key: cord-0295079-hfaiddki authors: M C, A.; Singh, A. K.; Pal, D.; Das, K.; Gajjala, A.; Venkateshan, M.; Mishra, B.; Patro, B. K.; Mohapatra, P. R.; Subba, S. H. title: Prevalence, characteristics, and predictors of Long COVID among diagnosed cases of COVID-19 date: 2022-01-08 journal: nan DOI: 10.1101/2022.01.04.21268536 sha: eeb526ced0f0cdd8ea0e334df2386fd76016265f doc_id: 295079 cord_uid: hfaiddki Background: Long COVID or long-term complication after COVID-19 has the ability to affect health and quality of life. Knowledge about the burden and predictors could aid in their prevention and management. Most of the studies are from high-income countries and focus on severe cases. We did this study to estimate the prevalence and identify the characteristics and predictors of Long COVID among our patients. Methodology: We recruited adult ([≥]18 years) patients who were diagnosed as Reverse Transcription Polymerase Chain Reaction (RTPCR) confirmed SARS-COV-2 infection and were either hospitalized or tested on outpatient basis. Eligible participants were followed up telephonically after four weeks of diagnosis of SARS-COV-2 infection to collect data on sociodemographic, clinical history, vaccination history, Cycle threshold (Ct) values during diagnosis and other variables. Characteristic of Long COVID were elicited, and multivariable logistic regression was done to find the predictors of Long COVID. Results: We have analyzed 487 individual data with a median follow-up of 44 days (Inter quartile range (IQR): 39,47). Overall, Long COVID was reported by 29.2% (95% Confidence interval (CI): 25.3%,33.4%) participants. Prevalence of Long COVID among patients with mild/moderate disease (n = 415) was 23.4% (95% CI: 19.5%,27.7%) as compared to 62.5% (95% CI: 50.7%,73%) in severe/critical cases(n=72). The most common Long COVID symptom was fatigue (64.8%) followed by cough (32.4%). Statistically significant predictors of Long COVID were - Pre-existing medical conditions (Adjusted Odds ratio (aOR)=2.00, 95% CI: 1.16,3.44), having a more significant number of symptoms during acute phase of COVID-19 disease (aOR=11.24, 95% CI: 4.00,31.51), two doses of COVID-19 vaccination (aOR=2.32, 95% CI: 1.17,4.58), the severity of illness (aOR=5.71, 95% CI: 3.00,10.89) and being admitted to hospital (Odds ratio (OR)=3.89, 95% CI: 2.49,6.08). Conclusion: A considerable proportion of COVID-19 cases reported Long COVID symptoms. More research is needed in Long COVID to objectively assess the symptoms and find the biological and radiological markers. 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 January 8, 2022. ; https://doi.org/10.1101/2022.01.04.21268536 doi: medRxiv preprint Studies have shown that Long COVID can affect almost all systems in the body. (12) The most described in the literature are respiratory disorders, cardiovascular disorders, neurocognitive disorders, mental health disorders, metabolic disorders etc. Symptoms are in multitudes: including fatigue, breathlessness, cough, anxiety, depression, palpitation, chest main, myalgia, cognitive dysfunction ("brain fog"), loss of smell, etc. Institute of Health in the USA has also launched new initiatives to study Long COVID and is expected to bring out more evidence. (6) Independent researchers are also working on understanding this phenomenon. In India, significantly less attention has been given to the burden of Long COVID. (16) During this study's conception, there were no research papers available in peerreviewed journals that measured the burden of Long COVID in India. As of today, India has more than 34 million cases of COVID-19. (17) 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 January 8, 2022. ; https://doi.org/10.1101/2022.01.04.21268536 doi: medRxiv preprint wide evidence gap on this condition in India. Thus, it is pertinent that we undertake a study to measure the burden, the characteristics, and the predictors of long COVID in India to bring much-needed insight into this condition. We estimated the prevalence, characteristics, and predictors of Long COVID by following up a cohort of patients who were Revere Transcription polymerase chain reaction (RTPCR) positive COVID-19 cases. The study was conducted at All India Institute of Medical Sciences (AIIMS) Bhubaneswar, a tertiary care government hospital and research institute. The study population included adult cases (age ≥18 years) of COVID-19 who were diagnosed with RTPCR test from AIIMS Bhubaneswar from April to September. Individuals less than 18 years and pregnant women were excluded. We accessed the AIIMS Bhubaneswar COVID-19 screening OPD database and records of patients admitted due to COVID-19. The database was cleaned by removing individuals with missing phone numbers, patients who expired, and those less than 18 years. As per the operational definition based on NICE guidelines, these individuals were contacted through telephone after four weeks from the date of their COVID-19 diagnosis. After taking verbal consent, a detailed telephonic interview was is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint . (21) All the data in this study was collected by the postgraduate student authors. Pre-testing of the questionnaire was done, and supervised calls were made before the beginning of actual data collection. The data collected during telephonic interviews were directly entered into EpiCollect5 app. An individual who could not be contacted after two attempts were excluded. The RTPCR cycle threshold (Ct) values during diagnosis of COVID-19 were retrieved from the hospital database to study its association with Long COVID symptoms. The sample size for the study was calculated separately for mild to moderate cases and severe cases. Based on previous estimates, a 20% prevalence was taken for mild to moderate cases and the required sample size was 400. (14) For severe cases we assumed a 50% prevalence of Long COVID, and the sample size was calculated to be 100. A relative precision of 20% was used in both calculations. Data was collected using EpiCollect5 and imported into Microsoft Excel for cleaning. The data was analyzed in statistical software R (version 3.6.3). Prevalence of Long COVID was determined by the number of participants who self-reported any of the Long COVID symptoms. The self-reported characteristics of symptoms were also given as proportions. The data were analyzed separately for mild to moderate and severe to critical patients. Logistics regression was used to find the predictive factors of Long COVID. Statistical significance for univariable analysis was set at p-value less . CC-BY-NC-ND 4.0 International license 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) preprint The copyright holder for this this version posted January 8, 2022. ; https://doi.org/10.1101/2022.01.04.21268536 doi: medRxiv preprint than 0.05. Multivariable logistic regression was done to obtain an adjusted odds ratio with a 95% confidence interval. Clinically significant, and variables which gives a pvalue less than 0.2 in univariable analysis were added to the multivariable model. The institutional ethics committee (IEC) of AIIMS Bhubaneswar granted ethical approval before starting the study (IEC Number: T/IM-NF/CM&FM/21/37). The study was explained to each individual, and a telephonic verbal consent was taken before starting data collection. The consent process was approved by the Institutional Ethics Committee. After data collection, if the participant was found to have symptoms of Long COVID, they were referred to Long COVID OPD in the department of Pulmonary Medicine, AIIMS Bhubaneswar. We listed 698 COVID-19 RTPCR positive cases from April to September 2021, out of which 189 patients could not be contacted. A total of 509 individuals were eligible to be included into the study. Consent was denied by nine participants, and thus a total of 500 interviews were conducted successfully. On the preliminary evaluation of data, thirteen entries had wrong dates and were dropped from the final analysis. 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 January 8, 2022. ; https://doi.org/10.1101/2022.01.04.21268536 doi: medRxiv preprint The mean age of the study participants was 39 years (SD=15 years), ranging from 18 to 88 years. Females were 199 (40.9%), and the majority of participants were graduates. Most of the participants were either unemployed or students or homemakers with no earnings. Thirty participants (6.2%) reported being in a job involving COVID-19 management. Majority of participants had normal BMI. (Table 1) Eighteen participants (3.7%) reported that they had COVID-19 before the current episode. Around 10% had a history of pre-existing Diabetes or Hypertension. Few participants gave the history of other comorbidities like asthma, tuberculosis, anxiety, cancer, or other chronic diseases, and none reported depression. A single question was used to record any type of self-reported substance use, and 54 (11.1%) • Did not give consent = 9 Total number of participants enrolled into the study = 500 • Error in dates = 13 . CC-BY-NC-ND 4.0 International license 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) preprint The copyright holder for this this version posted January 8, 2022. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint 19.5%,27.7%). (Figure 2 ) Among participants who were asymptomatic during the acute phase of COVID-19 (n=111), only six reported Long COVID symptoms. The most common symptom reported was Fatigue 92 (64.8%), followed by Cough 46 (32.4%). Only three participants reported cognitive dysfunction or Brain fog. Limitation of daily activity following Long COVID was not reported by the majority, but 41 (28.9%) participants reported having some activity limitation. Out of the 142 participants who self-reported Long COVID, 131 (92.3%) perceived the symptoms to be not severe, whereas 11 (7.7%) experienced the symptoms a lot. Health care practitioners were consulted for Long COVID by 49 (34.5%) participants. (Table 3) is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) 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 January 8, 2022. ; https://doi.org/10.1101/2022.01.04.21268536 doi: medRxiv preprint during the acute phase of disease was significantly associated with Long COVID (Odds ratio = 3.89 (95% CI: 2.49,6.08)); however, this variable was not included in the multivariable model due to the colinear relation with the severity of the disease. (Table 4 ) 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 January 8, 2022. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) 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 January 8, 2022. ; https://doi.org/10.1101/2022.01.04.21268536 doi: medRxiv preprint COVID in hospitalized patients of North India, gave an estimate of 40.3% after 4 to 6 weeks follow-up. (26) High prevalence of Long COVID symptoms in severe and hospitalized cases are reported from multiple studies from all over the world. (27, 28) The most common Long COVID symptoms found in our study were fatigue and cough. This is similar to other studies from India. (23, 24, 25, 29) The Self-reported symptoms in the COVID Symptom Study app and the National Coronavirus (COVID-19) Infection Survey (CIS) from the United Kingdom has also recorded that fatigue is the most common symptom reported. (14, 15) Multiple systematic reviews and meta-analysis on Long COVID have listed fatigue as the most common or among the first three Long COVID symptoms. (12, 22, 30, 31, 32) A recent study from India reported fatigue to be present even after three months of recovery from COVID-19. (33) Although fatigue was self-reported in this study, a consistent finding in multiple studies indicates that fatigue is, in fact, the most common of Long COVID symptoms. (34) Predictors of Long COVID are important because it helps to prioritize the at-risk population and design interventions. In our study, one of the strongest predictors of Long COVID was the severity of COVID-19 disease and hospital admission. This is intuitive because the chances of having persistent symptoms after four weeks postinfection can be higher if the disease is severe. This is backed up by a systematic 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 January 8, 2022. ; https://doi.org/10.1101/2022.01.04.21268536 doi: medRxiv preprint and a systematic review on this topic has found a similar and strong association between the pre-existing condition and Long COVID. (25, 30) An observational paradox in our study was that the participants who took two doses of COVID-19 vaccination had higher odds of developing Long COVID. It could be due to better survival in vaccinated individuals who may continue to exhibit symptoms of COVID-19 disease. But we could not find any literature on this association, and based on this study, we cannot imply causation. Age and sex, which was commonly found to be associated with Long COVID was not a significant predictor in our study. Cycle threshold (Ct) values of two genes were also not a significant predictor of Long COVID. The strength of this study is that all the cases of COVID-19 were diagnosed with RTPCR, and there is minimal risk of misclassification. The questionnaire used to capture the Long COVID was adapted from the standard case reporting format recommended by W.H.O. The data were collected by doctors involved in patient care and which improves the validity of the findings. Our study also had limitations. The Long COVID symptoms were all self-reported, and thus objective assessment of symptoms like fatigue was not done. Telephonic interviews precluded us from collecting additional information like clinical and radiological examination for correlating with the findings. We assessed Long COVID after four weeks, and no further follow-up was done. The cause of death of fifty-two individuals who were not alive during the time of data collection was not enquired, and their death may have been related to Long covid complications. In developed countries, many large-scale cohort studies are undertaken to understand this phenomenon. (35,36) Similar studies on Long COVID are lacking in India, and our research community should bridge this gap. We need more research into Long COVID to objectively assess the symptoms, to monitor the symptoms for a longer duration, 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 January 8, 2022. ; https://doi.org/10.1101/2022.01.04.21268536 doi: medRxiv preprint and to study the biological and radiological markers, which can lead to better treatment guidelines and comprehensive management of COVID-19 disease. WHO Director-General's opening remarks at the media briefing on COVID-19 -30 WHO Coronavirus (COVID-19) Dashboard | WHO Coronavirus Dashboard with Vaccination Data The latest on the covid-19 global situation & long-term sequelae How and why patients made Long Covid Chronic COVID syndrome: Need for an appropriate medical terminology for long-COVID and COVID long-haulers NIH launches new initiative to study "Long COVID" | National Institutes of Health (NIH) How Covid-19 Long Haulers Created a Movement | by Alexandra Sifferlin | Medium Coronavirus Blog NICE guideline on long COVID Overview | COVID-19 rapid guideline: managing the long-term effects of COVID A clinical case definition of post COVID-19 condition by a Delphi consensus More than 50 long-term effects of COVID-19: a systematic review and meta-analysis COVID postural tachycardia syndrome: an American Autonomic Society statement The prevalence of long COVID symptoms and COVID-19 complications -Office for National Statistics Attributes and predictors of long COVID Shadow of Long Covid: Why India needs to prepare for long-term effects of coronavirus -Coronavirus Outbreak News. INDIA TODAY Delhi's Post-Covid Clinic for Recovered Patients With Fresh Symptoms Opens. NDTV NEWS National Comprehensive Guidelines for Management of PostCovid Sequelae Global COVID-19 Clinical Platform Case Report Form (CRF) for Post COVID condition (Post COVID-19 CRF Characterising long COVID: a living systematic review Post COVID-19 sequelae: A prospective observational study from Northern India Elucidating Post-COVID-19 manifestations in India. medRxiv Long Term Health Consequences of COVID-19 in Hospitalized Patients from North India: A follow up study of upto 12 months Long-COVID': a cross-sectional study of persisting symptoms, biomarker and imaging abnormalities following hospitalisation for COVID-19 Covid-19 symptoms: Longitudinal evolution and persistence in outpatient settings Post-COVID-19 symptoms are not uncommon among recovered patients-A crosssectional online survey among the Indian population Frequency, signs and symptoms, and criteria adopted for long COVID-19: A systematic review Characteristics and predictors of acute and chronic post-COVID syndrome: A systematic review and meta-analysis The occurrence of long COVID: a rapid review Manifestations and risk factors of post COVID syndrome among COVID-19 patients presented with minimal symptoms -A study from Kerala Fatigue: Potential Contributing Factors 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) preprint 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 January 8, 2022. ; https://doi.org/10.1101/2022.01.04.21268536 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 January 8, 2022. ; https://doi.org/10.1101/2022.01.04.21268536 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 January 8, 2022. 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 January 8, 2022. ; https://doi.org/10.1101/2022.01.04.21268536 doi: medRxiv preprint