key: cord-0987904-qy1rmzjl authors: Klein, H.; Asseo, K.; Karni, N.; Benjamini, Y.; Nir-Paz, R.; Muszkat, M.; Israel, S.; Niv, M. Y. title: Onset, duration, and persistence of taste and smell changes and other COVID-19 symptoms: longitudinal study in Israeli patients date: 2020-09-27 journal: nan DOI: 10.1101/2020.09.25.20201343 sha: 4985565e3b2629c29663983138da968fc9953ef2 doc_id: 987904 cord_uid: qy1rmzjl Objectives: The multifaceted disease manifestation of COVID-19 requires longitudinal characterization of symptoms, to aid with screening and disease management. Methods: Phone interviews and follow-ups were completed with 112 mild COVID-19 RT-PCR-positive adult patients, over a 6 week period. Results: More than one symptom at disease onset was experienced by ~70 of the patients. Over one third of patients experienced fever, dry cough, headache, or muscle ache as the first symptom. If fatigue was reported, it was usually the first symptom to appear. Smell and taste changes had occurred 3.9 {+/-} 5.4 and 4.6 {+/-} 5.7 days (mean {+/-} SD) since disease onset and emerged as first symptoms in 15% and 18% of patients, respectively. Fever was the shortest lasting symptom (5.8 {+/-} 8.6 days (mean {+/-} SD), and smell and taste changes were the most long-lasting symptoms (24.3 {+/-} 22.9 days and 19.4 {+/-} 19.1 (mean {+/-} SD), respectively), with longer smell recovery correlated with smell change severity. In one third of patients who reported cough, smell and taste changes, these symptoms persisted after negative RT-PCR tests. Conclusions: Each symptom can occur as first or later, though some are more likely to appear as firsts, and typically more than one symptom occurs at disease onset. The severity of olfactory change is associated with its recovery time. Lack of chemosensory recuperation in recovered patients is common. These findings can aid patients through their illness and provide expected recovery patterns. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 27, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 27, 2020. Results: More than one symptom at disease onset was experienced by ~70 of the 33 patients. Over one third of the patients experienced fever, dry cough, headache, or 34 muscle ache as the first symptom. If fatigue was reported, it was usually the first 35 symptom to appear. Smell and taste changes had occurred 3.9 ± 5.4 and 4.6 ± 5.7 days 36 (mean ± SD) since disease onset and emerged as first symptoms in 15% and 18% of 37 patients, respectively. Fever was the shortest lasting symptom (5.8 ± 8.6 days (mean 38 ± SD), and smell and taste changes were the longest-lasting symptoms (24.3 ± 22.9 39 days and 19.4 ± 19.1 (mean ± SD), respectively), with longer smell recovery correlated 40 with smell change severity. In one third of patients who reported cough, smell and 41 taste changes, these symptoms persisted after negative RT-PCR tests. Conclusions: Each symptom can occur as first or later, some are more likely to appear 43 as firsts, and typically more than one symptom occurs at disease onset. The severity 44 of olfactory change is associated with its recovery time. Lack of chemosensory 45 recuperation in recovered patients is common. These findings can aid patients through 46 their illness and provide expected recovery patterns. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint reported symptoms also include sputum production, hemoptysis and lack of 61 appetite[6,7]. These symptoms' order of appearance, severity, durations and 62 persistence after recovery, is only partly studied [5, 8, 9] . 63 In this study, we address gaps in knowledge regarding the order of symptoms 64 appearance and their durations in RT-PCR positive patients. Establishing COVID-19 65 course of illness in terms of symptoms severity, order of appearance, duration, and 66 persistence after recovery, is of key importance for patients' ability to cope with the 67 mental, social, and financial challenges that accompany the disease. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 27, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 27, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 27, 2020. . https://doi.org/10.1101/2020.09.25.20201343 doi: medRxiv preprint 7 self-reported in 18% of patients under "any other symptoms" and was therefore 120 included in the final analysis. 121 Disease onset is defined as the first appearance of any of the COVID-19 symptoms 122 described above. Persistent symptoms are defined as symptoms that progress post 123 patient's recovery. 124 Order of appearance and duration were calculated from the dates reported by the 125 participants for each symptom. 126 Severity of smell (taste) change was calculated as the delta between the reported 127 ability to smell before the illness and its reported ability during the illness, rated on a 128 1-10 scale. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 27, 2020. . https://doi.org/10.1101/2020.09.25.20201343 doi: medRxiv preprint 8 their hospitalization and / or were hospitalized in the intensive care unit) and the 142 remaining 106 were classified as ambulatory patients. Order of symptoms' appearance 145 Symptoms appearance were usually concomitant: The first and second symptoms to 146 appear were accompanied by additional symptoms in 69% and 45% of the patients, 147 respectively. Headache, fever, dry cough, and muscle aches were the first symptoms 148 to appear in more than one third of the patients ( Figure 1A) . 149 Smell and taste changes were the first symptoms to appear (either solely or along with 150 other symptoms) in only 15% and 18% of patients, respectively. Fatigue was as a first 151 symptom in majority of patients who reported it( Figure 1A ). 152 Headache, fever, dry cough, and muscle aches appeared 1.99 ± 5, 1.79 ± 4.64, 1.35 ± 153 3.44 and 2.09 ± 5.07 days (mean ± SD), respectively, after illness onset. Smell and taste 154 changes appeared later, at 3.94 ± 5.45 and 4.58 ± 5.72 days (mean ± SD), respectively, 155 after illness onset ( Figure 1B) . Each of the 14 symptoms analyzed here could occur as 156 first or co-first, with vomiting, breathing difficulty and diarrhea being the most spread 157 out in the order in which they appear (see web-only Supplementary Figure S2) . Indeed, 158 vomiting and diarrhea appeared after the highest number of days since illness onset 159 (5.57 ± 8.26 and 5.85 ± 9.97, respectively) ( Figure 1B ). is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 27, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 27, 2020. (Figure 2A (6-7) ). As seen in Figure 2B is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 27, 2020. . https://doi.org/10.1101/2020.09.25.20201343 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 27, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 27, 2020. . https://doi.org/10.1101/2020.09.25.20201343 doi: medRxiv preprint We found that the durations of smell and taste were correlated (0.82, p<0.001) ( Figure 220 4A), with many patients reporting identical durations for these symptoms. 221 As seen in Figure 4B , a small yet significant correlation was found also between smell 222 change severity and its duration (correlation 0.34, p=0.003). The same was not 223 repeated for taste change severity and its duration (see web-only Supplementary 224 Figure S4 ). 225 Moreover, it is apparent that some patients' smell change had recovered 226 independently of how severe it was, while for another group of patients, it was closely 227 dependent on its severity ( Figure 4B ). We did not find any particular characteristics for 228 either of these subgroups. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 27, 2020. Our study reveals that smell and taste were the first symptoms to appear, in 15% and 248 18% of cases, respectively, either with or without other symptoms. We found duration 249 of smell change to be 24.3 ± 22.9 days and of taste change to be 19.4 ± 19.1 days 250 (mean ± SD), and their post-recovery persistence to be prevalent in 36% and 32% of 251 patients, respectively. 252 Concerning other COVID-19 symptoms, we found that even though their order of 253 appearance is varied, headache, fever, dry cough, and muscle aches were often (38% 254 -43%) the first symptoms to appear. Additionally, out of the 18% patients who 255 experienced fatigue, 80% had experienced it as a first symptom (14% of all patients). 256 Of note, the prevalence of fatigue in our study was lower than previously reported 257 36%-46%[17,18], probably because here it was not explicitly part of the questionnaire, 258 but rather self-reported under the "any other symptom" question. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 27, 2020. . https://doi.org/10.1101/2020.09.25.20201343 doi: medRxiv preprint Furthermore, we found COVID-19 illness to have a prolonged phase, in which 260 symptoms such as fever, lack of appetite, headache and muscle aches are recovered, 261 but dry cough, taste change and smell change persist, with the latter having the 262 longest duration of 24.3 ± 22.9 days (mean ± SD). The prolonged phase can persist 263 post-recovery: chemosensory changes were found to persist post recovery in about 264 one third of patients experiencing them (usually by the same patient). Additionally, 265 cough, either dry or productive, was also found to persist in approximately 27% of 266 patients who reported experiencing this symptom. However, patients experiencing 267 cough as persistent can be the same or different than those experiencing persistent 268 chemosensory changes. 269 We found slightly lower prevalence of patients who experienced chemosensory 270 changes as first symptoms than those previously reported [13] , which could be due to 271 recall bias caused by retrospective data collection. We also report a lower percentage 272 of persistence post recovery prevalence of smell and taste changes compared to 273 another study [16] , which could be sue to our larger sample size or other differences. 274 Longer duration of chemosensory changes that those reported by others[14,15,19] is 275 due to the longer follow-ups we conducted. Moreover, our detailed phone interviews 276 did not rely on online surveys, increasing the reliability of the data. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 27, 2020. . https://doi.org/10.1101/2020.09.25.20201343 doi: medRxiv preprint 16 We found smell (but not taste) change severity to be correlated with its time to 283 recover. However, while for a group of the patients, the time necessary for their smell 284 change recovery correlated with magnitude of the change they had experienced, for 285 another group, the severity of their smell change was unrelated to the time it took to 286 recover. Such grouping was also reported by Gerkin et al. [12] , but characteristics that 287 might account for the differences between these two groups could not yet be 288 unraveled and require future study. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 27, 2020. . https://doi.org/10.1101/2020.09.25.20201343 doi: medRxiv preprint 3.9 ± 5.4 and 4.6 ± 5.7 days (mean ± SD), respectively, since illness onset. Their 307 recovery can be either fast, or longer and gradual, sometimes accompanied by cough. 308 Interestingly, while dry cough appears early (1.35 ± 3.44 days (mean ± SD)) and 309 productive cough later (4.54 ± 7.09 days (mean ± SD)), both types of cough may persist is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 27, 2020. . https://doi.org/10.1101/2020.09.25.20201343 doi: medRxiv preprint The socio-economic implications of the 323 coronavirus pandemic (COVID-19): A review Clinical Characteristics of Coronavirus Disease 326 2019 in China Review of the Clinical 329 Characteristics of Coronavirus Disease 2019 (COVID-19) Symptom Duration and Risk Factors 332 for Delayed Return to Usual Health Among Outpatients with COVID-19 in a 333 Gemelli Against COVID-19 Post-Acute Care Study 337 Persistent Symptoms in Patients After Acute COVID-19 The epidemiology and pathogenesis of coronavirus 340 disease (COVID-19) outbreak Clinical Characteristics of COVID-19 Patients With 343 Digestive Symptoms in Hubei, China: A Descriptive Modeling the Onset of 347 Symptoms of COVID-19. Front Public Health Sedaghat 350 AR. Olfactory Dysfunction and Sinonasal Symptomatology in COVID Prevalence, Severity, Timing, and Associated Characteristics Self-rated smell ability enables highly specific 354 predictors of COVID-19 status: a case control study in Israel More than smell -COVID-19 is 358 associated with severe impairment of smell, taste, and chemesthesis 359 Recent smell loss is the best predictor 362 of COVID-19: a preregistered, cross-sectional study Alterations in Smell or Taste in Mildly 366 Symptomatic Outpatients With SARS-CoV-2 Infection is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted Disorders in COVID-19 Patients Persistent Smell Loss Following 376 COVID-19 patients' clinical characteristics, 380 discharge rate, and fatality rate of meta-analysis Clinical characteristics of 3062 COVID-19 patients: A 383 meta-analysis Prevalence and Duration of Acute Loss of Smell or 386 Taste in COVID-19 Patients Management of post-389 acute covid-19 in primary care It is made available under a perpetuity.is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprintThe copyright holder for this this version posted September 27, 2020. . https://doi.org/10.1101/2020.09.25.20201343 doi: medRxiv preprint It is made available under a perpetuity.is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprintThe copyright holder for this this version posted September 27, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprintThe copyright holder for this this version posted September 27, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprintThe copyright holder for this this version posted September 27, 2020. . https://doi.org/10.1101/2020.09.25.20201343 doi: medRxiv preprint