Intended for healthcare professionals

Practice Uncertainties

Which is the best model to assess risk for venous thromboembolism in hospitalised patients?

BMJ 2021; 373 doi: (Published 27 May 2021) Cite this as: BMJ 2021;373:n1106
  1. Daniel Horner, consultant in emergency and intensive care medicine1 2 3,
  2. Steve Goodacre, professor of emergency medicine3,
  3. Sarah Davis, senior lecturer in health economics3,
  4. Neil Burton, patient representative4,
  5. Beverley J Hunt, professor of thrombosis and haemostasis5
  1. 1Salford Royal NHS Foundation Trust, Salford, UK
  2. 2Division of Infection, Immunity and Respiratory Medicine, University of Manchester, Manchester, UK
  3. 3Centre for Urgent and Emergency Care Research (CURE), University of Sheffield, Sheffield, UK
  4. 4Thrombosis UK, Llanwrda, UK
  5. 5Kings Healthcare Partners & Thrombosis & Haemophilia Centre, Guy’s & St Thomas’ NHS Foundation Trust, London, UK
  1. Correspondence to D Horner danielhorner{at}

What you need to know

  • Venous thromboembolism in hospitalised patients can be potentially prevented through patient education and pharmacological thromboprophylaxis

  • Risk assessment models (RAMs) help clinicians decide who should be offered pharmacological thromboprophylaxis, but variation exists in their composition of risk factors and thresholds for high and low risk

  • Uncertainty exists over which RAM is optimal for hospitalised patients and whether any complex RAM outperforms simple criteria or subjective clinical opinion

Venous thromboembolism (VTE), which includes deep vein thrombosis (DVT) and pulmonary embolism (PE), is a major global health burden. North American data report a 30 day case fatality rate of 10.6% following VTE. Between 30% and 50% of survivors go on to have long term complications.12 About half of VTE episodes occur during hospitalisation for surgery or acute medical illness, or within 90 days from discharge. These events are classified as hospital acquired thrombosis (HAT).3

HAT events are potentially preventable through patient education and pharmacological thromboprophylaxis. A meta-analysis (seven trials, 15 095 hospitalised patients) showed greater than 50% risk reduction for VTE with heparins compared with control.4 In many elective surgical settings, thromboprophylaxis has become established practice.56

However, pharmacological thromboprophylaxis is not suitable for all patients admitted to hospital in an emergency. It can increase the baseline risk of major bleeding by approximately 0.4%.678910 When given inappropriately, the consequences can be potentially harmful, notably for patients with occult bleeding on admission or those undergoing emergency procedures.

VTE risk assessment models (RAMs) aim to minimise unnecessary pharmacological thromboprophylaxis and reduce the associated harm and costs. They can also potentially provide individualised and reproducible evaluation of VTE risk, independent of seniority, expertise or bias of the assessing clinician. Fifteen published RAMs were identified in a recent overview of systematic reviews.11 RAMs overlap on individual risk …

View Full Text

Log in

Log in through your institution


* For online subscription