key: cord-0977944-xytmitq5
authors: Gruber, R.; Montilva Ludewig, M. V.; Wessels, C.; Schlang, G.; Jedhoff, S.; Herbrandt, S.; Mattner, F.
title: Long-term symptoms after SARS-CoV-2 infection in a cohort of hospital employees: duration and predictive factors
date: 2022-03-22
journal: nan
DOI: 10.1101/2022.03.22.22272770
sha: 59918e295e707002651bcadb453107de96b39502
doc_id: 977944
cord_uid: xytmitq5

Objective: To evaluate the frequency, duration and patterns of long-term coronavirus disease 2019 (COVID-19) symptoms and to analyze risk factors for long-lasting COVID-19 sequelae among hospital employees (HEs). Design: Retrospective observational cohort study. Setting: Three medical centers in Cologne, Germany. Participants: We included HEs who had a positive severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) polymerase chain reaction (PCR) test between March 2020 and May 2021. Methods: We conducted a survey in mid-2021 with all HEs tested SARS-CoV-2 PCR positive during the study period and asked about the presence and duration of 24 different COVID-19 related symptoms. Chronological development and patterns of symptom complexes, probability of symptom persistence and possible risk factors for protracted COVID-19 course were analyzed. Results: Of 221 included HEs, a number of 104 HEs (47.1%) reported at least one persisting symptom for more than 90 days after initial SARS-CoV-2 detection. A symptom duration over 28 days was associated with multiple symptom complexes. The most common was the interrelated occurrence of shortness of breath, memory disorder, concentration disorders and fatigue. Each one cycle higher initial Ct value significantly increased the chances of overcoming symptoms (odds ratio [OR] = 1.05; p = 0.019). The occurrence of breathlessness within the first ten days (OR = 7.89; p = 0.008), an initial Ct value under 30 (OR = 3.36; p = 0.022) and a definitely nosocomial SARS-CoV-2 transmission (OR = 3.05; p = 0.049) showed a statistically significant association with increased odds of illness duration over 90 days. Conclusion: About half of the HEs suffered from long lasting symptoms over 90 days after almost entirely mild acute COVID-19. Different symptom complexes could be shown and predictive factors for long-term symptoms were identified. Predictive factors at the onset of the infection could possibly be used for early treatment to prevent development of long-term symptoms after COVID-19 in future.

After the third wave of severe acute respiratory syndrome coronavirus type 2 (SARS- infections in springtime 2021 in Germany, we observed at University Hospital Cologne-Merheim a high number of infected hospital employees (HE) with long-lasting symptoms after predominantly mild acute courses of coronavirus disease 2019 . 1,2 In contrast to more frequently studied long-lasting symptoms after severe COVID-19 courses, [3] [4] [5] we were surprised by the amount of symptoms that also occurred after mild courses not requiring hospitalization. Since the vast majority of people infected with SARS-CoV-2 do not need to be hospitalized, 6, 7 it is of great importance to also consider mild COVID-19 cases with regard to long-lasting COVID-19 symptoms.

There are numerous different terms and definitions for long-lasting COVID-19 symptoms.

The World Health Organization developed the following clinical case definition: "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". 8 Here, we had the opportunity to investigate the entire cohort of SARS-CoV-2 infected HEs.

HEs comprise a quasi-representative cohort of the employable population, here in Cologne in Germany. Furthermore, SARS-CoV-2 positive HEs were not only detected by testing due to respiratory or influenza-like symptoms but also by active screening of all suspected contacts to known SARS-CoV-2 cases in either patients, colleagues, friends, or family members. Therefore, we reasoned that the cohort would be appropriate to study the variety, combination, and duration of symptoms after non-severe courses of COVID-19. Hence, we is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted March 22, 2022. ;  https://doi.org/10.1101/2022.03. 22 .22272770 doi: medRxiv preprint

A retrospective observational cohort study was conducted among SARS-CoV-2 positive hospital employees (HEs) of three medical centers in Cologne, Germany.

All HEs who had a positive SARS-CoV-2 PCR test from the beginning of the pandemic in March 2020 until May 2021 were recorded by staff of the hospitals' Institute of Hygiene. The following data was collected for each HE after positive SARS-CoV-2 PCR detection: name, gender, date of birth, address, phone number and/or E-mail address, date of SARS-CoV-2 detection, Ct values of multiple PCR testings, initial symptoms, contact persons and contact situations, vaccination status, data on the inpatient course in in the event of hospital admission. This data was required for fulfilment of the reporting obligations to the city's public health department. From the beginning of the pandemic, the Occupational Health Service of the three medical centers looked after HEs who had been infected with SARS-CoV-2 and in some cases suffered from long-lasting COVID-19 symptoms. All HEs with previous SARS-CoV-2 infection were contacted by phone and asked about their willingness to participate in a systematic survey on possible long-term COVID-19 symptoms. HEs were given the option to cancel the survey at any time. The survey was conducted by staff of both the Occupational Health Service and the Institute of Hygiene and was carried out between June and October 2021. All HEs interviewed had at least a six-week interval from their first SARS-CoV-2 positive PCR result at the time of the interview. In preparation for the survey, a structured questionnaire was jointly developed that included 24 different long-term COVID-19 symptoms that were frequently described in previously published studies about long-lasting COVID-19 symptoms, 7, 9, 10 and adapted to individual symptoms of the HEs (supplemental material). HEs were asked only about symptoms that occurred after SARS-CoV-2 detection and were not asked to report symptoms that had previously existed. The self-assessed symptom severity (mild/medium/severe) and the respective time of symptom onset and end . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint 

Descriptive analysis of frequency and duration of long-term COVID-19 symptoms were conducted. Analysis was performed using the Statistical Software R. 11 Using co-occurrence network analysis, the formation, patterns and chronology of COVID-19 symptom complexes after SARS-CoV-2 infection were assessed. 12 Survival-time analysis was conducted in order to assess the probability of respective symptom persistence using Kaplan-Meier estimator. 13 Likewise, univariate analysis of symptom durations depending on professional group, age, initial Ct value and other influencing factors was carried out using Cox regression. 14 Possible risk factors for a protracted COVID-19 course with long-lasting COVID-19 symptoms were identified using logistic regression. 15 P values <.05 were considered statistically significant.

This study was approved by the ethics committee of the Witten/Herdecke University (S-273/2021). The study was core funded. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint After SARS-CoV-2 detection, 27 HEs (12.2%) reported no symptoms at all. In 33 HEs (14.9%), at least one symptom occurred for a minimum of one day and lasted no longer than 28 days. More than 90 days after initial SARS-CoV-2 detection, 104 HEs (47.1%) reported at least one persisting symptom. The most common symptoms were fatigue (n = 139, 62.9%), headache (n = 121, 54.8%), anosmia (n = 120, 54.3%) and ageusia (n = 118, 53.4%) ( Table   2 ). With a median symptom duration of six days (mean = 7.7 days; SD = 8.49), fever is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The probability of symptom persistence was plotted using Kaplan-Meier curves ( Figure 5 ).

The symptom least likely to still be present after 90 days was fever (0.0% of n = 81), whereas alopecia was the symptom most likely to still be present after 90 days (78.0% of n = 24). For

HEs who developed shortness of breath after SARS-CoV-2 detection (n = 105), the probability for persisting shortness of breath after 200 days was 39.0%. For HEs suffering from memory disorder (n = 57), the probability for persistence of memory disorder after 200 days was 58.4%.

Cox regression analysis showed that each one cycle higher initial Ct value significantly increased the chances of overcoming symptoms (odds ratio [OR] = 1.05; p = 0.019) ( Table 3 ). The chances of being symptom-free for HEs who had a possibly nosocomial infection were significantly lower than for HEs with a community acquired SARS-CoV-2 infection (OR = 0.48; p = 0.044). The Cox regression model also indicated that there was a higher chance of symptom resolving for medical staff than for nursing staff (OR = 2.13; p = 0.055).

Logistic regression was used to identify factors that increased the probability of illness duration lasting more than 90 days. The occurrence of breathlessness within the first ten days increased the probability of a prolonged illness duration over 90 days by a factor of OR = 7.89 (p = 0.008), the presence of anosmia or ageusia increased the probability by a factor of OR = 3.01 (p = 0.037). Initial Ct value < 30 (OR = 3.36; p = 0.022) and a definitely nosocomial SARS-CoV-2 transmission (OR = 3.05; p = 0.049) also showed a statistically significant correlation with increased odds of illness duration over 90 days. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint 

In our cohort of SARS-CoV-2 positive tested and almost entirely still unvaccinated hospital employees during the first three waves (wild-type and alpha variant of SARS-CoV-2), a standardized interview revealed a high proportion of hospital employees with long lasting symptoms. Only twelve percent reported no symptoms at all, whilst 73% had symptoms longer than 28 days and 47% longer than 90 days. This goes in line with the work of Carvalho-Schneider et al., based on a prospective follow-up of 150 adults with noncritical COVID-19, that 68% of adults reported at least one lasting symptom at day 30 after symptom onset. 16 Symptoms could be grouped in those lasting shorter, mid and longer times. Shortlasting symptoms were fever, rhinitis, sore throat and limb pain. Long-lasting symptoms were alopecia, concentration and memory disorder, palpitations, breathlessness, shortness of breath and sleep disorder. Symptoms could be grouped into symptom complexes. Two previous studies about long-lasting symptoms after SARS-CoV-2 infections in patient cohorts, using self-reporting questionnaire information, both described dry cough, diarrhea, confusion and disorientation, pain, anosmia, ageusia, and fatigue as long-lasting symptoms. 7, 9 Comparison of different observational studies upon long lasting symptoms after non-severe SARS-CoV-2 infection revealed differences in the studied cohorts, the items of the interviews and the follow up. Whereas the first studies included patients who were asked for convalescent plasma donation or patients who were surveyed in the context of a quality management system, 7,9 here we included all employees of the Cologne Municipal Hospitals, representing the employable population. Employees were not only tested for SARS-CoV-2 if symptoms were present, but also all direct contact persons were actively screened for SARS-CoV-2 for 14 days after the documented contact. Hence, we were able to also include HEs with an asymptomatic course. Interestingly, only twelve percent of all employees had no symptoms at all. A share of 28% of employees included in the study were proven direct contacts to SARS-CoV-2 positive cases within the hospital, defined as definite nosocomial, and in additional 20% of cases, direct contacts could not be excluded, defined as possible is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint 17 in the present study we were able to capture symptoms that were not necessarily linked to a physician's disease diagnosis. These symptoms may nevertheless be essential to Just recently, a nationwide cross-sectional study in the Danish population with over 60.000 SARS-CoV-2 positive persons and a SARS-CoV-2 negative control group revealed significant risk differences for anosmia, fatigue/exhaustion, dyspnea and reduced strengths in legs/arms. 6 Our data, which do not include control groups, are consistent with the findings of the Danish questionnaire study.

In addition, we asked our employees when each symptom started and how long it lasted. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint As an additional limitation of our study, it must be mentioned that we only asked about subjectively perceived symptoms that were not necessarily confirmed by a physician. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint However, the work of Roessler et al. confirms our findings and shows that also even more severe conditions that were diagnosed by a trained physician occur more often in the COVID-19 cohort than in the control cohort. 17 Recently, Cohen et al. detected an increase of 11% for SARS-CoV-2 infected patients over the age of 65 for other sequelae. 24 In our questionnaire, we explicitly asked only about new onset symptoms. Nevertheless, it must be mentioned that we can not exclude that some employees may have overstated their reported symptoms.

The socioeconomic impact of long lasting symptoms after a SARS-CoV-2 infection was early described by the surrogate sick leave of working people in Sweden. Median leave time from work was 35 days and 9% had sick leaves longer than four months. 10 For both medical and socioeconomic reasons, predictors for long lasting symptoms are essential. Predictors were first identified by the Swedish study investigating sick leaves, which were older age and sick leaves within the year prior to infection in a first rough analysis. 10 We identified a significant association between the occurrence of ageusia, anosmia or breathlessness during the first ten days, first Ct value <30 as well as a definitely nosocomial SARS-CoV-2 transmission, and a prolonged illness duration over 90 days.

Through the identification of risk factors for long-lasting COVID-19 course, individuals who are at increased risk can be selected and medically treated at an early stage. 5 An early identification of risk factors is important because drug therapy for COVID-19, for example monoclonal antibody therapy or antiviral therapy with Nirmatrelvir/Ritonavir is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint importance of full vaccination against COVID-19 must be emphasized also with regard to the prevention of long-lasting COVID-19 symptoms.

Surprisingly, the nosocomial acquisition was a significant predictor for a long lasting course.

We may speculate the acquisition in a nosocomial context as a surrogate for a higher infectious dose. This is consistent with our finding of the association between a lower initial Ct value and a prolonged course.

In line with recent literature we found that long-lasting symptoms after acute COVID-19 occur frequently, even after initially mild courses. About half of the HEs developed symptoms lasting more than 90 days after almost entirely mild courses of COVID-19. Different symptom complexes that developed and changed over time could be shown. According to our data, long-lasting COVID-19 symptoms can be either single symptoms or symptom complexes. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint 

All authors have no interests to declare. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint To avoid misleading interpretation and since the modelled outcome is a positive event (asymptomatic), we refrain from hazard terminology in favor of odds.

. CC-BY-NC-ND 4.0 International license It is made available under a perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint Model results for odds of illness duration over 90 days. Only hospital employees with available initial Ct value were included (N=150). For nominal scaled variables, odds ratio refers to the reference characteristic (male gender, nursing staff, work on the non-COVID ward, community acquired, initial Ct value >30). Odds ratios were calculated for the presence of the listed symptoms within the first ten days after initial SARS-CoV-2 detection compared to the absence of the respective symptoms. Symptoms that occurred rarely were not considered. If two symptoms correlate strongly with each other, they are combined and the combined symptom is called observed as soon as one of the symptoms occurs.

The accuracy data of logistic regression model were as followed: accuracy 0.73, crossvalidation accuracy 0.61, sensitivity 0.79, cross-validation sensitivity 0.53, specificity 0.64, cross-validation specificity 0.50. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint 

WHO Working Group on the Clinical Characterisation and Management of COVID-19 infection. A minimal common outcome measure set for COVID-19 clinical research. The Lancet infectious diseases 2020

6-month consequences of COVID-19 in patients discharged from hospital: a cohort study

Assessment of the Frequency and Variety of Persistent Symptoms Among Patients With COVID-19: A Systematic Review

Short-term and Long-term Rates of Postacute Sequelae of SARS-CoV-2 Infection: A Systematic Review

Post-acute symptoms, new onset diagnoses and health problems 6 to 12 months after SARS-CoV-2 infection: a nationwide questionnaire study in the adult Danish population

Post-COVID syndrome in nonhospitalised patients with COVID-19: a longitudinal prospective cohort study. The Lancet regional health

World Health Organization. A clinical case definition of post COVID-19 condition by a Delphi consensus

Risk Predictors and Symptom Features of Long COVID Within a Broad Primary Care Patient Population Including Both Tested and Untested Patients

Patterns and predictors of sick leave after Covid-19 and long Covid in a national Swedish cohort

R: A Language and Environment for Statistical Computing

Network Visualizations of Relationships in Psychometric Data

Regression Models and Life Tables

Categorical Data Analysis

Follow-up of adults with noncritical COVID-19 two months after symptom onset. Clinical microbiology and infection : the official publication of the

Post COVID-19 in children, adolescents, and adults: results of a matched cohort study including more than 150,000 individuals with COVID-19

Pathological sequelae of long-haul COVID

Non-cell-autonomous disruption of nuclear architecture as a potential cause of COVID-19-induced anosmia

Severe Neuro-COVID is associated with peripheral immune signatures, autoimmunity and signs of neurodegeneration: a prospective cross-sectional study

Long COVID: An overview

Long COVID or post-COVID-19 syndrome: putative pathophysiology, risk factors, and treatments

Long-term cardiovascular outcomes of COVID-19

Risk of persistent and new clinical sequelae among adults aged 65 years and older during the post-acute phase of SARS-CoV-2 infection: retrospective cohort study

Medikamentöse Therapie bei COVID-19 mit Bewertung durch die Fachgruppe COVRIIN beim Robert Koch-Institut. 2022

Association between vaccination status and reported incidence of post-acute COVID-19 symptoms in Israel: a cross-sectional study of patients tested between

The authors would like to thank all hospital employees for taking part in the survey.We also thank all staff of the Occupational Health Service as well as the Institute of Hygiene, 

The study was designed and initiated by: FM, GS.The questionnaire was designed by: FM, VM, CW.Collection of data including programming related to this: VM, CW, FM, RG, GS.Data analysis was done by: SJ, SH.The first draft was written by: RG, FM.All authors have critically revised the manuscript.All authors have approved the final version of the manuscript and agreed to be accountable for all aspects of the work.