Abstract: medRxiv preprint doi: https://doi.org/10.1101/2020.09.09.20184143; this version posted September 13, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. All rights reserved. No reuse allowed without permission. The Effect of Early Hydroxychloroquine-based Therapy in COVID-19 Patients in Ambulatory Care Settings: A Nationwide Prospective Cohort Study Authors: Tarek Sulaiman, Abdulrhman Mohana, Laila Alawdah, Nagla Mahmoud, Mustafa Hassanein, Tariq Wani, Amel Alfaifi, Eissa Alenazi, Nashwa Radwan, Nasser AlKhalifah, Ehab Elkady, Manwer Alanazi, Mohammed Alqahtani, Khalid Abdullah, Yousif Yousif, Fouad AboGazalah, Fuad Awwad, Khaled Alabdulkareem, Fahad AlGhofaili, Ahmad AlJedai, Hani Jokhdar, Fahad Alrabiah ABSTRACT BACKGROUND: Currently, there is no proven effective therapy nor vaccine for the treatment of SARS-CoV-2. Evidence regarding the potential benefit of early administration of hydroxychloroquine (HCQ) therapy in symptomatic patients with Coronavirus Disease (COVID-19) is not clear. METHODS: This observational prospective cohort study took place in 238 ambulatory fever clinics in Saudi Arabia, which followed the Ministry of Health (MOH) COVID-19 treatment guideline. This guideline included multiple treatment options for COVID-19 based on the best available evidence at the time, among which was Hydroxychloroquine (HCQ). Patients with confirmed COVD-19 (by reverse transcriptase polymerase chain reaction (PCR) test) who presented to these clinics with mild to moderate symptoms during the period from 5-26 June 2020 were included in this study. Our study looked at those who received HCQ-based therapy along with supportive care (SC) and compared them to patients who received SC alone. The primary outcome was hospital admission within 28-days of presentation. The secondary outcome was a composite of intensive care admission (ICU) and/or mortality during the followup period. Outcome data were assessed through a follow-up telephonic questionnaire at day 28 and were further verified with national hospitalisation and mortality registries. Multiple logistic regression model was used to control for prespecified confounders. RESULTS: Of the 7,892 symptomatic PCR-confirmed COVID-19 patients who visited the ambulatory fever clinics during the study period, 5,541 had verified clinical outcomes at day 28 (1,817 patients in the HCQ group vs 3,724 in the SC group). At baseline, patients who received HCQ therapy were more likely to be males who did not have hypertension or chronic lung disease compared to the SC group. No major differences were noted regarding other comorbid conditions. All patients were presenting with active complaints; however, the HCQ groups had higher rates of symptoms compared to the SC group (fever: 84% vs 66.3, headache: 49.8 vs 37.4, cough: 44.5 vs 35.6, respectively). Early HCQ-based therapy was associated with a lower hospital admission within 28-days compared to SC alone (9.4% compared to 16.6%, RRR 43%, p-value <0.001). The composite outcome of ICU admission and/or mortality at 28days was also lower in the HCQ group compared to the SC (1.2% compared to 2.6%, RRR 54%, p-value 0.001). Adjusting for age, gender, and major comorbid conditions, a multivariate logistic regression model showed a decrease in the odds of hospitalisation in patients who received HCQ compared to SC alone (adjusted OR..