Propensity Score Matching Analysis to Study the Effects of Low Molecular Weight Heparin in Retrospective COVID-19 Patient Real Life Data

Featuring Valeria Rolle, Biostatistician at ISPA

Abstract: 'Propensity Score Matching Analysis to Study the Effects of Low Molecular Weight Heparin in Retrospective COVID-19 Patient Real Life Data'.

Background: Coronavirus disease 2019 (COVID-19) infection is often associated with coagulopathies among hospitalized patients. Recent research found that treatment with anticoagulants improves the chances of survival of these patients. In this study, we aimed to assess the impact of treatment with low molecular weight heparin (LMWH) on the survival of COVID-19 patients, by using propensity score matching (PSM) to account for inter-patient variability.

Methods: 1937 patients with confirmed COVID-19 were hospitalized in the "HM Hospitales" network of hospitals in Spain. This database was retrospectively analyzed to evaluate the effect of LMWH. They were divided into LMWH-untreated and treated with only LMWH. A Cox proportional hazards model was used to estimate the effect of LMWH administration on in-hospital mortality. We adjusted for age, sex, history of hypertension, diabetes, heart failure and atrial fibrillation. Furthermore, a 1 :1 PSM analysis was performed to minimize the potential confounding factors, via the optimal method. All analysis were performed with R (version 3.6.1 ).

Results: In-hospital mortality for patients treated with LMWH was 12.6%, compared to 15.1 % in patients who did not receive it. In the multivariate Cox proportional hazards model, treatment with LMWH was associated with a reduced risk of mortality (adjusted HR of 0.38 per day, confidence interval 0.27-0.51, p<0.001 the results were further improved in the 1:1 PSM-matched group (adjusted HR of 0.17 per day confidence interval 0.09-0.31, p<0.001).

Conclusions: PSM effectively accounts for the high variability encountered in COVID-19 patients and supports the notion that LMWH as a treatment that improves in-hospital survival.

Interested in data services? Click on the button to read about our expertise

We are experts in study design, statistical analysis, data science, data capture and reporting for clinical trials