The algorithm uses lifestyle factors and medical history to predict the absolute risk of developing venous thromboembolism over one and five years, expressed as a percentage.
Researchers from Nottingham University said the tool could easily integrate into GP computer systems to identify patients most likely to benefit from early intervention.
This could include assessing patients' risk prior to hospital admission, long-haul flights, or starting a course of medicine with an increased clotting risk, they said.
There is no current algorithmic tool to predict clot risk in primary care, despite NICE guidance encouraging early detection of high-risk patients.
The Nottingham team used data on 3.5 million patients aged 25-84 years with no previous history of blood clots, gleaned from 564 practice systems.
They found that the risk of developing venous thromboembolism increased with factors including age, BMI, smoking status and presence of vascular, renal and respiratory disease.
Each year, around 15 cases occurred per 10,000 people.
They used the data to create risk models, which were evaluated for accuracy at one and five years.
GPs can input a patient’s history into the tool and it will return one- and five-year predictions of risk of developing venous thromboembolism, expressed as percentage scores.
The authors concluded: ‘The algorithm is based on simple clinical variables which the patient is likely to know or which are routinely recorded in GP computer systems.
‘The algorithm could be integrated into GP computer systems and used to risk assess patients prior to hospital admission or prior to the initiation of medication which might increase risk of venous thromboembolism.’
They added: ‘Further research is needed to assess how best to use the algorithm and whether, upon implementation, it has any impact on health outcomes.’
- The tool can be accessed online at www.qthrombosis.org
