key: cord-0846378-rgpyhb05
authors: Hall, M.; Group, ISARIC Clinical Characterisation
title: Ten months of temporal variation in the clinical journey of hospitalised patients with COVID-19: an observational cohort
date: 2021-06-03
journal: nan
DOI: 10.1101/2021.06.01.21258150
sha: 559d8069797b1e732495a3402f037395d1fd09ff
doc_id: 846378
cord_uid: rgpyhb05

There is potentially considerable variation in the nature and duration of the care provided to hospitalised patients during an ongoing infectious disease epidemic or pandemic. Improvements in care and clinician confidence may shorten the time spent as an inpatient, or the need for admission to an intensive care unit (ICU) or high density unit (HDU), while novel treatment modalities may reduce the time course of illness. On the other hand, limited resources at times of high demand may lead to rationing of resources, with less beneficial consequences. Despite little evidence on how the values of such variables change over the course of a crisis (such as the current COVID-19 pandemic), they may nevertheless be used as proxies for disease severity, outcome measures for clinical trials, and to inform planning and logistics. In this study, we investigate such time trends in an extremely large international cohort of 142,540 patients with symptom onset of, or hospital admission for, COVID-19 during 2020. The variables investigated are time from symptom onset to hospital admission, probability of ICU/HDU admission, time from hospital admission to ICU/HDU admission, case fatality ratio (CFR) and total length of hospital stay. Time from hospital symptom onset to hospital admission showed a rapid decline during the first months of the pandemic followed by peaks during August/September and December. ICU/HDU admission was more frequent from June to August, while there were only modest time trends in time from hospital admission to ICU/HDU. The CFR was lowest from June to August, a trend mostly driven by patients with no ICU/HDU admission. Raw numbers for overall hospital stay showed little overall variation over the time period, but further examination reveals a clear decline in time to discharge for ICU/HDU survivors. Our results establish that variables of these kinds have limitations when used as outcome measures in a rapidly-evolving situation.

During an epidemic or pandemic of a novel infectious disease, variations in the duration of understanding of the natural history of diseases improves with time (Docherty et al. 2021) , 53 and so too, confidence in safe discharge criteria or alternative models of care (Rojek and 54 Horby 2016), such as remote monitoring (Nunan et al. 2020 ; Bell et al. 2021 ). Moreover, the 55 introduction of effective treatments (Rochwerg et al. 2020 ) and standardisation of care may 56 rapidly reduce the severity or time course of illness (Dennis et al. 2021 ). However, decisions 57 about whether to admit or escalate care are also dependent on logistic factors such as the 58 availability of resources (e.g. ventilators, intensive care beds, staff) that may be rationed 59 during the peak of a pandemic, but abundant at other phases of an outbreak (Tyrrell et al. There may also be changes in policy to admit patients for indications that are not clinical -62 such as to facilitate effective quarantine (CMO Messaging 2021) or supervise provision of 63 treatments in clinical trials. We hypothesise that there is significant variation in the patient 64 journey over a pandemic period, and that this variability may limit the way these data can 65 be responsibly used. CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) Question 2: Variation in the proportion of patients being admitted to an ICU or HDU. 101 Patients were excluded if this variable was not available.

Question 3: Variation in the time from COVID-19 admission to ICU/HDU admission. Patients 103 were excluded if they were never admitted to an ICU or HDU, or if this variable was 104 otherwise not available.

Question 4: Variation in the overall case fatality rate. Patients were excluded if the final 106 outcome of their hospital stay was either not recorded or recorded as something other than 107 "death" or "discharge" (for example, transfer to another facility).

Question 5: Variation in the time from COVID-19 admission to death or discharge. (We 109 describe either as an "outcome".) Exclusions were as in question 4, as well as patients who 110 had a recorded outcome but no recorded outcome date.

Question 6: Variation in the status of patients (admitted, ICU/HDU admitted, dead, 112 discharged, or unknown outcome) on a given day after admission. Excluded here were 113 patients whose ICU/HDU status on the day of admission was unknown. For all analyses with a single outcome variable, we plotted its mean value against the 124 . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted June 3, 2021. We used two special categories of symptoms at admission: "common" symptoms (cough, 139 fatigue, fever, and shortness of breath) and gastrointestinal symptoms (abdominal pain, 140 diarrhoea, and vomiting). We introduced variables to the dataset counting the number of 141 each of these categories present for each patient. Missing data was disregarded here, so 142 these represent lower bounds. Multivariable linear regression was used to investigate factors associated with time from 147 onset of symptoms to hospital admission (question 1), time from COVID-19 admission to 148 . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted June 3, 2021. CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted June 3, 2021. ; https://doi.org/10.1101/2021.06.01.21258150 doi: medRxiv preprint unknown sex), median age 70 [IQR 56-82]), admitted at 620 sites in 47 countries. Table 1 173 shows a summary of baseline characteristics, and more detail, including country of origin 174 and cross tabulation by month of admission, can be found in table S1. 

. CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted June 3, 2021. gives the number of times the condition is recorded as present over the number of times its 185 presence or absence is recorded (i.e. the data is non-missing). Designated "common" This variable showed a marked decline during March, from a median of 9 days (IQR 5-14) for 197 patients with onset in the week beginning March 1 to 3 (IQR 0-7) in that beginning April 5.

Little further variation occurred until late July, when a gradual increase started, which then 199 peaked at a median of 6 (IQR 2-9) for the weeks in late August and early September before a 200 decline to a low of a median 4 (IQR 1-7) days in November; this was followed by another The four most frequent symptoms at admission were cough, fever, shortness of breath, and 206 fatigue; we class these as "common" (see Methods). The number of these that were present 

. CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 3, 2021. with the number of "common" symptoms (25% increase per symptom, 95% CI 24.4%-223 25.6%), the number of gastrointestinal symptoms (9.8% increase per symptom, 95% CI 9%-224 10.5%), male sex (4.4% increase, 95% CI 3.4%-5.5%) and discharge as the final outcome 225 (13% increase, 95% CI 11.6%-14.5%).

. CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 3, 2021. CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 3, 2021. ; https://doi.org/10.1101/2021.06.01.21258150 doi: medRxiv preprint (OR 1.46, 95% CI 1.14-1.88), hypertension (OR 1.26, 95% CI 1.2-1.32) and obesity (OR 1.74, 260 95% CI 1.65-1.83), whereas a wide variety of serious or chronic medical conditions were 261 associated with lower odds (see Table S5 ), as was smoking (OR 0.82, 95% CI 0.85-0.9). The 262 most extreme fitted odds ratio for a comorbidity with a positive association was 1.74 for 263 obesity, while that for an inverse association was 0.21 for dementia. chronic haematological disease (20.6% increase, 95% CI 12.8%-28.9%), and chronic kidney 282 disease (9.8% increase, 95% CI 5.9%-13.9%). In contrast, obesity (7.5% decrease, 95% CI 283 . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 3, 2021. ; https://doi.org/10.1101/2021.06.01.21258150 doi: medRxiv preprint 5.1%-9.9%), diabetes (3.5% decrease, 95% CI 0.7%-6.1%) and smoking (4.9% decrease, 95% 284 CI 0.1%-9.5%) were associated with shorter time to ICU. There was also evidence of a longer 285 time to ICU/HDU amongst pregnant patients (17.5%, 95% CI 3.4%-33.5%). is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 3, 2021. ; https://doi.org/10.1101/2021.06.01.21258150 doi: medRxiv preprint while there was a distinct peak around August and September in those without (top right).

When age is also considered (supplementary figures S5 and S6) a notable additional pattern 309 is the clear correlation of time to discharge and age in surviving non-ICU/HDU patients, 310 which is much less obvious, if present at all, in patients with an ICU/HDU admission. CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 3, 2021. (3.6% increase, 95% CI 2.1%-5.2%), and that with the shortest was August (10.8% decrease, 331 95% CI 6.4%-15%). 

There was no evidence that ICU/HDU patients admitted in March and May had a longer time 342 . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 3, 2021. CI 0.2%-4.8%) and shorter times to discharge (1.7% decrease, 95% CI 0.3%-3%).

. CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 3, 2021. from hospital admission to death, discharge or ICU/HDU admission amongst a smaller 407 cohort of UK patients. We confirm many of the trends that they identified, including the 408 lower CFRs over the summer and the increased odds of ICU/HDU admission in middle-aged 409 age groups. They did not, however, detect the increase in the proportion of patients with an 410 . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 3, 2021.

ICU/HDU admission during the summer, or the decline in time to discharge amongst non-411 ICU/HDU patients over the entire time period. As there were many fewer hospitals included 412 in that study than in ours (31 versus 620) this may be suggestive of variation in available 413 ICU/HDU capacity and usage amongst participating sites in the two studies. CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 3, 2021. ; https://doi.org/10.1101/2021.06.01.21258150 doi: medRxiv preprint partially be attributed to how case definitions are applied by physicians, or to the patient's 435 own perceptions, or to those of their families. Some presentations are likely to be more 436 alarming to the latter two groups than others; for example, individuals with none of the four 437 symptoms described above were admitted fastest of all and, amongst these, confusion was 438 the most prevalent other symptom. This also warns against using outcome data that are not adequately controlled to assess 474 efficacy and safety of treatments or other interventions, as effects may rather reflect 475 capacity of a system to provide high-quality care.

We found that shorter time to death is associated with female sex, lack of ICU/HDU 478 admission, and, amongst ICU/HDU patients, the extremes of age. Shorter time to discharge 479 is also associated with female sex and lack of ICU/HDU admission, and this variable 480 increases monotonically with age. CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 3, 2021. ; https://doi.org/10.1101/2021.06.01.21258150 doi: medRxiv preprint vary; and changes in transmissibility and virulence are expected to occur.

The observed variability should inform on the limitations of using observational data during 509 a long-lasting pandemic for management purposes in practice, and also question the use of 510 some variables, such as length of stay in hospital or in ICU, as clinical trial outcomes. This 511 demonstrates the importance of controlling for patient outcome data when designing 512 clinical trials; for example, using our data, assessing a new treatment during the months of 513 March to July will have shown a decrease in CFR from 33%, to 21% that may have been 514 falsely attributed to a treatment effect without a concurrent randomised control. CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted June 3, 2021. 

Implementation and Evaluation of a COVID-19 Rapid Follow-up Service 562 for Patients Discharged from the Emergency Department

Ggalluvial: Layered Grammar for Alluvial Plots

Wuhan Novel Coronavirus

Improving 568 Survival of Critical Care Patients With Coronavirus Disease 2019 in England: A National Cohort 569 Study

Changes in in-Hospital Mortality in the 572 First Wave of COVID-19: A Multicentre Prospective Observational Cohort Study Using the WHO 573 Clinical Characterisation Protocol UK

Emergence and 577 Spread of a SARS-CoV-2 Variant through Europe in the Summer of 2020

Global Outbreak Research: Harmony Not Hegemony

Symptoms at 582 Presentation for Patients Admitted to Hospital with Covid-19: Results from the ISARIC 583 Prospective Multinational Observational Study

Trends in Risks of Severe Events and Lengths of Stay 587 for COVID-19 Hospitalisations in England over the Pre-Vaccination Era: Results from the Public 588

COVID-19 Rapid Guideline: Managing 591 COVID-19

Triage Into the Community for COVID-19 (TICC-19) Patients Pathway -Service 594 Evaluation of the Virtual Monitoring of Patients with COVID Pneumonia

Covid-19: How to Triage Effectively in a Pandemic

R: A Language and Environment for Statistical Computing

A Living WHO Guideline on Drugs for Covid-19

Modernising Epidemic Science: Enabling Patient-605 Centred Research during Epidemics

Managing Intensive Care Admissions When 608

There Are Not Enough Beds during the COVID-19 Pandemic: A Systematic Review

Welcome to the Tidyverse

COVID-19 Clinical Management: Living Guidance