (original draft), writing (review and editing). OF-R—study design, data curation, formal analysis, software, visualisation, data interpretation, writing (original draft), writing (review and editing). SK—data curation, formal analysis, software, visualisation, data interpretation, writing (review and editing). IHC—study design, data interpretation, writing (original draft), writing (review and editing). JN, LT, PMM and NHR—study design, data interpretation, writing (review and editing). AG—study design, data interpretation, funding acquisition, writing (review and editing). ME—study design, data curation, formal analysis, software, visualisation, data interpretation, writing (review and editing). A-MFC—conceptualisation, data collection, study design, investigation, funding acquisition, methodology, supervision, project administration, literature review, data interpretation, writing (original draft), writing (review and editing). All authors had final responsibility for the decision to submit for publication. A-MFC had final responsibility for submission and is the guarantor.
Funding: This study was funded by the Swedish Research Council (2021-06536, A-MFC); Kempe Foundation (SMK-21-0014, A-MFC); Petrus and Augusta Hedlund’s Foundation (M-2022-1753, A-MFC); Västerbotten County Council (RV-967545; RV-967783, A-MFC); Horizon Europe (NEUROCOV–101057775, A-MFC and AG); Academy of Finland (#340539, HMO) and Research Council of Norway (262700, IHC and PMM). This project has received funding from the RIA HORIZON Research and Innovation Actions under GA No. NEUROCOV-101057775 (A-MFC and AG).
EC Disclaimer: Funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or European Health and Digital Agency (HADEA). Neither the European Union nor the granting authority can be held responsible for them.
The funding agencies had no role in study design, data collection, analysis of and interpretation of data, writing of the report or decision to submit the article for publication. The views and conclusions in this document are those of the authors and not influenced by the funding agencies.
Competing interests: None declared.
Patient and public involvement: Patients and/or the public were involved in the design, or conduct, or reporting, or dissemination plans of this research. Refer to the Methods section for further details.
Provenance and peer review: Not commissioned; externally peer reviewed.
Supplemental material: This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.

Data availability statement

Data may be obtained from a third party and are not publicly available. The study protocol (R script) for the SCCS and the matched cohort study are available upon request from the corresponding author. The study used secondary registry data, which are regulated by the Public Access to Information and Secrecy Act (2009:400) and are protected by strict confidentiality. However, for the purpose of research, after formal application to access personal data, the responsible authority can grant access to data, although this is contingent on vetting by the Ethical Review Authority of Sweden, according to the Act (2003:460) concerning the Ethical Review of Research Involving Humans. Synthetic (ie, depersonalised and jittered) data can be provided on request from the corresponding author.

Ethics statements

Patient consent for publication:

Ethics approval:

The study received ethical approval from the Swedish Ethical Review Authority (DNR 2020-02150).

Acknowledgements

We would like to acknowledge Swedish physicians with post-COVID syndrome, in particular Dr Alicia Edin; and Patient-Led Research Collaborative for constructive and enlightening feedback on the manuscript.

Derry HM, Lief L, Schenck EJ, et al. Peritraumatic stress among Caregivers of patients in the intensive care unit. Ann Am Thorac Soc 2020; 17:650–4.
doi:10.1513/AnnalsATS.201908-647RL•Google Scholar
Gandhi RT, Lynch JB, Del Rio C, et al. Mild or moderate COVID-19. N Engl J Med 2020; 383:1757–66.
doi:10.1056/NEJMcp2009249•Google Scholar
Desai AD, Lavelle M, Boursiquot BC, et al. Long-term complications of COVID-19. Am J Physiol Cell Physiol 2022; 322:C1–11.
doi:10.1152/ajpcell.00375.2021•Google Scholar
Crook H, Raza S, Nowell J, et al. Long Covid-mechanisms, risk factors, and management. BMJ 2021; 374:1648.
doi:10.1136/bmj.n1648•Google Scholar
Greenhalgh T, Sivan M, Delaney B, et al. Long Covid-an update for primary care. BMJ 2022;
doi:10.1136/bmj-2022-072117•Google Scholar
Al-Aly Z, Xie Y, Bowe B, et al. High-dimensional characterization of post-acute sequelae of COVID-19. Nature 2021; 594:259–64.
doi:10.1038/s41586-021-03553-9•Google Scholar
Sudre CH, Murray B, Varsavsky T, et al. Attributes and predictors of long COVID. Nat Med 2021; 27:626–31.
doi:10.1038/s41591-021-01292-y•Google Scholar
Martin-Loeches I, Motos A, Menéndez R, et al. ICU-acquired pneumonia is associated with poor health post-COVID-19 syndrome. J Clin Med 2021; 11.
Available: here
doi:10.3390/jcm11010224•Google Scholar
Al-Aly Z, Bowe B, Xie Y, et al. Long COVID after breakthrough SARS-Cov-2 infection. Nat Med 2022; 28:1461–7.
doi:10.1038/s41591-022-01840-0•Google Scholar
Brunvoll SH, Nygaard AB, Fagerland MW, et al. Post-acute symptoms 3-15 months after COVID-19 among Unvaccinated and vaccinated individuals with a breakthrough infection. Int J Infect Dis 2023; 126:10–3.
doi:10.1016/j.ijid.2022.11.009•Google Scholar
Bowe B, Xie Y, Al-Aly Z, et al. Acute and Postacute sequelae associated with SARS-Cov-2 Reinfection. Nat Med 2022; 28:2398–405.
doi:10.1038/s41591-022-02051-3•Google Scholar
Lopez-Leon S, Wegman-Ostrosky T, Perelman C, et al. More than 50 long-term effects of COVID-19: a systematic review and meta-analysis. Sci Rep 2021; 11.
doi:10.1038/s41598-021-95565-8•Google Scholar
Ceban F, Ling S, Lui LMW, et al. Fatigue and cognitive impairment in post-COVID-19 syndrome: A systematic review and meta-analysis. Brain Behav Immun 2022; 101:93–135.
doi:10.1016/j.bbi.2021.12.020•Google Scholar
Jennings G, Monaghan A, Xue F, et al. A systematic review of persistent symptoms and residual abnormal functioning following acute COVID-19: ongoing symptomatic phase vs. post-COVID-19 syndrome. J Clin Med 2021; 10.
doi:10.3390/jcm10245913•Google Scholar
Global Burden of Disease Long COVID Collaborators, Wulf Hanson S, Abbafati C, et al. Estimated global proportions of individuals with persistent fatigue, cognitive, and respiratory symptom clusters following symptomatic COVID-19 in 2020 and 2021. JAMA 2022; 328:1604–15.
doi:10.1001/jama.2022.18931•Google Scholar
Hiser SL, Fatima A, Ali M, et al. Post-intensive care syndrome (PICS): recent updates. J Intensive Care 2023; 11.
doi:10.1186/s40560-023-00670-7•Google Scholar
Welfare NBoHa. Statistik OM Tillstånd Efter COVID-19 I Primärvård och Specialiserad Vård. contract no.: art.NR: 2021-6-7495. 2021;
Google Scholar
Soriano JB, Murthy S, Marshall JC, et al. A clinical case definition of post-COVID-19 condition by a Delphi consensus. Lancet Infect Dis 2022; 22:e102–7.
doi:10.1016/S1473-3099(21)00703-9•Google Scholar
Katsoularis I, Fonseca-Rodríguez O, Farrington P, et al. Risks of deep vein thrombosis, pulmonary embolism, and bleeding after COVID-19: nationwide self-controlled cases series and matched cohort study. BMJ 2022; 377.
doi:10.1136/bmj-2021-069590•Google Scholar
Katsoularis I, Fonseca-Rodríguez O, Farrington P, et al. Risk of acute myocardial infarction and ischaemic stroke following COVID-19 in Sweden: a self-controlled case series and matched cohort study. Lancet 2021; 398:599–607.
doi:10.1016/S0140-6736(21)00896-5•Google Scholar
Gustafsson PE, San Sebastian M, Fonseca-Rodriguez O, et al. Inequitable impact of infection: social gradients in severe COVID-19 outcomes among all confirmed SARS-Cov-2 cases during the first pandemic wave in Sweden. J Epidemiol Community Health 2022; 76:261–7.
doi:10.1136/jech-2021-216778•Google Scholar
Fonseca-Rodríguez O, Gustafsson PE, San Sebastián M, et al. Spatial clustering and Contextual factors associated with Hospitalisation and deaths due to COVID-19 in Sweden: a Geospatial nationwide ecological study. BMJ Glob Health 2021; 6.
doi:10.1136/bmjgh-2021-006247•Google Scholar
Ludvigsson JF, Appelros P, Askling J, et al. Adaptation of the Charlson Comorbidity index for register-based research in Sweden. Clin Epidemiol 2021; 13:21–41.
doi:10.2147/CLEP.S282475•Google Scholar
Davis HE, McCorkell L, Vogel JM, et al. Long COVID: major findings, mechanisms and recommendations. Nat Rev Microbiol 2023; 21:133–46.
doi:10.1038/s41579-022-00846-2•Google Scholar
Fernández-de-Las-Peñas C, Pellicer-Valero OJ, Navarro-Pardo E, et al. Symptoms experienced at the acute phase of SARS-Cov-2 infection as risk factor of long-term post-COVID symptoms: the LONG-COVID-EXP-CM multicenter study. Int J Infect Dis 2022; 116:241–4.
doi:10.1016/j.ijid.2022.01.007•Google Scholar
Sigfrid L, Drake TM, Pauley E, et al. Long Covid in adults discharged from UK hospitals after COVID-19: A prospective, Multicentre cohort study using the ISARIC WHO clinical Characterisation protocol. Lancet Reg Health Eur 2021; 8.
doi:10.1016/j.lanepe.2021.100186•Google Scholar
Caspersen IH, Magnus P, Trogstad L, et al. Excess risk and clusters of symptoms after COVID-19 in a large Norwegian cohort. Eur J Epidemiol 2022; 37:539–48.
doi:10.1007/s10654-022-00847-8•Google Scholar
Ballering AV, van Zon SKR, olde Hartman TC, et al. Persistence of somatic symptoms after COVID-19 in the Netherlands: an observational cohort study. The Lancet 2022; 400:452–61.
doi:10.1016/S0140-6736(22)01214-4•Google Scholar
Mizrahi B, Sudry T, Flaks-Manov N, et al. Long Covid outcomes at one year after mild SARS-Cov-2 infection: nationwide cohort study. BMJ 2023; 380.
doi:10.1136/bmj-2022-072529•Google Scholar
Laugesen K, Ludvigsson JF, Schmidt M, et al. Nordic health Registry-based research: A review of health care systems and key registries. Clin Epidemiol 2021; 13:533–54.
doi:10.2147/CLEP.S314959•Google Scholar
Statistics UOfN. Prevalence of long COVID symptoms and COVID-19 complications office for National Statistics Uk2022. 2022;
Available: here
Google Scholar
Huang C, Huang L, Wang Y, et al. 6-month consequences of COVID-19 in patients discharged from hospital: a cohort study. Lancet 2021; 397:220–32.
Available: here
doi:10.1016/S0140-6736(20)32656-8•Google Scholar
Dupont A, Rauch A, Staessens S, et al. Vascular endothelial damage in the pathogenesis of organ injury in severe COVID-19. Arterioscler Thromb Vasc Biol 2021; 41:1760–73.
Available: here
doi:10.1161/ATVBAHA.120.315595•Google Scholar
Klein J, Wood J, Jaycox J, et al. Distinguishing features of long COVID identified through immune profiling. medRxiv 2022;
doi:10.1101/2022.08.09.22278592•Google Scholar
Su Y, Yuan D, Chen DG, et al. Multiple early factors anticipate post-acute COVID-19 sequelae. Cell 2022; 185:881–95.
doi:10.1016/j.cell.2022.01.014•Google Scholar

Overview

Abstract
Introduction
Methods
Results
Discussion
References

Supplementary files
Footnotes
Publication history
Rapid Responses

Article metrics

Altmetric

Dimensions