National French Survey of Coronavirus Disease (COVID-19) Symptoms in People Aged 70 and Over

Abstract The objective of this national French survey was to determine the coronavirus disease 2019 (COVID-19) semiology in seniors (n = 353; mean, 84.7 ± 7.0 years). A total of 57.8% of patients exhibited ≤3 symptoms, including thermal dysregulation (83.6%), cough (58.9%), asthenia (52.7%), polypnea (39.9%), and gastrointestinal signs (24.4%). Patients ≥80 years exhibited falls (P = .002) and asthenia (P = .002). Patients with neurocognitive disorders exhibited delirium (P < .001) and altered consciousness (P = .001). Clinical peculiarities of COVID-19 were reported in seniors. Clinical Trials Registration NCT04343781.

Since December 2019, coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) is spreading worldwide from China, affecting millions of people. Although older adults do not appear more prone than younger ones to be infected, they are more at risk of developing severe and lethal forms of COVID-19 [1][2][3]. The core question is thus to properly discuss the diagnosis of COVID-19 in older patients. It is commonly admitted that the semiology of older adults differs from that encountered in younger ones. Changes in the clinical expression of the diseases and difficulties in interpreting the clinical signs in older patients could blur the diagnosis process. If these peculiarities were also retrieved with COVID-19, it could be the cause of delayed diagnosis among older patients and be responsible for delayed care and isolation measures with subsequent higher risk of virus propagation. The objective of this national French survey was to describe and identify the symptoms most frequently encountered in people aged 70 years and older diagnosed with COVID-19.

METHODS
This cross-sectional study was conducted by the French Society of Geriatrics and Gerontology. An online standardized questionnaire was sent by email to all Society members and widely communicated through the professional networks in geriatrics and infectious diseases. Physicians were asked to report, between March 22 and April 5, 2020, their last 10 patients aged ≥70 years with confirmed SARS-CoV-2 infection (as defined as a positive reverse transcriptase-polymerase chain reaction [RT-PCR] test result). Those who had treated fewer than 10 diagnosed patients were asked to submit a questionnaire for each of them. The study was conducted in accordance with the ethical standards set forth in the Helsinki Declaration (1983), was declared to the National Commission for Information Technology and civil Liberties (ar20-0031v1), and was registered on clinicaltrials.gov under number NCT04343781.
The following characteristics were collected for each patient: demographic (age, sex, place where living, place of care, most recent disability score according to the Iso-Resource Group) [4], and medical history (major neurocognitive disorders [MND], hypertension, diabetes mellitus, asthma or chronic obstructive pulmonary disease, cardiomyopathy, severe chronic renal failure defined as creatinine clearance <30 mL/min, solid or hematological cancer).
The following symptoms observed within the first 72 hours of SARS-CoV-2 infection (ie, 72 hours from suspicion, possibly before diagnostic confirmation by RT-PCR test) were collected for each patient using yes/no questions: general signs (sudden deterioration of general condition, temperature, blood pressure), respiratory signs (cough, polypnea), ear nose and throat signs (rhinorrhea, odynophagia, otalgia, conjunctivitis, dysgeusia or ageusia, anosmia), gastrointestinal signs (diarrhea, nausea, vomiting) and geriatric syndromes (falls, hypo-or overactive delirium, altered consciousness). Changes in complete blood count (leukopenia, lymphopenia, thrombocytopenia) were also collected, with details when available.

Older patients' characteristics are presented in
Finally, the multiple correspondence analysis results distinguished between 2 profiles of older patients. The first profile matched with patients younger than age 80 years without MND, who exhibited more frequent hyperthermia and cough during the first 72 hours of the infection, but no fall, altered consciousness, or hypoactive delirium. In contrast, the second profile matched with patients aged 80 years and older with MND; the latter exhibiting more frequently no specific symptoms, and most often an absence of hyperthermia, polypnea, cough, and dysgeusia-ageusia.

DISCUSSION
This national French survey shows that older adults with COVID-19 exhibit a pauci-symptomatic clinical picture with less than 3 signs during the first 72 hours of the infection, generally combining general and respiratory signs (eg, Abbreviations: COPD, chronic obstructive pulmonary disease; ENT, ear nose and throat; GIR, Iso-Resource Group; MND, major neurocognitive disorders; SD, standard deviation. hyperthermia and cough) with peculiarities that should alert the clinician (eg, sudden deterioration of general condition, diarrhea, lymphopenia, geriatric syndromes including falls and delirium). Various clinical profiles were highlighted across older adults, especially among the oldest-old ≥80 years and those with chronic diseases such as MND.
Our survey provides the first description of the COVID-19 signs in older, and even oldest-old, adults with comorbidities [1][2][3]. Compared with previous meta-analyses in younger adults [5][6][7], we found that older adults with COVID-19 often exhibit thermal dysregulation, which, however, results less often in hyperthermia (56% here vs 82% [5] to 91% [6] in younger adults) and more often in subfebrile temperatures or alternations of hyperthermia and hypothermia (not described thus far to our knowledge). The prevalence of cough was similar (59% here vs 61% [5] to 72% [7] in younger adults). In contrast, the sudden deterioration of general condition, mostly illustrated by marked asthenia, was particularly frequent in older adults (53% here vs 36% [5] to 51% [6] in younger adults). Also, older adults exhibited more often dyspnea (40% here vs 26% [5] to 30% [6] in younger adults) and gastrointestinal signs (24.4% here with mostly diarrhea [21.8%] vs 10% in younger adults [5,8]). This should encourage clinicians to integrate the gastrointestinal signs into the diagnostic reasoning for SARS-CoV-2 infection in older adults. Older adults had less often anosmia (2% here vs 86% in younger adults [9]) and dysgeusia-ageusia (7% here vs 89% in younger adults [9]). The latter prevalence should be cautiously interpreted however because of olfactory and gustatory dysfunctions with advancing age [10]. Finally, we found a higher proportion of lymphopenia in older adults compared with the general population (75% here vs 55% [3]). Lymphopenia was more significant than that usually observed in the normal aging population (750/mm 3 vs 1432/mm 3 in the literature [11]), and may explain part of the excess mortality observed in older adults with COVID-19 [1].
Our study has a number of limitations. This is an observational cross-sectional study conducted on a panel of French older patients who may be not representative of the general older population. The 64 physicians who responded to the survey, however, came from all French regions. A reporting bias cannot be ruled out because the accuracy and completeness of the data were entirely reliant upon physicians' declarations, although the questionnaire was designed to limit variability in readers' interpretations by asking only factual data. Also, in the absence of mass screening policy in France, only patients for whom a biological test had been carried out because of suspected infection, for clinical reasons for example, could be included, which may have overestimated the prevalence of some signs. The lack of control group prevented to determine the average number of symptoms met in non-COVID-19 French older adults. Similarly, no data were available on the use of concomitant drugs, for example of antibiotics, which could partially explain increases in gastrointestinal signs. Finally, only patients diagnosed with RT-PCR test were included, although the sensitivity of this test presents a relatively high risk of false negatives (sensitivity of 72%) [12], which may have excluded a number of patients with COVID-19.
In conclusion, this national French survey revealed that the clinical picture of older adults with COVID-19 includes both general and respiratory signs like in younger adults (eg, hyperthermia and cough), but also more peculiar features such as marked asthenia, diarrhea, lymphopenia, and geriatric syndromes. We also reported various clinical profiles across older adults, notably in those aged 80 years and older and those with a history of MND who appeared particularly pauci-or asymptomatic during the first 72 hours of the infection. These findings should be integrated into the clinical reasoning in geriatric medicine and encourage the systematization of diagnostic tests for SARS-CoV-2 infection in older adults.