Doctors need reliable ways to spot patients at risk for dangerous blood clots, known as VTE. This review looked at fourteen different studies to see how well current prediction models work for people with multiple myeloma. The goal was to check if these tools could help prevent serious health problems.
The results showed that the models are not very good. When doctors used these tools, they found that about 7.9% of patients actually got blood clots. However, the models were not accurate enough to catch all the risks. The accuracy scores, called AUC, were low, ranging from 0.57 to 0.68. A score below 0.7 is generally considered poor for medical predictions.
Several problems were found with the research. Most of the studies had a high risk of errors in how they measured results. Only two studies checked if the models fit the real data well. Also, all the studies shared the same mathematical formulas, which might limit how well they work in different hospitals. Because of these issues, the review says these tools have limited use in real life.