Research Medical Officer University of Cape Town Cape Town, South Africa
Background: Due to its affordability, warfarin is the most widely used oral anticoagulant in Africa. Warfarin requires ongoing monitoring of the International Normalised Ratio (INR) to ensure that the INR remains in the therapeutic range.
Objectives: To quantify anticoagulation control in 9 sub-Saharan African countries and identify associations with adequate anticoagulation control.
Methods: We conducted a cross-sectional record review at 18 sites in 9 countries (Botswana, the Democratic Republic of Congo, Ethiopia, Gambia, Ghana, Mozambique, Nigeria, South Africa,Tanzania). We collected data on patient characteristics, indications for anticoagulation, mode of payment for healthcare, and INR results for up to 12 months of follow-up between January 2019 and December 2021. Inclusion criteria for this analysis were: At least 4 INR results with a gap of no more than 56 days between consecutive INRs, at least 28 days between 1st and 4th INR. We used the Rosendaal method to calculate time in Therapeutic Range (TTR). Adequate anticoagulation control was defined as TTR of ≥ 65%. To explore the impact of COVID-19, we categorized patients as treated in the “COVID-period” if any of their follow-up occurred during 2020 or 2021. We constructed a multivariable logistic regression model to identify associations with adequate INR control.
Results: Of 989 records reviewed, 659 patients fulfilled the inclusion criteria, median age 51 years (interquartile range (IQR) 39-64), 438/659 (66%) women. Anticoagulation control was poor: only 133/659 (20%) were adequately anticoagulated with a median TTR of 36% (IQR 16% to 57%). 51/198 (26%) with atrial fibrillation/flutter, 50/240 (21%) with venous thromboembolism, 24/149 (16%) with prosthetic valves and 8/72 (11%) with “other” indication for anticoagulation achieved TTR ≥ 65%. In a model adjusted for age, sex, and follow-up period (referent atrial fibrillation/flutter), patients with prosthetic valves and “other” indication were less likely to have adequate INR control, adjusted odds ratios (aORs) 0.50 (95% CI 0.26-0.93) and 0.27 (0.08-0.73) respectively. Compared to universal health coverage, patients who paid out-of-pocket for care, and those with health insurance were less likely to have adequate INR control, aORs 0.52 (95% CI 0.32-0.82) and 0.37 (95% CI 0.16-0.77) respectively.
Conclusions: INR control was suboptimal across all included countries and indications for anticoagulation. Control was better for those with universal health insurance, suggesting that payment for medication and monitoring poses a barrier to achieving good anticoagulation control. Access to affordable alternative oral anticoagulants where appropriate is an urgent need in Sub-Saharan Africa.
