Since Ms. Clinton will be 67.8 years old in July 2015, determining her major bleeding risk in the first three months of anticoagulation for VTE should she be treated beginning in July 2015 would first require knowing the risk of major bleeding in the literature for a particular group of VTE patients and knowing the mean age of the group. We also need to remember that the major bleeding risk increases by about 7%/year.(46) A meta-analysis by Carrier et. al of 69 studies (69 randomized trials and observational studies, n=19,021, mean age 60.0 years old) found major bleeding with the initial intravenous anticoagulant treatment occurred in about 4% of patients. Data from this study is used to estimate Ms. Clinton’s bleeding risk for the first three months of anticoagulation.(45) That bleeding rate needs to be adjusted for Ms. Clinton’s age to calculate her particular three month bleeding risk. Since Ms. Clinton is 7.8 years older than the average patient in this meta-analysis (67.8 – 60.0 = 7.8), her major bleeding risk in the first three months would be about 6.8% (4% x 1.07^7.8 = 6.8%).

After the initial three months of anticoagulation for VTE, Ms. Clinton’s subsequent bleeding risk would then need to be calculated from another meta-analysis of VTE randomized controlled trials. Linkins et al. analyzed the chances of bleeding of VTE patients with long-term oral anticoagulants (n= 3961, mean age = 58.6 years).(46) The factors going into the calculation of the long-term major bleeding risk are the following:

- The risk of major bleeding with oral anticoagulants increases by about 7% per year.(45, 46)
- Based on the Linkins et al. meta-analysis, major bleeding risk after the three months of initial anticoagulation averaged about 2.1%/year.
- At 67.8 years old, Ms. Clinton is 9.2 years older than the average patient in this study (58.6 years old).
- To adjust the bleeding risk in this study for her older age than patients in the meta-analysis of VTE patients, the monthly bleeding risk is multiplied by 1.07^9.2 (The major bleeding risk increases by 7% year. So since she is 9.2 years older than the average VTE patient, the rate of increase in yearly bleeding rate is raised to the 9.2th power.)
- 0.021 (yearly major bleeding risk in the meta-analysis for people of mean age=58.6-years-old) x 1.07^9.2 = 0.0391.
- Secondly, Ms. Clinton’s proposed long-term oral anticoagulation treatment until January 2025 would be 9.5 years (9 years and 6 months), whereas the duration of anticoagulation of VTE patients in the meta-analysis averaged only 1.2 years (i.e., 8.3 years shorter).
- To determine a mean monthly major bleeding risk over the proposed longer duration of oral anticoagulation (8.3 years), we further adjust the age-adjusted initial major bleeding risk (0.0391 per year) by multiplying by 1.07^4.15

(This represents the mean increase in yearly risk for the 8.3 additional years of anticoagulation (i.e., the average bleeding risk/year of the older Ms. Clinton is determined by the 8.3 additional years / 2 = 4.15 years.); 0.0391 x 1.07^4.15 = 0.0577/year. - To account for the 81% chance of a 67-year-old woman surviving until age 77 years (U.S. Census data(41)), we adjust the monthly risk number accordingly: 0.0577 (age and duration of anticoagulation adjusted monthly major bleeding risk) x 0.81 (81% mean 10 year survival for a 67 year-old woman) = 0.0467/year (age, duration of anticoagulation, and survival adjusted mean yearly bleeding risk).
- Since the total time of long-term oral anticoagulation before January 2025 would be 9.5 years, the major bleeding risk would be 0.0467/year (mean yearly risk) x 9.5 years = 0.443 (44.3%).
- Adding the major bleeding risk in the initial three months of anticoagulation (6.6%) to the risk of 9.5 years of oral anticoagulation (44.3%) and we get the total cumulative risk of major bleeding with anticoagulation treatment for a VTE or CVT from July 2015 – January 2025 ≈ 50%.

## Leave a Reply