Table 1 Simulation comparison of VLAD and RA-CUSUM approaches
From: Debate: what is the best method to monitor surgical performance?
Scenario: expected, actual mortality rate | Method | Proportion of chart exceeding the control limit at patient | ||||||
|---|---|---|---|---|---|---|---|---|
10 | 25 | 50 | 75 | 100 | 150 | 200 | ||
1: 5Â %, 5Â % | VLAD | 0.01 | 0.02 | 0.02 | 0.02 | 0.03 | 0.02 | 0.03 |
RA-CUSUM | 0.00 | 0.00 | 0.01 | 0.01 | 0.02 | 0.02 | 0.03 | |
2: 1Â %, 2Â % | VLAD | 0.01 | 0.08 | 0.08 | 0.06 | 0.14 | 0.18 | 0.21 |
RA-CUSUM | 0.00 | 0.01 | 0.05 | 0.08 | 0.15 | 0.22 | 0.32 | |
3: 10 %, 10 % but with 6 deaths in a row patients 94–99 | VLAD | 0.02 | 0.02 | 0.02 | 0.02 | 0.32 | 0.01 | 0.01 |
RA-CUSUM | 0.00 | 0.00 | 0.00 | 0.01 | 1.00 | 0.37 | 0.22 | |
4: 10Â %, 10Â % for first 100 patients, then 12.5Â % | VLAD | 0.01 | 0.01 | 0.02 | 0.02 | 0.02 | 0.06 | 0.07 |
RA-CUSUM | 0.00 | 0.00 | 0.01 | 0.01 | 0.01 | 0.07 | 0.12 | |