6-month mortality and readmissions of hospitalized COVID-19 patients: A nationwide cohort study of 8,679 patients in Germany

Christian Günster; Reinhard Busse; Melissa Spoden; Tanja Rombey; Gerhard Schillinger; Wolfgang Hoffmann; Steffen Weber-Carstens; Andreas Schuppert; Christian Karagiannidis

doi:10.1371/journal.pone.0255427

. 2021 Aug 5;16(8):e0255427. doi: 10.1371/journal.pone.0255427

6-month mortality and readmissions of hospitalized COVID-19 patients: A nationwide cohort study of 8,679 patients in Germany

Christian Günster ^1,^#, Reinhard Busse ^2,^#, Melissa Spoden ¹, Tanja Rombey ², Gerhard Schillinger ¹, Wolfgang Hoffmann ³, Steffen Weber-Carstens ⁴, Andreas Schuppert ⁵, Christian Karagiannidis ^6,^*

Editor: Aleksandar R Zivkovic⁷

¹Research Institute of the Local Health Care Funds, Federal Association of the Local Health Care Funds, Berlin, Germany

²Department of Health Care Management, Technische Universität Berlin, Berlin, Germany

³Section Epidemiology of Health Care and Community Health, Institute for Community Medicine, University Medicine Greifswald, Greifswald, Germany

⁴Department of Anesthesiology and Operative Intensive Care Medicine, Charité - Universitätsmedizin Berlin, Berlin, Germany

⁵Institute for Computational Biomedicine II, University Hospital RWTH Aachen University, Aachen, Germany

⁶Department of Pneumology and Critical Care Medicine, Cologne-Merheim Hospital, Kliniken der Stadt Köln, Witten/Herdecke University Hospital, Cologne, Germany

⁷Heidelberg University Hospital, GERMANY

Competing Interests: RB reports grants from Berlin University Alliance, during the conduct of the study; grants from Federal Ministry of Research and Education, grants from Federal Ministry of Health, grants from Innovation Fonds of the Federal Joint Committee, grants from World Health Organization, outside the submitted work, AS reports grants from Bayer AG, outside the submitted work. CK reports personal fees from Maquet, personal fees from Xenios, personal fees from Bayer, non-financial support from Speaker of the German register of ICUs, grants from German Ministry of Research and Education, during the conduct of the study. CG, MS, TR, GS, WH, and SWC have nothing to disclose.

^✉

* E-mail: Christian.Karagiannidis@uni-wh.de, karagiannidisc@kliniken-koeln.de

Contributed equally.

Roles

Christian Günster: Conceptualization, Data curation, Formal analysis, Methodology, Validation, Visualization, Writing – original draft

Reinhard Busse: Conceptualization, Formal analysis, Methodology, Resources, Supervision, Validation, Writing – original draft

Melissa Spoden: Data curation, Formal analysis, Methodology, Validation, Visualization

Tanja Rombey: Data curation, Methodology, Validation, Writing – original draft

Gerhard Schillinger: Conceptualization, Formal analysis, Methodology, Project administration, Resources, Supervision, Validation

Wolfgang Hoffmann: Formal analysis, Supervision, Writing – review & editing

Steffen Weber-Carstens: Formal analysis, Methodology, Supervision, Validation, Writing – review & editing

Andreas Schuppert: Methodology, Supervision, Validation, Writing – review & editing

Christian Karagiannidis: Conceptualization, Formal analysis, Methodology, Project administration, Supervision, Validation, Writing – original draft

Aleksandar R Zivkovic: Editor

PMCID: PMC8341502 PMID: 34351975

Abstract

Background

COVID-19 frequently necessitates in-patient treatment and in-patient mortality is high. Less is known about the long-term outcomes in terms of mortality and readmissions following in-patient treatment.

Aim

The aim of this paper is to provide a detailed account of hospitalized COVID-19 patients up to 180 days after their initial hospital admission.

Methods

An observational study with claims data from the German Local Health Care Funds of adult patients hospitalized in Germany between February 1 and April 30, 2020, with PCR-confirmed COVID-19 and a related principal diagnosis, for whom 6-month all-cause mortality and readmission rates for 180 days after admission or until death were available. A multivariable logistic regression model identified independent risk factors for 180-day all-cause mortality in this cohort.

Results

Of the 8,679 patients with a median age of 72 years, 2,161 (24.9%) died during the index hospitalization. The 30-day all-cause mortality rate was 23.9% (2,073/8,679), the 90-day rate was 27.9% (2,425/8,679), and the 180-day rate, 29.6% (2,566/8,679). The latter was 52.3% (1,472/2,817) for patients aged ≥80 years 23.6% (1,621/6,865) if not ventilated during index hospitalization, but 53.0% in case of those ventilated invasively (853/1,608). Risk factors for the 180-day all-cause mortality included coagulopathy, BMI ≥ 40, and age, while the female sex was a protective factor beyond a fewer prevalence of comorbidities. Of the 6,235 patients discharged alive, 1,668 were readmitted a total of 2,551 times within 180 days, resulting in an overall readmission rate of 26.8%.

Conclusions

The 180-day follow-up data of hospitalized COVID-19 patients in a nationwide cohort representing almost one-third of the German population show significant long-term, all-cause mortality and readmission rates, especially among patients with coagulopathy, whereas women have a profoundly better and long-lasting clinical outcome compared to men.

Introduction

Background

Within one year, the SARS-CoV-2 pandemic has affected more than 125 million people worldwide. During the first wave in spring 2020, hospitalization rates were high, reaching up to 70% in France, 55% in Spain, 50% in the UK and 20% in Germany until the end of April [1]. Mortality rates of hospitalized patients were also high at 12.5% in France [2], and more than 20% in the UK and Germany [3, 4], especially among patients requiring mechanical ventilation with up to 50%. Little is known about the long-term outcome of hospitalized COVID-19 patients, though, in general, there is increasing evidence of a long-COVID syndrome, affecting different organ systems.

Recently, the 6-month follow-up data of the first 1,733 hospitalized COVID-19 patients from Wuhan were published [5]. COVID-19 survivors suffered mainly from fatigue or muscle weakness, sleep difficulties, and anxiety or depression. These data are in line with recent data of an increased risk for neurological and psychiatric disorders after 6 months [6–10]. In case of severely ill COVID-19 patients, it has been found among 167 COVID-19 patients admitted to an intensive care unit (ICU) that older age and thrombocytopenia, among others, were significant risk factors for mortality at 28, 90 and 180 days [11]. Furthermore, in a multicentre-study from five European countries of hospitalized COVID-19 patients requiring extracorporeal membrane oxygenation (ECMO) showed that advanced age and low pH before ECMO were also associated with an increased risk of mortality within 6 months, with an overall rate of 53% (70/132) [12].

For large cohorts outside Wuhan, follow-up data have been reported up to a maximum of three months. Ninety-day post-admission outcomes have been reported from various European countries, albeit often for single hospitals. The 90-day mortality rate ranged from 11% in Spain [13] to 29% in Denmark [14] (both single-centre studies) for all hospitalized patients, and was 27% in Sweden [15], 31% in Belgium, France and Switzerland (both multi-centre studies) [16], and 35% in a Danish single-centre study for patients treated in the ICU [14]. An analysis of more than 100,000 hospitalized COVID-19 patients in the United States revealed a readmission rate of 9% to the same hospital 6 months post-discharge [17].

Objectives

Since a large cohort of 6-month follow-up data is currently lacking, the aim of this observational study was to determine the 6-month all-cause mortality and readmission rates of hospitalized COVID-19 patients who completed hospital treatment following a confirmed COVID-19 diagnosis, focusing particularly on patients requiring mechanical ventilation. Furthermore, factors associated with 6-month all-cause mortality were evaluated.

Materials and methods

Study design

This was a retrospective observational cohort study, using claims data. The study was investigator-initiated without any funding. It was approved by the Ethics Committee of the Witten/Herdecke University (research ethics board number 92/2020).

Setting

We used nationwide anonymous administrative claims data for in-patient episodes (including diagnoses, procedures, length of stay, transfers and discharge type), and core data (including age, gender, insurance status and survival status) of the sickness funds group “German local healthcare funds” (Allgemeine Ortskrankenkassen, AOK). In Germany, health insurance is obligatory and nearly every inhabitant has health insurance [18].

AOK is the largest sickness funds group within Germany’s statutory health insurance system. It provides statutory health insurance for roughly 32 percent of the German population, for whom it is representative when considering the factor age [19, 20]. Furthermore, membership in a sickness fund is open to anyone regardless of factors such as professional affiliation, income, or comorbidities. Sickness funds are obliged to accept any applicant and charge the same basic contribution rate [18]. According to the German accounting method for the health care system, all diagnoses, outcomes, and procedures must be reported to the sickness funds, as required by law. Strict legal requirements and verification by both hospitals and sickness funds aim to minimize any bias introduced by false coding and other systemic errors relating to the data. In addition, we checked data plausibility and consistency before analyzing them. Diagnoses were coded according to the 10^th revision of the International Classification of Diseases (ICD-10-GM) and procedures according to the International Classification of Procedures in Medicine, the “Operationen- und Prozedurenschlüssel” (OPS).

Study population

For the analyses, we included only patients with COVID-19 infection as confirmed by reverse transcription polymerase chain reaction (PCR) tests (diagnosis code U07.1), who were at least 18 years old, and were admitted to hospital for treatment of COVID-19 between February 1, 2020 and April 30, 2020, both dates included, and were discharged by June 30, 2020.

As COVID-19 cannot be coded as a principal diagnosis, we defined hospitalization for COVID-19 as an admission with a COVID-19-related principal diagnosis of respiratory failure, pulmonary embolism, viral infection, sepsis or renal failure during the initial (“index”) hospitalization with confirmed COVID-19 infection. A list of all included diagnoses is provided in S1 Table. The selection of patients was done to include only patients in whom COVID-19 was the primary reason for their hospital stay and to exclude those in whom COVID-19 was an incidental finding likely to be unrelated to their hospital stay. The unit of analysis was the individual patient. Since one person might have had several hospital stays during the observation period due to a transfer from one hospital to another, we grouped adjacent completed hospital stays into one patient. Each patient was followed up for 180 days after admission for survival status, and after discharge for readmissions. Patients had to be continuously insured with AOK for that period of time, unless they died earlier while still insured with AOK. This was a complete survey of all AOK insured.

Endpoints

All-cause mortality was measured as in-hospital mortality, i.e. death occurring during the index hospitalization, as well as in or out-of-hospital mortality within 30 days, 90 days and 180 days after the day of initial admission. The date of out-of-hospital death was extracted from the core data of the sickness funds.

For readmissions, we followed patients discharged alive from the index hospitalization for 180 days after the day of discharge. The discharge date was chosen because the patients’ length of index hospital stay varied widely. We determined both how many patients were readmitted at least once, and how many readmissions occurred in total for any particular cause, and for readmission with ventilation or potentially COVID-19-related systemic, respiratory, renal and neurological, gastrointestinal and liver as well as cardiovascular diagnoses only (S1 Table).

Statistical analysis

Baseline characteristics are described in terms of means and standard deviations (SD), and medians and inter-quartile ranges (IQR) for continuous variables, and in terms of proportions for categorical variables. Characteristics are shown for the whole study population and for patients with and without mechanical ventilation. For ventilated patients, two subgroups were formed: (a) patients with non-invasive mechanical ventilation only (NIV), and (b) patients with invasive mechanical ventilation (IMV), including those with non-invasive mechanical ventilation failure. Baseline characteristics are defined on the basis of conditions and interventions during index hospitalization and include gender, age groups (18–59 years, 60–69 years, 70–79 year, ≥80 years), selected Elixhauser comorbidities [21, 22], and type of ventilation. In addition, we report on other procedures, such as dialysis or ECMO, as well as further comorbidities and complications, such as septic shock, acute respiratory distress syndrome (ARDS), or renal failure post medical procedures. See S1 Table.

We report all-cause mortality (in-hospital, 30/90/180 days after admission) for patients grouped according to their baseline characteristics.

Kaplan-Meier curves for survival up to 180 days after the initial admission are presented by gender, age, type of ventilation, and selected Elixhauser comorbidities.

Owing to non-parallel Kaplan-Meier curves and a significant deviation from the proportional hazards assumption, no Cox-proportional hazard model could be estimated. The proportional hazards assumption was evaluated visually on plots of log (-log[survival]) versus log of survival time adjusted for covariates and by a global test based on Schoenfeld residuals. Instead, multivariable logistic regression was used to model the odds of the binary endpoint of 180-day all-cause mortality after admission as a function of age, sex, body mass index (BMI) categories (30≤34, 35≤39, ≥40 kg/m2) and Elixhauser comorbidities present at index hospitalization. We used cluster-robust standard errors to account for the clustering of patients in hospitals. Comorbidity conditions were defined as binary variables. A model including all potential risk factors was estimated first, with subsequent removal of risk factors that did not prove to be statistically significant (p ≥ 0.05). A Wald test was performed to confirm that the removal of these risk factors from the model did not result in any substantial loss in model fit. Adjusted odds ratios (OR) and 95% confidence intervals (CIs) were calculated and summarized in a forest plot. To evaluate the performance of the model, the area under the curve (AUC) was used as a measure of discrimination and the squared Pearson correlation (R²) between 180-day all-cause mortality and the log-odds of predicted mortality was used as a measure of the explained variation. All analyses were performed using STATA 16.0 (StataCorp, College Station, Texas).

Results

Study cohort

The patient flow is displayed in Fig 1. Out of the 1.075 million AOK-insured persons with at least one in-patient admission between February 1 and April 30, 2020, we identified 11,459 adult patients with PCR-confirmed COVID-19 diagnosis who were discharged before June 30, 2020. We excluded 2,572 patients who did not suffer from a COVID-19-related principal diagnosis but were admitted for diagnoses such as congestive heart disease (n = 161) or femur fracture (n = 139). Furthermore, we had to exclude 208 patients, who could not be followed for the full 6-month period after their initial admission or until death. In total, 8,679 patients were confirmed eligible and included in the analysis.

Demographics and comorbidity

The patients’ demographic characteristics are shown in Table 1. The cohort comprised slightly more men (4,641/8,679; 53.5%) than women (4,038/8,679; 46.5%). The median age was 72 years (IQR 57 to 82), the largest share of patients being in the age group of 80 years and older (2,817/8,679; 32.5%). The most observed comorbidities were hypertension (4,920/8,679; 56.7%), fluid and electrolyte disorders (4,641/8,679; 53.5%), diabetes mellitus (uncomplicated: 1,912/8,679; 22.0%; complicated: 734/8,679; 8.5%), cardiac arrhythmia (2,373/8,679; 27.3%), renal failure (1,994/8,679; 23.0%), and congestive heart failure (1,652/8,679; 19.0%).

Table 1. Patient characteristics and all-cause mortality.

Patients hospitalized for a COVID-19-related principal diagnosis	Total	In-hospital all-cause mortality	30-day all-cause mortality after hospital admission	90-day all-cause mortality after hospital admission	180-day all-cause mortality after hospital admission
	(n = 8,679)	(n = 2,161)	(n = 2,073)	(n = 2,425)	(n = 2,566)
		N (row percentage of total patients)
Total (N)	8,679 (100.0%)	2,161 (24.9%)	2,073 (23.9%)	2,425 (27.9%)	2,566 (29.6%)
Male	4,641 (53.5%)	1,293 (27.9%)	1,209 (26.1%)	1,416 (30.5%)	1,489 (32.1%)
Female	4,038 (46.5%)	868 (21.5%)	864 (21.4%)	1,009 (25.0%)	1,077 (26.8%)
Age, years		(value refers to column total)
Mean (SD)	68.6 (16.6)	78.4 (11.0)	79.2 (10.6)	78.8 (10.9)	78.9 (10.9)
Median (IQR)	72.0 (57.0–82.0)	81.0 (73.0–86.0)	81.0 (75.0–86.0)	81.0 (73.0–86.0)	81.0 (73.0–86.0)
Age groups, years		N (row percentage of total patients)
18–59 years	2,451 (28.2%)	152 (6.2%)	123 (5.0%)	158 (6.4%)	165 (6.7%)
60–69 years	1,506 (17.4%)	263 (17.5%)	221 (14.7%)	282 (18.7%)	291 (19.3%)
70–79 years	1,905 (21.9%)	548 (28.8%)	504 (26.5%)	598 (31.4%)	638 (33.5%)
80 years and older	2,817 (32.5%)	1,198 (42.5%)	1,225 (43.5%)	1,387 (49.2%)	1,472 (52.3%)
Elixhauser comorbidities		N (row percentage of total patients)
Hypertension	4,920 (56.7%)	1,296 (26.3%)	1,227 (24.9%)	1,478 (30.0%)	1,577 (32.1%)
Fluid and electrolyte disorders	4,641 (53.5%)	1,445 (31.1%)	1,343 (28.9%)	1,618 (34.9%)	1,710 (36.8%)
Cardiac arrhythmias	2,373 (27.3%)	940 (39.6%)	876 (36.9%)	1,035 (43.6%)	1,101 (46.4%)
Renal failure	1,994 (23.0%)	770 (38.6%)	746 (37.4%)	878 (44.0%)	942 (47.2%)
Diabetes, uncomplicated	1,912 (22.0%)	575 (30.1%)	537 (28.1%)	631 (33.0%)	664 (34.7%)
Congestive heart failure	1,652 (19.0%)	700 (42.4%)	631 (38.2%)	769 (46.5%)	823 (49.8%)
Chronic pulmonary disease	1,188 (13.7%)	360 (30.3%)	327 (27.5%)	389 (32.7%)	420 (35.4%)
Diabetes, complicated	734 (8.5%)	280 (38.1%)	261 (35.6%)	312 (42.5%)	333 (45.4%)
Other neurological disorders	671 (7.7%)	255 (38%)	235 (35.0%)	296 (44.1%)	312 (46.5%)
Coagulopathy	623 (7.2%)	285 (45.7%)	210 (33.7%)	290 (46.5%)	307 (49.3%)
Depression	601 (6.9%)	104 (17.3%)	101 (16.8%)	125 (20.8%)	144 (24.0%)
Liver disease	411 (4.7%)	184 (44.8%)	139 (33.8%)	187 (45.5%)	194 (47.2%)
Pulmonary circulation disorders	358 (4.1%)	157 (43.9%)	126 (35.2%)	164 (45.8%)	175 (48.9%)
Weight loss	351 (4.0%)	116 (33%)	98 (27.9%)	125 (35.6%)	142 (40.5%)
Metastatic cancer	53 (0.6%)	26 (49.1%)	23 (43.4%)	32 (60.4%)	37 (69.8%)
Further comorbidities		N (row percentage of total patients)
Cognitive impairment	2,207 (25.4%)	718 (32.5%)	719 (32.6%)	874 (39.6%)	947 (42.9%)
Delir, anoxia encephalopathy, somnolence, sopor and coma	1,058 (12.2%)	379 (35.8%)	330 (31.2%)	429 (40.5%)	459 (43.4%)
BMI ≥ 40	173 (2.0%)	59 (34.1%)	45 (26.0%)	60 (34.7%)	62 (35.8%)
Length of stay of index hospitalization		(value refers to column total)
Mean (SD)	16.5 (19.5)	12.8 (14.2)	9.6 (7.3)	13.0 (12.7)	14.0 (15.1)
Median (IQR)	10.0 (6.0–20.0)	8.0 (4.0–16.0)	7.0 (4.0–14.0)	9.0 (4.0–17.0)	9.0 (5.0–18.0)
Ventilation during index hospitalization		N (row percentage of total patients)
Not ventilated	6,865 (79.1%)	1,248 (18.2%)	1,304 (19.0%)	1,503 (21.9%)	1,621 (23.6%)
Invasive ventilation	1,608 (18.5%)	832 (51.7%)	695 (43.2%)	836 (52.0%)	853 (53.0%)
Only non-invasive ventilation	206 (2.4%)	81 (39.3%)	74 (35.9%)	86 (41.7%)	92 (44.7%)
Procedures during index hospitalization		N (row percentage of total patients)
Dialysis	838 (9.7%)	474 (56.6%)	382 (45.6%)	482 (57.5%)	508 (60.6%)
Tracheostomy	582 (6.7%)	221 (38.0%)	118 (20.3%)	217 (37.3%)	230 (39.5%)
Extracorporeal membrane oxygenation	160 (1.8%)	102 (63.8%)	68 (42.5%)	100 (62.5%)	102 (63.8%)
Haemofiltration	51 (0.6%)	37 (72.5%)	33 (64.7%)	37 (72.5%)	37 (72.5%)
Complications during index hospitalization		N (row percentage of total patients)
Septic shock	1,406 (16.2%)	746 (53.1%)	635 (45.2%)	764 (54.3%)	787 (56.0%)
ARDS	1,329 (15.3%)	692 (52.1%)	590 (44.4%)	696 (52.4%)	708 (53.3%)
Renal failure post procedure	1,252 (14.4%)	739 (59.0%)	637 (50.9%)	766 (61.2%)	795 (63.5%)
Lung embolism	188 (2.2%)	74 (39.4%)	61 (32.4%)	78 (41.5%)	81 (43.1%)
Intracerebral bleeding, cerebral infarction, stroke	122 (1.4%)	60 (49.2%)	44 (36.1%)	63 (51.6%)	65 (53.3%)
Acute myocardial infarction	112 (1.3%)	66 (58.9%)	60 (53.6%)	69 (61.6%)	70 (62.5%)
Deep vein thrombosis	88 (1.0%)	18 (20.5%)	15 (17.0%)	23 (26.1%)	27 (30.7%)
Myocarditis	38 (0.4%)	13 (34.2%)	11 (28.9%)	14 (36.8%)	14 (36.8%)
Lung edema	15 (0.2%)	2 (13.3%)	0 (0.0%)	2 (13.3%)	3 (20.0%)

Open in a new tab

Data are n (%), median (IQR) or mean (SD). BMI: Body Mass Index; ARDS: Acute respiratory distress syndrome.

Index hospitalization

Median length of stay of the index hospitalization was 10 days (IQR 6 to 20); the mean length of stay was 16.5 (SD 19.4) days (Table 1). 6,865 (79.1%) patients were treated without mechanical ventilation and 1,814 (20.9%) were treated with mechanical ventilation, of whom 1,608 (18.5%) received IMV and 206 (2.4%) received NIV only. The patients’ demographic characteristics stratified by ventilation status can be found in S2 Table. Dialysis was performed in 9.7% (838/8,679) and ECMO in 1.8% (160/8,679). Frequent complications during the index hospitalization included septic shock (1,406/8,679; 16.2%), ARDS (1,329/8,679; 15.3%), and acute renal failure (1,252/8,679; 14.4%).

All-cause mortality

Of the 8,679 included patients, 2,161(24.9%) died during the index hospitalization (Table 1). Measured from the day of initial admission, 30-day all-cause mortality was 23.9% (2,073/8,679), 90-day all-cause mortality was 27.9% (2,425/8,679), and 180-day all-cause mortality was 29.6% (2,566/8,679).

Women’s survival rates were about 5 percentage points higher than men’s at all three time points (Fig 2A). Survival was also strongly associated with age. Patients in the age group of 18–59 years had a 180-day all-cause mortality of 6.7% (1,65/2,451), which increased to 19.3% (291/1,506) in patients aged 60–69 years, 33.5% (638/1,905) in patients aged 70–79, and 52.3% (1,472/2,817) in patients aged 80 years or above (Fig 2B). The relative difference between in-hospital and 180-day all-cause mortality was also the highest in this age group with an additional 9.7% of patients dying during follow-up (Table 1). Patients who were not ventilated had a better survival record than ventilated patients (Fig 2C), their 180-day all-cause mortality being 23.6% (1,621/6,865) compared to 52.1% (945/1,814) of ventilated patients. Amongst ventilated patients, 30-, 90- and 180-day all-cause mortality rates were about 8% percentage points lower in patients treated with NIV compared to IMV.

Among the comorbidities present in at least 5% of the sample, the greatest difference between 30-day and 180-day all-cause mortality was observed for the following comorbidities: Coagulopathy, congestive heart failure, other neurological disorders, renal failure, and complicated diabetes mellitus (Table 1). Patients with coagulopathy had the largest increase in 180-day all-cause mortality, with an additional 15.6% (97/623) of patients dying between 30 and 180 days (Fig 2D). Survival was considerably better for patients without these comorbidities. For patients with coagulopathy, 180-day all-cause mortality was 49.3% (307/623; Fig 2D), with congestive heart failure 49.8% (823/1,652; Fig 2E), with other neurological disorders 46.5% (312/671; Fig 2F), with renal failure 47.2% (942/1,994; Fig 2G), and with complicated diabetes 45.4% (333/734; Fig 2H).

All-cause mortality among ventilated patients

In a subgroup analysis of ventilated patients only, the observed trends regarding gender, age, and ventilation type (non-invasive/invasive) persisted (Fig 3A–3C), while the differences narrowed (congestive heart failure and renal failure; Fig 3E and 3G) or vanished in the analyses related to comorbidity (Fig 3D, 3F and 3H). For patients with coagulopathy, the 30-day all-cause mortality was even lower than that of patients with no coagulopathy (Fig 3D); for patients with other neurological disorders, this was even observed beyond Day 90 (Fig 3F).

Factors associated with all-cause mortality

The results of the logistic regression model are presented for all covariates significantly associated with 180-day all-cause mortality in Fig 4. The model had good discrimination (AUC = 0.81; 95%-CI 0.80 to 0.82) and fit (R² = 0.23). Adjusted for age, sex and comorbidities, strong associations with increased odds of 180-day all-cause mortality (OR > 2) were observed for patients with a BMI ≥ 40 (OR 2.01, 95%-CI 1.33 to 3.05), liver disease (OR 2.45, 95%-CI 1.85 to 3.25), metastatic cancer (OR 8.02, 95%-CI 3.57 to 18.00), and coagulopathy (OR 2.31. 95%-CI 1.82 to 2.94). For age, the OR was 1.08 per year, indicating that the odds for 180-day all-cause mortality increase by the factor 2.21 per additional 10 years of age (1.082449^10), 4.88 per additional 20 years of age (1.082449^20), and so on. Conversely, a strong association for decreased odds of 180-day all-cause mortality was observed for female patients (OR 0.63, 95%-CI 0.56 to 0.70), and patients with depression (OR 0.46, 95%-CI 0.37 to 0.57).

Fig 4 — CI confidence interval, OR odds ratio, BMI body mass index. Only significant risk factors are included in the model. Each covariate has been adjusted for all other covariates displayed.

Readmissions

Of the 6,518 patients discharged alive from index hospitalization, 283 patients were lost to follow-up, as they died after 180 days of initial admission but before 180 days post discharge or were not insured with AOK anymore. Of the remaining 6,235 patients, 1,668 were readmitted a total of 2,551 times for some cause within180 days post discharge, resulting in an overall readmission rate of 26.8%.

Of the 6,518 patients discharged alive, 405 (6.2%) patients died within 180 days of the initial admission. Around half of them (201/405; 49.6%) were readmitted within 180 days of discharge, while the other half was not (204/405; 50.4%). Thus, the increase in 180-day all-cause mortality post discharge from the initial hospitalization was much higher in readmitted patients (201/1,668; 12.1%) than in those not readmitted (204/4,850; 4.2%).

Unlike mortality rates, readmission rates were only slightly higher among men (893/3,204; 27.9%) than women (775/3,031; 25.6%) (Table 2). Patients treated with IMV (231/717; 32.2%) had higher readmission rates than patients treated with NIV (34/121; 28.1%) or patients who were not ventilated (1,403/5,397; 26.0%).

Table 2. Readmission within 180-days of discharge.

Readmission among patients discharged alive from index admission	Total		Male		Female
	(n = 6,235)		(n = 3,204)		(n = 3,031)
All patients with at least one readmission	1,668 (26.8%)		893 (27.9%)		775 (25.6%)
Patients with ventilation during index admission
Exclusively non-invasive ventilation
34/121 (28.1%)		20/78 (25.6%)		14/43 (32.6%)
Invasive ventilation
231/717 (32.2%)		154/453 (34.0%)		77/264 (29.2%)
No ventilation
1,403/5,397 (26.0%)	719/2,673 (26.9%)		684/2,724 (25.1%)
COVID-19-related readmission within 180 days	Principal or secondary diagnosis (% of readmitted patients)
Any systemic/ respiratory/ renal/ neuro/ gastrointestinal and liver/ cardiovascular/ ventilation disorder or complication/ COVID-19	1,011 (60.6%)		550 (61.6%)		461 (59.5%)
Systemic disorders/complications	162 (9.7%)		104 (11.6%)		58 (7.5%)
Respiratory disorders/complications	601 (36.0%)		351 (39.3%)		250 (32.3%)
Renal disorders/complications	203 (12.2%)		119 (13.3%)		84 (10.8%)
Neurological disorders/complications	490 (29.4%)		244 (27.3%)		246 (31.7%)
Gastrointestinal and liver disorders/complications	117 (7.0%)		77 (8.6%)		40 (5.2%)
Cardiovascular disorders/complications	192 (11.5%)		115 (12.9%)		77 (9.9%)
Mechanical ventilation (non-invasive and invasive)	103 (6.2%)		66 (7.4%)		37 (4.8%)
COVID-19 (U07.1!)	212 (12.7%)		124 (13.9%)		88 (11.4%)

Open in a new tab

The majority of the readmitted patients (1,011/1,668; 60.6%), who had potentially COVID-19-related systemic, respiratory, renal, neurological, gastrointestinal liver, and cardio vascular principal/secondary diagnoses, were ventilated at readmission and/or had a positive PCR test for COVID-19 (Table 2). When viewed separately, the most frequent principal or secondary diagnoses for readmission were respiratory disorders or complications (601/1,668; 36.0%), and neurological disorders or complications (490/1,668; 29.4%). Less than ten percent of the patients received mechanical ventilation at readmission (103/1,668; 6.2%). 212 (12.7%) patients tested again positive for COVID-19 at readmission.

Discussion

This is the first study showing the 6-month all-cause mortality of hospitalized COVID-19 patients in a nationwide cohort, including readmissions within this time period. The major findings are the high 6-month all-cause mortality in COVID-19 patients, particularly among those requiring mechanical ventilation at 53%, that women have a sustained and profoundly beneficial 6-month outcome compared to men, and that patients aged above 80 years have the worst outcome with a 6-month all-cause mortality rate of 52% or up to 71% for those being ventilated. Furthermore, coagulopathy, liver diseases and severe obesity are associated with poor 6-month all-cause mortality. Lastly, readmission rates reached 27% with most of the patients with potentially COVID-19-related diagnoses being admitted for respiratory or neurological disorders or complications.

The initial in-hospital all-cause mortality of 25%, which is in line with many other European countries, increased to an all-cause mortality of 30% after 6 months, which demonstrates severe major prolonged implications of this disease, rather more than we would have expected. It is noticeable that ventilated patients have a poor overall outcome, especially patients over 70 years of age. In contrast, it is also evident that, in the younger age groups, there is only a slight increase in all-cause mortality after hospital discharge, although serious long-term consequences, having a huge impact on morbidity and quality of life may occur.

With regard to risk factors being associated with a poor or a more beneficial long-term outcome, several key factors became obvious. Overall, women show a better long-term outcome than men regardless of other confounding factors in terms of all-cause mortality, which may be due to their enhanced immune and inflammatory response to COVID-19 compared to men [23–26]. This is likely caused by their different genetic and endocrine mechanisms, including sex hormone actions, which might also influence the mechanisms of coagulopathy and thrombosis in COVID-19 [27]. On the other hand, factors contributing to increased all-cause mortality are especially disorders of the coagulation system, or liver disease, as already known from in-hospital mortality. These data are in line with our current understanding of COVID-19, particularly regarding the disorders of the coagulation or cerebrovascular system [28–30].

While diabetes generally worsens the outcome, this is not the case for patients being mechanically ventilated, whereas acute renal failure, congestive heart failure and age account for a worse outcome, independent of being ventilated. However, further data on patients’ diagnoses before admission with COVID-19 are absent in this study. In the light of the current analysis, it should be critically evaluated whether current intensive care therapy, including mechanical ventilation in patients over 80 years of age, is really effective in view of the very high mortality rate, or whether, in future, we should develop narrow criteria for eligible patients so that they could have a more favorable outcome. This certainly includes the frailty of the elderly [31–33].

Within 180 days of discharge, there was a high number of readmissions, representing of 27% of those discharged alive, primarily due to respiratory or neurological diagnoses, independent of gender. Also, all-cause mortality in readmitted patients remains rather high at 6% of all discharged patients. Besides the lung, readmissions with neurological complications may be a severe manifestation of the disease which may last for several months and lead to long-term complications, with their outcomes remaining unknown. Patients with coagulopathy had the highest 180-day all-cause mortality, which sheds a special light on the early detection of possible thrombosis or pulmonary embolism following the initial admission. In general, a close follow-up by general practitioners and the corresponding specialized disciplines is needed for early interventions regarding neurocognitive impairments. Notably, 13% of all readmitted patients were still or again positive for SARS-CoV-2, pointing out that virus elimination may take a long time in some severely ill patients.

Strengths and limitations

The major strength of our study is its size and length of follow-up, our cohort of being the largest cohort of patients hospitalized for COVID-19 that was followed up for six months after admission and discharge, respectively. The data source (claims data) also has several strengths. First, due to the administrative nature of the data, we present an almost complete cohort of 98% of the eligible patients with full follow-up data. Furthermore, information on readmission is independent of whether readmission was handled by the same or a different hospital. Second, since the choice of hospital is free under German statutory health insurance, our data set includes hospitals ranging from major tertiary referral centers to smaller regional hospitals, comprising real-world data unbiased by the degree of hospital specialization. Third, in-patient data is of high quality because disease and procedure codes are relevant for the amount of reimbursement and are, therefore, verified by hospitals and sickness funds. Lastly, unlike most studies using administrative data, we were able to assign data from different hospital stays to the individual patient so that the unit of analysis was the patient and not the hospital case.

There are also several limitations relating to the data source. First, it only includes patients from one group of German sickness fund. However, it is the largest group, which accounts for one-third of the total population, providing a large sample representative of the German population. Second, patient-specific data are limited to in-patient diagnoses, procedures, and initial characteristics, so some pre-existing conditions might have remained unknown. Third, we stratified by mechanical ventilation, and not by ICU treatment (as it is not coded separately), which sometimes includes high-flow oxygen therapy without mechanical ventilation. Fourth, detailed data such as laboratory values and information on patient preferences or clinical decision-making that may impact the initiation of invasive treatments are not available.

General limitations include that, given its observational nature, this study cannot determine causality between risk factors and long-term mortality, nor does it provide information on the patients’ cause of death. Lastly, the inclusion of patients admitted for COVID-19-related principal diagnoses might not be sufficient to distinguish between patients who were hospitalized for COVID-19 and patients with COVID-19 who were hospitalized for other reasons. Notably, 6-month all-cause mortality was similar between the in- and excluded patients.

Conclusions

In this nationwide cohort of patients hospitalized for COVID-19, considerable long-term all-cause mortality and readmission rates were observed. Patients with coagulopathy had the highest increase in 180-day all-cause mortality, followed by congestive heart failure, neurological diseases, and acute renal failure. However, the female sex is a profoundly protective factor in the COVID-19 disease.

Supporting information

S1 Table. Diagnosis codes of included patients.

(DOCX)

Click here for additional data file.^{(16.5KB, docx)}

S2 Table. Patient characteristics stratified by ventilation status.

(DOCX)

Click here for additional data file.^{(25KB, docx)}

Data Availability

The data used in this study cannot be made available in the manuscript, the supplemental files, or in a public repository due to German data protection laws (Bundesdatenschutzgesetz). Therefore, they are stored on a secure drive in the Wissenschaftliches Institut der AOK to facilitate replication of the results. Generally, access to data of statutory health insurance funds for research purposes is possible only under the conditions defined in the German Social Law (SGB V § 287). Requests for data access can be sent as a formal proposal, specifying the recipient and purpose of the data transfer, to the appropriate data protection agency. Access to the data used in this study can only be provided to external parties under the conditions of the cooperation contract of this research project and after written approval by the AOK. For assistance in obtaining access to the data, please contact wido@wido.bv.aok.de.

Funding Statement

Institutional support and physical resources were provided by the University Witten/ Herdecke and Kliniken der Stadt Köln, the Federal Association of the Local Health Care Funds and the Technical University of Berlin. The latter also received a grant from the Berlin University Alliance (112_PreEP_Corona). Article processing fees were funded by the authors. No funding source had a role in the design or conduct of the study; data collection, management, analysis, or interpretation; or the preparation, review, or approval of the manuscript.

References

1.Department Health Care Management. Data on Covid-19 hospitalisations /ICU treatments across European countries with data available as of 17/03/21 Berlin: Technische Universität Berlin; 2021. Available from: https://www.mig.tu-berlin.de/fileadmin/a38331600/sonstiges/COVID-19-STATS_170321_1945.pdf. Last accessed: 27 June 2021.
2.Bonnet G, Weizman O, Trimaille A, Pommier T, Cellier J, Geneste L, et al. Characteristics and outcomes of patients hospitalized for COVID-19 in France: The Critical COVID-19 France (CCF) study. Arch Cardiovasc Dis. 2021. doi: 10.1016/j.acvd.2021.01.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Karagiannidis C, Mostert C, Hentschker C, Voshaar T, Malzahn J, Schillinger G, et al. Case characteristics, resource use, and outcomes of 10 021 patients with COVID-19 admitted to 920 German hospitals: an observational study. Lancet Respir Med. 2020;8(9):853–62. doi: 10.1016/S2213-2600(20)30316-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Docherty AB, Harrison EM, Green CA, Hardwick HE, Pius R, Norman L, et al. Features of 20 133 UK patients in hospital with covid-19 using the ISARIC WHO Clinical Characterisation Protocol: prospective observational cohort study. Bmj. 2020;369:m1985. doi: 10.1136/bmj.m1985 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Huang C, Huang L, Wang Y, Li X, Ren L, Gu X, et al. 6-month consequences of COVID-19 in patients discharged from hospital: a cohort study. Lancet. 2021;397(10270):220–32. doi: 10.1016/S0140-6736(20)32656-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Taquet M, Geddes JR, Husain M, Luciano S, Harrison PJ. 6-month neurological and psychiatric outcomes in 236 379 survivors of COVID-19: a retrospective cohort study using electronic health records. The Lancet Psychiatry. doi: 10.1016/S2215-0366(21)00084-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Rogers JP, Chesney E, Oliver D, Pollak TA, McGuire P, Fusar-Poli P, et al. Psychiatric and neuropsychiatric presentations associated with severe coronavirus infections: a systematic review and meta-analysis with comparison to the COVID-19 pandemic. Lancet Psychiatry. 2020;7(7):611–27. doi: 10.1016/S2215-0366(20)30203-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Ellul MA, Benjamin L, Singh B, Lant S, Michael BD, Easton A, et al. Neurological associations of COVID-19. Lancet Neurol. 2020;19(9):767–83. doi: 10.1016/S1474-4422(20)30221-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Varatharaj A, Thomas N, Ellul MA, Davies NWS, Pollak TA, Tenorio EL, et al. Neurological and neuropsychiatric complications of COVID-19 in 153 patients: a UK-wide surveillance study. Lancet Psychiatry. 2020;7(10):875–82. doi: 10.1016/S2215-0366(20)30287-X [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Hernandez-Fernandez F, Sandoval Valencia H, Barbella-Aponte RA, Collado-Jimenez R, Ayo-Martin O, Barrena C, et al. Cerebrovascular disease in patients with COVID-19: neuroimaging, histological and clinical description. Brain. 2020;143(10):3089–103. doi: 10.1093/brain/awaa239 [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Zhu Y, Zhang J, Li Y, Liu F, Zhou Q, Peng Z. Association between thrombocytopenia and 180-day prognosis of COVID-19 patients in intensive care units: A two-center observational study. PLoS One. 2021;16(3):e0248671. doi: 10.1371/journal.pone.0248671 [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Biancari F, Mariscalco G, Dalén M, Settembre N, Welp H, Perrotti A, et al. Six-Month Survival After Extracorporeal Membrane Oxygenation for Severe COVID-19. J Cardiothorac Vasc Anesth. 2021. doi: 10.1053/j.jvca.2021.01.027 [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Garcia-Vidal C, Cózar-Llistó A, Meira F, Dueñas G, Puerta-Alcalde P, Cilloniz C, et al. Trends in mortality of hospitalised COVID-19 patients: A single centre observational cohort study from Spain. The Lancet Regional Health—Europe. 2021;3. doi: 10.1016/j.lanepe.2021.100041 [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Brieghel C, Ellekvist P, Lund ML, Soborg C, Walsted ES, Thomsen JJ, et al. Prognostic factors of 90-day mortality in patients hospitalised with COVID-19. Dan Med J. 2021;68(3). [PubMed] [Google Scholar]
15.Zettersten E, Engerstrom L, Bell M, Jaderling G, Martensson J, Block L, et al. Long-term outcome after intensive care for COVID-19: differences between men and women-a nationwide cohort study. Crit Care. 2021;25(1):86. doi: 10.1186/s13054-021-03511-x [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Network C-IGobotR, the C-ICUI. Clinical characteristics and day-90 outcomes of 4244 critically ill adults with COVID-19: a prospective cohort study. Intensive Care Med. 2021;47(1):60–73. doi: 10.1007/s00134-020-06294-x [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Kuehn BM. Hospital Readmission Is Common Among COVID-19 Survivors. JAMA. 2020;324(24):2477. doi: 10.1001/jama.2020.23910 [DOI] [PubMed] [Google Scholar]
18.Busse R, Blumel M, Knieps F, Barnighausen T. Statutory health insurance in Germany: a health system shaped by 135 years of solidarity, self-governance, and competition. Lancet. 2017;390(10097):882–97. doi: 10.1016/S0140-6736(17)31280-1 [DOI] [PubMed] [Google Scholar]
19.Bundesministerium für Gesundheit. Statistik über Versicherte, gegliedert nach Status, Alter, Wohnort und Kassenart (Stichtag: 1. Juli des jeweiligen Jahres). Bonn: Bundesministerium für Gesundheit; 2021. Available from: https://www.bundesgesundheitsministerium.de/fileadmin/Dateien/3_Downloads/Statistiken/GKV/Mitglieder_Versicherte/Versicherte_2020.xlsx. Last accessed: 26 June 2021.
20.Statistisches Bundesamt. Fortschreibung des Bevölkerungsstandes, 12411–0005: Bevölkerung: Deutschland, Stichtag, Altersjahre. Stichtag 31.12.2020. Wiesbaden: Statistisches Bundesamt; 2021. Available from: https://www.destatis.de/DE/Themen/Gesellschaft-Umwelt/Bevoelkerung/Bevoelkerungsstand/_inhalt.html. Last accessed: 26 June 2021.
21.Quan H, Sundararajan V, Halfon P, Fong A, Burnand B, Luthi JC, et al. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care. 2005;43(11):1130–9. doi: 10.1097/01.mlr.0000182534.19832.83 [DOI] [PubMed] [Google Scholar]
22.Elixhauser A, Steiner C, Harris DR, Coffey RM. Comorbidity measures for use with administrative data. Med Care. 1998;36(1):8–27. doi: 10.1097/00005650-199801000-00004 [DOI] [PubMed] [Google Scholar]
23.Takahashi T, Ellingson MK, Wong P, Israelow B, Lucas C, Klein J, et al. Sex differences in immune responses that underlie COVID-19 disease outcomes. Nature. 2020;588(7837):315–20. doi: 10.1038/s41586-020-2700-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Kim L, Garg S, O’Halloran A, Whitaker M, Pham H, Anderson EJ, et al. Risk Factors for Intensive Care Unit Admission and In-hospital Mortality among Hospitalized Adults Identified through the U.S. Coronavirus Disease 2019 (COVID-19)-Associated Hospitalization Surveillance Network (COVID-NET). Clin Infect Dis. 2020. doi: 10.1093/cid/ciaa1012 [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Qian J, Zhao L, Ye RZ, Li XJ, Liu YL. Age-dependent Gender Differences in COVID-19 in Mainland China: Comparative Study. Clin Infect Dis. 2020;71(9):2488–94. doi: 10.1093/cid/ciaa683 [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Yang P, Esper AM. Investigating the Sex Differences in COVID-19: Another Step Forward, But Many Unanswered Questions. Clin Infect Dis. 2020;71(9):2495–6. doi: 10.1093/cid/ciaa776 [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Pivonello R, Auriemma RS, Pivonello C, Isidori AM, Corona G, Colao A, et al. Sex disparities in COVID-19 severity and outcome: are men weaker or women stronger? Neuroendocrinology. 2020:1–20. doi: 10.1159/000513346 [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Frohlich GM, Jeschke E, Eichler U, Thiele H, Alhariri L, Reinthaler M, et al. Impact of oral anticoagulation on clinical outcomes of COVID-19: a nationwide cohort study of hospitalized patients in Germany. Clin Res Cardiol. 2021. doi: 10.1007/s00392-020-01783-x [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Iba T, Levy JH, Levi M, Thachil J. Coagulopathy in COVID-19. J Thromb Haemost. 2020;18(9):2103–9. doi: 10.1111/jth.14975 [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Ackermann M, Verleden SE, Kuehnel M, Haverich A, Welte T, Laenger F, et al. Pulmonary Vascular Endothelialitis, Thrombosis, and Angiogenesis in Covid-19. N Engl J Med. 2020;383(2):120–8. doi: 10.1056/NEJMoa2015432 [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Fernando SM, McIsaac DI, Rochwerg B, Bagshaw SM, Muscedere J, Munshi L, et al. Frailty and invasive mechanical ventilation: association with outcomes, extubation failure, and tracheostomy. Intensive Care Med. 2019;45(12):1742–52. doi: 10.1007/s00134-019-05795-8 [DOI] [PubMed] [Google Scholar]
32.Hewitt J, Carter B, Vilches-Moraga A, Quinn TJ, Braude P, Verduri A, et al. The effect of frailty on survival in patients with COVID-19 (COPE): a multicentre, European, observational cohort study. Lancet Public Health. 2020;5(8):e444–e51. doi: 10.1016/S2468-2667(20)30146-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Darvall JN, Bellomo R, Young PJ, Rockwood K, Pilcher D. Frailty and mortality in patients with COVID-19. Lancet Public Health. 2020;5(11):e580. doi: 10.1016/S2468-2667(20)30228-0 [DOI] [PMC free article] [PubMed] [Google Scholar]

PLoS One. doi: 10.1371/journal.pone.0255427.r001

Decision Letter 0

Aleksandar R Zivkovic

15 Jun 2021

PONE-D-21-15090

6-month follow up of hospitalised COVID-19 patients: A nationwide cohort study of 8,679 patients in Germany

PLOS ONE

Dear Dr. Karagiannidis,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Jul 30 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Aleksandar R. Zivkovic

Academic Editor

PLOS ONE

Journal requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. Please see http://www.bmj.com/content/340/bmj.c181.long for guidelines on how to de-identify and prepare clinical data for publication. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide

3. Thank you for stating the following in the Competing Interests section:

"RB reports grants from Berlin University Alliance, during the conduct of the study;

grants from Federal Ministry of Research and Education, grants from Federal Ministry

of Health, grants from Innovation Fonds of the Federal Joint Committee, grants from

World Health Organization, outside the submitted work, AS reports grants from Bayer

AG, outside the submitted work. CK reports personal fees from Maquet, personal fees

from Xenios, personal fees from Bayer, non-financial support from Speaker of the

German register of ICUs, grants from German Ministry of Research and Education,

during the conduct of the study. CG, MS, TR, GS, WH, and SWC have nothing to

disclose."

Please confirm that this does not alter your adherence to all PLOS ONE policies on sharing data and materials, by including the following statement: "This does not alter our adherence to PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests). If there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include your updated Competing Interests statement in your cover letter; we will change the online submission form on your behalf.

Please know it is PLOS ONE policy for corresponding authors to declare, on behalf of all authors, all potential competing interests for the purposes of transparency. PLOS defines a competing interest as anything that interferes with, or could reasonably be perceived as interfering with, the full and objective presentation, peer review, editorial decision-making, or publication of research or non-research articles submitted to one of the journals. Competing interests can be financial or non-financial, professional, or personal. Competing interests can arise in relationship to an organization or another person. Please follow this link to our website for more details on competing interests: http://journals.plos.org/plosone/s/competing-interests

4. Please include your tables as part of your main manuscript and remove the individual files. Please note that supplementary tables (should remain/ be uploaded) as separate "supporting information" files

5. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Reviewer #1: I assessed the manuscript by Günster and co-workers which aimed to assess factors associated to 6 months outcomes after discharged in COVID-19 hospitalized patients starting from the dataset of German Local Health Fundus. The topic of medium and long term outcomes of hospitalized COVID-19 patients is for sure of interest considering the magnitude of the pandemic with a high number of patients requiring intensive care with potential long term detrimental effect on health status. The manuscript is not easy to read and does not follow the STROBE checklist for observational study with a very poor method section. The main problem of the manuscript is a study design and outcome definition/selection which are unable to answer the research question posed by the authors.

Major comments:

- The major outcome of interest is death. Nevertheless, this is not stated by the authors nor in the abstract nor at the end of the introduction section.

- It is surprising to read “unselected and unbiased cohort” at the end of the introduction section. In particular, it is clear that the major problem of the study is a selection bias introduced by the authors with the exclusion into the analysis of 2,780 of patients who “does not suffer from a COVID-19 principal diagnosis”

- The dataset is insufficient to answer the research question. In particular, the authors were unable to distinguish between COVID-19 related and unrelated deaths.

- The definition of the effect modified entered into the model is poor. Several factors seem to be pre-hospital determinants of COVID-19 worse outcomes (i.e. age, gender, BMI) whereas other (i.e. mechanical ventilation, shock, etc) are events which happened during the first index events. Nevertheless, no mention across the manuscript was made about how time-to-events was handled into the final mode.

- It is surprising to see different median time of hospital stay in Table 1 for the “index hospital admission” at different time point (30-90-180 days). How is it possible if the baseline study population is the same?

Reviewer #2: The study includes a large amount of COVID-19 patients admitted to the hospital in Germany and analyses risk factors for in-hospital mortality and post-discharge mortality, as well as for readmission. This study addresses an important and current issue which goes beyond acute disease. Specifically, it shows a relevant impact of COVID-19 even after acute disease resolution and points to the importance of post-discharge patient monitoring to precede long-term mortality and readmission rates.

The large sample size is certainly an important advantage of this work. However, the limited length of stay needed for patient inclusion (February-April 2020) may create selection bias by excluding those patients with longer length of stay, who may be those with more severe disease or transferred to the intensive care unit.

There are few major issues to address:

- In the manuscript there a lot of English errors. Therefore, I suggest the manuscript to be thoroughly reviewed by a native English-speaker before resubmission.

- It is unclear why Cox regression was not used to build the predictive models and logistic regression was used instead. I would suggest to implement the methods section to better specify this. Time is an important factor to consider and a time-dependent analysis would be preferred, although logistic regression may still be fine since several time points are considered.

- It would be interesting to test the model in an external cohort. Alternatively, a method of internal validation should be considered to confirm the results.

- In the depicted Kaplan Meier curves in figure 2.

- In figure 3, it is unclear which are the variables which covariates are adjusted for.

Reviewer #3: In this nationwide observational study, the authors provide a detailed account of the follow-up of hospitalized COVID-19 patients until 6-months after their initial hospital admission. There is increasing information now in the literature about the outcome of hospitalized COVID-19 patients after discharge. However, the novelty of the present study rests on the large cohort considered with 6-month follow-up. Indeed, with the exception of China, so far published data on large cohorts of hospitalized COVID-19 patients limited their observation up to 3 months.

There are few areas, which are of concern for the authors considerations:

1. Here the focus is on mortality during 6-month follow-up, including death occurring during possible readmission to hospital after discharge. However, they did not analyze others outcomes such as long-term complications and incomplete recovery after hospital discharge. Therefore, the title of the manuscript should be more specific, including the reference to the parameter ‘mortality’ they have analyzed.

2. Data were from the German local health care funds, a health insurance system that approximately cover 32% of the German population. Although briefly mentioned in the limitations of the study, it would be useful to provide more details on the representativeness of the study cohort for all the German population, given the term ‘nationwide’ attributed to the study in the title, and before concluding for the generalization of the findings.

3. It would have been interesting to analyze separately the group of patients who were hospitalized for COVID-19 and that of COVID-19 patients hospitalized for other reasons. The authors, however, cannot distinguish these two groups. Therefore, this issue should be discussed in the limitations’ paragraph.

4. As shown in figure 2d, the 30-day mortality for patients with coagulopathy was lower than that of patients without coagulopathy. How the authors may explain this unexpected finding?

5. In the Discussion, they should elaborate more on the reason why female sex is a protective factor in terms of mortality in COVID-19 disease, as documented in the present study.

6. In the Discussion section, they need to highlight the strength and novelty of the study, as well as to elaborate more on the limitations as indicated above.

Minor

1. In the first paragraph of the Introduction, the authors reported the rates of hospitalization and mortality in France, Spain, UK and Germany. In support, they quoted reference #4 and #5 among others, which however refer to Italy. This discrepancy should be fixed.

2. For each panel of Figure 1 and 2, the Kaplan-Meier survival curves should include index number of patients at each time points.

Reviewer #4: The authors present a retrospective observational study in Germany on 6-months mortality rate and outcomes of patients with hospitalised Covid-19. The data is clearly presented.

Although retrospective, the study includes a high number of patients, representative of the german population and whose characteristics (age, male predominance) are in accordance with other published series on Covd-19 epidemiology. The authors find that 6 months mortality is high, higher in men than women.

One limit of the study is that it included patients during the first epidemic wave, when mortality rate was probably higher than now, as stated by several studies, especially for patients with coagulopathy. Still, this study remains of interest as it describes the course of severe Covid-19 and since mortality remains quite high in hospitalised patients and as it includes a representative population. The evolution of mortality over time could be better emphasised in the discussion section.

One might question the high readmission rate, especially for neurological and respiratory conditions, since this does not seem to be the case in all countries : authors should discuss this point in light of discharge conditions in Germany (e.g. are patients discharged home or do they benefit from in-hospital readmission with a transfer to another hospital ?) and in light of their personnal experience of the causes of readmission (what stands under neurological and respiratory conditions ?)

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

Reviewer #4: Yes: Justine Frija-Masson

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2021 Aug 5;16(8):e0255427. doi: 10.1371/journal.pone.0255427.r002

Author response to Decision Letter 0

13 Jul 2021

PLOS one

Editor in Chief

Prof. Zivkovic

Editor’s comments:

Journal requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

AUTHORS RESPONSE: Thank you for the opportunity to revise and resubmit our manuscript! We have formatted our manuscript according to the style requirements and renamed our files accordingly.

CHANGES TO THE MANUSCRIPT: To enhance the readability of our marked manuscript we did not track formatting changes. No formatting change had an influence on the written content.

In your revised cover letter, please address the following prompts:

We will update your Data Availability statement on your behalf to reflect the information you provide

AUTHORS RESPONSE: Unfortunately, there are legal restrictions on sharing the de-identified data set. We addressed this issue in our revised cover letter:

The data used in this study cannot be made available in the manuscript, the supplemental files, or in a public repository due to German data protection laws (Bundesdatenschutzgesetz). Therefore, they are stored on a secure drive in the Wissenschaftliches Institut der AOK to facilitate replication of the results. Generally, access to data of statutory health insurance funds for research purposes is possible only under the conditions defined in German Social Law (SGB V § 287). Requests for data access can be sent as a formal proposal specifying the recipient and purpose of the data transfer to the appropriate data protection agency. Access to the data used in this study can only be provided to external parties under the conditions of the cooperation contract of this research project and after written approval by the AOK. For assistance in obtaining access to the data, please contact wido@wido.bv.aok.de.

CHANGES TO THE MANUSCRIPT: We have added a data availability statement containing the information provided above (ll. X-xx).

3. Thank you for stating the following in the Competing Interests section:

"RB reports grants from Berlin University Alliance, during the conduct of the study; grants from Federal Ministry of Research and Education, grants from Federal Ministry of Health, grants from Innovation Fonds of the Federal Joint Committee, grants from World Health Organization, outside the submitted work, AS reports grants from Bayer AG, outside the submitted work. CK reports personal fees from Maquet, personal fees from Xenios, personal fees from Bayer, non-financial support from Speaker of the German register of ICUs, grants from German Ministry of Research and Education, during the conduct of the study. CG, MS, TR, GS, WH, and SWC have nothing to disclose."

AUTHORS RESPONSE: Thank you for pointing this out. We have added the suggested statement in the revised cover letter and manuscript. We also added a data availability statement (see our response to your previous comment).

CHANGES TO THE MANUSCRIPT: Added “This does not alter our adherence to PLOS ONE policies on sharing data and materials.” (ll. X-xx).

Please include your updated Competing Interests statement in your cover letter; we will change the online submission form on your behalf.

AUTHORS RESPONSE: Thank you. We confirm that we have understood the policy. We have included our updated Competing Interests statement.

CHANGES TO THE MANUSCRIPT: Changed according to the suggestions.

AUTHORS RESPONSE: Thank you, we followed your comment.

CHANGES TO THE MANUSCRIPT: Tables 1 and 2 are now included in the manuscript. Table S1 and S2 have been uploaded as a separate file.

AUTHORS RESPONSE: We have added captions for our Supporting Information files at the end of our manuscript and have updated in-text citations accordingly.

CHANGES TO THE MANUSCRIPT: Added “Supporting Information: Table S1: Diagnosis codes of included patients. Table S2: Patient characteristics stratified by ventilation status.” (ll. X-xx)

Reviewers' comments:

Reviewer's Responses to Questions

AUTHORS RESPONSE: Thank you very much for reviewing our manuscript! We have re-organized our manuscript to follow the STROBE checklist more closely. We particularly restructured and supplemented the methods section.

CHANGES TO THE MANUSCRIPT: We inserted additional subject headings throughout the manuscript, rearranged the methods section, and added the following sentences:

ll. X-x: “This was a retrospective observational cohort study using claims data.”

ll. X-x: “Bias: Thorough legal requirements and verification from both hospitals and sickness funds aim to minimize bias introduced by false coding and other systematic errors relating to the data. In addition, we checked data plausibility and consistency before analyzing the data.”

ll. x-x: “Study size: This was a complete survey of all AOK insured.”

ll. x-x: “Due to non-parallel Kaplan-Meier curves and a significant deviation from the proportional hazards assumption, no Cox-proportional hazard model could be estimated. The proportional hazards assumption was evaluated visually on plots of log(-log(survival)) versus log of survival time adjusted for covariates and by a global test based on Schoenfeld residuals. Instead, […].”

ll. x-x: “We used cluster-robust standard errors in order to account for clustering of patients in hospitals. Comorbidity conditions were defined as binary variables.”

ll. x-x: “[…] and summarized in a forest plot. To evaluate the performance of the model, the area under the curve (AUC) was used as measure of discrimination and the squared Pearson correlation (R2) between 180-day all-cause mortality and the log-odds of predicted mortality was used as measure of explained variation.”

Major comments:

- The major outcome of interest is death. Nevertheless, this is not stated by the authors nor in the abstract nor at the end of the introduction section.

AUTHORS RESPONSE: Thank you for pointing this out! We have now added this information in the abstract and end of introduction section. Furthermore, we changed the title to “6-month mortality and readmissions of hospitalised COVID-19 patients: a nationwide cohort study of 8,679 patients in Germany” to better reflect the focus of our study.

CHANGES TO THE MANUSCRIPT:

ll. 4-6 (Title): “6-month mortality and readmissions of hospitalised COVID-19 patients: a nationwide cohort study of 8,679 patients in Germany”

ll. X-x (Abstract, Methods): “…., for whom 6-month all-cause mortality and readmission rates for 180 days after admission or until death was available.”

ll. Xx-x (Introduction, Objectives): “[…], the aim of this observational study was to determine 6-month all-cause mortality and readmission rates of hospitalised COVID-19 patients with completed hospital treatments and a confirmed COVID-19 diagnosis, […] Furthermore factors associated with 6-month all-cause mortality were evaluated.”

AUTHORS RESPONSE: We agree with the reviewer’s point and have revised the objectives statement accordingly. Furthermore, we now explain and justify our selection of patients in the methods section and discuss the issue in the limitations section of our revised manuscript.

CHANGES TO THE MANUSCRIPT:

Deleted: “in a large, unselected and unbiased cohort of patients with”

Added:

ll. Xx-x (Methods, Participants): “The selection of patients was performed to include only patients in whom COVID-19 was the primary reason for their hospital stay and to exclude those in whom COVID-19 was an incidental finding likely to be unrelated to their hospital stay.”

ll. x-x (Discussion, Strength and limitations): “Lastly, the inclusion of patients admitted for COVID-19-related principal diagnoses might not be sufficient to distinguish between patients who were hospitalised for COVID-19 and those COVID-19 patients hospitalised for other reasons.”

- The dataset is insufficient to answer the research question. In particular, the authors were unable to distinguish between COVID-19 related and unrelated deaths.

AUTHORS RESPONSE: While we agree with the reviewer that it is generally very important distinguish between COVID-19-related and -unrelated deaths, we would like to stress that it was not our aim to assess the patients’ cause of death but to assess all-cause mortality in patients hospitalised for COVID-19. As already stated in the limitations section, our study was purely observational and does not allow for establishing causality. We have added a subclause to clarify this issue. Furthermore, we now consistently use the term “all-cause mortality” throughout the manuscript.

CHANGES TO THE MANUSCRIPT: Added: Ll. Xx.x: “[…] this study cannot determine causality between risk factors and long-term mortality, nor does it provide information on the patients’ cause of death.”

AUTHORS RESPONSE: Thank you for your thorough review, which has called our attention to interconnected weak spots in our presentation of methods and results. Patient characteristics and strata were solely defined by conditions and interventions during index hospitalization. Thereafter comorbidities at index hospitalization were entered into the multivariable modelling of 6-month all-cause mortality. Apart from the Kaplan-Meier analysis, we did use multivariable modelling for the binary endpoint of 180-day all-cause mortality after admission. Cox regression could not be applied to handle time-to-event more detailed because of a major deviation from the proportional hazards assumption. Therefore, we estimated a model by logistic regression. We are grateful for having had this pointed out and have sought to improve both sections in terms of clarity.

CHANGES TO THE MANUSCRIPT: We rearranged the methods section. We added

ll. Xx.x “Baseline characteristics are defined on conditions and interventions during index hospitalization […].”

We updated ll x-x: “Instead, multivariable logistic regression was used to model the odds of the binary endpoint 180-day all-cause mortality after admission as function of age, sex, body mass index (BMI) categories (30≤34, 35≤39, ≥40 kg/m2) and Elixhauser comorbidities present at index hospitalization.”

AUTHORS RESPONSE: Thank you for pointing this out! The values refer to the respective column total. We have added the missing row descriptions.

CHANGES TO THE MANUSCRIPT: Table 1: “Age, years: (value refers to column total)”, “Length of stay of index hospitalisation: (value refers to column total)"

AUTHORS RESPONSE: In our study, we included patients admitted to hospital between February and April 2020 with a length of stay not exceeding 30 June 2020. Accordingly, we were able to include patients with a hospital stay up to 2 months (when enrolled 30 April) to up to 5 months (when enrolled 1 February 2020).

The reason for the cut-off date being 30 June 2020 is that our analyses were performed in January 2021. Thus, we were able to include data up to 31 December 2020, which meant that – for a complete 6-month follow-up after discharge – we could include only patients discharged until 30 June 2020.

CHANGES TO THE MANUSCRIPT: Not applicable.

There are few major issues to address:

- In the manuscript there a lot of English errors. Therefore, I suggest the manuscript to be thoroughly reviewed by a native English-speaker before resubmission.

AUTHORS RESPONSE: Thank you for your suggestion. Our manuscript has now been reviewed by a professional company to correct any mistakes and improve its readability. Furthermore, the manuscript was adapted to American English.

CHANGES TO THE MANUSCRIPT: There have been various corrections/changes to improve language throughout the manuscript (all of which are tracked in the marked version of the manuscript and none of which had an influence on the written content).

Added: ll. X-x (Acknowledgements): The authors would like to thank NN for proof-reading and improving the language of our manuscript.

AUTHORS RESPONSE: We thank you for pointing this out. We would have preferred to conduct a Cox regression to handle time-to-event information more detailed. But unfortunately, Cox regression could not be applied due to a major deviation from the proportional hazards assumption. Therefore, we estimated a model for the binary endpoint of 180-day all-cause mortality after admission by logistic regression. Models for the other time points (in hospital-/30-day/90-day- all-cause mortality) were estimated too and showed similar results with more pronounced impact of metastatic cancer for longer follow up (not shown).

CHANGES TO THE MANUSCRIPT: We added ll. x-x: “Due to non-parallel Kaplan-Meier curves and a significant deviation from the proportional hazards assumption, no Cox-proportional hazard model could be estimated. The proportional hazards assumption was evaluated visually on plots of log(-log(survival)) versus log of survival time adjusted for covariates and by a global test based on Schoenfeld residuals.”

- It would be interesting to test the model in an external cohort. Alternatively, a method of internal validation should be considered to confirm the results.

- In the depicted Kaplan Meier curves in figure 2.

AUTHORS RESPONSE: Thank you for this interesting point. We wish to clarify that it was not our aim to develop a predictive model, but a multivariable logistic regression model of independent risk factors for 180-day all-cause mortality. We further specified the intent for and methods applying to our model in the abstract and methods section.

However, we are indeed planning compare this cohort to other cohorts, namely patients hospitalised during the “second wave” and “third wave”. It will be very interesting to see if the same independent risk factors for 180-day all-cause mortality will apply to these cohorts. Unfortunately, we are not able to perform these analyses as of now given the time frame until the respective data becomes available (see our response to your first comment).

CHANGES TO THE MANUSCRIPT:

ll. x-x (Abstract): “A multivariable logistic regression model identified independent risk factors for 180-day all-cause mortality in this cohort.”

- In figure 3, it is unclear which are the variables which covariates are adjusted for.

AUTHORS RESPONSE: Thank you for pointing us to this ambiguity. Each covariate has been adjusted for all other covariates displayed in figure 3. We have added this information in the figure legend.

CHANGES TO THE MANUSCRIPT: Added: Figure legend Fig 3: “Each covariate has been adjusted for all other covariates displayed.”

There are few areas, which are of concern for the authors considerations:

AUTHORS RESPONSE: Thank you very much for reviewing our manuscript! We agree with the reviewer and have changed the title accordingly.

CHANGES TO THE MANUSCRIPT: Title: “6-month mortality and readmissions of hospitalised COVID-19 patients: a nationwide cohort study of 8,679 patients in Germany”

AUTHORS RESPONSE: We followed the reviewer’s suggestion and have added more details on the German statutory health insurance system and the representativeness of the study cohort in the methods section.

CHANGES TO THE MANUSCRIPT:

ll. Xx-x: “AOK is the largest sickness fund group within Germany’s statutory health insurance system. It provides statutory health insurance for roughly 32 percent of the German population, for whom it is representative when considering the factor age [20, 21]. Furthermore, membership in a sickness fund is open to anyone regardless of factors such as professional affiliation, income, or comorbidities. At the same time, sickness funds are obliged to accept any new member and charge the same basic contribution rate [19].”

AUTHORS RESPONSE: Thank you for your valuable suggestion. We agree that this kind of analysis would have been very interesting and is certainly a point that future research should address. We now discuss the issue in the limitations section of our manuscript.

CHANGES TO THE MANUSCRIPT:

Added: ll. Xx-x: “Lastly, the inclusion of patients admitted for COVID-19-related principal diagnoses might not be sufficient to distinguish between patients who were hospitalized for COVID-19 and patients with COVID-19 who were hospitalized for other reasons. Of note, 6-months all-cause mortality was similar between the in- and excluded patients.”

4. As shown in figure 2d, the 30-day mortality for patients with coagulopathy was lower than that of patients without coagulopathy. How the authors may explain this unexpected finding?

AUTHORS RESPONSE: Thank you for raising this interesting question. We also wondered about the survival curves. However, our data don’t show a sufficient explanation. It might be that patient with recognized coagulopathy such as pulmonary embolism are under closer supervision within the first days of treatment. Therefore, initial survival might be somewhat better. Since this remains speculative, we didn’t integrate this point into the manuscript.

CHANGES TO THE MANUSCRIPT: Not applicable.

5. In the Discussion, they should elaborate more on the reason why female sex is a protective factor in terms of mortality in COVID-19 disease, as documented in the present study.

AUTHORS RESPONSE: We agree with the reviewer and have added another sentence and reference to discuss this important matter.

CHANGES TO THE MANUSCRIPT:

ll. X-x (Discussion): “Overall, women show a better long-term outcome than men regardless of other confounding factors in terms of all-cause mortality, which may be due to their enhanced immune and inflammatory response to COVID-19 compared to men [24-27]. This is likely caused by their different genetic and endocrine mechanisms, including sex hormone actions, which might also influence the mechanisms of coagulopathy and thrombosis in COVID-19 [28].”

6. In the Discussion section, they need to highlight the strength and novelty of the study, as well as to elaborate more on the limitations as indicated above.

AUTHORS RESPONSE: Thank you! We re-named the limitations section to “strengths and limitations” and have elaborated on both while considering your previous points.

CHANGES TO THE MANUSCRIPT:

Added: ll.x-x (Discussion, Strengths and limitations): “The major strength of our study is its size and length of follow-up, our cohort of being the largest cohort of patients hospitalized for COVID-19 that was followed up for six months after admission and discharge, respectively. The data source (claims data) also has several strengths. First, due to the administrative nature of the data, we describe an almost complete cohort of 98% of the eligible patients with full follow up data. Furthermore, information on readmission is independent of whether readmission was handled by the same or a different hospital. Second, since choice of hospital is free under German statutory health insurance, our data set includes hospitals ranging from major tertiary referral centers to smaller regional hospitals, which affords real-world data unbiased by the degree of hospital specialization. Third, inpatient data is of high quality because disease and procedure codes are relevant for the amount of remuneration and are therefore verified by hospitals and sickness funds. Lastly, unlike most studies using administrative data, we were able to assign data from different hospital stays to the individual patient so that the unit of analysis was the patient not the hospital case.

There are also several limitations relating to the data source. […] General limitations include that, given its observational nature, this study cannot determine causality between risk factors and long-term mortality, nor does it provide information on the patients’ cause of death. Lastly, the inclusion of patients admitted for COVID-19-related principal diagnoses might not be sufficient to distinguish between patients who were hospitalized for COVID-19 and patients with COVID-19 who were hospitalized for other reasons. Of note, 6-months all-cause mortality was similar between the in- and excluded patients.”

Minor

AUTHORS RESPONSE: Thank you very much for pointing this out! We have corrected the respective references.

CHANGES TO THE MANUSCRIPT:

ll. x-x (References):

“1. Department Health Care Management. Data on Covid-19 hospitalisations / ICU treatments across European countries with data available as of 17/03/21 Berlin: Technische Universität Berlin; 2021. Available from: https://www.mig.tu-berlin.de/fileadmin/a38331600/sonstiges/COVID-19-STATS_170321_1945.pdf. Last accessed: 27 June 2021.

2. Bonnet G, Weizman O, Trimaille A, Pommier T, Cellier J, Geneste L, et al. Characteristics and outcomes of patients hospitalized for COVID-19 in France: The Critical COVID-19 France (CCF) study. Arch Cardiovasc Dis. 2021. doi: 10.1016/j.acvd.2021.01.003.

3. Karagiannidis C, Mostert C, Hentschker C, Voshaar T, Malzahn J, Schillinger G, et al. Case characteristics, resource use, and outcomes of 10 021 patients with COVID-19 admitted to 920 German hospitals: an observational study. Lancet Respir Med. 2020;8(9):853-62. doi: 10.1016/S2213-2600(20)30316-7.

4. Docherty AB, Harrison EM, Green CA, Hardwick HE, Pius R, Norman L, et al. Features of 20 133 UK patients in hospital with covid-19 using the ISARIC WHO Clinical Characterisation Protocol: prospective observational cohort study. Bmj. 2020;369:m1985. doi: 10.1136/bmj.m1985.”

2. For each panel of Figure 1 and 2, the Kaplan-Meier survival curves should include index number of patients at each time points.

AUTHORS RESPONSE: Thank you for your suggestion. We have now added the number of patients at start, 90 and 180 days follow-up.

CHANGES TO THE MANUSCRIPT: Figure 1 and 2 (now 2 and 3): Changed accordingly.

Reviewer #4: The authors present a retrospective observational study in Germany on 6-months mortality rate and outcomes of patients with hospitalised Covid-19. The data is clearly presented.

AUTHORS RESPONSE: Thank you very much for reviewing our manuscript!

The reviewer raised an important point. We are indeed planning to compare this cohort to patients hospitalised during the “second wave” and “third wave”. However, the follow-up period of our study being 6 months, we are not able to analyze patients from the second wave (which in Germany lasted until mid Feburary 2021) before autumn. Therefore, we chose not to discuss this issue in our current manuscript.

However, we agree with the reviewer that it will be very interesting to see if mortality rates did in fact become lower in these subsequent waves. On the one hand, there is evidence that in-hospital mortality rates became lower. On the other hand, the later pandemic waves brought new variants of the virus with unknown long-term consequences. Furthermore, we found that there was no difference in in-hospital mortality rates between the first and second wave of ICU patients in Germany who we analysed as part of another study (Karagiannidis C, Windisch W, McAuley DF, Welte T, Busse R. Major differences in ICU admissions during the first and second COVID-19 wave in Germany. Lancet Respir Med. 2021;9(5):e47-e48. doi:10.1016/S2213-2600(21)00101-6).

CHANGES TO THE MANUSCRIPT: Not applicable.

One might question the high readmission rate, especially for neurological and respiratory conditions, since this does not seem to be the case in all countries: authors should discuss this point in light of discharge conditions in Germany (e.g. are patients discharged home or do they benefit from in-hospital readmission with a transfer to another hospital?) and in light of their personal experience of the causes of readmission (what stands under neurological and respiratory conditions ?)

AUTHORS RESPONSE: Thank you for this important comment. Compared to different health care systems, the German readmission rate is almost in average with other systems and with other diseases within our health care system with a broad variation. Since this is an everyday experience and we have only few data, we cannot add this to the manuscript, but we now added a new table with readmission reasons to the supplement.

CHANGES TO THE MANUSCRIPT: We now added a supplement table with definitions of all readmission groups.

Attachment

Submitted filename: Response to reviewers_R1 final.docx

Click here for additional data file.^{(38.1KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0255427.r003

Decision Letter 1

Aleksandar R Zivkovic

16 Jul 2021

6-month mortality and readmissions of hospitalized COVID-19 patients: a nationwide cohort study of 8,679 patients in Germany

PONE-D-21-15090R1

Dear Dr. Karagiannidis,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Aleksandar R. Zivkovic

Academic Editor

PLOS ONE

PLoS One. doi: 10.1371/journal.pone.0255427.r004

Acceptance letter

Aleksandar R Zivkovic

22 Jul 2021

PONE-D-21-15090R1

6-month mortality and readmissions of hospitalized COVID-19 patients: a nationwide cohort study of 8,679 patients in Germany

Dear Dr. Karagiannidis:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Aleksandar R. Zivkovic

Academic Editor

PLOS ONE

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1 Table. Diagnosis codes of included patients.

(DOCX)

Click here for additional data file.^{(16.5KB, docx)}

S2 Table. Patient characteristics stratified by ventilation status.

(DOCX)

Click here for additional data file.^{(25KB, docx)}

Attachment

Submitted filename: Response to reviewers_R1 final.docx

Click here for additional data file.^{(38.1KB, docx)}

Data Availability Statement

[pone.0255427.ref001] 1.Department Health Care Management. Data on Covid-19 hospitalisations /ICU treatments across European countries with data available as of 17/03/21 Berlin: Technische Universität Berlin; 2021. Available from: https://www.mig.tu-berlin.de/fileadmin/a38331600/sonstiges/COVID-19-STATS_170321_1945.pdf. Last accessed: 27 June 2021.

[pone.0255427.ref002] 2.Bonnet G, Weizman O, Trimaille A, Pommier T, Cellier J, Geneste L, et al. Characteristics and outcomes of patients hospitalized for COVID-19 in France: The Critical COVID-19 France (CCF) study. Arch Cardiovasc Dis. 2021. doi: 10.1016/j.acvd.2021.01.003 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255427.ref003] 3.Karagiannidis C, Mostert C, Hentschker C, Voshaar T, Malzahn J, Schillinger G, et al. Case characteristics, resource use, and outcomes of 10 021 patients with COVID-19 admitted to 920 German hospitals: an observational study. Lancet Respir Med. 2020;8(9):853–62. doi: 10.1016/S2213-2600(20)30316-7 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255427.ref004] 4.Docherty AB, Harrison EM, Green CA, Hardwick HE, Pius R, Norman L, et al. Features of 20 133 UK patients in hospital with covid-19 using the ISARIC WHO Clinical Characterisation Protocol: prospective observational cohort study. Bmj. 2020;369:m1985. doi: 10.1136/bmj.m1985 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255427.ref005] 5.Huang C, Huang L, Wang Y, Li X, Ren L, Gu X, et al. 6-month consequences of COVID-19 in patients discharged from hospital: a cohort study. Lancet. 2021;397(10270):220–32. doi: 10.1016/S0140-6736(20)32656-8 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255427.ref006] 6.Taquet M, Geddes JR, Husain M, Luciano S, Harrison PJ. 6-month neurological and psychiatric outcomes in 236 379 survivors of COVID-19: a retrospective cohort study using electronic health records. The Lancet Psychiatry. doi: 10.1016/S2215-0366(21)00084-5 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255427.ref007] 7.Rogers JP, Chesney E, Oliver D, Pollak TA, McGuire P, Fusar-Poli P, et al. Psychiatric and neuropsychiatric presentations associated with severe coronavirus infections: a systematic review and meta-analysis with comparison to the COVID-19 pandemic. Lancet Psychiatry. 2020;7(7):611–27. doi: 10.1016/S2215-0366(20)30203-0 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255427.ref008] 8.Ellul MA, Benjamin L, Singh B, Lant S, Michael BD, Easton A, et al. Neurological associations of COVID-19. Lancet Neurol. 2020;19(9):767–83. doi: 10.1016/S1474-4422(20)30221-0 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255427.ref009] 9.Varatharaj A, Thomas N, Ellul MA, Davies NWS, Pollak TA, Tenorio EL, et al. Neurological and neuropsychiatric complications of COVID-19 in 153 patients: a UK-wide surveillance study. Lancet Psychiatry. 2020;7(10):875–82. doi: 10.1016/S2215-0366(20)30287-X [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255427.ref010] 10.Hernandez-Fernandez F, Sandoval Valencia H, Barbella-Aponte RA, Collado-Jimenez R, Ayo-Martin O, Barrena C, et al. Cerebrovascular disease in patients with COVID-19: neuroimaging, histological and clinical description. Brain. 2020;143(10):3089–103. doi: 10.1093/brain/awaa239 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255427.ref011] 11.Zhu Y, Zhang J, Li Y, Liu F, Zhou Q, Peng Z. Association between thrombocytopenia and 180-day prognosis of COVID-19 patients in intensive care units: A two-center observational study. PLoS One. 2021;16(3):e0248671. doi: 10.1371/journal.pone.0248671 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255427.ref012] 12.Biancari F, Mariscalco G, Dalén M, Settembre N, Welp H, Perrotti A, et al. Six-Month Survival After Extracorporeal Membrane Oxygenation for Severe COVID-19. J Cardiothorac Vasc Anesth. 2021. doi: 10.1053/j.jvca.2021.01.027 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255427.ref013] 13.Garcia-Vidal C, Cózar-Llistó A, Meira F, Dueñas G, Puerta-Alcalde P, Cilloniz C, et al. Trends in mortality of hospitalised COVID-19 patients: A single centre observational cohort study from Spain. The Lancet Regional Health—Europe. 2021;3. doi: 10.1016/j.lanepe.2021.100041 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255427.ref014] 14.Brieghel C, Ellekvist P, Lund ML, Soborg C, Walsted ES, Thomsen JJ, et al. Prognostic factors of 90-day mortality in patients hospitalised with COVID-19. Dan Med J. 2021;68(3). [PubMed] [Google Scholar]

[pone.0255427.ref015] 15.Zettersten E, Engerstrom L, Bell M, Jaderling G, Martensson J, Block L, et al. Long-term outcome after intensive care for COVID-19: differences between men and women-a nationwide cohort study. Crit Care. 2021;25(1):86. doi: 10.1186/s13054-021-03511-x [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255427.ref016] 16.Network C-IGobotR, the C-ICUI. Clinical characteristics and day-90 outcomes of 4244 critically ill adults with COVID-19: a prospective cohort study. Intensive Care Med. 2021;47(1):60–73. doi: 10.1007/s00134-020-06294-x [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255427.ref017] 17.Kuehn BM. Hospital Readmission Is Common Among COVID-19 Survivors. JAMA. 2020;324(24):2477. doi: 10.1001/jama.2020.23910 [DOI] [PubMed] [Google Scholar]

[pone.0255427.ref018] 18.Busse R, Blumel M, Knieps F, Barnighausen T. Statutory health insurance in Germany: a health system shaped by 135 years of solidarity, self-governance, and competition. Lancet. 2017;390(10097):882–97. doi: 10.1016/S0140-6736(17)31280-1 [DOI] [PubMed] [Google Scholar]

[pone.0255427.ref019] 19.Bundesministerium für Gesundheit. Statistik über Versicherte, gegliedert nach Status, Alter, Wohnort und Kassenart (Stichtag: 1. Juli des jeweiligen Jahres). Bonn: Bundesministerium für Gesundheit; 2021. Available from: https://www.bundesgesundheitsministerium.de/fileadmin/Dateien/3_Downloads/Statistiken/GKV/Mitglieder_Versicherte/Versicherte_2020.xlsx. Last accessed: 26 June 2021.

[pone.0255427.ref020] 20.Statistisches Bundesamt. Fortschreibung des Bevölkerungsstandes, 12411–0005: Bevölkerung: Deutschland, Stichtag, Altersjahre. Stichtag 31.12.2020. Wiesbaden: Statistisches Bundesamt; 2021. Available from: https://www.destatis.de/DE/Themen/Gesellschaft-Umwelt/Bevoelkerung/Bevoelkerungsstand/_inhalt.html. Last accessed: 26 June 2021.

[pone.0255427.ref021] 21.Quan H, Sundararajan V, Halfon P, Fong A, Burnand B, Luthi JC, et al. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care. 2005;43(11):1130–9. doi: 10.1097/01.mlr.0000182534.19832.83 [DOI] [PubMed] [Google Scholar]

[pone.0255427.ref022] 22.Elixhauser A, Steiner C, Harris DR, Coffey RM. Comorbidity measures for use with administrative data. Med Care. 1998;36(1):8–27. doi: 10.1097/00005650-199801000-00004 [DOI] [PubMed] [Google Scholar]

[pone.0255427.ref023] 23.Takahashi T, Ellingson MK, Wong P, Israelow B, Lucas C, Klein J, et al. Sex differences in immune responses that underlie COVID-19 disease outcomes. Nature. 2020;588(7837):315–20. doi: 10.1038/s41586-020-2700-3 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255427.ref024] 24.Kim L, Garg S, O’Halloran A, Whitaker M, Pham H, Anderson EJ, et al. Risk Factors for Intensive Care Unit Admission and In-hospital Mortality among Hospitalized Adults Identified through the U.S. Coronavirus Disease 2019 (COVID-19)-Associated Hospitalization Surveillance Network (COVID-NET). Clin Infect Dis. 2020. doi: 10.1093/cid/ciaa1012 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255427.ref025] 25.Qian J, Zhao L, Ye RZ, Li XJ, Liu YL. Age-dependent Gender Differences in COVID-19 in Mainland China: Comparative Study. Clin Infect Dis. 2020;71(9):2488–94. doi: 10.1093/cid/ciaa683 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255427.ref026] 26.Yang P, Esper AM. Investigating the Sex Differences in COVID-19: Another Step Forward, But Many Unanswered Questions. Clin Infect Dis. 2020;71(9):2495–6. doi: 10.1093/cid/ciaa776 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255427.ref027] 27.Pivonello R, Auriemma RS, Pivonello C, Isidori AM, Corona G, Colao A, et al. Sex disparities in COVID-19 severity and outcome: are men weaker or women stronger? Neuroendocrinology. 2020:1–20. doi: 10.1159/000513346 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255427.ref028] 28.Frohlich GM, Jeschke E, Eichler U, Thiele H, Alhariri L, Reinthaler M, et al. Impact of oral anticoagulation on clinical outcomes of COVID-19: a nationwide cohort study of hospitalized patients in Germany. Clin Res Cardiol. 2021. doi: 10.1007/s00392-020-01783-x [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255427.ref029] 29.Iba T, Levy JH, Levi M, Thachil J. Coagulopathy in COVID-19. J Thromb Haemost. 2020;18(9):2103–9. doi: 10.1111/jth.14975 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255427.ref030] 30.Ackermann M, Verleden SE, Kuehnel M, Haverich A, Welte T, Laenger F, et al. Pulmonary Vascular Endothelialitis, Thrombosis, and Angiogenesis in Covid-19. N Engl J Med. 2020;383(2):120–8. doi: 10.1056/NEJMoa2015432 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255427.ref031] 31.Fernando SM, McIsaac DI, Rochwerg B, Bagshaw SM, Muscedere J, Munshi L, et al. Frailty and invasive mechanical ventilation: association with outcomes, extubation failure, and tracheostomy. Intensive Care Med. 2019;45(12):1742–52. doi: 10.1007/s00134-019-05795-8 [DOI] [PubMed] [Google Scholar]

[pone.0255427.ref032] 32.Hewitt J, Carter B, Vilches-Moraga A, Quinn TJ, Braude P, Verduri A, et al. The effect of frailty on survival in patients with COVID-19 (COPE): a multicentre, European, observational cohort study. Lancet Public Health. 2020;5(8):e444–e51. doi: 10.1016/S2468-2667(20)30146-8 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255427.ref033] 33.Darvall JN, Bellomo R, Young PJ, Rockwood K, Pilcher D. Frailty and mortality in patients with COVID-19. Lancet Public Health. 2020;5(11):e580. doi: 10.1016/S2468-2667(20)30228-0 [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

6-month mortality and readmissions of hospitalized COVID-19 patients: A nationwide cohort study of 8,679 patients in Germany

Christian Günster

Reinhard Busse

Melissa Spoden

Tanja Rombey

Gerhard Schillinger

Wolfgang Hoffmann

Steffen Weber-Carstens

Andreas Schuppert

Christian Karagiannidis

Roles

Abstract

Background

Aim

Methods

Results

Conclusions

Introduction

Background

Objectives

Materials and methods

Study design

Setting

Study population

Endpoints

Statistical analysis

Results

Study cohort

Fig 1. Patient flow diagram.

Demographics and comorbidity

Table 1. Patient characteristics and all-cause mortality.

Index hospitalization

All-cause mortality

Fig 2. Kaplan-Meier survival curves of all hospitalized COVID-19 patients followed for 180 days after hospital admission.

All-cause mortality among ventilated patients

Fig 3. Kaplan-Meier survival curves of all hospitalized COVID-19 patients on mechanical ventilation followed for 180 days after hospital admission.

Factors associated with all-cause mortality

Fig 4. Multivariable logistic regression analysis for 180-day all-cause mortality after hospital admission.

Readmissions

Table 2. Readmission within 180-days of discharge.

Discussion

Strengths and limitations

Conclusions

Supporting information

Data Availability

Funding Statement

References

Decision Letter 0

Aleksandar R Zivkovic

Roles

Author response to Decision Letter 0

Decision Letter 1

Aleksandar R Zivkovic

Roles

Acceptance letter

Aleksandar R Zivkovic

Roles

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases