Observational Study of the Efficiency of Treatments in Patients Hospitalized with Covid-19 in Madrid

Abstract

Background Many different treatments were heavily administered to patients with COVID-19 during the peak of the pandemic in Madrid without robust evidence supporting them. Methods We examined the association between sixteen treatments in four groups (steroids, antivirals, antibiotics and immunomodulators) and intubation or death. Data were obtained from patients that were admitted to an HM hospital with suspicion of COVID-19 until 24/04/2020, excluding unconfirmed diagnosis, those who were admitted before the epidemic started in Madrid, had an outcome that was not discharge or death or died within 24 hours of presentation. We compared outcomes between treated and untreated patients using propensity-score caliper matching. Results Of 2,307 patients in the dataset, 679 were excluded. Of the remaining 1,645 patients, 263 (16%) died and 311 (18.9%) died or were intubated. Except for hydroxychloroquine and prednisone, patients that were treated with any of the medications were more likely to go through an outcome of death or intubation at baseline. After propensity matching we found an association between treatment with hydroxychloroquine and prednisone and better outcomes (hazard ratios with 95% CI of 0.83 +- 0.06 and 0.85 +- 0.03). Results were similar in multiple sensitivity analyses. Conclusions In this multicenter study of patients admitted with COVID-19 hydroxychloroquine and prednisone administration was found to be associated with improved outcomes. Other treatments were associated with no effect or worse outcomes. Randomized, controlled trials of these medications in patients with COVID-19 are needed to avoid heavy administration of treatments with no strong evidence to support them.

Machine Learning for Mortality Analysis in Patients with COVID-19

“…In this paper, we analyze the anonymized dataset obtained from the HMhospitals’ network in Madrid (Spain), thanks to their project COVID DATA SAVE LIVES [ 34 ]. In [ 35 ], several hypothesis contrasts and the chi-squared and Wilcoxon tests were performed using this dataset. However, model-based statistical techniques and machine learning techniques were uncovered.…”

Randomized Controlled Trials of Early Ambulatory Hydroxychloroquine in the Prevention of COVID-19 Infection, Hospitalization, and Death: Meta-Analysis

“…4 5 6 7 8 9 In addition, a number of studies examining HCQ use in the hospitalized setting—particularly use starting within 24-48 hours of admission—have also shown benefit. 19 20 21 22 23 24 That modern, well-conducted non-randomized trials and observational studies would yield results similar to those of RCTs is not surprising as it is the common finding across numerous medical disciplines. 25 26…”

Hydroxychloroquine in the post-COVID-19 era: will this pandemic upset decades of clinical practice?

“…HCQ then was included in national recommendations in many countries as part of the strategy to fight the pandemic. Table 1 lists the main studies where HCQ was used in the treatment of COVID-19, e.g., retrospective studies of more than 1000 patients and randomized controlled studies [21][22][23][24][25][26][27][28][29][30][31][32][33][34][35][36][37][38][39][40]. Whilst most retrospective studies reported clinical improvement with HCQ, randomized controlled studies were stopped because of inefficacy and, eventually, a tendency towards increased overall mortality.…”