Prophylaxis of Venous Thromboembolism in Geriatric Settings: A Cluster-Randomized Multicomponent Interventional Trial

J Am Med Dir Assoc. 2018 Jun;19(6):497-503. doi: 10.1016/j.jamda.2018.02.004. Epub 2018 Mar 24.


Objectives: To evaluate the efficacy of an intervention on the practice of venous thromboembolism prevention.

Design: A multicenter, prospective, controlled, cluster-randomized, multifaceted intervention trial consisting of educational lectures, posters, and pocket cards reminding physicians of the guidelines for thromboprophylaxis use.

Settings: Twelve geriatric departments with 1861 beds total, of which 202, 803, and 856 in acute care, post-acute care, and long-term care wards, respectively.

Participants: Patients hospitalized between January 1 and May 31, 2015, in participating departments.

Measurements: The primary endpoint was the overall adequacy of thromboprophylaxis prescription at the patient level, defined as a composite endpoint consisting of indication, regimen, and duration of treatment. Geriatric departments were divided into an intervention group (6 departments) and control group (6 departments). The preintervention period was 1 month to provide baseline practice levels, the intervention period 2 months, and the postintervention period 1 month in acute care and post-acute care wards or 2 months in long-term care wards. Multivariable regression was used to analyze factors associated with the composite outcome.

Results: We included 2962 patients (1426 preintervention and 1536 postintervention), with median age 85 [79;90] years. For the overall 18.9% rate of inadequate thromboprophylaxis, 11.1% was attributable to underuse and 7.9% overuse. Intervention effects were more apparent in post-acute and long-term care wards although not significantly [odds ratio 1.44 (95% confidence interval 0.78;2.66), P = .241; and 1.44 (0.68, 3.06), P = .345]. Adequacy rates significantly improved in the postintervention period for the intervention group overall (from 78.9% to 83.4%; P = .027) and in post-acute care (from 75.4% to 86.3%; P = .004) and long-term care (from 87.0% to 91.7%; P = .050) wards, with no significant trend observed in the control group.

Conclusions/implications: This study failed to demonstrate improvement in prophylaxis adequacy with our intervention. However, the intervention seemed to improve practices in post-acute and long-term care but not acute care wards.

Keywords: Thromboprophylaxis; cluster-randomized trial; educational intervention; older patients.

Publication types

  • Multicenter Study
  • Randomized Controlled Trial

MeSH terms

  • Aged
  • Aged, 80 and over
  • Anticoagulants / therapeutic use*
  • Female
  • France
  • Humans
  • Inpatients*
  • Male
  • Outcome Assessment, Health Care
  • Prospective Studies
  • Venous Thromboembolism / prevention & control*


  • Anticoagulants