Nearly 1 in 4 patients with incident venous thromboembolism (VTE) were found not to initiate prescribed anticoagulation therapy in the month following hospital discharge, according to study results published in the American Journal of Medicine.
International guidelines recommend that patients who have experienced VTE be treated with anticoagulation therapy for at least 3 months. Yet, a substantial proportion of patients — particularly those prescribed warfarin regimens — never initiate outpatient pharmacotherapy after a VTE diagnosis. Identification of factors related to noninitiation could yield important information for clinicians in their efforts to improve compliance.
In this Danish nationwide cohort study, the data of 38,044 patients (median age, 66.1 years; 53.2% women) who were diagnosed with incident VTE (pulmonary embolism or incident deep vein thrombosis [DVT]) between 2003 and 2016 were analyzed. Noninitiation of therapy, the primary study outcome, was defined as the lack of claim for an anticoagulant medication prescription in the 30 days following hospital discharge.
In this cohort, there were 9,294 patients (24.1%) who failed to initiate anticoagulant therapy within 30 days of discharge, and 63.4% of patients who had incident DVT at presentation.
Noninitiation of anticoagulation was predicted by demographic variables including age <30 vs >65 years (relative risk [RR], 1.18; 95% CI, 1.13-1.33) and female sex (RR, 1.30; 95% CI, 1.25-1.34), and clinical factors including unprovoked VTE (RR, 1.13; 95% CI, 1.08-1.17), incident DVT (RR, 1.91; 95% CI, 1.81-2.01), and hospitalization lasting 0 to 3 days vs >3 days (RR, 1.96; 95% CI, 1.87-2.07). The impact on the absence of therapy initiation of socioeconomic variables vs above-mentioned factors was lower.
Chronic comorbidities that were found to predict noninitiation of anticoagulation were: ischemic heart disease (RR, 1.20; 95% CI, 1.13-1.28), congestive heart failure (RR, 1.27; 95% CI, 1.17-1.37), and liver disease (RR, 1.60; 95% CI, 1.42-1.81).
Study limitations include the possibility that some patients were incorrectly coded, the inability to rule out the occurrence of asymptomatic VTE not requiring therapy, the lack of information regarding physicians’ decisions not to prescribe treatment, a possible overestimation of the noninitiator population, and the lack of accounting for patients who filled but did not take their medication.
“Identification of clinical predictors of noninitiation may enable implementation of patient-tailored strategies to improve adherence and thereby potentially prevent venous thromboembolism morbidity, mortality, and recurrence,” noted the authors.
Albertsen IE, Goldhaber SZ, Piazza G, et al. Predictors of not initiating anticoagulation after incident venous thromboembolism: a Danish nationwide cohort study. Am J Med. 2019;133(4):463-472. doi:10.1016/j.amjmed.2019.08.051