key: cord-0842257-8a59ltcj
authors: Zazzara, M. B.; Penfold, R. S.; Roberts, A. L.; Lee, K.; Dooley, H.; Sudre, C. H.; Welch, C.; Bowyer, R. C. E.; Visconti, A.; Mangino, M.; Freydin, M. B.; El-Sayed Moustafa, J. S.; Small, K.; Murray, B.; Modat, M.; Wolf, J.; Ourselin, S.; Martin, F. C.; Steves, C. J.; Ni Lochlainn, M.
title: Delirium is a presenting symptom of COVID-19 in frail, older adults: a cohort study of 322 hospitalised and 535 community-based older adults
date: 2020-06-17
journal: nan
DOI: 10.1101/2020.06.15.20131722
sha: f76b0e465e94c6dbb7dd13601bf575b5644fa34a
doc_id: 842257
cord_uid: 8a59ltcj

Background: Frailty, increased vulnerability to physiological stressors, is associated with adverse outcomes. COVID-19 exhibits a more severe disease course in older, co-morbid adults. Awareness of atypical presentations is critical to facilitate early identification. Objective: To assess how frailty affects presenting COVID-19 symptoms in older adults. Design: Observational cohort study of hospitalised older patients and self-report data for community-based older adults. Setting: Admissions to St Thomas' Hospital, London with laboratory-confirmed COVID-19. Community-based data for 535 older adults using the COVID Symptom Study mobile application. Subjects: Hospital cohort: patients aged 65 and over (n=322); unscheduled hospital admission between March 1st, 2020 - May 5th, 2020; COVID-19 confirmed by RT-PCR of nasopharyngeal swab. Community-based cohort: participants aged 65 and over enrolled in the COVID Symptom Study (n=535); reported test-positive for COVID-19 from March 24th (application launch)- May 8th, 2020. Methods: Multivariate logistic regression analysis performed on age-matched samples from hospital and community-based cohorts to ascertain association of frailty with symptoms of confirmed COVID-19. Results: Hospital cohort: significantly higher prevalence of delirium in the frail sample, with no difference in fever or cough. Community-based cohort: significantly higher prevalence of probable delirium in frailer, older adults, and fatigue and shortness of breath. Conclusions: This is the first study demonstrating higher prevalence of delirium as a COVID-19 symptom in older adults with frailty compared to other older adults. This emphasises need for systematic frailty assessment and screening for delirium in acutely ill older patients in hospital and community settings. Clinicians should suspect COVID-19 in frail adults with delirium.

and screening for delirium in acutely ill older patients in hospital and community settings. Clinicians should suspect COVID-19 in frail adults with delirium.

COVID-19, an acute respiratory disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), poses a significant threat to the older population, who are at increased risk from infection, hospitalisation and death [1, 2] . In a study from Wuhan, China, over 70% of 339 patients over 60 years of age with COVID-19 were classified as severe or critical, with a case fatality rate of 19% and median survival of 5 days following admission [3] . In Italy, the highest case fatality rate was observed among older, male patients with multiple comorbidities [4] and in an American study, adults >65 years old represented 80% of COVID-19 fatalities [5] . Vulnerability within the older population might partially be explained by a greater number and severity of chronic diseases [6] and immunosenescence [7, 8] . It is imperative to rapidly identify and contain infection within this group.

Prevalent symptoms reported in patients with confirmed COVID-19 infection from a variety of international settings are summarised in Table 1 . However, it is well-documented that presenting symptoms vary with age for many diseases, including respiratory infections [9, 10, 11] . Symptoms including delirium, generalised weakness, malaise, anorexia, headache, dizziness, falls and functional decline are more common in older people. Less typical presentations of COVID-19 in older adults have been documented in case reports and small cohort studies (n<60), with symptoms including weakness, headache, delirium and an absence of fever or cough [12, 13, 14] . In work by our research group using point-of-care testing for COVID-19 in older adults, hypothermia was noted to be an early clinical sign in a minority of patients [15] . All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) 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 17, 2020. . https://doi.org/10.1101/2020.06. 15.20131722 doi: medRxiv preprint Frailty, an increased vulnerability to physiological stressors [16] , confers increased risk of adverse health outcomes including mortality, falls, fractures, comorbidities, hospitalisation, and disability [16, 17] . Frailty is also associated with increased likelihood of hospital-acquired complications, such as delirium, falls or deconditioning. Recent National Institute for Health and Care Excellence (NICE) guidelines in the UK recommend that all adult patients over 65 years old should be assessed for frailty on hospital admission to guide appropriate management and advanced care planning, using the Clinical Frailty Scale (CFS) where relevant [18] .

However, no study has reported on effects of frailty on symptoms of COVID-19, despite calls for data in this important area [19] .

We use point-of-care data from patients admitted to a large UK hospital trust, supported by community-based COVID Symptom Study mobile application ("app") data, to assess how frailty affects presentation of confirmed COVID-19 infection in older adults.

All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. questionnaire (a 7-item screening tool) [23] adapted for large-scale app usage, to assess level of frailty (Supplementary Table SP1 ). PRISMA-7 has been validated for use in older adults [24] and has good validity compared to the CFS [25] . The app was upgraded in April 2020 to allow family members or care-givers to submit data by-proxy, with consent. This has increased numbers of reporting older, frailer participants.

All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) 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 17, 2020. . https://doi.org/10.1101/2020.06. 15.20131722 doi: medRxiv preprint Consent to data analysis and sharing was obtained when individuals enrolled in the app. UK data up to May 8 th , 2020 were inspected and, as part of the quality control (QC) stage, invalid BMI (<15 and >55) and age entries (<18 years) removed. QC was performed by an automated python script developed independent of data analysis.

The total dataset comprised 2,848,396 individuals, 11,206 of whom had reported a positive COVID-19 test. This was filtered to include only those aged 65 years and above. All longitudinal entries were condensed per person to form a binary metric recording whether the person had experienced each of the 14 symptoms (below) during the study period.

Delirium was defined as a clinically documented delirium diagnosis on admission in the hospitalised cohort and "Yes" to the following question in the community-based cohort: 'Do you have any of the following symptoms:

confusion, disorientation or drowsiness?'

Descriptive statistics were used to describe demographic and key clinical characteristics of the study populations: overall and according to frailty. Frailty was classified as a binary variable: CFS≥ 5 = frail; CFS<5 = not frail for the hospital cohort and PRISMA7≥3 = frail; PRISMA7<3 = not frail in the community-based cohort. A binary temperature classification was used in the hospitalised cohort, based on a previous large observational cohort study [26] : ≥37.5°C = fever; 37.5 < afebrile.

Continuous variables were expressed as mean +/-standard deviation, and categorical variables as frequencies by absolute value and percentages (%). Differences in proportions and means of covariates between subjects with and without frailty were assessed using Wilcoxon's test for continuous variables and Fisher's exact test for categorical variables.

Following age-matching by frailty status in the hospitalised cohort (n=210), multivariable logistic regression analysis was performed to assess whether frailty predicts symptoms of confirmed COVID-19 infection on hospital admission, including fever, cough and delirium. BMI and sex were covariates. All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) 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 17, 2020. . https://doi.org/10.1101/2020.06.15.20131722 doi: medRxiv preprint Following age-matching by frailty status in the community-based cohort (n=238), multivariable logistic regression analysis was performed to assess whether frailty predicts symptoms of reported confirmed COVID-19 infection. Symptoms analysed included delirium, fever, persistent cough, fatigue, myalgia, shortness of breath, diarrhea, anorexia, abdominal pain, hoarse voice, anosmia/dysgeusia, headache, chest pain and pharyngodynia. BMI and sex were covariates.

Data analysis and graphics were performed in the R statistical environment (version 4.0) using the Tidyverse packages.

Hospital cohort:

Baseline characteristics of the unmatched hospital study cohort are summarised in Supplementary Table SP2 .

Mean age of 322 individuals with confirmed COVID-19 RT-PCR was 78.58 years (standard deviation (SD) +/-7.93); 154 (48%) were females. 165 (51%) patients were classified as frail.

After age-matching, difference in mean age between the frail (77.92 +/-SD 6.64 years) and non-frail group (77.85 +/-SD 7.04 years) was non-significant (Wilcoxon's p>0.05). The proportion of females was 39% in both groups. BMI differences between the frail and non-frail groups were non-significant (Figure 1 a-c and Table 2 ).

Comorbidities of the matched study cohort according to frailty status are summarised in Table 2 . There was a higher proportion of respiratory disease and dementia in the frail compared to non-frail group, but no significant differences in prevalence of cardiovascular disease, diabetes, psychiatric disease or cancer ( Table 2) .

After age-matching, delirium was reported in 40 (38%) of frail and 13 (12%) of non-frail patients with COVID-19 (Table 3) . Frailty was found to significantly predict delirium (P-value: 0.013; Odds Ratio (OR) (95% Confidence Interval (CI)) = 3.22 (1.44, 7.21)) after false discovery rate (FDR) correction for multiple testing ( Figure 2 and Table 3 ). There were no significant differences between frail and not frail for other symptoms (fever (temperature ≥ 37.5C) and cough). All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) 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 17, 2020. . https://doi.org/10.1101/2020.06.15.20131722 doi: medRxiv preprint

Community-based cohort:

The proportion of total COVID Symptom Study app users ≥65 years of age was 334,544/ 2,848,396 (11.7%).

Baseline characteristics of the unmatched study population are summarised in Supplementary Table SP2. 192 (36%) were classified as frail. 77% of the 192 frail participants with a reported positive COVID-19 test were reporting-by-proxy. In the frail group, delirium was present in 94/192 (49%) (self-reported 20/44 (45%);

reported-by-proxy 74/148 (50%)). Baseline characteristics of the age-matched cohort are summarised in Figure   1 (d-f) and Table 2 . Comorbidities are summarised in Table 2 .

After age-matching, frailty was found to significantly predict delirium (P-value 0.038; OR (95% CI) = 2.29 (1.33, 4.00)), after FDR correction for multiple testing (Table 3) . Frailty also predicted fatigue (P-value: 0.038; OR = 2.23 (1.27, 3.96)) and shortness of breath (P-value: 0.043; OR = 2.00 (1.19, 3.39)). There were no significant differences between frail and not frail for the other 11 symptoms analysed (Table 3 and   Supplementary Table SP3 ).

In this study, we demonstrate that prevalence of delirium and confusion is significantly higher in frail compared to non-frail older adults with COVID-19, highlighting both that a frailty assessment is fundamental, but also a systematic evaluation of change in mental status needs to be included when assessing this population. Delirium is associated with greater morbidity and mortality amongst older adults [27] and may be the only presenting symptom of COVID-19 in this highly vulnerable population.

COVID-19 disproportionately affects the older population, partly because of the increased prevalence of frailty.

Frail adults in residential settings may be at particular risk of transmission of respiratory illness. Case reports of COVID-19 outbreaks in nursing and residential homes have been published, highlighting the risks of infection in residential settings and a high case fatality [28] . Aside from residential facilities, many frail older adults are dependent on home care packages, with movement of care workers between residences. It is imperative that differences in COVID-19 presentation as a result of frailty are understood to prompt rapid diagnosis, isolation All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) 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 17, 2020. . https://doi.org/10.1101/2020.06.15.20131722 doi: medRxiv preprint and contact tracing to limit outbreaks, and to clinical assessment and treatment. Our findings alert clinicians to the need to identify frailty and look for atypical COVID-19 presentations.

We also identified self-reported fatigue and shortness of breath as more common amongst adults with frailty.

Higher prevalence of fatigue is consistent with a decrease in physiological reserve conferred by frailty.

Early identification of frailty facilitates a differential, targeted clinical approach, focusing not only on "typical" COVID-19 symptoms (e.g. fever and cough) but also on symptoms, such as delirium, demonstrated here to be more common in frail older adults. The importance of comprehensive frailty assessment in acute clinical settings and the community is nationally recognised [29, 30] . Rapid NICE guidance produced in response to the COVID-19 outbreak outlines the importance of identifying those at increased risk of poor outcomes and perhaps less likely to benefit from critical care admission [18, 30] . Delirium is frequently under-recognised in hospitalised populations [27] . Recognition of delirium necessitates a thorough investigation for underlying causes and management requires a tailored, patient-centred approach.

Delirium is a frequent complication of hospitalisation for older adults [31] and for patients admitted to critical care of all ages [32] . Whilst pathophysiology is not fully understood, neuroinflammation has been hypothesised to play a key role in its development [32] . In the context of COVID-19, few of the early descriptive studies included delirium or new confusion [33] , emphasising the importance of our finding. A recent case study described acute confusion in an older patient with no pre-existing cognitive impairment presenting with COVID-19 [14] . In a retrospective study of 214 patients from China, Mao et al. described neurological manifestations of COVID-19 and reported 'impaired consciousness' in 7.5% of patients, some of which may represent delirium. Other central nervous system (CNS) manifestations of COVID-19 infection reported were headache, dizziness, acute cerebrovascular disease, ataxia and seizures. These were significantly more common in severe infections and in older and comorbid patients [34] . In addition, non-neurological triggers may be important in delirium. Hypoxia, common in COVID-19 infections [35] , is a well-recognised cause of delirium [36] . In addition, electrolyte disturbance, dehydration, and tachycardia common to all viral infections may play a role. Some have suggested that reduced patient contact during the pandemic, for protection of staff and other patients and because of time taken to don personal protective equipment, may be contributing to missed or delayed diagnoses of fluctuant delirium and acute confusion in these patients [37, 38] . All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) 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 17, 2020. . https://doi.org/10.1101/2020.06.15.20131722 doi: medRxiv preprint A number of other coronaviruses have shown significant neurotropism in animal models [39, 40] ; it is plausible that SARS-CoV-2 displays this property. Indeed, SARS-CoV-2 was found in cerebrospinal fluid of a reported case of COVID-19 encephalitis in Beijing [41] . Different mechanisms have been proposed to explain neurological involvement in COVID-19. The virus could reach the CNS via a compromised blood-brain barrier, or migrate via neuronal pathways, infecting peripheral sensory or motor nerve endings [42] , leading to subsequent neurological damage and neuroinflammation. Some authors have proposed that angiotensinconverting enzyme 2 (ACE2), identified as the functional receptor for SARS-CoV-2, also expressed in the brain [43] , could play a crucial role in mediating the inflammatory response in COVID-19 infection [34, 44] .

There are several limitations to this study. For the hospital cohort, admissions data were collected retrospectively from electronic health records. CFS was not always documented systematically and, in approximately one-third of cases, had to be retrospectively calculated using relevant information from admission clerking.

In the community-based cohort, COVID-19 test result and symptoms were self-reported and entered directly by participants or by-proxy. The proportion of total app users ≥65 years of age is 11.7%, whereas only 4.8% of app users with a confirmed positive COVID-19 test were ≥65 years of age. Since 18.3% of the UK population are aged over 65 [45] , older adults overall, and especially laboratory-tested for COVID-19, are likely to be underrepresented. Sampling using a mobile application will under-represent individuals without mobile devices and is likely to under-represent those severely affected by the disease. These biases may mean we underestimate the differences in symptoms in frail people in the community study. The fact that the most significant finding from the community-based study was recapitulated in the clinical study is reassuring. Furthermore, of app users ≥65 years of age, 16.4% are categorised as frail. This is a similar proportion to the 15-20% of older adults categorised as moderate or severely frail during validation of the electronic Frailty Index [46] . All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. For community-based data, note that COVID-19 tests were not widely available in the UK at the time of study [47, 48] . Participants were likely tested because they displayed severe symptoms, were in contact with confirmed cases, were healthcare workers or had travelled abroad to a risk area early in the outbreak.

Additionally, the app captures data on whether participants have ever been diagnosed with COVID-19; some individuals may have previously tested positive but were no longer symptomatic when they enrolled.

This is the first study reporting that frail patients with COVID-19 present with a significantly higher prevalence of delirium than non-frail patients, using hospital and community-based cohorts. Our data emphasises the need for systematic assessment of frailty and delirium as critical components when assessing acutely ill patients in both hospital and community settings. Clinicians should suspect COVID-19 in frail patients presenting with delirium. This is especially important in care homes and long-term care facilities, where resident populations are vulnerable to respiratory disease outbreaks. Early detection facilitates implementation of infection control measures and appropriate personal protective equipment to avoid catastrophic spread and reduce mortality amongst frail older adults.

All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) 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 17, 2020. (which was not certified by peer review) 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 17, 2020. (which was not certified by peer review) 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 17, 2020. All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) 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 17, 2020. (which was not certified by peer review) 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 17, 2020. . https://doi.org/10.1101/2020.06.15.20131722 doi: medRxiv preprint (which was not certified by peer review) 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 17, 2020. . https://doi.org/10.1101/2020.06.15.20131722 doi: medRxiv preprint All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) 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 17, 2020. . https://doi.org/10.1101/2020.06.15.20131722 doi: medRxiv preprint

Comorbidity and its impact on 1590 patients with Covid-19 in China: A Nationwide Analysis

Predictors of Mortality for Patients with COVID-19 Pneumonia Caused by SARS-CoV-2: A Prospective Cohort Study

Coronavirus Disease 2019 in elderly patients: characteristics and prognostic factors based on 4-week follow-up

Case-Fatality Rate and Characteristics of Patients Dying in Relation to COVID-19 in Italy

COVID-19 )

The OpenSAFELY Collaborative

OpenSAFELY: factors associated with COVID-19-related hospital death in the linked electronic health records of 17 million adult NHS patients

Age and immunity: What is 'immunosenescence'?

Frailty and sarcopenia: The potential role of an aged immune system

Elderly patients with an atypical presentation of illness in the emergency department

COVID-19 in Older Adults: Key Points for Emergency Department Providers

Clinical Characteristics of 138 Hospitalized Patients with 2019 Novel Coronavirus-Infected Pneumonia in Wuhan, China

Clinical features of COVID-19 in elderly patients: A comparison with young and middle-aged patients

A comparative-descriptive analysis of clinical characteristics in 2019-Coronavirus-infected children and adults

Atypical presentation of COVID-19 in a frail older person

Detecting SARS-CoV-2 at point of care: Preliminary data comparing Loop-mediated isothermal amplification (LAMP) to PCR

Frailty in elderly people

Frailty in Older Adults: Evidence for a Phenotype

COVID-19 rapid guideline: critical care in adults

Delirium : a missing piece in the COVID-19

Geriatric Medicine Research Collaborative (GeMRC)

A global clinical measure of fitness and frailty in elderly people

PRISMA-7: A case-finding tool to identify older adults with moderate to severe disabilities

Fit for practice

Comparison of frailty screening instruments in the emergency department

Clinical Characteristics of Coronavirus Disease 2019 in China

Delirium is prevalent in older hospital inpatients and associated with adverse outcomes: Results of a prospective multi-centre study on World Delirium Awareness Day

Epidemiology of Covid-19 in a Long-Term Care Facility in King County, Washington

The NHS Long Term Plan

Frailty index predicts poor outcome in COVID -19 patients

Delirium in hospitalized older adults

Delirium in critically ill patients

Neurologic Features in Severe SARS-CoV-2 Infection

Neurologic Manifestations of Hospitalized Patients With Coronavirus Disease

Clinical characteristics of coronavirus disease 2019 (COVID-19) in China: A systematic review and meta-analysis

The major risk factors for delirium in a clinical setting

Collaborative Delirium Prevention in the Age of COVID-19

Delirium diagnosis, screening and management

Middle east respiratory syndrome coronavirus causes multiple organ damage and lethal disease in mice transgenic for human dipeptidyl peptidase 4

Mechanisms of Host Defense following Severe Acute Respiratory Syndrome-Coronavirus (SARS-CoV) Pulmonary Infection of Mice

Nervous system involvement after infection with COVID-19 and other coronaviruses

Axonal Transport Enables Neuron-to-Neuron Propagation of Human Coronavirus OC43

ACE2 in Brain Physiology and Pathophysiology: Evidence from Transgenic Animal Models

Evidence of the COVID-19 Virus Targeting the CNS: Tissue Distribution, Host-Virus Interaction, and Proposed Neurotropic Mechanisms

Office for National Statistics. Overview of the Population

Development and validation of an electronic frailty index using routine primary care electronic health record data

Testing of NHS staff and household members NHS England and NHS Improvement

COVID-19: investigation and initial clinical management of possible cases -GOV

Clinical Characteristics of 138 Hospitalized Patients with 2019 Novel Coronavirus-Infected Pneumonia in Wuhan, China

Clinical Features of 85 Fatal Cases of COVID-19 from Wuhan: A Retrospective Observational Study

Clinical characteristics of novel coronavirus cases in tertiary hospitals in Hubei Province

Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China

Clinical findings in a group of patients infected with the 2019 novel coronavirus (SARS-Cov-2) outside of Wuhan, China: retrospective case series

Surveillance Group I superiore di S. Characteristics of COVID-19 patients dying in Italy Report based on available data on

First cases of coronavirus disease 2019 (COVID-19) in the WHO European Region

Coronavirus disease 2019: new things to know